{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.5675160474615835,
  "eval_steps": 3000,
  "global_step": 33000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 7.428324794769287,
      "epoch": 0.00038902937171756465,
      "grad_norm": 1.3359375,
      "learning_rate": 2e-06,
      "loss": 7.0654,
      "mean_token_accuracy": 0.09049141183495521,
      "num_tokens": 8755.0,
      "step": 5
    },
    {
      "entropy": 7.406369161605835,
      "epoch": 0.0007780587434351293,
      "grad_norm": 1.5625,
      "learning_rate": 4.5e-06,
      "loss": 6.9787,
      "mean_token_accuracy": 0.09639345631003379,
      "num_tokens": 17641.0,
      "step": 10
    },
    {
      "entropy": 7.410000467300415,
      "epoch": 0.001167088115152694,
      "grad_norm": 1.53125,
      "learning_rate": 7e-06,
      "loss": 6.9408,
      "mean_token_accuracy": 0.09870931878685951,
      "num_tokens": 26293.0,
      "step": 15
    },
    {
      "entropy": 7.406883716583252,
      "epoch": 0.0015561174868702586,
      "grad_norm": 1.5,
      "learning_rate": 9.5e-06,
      "loss": 7.0309,
      "mean_token_accuracy": 0.08720004037022591,
      "num_tokens": 35627.0,
      "step": 20
    },
    {
      "entropy": 7.423058557510376,
      "epoch": 0.0019451468585878235,
      "grad_norm": 1.1328125,
      "learning_rate": 1.2e-05,
      "loss": 6.8901,
      "mean_token_accuracy": 0.09724865928292274,
      "num_tokens": 43861.0,
      "step": 25
    },
    {
      "entropy": 7.4552548885345455,
      "epoch": 0.002334176230305388,
      "grad_norm": 1.140625,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 7.0147,
      "mean_token_accuracy": 0.0973876304924488,
      "num_tokens": 52816.0,
      "step": 30
    },
    {
      "entropy": 7.438412618637085,
      "epoch": 0.0027232056020229526,
      "grad_norm": 1.15625,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 6.9921,
      "mean_token_accuracy": 0.09515059664845467,
      "num_tokens": 61370.0,
      "step": 35
    },
    {
      "entropy": 7.441454601287842,
      "epoch": 0.0031122349737405172,
      "grad_norm": 1.125,
      "learning_rate": 1.95e-05,
      "loss": 6.9487,
      "mean_token_accuracy": 0.10167206972837448,
      "num_tokens": 69444.0,
      "step": 40
    },
    {
      "entropy": 7.507920455932617,
      "epoch": 0.003501264345458082,
      "grad_norm": 1.046875,
      "learning_rate": 2.2e-05,
      "loss": 6.9887,
      "mean_token_accuracy": 0.09499939382076264,
      "num_tokens": 77892.0,
      "step": 45
    },
    {
      "entropy": 7.525271081924439,
      "epoch": 0.003890293717175647,
      "grad_norm": 1.03125,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 6.9882,
      "mean_token_accuracy": 0.10243152678012848,
      "num_tokens": 86154.0,
      "step": 50
    },
    {
      "entropy": 7.477654314041137,
      "epoch": 0.004279323088893211,
      "grad_norm": 1.0546875,
      "learning_rate": 2.7e-05,
      "loss": 7.0953,
      "mean_token_accuracy": 0.09470769762992859,
      "num_tokens": 95589.0,
      "step": 55
    },
    {
      "entropy": 7.454274988174438,
      "epoch": 0.004668352460610776,
      "grad_norm": 0.98046875,
      "learning_rate": 2.95e-05,
      "loss": 7.0717,
      "mean_token_accuracy": 0.09005340784788132,
      "num_tokens": 105268.0,
      "step": 60
    },
    {
      "entropy": 7.460178375244141,
      "epoch": 0.0050573818323283405,
      "grad_norm": 1.0390625,
      "learning_rate": 3.2e-05,
      "loss": 6.9674,
      "mean_token_accuracy": 0.10172970369458198,
      "num_tokens": 113347.0,
      "step": 65
    },
    {
      "entropy": 7.5014667987823485,
      "epoch": 0.005446411204045905,
      "grad_norm": 1.2265625,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 6.9797,
      "mean_token_accuracy": 0.0966665655374527,
      "num_tokens": 121235.0,
      "step": 70
    },
    {
      "entropy": 7.449297571182251,
      "epoch": 0.00583544057576347,
      "grad_norm": 1.0546875,
      "learning_rate": 3.7e-05,
      "loss": 6.9313,
      "mean_token_accuracy": 0.1067626990377903,
      "num_tokens": 129529.0,
      "step": 75
    },
    {
      "entropy": 7.443627548217774,
      "epoch": 0.0062244699474810344,
      "grad_norm": 1.1171875,
      "learning_rate": 3.95e-05,
      "loss": 6.9599,
      "mean_token_accuracy": 0.09893447831273079,
      "num_tokens": 137918.0,
      "step": 80
    },
    {
      "entropy": 7.364729690551758,
      "epoch": 0.006613499319198599,
      "grad_norm": 1.1796875,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 6.8697,
      "mean_token_accuracy": 0.10501775741577149,
      "num_tokens": 146445.0,
      "step": 85
    },
    {
      "entropy": 7.389777565002442,
      "epoch": 0.007002528690916164,
      "grad_norm": 1.078125,
      "learning_rate": 4.45e-05,
      "loss": 7.054,
      "mean_token_accuracy": 0.09568696841597557,
      "num_tokens": 155904.0,
      "step": 90
    },
    {
      "entropy": 7.473219966888427,
      "epoch": 0.007391558062633729,
      "grad_norm": 1.015625,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 7.0199,
      "mean_token_accuracy": 0.1002066396176815,
      "num_tokens": 164436.0,
      "step": 95
    },
    {
      "entropy": 7.500016736984253,
      "epoch": 0.007780587434351294,
      "grad_norm": 1.0,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 6.9907,
      "mean_token_accuracy": 0.0998419389128685,
      "num_tokens": 172985.0,
      "step": 100
    },
    {
      "entropy": 7.358818578720093,
      "epoch": 0.008169616806068859,
      "grad_norm": 0.9609375,
      "learning_rate": 5.2e-05,
      "loss": 6.8944,
      "mean_token_accuracy": 0.09846496134996414,
      "num_tokens": 182084.0,
      "step": 105
    },
    {
      "entropy": 7.398036003112793,
      "epoch": 0.008558646177786422,
      "grad_norm": 1.0703125,
      "learning_rate": 5.45e-05,
      "loss": 6.9349,
      "mean_token_accuracy": 0.10587768033146858,
      "num_tokens": 190742.0,
      "step": 110
    },
    {
      "entropy": 7.430330848693847,
      "epoch": 0.008947675549503988,
      "grad_norm": 1.171875,
      "learning_rate": 5.7e-05,
      "loss": 7.009,
      "mean_token_accuracy": 0.10080742314457894,
      "num_tokens": 199600.0,
      "step": 115
    },
    {
      "entropy": 7.456536912918091,
      "epoch": 0.009336704921221552,
      "grad_norm": 1.1328125,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 6.9662,
      "mean_token_accuracy": 0.10005040019750595,
      "num_tokens": 207883.0,
      "step": 120
    },
    {
      "entropy": 7.459503984451294,
      "epoch": 0.009725734292939117,
      "grad_norm": 1.03125,
      "learning_rate": 6.2e-05,
      "loss": 7.0087,
      "mean_token_accuracy": 0.09876146614551544,
      "num_tokens": 216216.0,
      "step": 125
    },
    {
      "entropy": 7.439739942550659,
      "epoch": 0.010114763664656681,
      "grad_norm": 1.0546875,
      "learning_rate": 6.450000000000001e-05,
      "loss": 6.998,
      "mean_token_accuracy": 0.10104034319519997,
      "num_tokens": 225250.0,
      "step": 130
    },
    {
      "entropy": 7.446091413497925,
      "epoch": 0.010503793036374246,
      "grad_norm": 1.0625,
      "learning_rate": 6.7e-05,
      "loss": 6.9821,
      "mean_token_accuracy": 0.0969055637717247,
      "num_tokens": 233748.0,
      "step": 135
    },
    {
      "entropy": 7.454025363922119,
      "epoch": 0.01089282240809181,
      "grad_norm": 1.0078125,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.0076,
      "mean_token_accuracy": 0.09575914740562438,
      "num_tokens": 242629.0,
      "step": 140
    },
    {
      "entropy": 7.417604207992554,
      "epoch": 0.011281851779809376,
      "grad_norm": 1.25,
      "learning_rate": 7.2e-05,
      "loss": 6.9827,
      "mean_token_accuracy": 0.0998191438615322,
      "num_tokens": 251458.0,
      "step": 145
    },
    {
      "entropy": 7.431661653518677,
      "epoch": 0.01167088115152694,
      "grad_norm": 1.0234375,
      "learning_rate": 7.45e-05,
      "loss": 6.946,
      "mean_token_accuracy": 0.10091756880283356,
      "num_tokens": 260557.0,
      "step": 150
    },
    {
      "entropy": 7.441375064849853,
      "epoch": 0.012059910523244505,
      "grad_norm": 1.09375,
      "learning_rate": 7.7e-05,
      "loss": 7.0783,
      "mean_token_accuracy": 0.09568855166435242,
      "num_tokens": 269262.0,
      "step": 155
    },
    {
      "entropy": 7.405449151992798,
      "epoch": 0.012448939894962069,
      "grad_norm": 1.1796875,
      "learning_rate": 7.950000000000001e-05,
      "loss": 6.9389,
      "mean_token_accuracy": 0.10365158095955848,
      "num_tokens": 277804.0,
      "step": 160
    },
    {
      "entropy": 7.441110897064209,
      "epoch": 0.012837969266679634,
      "grad_norm": 1.1953125,
      "learning_rate": 8.2e-05,
      "loss": 6.9126,
      "mean_token_accuracy": 0.10729183256626129,
      "num_tokens": 287450.0,
      "step": 165
    },
    {
      "entropy": 7.436047601699829,
      "epoch": 0.013226998638397198,
      "grad_norm": 1.078125,
      "learning_rate": 8.450000000000001e-05,
      "loss": 6.9786,
      "mean_token_accuracy": 0.09779882282018662,
      "num_tokens": 296497.0,
      "step": 170
    },
    {
      "entropy": 7.377359676361084,
      "epoch": 0.013616028010114764,
      "grad_norm": 1.078125,
      "learning_rate": 8.7e-05,
      "loss": 6.8905,
      "mean_token_accuracy": 0.10251178741455078,
      "num_tokens": 305414.0,
      "step": 175
    },
    {
      "entropy": 7.3786688327789305,
      "epoch": 0.014005057381832327,
      "grad_norm": 1.03125,
      "learning_rate": 8.95e-05,
      "loss": 6.9482,
      "mean_token_accuracy": 0.10084712728857995,
      "num_tokens": 313639.0,
      "step": 180
    },
    {
      "entropy": 7.403275775909424,
      "epoch": 0.014394086753549893,
      "grad_norm": 1.078125,
      "learning_rate": 9.2e-05,
      "loss": 6.9637,
      "mean_token_accuracy": 0.10111333504319191,
      "num_tokens": 322855.0,
      "step": 185
    },
    {
      "entropy": 7.512465000152588,
      "epoch": 0.014783116125267459,
      "grad_norm": 1.09375,
      "learning_rate": 9.45e-05,
      "loss": 7.1052,
      "mean_token_accuracy": 0.09858241751790046,
      "num_tokens": 332071.0,
      "step": 190
    },
    {
      "entropy": 7.459011697769165,
      "epoch": 0.015172145496985022,
      "grad_norm": 1.2421875,
      "learning_rate": 9.7e-05,
      "loss": 6.8957,
      "mean_token_accuracy": 0.1045821987092495,
      "num_tokens": 339990.0,
      "step": 195
    },
    {
      "entropy": 7.3208598613739015,
      "epoch": 0.015561174868702588,
      "grad_norm": 1.171875,
      "learning_rate": 9.95e-05,
      "loss": 6.9542,
      "mean_token_accuracy": 0.1042078159749508,
      "num_tokens": 348581.0,
      "step": 200
    },
    {
      "entropy": 7.4775693893432615,
      "epoch": 0.01595020424042015,
      "grad_norm": 1.171875,
      "learning_rate": 0.000102,
      "loss": 6.9543,
      "mean_token_accuracy": 0.10210119113326073,
      "num_tokens": 357032.0,
      "step": 205
    },
    {
      "entropy": 7.342398071289063,
      "epoch": 0.016339233612137717,
      "grad_norm": 1.03125,
      "learning_rate": 0.00010449999999999999,
      "loss": 6.9412,
      "mean_token_accuracy": 0.09677820205688477,
      "num_tokens": 366562.0,
      "step": 210
    },
    {
      "entropy": 7.483612537384033,
      "epoch": 0.016728262983855283,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000107,
      "loss": 6.9321,
      "mean_token_accuracy": 0.10381355360150338,
      "num_tokens": 374555.0,
      "step": 215
    },
    {
      "entropy": 7.41501030921936,
      "epoch": 0.017117292355572845,
      "grad_norm": 1.125,
      "learning_rate": 0.0001095,
      "loss": 6.9808,
      "mean_token_accuracy": 0.09787384644150735,
      "num_tokens": 383583.0,
      "step": 220
    },
    {
      "entropy": 7.477314519882202,
      "epoch": 0.01750632172729041,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000112,
      "loss": 7.0253,
      "mean_token_accuracy": 0.09879604578018189,
      "num_tokens": 392932.0,
      "step": 225
    },
    {
      "entropy": 7.3631621360778805,
      "epoch": 0.017895351099007976,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001145,
      "loss": 6.8465,
      "mean_token_accuracy": 0.10572261735796928,
      "num_tokens": 401653.0,
      "step": 230
    },
    {
      "entropy": 7.390638160705566,
      "epoch": 0.01828438047072554,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00011700000000000001,
      "loss": 6.9697,
      "mean_token_accuracy": 0.1092008151113987,
      "num_tokens": 410432.0,
      "step": 235
    },
    {
      "entropy": 7.437095785140992,
      "epoch": 0.018673409842443103,
      "grad_norm": 1.078125,
      "learning_rate": 0.00011949999999999999,
      "loss": 6.9627,
      "mean_token_accuracy": 0.09653985649347305,
      "num_tokens": 419380.0,
      "step": 240
    },
    {
      "entropy": 7.364046430587768,
      "epoch": 0.01906243921416067,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000122,
      "loss": 6.8292,
      "mean_token_accuracy": 0.10461175367236138,
      "num_tokens": 427682.0,
      "step": 245
    },
    {
      "entropy": 7.408738136291504,
      "epoch": 0.019451468585878234,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001245,
      "loss": 6.9587,
      "mean_token_accuracy": 0.09456864520907401,
      "num_tokens": 437708.0,
      "step": 250
    },
    {
      "entropy": 7.433396196365356,
      "epoch": 0.0198404979575958,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000127,
      "loss": 6.9371,
      "mean_token_accuracy": 0.09942449256777763,
      "num_tokens": 446480.0,
      "step": 255
    },
    {
      "entropy": 7.379309272766113,
      "epoch": 0.020229527329313362,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001295,
      "loss": 6.9721,
      "mean_token_accuracy": 0.09638917818665504,
      "num_tokens": 455273.0,
      "step": 260
    },
    {
      "entropy": 7.356784772872925,
      "epoch": 0.020618556701030927,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000132,
      "loss": 6.883,
      "mean_token_accuracy": 0.10503410249948501,
      "num_tokens": 464306.0,
      "step": 265
    },
    {
      "entropy": 7.388022518157959,
      "epoch": 0.021007586072748493,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00013450000000000002,
      "loss": 6.8831,
      "mean_token_accuracy": 0.10173421129584312,
      "num_tokens": 472812.0,
      "step": 270
    },
    {
      "entropy": 7.331966209411621,
      "epoch": 0.02139661544446606,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00013700000000000002,
      "loss": 6.8727,
      "mean_token_accuracy": 0.10593024790287017,
      "num_tokens": 481326.0,
      "step": 275
    },
    {
      "entropy": 7.374731874465942,
      "epoch": 0.02178564481618362,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001395,
      "loss": 6.8428,
      "mean_token_accuracy": 0.10759447142481804,
      "num_tokens": 489147.0,
      "step": 280
    },
    {
      "entropy": 7.36668381690979,
      "epoch": 0.022174674187901186,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014199999999999998,
      "loss": 6.8504,
      "mean_token_accuracy": 0.10656936541199684,
      "num_tokens": 497455.0,
      "step": 285
    },
    {
      "entropy": 7.356879997253418,
      "epoch": 0.02256370355961875,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001445,
      "loss": 6.758,
      "mean_token_accuracy": 0.10305927395820617,
      "num_tokens": 506735.0,
      "step": 290
    },
    {
      "entropy": 7.270456027984619,
      "epoch": 0.022952732931336317,
      "grad_norm": 1.265625,
      "learning_rate": 0.000147,
      "loss": 6.8849,
      "mean_token_accuracy": 0.09658203423023223,
      "num_tokens": 515976.0,
      "step": 295
    },
    {
      "entropy": 7.3423985004425045,
      "epoch": 0.02334176230305388,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001495,
      "loss": 6.866,
      "mean_token_accuracy": 0.10728994905948638,
      "num_tokens": 524986.0,
      "step": 300
    },
    {
      "entropy": 7.408446359634399,
      "epoch": 0.023730791674771445,
      "grad_norm": 1.390625,
      "learning_rate": 0.000152,
      "loss": 6.8928,
      "mean_token_accuracy": 0.110896135866642,
      "num_tokens": 533146.0,
      "step": 305
    },
    {
      "entropy": 7.319546413421631,
      "epoch": 0.02411982104648901,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.785,
      "mean_token_accuracy": 0.1116275131702423,
      "num_tokens": 541704.0,
      "step": 310
    },
    {
      "entropy": 7.265841770172119,
      "epoch": 0.024508850418206576,
      "grad_norm": 1.296875,
      "learning_rate": 0.000157,
      "loss": 6.7781,
      "mean_token_accuracy": 0.10793708562850952,
      "num_tokens": 550088.0,
      "step": 315
    },
    {
      "entropy": 7.317618799209595,
      "epoch": 0.024897879789924138,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001595,
      "loss": 6.7721,
      "mean_token_accuracy": 0.11174670755863189,
      "num_tokens": 558763.0,
      "step": 320
    },
    {
      "entropy": 7.293882560729981,
      "epoch": 0.025286909161641703,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000162,
      "loss": 6.8912,
      "mean_token_accuracy": 0.10175719037652016,
      "num_tokens": 567538.0,
      "step": 325
    },
    {
      "entropy": 7.340778779983521,
      "epoch": 0.02567593853335927,
      "grad_norm": 1.078125,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.8854,
      "mean_token_accuracy": 0.09728848040103913,
      "num_tokens": 577626.0,
      "step": 330
    },
    {
      "entropy": 7.330549144744873,
      "epoch": 0.026064967905076834,
      "grad_norm": 1.078125,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.8596,
      "mean_token_accuracy": 0.096848464012146,
      "num_tokens": 587163.0,
      "step": 335
    },
    {
      "entropy": 7.255002021789551,
      "epoch": 0.026453997276794396,
      "grad_norm": 1.109375,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.8158,
      "mean_token_accuracy": 0.10388773456215858,
      "num_tokens": 595896.0,
      "step": 340
    },
    {
      "entropy": 7.4087928295135494,
      "epoch": 0.026843026648511962,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.8486,
      "mean_token_accuracy": 0.10652071833610535,
      "num_tokens": 603815.0,
      "step": 345
    },
    {
      "entropy": 7.324870347976685,
      "epoch": 0.027232056020229527,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.8768,
      "mean_token_accuracy": 0.10469710528850555,
      "num_tokens": 612208.0,
      "step": 350
    },
    {
      "entropy": 7.263567209243774,
      "epoch": 0.027621085391947093,
      "grad_norm": 1.234375,
      "learning_rate": 0.000177,
      "loss": 6.836,
      "mean_token_accuracy": 0.10391956865787506,
      "num_tokens": 621840.0,
      "step": 355
    },
    {
      "entropy": 7.413131141662598,
      "epoch": 0.028010114763664655,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001795,
      "loss": 6.8576,
      "mean_token_accuracy": 0.10285800844430923,
      "num_tokens": 630770.0,
      "step": 360
    },
    {
      "entropy": 7.226842975616455,
      "epoch": 0.02839914413538222,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000182,
      "loss": 6.8735,
      "mean_token_accuracy": 0.10087347999215127,
      "num_tokens": 640313.0,
      "step": 365
    },
    {
      "entropy": 7.4696742534637455,
      "epoch": 0.028788173507099786,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001845,
      "loss": 6.8818,
      "mean_token_accuracy": 0.10657392218708991,
      "num_tokens": 648594.0,
      "step": 370
    },
    {
      "entropy": 7.3582908630371096,
      "epoch": 0.02917720287881735,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000187,
      "loss": 6.8691,
      "mean_token_accuracy": 0.11030346155166626,
      "num_tokens": 656478.0,
      "step": 375
    },
    {
      "entropy": 7.226627206802368,
      "epoch": 0.029566232250534917,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001895,
      "loss": 6.8622,
      "mean_token_accuracy": 0.10462825298309326,
      "num_tokens": 665405.0,
      "step": 380
    },
    {
      "entropy": 7.430356884002686,
      "epoch": 0.02995526162225248,
      "grad_norm": 1.34375,
      "learning_rate": 0.000192,
      "loss": 6.8824,
      "mean_token_accuracy": 0.10508352369070054,
      "num_tokens": 674301.0,
      "step": 385
    },
    {
      "entropy": 7.297737455368042,
      "epoch": 0.030344290993970045,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001945,
      "loss": 6.9,
      "mean_token_accuracy": 0.10454680025577545,
      "num_tokens": 683159.0,
      "step": 390
    },
    {
      "entropy": 7.246564197540283,
      "epoch": 0.03073332036568761,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.7087,
      "mean_token_accuracy": 0.11614901497960091,
      "num_tokens": 690992.0,
      "step": 395
    },
    {
      "entropy": 7.255606651306152,
      "epoch": 0.031122349737405176,
      "grad_norm": 1.1875,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.8127,
      "mean_token_accuracy": 0.10583920776844025,
      "num_tokens": 699934.0,
      "step": 400
    },
    {
      "entropy": 7.349058151245117,
      "epoch": 0.03151137910912274,
      "grad_norm": 1.28125,
      "learning_rate": 0.000202,
      "loss": 6.8701,
      "mean_token_accuracy": 0.10579104423522949,
      "num_tokens": 708844.0,
      "step": 405
    },
    {
      "entropy": 7.31098198890686,
      "epoch": 0.0319004084808403,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.9175,
      "mean_token_accuracy": 0.10112065151333809,
      "num_tokens": 718148.0,
      "step": 410
    },
    {
      "entropy": 7.361715412139892,
      "epoch": 0.03228943785255787,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000207,
      "loss": 6.8723,
      "mean_token_accuracy": 0.09706154093146324,
      "num_tokens": 726859.0,
      "step": 415
    },
    {
      "entropy": 7.335133504867554,
      "epoch": 0.032678467224275434,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002095,
      "loss": 6.8699,
      "mean_token_accuracy": 0.10610849931836128,
      "num_tokens": 735045.0,
      "step": 420
    },
    {
      "entropy": 7.167201137542724,
      "epoch": 0.033067496595993,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000212,
      "loss": 6.7586,
      "mean_token_accuracy": 0.10604033321142196,
      "num_tokens": 743538.0,
      "step": 425
    },
    {
      "entropy": 7.356954431533813,
      "epoch": 0.033456525967710565,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002145,
      "loss": 6.8011,
      "mean_token_accuracy": 0.10508049800992011,
      "num_tokens": 752518.0,
      "step": 430
    },
    {
      "entropy": 7.231763410568237,
      "epoch": 0.033845555339428124,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.8318,
      "mean_token_accuracy": 0.1046580545604229,
      "num_tokens": 760354.0,
      "step": 435
    },
    {
      "entropy": 7.280563879013061,
      "epoch": 0.03423458471114569,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002195,
      "loss": 6.8757,
      "mean_token_accuracy": 0.10113996267318726,
      "num_tokens": 769522.0,
      "step": 440
    },
    {
      "entropy": 7.253572177886963,
      "epoch": 0.034623614082863255,
      "grad_norm": 1.34375,
      "learning_rate": 0.000222,
      "loss": 6.7992,
      "mean_token_accuracy": 0.11093137413263321,
      "num_tokens": 778548.0,
      "step": 445
    },
    {
      "entropy": 7.248162889480591,
      "epoch": 0.03501264345458082,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002245,
      "loss": 6.8203,
      "mean_token_accuracy": 0.10633306801319123,
      "num_tokens": 787666.0,
      "step": 450
    },
    {
      "entropy": 7.291629838943481,
      "epoch": 0.035401672826298386,
      "grad_norm": 1.25,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.761,
      "mean_token_accuracy": 0.11031401231884956,
      "num_tokens": 796550.0,
      "step": 455
    },
    {
      "entropy": 7.261716270446778,
      "epoch": 0.03579070219801595,
      "grad_norm": 1.5,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.8326,
      "mean_token_accuracy": 0.10464341267943382,
      "num_tokens": 804549.0,
      "step": 460
    },
    {
      "entropy": 7.214675569534302,
      "epoch": 0.03617973156973352,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.796,
      "mean_token_accuracy": 0.10827421844005584,
      "num_tokens": 813328.0,
      "step": 465
    },
    {
      "entropy": 7.345154762268066,
      "epoch": 0.03656876094145108,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.8591,
      "mean_token_accuracy": 0.10106633976101875,
      "num_tokens": 821733.0,
      "step": 470
    },
    {
      "entropy": 7.146458864212036,
      "epoch": 0.03695779031316864,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000237,
      "loss": 6.788,
      "mean_token_accuracy": 0.10371901914477348,
      "num_tokens": 830243.0,
      "step": 475
    },
    {
      "entropy": 7.294699192047119,
      "epoch": 0.03734681968488621,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002395,
      "loss": 6.7664,
      "mean_token_accuracy": 0.10468717440962791,
      "num_tokens": 839059.0,
      "step": 480
    },
    {
      "entropy": 7.164042663574219,
      "epoch": 0.03773584905660377,
      "grad_norm": 1.25,
      "learning_rate": 0.000242,
      "loss": 6.6576,
      "mean_token_accuracy": 0.11653261855244637,
      "num_tokens": 846932.0,
      "step": 485
    },
    {
      "entropy": 7.279664421081543,
      "epoch": 0.03812487842832134,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002445,
      "loss": 6.8641,
      "mean_token_accuracy": 0.09954719170928002,
      "num_tokens": 855333.0,
      "step": 490
    },
    {
      "entropy": 7.079974603652954,
      "epoch": 0.0385139078000389,
      "grad_norm": 1.34375,
      "learning_rate": 0.000247,
      "loss": 6.6271,
      "mean_token_accuracy": 0.11037061735987663,
      "num_tokens": 863571.0,
      "step": 495
    },
    {
      "entropy": 7.1013068675994875,
      "epoch": 0.03890293717175647,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002495,
      "loss": 6.7244,
      "mean_token_accuracy": 0.11674270778894424,
      "num_tokens": 872145.0,
      "step": 500
    },
    {
      "entropy": 7.247890615463257,
      "epoch": 0.039291966543474034,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000252,
      "loss": 6.7569,
      "mean_token_accuracy": 0.1087301380932331,
      "num_tokens": 880634.0,
      "step": 505
    },
    {
      "entropy": 7.207820320129395,
      "epoch": 0.0396809959151916,
      "grad_norm": 1.25,
      "learning_rate": 0.0002545,
      "loss": 6.7062,
      "mean_token_accuracy": 0.10853548645973206,
      "num_tokens": 889178.0,
      "step": 510
    },
    {
      "entropy": 7.169516324996948,
      "epoch": 0.04007002528690916,
      "grad_norm": 1.15625,
      "learning_rate": 0.000257,
      "loss": 6.7134,
      "mean_token_accuracy": 0.10897102132439614,
      "num_tokens": 898255.0,
      "step": 515
    },
    {
      "entropy": 7.152969646453857,
      "epoch": 0.040459054658626724,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002595,
      "loss": 6.774,
      "mean_token_accuracy": 0.10289896130561829,
      "num_tokens": 907369.0,
      "step": 520
    },
    {
      "entropy": 7.14563512802124,
      "epoch": 0.04084808403034429,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000262,
      "loss": 6.6889,
      "mean_token_accuracy": 0.10718784034252167,
      "num_tokens": 915946.0,
      "step": 525
    },
    {
      "entropy": 7.151228046417236,
      "epoch": 0.041237113402061855,
      "grad_norm": 1.09375,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.7328,
      "mean_token_accuracy": 0.10271384865045548,
      "num_tokens": 925579.0,
      "step": 530
    },
    {
      "entropy": 7.176724624633789,
      "epoch": 0.04162614277377942,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.7277,
      "mean_token_accuracy": 0.10272877439856529,
      "num_tokens": 934732.0,
      "step": 535
    },
    {
      "entropy": 7.142701244354248,
      "epoch": 0.042015172145496986,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.8021,
      "mean_token_accuracy": 0.10387543737888336,
      "num_tokens": 943371.0,
      "step": 540
    },
    {
      "entropy": 7.137252473831177,
      "epoch": 0.04240420151721455,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.6423,
      "mean_token_accuracy": 0.11301440745592117,
      "num_tokens": 951977.0,
      "step": 545
    },
    {
      "entropy": 7.172152137756347,
      "epoch": 0.04279323088893212,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002745,
      "loss": 6.6448,
      "mean_token_accuracy": 0.11049823835492134,
      "num_tokens": 960474.0,
      "step": 550
    },
    {
      "entropy": 7.159184741973877,
      "epoch": 0.04318226026064968,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000277,
      "loss": 6.754,
      "mean_token_accuracy": 0.10716015249490737,
      "num_tokens": 969823.0,
      "step": 555
    },
    {
      "entropy": 7.191059017181397,
      "epoch": 0.04357128963236724,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002795,
      "loss": 6.682,
      "mean_token_accuracy": 0.12077577039599419,
      "num_tokens": 977417.0,
      "step": 560
    },
    {
      "entropy": 7.131459283828735,
      "epoch": 0.04396031900408481,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.6929,
      "mean_token_accuracy": 0.11818830072879791,
      "num_tokens": 985519.0,
      "step": 565
    },
    {
      "entropy": 7.157332468032837,
      "epoch": 0.04434934837580237,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002845,
      "loss": 6.6985,
      "mean_token_accuracy": 0.11225624307990074,
      "num_tokens": 994136.0,
      "step": 570
    },
    {
      "entropy": 7.024613523483277,
      "epoch": 0.04473837774751994,
      "grad_norm": 1.3125,
      "learning_rate": 0.000287,
      "loss": 6.6678,
      "mean_token_accuracy": 0.11105563789606095,
      "num_tokens": 1002896.0,
      "step": 575
    },
    {
      "entropy": 7.242383337020874,
      "epoch": 0.0451274071192375,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002895,
      "loss": 6.7489,
      "mean_token_accuracy": 0.11493279933929443,
      "num_tokens": 1011032.0,
      "step": 580
    },
    {
      "entropy": 7.015460014343262,
      "epoch": 0.04551643649095507,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000292,
      "loss": 6.7204,
      "mean_token_accuracy": 0.10362800061702729,
      "num_tokens": 1019901.0,
      "step": 585
    },
    {
      "entropy": 7.18277530670166,
      "epoch": 0.045905465862672634,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002945,
      "loss": 6.7207,
      "mean_token_accuracy": 0.10517120659351349,
      "num_tokens": 1029049.0,
      "step": 590
    },
    {
      "entropy": 7.049787425994873,
      "epoch": 0.0462944952343902,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000297,
      "loss": 6.6798,
      "mean_token_accuracy": 0.1185940608382225,
      "num_tokens": 1037580.0,
      "step": 595
    },
    {
      "entropy": 7.167644882202149,
      "epoch": 0.04668352460610776,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002995,
      "loss": 6.7261,
      "mean_token_accuracy": 0.1120360367000103,
      "num_tokens": 1045804.0,
      "step": 600
    },
    {
      "entropy": 7.098359155654907,
      "epoch": 0.047072553977825324,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000302,
      "loss": 6.694,
      "mean_token_accuracy": 0.10400344282388688,
      "num_tokens": 1054719.0,
      "step": 605
    },
    {
      "entropy": 7.160435199737549,
      "epoch": 0.04746158334954289,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003045,
      "loss": 6.6834,
      "mean_token_accuracy": 0.11288492232561112,
      "num_tokens": 1063373.0,
      "step": 610
    },
    {
      "entropy": 7.067417621612549,
      "epoch": 0.047850612721260455,
      "grad_norm": 1.234375,
      "learning_rate": 0.000307,
      "loss": 6.6745,
      "mean_token_accuracy": 0.1108371913433075,
      "num_tokens": 1073021.0,
      "step": 615
    },
    {
      "entropy": 7.10958251953125,
      "epoch": 0.04823964209297802,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003095,
      "loss": 6.7372,
      "mean_token_accuracy": 0.10155223831534385,
      "num_tokens": 1081547.0,
      "step": 620
    },
    {
      "entropy": 7.040739250183106,
      "epoch": 0.048628671464695586,
      "grad_norm": 1.09375,
      "learning_rate": 0.000312,
      "loss": 6.7017,
      "mean_token_accuracy": 0.10570611953735351,
      "num_tokens": 1090695.0,
      "step": 625
    },
    {
      "entropy": 7.087480449676514,
      "epoch": 0.04901770083641315,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003145,
      "loss": 6.5995,
      "mean_token_accuracy": 0.11334518268704415,
      "num_tokens": 1099563.0,
      "step": 630
    },
    {
      "entropy": 7.126954460144043,
      "epoch": 0.04940673020813072,
      "grad_norm": 1.25,
      "learning_rate": 0.000317,
      "loss": 6.6643,
      "mean_token_accuracy": 0.10861440300941468,
      "num_tokens": 1109021.0,
      "step": 635
    },
    {
      "entropy": 7.063493824005127,
      "epoch": 0.049795759579848276,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003195,
      "loss": 6.6672,
      "mean_token_accuracy": 0.11145636141300201,
      "num_tokens": 1118281.0,
      "step": 640
    },
    {
      "entropy": 7.050823497772217,
      "epoch": 0.05018478895156584,
      "grad_norm": 1.1875,
      "learning_rate": 0.000322,
      "loss": 6.7071,
      "mean_token_accuracy": 0.10184085294604302,
      "num_tokens": 1127513.0,
      "step": 645
    },
    {
      "entropy": 6.994142723083496,
      "epoch": 0.05057381832328341,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.5873,
      "mean_token_accuracy": 0.113088721036911,
      "num_tokens": 1136140.0,
      "step": 650
    },
    {
      "entropy": 7.076479101181031,
      "epoch": 0.05096284769500097,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.6142,
      "mean_token_accuracy": 0.10685904398560524,
      "num_tokens": 1144297.0,
      "step": 655
    },
    {
      "entropy": 7.086294746398925,
      "epoch": 0.05135187706671854,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.6877,
      "mean_token_accuracy": 0.1043604202568531,
      "num_tokens": 1153639.0,
      "step": 660
    },
    {
      "entropy": 7.073908519744873,
      "epoch": 0.0517409064384361,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.6662,
      "mean_token_accuracy": 0.10644880160689354,
      "num_tokens": 1162825.0,
      "step": 665
    },
    {
      "entropy": 7.004607772827148,
      "epoch": 0.05212993581015367,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.6953,
      "mean_token_accuracy": 0.11124756038188935,
      "num_tokens": 1171887.0,
      "step": 670
    },
    {
      "entropy": 7.069401454925537,
      "epoch": 0.052518965181871234,
      "grad_norm": 1.25,
      "learning_rate": 0.000337,
      "loss": 6.6135,
      "mean_token_accuracy": 0.11303046569228173,
      "num_tokens": 1179963.0,
      "step": 675
    },
    {
      "entropy": 6.96059250831604,
      "epoch": 0.05290799455358879,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003395,
      "loss": 6.5779,
      "mean_token_accuracy": 0.11756718456745148,
      "num_tokens": 1188408.0,
      "step": 680
    },
    {
      "entropy": 7.045384645462036,
      "epoch": 0.05329702392530636,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000342,
      "loss": 6.6249,
      "mean_token_accuracy": 0.11092553362250328,
      "num_tokens": 1197751.0,
      "step": 685
    },
    {
      "entropy": 6.961059188842773,
      "epoch": 0.053686053297023924,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.6564,
      "mean_token_accuracy": 0.11087723523378372,
      "num_tokens": 1207101.0,
      "step": 690
    },
    {
      "entropy": 7.06245789527893,
      "epoch": 0.05407508266874149,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000347,
      "loss": 6.6611,
      "mean_token_accuracy": 0.10940892472863198,
      "num_tokens": 1216395.0,
      "step": 695
    },
    {
      "entropy": 7.076578807830811,
      "epoch": 0.054464112040459055,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003495,
      "loss": 6.6477,
      "mean_token_accuracy": 0.1117068886756897,
      "num_tokens": 1225496.0,
      "step": 700
    },
    {
      "entropy": 7.017858028411865,
      "epoch": 0.05485314141217662,
      "grad_norm": 1.125,
      "learning_rate": 0.000352,
      "loss": 6.7116,
      "mean_token_accuracy": 0.11090820804238319,
      "num_tokens": 1233934.0,
      "step": 705
    },
    {
      "entropy": 6.977232789993286,
      "epoch": 0.055242170783894186,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003545,
      "loss": 6.5591,
      "mean_token_accuracy": 0.1176533691585064,
      "num_tokens": 1242165.0,
      "step": 710
    },
    {
      "entropy": 7.053106498718262,
      "epoch": 0.05563120015561175,
      "grad_norm": 1.296875,
      "learning_rate": 0.000357,
      "loss": 6.6346,
      "mean_token_accuracy": 0.11234478205442429,
      "num_tokens": 1250021.0,
      "step": 715
    },
    {
      "entropy": 7.054070568084716,
      "epoch": 0.05602022952732931,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003595,
      "loss": 6.6627,
      "mean_token_accuracy": 0.10676235780119896,
      "num_tokens": 1258397.0,
      "step": 720
    },
    {
      "entropy": 6.966783428192139,
      "epoch": 0.056409258899046875,
      "grad_norm": 1.34375,
      "learning_rate": 0.000362,
      "loss": 6.6435,
      "mean_token_accuracy": 0.1090509571135044,
      "num_tokens": 1266764.0,
      "step": 725
    },
    {
      "entropy": 7.071456336975098,
      "epoch": 0.05679828827076444,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003645,
      "loss": 6.6525,
      "mean_token_accuracy": 0.10880729258060455,
      "num_tokens": 1275643.0,
      "step": 730
    },
    {
      "entropy": 6.989516830444336,
      "epoch": 0.057187317642482006,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000367,
      "loss": 6.6201,
      "mean_token_accuracy": 0.11093898415565491,
      "num_tokens": 1284379.0,
      "step": 735
    },
    {
      "entropy": 6.985952568054199,
      "epoch": 0.05757634701419957,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003695,
      "loss": 6.4795,
      "mean_token_accuracy": 0.11814915612339974,
      "num_tokens": 1293115.0,
      "step": 740
    },
    {
      "entropy": 7.051576614379883,
      "epoch": 0.05796537638591714,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000372,
      "loss": 6.682,
      "mean_token_accuracy": 0.10519764646887779,
      "num_tokens": 1301583.0,
      "step": 745
    },
    {
      "entropy": 6.9952661991119385,
      "epoch": 0.0583544057576347,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003745,
      "loss": 6.5803,
      "mean_token_accuracy": 0.11081231012940407,
      "num_tokens": 1310721.0,
      "step": 750
    },
    {
      "entropy": 7.042070531845093,
      "epoch": 0.05874343512935227,
      "grad_norm": 1.296875,
      "learning_rate": 0.000377,
      "loss": 6.7291,
      "mean_token_accuracy": 0.10293859988451004,
      "num_tokens": 1319056.0,
      "step": 755
    },
    {
      "entropy": 6.974556493759155,
      "epoch": 0.059132464501069834,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003795,
      "loss": 6.6182,
      "mean_token_accuracy": 0.1109778843820095,
      "num_tokens": 1327259.0,
      "step": 760
    },
    {
      "entropy": 6.920253753662109,
      "epoch": 0.05952149387278739,
      "grad_norm": 1.171875,
      "learning_rate": 0.000382,
      "loss": 6.5178,
      "mean_token_accuracy": 0.11307926774024964,
      "num_tokens": 1335906.0,
      "step": 765
    },
    {
      "entropy": 6.978034830093383,
      "epoch": 0.05991052324450496,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003845,
      "loss": 6.5646,
      "mean_token_accuracy": 0.10813911706209182,
      "num_tokens": 1344743.0,
      "step": 770
    },
    {
      "entropy": 6.963363122940064,
      "epoch": 0.060299552616222524,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.5806,
      "mean_token_accuracy": 0.1119392916560173,
      "num_tokens": 1353211.0,
      "step": 775
    },
    {
      "entropy": 7.029351425170899,
      "epoch": 0.06068858198794009,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.6888,
      "mean_token_accuracy": 0.10860950574278831,
      "num_tokens": 1361849.0,
      "step": 780
    },
    {
      "entropy": 7.0644081115722654,
      "epoch": 0.061077611359657655,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.6819,
      "mean_token_accuracy": 0.11109407320618629,
      "num_tokens": 1369804.0,
      "step": 785
    },
    {
      "entropy": 6.8550738334655765,
      "epoch": 0.06146664073137522,
      "grad_norm": 1.125,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.4614,
      "mean_token_accuracy": 0.11694267019629478,
      "num_tokens": 1378478.0,
      "step": 790
    },
    {
      "entropy": 6.972127914428711,
      "epoch": 0.061855670103092786,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.5706,
      "mean_token_accuracy": 0.1196007214486599,
      "num_tokens": 1387225.0,
      "step": 795
    },
    {
      "entropy": 6.917468309402466,
      "epoch": 0.06224469947481035,
      "grad_norm": 1.25,
      "learning_rate": 0.0003995,
      "loss": 6.5368,
      "mean_token_accuracy": 0.11665247678756714,
      "num_tokens": 1395338.0,
      "step": 800
    },
    {
      "entropy": 6.90943489074707,
      "epoch": 0.06263372884652792,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000402,
      "loss": 6.5161,
      "mean_token_accuracy": 0.11467550322413445,
      "num_tokens": 1403794.0,
      "step": 805
    },
    {
      "entropy": 6.851684522628784,
      "epoch": 0.06302275821824548,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004045,
      "loss": 6.594,
      "mean_token_accuracy": 0.10971078649163246,
      "num_tokens": 1412522.0,
      "step": 810
    },
    {
      "entropy": 6.944258499145508,
      "epoch": 0.06341178758996305,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.586,
      "mean_token_accuracy": 0.11532295569777488,
      "num_tokens": 1421347.0,
      "step": 815
    },
    {
      "entropy": 6.928653335571289,
      "epoch": 0.0638008169616806,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004095,
      "loss": 6.5611,
      "mean_token_accuracy": 0.11696547418832778,
      "num_tokens": 1429788.0,
      "step": 820
    },
    {
      "entropy": 6.896436071395874,
      "epoch": 0.06418984633339817,
      "grad_norm": 1.3125,
      "learning_rate": 0.000412,
      "loss": 6.5596,
      "mean_token_accuracy": 0.11556609123945236,
      "num_tokens": 1437442.0,
      "step": 825
    },
    {
      "entropy": 6.894798803329468,
      "epoch": 0.06457887570511574,
      "grad_norm": 1.125,
      "learning_rate": 0.0004145,
      "loss": 6.5008,
      "mean_token_accuracy": 0.11221573501825333,
      "num_tokens": 1445604.0,
      "step": 830
    },
    {
      "entropy": 6.868392038345337,
      "epoch": 0.0649679050768333,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000417,
      "loss": 6.5733,
      "mean_token_accuracy": 0.11458783373236656,
      "num_tokens": 1454342.0,
      "step": 835
    },
    {
      "entropy": 6.891533136367798,
      "epoch": 0.06535693444855087,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004195,
      "loss": 6.5434,
      "mean_token_accuracy": 0.11737344339489937,
      "num_tokens": 1463189.0,
      "step": 840
    },
    {
      "entropy": 6.914099884033203,
      "epoch": 0.06574596382026843,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000422,
      "loss": 6.6141,
      "mean_token_accuracy": 0.10967839807271958,
      "num_tokens": 1471608.0,
      "step": 845
    },
    {
      "entropy": 6.9978296756744385,
      "epoch": 0.066134993191986,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004245,
      "loss": 6.6144,
      "mean_token_accuracy": 0.1100444920361042,
      "num_tokens": 1480452.0,
      "step": 850
    },
    {
      "entropy": 6.94025354385376,
      "epoch": 0.06652402256370356,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000427,
      "loss": 6.5336,
      "mean_token_accuracy": 0.11354602947831154,
      "num_tokens": 1489118.0,
      "step": 855
    },
    {
      "entropy": 6.8302192211151125,
      "epoch": 0.06691305193542113,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004295,
      "loss": 6.535,
      "mean_token_accuracy": 0.10949739366769791,
      "num_tokens": 1498961.0,
      "step": 860
    },
    {
      "entropy": 6.9123735427856445,
      "epoch": 0.06730208130713869,
      "grad_norm": 1.265625,
      "learning_rate": 0.000432,
      "loss": 6.5356,
      "mean_token_accuracy": 0.11317324861884118,
      "num_tokens": 1506772.0,
      "step": 865
    },
    {
      "entropy": 6.865458965301514,
      "epoch": 0.06769111067885625,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004345,
      "loss": 6.4228,
      "mean_token_accuracy": 0.12698905244469644,
      "num_tokens": 1515095.0,
      "step": 870
    },
    {
      "entropy": 6.808365058898926,
      "epoch": 0.06808014005057382,
      "grad_norm": 1.125,
      "learning_rate": 0.000437,
      "loss": 6.4697,
      "mean_token_accuracy": 0.12296628281474113,
      "num_tokens": 1523477.0,
      "step": 875
    },
    {
      "entropy": 6.824631023406982,
      "epoch": 0.06846916942229138,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004395,
      "loss": 6.5022,
      "mean_token_accuracy": 0.11501481533050537,
      "num_tokens": 1531786.0,
      "step": 880
    },
    {
      "entropy": 6.805560159683227,
      "epoch": 0.06885819879400895,
      "grad_norm": 1.234375,
      "learning_rate": 0.000442,
      "loss": 6.5989,
      "mean_token_accuracy": 0.10919139832258225,
      "num_tokens": 1540737.0,
      "step": 885
    },
    {
      "entropy": 6.883869266510009,
      "epoch": 0.06924722816572651,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004445,
      "loss": 6.4469,
      "mean_token_accuracy": 0.11882291138172149,
      "num_tokens": 1548923.0,
      "step": 890
    },
    {
      "entropy": 6.784322309494018,
      "epoch": 0.06963625753744408,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000447,
      "loss": 6.4067,
      "mean_token_accuracy": 0.12390651702880859,
      "num_tokens": 1557043.0,
      "step": 895
    },
    {
      "entropy": 6.771895265579223,
      "epoch": 0.07002528690916164,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.4486,
      "mean_token_accuracy": 0.1224091038107872,
      "num_tokens": 1564994.0,
      "step": 900
    },
    {
      "entropy": 6.809636068344116,
      "epoch": 0.0704143162808792,
      "grad_norm": 1.125,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.4563,
      "mean_token_accuracy": 0.1136368915438652,
      "num_tokens": 1573859.0,
      "step": 905
    },
    {
      "entropy": 6.830642652511597,
      "epoch": 0.07080334565259677,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.429,
      "mean_token_accuracy": 0.11435260772705078,
      "num_tokens": 1581947.0,
      "step": 910
    },
    {
      "entropy": 6.772935581207276,
      "epoch": 0.07119237502431433,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.4005,
      "mean_token_accuracy": 0.11757316365838051,
      "num_tokens": 1589971.0,
      "step": 915
    },
    {
      "entropy": 6.689200162887573,
      "epoch": 0.0715814043960319,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.3619,
      "mean_token_accuracy": 0.11918411031365395,
      "num_tokens": 1598612.0,
      "step": 920
    },
    {
      "entropy": 6.844861602783203,
      "epoch": 0.07197043376774946,
      "grad_norm": 1.15625,
      "learning_rate": 0.000462,
      "loss": 6.5099,
      "mean_token_accuracy": 0.11082213446497917,
      "num_tokens": 1607378.0,
      "step": 925
    },
    {
      "entropy": 6.754356575012207,
      "epoch": 0.07235946313946703,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004645,
      "loss": 6.5073,
      "mean_token_accuracy": 0.11739620044827462,
      "num_tokens": 1615987.0,
      "step": 930
    },
    {
      "entropy": 6.726682090759278,
      "epoch": 0.07274849251118459,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000467,
      "loss": 6.3035,
      "mean_token_accuracy": 0.12486169338226319,
      "num_tokens": 1625366.0,
      "step": 935
    },
    {
      "entropy": 6.736809110641479,
      "epoch": 0.07313752188290217,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004695,
      "loss": 6.4848,
      "mean_token_accuracy": 0.11880016550421715,
      "num_tokens": 1633020.0,
      "step": 940
    },
    {
      "entropy": 6.754602718353271,
      "epoch": 0.07352655125461972,
      "grad_norm": 1.15625,
      "learning_rate": 0.000472,
      "loss": 6.3883,
      "mean_token_accuracy": 0.11748389154672623,
      "num_tokens": 1641518.0,
      "step": 945
    },
    {
      "entropy": 6.705056047439575,
      "epoch": 0.07391558062633728,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004745,
      "loss": 6.4726,
      "mean_token_accuracy": 0.11781010702252388,
      "num_tokens": 1650472.0,
      "step": 950
    },
    {
      "entropy": 6.785799169540406,
      "epoch": 0.07430460999805485,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000477,
      "loss": 6.507,
      "mean_token_accuracy": 0.11363615766167641,
      "num_tokens": 1659037.0,
      "step": 955
    },
    {
      "entropy": 6.829026794433593,
      "epoch": 0.07469363936977241,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004795,
      "loss": 6.5735,
      "mean_token_accuracy": 0.11651240065693855,
      "num_tokens": 1667897.0,
      "step": 960
    },
    {
      "entropy": 6.732403898239136,
      "epoch": 0.07508266874148999,
      "grad_norm": 1.1875,
      "learning_rate": 0.000482,
      "loss": 6.4198,
      "mean_token_accuracy": 0.11139412000775337,
      "num_tokens": 1676535.0,
      "step": 965
    },
    {
      "entropy": 6.8594677448272705,
      "epoch": 0.07547169811320754,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004845,
      "loss": 6.4466,
      "mean_token_accuracy": 0.11887699663639069,
      "num_tokens": 1684415.0,
      "step": 970
    },
    {
      "entropy": 6.723340702056885,
      "epoch": 0.07586072748492512,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000487,
      "loss": 6.3959,
      "mean_token_accuracy": 0.12453912273049354,
      "num_tokens": 1692443.0,
      "step": 975
    },
    {
      "entropy": 6.670280694961548,
      "epoch": 0.07624975685664268,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004895,
      "loss": 6.4114,
      "mean_token_accuracy": 0.12035818472504616,
      "num_tokens": 1700814.0,
      "step": 980
    },
    {
      "entropy": 6.637305498123169,
      "epoch": 0.07663878622836025,
      "grad_norm": 1.171875,
      "learning_rate": 0.000492,
      "loss": 6.2552,
      "mean_token_accuracy": 0.12497311756014824,
      "num_tokens": 1709591.0,
      "step": 985
    },
    {
      "entropy": 6.758705091476441,
      "epoch": 0.0770278156000778,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004945,
      "loss": 6.491,
      "mean_token_accuracy": 0.11199909374117852,
      "num_tokens": 1718951.0,
      "step": 990
    },
    {
      "entropy": 6.755342149734497,
      "epoch": 0.07741684497179536,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000497,
      "loss": 6.4487,
      "mean_token_accuracy": 0.11627254337072372,
      "num_tokens": 1727481.0,
      "step": 995
    },
    {
      "entropy": 6.776053047180175,
      "epoch": 0.07780587434351294,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004995,
      "loss": 6.4373,
      "mean_token_accuracy": 0.1180419959127903,
      "num_tokens": 1736985.0,
      "step": 1000
    },
    {
      "entropy": 6.759429550170898,
      "epoch": 0.0781949037152305,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999999989075146,
      "loss": 6.5256,
      "mean_token_accuracy": 0.1146937295794487,
      "num_tokens": 1745470.0,
      "step": 1005
    },
    {
      "entropy": 6.769218778610229,
      "epoch": 0.07858393308694807,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004999999944692927,
      "loss": 6.4193,
      "mean_token_accuracy": 0.11607462614774704,
      "num_tokens": 1754185.0,
      "step": 1010
    },
    {
      "entropy": 6.672504806518555,
      "epoch": 0.07897296245866563,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000499999986617054,
      "loss": 6.4365,
      "mean_token_accuracy": 0.11394271180033684,
      "num_tokens": 1762100.0,
      "step": 1015
    },
    {
      "entropy": 6.805412864685058,
      "epoch": 0.0793619918303832,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004999999753507986,
      "loss": 6.4533,
      "mean_token_accuracy": 0.11515071243047714,
      "num_tokens": 1771070.0,
      "step": 1020
    },
    {
      "entropy": 6.658144235610962,
      "epoch": 0.07975102120210076,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999999606705267,
      "loss": 6.4549,
      "mean_token_accuracy": 0.11719758063554764,
      "num_tokens": 1780085.0,
      "step": 1025
    },
    {
      "entropy": 6.757973384857178,
      "epoch": 0.08014005057381832,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999999425762385,
      "loss": 6.4477,
      "mean_token_accuracy": 0.12243591025471687,
      "num_tokens": 1789316.0,
      "step": 1030
    },
    {
      "entropy": 6.678041696548462,
      "epoch": 0.08052907994553589,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999999210679344,
      "loss": 6.3704,
      "mean_token_accuracy": 0.1213125005364418,
      "num_tokens": 1797975.0,
      "step": 1035
    },
    {
      "entropy": 6.65592589378357,
      "epoch": 0.08091810931725345,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999998961456145,
      "loss": 6.4589,
      "mean_token_accuracy": 0.12353531643748283,
      "num_tokens": 1806129.0,
      "step": 1040
    },
    {
      "entropy": 6.8409473419189455,
      "epoch": 0.08130713868897102,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999998678092794,
      "loss": 6.4574,
      "mean_token_accuracy": 0.11680945232510567,
      "num_tokens": 1813956.0,
      "step": 1045
    },
    {
      "entropy": 6.737816333770752,
      "epoch": 0.08169616806068858,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999998360589293,
      "loss": 6.481,
      "mean_token_accuracy": 0.11468739807605743,
      "num_tokens": 1822572.0,
      "step": 1050
    },
    {
      "entropy": 6.6663189888000485,
      "epoch": 0.08208519743240615,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499999800894565,
      "loss": 6.3814,
      "mean_token_accuracy": 0.11563698872923851,
      "num_tokens": 1831870.0,
      "step": 1055
    },
    {
      "entropy": 6.598942708969116,
      "epoch": 0.08247422680412371,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004999997623161866,
      "loss": 6.4276,
      "mean_token_accuracy": 0.11558841541409492,
      "num_tokens": 1840414.0,
      "step": 1060
    },
    {
      "entropy": 6.852775192260742,
      "epoch": 0.08286325617584128,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999997203237951,
      "loss": 6.4426,
      "mean_token_accuracy": 0.11738709583878518,
      "num_tokens": 1850385.0,
      "step": 1065
    },
    {
      "entropy": 6.492261695861816,
      "epoch": 0.08325228554755884,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499999674917391,
      "loss": 6.3487,
      "mean_token_accuracy": 0.12343356981873513,
      "num_tokens": 1858479.0,
      "step": 1070
    },
    {
      "entropy": 6.708926773071289,
      "epoch": 0.0836413149192764,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999996260969748,
      "loss": 6.4097,
      "mean_token_accuracy": 0.11067359670996665,
      "num_tokens": 1867357.0,
      "step": 1075
    },
    {
      "entropy": 6.690218782424926,
      "epoch": 0.08403034429099397,
      "grad_norm": 1.25,
      "learning_rate": 0.0004999995738625475,
      "loss": 6.3768,
      "mean_token_accuracy": 0.1146676443517208,
      "num_tokens": 1876211.0,
      "step": 1080
    },
    {
      "entropy": 6.6100105285644535,
      "epoch": 0.08441937366271153,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999995182141099,
      "loss": 6.2786,
      "mean_token_accuracy": 0.11257534921169281,
      "num_tokens": 1884301.0,
      "step": 1085
    },
    {
      "entropy": 6.706263875961303,
      "epoch": 0.0848084030344291,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004999994591516626,
      "loss": 6.461,
      "mean_token_accuracy": 0.1196840226650238,
      "num_tokens": 1893191.0,
      "step": 1090
    },
    {
      "entropy": 6.564361143112182,
      "epoch": 0.08519743240614666,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999993966752066,
      "loss": 6.3293,
      "mean_token_accuracy": 0.11676682159304619,
      "num_tokens": 1901434.0,
      "step": 1095
    },
    {
      "entropy": 6.664592123031616,
      "epoch": 0.08558646177786423,
      "grad_norm": 1.203125,
      "learning_rate": 0.000499999330784743,
      "loss": 6.3132,
      "mean_token_accuracy": 0.1269494652748108,
      "num_tokens": 1910054.0,
      "step": 1100
    },
    {
      "entropy": 6.6648547649383545,
      "epoch": 0.08597549114958179,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999992614802725,
      "loss": 6.4633,
      "mean_token_accuracy": 0.11476121544837951,
      "num_tokens": 1918607.0,
      "step": 1105
    },
    {
      "entropy": 6.6421819686889645,
      "epoch": 0.08636452052129936,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999991887617966,
      "loss": 6.3989,
      "mean_token_accuracy": 0.11885848641395569,
      "num_tokens": 1927610.0,
      "step": 1110
    },
    {
      "entropy": 6.606335258483886,
      "epoch": 0.08675354989301692,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004999991126293158,
      "loss": 6.3591,
      "mean_token_accuracy": 0.11808456256985664,
      "num_tokens": 1936174.0,
      "step": 1115
    },
    {
      "entropy": 6.620585012435913,
      "epoch": 0.08714257926473448,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999990330828318,
      "loss": 6.285,
      "mean_token_accuracy": 0.12242377102375031,
      "num_tokens": 1944228.0,
      "step": 1120
    },
    {
      "entropy": 6.607644987106323,
      "epoch": 0.08753160863645205,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999989501223456,
      "loss": 6.3877,
      "mean_token_accuracy": 0.12339339479804039,
      "num_tokens": 1953144.0,
      "step": 1125
    },
    {
      "entropy": 6.704639625549317,
      "epoch": 0.08792063800816961,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999988637478584,
      "loss": 6.4318,
      "mean_token_accuracy": 0.11419345438480377,
      "num_tokens": 1961947.0,
      "step": 1130
    },
    {
      "entropy": 6.676687431335449,
      "epoch": 0.08830966737988719,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999987739593716,
      "loss": 6.4129,
      "mean_token_accuracy": 0.11595526710152626,
      "num_tokens": 1970674.0,
      "step": 1135
    },
    {
      "entropy": 6.671134376525879,
      "epoch": 0.08869869675160474,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999986807568865,
      "loss": 6.3788,
      "mean_token_accuracy": 0.12137411832809449,
      "num_tokens": 1979173.0,
      "step": 1140
    },
    {
      "entropy": 6.589903163909912,
      "epoch": 0.08908772612332232,
      "grad_norm": 1.25,
      "learning_rate": 0.0004999985841404045,
      "loss": 6.3583,
      "mean_token_accuracy": 0.12250440344214439,
      "num_tokens": 1988134.0,
      "step": 1145
    },
    {
      "entropy": 6.666301631927491,
      "epoch": 0.08947675549503988,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999984841099271,
      "loss": 6.3107,
      "mean_token_accuracy": 0.12049696892499924,
      "num_tokens": 1996977.0,
      "step": 1150
    },
    {
      "entropy": 6.652774333953857,
      "epoch": 0.08986578486675743,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499998380665456,
      "loss": 6.3871,
      "mean_token_accuracy": 0.11820772588253022,
      "num_tokens": 2005946.0,
      "step": 1155
    },
    {
      "entropy": 6.566985940933227,
      "epoch": 0.090254814238475,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004999982738069924,
      "loss": 6.2172,
      "mean_token_accuracy": 0.12243169024586678,
      "num_tokens": 2014882.0,
      "step": 1160
    },
    {
      "entropy": 6.548052835464477,
      "epoch": 0.09064384361019256,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999981635345382,
      "loss": 6.407,
      "mean_token_accuracy": 0.11777823343873024,
      "num_tokens": 2024858.0,
      "step": 1165
    },
    {
      "entropy": 6.58183331489563,
      "epoch": 0.09103287298191014,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004999980498480949,
      "loss": 6.3198,
      "mean_token_accuracy": 0.12247655168175697,
      "num_tokens": 2034072.0,
      "step": 1170
    },
    {
      "entropy": 6.619194364547729,
      "epoch": 0.0914219023536277,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999979327476644,
      "loss": 6.3244,
      "mean_token_accuracy": 0.1233485259115696,
      "num_tokens": 2042649.0,
      "step": 1175
    },
    {
      "entropy": 6.5773601055145265,
      "epoch": 0.09181093172534527,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999978122332485,
      "loss": 6.3418,
      "mean_token_accuracy": 0.12649165764451026,
      "num_tokens": 2050866.0,
      "step": 1180
    },
    {
      "entropy": 6.624700975418091,
      "epoch": 0.09219996109706283,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999976883048487,
      "loss": 6.356,
      "mean_token_accuracy": 0.12287835404276848,
      "num_tokens": 2058914.0,
      "step": 1185
    },
    {
      "entropy": 6.566097354888916,
      "epoch": 0.0925889904687804,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999975609624673,
      "loss": 6.287,
      "mean_token_accuracy": 0.12197296395897865,
      "num_tokens": 2068098.0,
      "step": 1190
    },
    {
      "entropy": 6.562694406509399,
      "epoch": 0.09297801984049796,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999974302061059,
      "loss": 6.2968,
      "mean_token_accuracy": 0.11417963430285454,
      "num_tokens": 2077299.0,
      "step": 1195
    },
    {
      "entropy": 6.573287296295166,
      "epoch": 0.09336704921221552,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999972960357666,
      "loss": 6.3798,
      "mean_token_accuracy": 0.11859698817133904,
      "num_tokens": 2087868.0,
      "step": 1200
    },
    {
      "entropy": 6.565402412414551,
      "epoch": 0.09375607858393309,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999971584514516,
      "loss": 6.3108,
      "mean_token_accuracy": 0.1254733011126518,
      "num_tokens": 2096553.0,
      "step": 1205
    },
    {
      "entropy": 6.652924728393555,
      "epoch": 0.09414510795565065,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999970174531627,
      "loss": 6.2801,
      "mean_token_accuracy": 0.1194481149315834,
      "num_tokens": 2104809.0,
      "step": 1210
    },
    {
      "entropy": 6.454768133163452,
      "epoch": 0.09453413732736822,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999968730409022,
      "loss": 6.2644,
      "mean_token_accuracy": 0.12134657055139542,
      "num_tokens": 2113869.0,
      "step": 1215
    },
    {
      "entropy": 6.491008758544922,
      "epoch": 0.09492316669908578,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999967252146723,
      "loss": 6.0974,
      "mean_token_accuracy": 0.11863267570734023,
      "num_tokens": 2121749.0,
      "step": 1220
    },
    {
      "entropy": 6.488491821289062,
      "epoch": 0.09531219607080335,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999965739744752,
      "loss": 6.3197,
      "mean_token_accuracy": 0.11957874894142151,
      "num_tokens": 2130828.0,
      "step": 1225
    },
    {
      "entropy": 6.629486036300659,
      "epoch": 0.09570122544252091,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999964193203132,
      "loss": 6.2794,
      "mean_token_accuracy": 0.12205140367150306,
      "num_tokens": 2138928.0,
      "step": 1230
    },
    {
      "entropy": 6.551081609725952,
      "epoch": 0.09609025481423847,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999962612521888,
      "loss": 6.334,
      "mean_token_accuracy": 0.12222919836640359,
      "num_tokens": 2147643.0,
      "step": 1235
    },
    {
      "entropy": 6.581508445739746,
      "epoch": 0.09647928418595604,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999960997701041,
      "loss": 6.2547,
      "mean_token_accuracy": 0.12721101716160774,
      "num_tokens": 2156766.0,
      "step": 1240
    },
    {
      "entropy": 6.509366512298584,
      "epoch": 0.0968683135576736,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999959348740618,
      "loss": 6.3083,
      "mean_token_accuracy": 0.12237787693738937,
      "num_tokens": 2165517.0,
      "step": 1245
    },
    {
      "entropy": 6.485399150848389,
      "epoch": 0.09725734292939117,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999957665640641,
      "loss": 6.1935,
      "mean_token_accuracy": 0.1266528323292732,
      "num_tokens": 2174812.0,
      "step": 1250
    },
    {
      "entropy": 6.535298681259155,
      "epoch": 0.09764637230110873,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499995594840114,
      "loss": 6.3305,
      "mean_token_accuracy": 0.11914158836007119,
      "num_tokens": 2183968.0,
      "step": 1255
    },
    {
      "entropy": 6.618254613876343,
      "epoch": 0.0980354016728263,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999954197022138,
      "loss": 6.3459,
      "mean_token_accuracy": 0.12755375280976294,
      "num_tokens": 2191947.0,
      "step": 1260
    },
    {
      "entropy": 6.528839731216431,
      "epoch": 0.09842443104454386,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999952411503661,
      "loss": 6.1907,
      "mean_token_accuracy": 0.1307717591524124,
      "num_tokens": 2199951.0,
      "step": 1265
    },
    {
      "entropy": 6.479862356185913,
      "epoch": 0.09881346041626143,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999950591845738,
      "loss": 6.3421,
      "mean_token_accuracy": 0.12372912466526031,
      "num_tokens": 2208714.0,
      "step": 1270
    },
    {
      "entropy": 6.498292541503906,
      "epoch": 0.09920248978797899,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999948738048397,
      "loss": 6.2,
      "mean_token_accuracy": 0.1264264166355133,
      "num_tokens": 2217250.0,
      "step": 1275
    },
    {
      "entropy": 6.5420334815979,
      "epoch": 0.09959151915969655,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004999946850111663,
      "loss": 6.253,
      "mean_token_accuracy": 0.12835126966238022,
      "num_tokens": 2225468.0,
      "step": 1280
    },
    {
      "entropy": 6.4549304962158205,
      "epoch": 0.09998054853141412,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999944928035569,
      "loss": 6.2102,
      "mean_token_accuracy": 0.1315177209675312,
      "num_tokens": 2233893.0,
      "step": 1285
    },
    {
      "entropy": 6.566325521469116,
      "epoch": 0.10036957790313168,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000499994297182014,
      "loss": 6.2795,
      "mean_token_accuracy": 0.12629038318991662,
      "num_tokens": 2242487.0,
      "step": 1290
    },
    {
      "entropy": 6.4853943347930905,
      "epoch": 0.10075860727484925,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999940981465408,
      "loss": 6.1073,
      "mean_token_accuracy": 0.1271251358091831,
      "num_tokens": 2250518.0,
      "step": 1295
    },
    {
      "entropy": 6.446168088912964,
      "epoch": 0.10114763664656681,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004999938956971404,
      "loss": 6.2792,
      "mean_token_accuracy": 0.11826291009783745,
      "num_tokens": 2259024.0,
      "step": 1300
    },
    {
      "entropy": 6.553367996215821,
      "epoch": 0.10153666601828439,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999936898338156,
      "loss": 6.3272,
      "mean_token_accuracy": 0.1139446422457695,
      "num_tokens": 2267832.0,
      "step": 1305
    },
    {
      "entropy": 6.4859734058380125,
      "epoch": 0.10192569539000194,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999934805565698,
      "loss": 6.3817,
      "mean_token_accuracy": 0.1167535126209259,
      "num_tokens": 2276834.0,
      "step": 1310
    },
    {
      "entropy": 6.623997735977173,
      "epoch": 0.10231472476171952,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499993267865406,
      "loss": 6.3096,
      "mean_token_accuracy": 0.12032428234815598,
      "num_tokens": 2285736.0,
      "step": 1315
    },
    {
      "entropy": 6.567375421524048,
      "epoch": 0.10270375413343708,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999930517603275,
      "loss": 6.256,
      "mean_token_accuracy": 0.13201069533824922,
      "num_tokens": 2294177.0,
      "step": 1320
    },
    {
      "entropy": 6.438686847686768,
      "epoch": 0.10309278350515463,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004999928322413374,
      "loss": 6.0542,
      "mean_token_accuracy": 0.13524992465972902,
      "num_tokens": 2302784.0,
      "step": 1325
    },
    {
      "entropy": 6.362113523483276,
      "epoch": 0.1034818128768722,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999926093084394,
      "loss": 6.171,
      "mean_token_accuracy": 0.12782380506396293,
      "num_tokens": 2311650.0,
      "step": 1330
    },
    {
      "entropy": 6.510524940490723,
      "epoch": 0.10387084224858976,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999923829616365,
      "loss": 6.2756,
      "mean_token_accuracy": 0.12364537864923478,
      "num_tokens": 2319828.0,
      "step": 1335
    },
    {
      "entropy": 6.539057922363281,
      "epoch": 0.10425987162030734,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999921532009323,
      "loss": 6.2842,
      "mean_token_accuracy": 0.12230362668633461,
      "num_tokens": 2328636.0,
      "step": 1340
    },
    {
      "entropy": 6.401849460601807,
      "epoch": 0.1046489009920249,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999919200263304,
      "loss": 6.17,
      "mean_token_accuracy": 0.12805957421660424,
      "num_tokens": 2336771.0,
      "step": 1345
    },
    {
      "entropy": 6.524014139175415,
      "epoch": 0.10503793036374247,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999916834378342,
      "loss": 6.3125,
      "mean_token_accuracy": 0.12446440905332565,
      "num_tokens": 2346106.0,
      "step": 1350
    },
    {
      "entropy": 6.554375123977661,
      "epoch": 0.10542695973546003,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999914434354472,
      "loss": 6.3107,
      "mean_token_accuracy": 0.116956078261137,
      "num_tokens": 2354943.0,
      "step": 1355
    },
    {
      "entropy": 6.330786418914795,
      "epoch": 0.10581598910717759,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999912000191733,
      "loss": 6.2156,
      "mean_token_accuracy": 0.12508122622966766,
      "num_tokens": 2363699.0,
      "step": 1360
    },
    {
      "entropy": 6.5141215324401855,
      "epoch": 0.10620501847889516,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499990953189016,
      "loss": 6.3486,
      "mean_token_accuracy": 0.11765544191002845,
      "num_tokens": 2373316.0,
      "step": 1365
    },
    {
      "entropy": 6.683814477920532,
      "epoch": 0.10659404785061272,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999907029449791,
      "loss": 6.4201,
      "mean_token_accuracy": 0.11839015111327171,
      "num_tokens": 2381550.0,
      "step": 1370
    },
    {
      "entropy": 6.481244421005249,
      "epoch": 0.10698307722233029,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999904492870663,
      "loss": 6.221,
      "mean_token_accuracy": 0.12212294861674308,
      "num_tokens": 2389546.0,
      "step": 1375
    },
    {
      "entropy": 6.468661642074585,
      "epoch": 0.10737210659404785,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999901922152816,
      "loss": 6.1852,
      "mean_token_accuracy": 0.1323728546500206,
      "num_tokens": 2398852.0,
      "step": 1380
    },
    {
      "entropy": 6.469674444198608,
      "epoch": 0.10776113596576542,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999899317296288,
      "loss": 6.2775,
      "mean_token_accuracy": 0.11976648792624474,
      "num_tokens": 2407815.0,
      "step": 1385
    },
    {
      "entropy": 6.445081806182861,
      "epoch": 0.10815016533748298,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999896678301121,
      "loss": 6.1412,
      "mean_token_accuracy": 0.12592858001589774,
      "num_tokens": 2417300.0,
      "step": 1390
    },
    {
      "entropy": 6.435740280151367,
      "epoch": 0.10853919470920055,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999894005167352,
      "loss": 6.2916,
      "mean_token_accuracy": 0.1248296931385994,
      "num_tokens": 2426257.0,
      "step": 1395
    },
    {
      "entropy": 6.594287014007568,
      "epoch": 0.10892822408091811,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499989129789502,
      "loss": 6.3049,
      "mean_token_accuracy": 0.12577738165855407,
      "num_tokens": 2434595.0,
      "step": 1400
    },
    {
      "entropy": 6.599342203140258,
      "epoch": 0.10931725345263567,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999888556484172,
      "loss": 6.3221,
      "mean_token_accuracy": 0.12230012863874436,
      "num_tokens": 2443945.0,
      "step": 1405
    },
    {
      "entropy": 6.375091934204102,
      "epoch": 0.10970628282435324,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999885780934845,
      "loss": 6.1907,
      "mean_token_accuracy": 0.12613629177212715,
      "num_tokens": 2453141.0,
      "step": 1410
    },
    {
      "entropy": 6.551033687591553,
      "epoch": 0.1100953121960708,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999882971247081,
      "loss": 6.2644,
      "mean_token_accuracy": 0.12420548647642135,
      "num_tokens": 2461873.0,
      "step": 1415
    },
    {
      "entropy": 6.461436748504639,
      "epoch": 0.11048434156778837,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999880127420926,
      "loss": 6.2966,
      "mean_token_accuracy": 0.12048030272126198,
      "num_tokens": 2471238.0,
      "step": 1420
    },
    {
      "entropy": 6.48109073638916,
      "epoch": 0.11087337093950593,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999877249456421,
      "loss": 6.1976,
      "mean_token_accuracy": 0.12889449149370194,
      "num_tokens": 2479412.0,
      "step": 1425
    },
    {
      "entropy": 6.289962291717529,
      "epoch": 0.1112624003112235,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004999874337353609,
      "loss": 6.0753,
      "mean_token_accuracy": 0.12814749032258987,
      "num_tokens": 2488150.0,
      "step": 1430
    },
    {
      "entropy": 6.453941392898559,
      "epoch": 0.11165142968294106,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999871391112535,
      "loss": 6.2458,
      "mean_token_accuracy": 0.1274418205022812,
      "num_tokens": 2498103.0,
      "step": 1435
    },
    {
      "entropy": 6.403725814819336,
      "epoch": 0.11204045905465862,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999868410733244,
      "loss": 6.1165,
      "mean_token_accuracy": 0.12846592366695403,
      "num_tokens": 2506211.0,
      "step": 1440
    },
    {
      "entropy": 6.322660779953003,
      "epoch": 0.11242948842637619,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499986539621578,
      "loss": 6.1592,
      "mean_token_accuracy": 0.12967515960335732,
      "num_tokens": 2514743.0,
      "step": 1445
    },
    {
      "entropy": 6.377335548400879,
      "epoch": 0.11281851779809375,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999862347560191,
      "loss": 6.143,
      "mean_token_accuracy": 0.12324637621641159,
      "num_tokens": 2523824.0,
      "step": 1450
    },
    {
      "entropy": 6.491170978546142,
      "epoch": 0.11320754716981132,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999859264766521,
      "loss": 6.2174,
      "mean_token_accuracy": 0.12356971949338913,
      "num_tokens": 2532360.0,
      "step": 1455
    },
    {
      "entropy": 6.422462749481201,
      "epoch": 0.11359657654152888,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999856147834817,
      "loss": 6.1649,
      "mean_token_accuracy": 0.13105847761034967,
      "num_tokens": 2540547.0,
      "step": 1460
    },
    {
      "entropy": 6.408833074569702,
      "epoch": 0.11398560591324645,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999852996765129,
      "loss": 6.1838,
      "mean_token_accuracy": 0.12942414954304696,
      "num_tokens": 2549193.0,
      "step": 1465
    },
    {
      "entropy": 6.430090045928955,
      "epoch": 0.11437463528496401,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999849811557502,
      "loss": 6.2007,
      "mean_token_accuracy": 0.12359017059206963,
      "num_tokens": 2557553.0,
      "step": 1470
    },
    {
      "entropy": 6.530663251876831,
      "epoch": 0.11476366465668159,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999846592211984,
      "loss": 6.2997,
      "mean_token_accuracy": 0.12577586844563485,
      "num_tokens": 2565945.0,
      "step": 1475
    },
    {
      "entropy": 6.482461261749267,
      "epoch": 0.11515269402839914,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004999843338728626,
      "loss": 6.1625,
      "mean_token_accuracy": 0.12412880286574364,
      "num_tokens": 2574364.0,
      "step": 1480
    },
    {
      "entropy": 6.351671361923218,
      "epoch": 0.1155417234001167,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999840051107476,
      "loss": 6.1036,
      "mean_token_accuracy": 0.1279638461768627,
      "num_tokens": 2582825.0,
      "step": 1485
    },
    {
      "entropy": 6.4820616245269775,
      "epoch": 0.11593075277183428,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999836729348584,
      "loss": 6.2311,
      "mean_token_accuracy": 0.12474912479519844,
      "num_tokens": 2591628.0,
      "step": 1490
    },
    {
      "entropy": 6.486547756195068,
      "epoch": 0.11631978214355183,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999833373452,
      "loss": 6.2322,
      "mean_token_accuracy": 0.11734669506549836,
      "num_tokens": 2600436.0,
      "step": 1495
    },
    {
      "entropy": 6.3622339248657225,
      "epoch": 0.1167088115152694,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999829983417778,
      "loss": 6.1925,
      "mean_token_accuracy": 0.12901659011840821,
      "num_tokens": 2609709.0,
      "step": 1500
    },
    {
      "entropy": 6.497528409957885,
      "epoch": 0.11709784088698696,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999826559245965,
      "loss": 6.1939,
      "mean_token_accuracy": 0.12794017940759658,
      "num_tokens": 2617992.0,
      "step": 1505
    },
    {
      "entropy": 6.297898578643799,
      "epoch": 0.11748687025870454,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999823100936615,
      "loss": 6.1081,
      "mean_token_accuracy": 0.12741497829556464,
      "num_tokens": 2626706.0,
      "step": 1510
    },
    {
      "entropy": 6.547286319732666,
      "epoch": 0.1178758996304221,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999819608489781,
      "loss": 6.1859,
      "mean_token_accuracy": 0.13434533700346946,
      "num_tokens": 2634642.0,
      "step": 1515
    },
    {
      "entropy": 6.450995445251465,
      "epoch": 0.11826492900213967,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999816081905515,
      "loss": 6.2134,
      "mean_token_accuracy": 0.12014995068311692,
      "num_tokens": 2644160.0,
      "step": 1520
    },
    {
      "entropy": 6.299431896209716,
      "epoch": 0.11865395837385723,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999812521183872,
      "loss": 6.1913,
      "mean_token_accuracy": 0.12312416806817054,
      "num_tokens": 2653390.0,
      "step": 1525
    },
    {
      "entropy": 6.493148279190064,
      "epoch": 0.11904298774557479,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999808926324903,
      "loss": 6.2347,
      "mean_token_accuracy": 0.1258779875934124,
      "num_tokens": 2662561.0,
      "step": 1530
    },
    {
      "entropy": 6.428391170501709,
      "epoch": 0.11943201711729236,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004999805297328665,
      "loss": 6.1935,
      "mean_token_accuracy": 0.12642016857862473,
      "num_tokens": 2670694.0,
      "step": 1535
    },
    {
      "entropy": 6.3111365795135494,
      "epoch": 0.11982104648900992,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999801634195214,
      "loss": 6.078,
      "mean_token_accuracy": 0.13140376582741736,
      "num_tokens": 2679583.0,
      "step": 1540
    },
    {
      "entropy": 6.370851612091064,
      "epoch": 0.12021007586072749,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999797936924603,
      "loss": 6.0818,
      "mean_token_accuracy": 0.130515868216753,
      "num_tokens": 2688429.0,
      "step": 1545
    },
    {
      "entropy": 6.420472621917725,
      "epoch": 0.12059910523244505,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499979420551689,
      "loss": 6.1548,
      "mean_token_accuracy": 0.12875235974788665,
      "num_tokens": 2697134.0,
      "step": 1550
    },
    {
      "entropy": 6.406427812576294,
      "epoch": 0.12098813460416262,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499979043997213,
      "loss": 6.2798,
      "mean_token_accuracy": 0.1203388474881649,
      "num_tokens": 2705641.0,
      "step": 1555
    },
    {
      "entropy": 6.430809783935547,
      "epoch": 0.12137716397588018,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999786640290381,
      "loss": 6.2262,
      "mean_token_accuracy": 0.13087129667401315,
      "num_tokens": 2714977.0,
      "step": 1560
    },
    {
      "entropy": 6.429128837585449,
      "epoch": 0.12176619334759774,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999782806471701,
      "loss": 6.0765,
      "mean_token_accuracy": 0.1384890541434288,
      "num_tokens": 2723120.0,
      "step": 1565
    },
    {
      "entropy": 6.230881023406982,
      "epoch": 0.12215522271931531,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999778938516147,
      "loss": 6.0843,
      "mean_token_accuracy": 0.13497595563530923,
      "num_tokens": 2731964.0,
      "step": 1570
    },
    {
      "entropy": 6.525674676895141,
      "epoch": 0.12254425209103287,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499977503642378,
      "loss": 6.1425,
      "mean_token_accuracy": 0.12706299051642417,
      "num_tokens": 2741537.0,
      "step": 1575
    },
    {
      "entropy": 6.336614084243775,
      "epoch": 0.12293328146275044,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999771100194656,
      "loss": 6.2488,
      "mean_token_accuracy": 0.1276876747608185,
      "num_tokens": 2751522.0,
      "step": 1580
    },
    {
      "entropy": 6.4886940002441404,
      "epoch": 0.123322310834468,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999767129828837,
      "loss": 6.2514,
      "mean_token_accuracy": 0.11896501258015632,
      "num_tokens": 2761100.0,
      "step": 1585
    },
    {
      "entropy": 6.379914951324463,
      "epoch": 0.12371134020618557,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999763125326383,
      "loss": 6.1805,
      "mean_token_accuracy": 0.12824108004570006,
      "num_tokens": 2769171.0,
      "step": 1590
    },
    {
      "entropy": 6.368760776519776,
      "epoch": 0.12410036957790313,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999759086687355,
      "loss": 6.1265,
      "mean_token_accuracy": 0.12826206013560296,
      "num_tokens": 2778021.0,
      "step": 1595
    },
    {
      "entropy": 6.2863696098327635,
      "epoch": 0.1244893989496207,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999755013911813,
      "loss": 6.0772,
      "mean_token_accuracy": 0.1298437587916851,
      "num_tokens": 2787206.0,
      "step": 1600
    },
    {
      "entropy": 6.341671276092529,
      "epoch": 0.12487842832133826,
      "grad_norm": 1.125,
      "learning_rate": 0.000499975090699982,
      "loss": 6.1316,
      "mean_token_accuracy": 0.1293750949203968,
      "num_tokens": 2796051.0,
      "step": 1605
    },
    {
      "entropy": 6.3663938522338865,
      "epoch": 0.12526745769305583,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999746765951438,
      "loss": 6.0874,
      "mean_token_accuracy": 0.13239880800247192,
      "num_tokens": 2804961.0,
      "step": 1610
    },
    {
      "entropy": 6.326526165008545,
      "epoch": 0.1256564870647734,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999742590766729,
      "loss": 6.0454,
      "mean_token_accuracy": 0.13350202292203903,
      "num_tokens": 2813691.0,
      "step": 1615
    },
    {
      "entropy": 6.385087394714356,
      "epoch": 0.12604551643649095,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999738381445757,
      "loss": 6.1812,
      "mean_token_accuracy": 0.12938434183597564,
      "num_tokens": 2822248.0,
      "step": 1620
    },
    {
      "entropy": 6.396917104721069,
      "epoch": 0.1264345458082085,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999734137988586,
      "loss": 6.142,
      "mean_token_accuracy": 0.130577003210783,
      "num_tokens": 2830826.0,
      "step": 1625
    },
    {
      "entropy": 6.3590367317199705,
      "epoch": 0.1268235751799261,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499972986039528,
      "loss": 6.1216,
      "mean_token_accuracy": 0.12951069474220275,
      "num_tokens": 2839766.0,
      "step": 1630
    },
    {
      "entropy": 6.207428693771362,
      "epoch": 0.12721260455164365,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999725548665904,
      "loss": 6.0657,
      "mean_token_accuracy": 0.13443321511149406,
      "num_tokens": 2848728.0,
      "step": 1635
    },
    {
      "entropy": 6.306617021560669,
      "epoch": 0.1276016339233612,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999721202800524,
      "loss": 6.0904,
      "mean_token_accuracy": 0.13172278329730033,
      "num_tokens": 2857221.0,
      "step": 1640
    },
    {
      "entropy": 6.388878870010376,
      "epoch": 0.12799066329507877,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999716822799206,
      "loss": 6.1457,
      "mean_token_accuracy": 0.13395920395851135,
      "num_tokens": 2865314.0,
      "step": 1645
    },
    {
      "entropy": 6.261692237854004,
      "epoch": 0.12837969266679633,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999712408662016,
      "loss": 6.0185,
      "mean_token_accuracy": 0.13654017597436904,
      "num_tokens": 2873654.0,
      "step": 1650
    },
    {
      "entropy": 6.265604686737061,
      "epoch": 0.12876872203851392,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999707960389021,
      "loss": 5.9357,
      "mean_token_accuracy": 0.14313312098383904,
      "num_tokens": 2881604.0,
      "step": 1655
    },
    {
      "entropy": 6.309537696838379,
      "epoch": 0.12915775141023147,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999703477980288,
      "loss": 6.1277,
      "mean_token_accuracy": 0.12997377887368203,
      "num_tokens": 2890427.0,
      "step": 1660
    },
    {
      "entropy": 6.385668992996216,
      "epoch": 0.12954678078194903,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999698961435885,
      "loss": 6.0921,
      "mean_token_accuracy": 0.13262823969125748,
      "num_tokens": 2899206.0,
      "step": 1665
    },
    {
      "entropy": 6.25797815322876,
      "epoch": 0.1299358101536666,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999694410755882,
      "loss": 6.0263,
      "mean_token_accuracy": 0.13411636501550675,
      "num_tokens": 2907738.0,
      "step": 1670
    },
    {
      "entropy": 6.371025228500367,
      "epoch": 0.13032483952538418,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999689825940347,
      "loss": 6.2274,
      "mean_token_accuracy": 0.12602662891149521,
      "num_tokens": 2916881.0,
      "step": 1675
    },
    {
      "entropy": 6.40340576171875,
      "epoch": 0.13071386889710174,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999685206989349,
      "loss": 6.1783,
      "mean_token_accuracy": 0.12609613835811614,
      "num_tokens": 2926467.0,
      "step": 1680
    },
    {
      "entropy": 6.337603616714477,
      "epoch": 0.1311028982688193,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999680553902959,
      "loss": 6.1299,
      "mean_token_accuracy": 0.13105945885181428,
      "num_tokens": 2934671.0,
      "step": 1685
    },
    {
      "entropy": 6.33728141784668,
      "epoch": 0.13149192764053685,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999675866681247,
      "loss": 6.1511,
      "mean_token_accuracy": 0.12666159719228745,
      "num_tokens": 2943925.0,
      "step": 1690
    },
    {
      "entropy": 6.37947564125061,
      "epoch": 0.1318809570122544,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004999671145324286,
      "loss": 6.1889,
      "mean_token_accuracy": 0.1274726241827011,
      "num_tokens": 2952118.0,
      "step": 1695
    },
    {
      "entropy": 6.279305553436279,
      "epoch": 0.132269986383972,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999666389832144,
      "loss": 6.1581,
      "mean_token_accuracy": 0.12846080139279364,
      "num_tokens": 2961110.0,
      "step": 1700
    },
    {
      "entropy": 6.348481130599976,
      "epoch": 0.13265901575568956,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999661600204897,
      "loss": 6.0874,
      "mean_token_accuracy": 0.13976106271147729,
      "num_tokens": 2968877.0,
      "step": 1705
    },
    {
      "entropy": 6.393802261352539,
      "epoch": 0.13304804512740712,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999656776442615,
      "loss": 6.1459,
      "mean_token_accuracy": 0.12958939224481583,
      "num_tokens": 2977346.0,
      "step": 1710
    },
    {
      "entropy": 6.370845556259155,
      "epoch": 0.13343707449912467,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999651918545372,
      "loss": 6.1958,
      "mean_token_accuracy": 0.12310019731521607,
      "num_tokens": 2987494.0,
      "step": 1715
    },
    {
      "entropy": 6.3494658946990965,
      "epoch": 0.13382610387084226,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999647026513243,
      "loss": 6.0994,
      "mean_token_accuracy": 0.13097773119807243,
      "num_tokens": 2996728.0,
      "step": 1720
    },
    {
      "entropy": 6.2700177192687985,
      "epoch": 0.13421513324255982,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00049996421003463,
      "loss": 6.1329,
      "mean_token_accuracy": 0.12893041744828224,
      "num_tokens": 3005810.0,
      "step": 1725
    },
    {
      "entropy": 6.3634881496429445,
      "epoch": 0.13460416261427738,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004999637140044619,
      "loss": 6.1633,
      "mean_token_accuracy": 0.12823313996195793,
      "num_tokens": 3014603.0,
      "step": 1730
    },
    {
      "entropy": 6.431996059417725,
      "epoch": 0.13499319198599494,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999632145608275,
      "loss": 6.1479,
      "mean_token_accuracy": 0.13290574923157691,
      "num_tokens": 3023170.0,
      "step": 1735
    },
    {
      "entropy": 6.219868755340576,
      "epoch": 0.1353822213577125,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999627117037343,
      "loss": 5.9015,
      "mean_token_accuracy": 0.14494576007127763,
      "num_tokens": 3030565.0,
      "step": 1740
    },
    {
      "entropy": 6.2956647872924805,
      "epoch": 0.13577125072943008,
      "grad_norm": 1.03125,
      "learning_rate": 0.00049996220543319,
      "loss": 6.1691,
      "mean_token_accuracy": 0.13108205273747445,
      "num_tokens": 3040124.0,
      "step": 1745
    },
    {
      "entropy": 6.294409704208374,
      "epoch": 0.13616028010114764,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999616957492025,
      "loss": 6.0324,
      "mean_token_accuracy": 0.13331486508250237,
      "num_tokens": 3048458.0,
      "step": 1750
    },
    {
      "entropy": 6.263208246231079,
      "epoch": 0.1365493094728652,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999611826517793,
      "loss": 6.006,
      "mean_token_accuracy": 0.12804921418428422,
      "num_tokens": 3057926.0,
      "step": 1755
    },
    {
      "entropy": 6.294978046417237,
      "epoch": 0.13693833884458276,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999606661409281,
      "loss": 6.033,
      "mean_token_accuracy": 0.13218592926859857,
      "num_tokens": 3066683.0,
      "step": 1760
    },
    {
      "entropy": 6.193322563171387,
      "epoch": 0.13732736821630034,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999601462166569,
      "loss": 6.1126,
      "mean_token_accuracy": 0.12744091004133223,
      "num_tokens": 3075117.0,
      "step": 1765
    },
    {
      "entropy": 6.409311723709107,
      "epoch": 0.1377163975880179,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999596228789736,
      "loss": 6.1632,
      "mean_token_accuracy": 0.12578216195106506,
      "num_tokens": 3084399.0,
      "step": 1770
    },
    {
      "entropy": 6.222201776504517,
      "epoch": 0.13810542695973546,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999590961278859,
      "loss": 5.9466,
      "mean_token_accuracy": 0.13312947526574134,
      "num_tokens": 3093030.0,
      "step": 1775
    },
    {
      "entropy": 6.30834550857544,
      "epoch": 0.13849445633145302,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999585659634021,
      "loss": 5.9956,
      "mean_token_accuracy": 0.13578457087278367,
      "num_tokens": 3101670.0,
      "step": 1780
    },
    {
      "entropy": 6.144382953643799,
      "epoch": 0.13888348570317058,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999580323855301,
      "loss": 5.9948,
      "mean_token_accuracy": 0.13799581676721573,
      "num_tokens": 3109703.0,
      "step": 1785
    },
    {
      "entropy": 6.229348421096802,
      "epoch": 0.13927251507488816,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999574953942781,
      "loss": 5.9804,
      "mean_token_accuracy": 0.1379885770380497,
      "num_tokens": 3118388.0,
      "step": 1790
    },
    {
      "entropy": 6.2648790836334225,
      "epoch": 0.13966154444660572,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999569549896541,
      "loss": 5.9672,
      "mean_token_accuracy": 0.1371411792933941,
      "num_tokens": 3126466.0,
      "step": 1795
    },
    {
      "entropy": 6.259693765640259,
      "epoch": 0.14005057381832328,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999564111716665,
      "loss": 6.0299,
      "mean_token_accuracy": 0.13593954741954803,
      "num_tokens": 3135322.0,
      "step": 1800
    },
    {
      "entropy": 6.272290372848511,
      "epoch": 0.14043960319004084,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999558639403232,
      "loss": 6.0157,
      "mean_token_accuracy": 0.14133850261569023,
      "num_tokens": 3144498.0,
      "step": 1805
    },
    {
      "entropy": 6.185325717926025,
      "epoch": 0.1408286325617584,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999553132956328,
      "loss": 5.9996,
      "mean_token_accuracy": 0.12976517975330354,
      "num_tokens": 3153573.0,
      "step": 1810
    },
    {
      "entropy": 6.203995656967163,
      "epoch": 0.14121766193347599,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999547592376035,
      "loss": 5.9301,
      "mean_token_accuracy": 0.13893781676888467,
      "num_tokens": 3161705.0,
      "step": 1815
    },
    {
      "entropy": 6.24528398513794,
      "epoch": 0.14160669130519354,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999542017662438,
      "loss": 6.0915,
      "mean_token_accuracy": 0.13699387311935424,
      "num_tokens": 3170068.0,
      "step": 1820
    },
    {
      "entropy": 6.246503305435181,
      "epoch": 0.1419957206769111,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999536408815623,
      "loss": 6.0929,
      "mean_token_accuracy": 0.13167436718940734,
      "num_tokens": 3178140.0,
      "step": 1825
    },
    {
      "entropy": 6.2965551853179935,
      "epoch": 0.14238475004862866,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999530765835672,
      "loss": 6.0795,
      "mean_token_accuracy": 0.13575641214847564,
      "num_tokens": 3186518.0,
      "step": 1830
    },
    {
      "entropy": 6.261485242843628,
      "epoch": 0.14277377942034625,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999525088722673,
      "loss": 6.0548,
      "mean_token_accuracy": 0.13169616609811782,
      "num_tokens": 3195246.0,
      "step": 1835
    },
    {
      "entropy": 6.323240137100219,
      "epoch": 0.1431628087920638,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999519377476711,
      "loss": 6.0778,
      "mean_token_accuracy": 0.13250553533434867,
      "num_tokens": 3203946.0,
      "step": 1840
    },
    {
      "entropy": 6.248462677001953,
      "epoch": 0.14355183816378136,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999513632097872,
      "loss": 6.0928,
      "mean_token_accuracy": 0.13283114582300187,
      "num_tokens": 3212208.0,
      "step": 1845
    },
    {
      "entropy": 6.229254961013794,
      "epoch": 0.14394086753549892,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999507852586245,
      "loss": 6.0241,
      "mean_token_accuracy": 0.13466759771108627,
      "num_tokens": 3221572.0,
      "step": 1850
    },
    {
      "entropy": 6.36988377571106,
      "epoch": 0.14432989690721648,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999502038941917,
      "loss": 6.1306,
      "mean_token_accuracy": 0.12890521064400673,
      "num_tokens": 3230898.0,
      "step": 1855
    },
    {
      "entropy": 6.229429626464844,
      "epoch": 0.14471892627893407,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999496191164975,
      "loss": 6.1211,
      "mean_token_accuracy": 0.1284354142844677,
      "num_tokens": 3240524.0,
      "step": 1860
    },
    {
      "entropy": 6.201262140274048,
      "epoch": 0.14510795565065163,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499949030925551,
      "loss": 5.8856,
      "mean_token_accuracy": 0.13885571360588073,
      "num_tokens": 3248864.0,
      "step": 1865
    },
    {
      "entropy": 6.266730690002442,
      "epoch": 0.14549698502236919,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999484393213609,
      "loss": 6.068,
      "mean_token_accuracy": 0.1342063754796982,
      "num_tokens": 3256823.0,
      "step": 1870
    },
    {
      "entropy": 6.170843505859375,
      "epoch": 0.14588601439408674,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999478443039363,
      "loss": 5.9629,
      "mean_token_accuracy": 0.14102984592318535,
      "num_tokens": 3265621.0,
      "step": 1875
    },
    {
      "entropy": 6.1978428840637205,
      "epoch": 0.14627504376580433,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999472458732862,
      "loss": 5.9998,
      "mean_token_accuracy": 0.1362139955163002,
      "num_tokens": 3273515.0,
      "step": 1880
    },
    {
      "entropy": 6.338559913635254,
      "epoch": 0.1466640731375219,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999466440294198,
      "loss": 6.0745,
      "mean_token_accuracy": 0.13152379617094995,
      "num_tokens": 3282283.0,
      "step": 1885
    },
    {
      "entropy": 6.185009860992432,
      "epoch": 0.14705310250923945,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499946038772346,
      "loss": 6.0647,
      "mean_token_accuracy": 0.13816771656274796,
      "num_tokens": 3290792.0,
      "step": 1890
    },
    {
      "entropy": 6.146698713302612,
      "epoch": 0.147442131880957,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999454301020741,
      "loss": 5.9306,
      "mean_token_accuracy": 0.1381203569471836,
      "num_tokens": 3299430.0,
      "step": 1895
    },
    {
      "entropy": 6.161927223205566,
      "epoch": 0.14783116125267456,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999448180186133,
      "loss": 5.9877,
      "mean_token_accuracy": 0.14127052575349808,
      "num_tokens": 3307949.0,
      "step": 1900
    },
    {
      "entropy": 6.291336107254028,
      "epoch": 0.14822019062439215,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499944202521973,
      "loss": 6.0399,
      "mean_token_accuracy": 0.13172024860978127,
      "num_tokens": 3316900.0,
      "step": 1905
    },
    {
      "entropy": 6.205208778381348,
      "epoch": 0.1486092199961097,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999435836121624,
      "loss": 5.9569,
      "mean_token_accuracy": 0.1432355299592018,
      "num_tokens": 3325656.0,
      "step": 1910
    },
    {
      "entropy": 6.190450620651245,
      "epoch": 0.14899824936782727,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499942961289191,
      "loss": 6.198,
      "mean_token_accuracy": 0.1225706160068512,
      "num_tokens": 3334219.0,
      "step": 1915
    },
    {
      "entropy": 6.228388118743896,
      "epoch": 0.14938727873954483,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999423355530683,
      "loss": 5.9451,
      "mean_token_accuracy": 0.13421944826841353,
      "num_tokens": 3343637.0,
      "step": 1920
    },
    {
      "entropy": 6.189314556121826,
      "epoch": 0.1497763081112624,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999417064038035,
      "loss": 6.0871,
      "mean_token_accuracy": 0.13039903715252876,
      "num_tokens": 3352624.0,
      "step": 1925
    },
    {
      "entropy": 6.1881711959838865,
      "epoch": 0.15016533748297997,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999410738414065,
      "loss": 6.0079,
      "mean_token_accuracy": 0.14228348284959794,
      "num_tokens": 3361394.0,
      "step": 1930
    },
    {
      "entropy": 6.3036641597747805,
      "epoch": 0.15055436685469753,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999404378658866,
      "loss": 6.0535,
      "mean_token_accuracy": 0.13427391052246093,
      "num_tokens": 3369406.0,
      "step": 1935
    },
    {
      "entropy": 6.243283081054687,
      "epoch": 0.1509433962264151,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999397984772536,
      "loss": 5.9956,
      "mean_token_accuracy": 0.13894839733839034,
      "num_tokens": 3377737.0,
      "step": 1940
    },
    {
      "entropy": 6.194996166229248,
      "epoch": 0.15133242559813265,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999391556755172,
      "loss": 6.051,
      "mean_token_accuracy": 0.1303106352686882,
      "num_tokens": 3385514.0,
      "step": 1945
    },
    {
      "entropy": 6.298865890502929,
      "epoch": 0.15172145496985023,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999385094606872,
      "loss": 5.9586,
      "mean_token_accuracy": 0.13934986293315887,
      "num_tokens": 3394500.0,
      "step": 1950
    },
    {
      "entropy": 6.167548418045044,
      "epoch": 0.1521104843415678,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999378598327733,
      "loss": 5.9497,
      "mean_token_accuracy": 0.1368454597890377,
      "num_tokens": 3403289.0,
      "step": 1955
    },
    {
      "entropy": 6.102401065826416,
      "epoch": 0.15249951371328535,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999372067917854,
      "loss": 5.881,
      "mean_token_accuracy": 0.14984163641929626,
      "num_tokens": 3410892.0,
      "step": 1960
    },
    {
      "entropy": 6.029754686355591,
      "epoch": 0.1528885430850029,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999365503377335,
      "loss": 5.9761,
      "mean_token_accuracy": 0.13527074232697486,
      "num_tokens": 3420329.0,
      "step": 1965
    },
    {
      "entropy": 6.220583820343018,
      "epoch": 0.1532775724567205,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999358904706273,
      "loss": 5.9717,
      "mean_token_accuracy": 0.13858482763171195,
      "num_tokens": 3429429.0,
      "step": 1970
    },
    {
      "entropy": 6.295728540420532,
      "epoch": 0.15366660182843805,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999352271904771,
      "loss": 6.0824,
      "mean_token_accuracy": 0.13455277010798455,
      "num_tokens": 3438198.0,
      "step": 1975
    },
    {
      "entropy": 6.339246654510498,
      "epoch": 0.1540556312001556,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999345604972928,
      "loss": 6.2075,
      "mean_token_accuracy": 0.13310084342956544,
      "num_tokens": 3447732.0,
      "step": 1980
    },
    {
      "entropy": 6.231119155883789,
      "epoch": 0.15444466057187317,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999338903910846,
      "loss": 5.9915,
      "mean_token_accuracy": 0.1366339311003685,
      "num_tokens": 3456392.0,
      "step": 1985
    },
    {
      "entropy": 6.148554086685181,
      "epoch": 0.15483368994359073,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999332168718626,
      "loss": 6.043,
      "mean_token_accuracy": 0.13072005808353424,
      "num_tokens": 3464473.0,
      "step": 1990
    },
    {
      "entropy": 6.224080324172974,
      "epoch": 0.15522271931530832,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999325399396371,
      "loss": 5.9976,
      "mean_token_accuracy": 0.13627456203103067,
      "num_tokens": 3473740.0,
      "step": 1995
    },
    {
      "entropy": 6.042808628082275,
      "epoch": 0.15561174868702587,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999318595944183,
      "loss": 5.9012,
      "mean_token_accuracy": 0.1403045043349266,
      "num_tokens": 3482249.0,
      "step": 2000
    },
    {
      "entropy": 6.339179420471192,
      "epoch": 0.15600077805874343,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999311758362166,
      "loss": 6.1273,
      "mean_token_accuracy": 0.1341732420027256,
      "num_tokens": 3490753.0,
      "step": 2005
    },
    {
      "entropy": 6.26419415473938,
      "epoch": 0.156389807430461,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999304886650422,
      "loss": 6.0787,
      "mean_token_accuracy": 0.13692996352910997,
      "num_tokens": 3499373.0,
      "step": 2010
    },
    {
      "entropy": 6.20645318031311,
      "epoch": 0.15677883680217858,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999297980809058,
      "loss": 6.0476,
      "mean_token_accuracy": 0.13018234223127365,
      "num_tokens": 3508344.0,
      "step": 2015
    },
    {
      "entropy": 6.330630493164063,
      "epoch": 0.15716786617389614,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999291040838176,
      "loss": 6.0069,
      "mean_token_accuracy": 0.13502927049994468,
      "num_tokens": 3517009.0,
      "step": 2020
    },
    {
      "entropy": 6.123447847366333,
      "epoch": 0.1575568955456137,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999284066737884,
      "loss": 6.0392,
      "mean_token_accuracy": 0.13788192793726922,
      "num_tokens": 3525989.0,
      "step": 2025
    },
    {
      "entropy": 6.272771978378296,
      "epoch": 0.15794592491733125,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999277058508285,
      "loss": 5.9269,
      "mean_token_accuracy": 0.1317255400121212,
      "num_tokens": 3533947.0,
      "step": 2030
    },
    {
      "entropy": 6.155836677551269,
      "epoch": 0.1583349542890488,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999270016149486,
      "loss": 5.9925,
      "mean_token_accuracy": 0.1387178160250187,
      "num_tokens": 3541974.0,
      "step": 2035
    },
    {
      "entropy": 6.133522939682007,
      "epoch": 0.1587239836607664,
      "grad_norm": 1.25,
      "learning_rate": 0.0004999262939661596,
      "loss": 5.9342,
      "mean_token_accuracy": 0.14313246235251426,
      "num_tokens": 3550436.0,
      "step": 2040
    },
    {
      "entropy": 6.209865283966065,
      "epoch": 0.15911301303248396,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999255829044721,
      "loss": 5.9338,
      "mean_token_accuracy": 0.14274076968431473,
      "num_tokens": 3558283.0,
      "step": 2045
    },
    {
      "entropy": 6.033965444564819,
      "epoch": 0.15950204240420152,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999248684298968,
      "loss": 5.8505,
      "mean_token_accuracy": 0.14578440189361572,
      "num_tokens": 3566968.0,
      "step": 2050
    },
    {
      "entropy": 6.2075746059417725,
      "epoch": 0.15989107177591907,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999241505424447,
      "loss": 5.974,
      "mean_token_accuracy": 0.13596091344952582,
      "num_tokens": 3575580.0,
      "step": 2055
    },
    {
      "entropy": 6.221923780441284,
      "epoch": 0.16028010114763663,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999234292421265,
      "loss": 6.0571,
      "mean_token_accuracy": 0.13546983748674393,
      "num_tokens": 3584389.0,
      "step": 2060
    },
    {
      "entropy": 6.1582318305969235,
      "epoch": 0.16066913051935422,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999227045289535,
      "loss": 5.883,
      "mean_token_accuracy": 0.13753796219825745,
      "num_tokens": 3593009.0,
      "step": 2065
    },
    {
      "entropy": 6.11220064163208,
      "epoch": 0.16105815989107178,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999219764029363,
      "loss": 5.9179,
      "mean_token_accuracy": 0.13649867624044418,
      "num_tokens": 3601914.0,
      "step": 2070
    },
    {
      "entropy": 6.034575176239014,
      "epoch": 0.16144718926278934,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999212448640861,
      "loss": 5.8876,
      "mean_token_accuracy": 0.14072516635060311,
      "num_tokens": 3609883.0,
      "step": 2075
    },
    {
      "entropy": 6.121336221694946,
      "epoch": 0.1618362186345069,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999205099124141,
      "loss": 5.8222,
      "mean_token_accuracy": 0.14431814327836037,
      "num_tokens": 3618103.0,
      "step": 2080
    },
    {
      "entropy": 6.2555018901824955,
      "epoch": 0.16222524800622448,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999197715479313,
      "loss": 6.1008,
      "mean_token_accuracy": 0.13108571246266365,
      "num_tokens": 3625861.0,
      "step": 2085
    },
    {
      "entropy": 6.118966913223266,
      "epoch": 0.16261427737794204,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999190297706489,
      "loss": 5.883,
      "mean_token_accuracy": 0.13489673435688018,
      "num_tokens": 3634498.0,
      "step": 2090
    },
    {
      "entropy": 6.024101257324219,
      "epoch": 0.1630033067496596,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999182845805783,
      "loss": 5.8272,
      "mean_token_accuracy": 0.13758305013179778,
      "num_tokens": 3644186.0,
      "step": 2095
    },
    {
      "entropy": 6.0807852268219,
      "epoch": 0.16339233612137716,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999175359777306,
      "loss": 5.9728,
      "mean_token_accuracy": 0.13705984950065614,
      "num_tokens": 3652831.0,
      "step": 2100
    },
    {
      "entropy": 6.256798458099365,
      "epoch": 0.16378136549309472,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999167839621174,
      "loss": 5.8406,
      "mean_token_accuracy": 0.14277086704969405,
      "num_tokens": 3661230.0,
      "step": 2105
    },
    {
      "entropy": 6.076580953598023,
      "epoch": 0.1641703948648123,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999160285337501,
      "loss": 5.9843,
      "mean_token_accuracy": 0.13940260484814643,
      "num_tokens": 3669250.0,
      "step": 2110
    },
    {
      "entropy": 6.178564977645874,
      "epoch": 0.16455942423652986,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999152696926399,
      "loss": 5.9566,
      "mean_token_accuracy": 0.13556850478053092,
      "num_tokens": 3677568.0,
      "step": 2115
    },
    {
      "entropy": 6.145662260055542,
      "epoch": 0.16494845360824742,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999145074387985,
      "loss": 5.9574,
      "mean_token_accuracy": 0.1433981440961361,
      "num_tokens": 3685488.0,
      "step": 2120
    },
    {
      "entropy": 6.112370824813842,
      "epoch": 0.16533748297996498,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999137417722374,
      "loss": 5.9228,
      "mean_token_accuracy": 0.1389925181865692,
      "num_tokens": 3694357.0,
      "step": 2125
    },
    {
      "entropy": 6.203349447250366,
      "epoch": 0.16572651235168256,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999129726929684,
      "loss": 5.9896,
      "mean_token_accuracy": 0.1359809495508671,
      "num_tokens": 3703173.0,
      "step": 2130
    },
    {
      "entropy": 6.132759428024292,
      "epoch": 0.16611554172340012,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999122002010029,
      "loss": 5.9595,
      "mean_token_accuracy": 0.1417309373617172,
      "num_tokens": 3712070.0,
      "step": 2135
    },
    {
      "entropy": 6.10051646232605,
      "epoch": 0.16650457109511768,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999114242963527,
      "loss": 5.8551,
      "mean_token_accuracy": 0.14509744718670844,
      "num_tokens": 3720245.0,
      "step": 2140
    },
    {
      "entropy": 6.081886148452758,
      "epoch": 0.16689360046683524,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999106449790298,
      "loss": 5.9421,
      "mean_token_accuracy": 0.13998848721385002,
      "num_tokens": 3730064.0,
      "step": 2145
    },
    {
      "entropy": 6.055927038192749,
      "epoch": 0.1672826298385528,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999098622490458,
      "loss": 5.9073,
      "mean_token_accuracy": 0.1380058228969574,
      "num_tokens": 3738741.0,
      "step": 2150
    },
    {
      "entropy": 6.19128189086914,
      "epoch": 0.16767165921027039,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999090761064125,
      "loss": 5.973,
      "mean_token_accuracy": 0.1361282914876938,
      "num_tokens": 3747472.0,
      "step": 2155
    },
    {
      "entropy": 6.153685426712036,
      "epoch": 0.16806068858198794,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499908286551142,
      "loss": 5.9996,
      "mean_token_accuracy": 0.1330367848277092,
      "num_tokens": 3756643.0,
      "step": 2160
    },
    {
      "entropy": 6.19166932106018,
      "epoch": 0.1684497179537055,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999074935832463,
      "loss": 6.1118,
      "mean_token_accuracy": 0.12960731536149978,
      "num_tokens": 3765490.0,
      "step": 2165
    },
    {
      "entropy": 6.169116973876953,
      "epoch": 0.16883874732542306,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999066972027373,
      "loss": 5.8612,
      "mean_token_accuracy": 0.13973119109869003,
      "num_tokens": 3773063.0,
      "step": 2170
    },
    {
      "entropy": 6.154691886901856,
      "epoch": 0.16922777669714065,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999058974096271,
      "loss": 6.0261,
      "mean_token_accuracy": 0.1344061642885208,
      "num_tokens": 3781763.0,
      "step": 2175
    },
    {
      "entropy": 6.188900423049927,
      "epoch": 0.1696168060688582,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499905094203928,
      "loss": 6.0189,
      "mean_token_accuracy": 0.1358758971095085,
      "num_tokens": 3790582.0,
      "step": 2180
    },
    {
      "entropy": 6.089183139801025,
      "epoch": 0.17000583544057576,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499904287585652,
      "loss": 5.827,
      "mean_token_accuracy": 0.14113157913088797,
      "num_tokens": 3798745.0,
      "step": 2185
    },
    {
      "entropy": 6.06810245513916,
      "epoch": 0.17039486481229332,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999034775548113,
      "loss": 5.9187,
      "mean_token_accuracy": 0.1380026899278164,
      "num_tokens": 3806901.0,
      "step": 2190
    },
    {
      "entropy": 6.149272918701172,
      "epoch": 0.17078389418401088,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004999026641114185,
      "loss": 5.8776,
      "mean_token_accuracy": 0.14875079840421676,
      "num_tokens": 3815742.0,
      "step": 2195
    },
    {
      "entropy": 6.092239236831665,
      "epoch": 0.17117292355572847,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999018472554857,
      "loss": 5.9658,
      "mean_token_accuracy": 0.13768358528614044,
      "num_tokens": 3824468.0,
      "step": 2200
    },
    {
      "entropy": 6.058383893966675,
      "epoch": 0.17156195292744603,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999010269870253,
      "loss": 5.9261,
      "mean_token_accuracy": 0.14319675862789155,
      "num_tokens": 3832834.0,
      "step": 2205
    },
    {
      "entropy": 6.115259170532227,
      "epoch": 0.17195098229916359,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999002033060498,
      "loss": 5.9802,
      "mean_token_accuracy": 0.1424368679523468,
      "num_tokens": 3841780.0,
      "step": 2210
    },
    {
      "entropy": 6.149002027511597,
      "epoch": 0.17234001167088114,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998993762125716,
      "loss": 6.0616,
      "mean_token_accuracy": 0.13416145145893096,
      "num_tokens": 3849840.0,
      "step": 2215
    },
    {
      "entropy": 6.18777437210083,
      "epoch": 0.17272904104259873,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998985457066034,
      "loss": 5.9945,
      "mean_token_accuracy": 0.13655759170651435,
      "num_tokens": 3858204.0,
      "step": 2220
    },
    {
      "entropy": 6.167106294631958,
      "epoch": 0.1731180704143163,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998977117881576,
      "loss": 5.9472,
      "mean_token_accuracy": 0.13820890188217164,
      "num_tokens": 3866615.0,
      "step": 2225
    },
    {
      "entropy": 6.092865514755249,
      "epoch": 0.17350709978603385,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998968744572472,
      "loss": 5.9817,
      "mean_token_accuracy": 0.1438558429479599,
      "num_tokens": 3876003.0,
      "step": 2230
    },
    {
      "entropy": 6.155641174316406,
      "epoch": 0.1738961291577514,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998960337138845,
      "loss": 5.8931,
      "mean_token_accuracy": 0.14134934842586516,
      "num_tokens": 3884362.0,
      "step": 2235
    },
    {
      "entropy": 6.0395012378692625,
      "epoch": 0.17428515852946896,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998951895580826,
      "loss": 5.8524,
      "mean_token_accuracy": 0.14561742097139357,
      "num_tokens": 3893177.0,
      "step": 2240
    },
    {
      "entropy": 6.08151388168335,
      "epoch": 0.17467418790118655,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998943419898542,
      "loss": 5.9069,
      "mean_token_accuracy": 0.14305766448378562,
      "num_tokens": 3903006.0,
      "step": 2245
    },
    {
      "entropy": 6.071657657623291,
      "epoch": 0.1750632172729041,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998934910092119,
      "loss": 5.7658,
      "mean_token_accuracy": 0.14124478846788407,
      "num_tokens": 3912256.0,
      "step": 2250
    },
    {
      "entropy": 6.0484153747558596,
      "epoch": 0.17545224664462167,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998926366161689,
      "loss": 5.9728,
      "mean_token_accuracy": 0.1354946233332157,
      "num_tokens": 3921121.0,
      "step": 2255
    },
    {
      "entropy": 6.089042663574219,
      "epoch": 0.17584127601633923,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998917788107383,
      "loss": 5.8974,
      "mean_token_accuracy": 0.1439758561551571,
      "num_tokens": 3929348.0,
      "step": 2260
    },
    {
      "entropy": 6.124771785736084,
      "epoch": 0.17623030538805678,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998909175929326,
      "loss": 6.0547,
      "mean_token_accuracy": 0.13560474514961243,
      "num_tokens": 3938073.0,
      "step": 2265
    },
    {
      "entropy": 6.176717805862427,
      "epoch": 0.17661933475977437,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998900529627653,
      "loss": 6.0357,
      "mean_token_accuracy": 0.13639540523290633,
      "num_tokens": 3946641.0,
      "step": 2270
    },
    {
      "entropy": 6.192051935195923,
      "epoch": 0.17700836413149193,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998891849202495,
      "loss": 5.8985,
      "mean_token_accuracy": 0.13934031128883362,
      "num_tokens": 3954586.0,
      "step": 2275
    },
    {
      "entropy": 6.054885196685791,
      "epoch": 0.1773973935032095,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499888313465398,
      "loss": 5.9597,
      "mean_token_accuracy": 0.14134562537074088,
      "num_tokens": 3963745.0,
      "step": 2280
    },
    {
      "entropy": 6.0358985424041744,
      "epoch": 0.17778642287492705,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998874385982247,
      "loss": 5.7746,
      "mean_token_accuracy": 0.14625951200723647,
      "num_tokens": 3972452.0,
      "step": 2285
    },
    {
      "entropy": 6.034676504135132,
      "epoch": 0.17817545224664463,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998865603187421,
      "loss": 5.8922,
      "mean_token_accuracy": 0.14618094712495805,
      "num_tokens": 3980770.0,
      "step": 2290
    },
    {
      "entropy": 6.117841482162476,
      "epoch": 0.1785644816183622,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499885678626964,
      "loss": 5.9268,
      "mean_token_accuracy": 0.1389491453766823,
      "num_tokens": 3989545.0,
      "step": 2295
    },
    {
      "entropy": 6.102830028533935,
      "epoch": 0.17895351099007975,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998847935229038,
      "loss": 5.8846,
      "mean_token_accuracy": 0.1403166502714157,
      "num_tokens": 3998034.0,
      "step": 2300
    },
    {
      "entropy": 6.014950513839722,
      "epoch": 0.1793425403617973,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998839050065746,
      "loss": 5.8995,
      "mean_token_accuracy": 0.1459874078631401,
      "num_tokens": 4007110.0,
      "step": 2305
    },
    {
      "entropy": 6.135031366348267,
      "epoch": 0.17973156973351487,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998830130779902,
      "loss": 5.9831,
      "mean_token_accuracy": 0.1331685684621334,
      "num_tokens": 4015930.0,
      "step": 2310
    },
    {
      "entropy": 6.003888893127441,
      "epoch": 0.18012059910523245,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499882117737164,
      "loss": 5.7744,
      "mean_token_accuracy": 0.14710479527711867,
      "num_tokens": 4024510.0,
      "step": 2315
    },
    {
      "entropy": 6.005897569656372,
      "epoch": 0.18050962847695,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998812189841096,
      "loss": 5.8876,
      "mean_token_accuracy": 0.14109528660774232,
      "num_tokens": 4033714.0,
      "step": 2320
    },
    {
      "entropy": 6.209692478179932,
      "epoch": 0.18089865784866757,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998803168188407,
      "loss": 6.0785,
      "mean_token_accuracy": 0.1327739931643009,
      "num_tokens": 4042376.0,
      "step": 2325
    },
    {
      "entropy": 6.037041902542114,
      "epoch": 0.18128768722038513,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998794112413708,
      "loss": 5.8403,
      "mean_token_accuracy": 0.14402748942375182,
      "num_tokens": 4050521.0,
      "step": 2330
    },
    {
      "entropy": 6.196396255493164,
      "epoch": 0.18167671659210272,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998785022517137,
      "loss": 6.01,
      "mean_token_accuracy": 0.13692177534103395,
      "num_tokens": 4059384.0,
      "step": 2335
    },
    {
      "entropy": 6.081640815734863,
      "epoch": 0.18206574596382027,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998775898498835,
      "loss": 5.9044,
      "mean_token_accuracy": 0.1395252026617527,
      "num_tokens": 4067655.0,
      "step": 2340
    },
    {
      "entropy": 6.1328636646270756,
      "epoch": 0.18245477533553783,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998766740358936,
      "loss": 5.8762,
      "mean_token_accuracy": 0.14391948282718658,
      "num_tokens": 4075828.0,
      "step": 2345
    },
    {
      "entropy": 6.070217180252075,
      "epoch": 0.1828438047072554,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998757548097582,
      "loss": 5.8281,
      "mean_token_accuracy": 0.14209648966789246,
      "num_tokens": 4083796.0,
      "step": 2350
    },
    {
      "entropy": 6.027770709991455,
      "epoch": 0.18323283407897295,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998748321714911,
      "loss": 5.9639,
      "mean_token_accuracy": 0.14366325587034226,
      "num_tokens": 4092981.0,
      "step": 2355
    },
    {
      "entropy": 6.0527486324310305,
      "epoch": 0.18362186345069054,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998739061211065,
      "loss": 5.7937,
      "mean_token_accuracy": 0.14437015652656554,
      "num_tokens": 4101363.0,
      "step": 2360
    },
    {
      "entropy": 6.061078214645386,
      "epoch": 0.1840108928224081,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998729766586181,
      "loss": 5.8782,
      "mean_token_accuracy": 0.13865672498941423,
      "num_tokens": 4110164.0,
      "step": 2365
    },
    {
      "entropy": 6.193390893936157,
      "epoch": 0.18439992219412565,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998720437840403,
      "loss": 5.9812,
      "mean_token_accuracy": 0.1382698208093643,
      "num_tokens": 4119098.0,
      "step": 2370
    },
    {
      "entropy": 6.045206737518311,
      "epoch": 0.1847889515658432,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998711074973871,
      "loss": 5.8919,
      "mean_token_accuracy": 0.14405231848359107,
      "num_tokens": 4127522.0,
      "step": 2375
    },
    {
      "entropy": 6.047341442108154,
      "epoch": 0.1851779809375608,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998701677986728,
      "loss": 5.9272,
      "mean_token_accuracy": 0.1469861164689064,
      "num_tokens": 4136311.0,
      "step": 2380
    },
    {
      "entropy": 6.066590690612793,
      "epoch": 0.18556701030927836,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998692246879116,
      "loss": 5.8304,
      "mean_token_accuracy": 0.1474180541932583,
      "num_tokens": 4145160.0,
      "step": 2385
    },
    {
      "entropy": 6.014827108383178,
      "epoch": 0.18595603968099592,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998682781651178,
      "loss": 5.8197,
      "mean_token_accuracy": 0.14966430738568307,
      "num_tokens": 4153921.0,
      "step": 2390
    },
    {
      "entropy": 6.020003175735473,
      "epoch": 0.18634506905271347,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998673282303059,
      "loss": 5.9001,
      "mean_token_accuracy": 0.13874251693487166,
      "num_tokens": 4162580.0,
      "step": 2395
    },
    {
      "entropy": 6.064752817153931,
      "epoch": 0.18673409842443103,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998663748834901,
      "loss": 5.8882,
      "mean_token_accuracy": 0.1460046499967575,
      "num_tokens": 4170623.0,
      "step": 2400
    },
    {
      "entropy": 6.044636297225952,
      "epoch": 0.18712312779614862,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998654181246851,
      "loss": 5.8728,
      "mean_token_accuracy": 0.14949819520115853,
      "num_tokens": 4178724.0,
      "step": 2405
    },
    {
      "entropy": 6.043307685852051,
      "epoch": 0.18751215716786618,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998644579539052,
      "loss": 5.8736,
      "mean_token_accuracy": 0.1448787271976471,
      "num_tokens": 4187773.0,
      "step": 2410
    },
    {
      "entropy": 6.046842908859253,
      "epoch": 0.18790118653958374,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499863494371165,
      "loss": 5.8886,
      "mean_token_accuracy": 0.1425296075642109,
      "num_tokens": 4196628.0,
      "step": 2415
    },
    {
      "entropy": 6.044779539108276,
      "epoch": 0.1882902159113013,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998625273764793,
      "loss": 5.9093,
      "mean_token_accuracy": 0.13880138099193573,
      "num_tokens": 4205090.0,
      "step": 2420
    },
    {
      "entropy": 6.079436826705932,
      "epoch": 0.18867924528301888,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998615569698626,
      "loss": 5.8185,
      "mean_token_accuracy": 0.14609641581773758,
      "num_tokens": 4212757.0,
      "step": 2425
    },
    {
      "entropy": 5.997606372833252,
      "epoch": 0.18906827465473644,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998605831513296,
      "loss": 5.9013,
      "mean_token_accuracy": 0.13931182399392128,
      "num_tokens": 4221556.0,
      "step": 2430
    },
    {
      "entropy": 5.99919490814209,
      "epoch": 0.189457304026454,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998596059208953,
      "loss": 5.8797,
      "mean_token_accuracy": 0.1436249643564224,
      "num_tokens": 4231524.0,
      "step": 2435
    },
    {
      "entropy": 6.080678129196167,
      "epoch": 0.18984633339817156,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998586252785743,
      "loss": 5.9193,
      "mean_token_accuracy": 0.14638482481241227,
      "num_tokens": 4240229.0,
      "step": 2440
    },
    {
      "entropy": 6.062763833999634,
      "epoch": 0.19023536276988912,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998576412243816,
      "loss": 5.7908,
      "mean_token_accuracy": 0.14704818427562713,
      "num_tokens": 4248053.0,
      "step": 2445
    },
    {
      "entropy": 6.035331726074219,
      "epoch": 0.1906243921416067,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998566537583321,
      "loss": 5.8422,
      "mean_token_accuracy": 0.14500436931848526,
      "num_tokens": 4256019.0,
      "step": 2450
    },
    {
      "entropy": 5.968196296691895,
      "epoch": 0.19101342151332426,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998556628804408,
      "loss": 5.8189,
      "mean_token_accuracy": 0.1476341411471367,
      "num_tokens": 4263805.0,
      "step": 2455
    },
    {
      "entropy": 6.042609071731567,
      "epoch": 0.19140245088504182,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998546685907225,
      "loss": 5.8832,
      "mean_token_accuracy": 0.14107748866081238,
      "num_tokens": 4272712.0,
      "step": 2460
    },
    {
      "entropy": 6.127231884002685,
      "epoch": 0.19179148025675938,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998536708891927,
      "loss": 5.8106,
      "mean_token_accuracy": 0.14344534873962403,
      "num_tokens": 4281222.0,
      "step": 2465
    },
    {
      "entropy": 5.9839294910430905,
      "epoch": 0.19218050962847694,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998526697758663,
      "loss": 5.9239,
      "mean_token_accuracy": 0.13669644743204118,
      "num_tokens": 4290558.0,
      "step": 2470
    },
    {
      "entropy": 6.183317375183106,
      "epoch": 0.19256953900019452,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998516652507585,
      "loss": 5.9632,
      "mean_token_accuracy": 0.13987561762332917,
      "num_tokens": 4299684.0,
      "step": 2475
    },
    {
      "entropy": 6.019082832336426,
      "epoch": 0.19295856837191208,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998506573138846,
      "loss": 5.8976,
      "mean_token_accuracy": 0.14164622724056244,
      "num_tokens": 4308938.0,
      "step": 2480
    },
    {
      "entropy": 5.95548644065857,
      "epoch": 0.19334759774362964,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998496459652598,
      "loss": 5.7632,
      "mean_token_accuracy": 0.1458887979388237,
      "num_tokens": 4317646.0,
      "step": 2485
    },
    {
      "entropy": 5.9978094577789305,
      "epoch": 0.1937366271153472,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998486312048995,
      "loss": 5.9002,
      "mean_token_accuracy": 0.13895228877663612,
      "num_tokens": 4326876.0,
      "step": 2490
    },
    {
      "entropy": 6.007184743881226,
      "epoch": 0.19412565648706478,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998476130328193,
      "loss": 5.8647,
      "mean_token_accuracy": 0.14383746013045312,
      "num_tokens": 4335391.0,
      "step": 2495
    },
    {
      "entropy": 6.137946605682373,
      "epoch": 0.19451468585878234,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998465914490343,
      "loss": 5.8362,
      "mean_token_accuracy": 0.14361774623394014,
      "num_tokens": 4343253.0,
      "step": 2500
    },
    {
      "entropy": 5.859136152267456,
      "epoch": 0.1949037152304999,
      "grad_norm": 1.046875,
      "learning_rate": 0.00049984556645356,
      "loss": 5.8093,
      "mean_token_accuracy": 0.14489779770374298,
      "num_tokens": 4352161.0,
      "step": 2505
    },
    {
      "entropy": 6.0031920909881595,
      "epoch": 0.19529274460221746,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998445380464124,
      "loss": 5.7652,
      "mean_token_accuracy": 0.1499246060848236,
      "num_tokens": 4360414.0,
      "step": 2510
    },
    {
      "entropy": 6.0329090595245365,
      "epoch": 0.19568177397393502,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998435062276066,
      "loss": 5.7595,
      "mean_token_accuracy": 0.15170303583145142,
      "num_tokens": 4369375.0,
      "step": 2515
    },
    {
      "entropy": 5.8901123046875,
      "epoch": 0.1960708033456526,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998424709971586,
      "loss": 5.789,
      "mean_token_accuracy": 0.1448018729686737,
      "num_tokens": 4378638.0,
      "step": 2520
    },
    {
      "entropy": 6.047815895080566,
      "epoch": 0.19645983271737016,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998414323550839,
      "loss": 5.7997,
      "mean_token_accuracy": 0.1467505469918251,
      "num_tokens": 4388262.0,
      "step": 2525
    },
    {
      "entropy": 6.0011697769165036,
      "epoch": 0.19684886208908772,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998403903013984,
      "loss": 5.7862,
      "mean_token_accuracy": 0.14499384984374047,
      "num_tokens": 4396894.0,
      "step": 2530
    },
    {
      "entropy": 6.04869327545166,
      "epoch": 0.19723789146080528,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998393448361179,
      "loss": 5.9252,
      "mean_token_accuracy": 0.14205848649144173,
      "num_tokens": 4405455.0,
      "step": 2535
    },
    {
      "entropy": 6.017027759552002,
      "epoch": 0.19762692083252287,
      "grad_norm": 1.234375,
      "learning_rate": 0.000499838295959258,
      "loss": 5.8164,
      "mean_token_accuracy": 0.1469386860728264,
      "num_tokens": 4414355.0,
      "step": 2540
    },
    {
      "entropy": 5.904364156723022,
      "epoch": 0.19801595020424043,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499837243670835,
      "loss": 5.7141,
      "mean_token_accuracy": 0.1543835937976837,
      "num_tokens": 4422305.0,
      "step": 2545
    },
    {
      "entropy": 6.0292998313903805,
      "epoch": 0.19840497957595798,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998361879708646,
      "loss": 5.8479,
      "mean_token_accuracy": 0.1407075747847557,
      "num_tokens": 4430813.0,
      "step": 2550
    },
    {
      "entropy": 6.011082410812378,
      "epoch": 0.19879400894767554,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499835128859363,
      "loss": 5.8193,
      "mean_token_accuracy": 0.1463027387857437,
      "num_tokens": 4440229.0,
      "step": 2555
    },
    {
      "entropy": 5.949166488647461,
      "epoch": 0.1991830383193931,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998340663363461,
      "loss": 5.8318,
      "mean_token_accuracy": 0.14452947080135345,
      "num_tokens": 4449417.0,
      "step": 2560
    },
    {
      "entropy": 6.0368249893188475,
      "epoch": 0.1995720676911107,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998330004018301,
      "loss": 5.8755,
      "mean_token_accuracy": 0.14237613081932068,
      "num_tokens": 4457377.0,
      "step": 2565
    },
    {
      "entropy": 6.04552845954895,
      "epoch": 0.19996109706282825,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998319310558312,
      "loss": 5.793,
      "mean_token_accuracy": 0.1423659287393093,
      "num_tokens": 4466839.0,
      "step": 2570
    },
    {
      "entropy": 5.96616497039795,
      "epoch": 0.2003501264345458,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998308582983657,
      "loss": 5.8135,
      "mean_token_accuracy": 0.14745612293481827,
      "num_tokens": 4476369.0,
      "step": 2575
    },
    {
      "entropy": 5.971952486038208,
      "epoch": 0.20073915580626336,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998297821294497,
      "loss": 5.9056,
      "mean_token_accuracy": 0.14230666756629945,
      "num_tokens": 4485609.0,
      "step": 2580
    },
    {
      "entropy": 6.022813320159912,
      "epoch": 0.20112818517798095,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998287025490995,
      "loss": 5.8464,
      "mean_token_accuracy": 0.14304948300123216,
      "num_tokens": 4494535.0,
      "step": 2585
    },
    {
      "entropy": 6.004377222061157,
      "epoch": 0.2015172145496985,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998276195573317,
      "loss": 5.8093,
      "mean_token_accuracy": 0.14025293588638305,
      "num_tokens": 4503133.0,
      "step": 2590
    },
    {
      "entropy": 5.923559379577637,
      "epoch": 0.20190624392141607,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998265331541627,
      "loss": 5.7086,
      "mean_token_accuracy": 0.1545959234237671,
      "num_tokens": 4511465.0,
      "step": 2595
    },
    {
      "entropy": 5.973448705673218,
      "epoch": 0.20229527329313363,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998254433396088,
      "loss": 5.7744,
      "mean_token_accuracy": 0.1495731770992279,
      "num_tokens": 4520177.0,
      "step": 2600
    },
    {
      "entropy": 6.008311653137207,
      "epoch": 0.20268430266485118,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998243501136867,
      "loss": 5.934,
      "mean_token_accuracy": 0.14287928119301796,
      "num_tokens": 4528370.0,
      "step": 2605
    },
    {
      "entropy": 6.135716485977173,
      "epoch": 0.20307333203656877,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998232534764129,
      "loss": 5.9736,
      "mean_token_accuracy": 0.13521070554852485,
      "num_tokens": 4538258.0,
      "step": 2610
    },
    {
      "entropy": 6.009545373916626,
      "epoch": 0.20346236140828633,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998221534278041,
      "loss": 5.856,
      "mean_token_accuracy": 0.14092498347163201,
      "num_tokens": 4547300.0,
      "step": 2615
    },
    {
      "entropy": 6.087115001678467,
      "epoch": 0.2038513907800039,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004998210499678769,
      "loss": 5.9185,
      "mean_token_accuracy": 0.13737687543034555,
      "num_tokens": 4556877.0,
      "step": 2620
    },
    {
      "entropy": 5.95903000831604,
      "epoch": 0.20424042015172145,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998199430966481,
      "loss": 5.7669,
      "mean_token_accuracy": 0.15062909051775933,
      "num_tokens": 4565999.0,
      "step": 2625
    },
    {
      "entropy": 5.973028469085693,
      "epoch": 0.20462944952343903,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998188328141346,
      "loss": 5.8182,
      "mean_token_accuracy": 0.14192517548799516,
      "num_tokens": 4574027.0,
      "step": 2630
    },
    {
      "entropy": 5.941861963272094,
      "epoch": 0.2050184788951566,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998177191203531,
      "loss": 5.8217,
      "mean_token_accuracy": 0.1367108166217804,
      "num_tokens": 4582402.0,
      "step": 2635
    },
    {
      "entropy": 5.983818292617798,
      "epoch": 0.20540750826687415,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998166020153204,
      "loss": 5.7649,
      "mean_token_accuracy": 0.1477700263261795,
      "num_tokens": 4590381.0,
      "step": 2640
    },
    {
      "entropy": 6.012215900421142,
      "epoch": 0.2057965376385917,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998154814990538,
      "loss": 5.8301,
      "mean_token_accuracy": 0.14014950394630432,
      "num_tokens": 4598910.0,
      "step": 2645
    },
    {
      "entropy": 6.1034973621368405,
      "epoch": 0.20618556701030927,
      "grad_norm": 1.0625,
      "learning_rate": 0.00049981435757157,
      "loss": 5.9362,
      "mean_token_accuracy": 0.13985663801431655,
      "num_tokens": 4607400.0,
      "step": 2650
    },
    {
      "entropy": 5.96677680015564,
      "epoch": 0.20657459638202685,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004998132302328862,
      "loss": 5.8766,
      "mean_token_accuracy": 0.14142798110842705,
      "num_tokens": 4616510.0,
      "step": 2655
    },
    {
      "entropy": 6.054764366149902,
      "epoch": 0.2069636257537444,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998120994830195,
      "loss": 5.8317,
      "mean_token_accuracy": 0.14837341755628586,
      "num_tokens": 4624529.0,
      "step": 2660
    },
    {
      "entropy": 5.982214736938476,
      "epoch": 0.20735265512546197,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499810965321987,
      "loss": 5.7662,
      "mean_token_accuracy": 0.14475079774856567,
      "num_tokens": 4632141.0,
      "step": 2665
    },
    {
      "entropy": 6.031361770629883,
      "epoch": 0.20774168449717953,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998098277498059,
      "loss": 5.9592,
      "mean_token_accuracy": 0.14176981151103973,
      "num_tokens": 4640378.0,
      "step": 2670
    },
    {
      "entropy": 5.981580209732056,
      "epoch": 0.2081307138688971,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998086867664934,
      "loss": 5.735,
      "mean_token_accuracy": 0.1444416418671608,
      "num_tokens": 4649313.0,
      "step": 2675
    },
    {
      "entropy": 6.000965023040772,
      "epoch": 0.20851974324061467,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499807542372067,
      "loss": 5.8336,
      "mean_token_accuracy": 0.14503469914197922,
      "num_tokens": 4658272.0,
      "step": 2680
    },
    {
      "entropy": 5.966818475723267,
      "epoch": 0.20890877261233223,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998063945665439,
      "loss": 5.7057,
      "mean_token_accuracy": 0.1511964574456215,
      "num_tokens": 4665970.0,
      "step": 2685
    },
    {
      "entropy": 6.062405920028686,
      "epoch": 0.2092978019840498,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998052433499416,
      "loss": 5.9654,
      "mean_token_accuracy": 0.1380128502845764,
      "num_tokens": 4675308.0,
      "step": 2690
    },
    {
      "entropy": 5.959730434417724,
      "epoch": 0.20968683135576735,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998040887222776,
      "loss": 5.7913,
      "mean_token_accuracy": 0.14492155313491822,
      "num_tokens": 4684316.0,
      "step": 2695
    },
    {
      "entropy": 5.986378192901611,
      "epoch": 0.21007586072748494,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998029306835693,
      "loss": 5.7096,
      "mean_token_accuracy": 0.1470434159040451,
      "num_tokens": 4693029.0,
      "step": 2700
    },
    {
      "entropy": 5.935003757476807,
      "epoch": 0.2104648900992025,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998017692338344,
      "loss": 5.9259,
      "mean_token_accuracy": 0.1361936129629612,
      "num_tokens": 4702404.0,
      "step": 2705
    },
    {
      "entropy": 6.070126581192016,
      "epoch": 0.21085391947092005,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998006043730905,
      "loss": 5.8719,
      "mean_token_accuracy": 0.1365981914103031,
      "num_tokens": 4711126.0,
      "step": 2710
    },
    {
      "entropy": 5.990424919128418,
      "epoch": 0.2112429488426376,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997994361013551,
      "loss": 5.8809,
      "mean_token_accuracy": 0.14911534413695335,
      "num_tokens": 4719816.0,
      "step": 2715
    },
    {
      "entropy": 6.009741735458374,
      "epoch": 0.21163197821435517,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997982644186461,
      "loss": 5.8253,
      "mean_token_accuracy": 0.14253749400377275,
      "num_tokens": 4728294.0,
      "step": 2720
    },
    {
      "entropy": 6.006847667694092,
      "epoch": 0.21202100758607276,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997970893249813,
      "loss": 5.753,
      "mean_token_accuracy": 0.1469218373298645,
      "num_tokens": 4736708.0,
      "step": 2725
    },
    {
      "entropy": 5.951018142700195,
      "epoch": 0.21241003695779032,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997959108203785,
      "loss": 5.9443,
      "mean_token_accuracy": 0.13391065895557402,
      "num_tokens": 4744893.0,
      "step": 2730
    },
    {
      "entropy": 6.096749687194825,
      "epoch": 0.21279906632950787,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997947289048554,
      "loss": 5.8831,
      "mean_token_accuracy": 0.146332485973835,
      "num_tokens": 4753246.0,
      "step": 2735
    },
    {
      "entropy": 5.942298698425293,
      "epoch": 0.21318809570122543,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997935435784302,
      "loss": 5.8051,
      "mean_token_accuracy": 0.14182316064834594,
      "num_tokens": 4762229.0,
      "step": 2740
    },
    {
      "entropy": 5.863367128372192,
      "epoch": 0.21357712507294302,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997923548411208,
      "loss": 5.7539,
      "mean_token_accuracy": 0.14836178421974183,
      "num_tokens": 4771315.0,
      "step": 2745
    },
    {
      "entropy": 5.856457233428955,
      "epoch": 0.21396615444466058,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499791162692945,
      "loss": 5.6993,
      "mean_token_accuracy": 0.15656755715608597,
      "num_tokens": 4779848.0,
      "step": 2750
    },
    {
      "entropy": 6.063330602645874,
      "epoch": 0.21435518381637814,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997899671339214,
      "loss": 5.9217,
      "mean_token_accuracy": 0.13790412321686746,
      "num_tokens": 4788939.0,
      "step": 2755
    },
    {
      "entropy": 6.016645288467407,
      "epoch": 0.2147442131880957,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997887681640675,
      "loss": 5.819,
      "mean_token_accuracy": 0.146475438028574,
      "num_tokens": 4797719.0,
      "step": 2760
    },
    {
      "entropy": 6.004975509643555,
      "epoch": 0.21513324255981325,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997875657834021,
      "loss": 5.871,
      "mean_token_accuracy": 0.14507240653038025,
      "num_tokens": 4805601.0,
      "step": 2765
    },
    {
      "entropy": 6.079917144775391,
      "epoch": 0.21552227193153084,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997863599919432,
      "loss": 5.8329,
      "mean_token_accuracy": 0.13957922384142876,
      "num_tokens": 4814993.0,
      "step": 2770
    },
    {
      "entropy": 6.004630899429321,
      "epoch": 0.2159113013032484,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997851507897089,
      "loss": 5.7945,
      "mean_token_accuracy": 0.14447189047932624,
      "num_tokens": 4824156.0,
      "step": 2775
    },
    {
      "entropy": 5.940205669403076,
      "epoch": 0.21630033067496596,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997839381767178,
      "loss": 5.8904,
      "mean_token_accuracy": 0.14048577919602395,
      "num_tokens": 4833827.0,
      "step": 2780
    },
    {
      "entropy": 5.976140832901001,
      "epoch": 0.21668936004668352,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997827221529882,
      "loss": 5.8214,
      "mean_token_accuracy": 0.14300673305988312,
      "num_tokens": 4842801.0,
      "step": 2785
    },
    {
      "entropy": 5.935723829269409,
      "epoch": 0.2170783894184011,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997815027185386,
      "loss": 5.6989,
      "mean_token_accuracy": 0.14900791123509408,
      "num_tokens": 4850928.0,
      "step": 2790
    },
    {
      "entropy": 5.938062143325806,
      "epoch": 0.21746741879011866,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997802798733874,
      "loss": 5.8148,
      "mean_token_accuracy": 0.14658517614006997,
      "num_tokens": 4859510.0,
      "step": 2795
    },
    {
      "entropy": 6.017436552047729,
      "epoch": 0.21785644816183622,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004997790536175534,
      "loss": 5.8329,
      "mean_token_accuracy": 0.14880072325468063,
      "num_tokens": 4868201.0,
      "step": 2800
    },
    {
      "entropy": 5.924137544631958,
      "epoch": 0.21824547753355378,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997778239510548,
      "loss": 5.7554,
      "mean_token_accuracy": 0.1540842115879059,
      "num_tokens": 4876896.0,
      "step": 2805
    },
    {
      "entropy": 5.861286544799805,
      "epoch": 0.21863450690527134,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997765908739106,
      "loss": 5.7201,
      "mean_token_accuracy": 0.15085181072354317,
      "num_tokens": 4885136.0,
      "step": 2810
    },
    {
      "entropy": 5.949110889434815,
      "epoch": 0.21902353627698892,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997753543861395,
      "loss": 5.7256,
      "mean_token_accuracy": 0.15433648526668547,
      "num_tokens": 4893277.0,
      "step": 2815
    },
    {
      "entropy": 5.893832397460938,
      "epoch": 0.21941256564870648,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00049977411448776,
      "loss": 5.8811,
      "mean_token_accuracy": 0.14084682166576384,
      "num_tokens": 4901977.0,
      "step": 2820
    },
    {
      "entropy": 6.035731315612793,
      "epoch": 0.21980159502042404,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997728711787912,
      "loss": 5.8857,
      "mean_token_accuracy": 0.13519714698195456,
      "num_tokens": 4910888.0,
      "step": 2825
    },
    {
      "entropy": 5.9839459419250485,
      "epoch": 0.2201906243921416,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997716244592518,
      "loss": 5.873,
      "mean_token_accuracy": 0.1393221788108349,
      "num_tokens": 4920369.0,
      "step": 2830
    },
    {
      "entropy": 6.038845252990723,
      "epoch": 0.22057965376385918,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997703743291607,
      "loss": 5.852,
      "mean_token_accuracy": 0.1393017925322056,
      "num_tokens": 4928799.0,
      "step": 2835
    },
    {
      "entropy": 6.034655618667602,
      "epoch": 0.22096868313557674,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499769120788537,
      "loss": 5.7549,
      "mean_token_accuracy": 0.1501436188817024,
      "num_tokens": 4937194.0,
      "step": 2840
    },
    {
      "entropy": 5.931065654754638,
      "epoch": 0.2213577125072943,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997678638373995,
      "loss": 5.8327,
      "mean_token_accuracy": 0.14632293730974197,
      "num_tokens": 4946174.0,
      "step": 2845
    },
    {
      "entropy": 5.936599540710449,
      "epoch": 0.22174674187901186,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004997666034757676,
      "loss": 5.7925,
      "mean_token_accuracy": 0.13978436663746835,
      "num_tokens": 4955376.0,
      "step": 2850
    },
    {
      "entropy": 5.937894487380982,
      "epoch": 0.22213577125072942,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004997653397036603,
      "loss": 5.7927,
      "mean_token_accuracy": 0.15235631987452508,
      "num_tokens": 4964138.0,
      "step": 2855
    },
    {
      "entropy": 5.9808601379394535,
      "epoch": 0.222524800622447,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997640725210965,
      "loss": 5.8731,
      "mean_token_accuracy": 0.14401781558990479,
      "num_tokens": 4973868.0,
      "step": 2860
    },
    {
      "entropy": 5.941016101837159,
      "epoch": 0.22291382999416456,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997628019280958,
      "loss": 5.7673,
      "mean_token_accuracy": 0.1464352160692215,
      "num_tokens": 4982687.0,
      "step": 2865
    },
    {
      "entropy": 5.9928521633148195,
      "epoch": 0.22330285936588212,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997615279246773,
      "loss": 5.8145,
      "mean_token_accuracy": 0.1418926164507866,
      "num_tokens": 4992072.0,
      "step": 2870
    },
    {
      "entropy": 5.967259168624878,
      "epoch": 0.22369188873759968,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997602505108603,
      "loss": 5.7397,
      "mean_token_accuracy": 0.15427607148885727,
      "num_tokens": 5000551.0,
      "step": 2875
    },
    {
      "entropy": 5.983960723876953,
      "epoch": 0.22408091810931724,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997589696866644,
      "loss": 5.9152,
      "mean_token_accuracy": 0.14138678908348085,
      "num_tokens": 5009292.0,
      "step": 2880
    },
    {
      "entropy": 5.977126598358154,
      "epoch": 0.22446994748103483,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997576854521088,
      "loss": 5.7549,
      "mean_token_accuracy": 0.1445073001086712,
      "num_tokens": 5018407.0,
      "step": 2885
    },
    {
      "entropy": 5.952081108093262,
      "epoch": 0.22485897685275238,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499756397807213,
      "loss": 5.8266,
      "mean_token_accuracy": 0.1417175605893135,
      "num_tokens": 5027353.0,
      "step": 2890
    },
    {
      "entropy": 6.0577531337738035,
      "epoch": 0.22524800622446994,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997551067519966,
      "loss": 5.8279,
      "mean_token_accuracy": 0.14191178530454635,
      "num_tokens": 5036476.0,
      "step": 2895
    },
    {
      "entropy": 5.93521409034729,
      "epoch": 0.2256370355961875,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997538122864792,
      "loss": 5.7732,
      "mean_token_accuracy": 0.1461431086063385,
      "num_tokens": 5045132.0,
      "step": 2900
    },
    {
      "entropy": 5.991693019866943,
      "epoch": 0.2260260649679051,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997525144106804,
      "loss": 5.7998,
      "mean_token_accuracy": 0.14501917809247972,
      "num_tokens": 5053665.0,
      "step": 2905
    },
    {
      "entropy": 5.985103130340576,
      "epoch": 0.22641509433962265,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00049975121312462,
      "loss": 5.7395,
      "mean_token_accuracy": 0.1550862744450569,
      "num_tokens": 5062516.0,
      "step": 2910
    },
    {
      "entropy": 5.932449817657471,
      "epoch": 0.2268041237113402,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997499084283177,
      "loss": 5.7869,
      "mean_token_accuracy": 0.1438561961054802,
      "num_tokens": 5071260.0,
      "step": 2915
    },
    {
      "entropy": 5.883814859390259,
      "epoch": 0.22719315308305776,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997486003217932,
      "loss": 5.7513,
      "mean_token_accuracy": 0.15051764249801636,
      "num_tokens": 5079215.0,
      "step": 2920
    },
    {
      "entropy": 5.976101589202881,
      "epoch": 0.22758218245477532,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997472888050664,
      "loss": 5.8134,
      "mean_token_accuracy": 0.14436266273260118,
      "num_tokens": 5088404.0,
      "step": 2925
    },
    {
      "entropy": 5.9055495262146,
      "epoch": 0.2279712118264929,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997459738781573,
      "loss": 5.7343,
      "mean_token_accuracy": 0.1516444243490696,
      "num_tokens": 5096756.0,
      "step": 2930
    },
    {
      "entropy": 5.991645336151123,
      "epoch": 0.22836024119821047,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004997446555410857,
      "loss": 5.91,
      "mean_token_accuracy": 0.14260229244828224,
      "num_tokens": 5106160.0,
      "step": 2935
    },
    {
      "entropy": 5.823283624649048,
      "epoch": 0.22874927056992803,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997433337938716,
      "loss": 5.8332,
      "mean_token_accuracy": 0.14509238600730895,
      "num_tokens": 5114915.0,
      "step": 2940
    },
    {
      "entropy": 6.0273322582244875,
      "epoch": 0.22913829994164558,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997420086365351,
      "loss": 5.7454,
      "mean_token_accuracy": 0.1477459713816643,
      "num_tokens": 5123249.0,
      "step": 2945
    },
    {
      "entropy": 5.918133068084717,
      "epoch": 0.22952732931336317,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997406800690966,
      "loss": 5.7409,
      "mean_token_accuracy": 0.1458730474114418,
      "num_tokens": 5131906.0,
      "step": 2950
    },
    {
      "entropy": 5.9432145118713375,
      "epoch": 0.22991635868508073,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997393480915758,
      "loss": 5.7206,
      "mean_token_accuracy": 0.1492404267191887,
      "num_tokens": 5140740.0,
      "step": 2955
    },
    {
      "entropy": 6.023179388046264,
      "epoch": 0.2303053880567983,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997380127039931,
      "loss": 5.9131,
      "mean_token_accuracy": 0.1447205662727356,
      "num_tokens": 5149748.0,
      "step": 2960
    },
    {
      "entropy": 5.831229782104492,
      "epoch": 0.23069441742851585,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997366739063688,
      "loss": 5.6358,
      "mean_token_accuracy": 0.1557387501001358,
      "num_tokens": 5158475.0,
      "step": 2965
    },
    {
      "entropy": 5.921968793869018,
      "epoch": 0.2310834468002334,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997353316987231,
      "loss": 5.8501,
      "mean_token_accuracy": 0.13665784075856208,
      "num_tokens": 5166593.0,
      "step": 2970
    },
    {
      "entropy": 6.01880030632019,
      "epoch": 0.231472476171951,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997339860810765,
      "loss": 5.8048,
      "mean_token_accuracy": 0.1443267844617367,
      "num_tokens": 5175557.0,
      "step": 2975
    },
    {
      "entropy": 5.942876625061035,
      "epoch": 0.23186150554366855,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997326370534495,
      "loss": 5.7114,
      "mean_token_accuracy": 0.1499389424920082,
      "num_tokens": 5184140.0,
      "step": 2980
    },
    {
      "entropy": 5.920814514160156,
      "epoch": 0.2322505349153861,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997312846158622,
      "loss": 5.7985,
      "mean_token_accuracy": 0.14361362606287004,
      "num_tokens": 5192347.0,
      "step": 2985
    },
    {
      "entropy": 5.987712621688843,
      "epoch": 0.23263956428710367,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997299287683355,
      "loss": 5.7626,
      "mean_token_accuracy": 0.1525233119726181,
      "num_tokens": 5201327.0,
      "step": 2990
    },
    {
      "entropy": 5.954172992706299,
      "epoch": 0.23302859365882125,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997285695108898,
      "loss": 5.8512,
      "mean_token_accuracy": 0.14315994083881378,
      "num_tokens": 5209873.0,
      "step": 2995
    },
    {
      "entropy": 5.992607831954956,
      "epoch": 0.2334176230305388,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997272068435458,
      "loss": 5.7687,
      "mean_token_accuracy": 0.148164002597332,
      "num_tokens": 5219176.0,
      "step": 3000
    },
    {
      "epoch": 0.2334176230305388,
      "eval_entropy": 5.708837842803603,
      "eval_loss": 5.805070877075195,
      "eval_mean_token_accuracy": 0.15528924172677824,
      "eval_num_tokens": 5219176.0,
      "eval_runtime": 21.7192,
      "eval_samples_per_second": 1913.423,
      "eval_steps_per_second": 239.189,
      "step": 3000
    },
    {
      "entropy": 5.8470841407775875,
      "epoch": 0.23380665240225637,
      "grad_norm": 1.125,
      "learning_rate": 0.000499725840766324,
      "loss": 5.6901,
      "mean_token_accuracy": 0.16298863142728806,
      "num_tokens": 5227310.0,
      "step": 3005
    },
    {
      "entropy": 5.977334976196289,
      "epoch": 0.23419568177397393,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997244712792453,
      "loss": 5.7837,
      "mean_token_accuracy": 0.1502714529633522,
      "num_tokens": 5235882.0,
      "step": 3010
    },
    {
      "entropy": 5.930918598175049,
      "epoch": 0.2345847111456915,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997230983823305,
      "loss": 5.7128,
      "mean_token_accuracy": 0.14631212800741195,
      "num_tokens": 5244149.0,
      "step": 3015
    },
    {
      "entropy": 5.768956327438355,
      "epoch": 0.23497374051740907,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997217220756003,
      "loss": 5.5696,
      "mean_token_accuracy": 0.15455945283174516,
      "num_tokens": 5252473.0,
      "step": 3020
    },
    {
      "entropy": 5.900669765472412,
      "epoch": 0.23536276988912663,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997203423590757,
      "loss": 5.7218,
      "mean_token_accuracy": 0.15821771323680878,
      "num_tokens": 5260972.0,
      "step": 3025
    },
    {
      "entropy": 5.909393644332885,
      "epoch": 0.2357517992608442,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997189592327775,
      "loss": 5.7513,
      "mean_token_accuracy": 0.14254340678453445,
      "num_tokens": 5270513.0,
      "step": 3030
    },
    {
      "entropy": 5.9588981628417965,
      "epoch": 0.23614082863256175,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997175726967268,
      "loss": 5.7111,
      "mean_token_accuracy": 0.1438601642847061,
      "num_tokens": 5278433.0,
      "step": 3035
    },
    {
      "entropy": 5.870810604095459,
      "epoch": 0.23652985800427934,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997161827509447,
      "loss": 5.7325,
      "mean_token_accuracy": 0.14383522644639016,
      "num_tokens": 5287389.0,
      "step": 3040
    },
    {
      "entropy": 5.874240207672119,
      "epoch": 0.2369188873759969,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004997147893954521,
      "loss": 5.8365,
      "mean_token_accuracy": 0.14417727142572404,
      "num_tokens": 5296220.0,
      "step": 3045
    },
    {
      "entropy": 5.968716287612915,
      "epoch": 0.23730791674771445,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997133926302702,
      "loss": 5.7495,
      "mean_token_accuracy": 0.15066867470741271,
      "num_tokens": 5304152.0,
      "step": 3050
    },
    {
      "entropy": 5.972426891326904,
      "epoch": 0.237696946119432,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997119924554204,
      "loss": 5.7685,
      "mean_token_accuracy": 0.14622094333171845,
      "num_tokens": 5312374.0,
      "step": 3055
    },
    {
      "entropy": 5.836151027679444,
      "epoch": 0.23808597549114957,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004997105888709236,
      "loss": 5.7113,
      "mean_token_accuracy": 0.14792220890522004,
      "num_tokens": 5320696.0,
      "step": 3060
    },
    {
      "entropy": 5.961451530456543,
      "epoch": 0.23847500486286716,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997091818768015,
      "loss": 5.8176,
      "mean_token_accuracy": 0.1443133182823658,
      "num_tokens": 5328754.0,
      "step": 3065
    },
    {
      "entropy": 5.941938495635986,
      "epoch": 0.23886403423458472,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499707771473075,
      "loss": 5.8091,
      "mean_token_accuracy": 0.14472589865326882,
      "num_tokens": 5337553.0,
      "step": 3070
    },
    {
      "entropy": 5.9362963199615475,
      "epoch": 0.23925306360630227,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997063576597659,
      "loss": 5.7188,
      "mean_token_accuracy": 0.1526940643787384,
      "num_tokens": 5346063.0,
      "step": 3075
    },
    {
      "entropy": 5.906724262237549,
      "epoch": 0.23964209297801983,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997049404368954,
      "loss": 5.7852,
      "mean_token_accuracy": 0.1506700687110424,
      "num_tokens": 5354187.0,
      "step": 3080
    },
    {
      "entropy": 5.951892757415772,
      "epoch": 0.2400311223497374,
      "grad_norm": 1.1875,
      "learning_rate": 0.000499703519804485,
      "loss": 5.8262,
      "mean_token_accuracy": 0.14107102453708648,
      "num_tokens": 5362819.0,
      "step": 3085
    },
    {
      "entropy": 5.90645399093628,
      "epoch": 0.24042015172145498,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997020957625564,
      "loss": 5.6952,
      "mean_token_accuracy": 0.15633721873164177,
      "num_tokens": 5371249.0,
      "step": 3090
    },
    {
      "entropy": 5.86179895401001,
      "epoch": 0.24080918109317254,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997006683111312,
      "loss": 5.7164,
      "mean_token_accuracy": 0.15006546974182128,
      "num_tokens": 5379284.0,
      "step": 3095
    },
    {
      "entropy": 5.949423742294312,
      "epoch": 0.2411982104648901,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499699237450231,
      "loss": 5.8026,
      "mean_token_accuracy": 0.14799220338463784,
      "num_tokens": 5388541.0,
      "step": 3100
    },
    {
      "entropy": 5.940669870376587,
      "epoch": 0.24158723983660765,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996978031798774,
      "loss": 5.8654,
      "mean_token_accuracy": 0.14638842642307281,
      "num_tokens": 5397226.0,
      "step": 3105
    },
    {
      "entropy": 5.933017158508301,
      "epoch": 0.24197626920832524,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996963655000924,
      "loss": 5.7814,
      "mean_token_accuracy": 0.1445136360824108,
      "num_tokens": 5405579.0,
      "step": 3110
    },
    {
      "entropy": 5.975747013092041,
      "epoch": 0.2423652985800428,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004996949244108977,
      "loss": 5.8878,
      "mean_token_accuracy": 0.1370643675327301,
      "num_tokens": 5414529.0,
      "step": 3115
    },
    {
      "entropy": 6.017865705490112,
      "epoch": 0.24275432795176036,
      "grad_norm": 1.0,
      "learning_rate": 0.000499693479912315,
      "loss": 5.8519,
      "mean_token_accuracy": 0.15119654461741447,
      "num_tokens": 5423641.0,
      "step": 3120
    },
    {
      "entropy": 5.889483213424683,
      "epoch": 0.24314335732347792,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996920320043666,
      "loss": 5.7358,
      "mean_token_accuracy": 0.15109173879027366,
      "num_tokens": 5432558.0,
      "step": 3125
    },
    {
      "entropy": 5.896158361434937,
      "epoch": 0.24353238669519547,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996905806870742,
      "loss": 5.692,
      "mean_token_accuracy": 0.15060517191886902,
      "num_tokens": 5441449.0,
      "step": 3130
    },
    {
      "entropy": 5.880832290649414,
      "epoch": 0.24392141606691306,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004996891259604598,
      "loss": 5.7285,
      "mean_token_accuracy": 0.14988522976636887,
      "num_tokens": 5451142.0,
      "step": 3135
    },
    {
      "entropy": 5.923013973236084,
      "epoch": 0.24431044543863062,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004996876678245455,
      "loss": 5.6773,
      "mean_token_accuracy": 0.14793565645813941,
      "num_tokens": 5459290.0,
      "step": 3140
    },
    {
      "entropy": 5.868299388885498,
      "epoch": 0.24469947481034818,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996862062793536,
      "loss": 5.6931,
      "mean_token_accuracy": 0.1527027264237404,
      "num_tokens": 5467776.0,
      "step": 3145
    },
    {
      "entropy": 5.7914379119873045,
      "epoch": 0.24508850418206574,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996847413249061,
      "loss": 5.5762,
      "mean_token_accuracy": 0.15867648273706436,
      "num_tokens": 5476005.0,
      "step": 3150
    },
    {
      "entropy": 5.935610818862915,
      "epoch": 0.24547753355378332,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996832729612252,
      "loss": 5.7653,
      "mean_token_accuracy": 0.14748554229736327,
      "num_tokens": 5484225.0,
      "step": 3155
    },
    {
      "entropy": 5.774376010894775,
      "epoch": 0.24586656292550088,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996818011883333,
      "loss": 5.6793,
      "mean_token_accuracy": 0.15868545621633529,
      "num_tokens": 5493111.0,
      "step": 3160
    },
    {
      "entropy": 5.852526235580444,
      "epoch": 0.24625559229721844,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996803260062527,
      "loss": 5.7066,
      "mean_token_accuracy": 0.15420344173908235,
      "num_tokens": 5501574.0,
      "step": 3165
    },
    {
      "entropy": 5.928380393981934,
      "epoch": 0.246644621668936,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996788474150057,
      "loss": 5.7265,
      "mean_token_accuracy": 0.14918727576732635,
      "num_tokens": 5511136.0,
      "step": 3170
    },
    {
      "entropy": 5.81124529838562,
      "epoch": 0.24703365104065356,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996773654146149,
      "loss": 5.7172,
      "mean_token_accuracy": 0.15357716828584672,
      "num_tokens": 5520324.0,
      "step": 3175
    },
    {
      "entropy": 5.940634679794312,
      "epoch": 0.24742268041237114,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996758800051026,
      "loss": 5.6802,
      "mean_token_accuracy": 0.1523045241832733,
      "num_tokens": 5528960.0,
      "step": 3180
    },
    {
      "entropy": 5.899743127822876,
      "epoch": 0.2478117097840887,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004996743911864916,
      "loss": 5.8265,
      "mean_token_accuracy": 0.1465790882706642,
      "num_tokens": 5537391.0,
      "step": 3185
    },
    {
      "entropy": 5.9081946849823,
      "epoch": 0.24820073915580626,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996728989588041,
      "loss": 5.7762,
      "mean_token_accuracy": 0.14774141162633897,
      "num_tokens": 5546577.0,
      "step": 3190
    },
    {
      "entropy": 6.005410671234131,
      "epoch": 0.24858976852752382,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996714033220632,
      "loss": 5.6988,
      "mean_token_accuracy": 0.14901615232229232,
      "num_tokens": 5555247.0,
      "step": 3195
    },
    {
      "entropy": 5.847511529922485,
      "epoch": 0.2489787978992414,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996699042762911,
      "loss": 5.6683,
      "mean_token_accuracy": 0.15640627592802048,
      "num_tokens": 5563598.0,
      "step": 3200
    },
    {
      "entropy": 5.894534111022949,
      "epoch": 0.24936782727095896,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996684018215111,
      "loss": 5.6906,
      "mean_token_accuracy": 0.14715228825807572,
      "num_tokens": 5572734.0,
      "step": 3205
    },
    {
      "entropy": 5.902574157714843,
      "epoch": 0.24975685664267652,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996668959577453,
      "loss": 5.8265,
      "mean_token_accuracy": 0.1428134858608246,
      "num_tokens": 5581224.0,
      "step": 3210
    },
    {
      "entropy": 5.957176971435547,
      "epoch": 0.2501458860143941,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996653866850173,
      "loss": 5.8056,
      "mean_token_accuracy": 0.14543875753879548,
      "num_tokens": 5590138.0,
      "step": 3215
    },
    {
      "entropy": 5.899808120727539,
      "epoch": 0.25053491538611167,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004996638740033495,
      "loss": 5.7449,
      "mean_token_accuracy": 0.15667717456817626,
      "num_tokens": 5598654.0,
      "step": 3220
    },
    {
      "entropy": 5.809116697311401,
      "epoch": 0.2509239447578292,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996623579127651,
      "loss": 5.6802,
      "mean_token_accuracy": 0.1541752725839615,
      "num_tokens": 5607404.0,
      "step": 3225
    },
    {
      "entropy": 5.900450563430786,
      "epoch": 0.2513129741295468,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996608384132868,
      "loss": 5.7177,
      "mean_token_accuracy": 0.14940476417541504,
      "num_tokens": 5617133.0,
      "step": 3230
    },
    {
      "entropy": 5.859549045562744,
      "epoch": 0.25170200350126437,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499659315504938,
      "loss": 5.7358,
      "mean_token_accuracy": 0.15141356587409974,
      "num_tokens": 5625389.0,
      "step": 3235
    },
    {
      "entropy": 5.836001014709472,
      "epoch": 0.2520910328729819,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996577891877417,
      "loss": 5.615,
      "mean_token_accuracy": 0.1595832273364067,
      "num_tokens": 5633985.0,
      "step": 3240
    },
    {
      "entropy": 5.941919231414795,
      "epoch": 0.2524800622446995,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996562594617209,
      "loss": 5.7245,
      "mean_token_accuracy": 0.147165247797966,
      "num_tokens": 5642524.0,
      "step": 3245
    },
    {
      "entropy": 5.855136919021606,
      "epoch": 0.252869091616417,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996547263268991,
      "loss": 5.6914,
      "mean_token_accuracy": 0.15046496391296388,
      "num_tokens": 5651788.0,
      "step": 3250
    },
    {
      "entropy": 5.824191331863403,
      "epoch": 0.2532581209881346,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996531897832993,
      "loss": 5.7323,
      "mean_token_accuracy": 0.14551326930522918,
      "num_tokens": 5660757.0,
      "step": 3255
    },
    {
      "entropy": 5.995775270462036,
      "epoch": 0.2536471503598522,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499651649830945,
      "loss": 5.8321,
      "mean_token_accuracy": 0.15186033323407172,
      "num_tokens": 5669760.0,
      "step": 3260
    },
    {
      "entropy": 5.861927843093872,
      "epoch": 0.2540361797315697,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996501064698593,
      "loss": 5.7555,
      "mean_token_accuracy": 0.15121082663536073,
      "num_tokens": 5678313.0,
      "step": 3265
    },
    {
      "entropy": 5.893115282058716,
      "epoch": 0.2544252091032873,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499648559700066,
      "loss": 5.6689,
      "mean_token_accuracy": 0.15300092101097107,
      "num_tokens": 5686942.0,
      "step": 3270
    },
    {
      "entropy": 5.862796831130981,
      "epoch": 0.25481423847500484,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996470095215884,
      "loss": 5.6618,
      "mean_token_accuracy": 0.15529074147343636,
      "num_tokens": 5695612.0,
      "step": 3275
    },
    {
      "entropy": 5.916653394699097,
      "epoch": 0.2552032678467224,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996454559344498,
      "loss": 5.836,
      "mean_token_accuracy": 0.14220810383558274,
      "num_tokens": 5705067.0,
      "step": 3280
    },
    {
      "entropy": 5.9034261226654055,
      "epoch": 0.25559229721844,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996438989386741,
      "loss": 5.757,
      "mean_token_accuracy": 0.1522183448076248,
      "num_tokens": 5714092.0,
      "step": 3285
    },
    {
      "entropy": 5.900182867050171,
      "epoch": 0.25598132659015754,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996423385342847,
      "loss": 5.7082,
      "mean_token_accuracy": 0.15092059820890427,
      "num_tokens": 5722638.0,
      "step": 3290
    },
    {
      "entropy": 5.842333030700684,
      "epoch": 0.25637035596187513,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996407747213055,
      "loss": 5.7133,
      "mean_token_accuracy": 0.14781118407845498,
      "num_tokens": 5731511.0,
      "step": 3295
    },
    {
      "entropy": 5.86269474029541,
      "epoch": 0.25675938533359266,
      "grad_norm": 0.9375,
      "learning_rate": 0.00049963920749976,
      "loss": 5.7235,
      "mean_token_accuracy": 0.15076699554920198,
      "num_tokens": 5740950.0,
      "step": 3300
    },
    {
      "entropy": 5.866805219650269,
      "epoch": 0.25714841470531025,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996376368696721,
      "loss": 5.6751,
      "mean_token_accuracy": 0.15006920993328093,
      "num_tokens": 5749163.0,
      "step": 3305
    },
    {
      "entropy": 5.9052917003631595,
      "epoch": 0.25753744407702783,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996360628310656,
      "loss": 5.7216,
      "mean_token_accuracy": 0.15033655017614364,
      "num_tokens": 5758321.0,
      "step": 3310
    },
    {
      "entropy": 5.907117176055908,
      "epoch": 0.25792647344874536,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996344853839644,
      "loss": 5.7838,
      "mean_token_accuracy": 0.1547563076019287,
      "num_tokens": 5767277.0,
      "step": 3315
    },
    {
      "entropy": 5.861928462982178,
      "epoch": 0.25831550282046295,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996329045283924,
      "loss": 5.6757,
      "mean_token_accuracy": 0.15904149860143663,
      "num_tokens": 5775694.0,
      "step": 3320
    },
    {
      "entropy": 5.760580205917359,
      "epoch": 0.25870453219218054,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004996313202643737,
      "loss": 5.587,
      "mean_token_accuracy": 0.15962285846471785,
      "num_tokens": 5783932.0,
      "step": 3325
    },
    {
      "entropy": 5.855116939544677,
      "epoch": 0.25909356156389807,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996297325919321,
      "loss": 5.7234,
      "mean_token_accuracy": 0.15485287457704544,
      "num_tokens": 5792100.0,
      "step": 3330
    },
    {
      "entropy": 5.87807788848877,
      "epoch": 0.25948259093561565,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996281415110919,
      "loss": 5.7065,
      "mean_token_accuracy": 0.1520189329981804,
      "num_tokens": 5801481.0,
      "step": 3335
    },
    {
      "entropy": 5.915521335601807,
      "epoch": 0.2598716203073332,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996265470218773,
      "loss": 5.7913,
      "mean_token_accuracy": 0.14595863968133926,
      "num_tokens": 5810053.0,
      "step": 3340
    },
    {
      "entropy": 5.808931303024292,
      "epoch": 0.26026064967905077,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996249491243122,
      "loss": 5.5755,
      "mean_token_accuracy": 0.15734014958143233,
      "num_tokens": 5818454.0,
      "step": 3345
    },
    {
      "entropy": 5.849155044555664,
      "epoch": 0.26064967905076836,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996233478184211,
      "loss": 5.7603,
      "mean_token_accuracy": 0.14564188867807387,
      "num_tokens": 5827013.0,
      "step": 3350
    },
    {
      "entropy": 5.797790098190307,
      "epoch": 0.2610387084224859,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996217431042282,
      "loss": 5.7568,
      "mean_token_accuracy": 0.1458510808646679,
      "num_tokens": 5835862.0,
      "step": 3355
    },
    {
      "entropy": 5.792097759246826,
      "epoch": 0.2614277377942035,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499620134981758,
      "loss": 5.53,
      "mean_token_accuracy": 0.15552110224962234,
      "num_tokens": 5843827.0,
      "step": 3360
    },
    {
      "entropy": 5.825706052780151,
      "epoch": 0.261816767165921,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996185234510346,
      "loss": 5.7161,
      "mean_token_accuracy": 0.1480458125472069,
      "num_tokens": 5852677.0,
      "step": 3365
    },
    {
      "entropy": 5.8807507991790775,
      "epoch": 0.2622057965376386,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996169085120828,
      "loss": 5.6675,
      "mean_token_accuracy": 0.15776466727256774,
      "num_tokens": 5861686.0,
      "step": 3370
    },
    {
      "entropy": 5.7480799674987795,
      "epoch": 0.2625948259093562,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996152901649268,
      "loss": 5.6499,
      "mean_token_accuracy": 0.14646029621362686,
      "num_tokens": 5870690.0,
      "step": 3375
    },
    {
      "entropy": 5.8425665378570555,
      "epoch": 0.2629838552810737,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996136684095913,
      "loss": 5.7231,
      "mean_token_accuracy": 0.15138084143400193,
      "num_tokens": 5879804.0,
      "step": 3380
    },
    {
      "entropy": 5.917961263656617,
      "epoch": 0.2633728846527913,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996120432461009,
      "loss": 5.8153,
      "mean_token_accuracy": 0.14202643558382988,
      "num_tokens": 5888781.0,
      "step": 3385
    },
    {
      "entropy": 5.930140781402588,
      "epoch": 0.2637619140245088,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004996104146744804,
      "loss": 5.7168,
      "mean_token_accuracy": 0.1495570257306099,
      "num_tokens": 5897439.0,
      "step": 3390
    },
    {
      "entropy": 5.8434840679168705,
      "epoch": 0.2641509433962264,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996087826947541,
      "loss": 5.6682,
      "mean_token_accuracy": 0.1641407147049904,
      "num_tokens": 5905471.0,
      "step": 3395
    },
    {
      "entropy": 5.766514444351197,
      "epoch": 0.264539972767944,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996071473069471,
      "loss": 5.7031,
      "mean_token_accuracy": 0.15298082679510117,
      "num_tokens": 5913795.0,
      "step": 3400
    },
    {
      "entropy": 5.9807665824890135,
      "epoch": 0.26492900213966153,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996055085110842,
      "loss": 5.7697,
      "mean_token_accuracy": 0.14786940217018127,
      "num_tokens": 5922119.0,
      "step": 3405
    },
    {
      "entropy": 5.797083187103271,
      "epoch": 0.2653180315113791,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996038663071902,
      "loss": 5.6452,
      "mean_token_accuracy": 0.15326521545648575,
      "num_tokens": 5930759.0,
      "step": 3410
    },
    {
      "entropy": 5.855861043930053,
      "epoch": 0.26570706088309665,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996022206952901,
      "loss": 5.6921,
      "mean_token_accuracy": 0.15083688646554946,
      "num_tokens": 5939830.0,
      "step": 3415
    },
    {
      "entropy": 5.9113367080688475,
      "epoch": 0.26609609025481423,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996005716754086,
      "loss": 5.6936,
      "mean_token_accuracy": 0.15180784612894058,
      "num_tokens": 5947737.0,
      "step": 3420
    },
    {
      "entropy": 5.746157121658325,
      "epoch": 0.2664851196265318,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499598919247571,
      "loss": 5.6511,
      "mean_token_accuracy": 0.15976583808660508,
      "num_tokens": 5955905.0,
      "step": 3425
    },
    {
      "entropy": 5.814898061752319,
      "epoch": 0.26687414899824935,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995972634118023,
      "loss": 5.703,
      "mean_token_accuracy": 0.15062872618436812,
      "num_tokens": 5965223.0,
      "step": 3430
    },
    {
      "entropy": 5.902707624435425,
      "epoch": 0.26726317836996694,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995956041681277,
      "loss": 5.7254,
      "mean_token_accuracy": 0.15757619440555573,
      "num_tokens": 5973615.0,
      "step": 3435
    },
    {
      "entropy": 5.814543104171753,
      "epoch": 0.2676522077416845,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995939415165722,
      "loss": 5.6472,
      "mean_token_accuracy": 0.15822276771068572,
      "num_tokens": 5981600.0,
      "step": 3440
    },
    {
      "entropy": 5.832520580291748,
      "epoch": 0.26804123711340205,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995922754571611,
      "loss": 5.6538,
      "mean_token_accuracy": 0.15374368280172349,
      "num_tokens": 5989928.0,
      "step": 3445
    },
    {
      "entropy": 5.787254190444946,
      "epoch": 0.26843026648511964,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995906059899197,
      "loss": 5.7136,
      "mean_token_accuracy": 0.15472160428762435,
      "num_tokens": 5998704.0,
      "step": 3450
    },
    {
      "entropy": 5.824487972259521,
      "epoch": 0.26881929585683717,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004995889331148733,
      "loss": 5.6914,
      "mean_token_accuracy": 0.15172771811485292,
      "num_tokens": 6007476.0,
      "step": 3455
    },
    {
      "entropy": 5.947670650482178,
      "epoch": 0.26920832522855476,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004995872568320474,
      "loss": 5.7556,
      "mean_token_accuracy": 0.15055454671382903,
      "num_tokens": 6016324.0,
      "step": 3460
    },
    {
      "entropy": 5.85786542892456,
      "epoch": 0.26959735460027234,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995855771414673,
      "loss": 5.7097,
      "mean_token_accuracy": 0.1530899092555046,
      "num_tokens": 6024709.0,
      "step": 3465
    },
    {
      "entropy": 5.800860786437989,
      "epoch": 0.2699863839719899,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995838940431585,
      "loss": 5.5442,
      "mean_token_accuracy": 0.1590125173330307,
      "num_tokens": 6033131.0,
      "step": 3470
    },
    {
      "entropy": 5.785799980163574,
      "epoch": 0.27037541334370746,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995822075371466,
      "loss": 5.7525,
      "mean_token_accuracy": 0.15026821494102477,
      "num_tokens": 6042310.0,
      "step": 3475
    },
    {
      "entropy": 5.797628450393677,
      "epoch": 0.270764442715425,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004995805176234571,
      "loss": 5.5729,
      "mean_token_accuracy": 0.16002072095870973,
      "num_tokens": 6050617.0,
      "step": 3480
    },
    {
      "entropy": 5.856146669387817,
      "epoch": 0.2711534720871426,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004995788243021158,
      "loss": 5.7454,
      "mean_token_accuracy": 0.14996809065341948,
      "num_tokens": 6060090.0,
      "step": 3485
    },
    {
      "entropy": 5.894637155532837,
      "epoch": 0.27154250145886016,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004995771275731483,
      "loss": 5.6853,
      "mean_token_accuracy": 0.1579849451780319,
      "num_tokens": 6068646.0,
      "step": 3490
    },
    {
      "entropy": 5.842238807678223,
      "epoch": 0.2719315308305777,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995754274365802,
      "loss": 5.7378,
      "mean_token_accuracy": 0.15850494801998138,
      "num_tokens": 6077896.0,
      "step": 3495
    },
    {
      "entropy": 5.848576164245605,
      "epoch": 0.2723205602022953,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004995737238924376,
      "loss": 5.6541,
      "mean_token_accuracy": 0.15065545886754989,
      "num_tokens": 6087189.0,
      "step": 3500
    },
    {
      "entropy": 5.827096271514892,
      "epoch": 0.2727095895740128,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995720169407461,
      "loss": 5.6674,
      "mean_token_accuracy": 0.15255527347326278,
      "num_tokens": 6095394.0,
      "step": 3505
    },
    {
      "entropy": 5.859431648254395,
      "epoch": 0.2730986189457304,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995703065815317,
      "loss": 5.6283,
      "mean_token_accuracy": 0.15823251754045486,
      "num_tokens": 6104135.0,
      "step": 3510
    },
    {
      "entropy": 5.830027008056641,
      "epoch": 0.273487648317448,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995685928148203,
      "loss": 5.7462,
      "mean_token_accuracy": 0.14713549017906188,
      "num_tokens": 6112160.0,
      "step": 3515
    },
    {
      "entropy": 5.920894908905029,
      "epoch": 0.2738766776891655,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499566875640638,
      "loss": 5.7406,
      "mean_token_accuracy": 0.1555536597967148,
      "num_tokens": 6120575.0,
      "step": 3520
    },
    {
      "entropy": 5.913124990463257,
      "epoch": 0.2742657070608831,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004995651550590108,
      "loss": 5.7516,
      "mean_token_accuracy": 0.15152862668037415,
      "num_tokens": 6129367.0,
      "step": 3525
    },
    {
      "entropy": 5.8409215927124025,
      "epoch": 0.2746547364326007,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995634310699647,
      "loss": 5.72,
      "mean_token_accuracy": 0.15540992468595505,
      "num_tokens": 6137440.0,
      "step": 3530
    },
    {
      "entropy": 5.802550172805786,
      "epoch": 0.2750437658043182,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004995617036735261,
      "loss": 5.6392,
      "mean_token_accuracy": 0.15535928457975387,
      "num_tokens": 6146770.0,
      "step": 3535
    },
    {
      "entropy": 5.972483587265015,
      "epoch": 0.2754327951760358,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004995599728697211,
      "loss": 5.7402,
      "mean_token_accuracy": 0.15305006504058838,
      "num_tokens": 6155994.0,
      "step": 3540
    },
    {
      "entropy": 5.759503173828125,
      "epoch": 0.27582182454775334,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995582386585759,
      "loss": 5.6331,
      "mean_token_accuracy": 0.15320900082588196,
      "num_tokens": 6163941.0,
      "step": 3545
    },
    {
      "entropy": 5.798756551742554,
      "epoch": 0.2762108539194709,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995565010401167,
      "loss": 5.7032,
      "mean_token_accuracy": 0.14819759875535965,
      "num_tokens": 6172681.0,
      "step": 3550
    },
    {
      "entropy": 5.87427248954773,
      "epoch": 0.2765998832911885,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004995547600143702,
      "loss": 5.6763,
      "mean_token_accuracy": 0.1468235284090042,
      "num_tokens": 6181449.0,
      "step": 3555
    },
    {
      "entropy": 5.8288336277008055,
      "epoch": 0.27698891266290604,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004995530155813625,
      "loss": 5.6704,
      "mean_token_accuracy": 0.15640894025564195,
      "num_tokens": 6190275.0,
      "step": 3560
    },
    {
      "entropy": 5.878096294403076,
      "epoch": 0.2773779420346236,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004995512677411203,
      "loss": 5.6944,
      "mean_token_accuracy": 0.1533144235610962,
      "num_tokens": 6198521.0,
      "step": 3565
    },
    {
      "entropy": 5.882464456558227,
      "epoch": 0.27776697140634116,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004995495164936698,
      "loss": 5.7398,
      "mean_token_accuracy": 0.14755897372961044,
      "num_tokens": 6207318.0,
      "step": 3570
    },
    {
      "entropy": 5.817657423019409,
      "epoch": 0.27815600077805874,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004995477618390381,
      "loss": 5.6161,
      "mean_token_accuracy": 0.15557516366243362,
      "num_tokens": 6216693.0,
      "step": 3575
    },
    {
      "entropy": 5.801495313644409,
      "epoch": 0.27854503014977633,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995460037772513,
      "loss": 5.6811,
      "mean_token_accuracy": 0.15199395269155502,
      "num_tokens": 6225207.0,
      "step": 3580
    },
    {
      "entropy": 5.822018051147461,
      "epoch": 0.27893405952149386,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995442423083365,
      "loss": 5.6166,
      "mean_token_accuracy": 0.16307127773761748,
      "num_tokens": 6232653.0,
      "step": 3585
    },
    {
      "entropy": 5.78550910949707,
      "epoch": 0.27932308889321145,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995424774323201,
      "loss": 5.6199,
      "mean_token_accuracy": 0.16146358624100685,
      "num_tokens": 6241093.0,
      "step": 3590
    },
    {
      "entropy": 5.737304973602295,
      "epoch": 0.279712118264929,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499540709149229,
      "loss": 5.541,
      "mean_token_accuracy": 0.16245075911283494,
      "num_tokens": 6249144.0,
      "step": 3595
    },
    {
      "entropy": 5.807320785522461,
      "epoch": 0.28010114763664656,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004995389374590901,
      "loss": 5.6141,
      "mean_token_accuracy": 0.15612534880638124,
      "num_tokens": 6257622.0,
      "step": 3600
    },
    {
      "entropy": 5.836864995956421,
      "epoch": 0.28049017700836415,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995371623619301,
      "loss": 5.6374,
      "mean_token_accuracy": 0.15557539314031602,
      "num_tokens": 6266435.0,
      "step": 3605
    },
    {
      "entropy": 5.805403709411621,
      "epoch": 0.2808792063800817,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499535383857776,
      "loss": 5.7716,
      "mean_token_accuracy": 0.14686519801616668,
      "num_tokens": 6275053.0,
      "step": 3610
    },
    {
      "entropy": 5.825790929794311,
      "epoch": 0.28126823575179927,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499533601946655,
      "loss": 5.6419,
      "mean_token_accuracy": 0.15174684375524522,
      "num_tokens": 6283240.0,
      "step": 3615
    },
    {
      "entropy": 5.7599629878997805,
      "epoch": 0.2816572651235168,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995318166285938,
      "loss": 5.5906,
      "mean_token_accuracy": 0.16161544173955916,
      "num_tokens": 6292133.0,
      "step": 3620
    },
    {
      "entropy": 5.852939414978027,
      "epoch": 0.2820462944952344,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995300279036199,
      "loss": 5.7118,
      "mean_token_accuracy": 0.15506285429000854,
      "num_tokens": 6300954.0,
      "step": 3625
    },
    {
      "entropy": 5.764850902557373,
      "epoch": 0.28243532386695197,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00049952823577176,
      "loss": 5.5884,
      "mean_token_accuracy": 0.1620279625058174,
      "num_tokens": 6309637.0,
      "step": 3630
    },
    {
      "entropy": 5.811691093444824,
      "epoch": 0.2828243532386695,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995264402330416,
      "loss": 5.6501,
      "mean_token_accuracy": 0.15416521728038787,
      "num_tokens": 6318539.0,
      "step": 3635
    },
    {
      "entropy": 5.823588705062866,
      "epoch": 0.2832133826103871,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995246412874919,
      "loss": 5.5864,
      "mean_token_accuracy": 0.1556072413921356,
      "num_tokens": 6327109.0,
      "step": 3640
    },
    {
      "entropy": 5.76607027053833,
      "epoch": 0.2836024119821047,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499522838935138,
      "loss": 5.5816,
      "mean_token_accuracy": 0.15582408532500266,
      "num_tokens": 6335555.0,
      "step": 3645
    },
    {
      "entropy": 5.843725967407226,
      "epoch": 0.2839914413538222,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004995210331760074,
      "loss": 5.749,
      "mean_token_accuracy": 0.15214236080646515,
      "num_tokens": 6344716.0,
      "step": 3650
    },
    {
      "entropy": 5.800505876541138,
      "epoch": 0.2843804707255398,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995192240101275,
      "loss": 5.5952,
      "mean_token_accuracy": 0.16312976628541948,
      "num_tokens": 6353618.0,
      "step": 3655
    },
    {
      "entropy": 5.8812049388885494,
      "epoch": 0.2847695000972573,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995174114375258,
      "loss": 5.7227,
      "mean_token_accuracy": 0.14763198420405388,
      "num_tokens": 6362039.0,
      "step": 3660
    },
    {
      "entropy": 5.870044326782226,
      "epoch": 0.2851585294689749,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995155954582297,
      "loss": 5.6442,
      "mean_token_accuracy": 0.1539861425757408,
      "num_tokens": 6370331.0,
      "step": 3665
    },
    {
      "entropy": 5.775976467132568,
      "epoch": 0.2855475588406925,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995137760722668,
      "loss": 5.6273,
      "mean_token_accuracy": 0.1601376011967659,
      "num_tokens": 6378943.0,
      "step": 3670
    },
    {
      "entropy": 5.825973987579346,
      "epoch": 0.28593658821241,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995119532796646,
      "loss": 5.691,
      "mean_token_accuracy": 0.1445460721850395,
      "num_tokens": 6387745.0,
      "step": 3675
    },
    {
      "entropy": 5.7106483459472654,
      "epoch": 0.2863256175841276,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499510127080451,
      "loss": 5.5309,
      "mean_token_accuracy": 0.1584558069705963,
      "num_tokens": 6396915.0,
      "step": 3680
    },
    {
      "entropy": 5.7301453113555905,
      "epoch": 0.28671464695584514,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004995082974746535,
      "loss": 5.58,
      "mean_token_accuracy": 0.15873111188411712,
      "num_tokens": 6405453.0,
      "step": 3685
    },
    {
      "entropy": 5.8472319602966305,
      "epoch": 0.28710367632756273,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004995064644622999,
      "loss": 5.6586,
      "mean_token_accuracy": 0.15374509990215302,
      "num_tokens": 6413693.0,
      "step": 3690
    },
    {
      "entropy": 5.769003009796142,
      "epoch": 0.2874927056992803,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995046280434181,
      "loss": 5.6595,
      "mean_token_accuracy": 0.1512232981622219,
      "num_tokens": 6422738.0,
      "step": 3695
    },
    {
      "entropy": 5.813616132736206,
      "epoch": 0.28788173507099785,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995027882180358,
      "loss": 5.5907,
      "mean_token_accuracy": 0.1563073292374611,
      "num_tokens": 6431743.0,
      "step": 3700
    },
    {
      "entropy": 5.818098640441894,
      "epoch": 0.28827076444271543,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004995009449861812,
      "loss": 5.6324,
      "mean_token_accuracy": 0.15437836945056915,
      "num_tokens": 6440586.0,
      "step": 3705
    },
    {
      "entropy": 5.8015789031982425,
      "epoch": 0.28865979381443296,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499499098347882,
      "loss": 5.6886,
      "mean_token_accuracy": 0.1576991096138954,
      "num_tokens": 6449644.0,
      "step": 3710
    },
    {
      "entropy": 5.7611973762512205,
      "epoch": 0.28904882318615055,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004994972483031662,
      "loss": 5.5788,
      "mean_token_accuracy": 0.15928053706884385,
      "num_tokens": 6458752.0,
      "step": 3715
    },
    {
      "entropy": 5.740774345397949,
      "epoch": 0.28943785255786814,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000499495394852062,
      "loss": 5.6407,
      "mean_token_accuracy": 0.15863003581762314,
      "num_tokens": 6468333.0,
      "step": 3720
    },
    {
      "entropy": 5.665927886962891,
      "epoch": 0.28982688192958567,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994935379945977,
      "loss": 5.5011,
      "mean_token_accuracy": 0.15861316472291948,
      "num_tokens": 6477406.0,
      "step": 3725
    },
    {
      "entropy": 5.827551364898682,
      "epoch": 0.29021591130130325,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994916777308012,
      "loss": 5.659,
      "mean_token_accuracy": 0.15557901114225386,
      "num_tokens": 6486399.0,
      "step": 3730
    },
    {
      "entropy": 5.741009759902954,
      "epoch": 0.29060494067302084,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994898140607007,
      "loss": 5.5645,
      "mean_token_accuracy": 0.15651842430233956,
      "num_tokens": 6494853.0,
      "step": 3735
    },
    {
      "entropy": 5.797096490859985,
      "epoch": 0.29099397004473837,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994879469843247,
      "loss": 5.7275,
      "mean_token_accuracy": 0.15631626546382904,
      "num_tokens": 6503301.0,
      "step": 3740
    },
    {
      "entropy": 5.826006078720093,
      "epoch": 0.29138299941645596,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004994860765017013,
      "loss": 5.6201,
      "mean_token_accuracy": 0.15773724168539047,
      "num_tokens": 6512685.0,
      "step": 3745
    },
    {
      "entropy": 5.798263502120972,
      "epoch": 0.2917720287881735,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499484202612859,
      "loss": 5.7223,
      "mean_token_accuracy": 0.15036529451608657,
      "num_tokens": 6520673.0,
      "step": 3750
    },
    {
      "entropy": 5.807061958312988,
      "epoch": 0.2921610581598911,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004994823253178264,
      "loss": 5.6301,
      "mean_token_accuracy": 0.1517362654209137,
      "num_tokens": 6529039.0,
      "step": 3755
    },
    {
      "entropy": 5.808193635940552,
      "epoch": 0.29255008753160866,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994804446166317,
      "loss": 5.7114,
      "mean_token_accuracy": 0.15373645722866058,
      "num_tokens": 6537932.0,
      "step": 3760
    },
    {
      "entropy": 5.8420493602752686,
      "epoch": 0.2929391169033262,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004994785605093035,
      "loss": 5.5944,
      "mean_token_accuracy": 0.1566125750541687,
      "num_tokens": 6546278.0,
      "step": 3765
    },
    {
      "entropy": 5.7808938980102536,
      "epoch": 0.2933281462750438,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994766729958705,
      "loss": 5.6484,
      "mean_token_accuracy": 0.15963713079690933,
      "num_tokens": 6554449.0,
      "step": 3770
    },
    {
      "entropy": 5.7649589538574215,
      "epoch": 0.2937171756467613,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004994747820763612,
      "loss": 5.6254,
      "mean_token_accuracy": 0.15676820427179336,
      "num_tokens": 6563272.0,
      "step": 3775
    },
    {
      "entropy": 5.8970057487487795,
      "epoch": 0.2941062050184789,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004994728877508046,
      "loss": 5.6457,
      "mean_token_accuracy": 0.15577749907970428,
      "num_tokens": 6571864.0,
      "step": 3780
    },
    {
      "entropy": 5.765396070480347,
      "epoch": 0.2944952343901965,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004994709900192289,
      "loss": 5.6518,
      "mean_token_accuracy": 0.1575314909219742,
      "num_tokens": 6580435.0,
      "step": 3785
    },
    {
      "entropy": 5.855135822296143,
      "epoch": 0.294884263761914,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994690888816635,
      "loss": 5.6301,
      "mean_token_accuracy": 0.15876953154802323,
      "num_tokens": 6589261.0,
      "step": 3790
    },
    {
      "entropy": 5.823793649673462,
      "epoch": 0.2952732931336316,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004994671843381367,
      "loss": 5.7259,
      "mean_token_accuracy": 0.1586314097046852,
      "num_tokens": 6598057.0,
      "step": 3795
    },
    {
      "entropy": 5.912685489654541,
      "epoch": 0.29566232250534913,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004994652763886778,
      "loss": 5.6863,
      "mean_token_accuracy": 0.15288203358650207,
      "num_tokens": 6606439.0,
      "step": 3800
    },
    {
      "entropy": 5.812867593765259,
      "epoch": 0.2960513518770667,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994633650333155,
      "loss": 5.6928,
      "mean_token_accuracy": 0.15570134967565535,
      "num_tokens": 6615350.0,
      "step": 3805
    },
    {
      "entropy": 5.81829686164856,
      "epoch": 0.2964403812487843,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994614502720792,
      "loss": 5.729,
      "mean_token_accuracy": 0.1526700034737587,
      "num_tokens": 6624258.0,
      "step": 3810
    },
    {
      "entropy": 5.933809518814087,
      "epoch": 0.29682941062050183,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994595321049974,
      "loss": 5.73,
      "mean_token_accuracy": 0.14964744150638581,
      "num_tokens": 6632382.0,
      "step": 3815
    },
    {
      "entropy": 5.775454378128051,
      "epoch": 0.2972184399922194,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004994576105320994,
      "loss": 5.6314,
      "mean_token_accuracy": 0.15691597014665604,
      "num_tokens": 6640482.0,
      "step": 3820
    },
    {
      "entropy": 5.7647865295410154,
      "epoch": 0.29760746936393695,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994556855534146,
      "loss": 5.6172,
      "mean_token_accuracy": 0.15665874034166336,
      "num_tokens": 6648912.0,
      "step": 3825
    },
    {
      "entropy": 5.802655172348023,
      "epoch": 0.29799649873565454,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994537571689719,
      "loss": 5.5879,
      "mean_token_accuracy": 0.16325612962245942,
      "num_tokens": 6656912.0,
      "step": 3830
    },
    {
      "entropy": 5.689533519744873,
      "epoch": 0.2983855281073721,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004994518253788007,
      "loss": 5.6279,
      "mean_token_accuracy": 0.15841389894485475,
      "num_tokens": 6665900.0,
      "step": 3835
    },
    {
      "entropy": 5.833044481277466,
      "epoch": 0.29877455747908965,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994498901829304,
      "loss": 5.6876,
      "mean_token_accuracy": 0.15316616296768187,
      "num_tokens": 6674426.0,
      "step": 3840
    },
    {
      "entropy": 5.8335991382598875,
      "epoch": 0.29916358685080724,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994479515813903,
      "loss": 5.6285,
      "mean_token_accuracy": 0.15730705857276917,
      "num_tokens": 6683566.0,
      "step": 3845
    },
    {
      "entropy": 5.746164703369141,
      "epoch": 0.2995526162225248,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994460095742096,
      "loss": 5.651,
      "mean_token_accuracy": 0.1572297602891922,
      "num_tokens": 6693341.0,
      "step": 3850
    },
    {
      "entropy": 5.787024450302124,
      "epoch": 0.29994164559424236,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499444064161418,
      "loss": 5.5538,
      "mean_token_accuracy": 0.15665306150913239,
      "num_tokens": 6701734.0,
      "step": 3855
    },
    {
      "entropy": 5.753479909896851,
      "epoch": 0.30033067496595994,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499442115343045,
      "loss": 5.5831,
      "mean_token_accuracy": 0.15792362838983537,
      "num_tokens": 6710572.0,
      "step": 3860
    },
    {
      "entropy": 5.7785179138183596,
      "epoch": 0.3007197043376775,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994401631191202,
      "loss": 5.7056,
      "mean_token_accuracy": 0.15263333171606064,
      "num_tokens": 6719355.0,
      "step": 3865
    },
    {
      "entropy": 5.929450035095215,
      "epoch": 0.30110873370939506,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004994382074896731,
      "loss": 5.7344,
      "mean_token_accuracy": 0.1500555157661438,
      "num_tokens": 6729279.0,
      "step": 3870
    },
    {
      "entropy": 5.789097547531128,
      "epoch": 0.30149776308111265,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994362484547335,
      "loss": 5.5582,
      "mean_token_accuracy": 0.156083644926548,
      "num_tokens": 6737318.0,
      "step": 3875
    },
    {
      "entropy": 5.763490629196167,
      "epoch": 0.3018867924528302,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004994342860143309,
      "loss": 5.6434,
      "mean_token_accuracy": 0.1571171462535858,
      "num_tokens": 6746520.0,
      "step": 3880
    },
    {
      "entropy": 5.786573886871338,
      "epoch": 0.30227582182454776,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994323201684953,
      "loss": 5.5616,
      "mean_token_accuracy": 0.16137573271989822,
      "num_tokens": 6755685.0,
      "step": 3885
    },
    {
      "entropy": 5.813833856582642,
      "epoch": 0.3026648511962653,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004994303509172566,
      "loss": 5.7087,
      "mean_token_accuracy": 0.15270676612854003,
      "num_tokens": 6764347.0,
      "step": 3890
    },
    {
      "entropy": 5.8389809131622314,
      "epoch": 0.3030538805679829,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994283782606444,
      "loss": 5.6769,
      "mean_token_accuracy": 0.15980001389980317,
      "num_tokens": 6772887.0,
      "step": 3895
    },
    {
      "entropy": 5.747214937210083,
      "epoch": 0.30344290993970047,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994264021986888,
      "loss": 5.6378,
      "mean_token_accuracy": 0.1533655397593975,
      "num_tokens": 6780860.0,
      "step": 3900
    },
    {
      "entropy": 5.781130027770996,
      "epoch": 0.303831939311418,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994244227314197,
      "loss": 5.6698,
      "mean_token_accuracy": 0.15610713958740235,
      "num_tokens": 6789491.0,
      "step": 3905
    },
    {
      "entropy": 5.861451292037964,
      "epoch": 0.3042209686831356,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994224398588672,
      "loss": 5.5945,
      "mean_token_accuracy": 0.15678352415561675,
      "num_tokens": 6797128.0,
      "step": 3910
    },
    {
      "entropy": 5.7906858921051025,
      "epoch": 0.3046099980548531,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004994204535810615,
      "loss": 5.6269,
      "mean_token_accuracy": 0.15614872425794601,
      "num_tokens": 6805235.0,
      "step": 3915
    },
    {
      "entropy": 5.724874782562256,
      "epoch": 0.3049990274265707,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994184638980326,
      "loss": 5.5355,
      "mean_token_accuracy": 0.15990695655345916,
      "num_tokens": 6813863.0,
      "step": 3920
    },
    {
      "entropy": 5.7675896167755125,
      "epoch": 0.3053880567982883,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994164708098107,
      "loss": 5.6011,
      "mean_token_accuracy": 0.1568060576915741,
      "num_tokens": 6822015.0,
      "step": 3925
    },
    {
      "entropy": 5.847005462646484,
      "epoch": 0.3057770861700058,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499414474316426,
      "loss": 5.672,
      "mean_token_accuracy": 0.15424158275127411,
      "num_tokens": 6830598.0,
      "step": 3930
    },
    {
      "entropy": 5.797545289993286,
      "epoch": 0.3061661155417234,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004994124744179088,
      "loss": 5.6619,
      "mean_token_accuracy": 0.1645057313144207,
      "num_tokens": 6839129.0,
      "step": 3935
    },
    {
      "entropy": 5.808172369003296,
      "epoch": 0.306555144913441,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004994104711142896,
      "loss": 5.6101,
      "mean_token_accuracy": 0.15196753293275833,
      "num_tokens": 6848199.0,
      "step": 3940
    },
    {
      "entropy": 5.799759769439698,
      "epoch": 0.3069441742851585,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004994084644055986,
      "loss": 5.6313,
      "mean_token_accuracy": 0.15993681401014329,
      "num_tokens": 6857340.0,
      "step": 3945
    },
    {
      "entropy": 5.753718996047974,
      "epoch": 0.3073332036568761,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994064542918664,
      "loss": 5.6657,
      "mean_token_accuracy": 0.15309109836816787,
      "num_tokens": 6865830.0,
      "step": 3950
    },
    {
      "entropy": 5.8039764881134035,
      "epoch": 0.30772223302859364,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004994044407731235,
      "loss": 5.7128,
      "mean_token_accuracy": 0.15277842581272125,
      "num_tokens": 6874318.0,
      "step": 3955
    },
    {
      "entropy": 5.9370396614074705,
      "epoch": 0.3081112624003112,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004994024238494002,
      "loss": 5.7105,
      "mean_token_accuracy": 0.14689986407756805,
      "num_tokens": 6882008.0,
      "step": 3960
    },
    {
      "entropy": 5.783172225952148,
      "epoch": 0.3085002917720288,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994004035207274,
      "loss": 5.5668,
      "mean_token_accuracy": 0.1621098205447197,
      "num_tokens": 6890411.0,
      "step": 3965
    },
    {
      "entropy": 5.747785043716431,
      "epoch": 0.30888932114374634,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004993983797871356,
      "loss": 5.6063,
      "mean_token_accuracy": 0.15390863865613938,
      "num_tokens": 6899201.0,
      "step": 3970
    },
    {
      "entropy": 5.75714340209961,
      "epoch": 0.30927835051546393,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004993963526486555,
      "loss": 5.6242,
      "mean_token_accuracy": 0.1541016474366188,
      "num_tokens": 6907470.0,
      "step": 3975
    },
    {
      "entropy": 5.756861925125122,
      "epoch": 0.30966737988718146,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499394322105318,
      "loss": 5.6563,
      "mean_token_accuracy": 0.15384358018636704,
      "num_tokens": 6916542.0,
      "step": 3980
    },
    {
      "entropy": 5.796830177307129,
      "epoch": 0.31005640925889905,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004993922881571538,
      "loss": 5.6515,
      "mean_token_accuracy": 0.15680348873138428,
      "num_tokens": 6925974.0,
      "step": 3985
    },
    {
      "entropy": 5.869470882415771,
      "epoch": 0.31044543863061663,
      "grad_norm": 1.0,
      "learning_rate": 0.0004993902508041939,
      "loss": 5.6726,
      "mean_token_accuracy": 0.14945410192012787,
      "num_tokens": 6934987.0,
      "step": 3990
    },
    {
      "entropy": 5.743036556243896,
      "epoch": 0.31083446800233416,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000499388210046469,
      "loss": 5.5603,
      "mean_token_accuracy": 0.15534475296735764,
      "num_tokens": 6944497.0,
      "step": 3995
    },
    {
      "entropy": 5.8093039989471436,
      "epoch": 0.31122349737405175,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004993861658840102,
      "loss": 5.6948,
      "mean_token_accuracy": 0.15089739114046097,
      "num_tokens": 6953394.0,
      "step": 4000
    },
    {
      "entropy": 5.807778215408325,
      "epoch": 0.3116125267457693,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004993841183168484,
      "loss": 5.6118,
      "mean_token_accuracy": 0.15972677171230315,
      "num_tokens": 6961715.0,
      "step": 4005
    },
    {
      "entropy": 5.731346893310547,
      "epoch": 0.31200155611748687,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004993820673450148,
      "loss": 5.6121,
      "mean_token_accuracy": 0.15898443907499313,
      "num_tokens": 6970041.0,
      "step": 4010
    },
    {
      "entropy": 5.829005384445191,
      "epoch": 0.31239058548920445,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993800129685404,
      "loss": 5.7512,
      "mean_token_accuracy": 0.15160393118858337,
      "num_tokens": 6979522.0,
      "step": 4015
    },
    {
      "entropy": 5.854546594619751,
      "epoch": 0.312779614860922,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993779551874565,
      "loss": 5.7053,
      "mean_token_accuracy": 0.1508307620882988,
      "num_tokens": 6988470.0,
      "step": 4020
    },
    {
      "entropy": 5.915190696716309,
      "epoch": 0.31316864423263957,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993758940017943,
      "loss": 5.7187,
      "mean_token_accuracy": 0.15690406635403634,
      "num_tokens": 6996696.0,
      "step": 4025
    },
    {
      "entropy": 5.816199493408203,
      "epoch": 0.31355767360435716,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499373829411585,
      "loss": 5.5878,
      "mean_token_accuracy": 0.15647319704294205,
      "num_tokens": 7005297.0,
      "step": 4030
    },
    {
      "entropy": 5.758253908157348,
      "epoch": 0.3139467029760747,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00049937176141686,
      "loss": 5.5895,
      "mean_token_accuracy": 0.15717500299215317,
      "num_tokens": 7013569.0,
      "step": 4035
    },
    {
      "entropy": 5.738769006729126,
      "epoch": 0.3143357323477923,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004993696900176506,
      "loss": 5.5396,
      "mean_token_accuracy": 0.15976371392607688,
      "num_tokens": 7021449.0,
      "step": 4040
    },
    {
      "entropy": 5.777108478546142,
      "epoch": 0.3147247617195098,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004993676152139883,
      "loss": 5.6578,
      "mean_token_accuracy": 0.15396279245615005,
      "num_tokens": 7029663.0,
      "step": 4045
    },
    {
      "entropy": 5.760292291641235,
      "epoch": 0.3151137910912274,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004993655370059046,
      "loss": 5.6574,
      "mean_token_accuracy": 0.15446390137076377,
      "num_tokens": 7039038.0,
      "step": 4050
    },
    {
      "entropy": 5.78679347038269,
      "epoch": 0.315502820462945,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004993634553934309,
      "loss": 5.6006,
      "mean_token_accuracy": 0.15957597270607948,
      "num_tokens": 7047595.0,
      "step": 4055
    },
    {
      "entropy": 5.786748456954956,
      "epoch": 0.3158918498346625,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993613703765989,
      "loss": 5.6223,
      "mean_token_accuracy": 0.15329041033983232,
      "num_tokens": 7055711.0,
      "step": 4060
    },
    {
      "entropy": 5.786895799636841,
      "epoch": 0.3162808792063801,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004993592819554402,
      "loss": 5.597,
      "mean_token_accuracy": 0.1581099197268486,
      "num_tokens": 7064100.0,
      "step": 4065
    },
    {
      "entropy": 5.783665037155151,
      "epoch": 0.3166699085780976,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993571901299866,
      "loss": 5.5626,
      "mean_token_accuracy": 0.16303355991840363,
      "num_tokens": 7073040.0,
      "step": 4070
    },
    {
      "entropy": 5.785036230087281,
      "epoch": 0.3170589379498152,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993550949002697,
      "loss": 5.6739,
      "mean_token_accuracy": 0.1541762962937355,
      "num_tokens": 7081513.0,
      "step": 4075
    },
    {
      "entropy": 5.741682720184326,
      "epoch": 0.3174479673215328,
      "grad_norm": 1.0,
      "learning_rate": 0.0004993529962663213,
      "loss": 5.5549,
      "mean_token_accuracy": 0.16333828270435333,
      "num_tokens": 7090704.0,
      "step": 4080
    },
    {
      "entropy": 5.749197149276734,
      "epoch": 0.31783699669325033,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004993508942281732,
      "loss": 5.7181,
      "mean_token_accuracy": 0.14728385508060454,
      "num_tokens": 7099692.0,
      "step": 4085
    },
    {
      "entropy": 5.7576361179351805,
      "epoch": 0.3182260260649679,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004993487887858575,
      "loss": 5.5516,
      "mean_token_accuracy": 0.15469060242176055,
      "num_tokens": 7108021.0,
      "step": 4090
    },
    {
      "entropy": 5.791795110702514,
      "epoch": 0.31861505543668545,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499346679939406,
      "loss": 5.6172,
      "mean_token_accuracy": 0.15831730961799623,
      "num_tokens": 7116689.0,
      "step": 4095
    },
    {
      "entropy": 5.788445377349854,
      "epoch": 0.31900408480840303,
      "grad_norm": 1.125,
      "learning_rate": 0.0004993445676888507,
      "loss": 5.7264,
      "mean_token_accuracy": 0.15102792978286744,
      "num_tokens": 7125453.0,
      "step": 4100
    },
    {
      "entropy": 5.898724317550659,
      "epoch": 0.3193931141801206,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993424520342236,
      "loss": 5.7438,
      "mean_token_accuracy": 0.1532034158706665,
      "num_tokens": 7133692.0,
      "step": 4105
    },
    {
      "entropy": 5.763997888565063,
      "epoch": 0.31978214355183815,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004993403329755569,
      "loss": 5.5763,
      "mean_token_accuracy": 0.16306018978357315,
      "num_tokens": 7142362.0,
      "step": 4110
    },
    {
      "entropy": 5.711903047561646,
      "epoch": 0.32017117292355574,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993382105128828,
      "loss": 5.5877,
      "mean_token_accuracy": 0.15780982971191407,
      "num_tokens": 7151020.0,
      "step": 4115
    },
    {
      "entropy": 5.738257598876953,
      "epoch": 0.32056020229527327,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004993360846462333,
      "loss": 5.4984,
      "mean_token_accuracy": 0.16021377146244048,
      "num_tokens": 7159453.0,
      "step": 4120
    },
    {
      "entropy": 5.821284532546997,
      "epoch": 0.32094923166699085,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004993339553756408,
      "loss": 5.7802,
      "mean_token_accuracy": 0.14731935635209084,
      "num_tokens": 7168758.0,
      "step": 4125
    },
    {
      "entropy": 5.792877912521362,
      "epoch": 0.32133826103870844,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993318227011378,
      "loss": 5.5758,
      "mean_token_accuracy": 0.1596775844693184,
      "num_tokens": 7177835.0,
      "step": 4130
    },
    {
      "entropy": 5.755284166336059,
      "epoch": 0.32172729041042597,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004993296866227562,
      "loss": 5.6175,
      "mean_token_accuracy": 0.1625669404864311,
      "num_tokens": 7186424.0,
      "step": 4135
    },
    {
      "entropy": 5.801063823699951,
      "epoch": 0.32211631978214356,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993275471405288,
      "loss": 5.5849,
      "mean_token_accuracy": 0.16332011371850969,
      "num_tokens": 7194800.0,
      "step": 4140
    },
    {
      "entropy": 5.707626724243164,
      "epoch": 0.32250534915386114,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004993254042544879,
      "loss": 5.5368,
      "mean_token_accuracy": 0.16244402751326562,
      "num_tokens": 7203820.0,
      "step": 4145
    },
    {
      "entropy": 5.7076304912567135,
      "epoch": 0.3228943785255787,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499323257964666,
      "loss": 5.5251,
      "mean_token_accuracy": 0.16087611466646196,
      "num_tokens": 7211793.0,
      "step": 4150
    },
    {
      "entropy": 5.765037918090821,
      "epoch": 0.32328340789729626,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004993211082710957,
      "loss": 5.5877,
      "mean_token_accuracy": 0.16022843420505523,
      "num_tokens": 7220471.0,
      "step": 4155
    },
    {
      "entropy": 5.769202852249146,
      "epoch": 0.3236724372690138,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993189551738097,
      "loss": 5.6408,
      "mean_token_accuracy": 0.15618563294410706,
      "num_tokens": 7228788.0,
      "step": 4160
    },
    {
      "entropy": 5.827218103408813,
      "epoch": 0.3240614666407314,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993167986728405,
      "loss": 5.7919,
      "mean_token_accuracy": 0.15401089787483216,
      "num_tokens": 7237707.0,
      "step": 4165
    },
    {
      "entropy": 5.81566367149353,
      "epoch": 0.32445049601244896,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499314638768221,
      "loss": 5.5426,
      "mean_token_accuracy": 0.16399593651294708,
      "num_tokens": 7246296.0,
      "step": 4170
    },
    {
      "entropy": 5.716956663131714,
      "epoch": 0.3248395253841665,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993124754599839,
      "loss": 5.4757,
      "mean_token_accuracy": 0.16270810216665268,
      "num_tokens": 7255076.0,
      "step": 4175
    },
    {
      "entropy": 5.584178733825683,
      "epoch": 0.3252285547558841,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004993103087481619,
      "loss": 5.5717,
      "mean_token_accuracy": 0.16209740787744523,
      "num_tokens": 7263169.0,
      "step": 4180
    },
    {
      "entropy": 5.807580137252808,
      "epoch": 0.3256175841276016,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004993081386327881,
      "loss": 5.6369,
      "mean_token_accuracy": 0.16165959537029267,
      "num_tokens": 7271748.0,
      "step": 4185
    },
    {
      "entropy": 5.74710111618042,
      "epoch": 0.3260066134993192,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004993059651138953,
      "loss": 5.4493,
      "mean_token_accuracy": 0.16019158512353898,
      "num_tokens": 7280291.0,
      "step": 4190
    },
    {
      "entropy": 5.666503047943115,
      "epoch": 0.3263956428710368,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993037881915165,
      "loss": 5.6308,
      "mean_token_accuracy": 0.15489906072616577,
      "num_tokens": 7290307.0,
      "step": 4195
    },
    {
      "entropy": 5.827814769744873,
      "epoch": 0.3267846722427543,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993016078656847,
      "loss": 5.7312,
      "mean_token_accuracy": 0.1512381076812744,
      "num_tokens": 7297942.0,
      "step": 4200
    },
    {
      "entropy": 5.732455778121948,
      "epoch": 0.3271737016144719,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499299424136433,
      "loss": 5.6003,
      "mean_token_accuracy": 0.1560288652777672,
      "num_tokens": 7307378.0,
      "step": 4205
    },
    {
      "entropy": 5.762440347671509,
      "epoch": 0.32756273098618943,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004992972370037946,
      "loss": 5.5543,
      "mean_token_accuracy": 0.16346563994884492,
      "num_tokens": 7315985.0,
      "step": 4210
    },
    {
      "entropy": 5.812780046463013,
      "epoch": 0.327951760357907,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004992950464678026,
      "loss": 5.6665,
      "mean_token_accuracy": 0.15801236927509307,
      "num_tokens": 7324142.0,
      "step": 4215
    },
    {
      "entropy": 5.691588354110718,
      "epoch": 0.3283407897296246,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004992928525284903,
      "loss": 5.5215,
      "mean_token_accuracy": 0.15780295431613922,
      "num_tokens": 7333012.0,
      "step": 4220
    },
    {
      "entropy": 5.744389343261719,
      "epoch": 0.32872981910134214,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499290655185891,
      "loss": 5.5711,
      "mean_token_accuracy": 0.163132381439209,
      "num_tokens": 7341838.0,
      "step": 4225
    },
    {
      "entropy": 5.7133368968963625,
      "epoch": 0.3291188484730597,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499288454440038,
      "loss": 5.5061,
      "mean_token_accuracy": 0.16726227551698686,
      "num_tokens": 7349986.0,
      "step": 4230
    },
    {
      "entropy": 5.8127552509307865,
      "epoch": 0.3295078778447773,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004992862502909645,
      "loss": 5.7528,
      "mean_token_accuracy": 0.1513574466109276,
      "num_tokens": 7359428.0,
      "step": 4235
    },
    {
      "entropy": 5.859197282791138,
      "epoch": 0.32989690721649484,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992840427387043,
      "loss": 5.6522,
      "mean_token_accuracy": 0.15405798852443695,
      "num_tokens": 7368524.0,
      "step": 4240
    },
    {
      "entropy": 5.789969062805175,
      "epoch": 0.3302859365882124,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004992818317832908,
      "loss": 5.5471,
      "mean_token_accuracy": 0.1576457753777504,
      "num_tokens": 7377979.0,
      "step": 4245
    },
    {
      "entropy": 5.635087633132935,
      "epoch": 0.33067496595992996,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992796174247574,
      "loss": 5.4541,
      "mean_token_accuracy": 0.16037447601556779,
      "num_tokens": 7386283.0,
      "step": 4250
    },
    {
      "entropy": 5.664917135238648,
      "epoch": 0.33106399533164754,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004992773996631378,
      "loss": 5.596,
      "mean_token_accuracy": 0.15014797449111938,
      "num_tokens": 7395409.0,
      "step": 4255
    },
    {
      "entropy": 5.80849256515503,
      "epoch": 0.33145302470336513,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004992751784984656,
      "loss": 5.6695,
      "mean_token_accuracy": 0.1540595680475235,
      "num_tokens": 7403389.0,
      "step": 4260
    },
    {
      "entropy": 5.758498573303223,
      "epoch": 0.33184205407508266,
      "grad_norm": 1.0,
      "learning_rate": 0.0004992729539307745,
      "loss": 5.5814,
      "mean_token_accuracy": 0.15872332602739334,
      "num_tokens": 7411506.0,
      "step": 4265
    },
    {
      "entropy": 5.792923545837402,
      "epoch": 0.33223108344680025,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004992707259600984,
      "loss": 5.6829,
      "mean_token_accuracy": 0.15659109950065614,
      "num_tokens": 7420219.0,
      "step": 4270
    },
    {
      "entropy": 5.784406566619873,
      "epoch": 0.3326201128185178,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004992684945864709,
      "loss": 5.5855,
      "mean_token_accuracy": 0.16268359273672103,
      "num_tokens": 7428442.0,
      "step": 4275
    },
    {
      "entropy": 5.741096687316895,
      "epoch": 0.33300914219023536,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004992662598099261,
      "loss": 5.5533,
      "mean_token_accuracy": 0.16411196142435075,
      "num_tokens": 7436364.0,
      "step": 4280
    },
    {
      "entropy": 5.7426940441131595,
      "epoch": 0.33339817156195295,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004992640216304975,
      "loss": 5.6203,
      "mean_token_accuracy": 0.15268845930695535,
      "num_tokens": 7444122.0,
      "step": 4285
    },
    {
      "entropy": 5.815378856658936,
      "epoch": 0.3337872009336705,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992617800482196,
      "loss": 5.6274,
      "mean_token_accuracy": 0.15351774916052818,
      "num_tokens": 7454122.0,
      "step": 4290
    },
    {
      "entropy": 5.75958480834961,
      "epoch": 0.33417623030538807,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499259535063126,
      "loss": 5.6329,
      "mean_token_accuracy": 0.15966753661632538,
      "num_tokens": 7463057.0,
      "step": 4295
    },
    {
      "entropy": 5.759614610671997,
      "epoch": 0.3345652596771056,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499257286675251,
      "loss": 5.6019,
      "mean_token_accuracy": 0.15679196566343306,
      "num_tokens": 7471675.0,
      "step": 4300
    },
    {
      "entropy": 5.660829019546509,
      "epoch": 0.3349542890488232,
      "grad_norm": 1.0,
      "learning_rate": 0.0004992550348846285,
      "loss": 5.543,
      "mean_token_accuracy": 0.15981356799602509,
      "num_tokens": 7479870.0,
      "step": 4305
    },
    {
      "entropy": 5.7939084529876705,
      "epoch": 0.33534331842054077,
      "grad_norm": 1.0,
      "learning_rate": 0.0004992527796912928,
      "loss": 5.6133,
      "mean_token_accuracy": 0.15689725428819656,
      "num_tokens": 7489830.0,
      "step": 4310
    },
    {
      "entropy": 5.739326524734497,
      "epoch": 0.3357323477922583,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992505210952782,
      "loss": 5.5453,
      "mean_token_accuracy": 0.16509315073490144,
      "num_tokens": 7497852.0,
      "step": 4315
    },
    {
      "entropy": 5.7341211318969725,
      "epoch": 0.3361213771639759,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004992482590966188,
      "loss": 5.6712,
      "mean_token_accuracy": 0.1600148931145668,
      "num_tokens": 7506419.0,
      "step": 4320
    },
    {
      "entropy": 5.793564033508301,
      "epoch": 0.3365104065356934,
      "grad_norm": 1.0,
      "learning_rate": 0.0004992459936953489,
      "loss": 5.5826,
      "mean_token_accuracy": 0.164952789247036,
      "num_tokens": 7514757.0,
      "step": 4325
    },
    {
      "entropy": 5.7567079067230225,
      "epoch": 0.336899435907411,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004992437248915032,
      "loss": 5.6338,
      "mean_token_accuracy": 0.15620678365230561,
      "num_tokens": 7523912.0,
      "step": 4330
    },
    {
      "entropy": 5.675789833068848,
      "epoch": 0.3372884652791286,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992414526851157,
      "loss": 5.5141,
      "mean_token_accuracy": 0.16389839053153993,
      "num_tokens": 7532401.0,
      "step": 4335
    },
    {
      "entropy": 5.65968542098999,
      "epoch": 0.3376774946508461,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004992391770762212,
      "loss": 5.4872,
      "mean_token_accuracy": 0.16425893306732178,
      "num_tokens": 7541257.0,
      "step": 4340
    },
    {
      "entropy": 5.682751226425171,
      "epoch": 0.3380665240225637,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000499236898064854,
      "loss": 5.4357,
      "mean_token_accuracy": 0.16272755414247514,
      "num_tokens": 7549812.0,
      "step": 4345
    },
    {
      "entropy": 5.66465916633606,
      "epoch": 0.3384555533942813,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004992346156510488,
      "loss": 5.5252,
      "mean_token_accuracy": 0.1634671851992607,
      "num_tokens": 7559319.0,
      "step": 4350
    },
    {
      "entropy": 5.7516988754272464,
      "epoch": 0.3388445827659988,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004992323298348403,
      "loss": 5.6792,
      "mean_token_accuracy": 0.1523505702614784,
      "num_tokens": 7567900.0,
      "step": 4355
    },
    {
      "entropy": 5.835768604278565,
      "epoch": 0.3392336121377164,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499230040616263,
      "loss": 5.5907,
      "mean_token_accuracy": 0.15399069488048553,
      "num_tokens": 7576163.0,
      "step": 4360
    },
    {
      "entropy": 5.735501480102539,
      "epoch": 0.33962264150943394,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004992277479953518,
      "loss": 5.5239,
      "mean_token_accuracy": 0.16011153161525726,
      "num_tokens": 7584221.0,
      "step": 4365
    },
    {
      "entropy": 5.692150402069092,
      "epoch": 0.34001167088115153,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004992254519721414,
      "loss": 5.522,
      "mean_token_accuracy": 0.16146405190229415,
      "num_tokens": 7593569.0,
      "step": 4370
    },
    {
      "entropy": 5.694510746002197,
      "epoch": 0.3404007002528691,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004992231525466666,
      "loss": 5.58,
      "mean_token_accuracy": 0.1608036696910858,
      "num_tokens": 7602666.0,
      "step": 4375
    },
    {
      "entropy": 5.709047651290893,
      "epoch": 0.34078972962458665,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004992208497189624,
      "loss": 5.6591,
      "mean_token_accuracy": 0.15225481390953063,
      "num_tokens": 7611874.0,
      "step": 4380
    },
    {
      "entropy": 5.796590566635132,
      "epoch": 0.34117875899630423,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004992185434890637,
      "loss": 5.607,
      "mean_token_accuracy": 0.1618656650185585,
      "num_tokens": 7620385.0,
      "step": 4385
    },
    {
      "entropy": 5.819919538497925,
      "epoch": 0.34156778836802176,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004992162338570055,
      "loss": 5.6732,
      "mean_token_accuracy": 0.14887086302042007,
      "num_tokens": 7628786.0,
      "step": 4390
    },
    {
      "entropy": 5.7014552593231205,
      "epoch": 0.34195681773973935,
      "grad_norm": 1.0,
      "learning_rate": 0.0004992139208228227,
      "loss": 5.4761,
      "mean_token_accuracy": 0.16517678499221802,
      "num_tokens": 7637389.0,
      "step": 4395
    },
    {
      "entropy": 5.685795831680298,
      "epoch": 0.34234584711145694,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004992116043865506,
      "loss": 5.6409,
      "mean_token_accuracy": 0.15962141305208205,
      "num_tokens": 7645793.0,
      "step": 4400
    },
    {
      "entropy": 5.836361122131348,
      "epoch": 0.34273487648317447,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992092845482244,
      "loss": 5.5901,
      "mean_token_accuracy": 0.16014044135808944,
      "num_tokens": 7653951.0,
      "step": 4405
    },
    {
      "entropy": 5.672879076004028,
      "epoch": 0.34312390585489205,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004992069613078791,
      "loss": 5.479,
      "mean_token_accuracy": 0.16533271372318267,
      "num_tokens": 7662753.0,
      "step": 4410
    },
    {
      "entropy": 5.648221015930176,
      "epoch": 0.3435129352266096,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004992046346655499,
      "loss": 5.4363,
      "mean_token_accuracy": 0.16223288476467132,
      "num_tokens": 7671832.0,
      "step": 4415
    },
    {
      "entropy": 5.7435859680175785,
      "epoch": 0.34390196459832717,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004992023046212723,
      "loss": 5.6066,
      "mean_token_accuracy": 0.1573074996471405,
      "num_tokens": 7680969.0,
      "step": 4420
    },
    {
      "entropy": 5.7971031188964846,
      "epoch": 0.34429099397004476,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991999711750816,
      "loss": 5.6109,
      "mean_token_accuracy": 0.153276427090168,
      "num_tokens": 7689277.0,
      "step": 4425
    },
    {
      "entropy": 5.703974771499634,
      "epoch": 0.3446800233417623,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004991976343270133,
      "loss": 5.5229,
      "mean_token_accuracy": 0.16756367832422256,
      "num_tokens": 7697373.0,
      "step": 4430
    },
    {
      "entropy": 5.802385854721069,
      "epoch": 0.3450690527134799,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991952940771026,
      "loss": 5.597,
      "mean_token_accuracy": 0.15816881507635117,
      "num_tokens": 7705767.0,
      "step": 4435
    },
    {
      "entropy": 5.67456431388855,
      "epoch": 0.34545808208519746,
      "grad_norm": 1.125,
      "learning_rate": 0.0004991929504253852,
      "loss": 5.4795,
      "mean_token_accuracy": 0.16529733687639236,
      "num_tokens": 7713746.0,
      "step": 4440
    },
    {
      "entropy": 5.700855445861817,
      "epoch": 0.345847111456915,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004991906033718966,
      "loss": 5.6177,
      "mean_token_accuracy": 0.15523315817117692,
      "num_tokens": 7721829.0,
      "step": 4445
    },
    {
      "entropy": 5.684748125076294,
      "epoch": 0.3462361408286326,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004991882529166724,
      "loss": 5.4995,
      "mean_token_accuracy": 0.16088906228542327,
      "num_tokens": 7730479.0,
      "step": 4450
    },
    {
      "entropy": 5.747715044021606,
      "epoch": 0.3466251702003501,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991858990597483,
      "loss": 5.4645,
      "mean_token_accuracy": 0.16631900370121003,
      "num_tokens": 7738659.0,
      "step": 4455
    },
    {
      "entropy": 5.775031328201294,
      "epoch": 0.3470141995720677,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991835418011601,
      "loss": 5.7453,
      "mean_token_accuracy": 0.15679385513067245,
      "num_tokens": 7747597.0,
      "step": 4460
    },
    {
      "entropy": 5.736643743515015,
      "epoch": 0.3474032289437853,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004991811811409434,
      "loss": 5.5906,
      "mean_token_accuracy": 0.1590351827442646,
      "num_tokens": 7756666.0,
      "step": 4465
    },
    {
      "entropy": 5.719669580459595,
      "epoch": 0.3477922583155028,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991788170791341,
      "loss": 5.6122,
      "mean_token_accuracy": 0.157757031917572,
      "num_tokens": 7765621.0,
      "step": 4470
    },
    {
      "entropy": 5.711370182037354,
      "epoch": 0.3481812876872204,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991764496157682,
      "loss": 5.5161,
      "mean_token_accuracy": 0.15926966667175294,
      "num_tokens": 7774137.0,
      "step": 4475
    },
    {
      "entropy": 5.79927077293396,
      "epoch": 0.34857031705893793,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004991740787508814,
      "loss": 5.6149,
      "mean_token_accuracy": 0.1526617646217346,
      "num_tokens": 7783144.0,
      "step": 4480
    },
    {
      "entropy": 5.7292121887207035,
      "epoch": 0.3489593464306555,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991717044845097,
      "loss": 5.6553,
      "mean_token_accuracy": 0.15161389783024787,
      "num_tokens": 7791530.0,
      "step": 4485
    },
    {
      "entropy": 5.7438078880310055,
      "epoch": 0.3493483758023731,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004991693268166892,
      "loss": 5.5084,
      "mean_token_accuracy": 0.16316290348768234,
      "num_tokens": 7800869.0,
      "step": 4490
    },
    {
      "entropy": 5.714588069915772,
      "epoch": 0.34973740517409063,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004991669457474559,
      "loss": 5.5936,
      "mean_token_accuracy": 0.1608248233795166,
      "num_tokens": 7810201.0,
      "step": 4495
    },
    {
      "entropy": 5.6973871231079105,
      "epoch": 0.3501264345458082,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004991645612768461,
      "loss": 5.4826,
      "mean_token_accuracy": 0.16842566281557084,
      "num_tokens": 7818529.0,
      "step": 4500
    },
    {
      "entropy": 5.698172330856323,
      "epoch": 0.35051546391752575,
      "grad_norm": 1.0,
      "learning_rate": 0.0004991621734048958,
      "loss": 5.5308,
      "mean_token_accuracy": 0.16790317595005036,
      "num_tokens": 7827254.0,
      "step": 4505
    },
    {
      "entropy": 5.725019407272339,
      "epoch": 0.35090449328924334,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004991597821316415,
      "loss": 5.544,
      "mean_token_accuracy": 0.1619415909051895,
      "num_tokens": 7836437.0,
      "step": 4510
    },
    {
      "entropy": 5.697857475280761,
      "epoch": 0.3512935226609609,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004991573874571192,
      "loss": 5.5437,
      "mean_token_accuracy": 0.16284422427415848,
      "num_tokens": 7844939.0,
      "step": 4515
    },
    {
      "entropy": 5.720762729644775,
      "epoch": 0.35168255203267845,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991549893813653,
      "loss": 5.5216,
      "mean_token_accuracy": 0.15868689864873886,
      "num_tokens": 7852889.0,
      "step": 4520
    },
    {
      "entropy": 5.77600417137146,
      "epoch": 0.35207158140439604,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991525879044162,
      "loss": 5.6742,
      "mean_token_accuracy": 0.15549049526453018,
      "num_tokens": 7861867.0,
      "step": 4525
    },
    {
      "entropy": 5.702154541015625,
      "epoch": 0.35246061077611357,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991501830263084,
      "loss": 5.4717,
      "mean_token_accuracy": 0.1698843866586685,
      "num_tokens": 7869343.0,
      "step": 4530
    },
    {
      "entropy": 5.729362964630127,
      "epoch": 0.35284964014783116,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004991477747470783,
      "loss": 5.5225,
      "mean_token_accuracy": 0.16422102302312852,
      "num_tokens": 7877142.0,
      "step": 4535
    },
    {
      "entropy": 5.67864956855774,
      "epoch": 0.35323866951954874,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004991453630667625,
      "loss": 5.5478,
      "mean_token_accuracy": 0.16405916661024095,
      "num_tokens": 7885183.0,
      "step": 4540
    },
    {
      "entropy": 5.5773327350616455,
      "epoch": 0.3536276988912663,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991429479853976,
      "loss": 5.3768,
      "mean_token_accuracy": 0.16831792742013932,
      "num_tokens": 7893848.0,
      "step": 4545
    },
    {
      "entropy": 5.6918010234832765,
      "epoch": 0.35401672826298386,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004991405295030202,
      "loss": 5.5552,
      "mean_token_accuracy": 0.16068892925977707,
      "num_tokens": 7902438.0,
      "step": 4550
    },
    {
      "entropy": 5.708292245864868,
      "epoch": 0.35440575763470145,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004991381076196671,
      "loss": 5.5235,
      "mean_token_accuracy": 0.1578003153204918,
      "num_tokens": 7910951.0,
      "step": 4555
    },
    {
      "entropy": 5.700537490844726,
      "epoch": 0.354794787006419,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004991356823353748,
      "loss": 5.6356,
      "mean_token_accuracy": 0.15564948841929435,
      "num_tokens": 7920375.0,
      "step": 4560
    },
    {
      "entropy": 5.666615581512451,
      "epoch": 0.35518381637813656,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004991332536501804,
      "loss": 5.4984,
      "mean_token_accuracy": 0.1664597824215889,
      "num_tokens": 7928656.0,
      "step": 4565
    },
    {
      "entropy": 5.829252243041992,
      "epoch": 0.3555728457498541,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004991308215641205,
      "loss": 5.7335,
      "mean_token_accuracy": 0.149640491604805,
      "num_tokens": 7936929.0,
      "step": 4570
    },
    {
      "entropy": 5.759623050689697,
      "epoch": 0.3559618751215717,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991283860772323,
      "loss": 5.5434,
      "mean_token_accuracy": 0.1643831327557564,
      "num_tokens": 7945592.0,
      "step": 4575
    },
    {
      "entropy": 5.700624799728393,
      "epoch": 0.35635090449328927,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004991259471895525,
      "loss": 5.4988,
      "mean_token_accuracy": 0.1662057653069496,
      "num_tokens": 7953656.0,
      "step": 4580
    },
    {
      "entropy": 5.717899131774902,
      "epoch": 0.3567399338650068,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004991235049011182,
      "loss": 5.5573,
      "mean_token_accuracy": 0.1631222665309906,
      "num_tokens": 7962038.0,
      "step": 4585
    },
    {
      "entropy": 5.733677768707276,
      "epoch": 0.3571289632367244,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004991210592119663,
      "loss": 5.6425,
      "mean_token_accuracy": 0.15202119797468186,
      "num_tokens": 7970338.0,
      "step": 4590
    },
    {
      "entropy": 5.7634436130523685,
      "epoch": 0.3575179926084419,
      "grad_norm": 1.0,
      "learning_rate": 0.0004991186101221342,
      "loss": 5.5692,
      "mean_token_accuracy": 0.16211371570825578,
      "num_tokens": 7979173.0,
      "step": 4595
    },
    {
      "entropy": 5.686269044876099,
      "epoch": 0.3579070219801595,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499116157631659,
      "loss": 5.4418,
      "mean_token_accuracy": 0.1698352262377739,
      "num_tokens": 7987245.0,
      "step": 4600
    },
    {
      "entropy": 5.736973428726197,
      "epoch": 0.3582960513518771,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991137017405777,
      "loss": 5.6469,
      "mean_token_accuracy": 0.1493039160966873,
      "num_tokens": 7995551.0,
      "step": 4605
    },
    {
      "entropy": 5.636900091171265,
      "epoch": 0.3586850807235946,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004991112424489277,
      "loss": 5.4519,
      "mean_token_accuracy": 0.16245980858802794,
      "num_tokens": 8004069.0,
      "step": 4610
    },
    {
      "entropy": 5.685174226760864,
      "epoch": 0.3590741100953122,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004991087797567462,
      "loss": 5.5399,
      "mean_token_accuracy": 0.16341051310300828,
      "num_tokens": 8012537.0,
      "step": 4615
    },
    {
      "entropy": 5.761553335189819,
      "epoch": 0.35946313946702974,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004991063136640709,
      "loss": 5.5287,
      "mean_token_accuracy": 0.1633700117468834,
      "num_tokens": 8021387.0,
      "step": 4620
    },
    {
      "entropy": 5.706650924682617,
      "epoch": 0.3598521688387473,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004991038441709388,
      "loss": 5.4672,
      "mean_token_accuracy": 0.16015367358922958,
      "num_tokens": 8030261.0,
      "step": 4625
    },
    {
      "entropy": 5.767949962615967,
      "epoch": 0.3602411982104649,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991013712773877,
      "loss": 5.5355,
      "mean_token_accuracy": 0.15736105889081956,
      "num_tokens": 8038161.0,
      "step": 4630
    },
    {
      "entropy": 5.666050863265991,
      "epoch": 0.36063022758218244,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004990988949834549,
      "loss": 5.6315,
      "mean_token_accuracy": 0.1598365470767021,
      "num_tokens": 8046340.0,
      "step": 4635
    },
    {
      "entropy": 5.749620676040649,
      "epoch": 0.3610192569539,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499096415289178,
      "loss": 5.5497,
      "mean_token_accuracy": 0.15858872681856157,
      "num_tokens": 8054997.0,
      "step": 4640
    },
    {
      "entropy": 5.681770181655883,
      "epoch": 0.3614082863256176,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004990939321945948,
      "loss": 5.4899,
      "mean_token_accuracy": 0.16831997483968736,
      "num_tokens": 8063686.0,
      "step": 4645
    },
    {
      "entropy": 5.703184795379639,
      "epoch": 0.36179731569733514,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004990914456997427,
      "loss": 5.6065,
      "mean_token_accuracy": 0.15740725249052048,
      "num_tokens": 8072999.0,
      "step": 4650
    },
    {
      "entropy": 5.688112354278564,
      "epoch": 0.36218634506905273,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004990889558046598,
      "loss": 5.4586,
      "mean_token_accuracy": 0.16635777801275253,
      "num_tokens": 8081596.0,
      "step": 4655
    },
    {
      "entropy": 5.723988056182861,
      "epoch": 0.36257537444077026,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004990864625093836,
      "loss": 5.6833,
      "mean_token_accuracy": 0.15422971099615096,
      "num_tokens": 8090948.0,
      "step": 4660
    },
    {
      "entropy": 5.691533470153809,
      "epoch": 0.36296440381248785,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990839658139519,
      "loss": 5.6917,
      "mean_token_accuracy": 0.1527736008167267,
      "num_tokens": 8099445.0,
      "step": 4665
    },
    {
      "entropy": 5.79793004989624,
      "epoch": 0.36335343318420543,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004990814657184028,
      "loss": 5.5416,
      "mean_token_accuracy": 0.1518623486161232,
      "num_tokens": 8108119.0,
      "step": 4670
    },
    {
      "entropy": 5.735276746749878,
      "epoch": 0.36374246255592296,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004990789622227741,
      "loss": 5.5632,
      "mean_token_accuracy": 0.16103332489728928,
      "num_tokens": 8117060.0,
      "step": 4675
    },
    {
      "entropy": 5.661766004562378,
      "epoch": 0.36413149192764055,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004990764553271038,
      "loss": 5.5396,
      "mean_token_accuracy": 0.16326614171266557,
      "num_tokens": 8126229.0,
      "step": 4680
    },
    {
      "entropy": 5.735515117645264,
      "epoch": 0.3645205212993581,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004990739450314299,
      "loss": 5.497,
      "mean_token_accuracy": 0.17422327995300294,
      "num_tokens": 8134694.0,
      "step": 4685
    },
    {
      "entropy": 5.692343664169312,
      "epoch": 0.36490955067107567,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004990714313357906,
      "loss": 5.5698,
      "mean_token_accuracy": 0.16217846274375916,
      "num_tokens": 8143329.0,
      "step": 4690
    },
    {
      "entropy": 5.705916833877564,
      "epoch": 0.36529858004279325,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499068914240224,
      "loss": 5.5521,
      "mean_token_accuracy": 0.1667526125907898,
      "num_tokens": 8151432.0,
      "step": 4695
    },
    {
      "entropy": 5.7003473281860355,
      "epoch": 0.3656876094145108,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004990663937447682,
      "loss": 5.5156,
      "mean_token_accuracy": 0.16553931534290314,
      "num_tokens": 8159677.0,
      "step": 4700
    },
    {
      "entropy": 5.665900373458863,
      "epoch": 0.36607663878622837,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004990638698494615,
      "loss": 5.5811,
      "mean_token_accuracy": 0.157049497961998,
      "num_tokens": 8168710.0,
      "step": 4705
    },
    {
      "entropy": 5.718477869033814,
      "epoch": 0.3664656681579459,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004990613425543423,
      "loss": 5.5621,
      "mean_token_accuracy": 0.16302944272756575,
      "num_tokens": 8177501.0,
      "step": 4710
    },
    {
      "entropy": 5.824956321716309,
      "epoch": 0.3668546975296635,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004990588118594487,
      "loss": 5.646,
      "mean_token_accuracy": 0.15586332529783248,
      "num_tokens": 8185791.0,
      "step": 4715
    },
    {
      "entropy": 5.770373153686523,
      "epoch": 0.3672437269013811,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004990562777648194,
      "loss": 5.6626,
      "mean_token_accuracy": 0.16009435802698135,
      "num_tokens": 8195020.0,
      "step": 4720
    },
    {
      "entropy": 5.6913634777069095,
      "epoch": 0.3676327562730986,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004990537402704928,
      "loss": 5.5344,
      "mean_token_accuracy": 0.1582164779305458,
      "num_tokens": 8203540.0,
      "step": 4725
    },
    {
      "entropy": 5.771582317352295,
      "epoch": 0.3680217856448162,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004990511993765071,
      "loss": 5.5201,
      "mean_token_accuracy": 0.15834065079689025,
      "num_tokens": 8211985.0,
      "step": 4730
    },
    {
      "entropy": 5.6868315696716305,
      "epoch": 0.3684108150165337,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004990486550829011,
      "loss": 5.5977,
      "mean_token_accuracy": 0.16268818974494934,
      "num_tokens": 8220289.0,
      "step": 4735
    },
    {
      "entropy": 5.8069891929626465,
      "epoch": 0.3687998443882513,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004990461073897134,
      "loss": 5.6219,
      "mean_token_accuracy": 0.15378663092851638,
      "num_tokens": 8229652.0,
      "step": 4740
    },
    {
      "entropy": 5.790693902969361,
      "epoch": 0.3691888737599689,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004990435562969827,
      "loss": 5.6213,
      "mean_token_accuracy": 0.15782863050699233,
      "num_tokens": 8237728.0,
      "step": 4745
    },
    {
      "entropy": 5.6777466297149655,
      "epoch": 0.3695779031316864,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004990410018047475,
      "loss": 5.5061,
      "mean_token_accuracy": 0.16452137678861617,
      "num_tokens": 8246393.0,
      "step": 4750
    },
    {
      "entropy": 5.654371023178101,
      "epoch": 0.369966932503404,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004990384439130467,
      "loss": 5.4701,
      "mean_token_accuracy": 0.16525126546621322,
      "num_tokens": 8256736.0,
      "step": 4755
    },
    {
      "entropy": 5.737254905700683,
      "epoch": 0.3703559618751216,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004990358826219192,
      "loss": 5.5647,
      "mean_token_accuracy": 0.1631600394845009,
      "num_tokens": 8265126.0,
      "step": 4760
    },
    {
      "entropy": 5.716107511520386,
      "epoch": 0.37074499124683913,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004990333179314038,
      "loss": 5.4713,
      "mean_token_accuracy": 0.1658305548131466,
      "num_tokens": 8273136.0,
      "step": 4765
    },
    {
      "entropy": 5.576092529296875,
      "epoch": 0.3711340206185567,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004990307498415393,
      "loss": 5.393,
      "mean_token_accuracy": 0.16830165013670922,
      "num_tokens": 8281697.0,
      "step": 4770
    },
    {
      "entropy": 5.625568485260009,
      "epoch": 0.37152304999027425,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004990281783523648,
      "loss": 5.5292,
      "mean_token_accuracy": 0.17112733125686647,
      "num_tokens": 8290765.0,
      "step": 4775
    },
    {
      "entropy": 5.794410514831543,
      "epoch": 0.37191207936199183,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990256034639192,
      "loss": 5.6345,
      "mean_token_accuracy": 0.15772489309310914,
      "num_tokens": 8299875.0,
      "step": 4780
    },
    {
      "entropy": 5.699645185470581,
      "epoch": 0.3723011087337094,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004990230251762418,
      "loss": 5.497,
      "mean_token_accuracy": 0.16678553968667983,
      "num_tokens": 8308515.0,
      "step": 4785
    },
    {
      "entropy": 5.633892059326172,
      "epoch": 0.37269013810542695,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004990204434893713,
      "loss": 5.4212,
      "mean_token_accuracy": 0.17109153419733047,
      "num_tokens": 8317260.0,
      "step": 4790
    },
    {
      "entropy": 5.723616027832032,
      "epoch": 0.37307916747714454,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004990178584033474,
      "loss": 5.5903,
      "mean_token_accuracy": 0.15471651703119277,
      "num_tokens": 8325887.0,
      "step": 4795
    },
    {
      "entropy": 5.791163778305053,
      "epoch": 0.37346819684886207,
      "grad_norm": 1.0,
      "learning_rate": 0.0004990152699182089,
      "loss": 5.5933,
      "mean_token_accuracy": 0.15361500382423401,
      "num_tokens": 8333974.0,
      "step": 4800
    },
    {
      "entropy": 5.673134469985962,
      "epoch": 0.37385722622057965,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004990126780339953,
      "loss": 5.5633,
      "mean_token_accuracy": 0.1595918729901314,
      "num_tokens": 8343199.0,
      "step": 4805
    },
    {
      "entropy": 5.618671321868897,
      "epoch": 0.37424625559229724,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004990100827507459,
      "loss": 5.3647,
      "mean_token_accuracy": 0.17112764716148376,
      "num_tokens": 8351202.0,
      "step": 4810
    },
    {
      "entropy": 5.743080711364746,
      "epoch": 0.37463528496401477,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004990074840684999,
      "loss": 5.7062,
      "mean_token_accuracy": 0.1510538563132286,
      "num_tokens": 8360011.0,
      "step": 4815
    },
    {
      "entropy": 5.719053077697754,
      "epoch": 0.37502431433573236,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004990048819872969,
      "loss": 5.5776,
      "mean_token_accuracy": 0.16062003821134568,
      "num_tokens": 8367874.0,
      "step": 4820
    },
    {
      "entropy": 5.719112777709961,
      "epoch": 0.3754133437074499,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004990022765071765,
      "loss": 5.5265,
      "mean_token_accuracy": 0.16080788075923919,
      "num_tokens": 8376708.0,
      "step": 4825
    },
    {
      "entropy": 5.742806720733642,
      "epoch": 0.3758023730791675,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000498999667628178,
      "loss": 5.5579,
      "mean_token_accuracy": 0.16359954476356506,
      "num_tokens": 8385467.0,
      "step": 4830
    },
    {
      "entropy": 5.691323232650757,
      "epoch": 0.37619140245088506,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004989970553503411,
      "loss": 5.5159,
      "mean_token_accuracy": 0.1591488540172577,
      "num_tokens": 8393574.0,
      "step": 4835
    },
    {
      "entropy": 5.662193822860718,
      "epoch": 0.3765804318226026,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004989944396737054,
      "loss": 5.5633,
      "mean_token_accuracy": 0.16046745032072068,
      "num_tokens": 8402857.0,
      "step": 4840
    },
    {
      "entropy": 5.703533458709717,
      "epoch": 0.3769694611943202,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004989918205983106,
      "loss": 5.5743,
      "mean_token_accuracy": 0.17153286933898926,
      "num_tokens": 8411728.0,
      "step": 4845
    },
    {
      "entropy": 5.758769559860229,
      "epoch": 0.37735849056603776,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004989891981241964,
      "loss": 5.556,
      "mean_token_accuracy": 0.16337950974702836,
      "num_tokens": 8420307.0,
      "step": 4850
    },
    {
      "entropy": 5.714922618865967,
      "epoch": 0.3777475199377553,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004989865722514027,
      "loss": 5.5571,
      "mean_token_accuracy": 0.15444338619709014,
      "num_tokens": 8429442.0,
      "step": 4855
    },
    {
      "entropy": 5.666832208633423,
      "epoch": 0.3781365493094729,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004989839429799692,
      "loss": 5.4536,
      "mean_token_accuracy": 0.16823951601982118,
      "num_tokens": 8437945.0,
      "step": 4860
    },
    {
      "entropy": 5.644245529174805,
      "epoch": 0.3785255786811904,
      "grad_norm": 1.09375,
      "learning_rate": 0.000498981310309936,
      "loss": 5.4404,
      "mean_token_accuracy": 0.16453132629394532,
      "num_tokens": 8445743.0,
      "step": 4865
    },
    {
      "entropy": 5.654577589035034,
      "epoch": 0.378914608052908,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004989786742413428,
      "loss": 5.5307,
      "mean_token_accuracy": 0.15743673890829085,
      "num_tokens": 8454091.0,
      "step": 4870
    },
    {
      "entropy": 5.790392732620239,
      "epoch": 0.3793036374246256,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004989760347742298,
      "loss": 5.5369,
      "mean_token_accuracy": 0.15999819785356523,
      "num_tokens": 8461992.0,
      "step": 4875
    },
    {
      "entropy": 5.628113794326782,
      "epoch": 0.3796926667963431,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004989733919086369,
      "loss": 5.4858,
      "mean_token_accuracy": 0.16528079956769942,
      "num_tokens": 8470730.0,
      "step": 4880
    },
    {
      "entropy": 5.649323844909668,
      "epoch": 0.3800816961680607,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004989707456446043,
      "loss": 5.5329,
      "mean_token_accuracy": 0.15793396532535553,
      "num_tokens": 8479947.0,
      "step": 4885
    },
    {
      "entropy": 5.7841041564941404,
      "epoch": 0.38047072553977823,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004989680959821721,
      "loss": 5.6132,
      "mean_token_accuracy": 0.16241377145051955,
      "num_tokens": 8488237.0,
      "step": 4890
    },
    {
      "entropy": 5.6983592987060545,
      "epoch": 0.3808597549114958,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004989654429213805,
      "loss": 5.4721,
      "mean_token_accuracy": 0.1697806179523468,
      "num_tokens": 8497146.0,
      "step": 4895
    },
    {
      "entropy": 5.713892698287964,
      "epoch": 0.3812487842832134,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004989627864622699,
      "loss": 5.5602,
      "mean_token_accuracy": 0.16247230917215347,
      "num_tokens": 8505736.0,
      "step": 4900
    },
    {
      "entropy": 5.7166149616241455,
      "epoch": 0.38163781365493094,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004989601266048804,
      "loss": 5.5634,
      "mean_token_accuracy": 0.160552416741848,
      "num_tokens": 8515368.0,
      "step": 4905
    },
    {
      "entropy": 5.698330879211426,
      "epoch": 0.3820268430266485,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004989574633492525,
      "loss": 5.5375,
      "mean_token_accuracy": 0.16279081553220748,
      "num_tokens": 8524141.0,
      "step": 4910
    },
    {
      "entropy": 5.688110256195069,
      "epoch": 0.38241587239836605,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004989547966954266,
      "loss": 5.4532,
      "mean_token_accuracy": 0.1611531913280487,
      "num_tokens": 8533116.0,
      "step": 4915
    },
    {
      "entropy": 5.646517324447632,
      "epoch": 0.38280490177008364,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000498952126643443,
      "loss": 5.5777,
      "mean_token_accuracy": 0.16830433458089827,
      "num_tokens": 8541919.0,
      "step": 4920
    },
    {
      "entropy": 5.6638476848602295,
      "epoch": 0.3831939311418012,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004989494531933424,
      "loss": 5.4598,
      "mean_token_accuracy": 0.1696254074573517,
      "num_tokens": 8550372.0,
      "step": 4925
    },
    {
      "entropy": 5.71760311126709,
      "epoch": 0.38358296051351876,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004989467763451652,
      "loss": 5.5636,
      "mean_token_accuracy": 0.1605513334274292,
      "num_tokens": 8559246.0,
      "step": 4930
    },
    {
      "entropy": 5.625193119049072,
      "epoch": 0.38397198988523634,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004989440960989523,
      "loss": 5.3301,
      "mean_token_accuracy": 0.1718791589140892,
      "num_tokens": 8567005.0,
      "step": 4935
    },
    {
      "entropy": 5.594034099578858,
      "epoch": 0.3843610192569539,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000498941412454744,
      "loss": 5.5302,
      "mean_token_accuracy": 0.15369682461023332,
      "num_tokens": 8575399.0,
      "step": 4940
    },
    {
      "entropy": 5.778599739074707,
      "epoch": 0.38475004862867146,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004989387254125813,
      "loss": 5.6007,
      "mean_token_accuracy": 0.15802320688962937,
      "num_tokens": 8583994.0,
      "step": 4945
    },
    {
      "entropy": 5.684764814376831,
      "epoch": 0.38513907800038905,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004989360349725049,
      "loss": 5.5293,
      "mean_token_accuracy": 0.16684771031141282,
      "num_tokens": 8593446.0,
      "step": 4950
    },
    {
      "entropy": 5.687840461730957,
      "epoch": 0.3855281073721066,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004989333411345555,
      "loss": 5.5257,
      "mean_token_accuracy": 0.16687152683734893,
      "num_tokens": 8602821.0,
      "step": 4955
    },
    {
      "entropy": 5.696226263046265,
      "epoch": 0.38591713674382416,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004989306438987742,
      "loss": 5.6207,
      "mean_token_accuracy": 0.15970443338155746,
      "num_tokens": 8611831.0,
      "step": 4960
    },
    {
      "entropy": 5.7406542778015135,
      "epoch": 0.38630616611554175,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004989279432652018,
      "loss": 5.5119,
      "mean_token_accuracy": 0.16554239988327027,
      "num_tokens": 8620986.0,
      "step": 4965
    },
    {
      "entropy": 5.686248922348023,
      "epoch": 0.3866951954872593,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004989252392338791,
      "loss": 5.453,
      "mean_token_accuracy": 0.16537498086690902,
      "num_tokens": 8629822.0,
      "step": 4970
    },
    {
      "entropy": 5.690024566650391,
      "epoch": 0.38708422485897687,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004989225318048475,
      "loss": 5.5705,
      "mean_token_accuracy": 0.15630877017974854,
      "num_tokens": 8637679.0,
      "step": 4975
    },
    {
      "entropy": 5.737256383895874,
      "epoch": 0.3874732542306944,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004989198209781478,
      "loss": 5.5923,
      "mean_token_accuracy": 0.158028045296669,
      "num_tokens": 8646705.0,
      "step": 4980
    },
    {
      "entropy": 5.754758739471436,
      "epoch": 0.387862283602412,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004989171067538211,
      "loss": 5.519,
      "mean_token_accuracy": 0.16303036659955977,
      "num_tokens": 8655599.0,
      "step": 4985
    },
    {
      "entropy": 5.646867895126343,
      "epoch": 0.38825131297412957,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000498914389131909,
      "loss": 5.5341,
      "mean_token_accuracy": 0.1682754009962082,
      "num_tokens": 8665025.0,
      "step": 4990
    },
    {
      "entropy": 5.649749040603638,
      "epoch": 0.3886403423458471,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004989116681124523,
      "loss": 5.4188,
      "mean_token_accuracy": 0.1704274147748947,
      "num_tokens": 8672994.0,
      "step": 4995
    },
    {
      "entropy": 5.624281358718872,
      "epoch": 0.3890293717175647,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004989089436954924,
      "loss": 5.5622,
      "mean_token_accuracy": 0.15898848474025726,
      "num_tokens": 8682457.0,
      "step": 5000
    },
    {
      "entropy": 5.739774608612061,
      "epoch": 0.3894184010892822,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004989062158810706,
      "loss": 5.5326,
      "mean_token_accuracy": 0.16346432864665986,
      "num_tokens": 8690888.0,
      "step": 5005
    },
    {
      "entropy": 5.787450122833252,
      "epoch": 0.3898074304609998,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004989034846692284,
      "loss": 5.6305,
      "mean_token_accuracy": 0.16038641184568406,
      "num_tokens": 8699632.0,
      "step": 5010
    },
    {
      "entropy": 5.636838054656982,
      "epoch": 0.3901964598327174,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004989007500600073,
      "loss": 5.4726,
      "mean_token_accuracy": 0.16696102768182755,
      "num_tokens": 8707955.0,
      "step": 5015
    },
    {
      "entropy": 5.6930876731872555,
      "epoch": 0.3905854892044349,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004988980120534486,
      "loss": 5.6131,
      "mean_token_accuracy": 0.163053660094738,
      "num_tokens": 8716103.0,
      "step": 5020
    },
    {
      "entropy": 5.659214639663697,
      "epoch": 0.3909745185761525,
      "grad_norm": 1.015625,
      "learning_rate": 0.000498895270649594,
      "loss": 5.4648,
      "mean_token_accuracy": 0.17025580406188964,
      "num_tokens": 8724499.0,
      "step": 5025
    },
    {
      "entropy": 5.598870706558228,
      "epoch": 0.39136354794787004,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000498892525848485,
      "loss": 5.4783,
      "mean_token_accuracy": 0.17237136960029603,
      "num_tokens": 8732829.0,
      "step": 5030
    },
    {
      "entropy": 5.5582152843475345,
      "epoch": 0.3917525773195876,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004988897776501633,
      "loss": 5.3546,
      "mean_token_accuracy": 0.17332328334450722,
      "num_tokens": 8741617.0,
      "step": 5035
    },
    {
      "entropy": 5.609436273574829,
      "epoch": 0.3921416066913052,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004988870260546705,
      "loss": 5.342,
      "mean_token_accuracy": 0.1763409346342087,
      "num_tokens": 8750146.0,
      "step": 5040
    },
    {
      "entropy": 5.682952070236206,
      "epoch": 0.39253063606302274,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004988842710620486,
      "loss": 5.5173,
      "mean_token_accuracy": 0.16273482590913774,
      "num_tokens": 8758744.0,
      "step": 5045
    },
    {
      "entropy": 5.594632911682129,
      "epoch": 0.39291966543474033,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004988815126723391,
      "loss": 5.5273,
      "mean_token_accuracy": 0.15988082885742189,
      "num_tokens": 8768219.0,
      "step": 5050
    },
    {
      "entropy": 5.68524203300476,
      "epoch": 0.3933086948064579,
      "grad_norm": 1.0,
      "learning_rate": 0.0004988787508855841,
      "loss": 5.5299,
      "mean_token_accuracy": 0.1616822063922882,
      "num_tokens": 8777047.0,
      "step": 5055
    },
    {
      "entropy": 5.64661750793457,
      "epoch": 0.39369772417817545,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004988759857018253,
      "loss": 5.4171,
      "mean_token_accuracy": 0.16649373024702072,
      "num_tokens": 8785918.0,
      "step": 5060
    },
    {
      "entropy": 5.602305221557617,
      "epoch": 0.39408675354989303,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988732171211048,
      "loss": 5.5035,
      "mean_token_accuracy": 0.15666143596172333,
      "num_tokens": 8794253.0,
      "step": 5065
    },
    {
      "entropy": 5.680363178253174,
      "epoch": 0.39447578292161056,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988704451434644,
      "loss": 5.549,
      "mean_token_accuracy": 0.1658593460917473,
      "num_tokens": 8802352.0,
      "step": 5070
    },
    {
      "entropy": 5.6955273151397705,
      "epoch": 0.39486481229332815,
      "grad_norm": 1.0,
      "learning_rate": 0.0004988676697689465,
      "loss": 5.5196,
      "mean_token_accuracy": 0.1587934285402298,
      "num_tokens": 8810956.0,
      "step": 5075
    },
    {
      "entropy": 5.531219434738159,
      "epoch": 0.39525384166504574,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000498864890997593,
      "loss": 5.4296,
      "mean_token_accuracy": 0.16261873245239258,
      "num_tokens": 8819062.0,
      "step": 5080
    },
    {
      "entropy": 5.654786825180054,
      "epoch": 0.39564287103676327,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004988621088294461,
      "loss": 5.5314,
      "mean_token_accuracy": 0.15552457123994828,
      "num_tokens": 8827470.0,
      "step": 5085
    },
    {
      "entropy": 5.707628393173218,
      "epoch": 0.39603190040848085,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000498859323264548,
      "loss": 5.5581,
      "mean_token_accuracy": 0.16062256842851638,
      "num_tokens": 8837003.0,
      "step": 5090
    },
    {
      "entropy": 5.6874775886535645,
      "epoch": 0.3964209297801984,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000498856534302941,
      "loss": 5.5608,
      "mean_token_accuracy": 0.15407982915639878,
      "num_tokens": 8846605.0,
      "step": 5095
    },
    {
      "entropy": 5.72544584274292,
      "epoch": 0.39680995915191597,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004988537419446673,
      "loss": 5.5834,
      "mean_token_accuracy": 0.1655305802822113,
      "num_tokens": 8855232.0,
      "step": 5100
    },
    {
      "entropy": 5.723621511459351,
      "epoch": 0.39719898852363356,
      "grad_norm": 1.0,
      "learning_rate": 0.0004988509461897694,
      "loss": 5.4733,
      "mean_token_accuracy": 0.1710292637348175,
      "num_tokens": 8864176.0,
      "step": 5105
    },
    {
      "entropy": 5.634792900085449,
      "epoch": 0.3975880178953511,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004988481470382897,
      "loss": 5.5572,
      "mean_token_accuracy": 0.1697572275996208,
      "num_tokens": 8872671.0,
      "step": 5110
    },
    {
      "entropy": 5.580335474014282,
      "epoch": 0.3979770472670687,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004988453444902708,
      "loss": 5.4439,
      "mean_token_accuracy": 0.1702474907040596,
      "num_tokens": 8881316.0,
      "step": 5115
    },
    {
      "entropy": 5.734807109832763,
      "epoch": 0.3983660766387862,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988425385457549,
      "loss": 5.5739,
      "mean_token_accuracy": 0.16411821097135543,
      "num_tokens": 8890114.0,
      "step": 5120
    },
    {
      "entropy": 5.723512649536133,
      "epoch": 0.3987551060105038,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004988397292047848,
      "loss": 5.5085,
      "mean_token_accuracy": 0.1680221065878868,
      "num_tokens": 8899598.0,
      "step": 5125
    },
    {
      "entropy": 5.563539695739746,
      "epoch": 0.3991441353822214,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004988369164674032,
      "loss": 5.3636,
      "mean_token_accuracy": 0.16728202253580093,
      "num_tokens": 8908244.0,
      "step": 5130
    },
    {
      "entropy": 5.621714067459107,
      "epoch": 0.3995331647539389,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004988341003336526,
      "loss": 5.4206,
      "mean_token_accuracy": 0.1688075691461563,
      "num_tokens": 8916280.0,
      "step": 5135
    },
    {
      "entropy": 5.721775341033935,
      "epoch": 0.3999221941256565,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004988312808035757,
      "loss": 5.4744,
      "mean_token_accuracy": 0.16042978763580323,
      "num_tokens": 8923947.0,
      "step": 5140
    },
    {
      "entropy": 5.60654878616333,
      "epoch": 0.400311223497374,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004988284578772155,
      "loss": 5.5254,
      "mean_token_accuracy": 0.16021703332662582,
      "num_tokens": 8933374.0,
      "step": 5145
    },
    {
      "entropy": 5.612744903564453,
      "epoch": 0.4007002528690916,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004988256315546146,
      "loss": 5.4509,
      "mean_token_accuracy": 0.16805752664804458,
      "num_tokens": 8941757.0,
      "step": 5150
    },
    {
      "entropy": 5.625841426849365,
      "epoch": 0.4010892822408092,
      "grad_norm": 1.0,
      "learning_rate": 0.0004988228018358161,
      "loss": 5.3531,
      "mean_token_accuracy": 0.17098772823810576,
      "num_tokens": 8950128.0,
      "step": 5155
    },
    {
      "entropy": 5.682161426544189,
      "epoch": 0.40147831161252673,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988199687208628,
      "loss": 5.5001,
      "mean_token_accuracy": 0.1601304143667221,
      "num_tokens": 8958939.0,
      "step": 5160
    },
    {
      "entropy": 5.6470198154449465,
      "epoch": 0.4018673409842443,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988171322097977,
      "loss": 5.4581,
      "mean_token_accuracy": 0.16841353476047516,
      "num_tokens": 8967519.0,
      "step": 5165
    },
    {
      "entropy": 5.540816164016723,
      "epoch": 0.4022563703559619,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004988142923026638,
      "loss": 5.3437,
      "mean_token_accuracy": 0.17257575541734696,
      "num_tokens": 8976215.0,
      "step": 5170
    },
    {
      "entropy": 5.704356336593628,
      "epoch": 0.40264539972767943,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004988114489995044,
      "loss": 5.5729,
      "mean_token_accuracy": 0.15235375314950944,
      "num_tokens": 8984060.0,
      "step": 5175
    },
    {
      "entropy": 5.702901840209961,
      "epoch": 0.403034429099397,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004988086023003624,
      "loss": 5.5069,
      "mean_token_accuracy": 0.16046454161405563,
      "num_tokens": 8992301.0,
      "step": 5180
    },
    {
      "entropy": 5.702128458023071,
      "epoch": 0.40342345847111455,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004988057522052811,
      "loss": 5.5246,
      "mean_token_accuracy": 0.16668639332056046,
      "num_tokens": 9000517.0,
      "step": 5185
    },
    {
      "entropy": 5.628679847717285,
      "epoch": 0.40381248784283214,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004988028987143037,
      "loss": 5.4778,
      "mean_token_accuracy": 0.16389556378126144,
      "num_tokens": 9008826.0,
      "step": 5190
    },
    {
      "entropy": 5.691676425933838,
      "epoch": 0.4042015172145497,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004988000418274736,
      "loss": 5.4805,
      "mean_token_accuracy": 0.15906048715114593,
      "num_tokens": 9017952.0,
      "step": 5195
    },
    {
      "entropy": 5.6584619045257565,
      "epoch": 0.40459054658626725,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004987971815448341,
      "loss": 5.4767,
      "mean_token_accuracy": 0.1735681936144829,
      "num_tokens": 9026688.0,
      "step": 5200
    },
    {
      "entropy": 5.610867261886597,
      "epoch": 0.40497957595798484,
      "grad_norm": 1.0,
      "learning_rate": 0.0004987943178664285,
      "loss": 5.4362,
      "mean_token_accuracy": 0.1688633918762207,
      "num_tokens": 9035761.0,
      "step": 5205
    },
    {
      "entropy": 5.649522590637207,
      "epoch": 0.40536860532970237,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004987914507923004,
      "loss": 5.5744,
      "mean_token_accuracy": 0.15566423609852792,
      "num_tokens": 9044510.0,
      "step": 5210
    },
    {
      "entropy": 5.657856607437134,
      "epoch": 0.40575763470141996,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004987885803224931,
      "loss": 5.4375,
      "mean_token_accuracy": 0.1625622183084488,
      "num_tokens": 9053031.0,
      "step": 5215
    },
    {
      "entropy": 5.595879364013672,
      "epoch": 0.40614666407313754,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004987857064570505,
      "loss": 5.4145,
      "mean_token_accuracy": 0.16717963814735412,
      "num_tokens": 9061580.0,
      "step": 5220
    },
    {
      "entropy": 5.6951171875,
      "epoch": 0.4065356934448551,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004987828291960158,
      "loss": 5.5465,
      "mean_token_accuracy": 0.16043514013290405,
      "num_tokens": 9069741.0,
      "step": 5225
    },
    {
      "entropy": 5.774275636672973,
      "epoch": 0.40692472281657266,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000498779948539433,
      "loss": 5.5675,
      "mean_token_accuracy": 0.15380849987268447,
      "num_tokens": 9079214.0,
      "step": 5230
    },
    {
      "entropy": 5.7064296245574955,
      "epoch": 0.4073137521882902,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004987770644873455,
      "loss": 5.5098,
      "mean_token_accuracy": 0.1690395548939705,
      "num_tokens": 9088201.0,
      "step": 5235
    },
    {
      "entropy": 5.667363595962525,
      "epoch": 0.4077027815600078,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004987741770397974,
      "loss": 5.5489,
      "mean_token_accuracy": 0.16119512617588044,
      "num_tokens": 9096829.0,
      "step": 5240
    },
    {
      "entropy": 5.649460363388061,
      "epoch": 0.40809181093172536,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004987712861968323,
      "loss": 5.5065,
      "mean_token_accuracy": 0.16805324405431749,
      "num_tokens": 9105515.0,
      "step": 5245
    },
    {
      "entropy": 5.689359188079834,
      "epoch": 0.4084808403034429,
      "grad_norm": 1.03125,
      "learning_rate": 0.000498768391958494,
      "loss": 5.4731,
      "mean_token_accuracy": 0.1702303782105446,
      "num_tokens": 9113614.0,
      "step": 5250
    },
    {
      "entropy": 5.667102193832397,
      "epoch": 0.4088698696751605,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004987654943248266,
      "loss": 5.5491,
      "mean_token_accuracy": 0.15714831948280333,
      "num_tokens": 9122337.0,
      "step": 5255
    },
    {
      "entropy": 5.698009967803955,
      "epoch": 0.40925889904687807,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004987625932958739,
      "loss": 5.4528,
      "mean_token_accuracy": 0.16894147098064421,
      "num_tokens": 9130417.0,
      "step": 5260
    },
    {
      "entropy": 5.656760168075562,
      "epoch": 0.4096479284185956,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004987596888716801,
      "loss": 5.526,
      "mean_token_accuracy": 0.16036569625139235,
      "num_tokens": 9139317.0,
      "step": 5265
    },
    {
      "entropy": 5.640413570404053,
      "epoch": 0.4100369577903132,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987567810522892,
      "loss": 5.4927,
      "mean_token_accuracy": 0.16545147001743316,
      "num_tokens": 9147808.0,
      "step": 5270
    },
    {
      "entropy": 5.760292911529541,
      "epoch": 0.4104259871620307,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004987538698377451,
      "loss": 5.5919,
      "mean_token_accuracy": 0.15634579882025718,
      "num_tokens": 9157117.0,
      "step": 5275
    },
    {
      "entropy": 5.695358037948608,
      "epoch": 0.4108150165337483,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004987509552280924,
      "loss": 5.4404,
      "mean_token_accuracy": 0.16127725690603256,
      "num_tokens": 9165297.0,
      "step": 5280
    },
    {
      "entropy": 5.688448429107666,
      "epoch": 0.4112040459054659,
      "grad_norm": 1.0625,
      "learning_rate": 0.000498748037223375,
      "loss": 5.5799,
      "mean_token_accuracy": 0.15950480997562408,
      "num_tokens": 9172945.0,
      "step": 5285
    },
    {
      "entropy": 5.684661340713501,
      "epoch": 0.4115930752771834,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004987451158236372,
      "loss": 5.503,
      "mean_token_accuracy": 0.16800302863121033,
      "num_tokens": 9181445.0,
      "step": 5290
    },
    {
      "entropy": 5.619538497924805,
      "epoch": 0.411982104648901,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004987421910289234,
      "loss": 5.3565,
      "mean_token_accuracy": 0.1684710130095482,
      "num_tokens": 9189535.0,
      "step": 5295
    },
    {
      "entropy": 5.6428838729858395,
      "epoch": 0.41237113402061853,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004987392628392781,
      "loss": 5.56,
      "mean_token_accuracy": 0.15575477182865144,
      "num_tokens": 9198328.0,
      "step": 5300
    },
    {
      "entropy": 5.700162172317505,
      "epoch": 0.4127601633923361,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004987363312547456,
      "loss": 5.4656,
      "mean_token_accuracy": 0.16962231397628785,
      "num_tokens": 9206840.0,
      "step": 5305
    },
    {
      "entropy": 5.582920360565185,
      "epoch": 0.4131491927640537,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987333962753703,
      "loss": 5.3699,
      "mean_token_accuracy": 0.16728675216436387,
      "num_tokens": 9215155.0,
      "step": 5310
    },
    {
      "entropy": 5.6116234302520756,
      "epoch": 0.41353822213577124,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004987304579011967,
      "loss": 5.544,
      "mean_token_accuracy": 0.16376285403966903,
      "num_tokens": 9224426.0,
      "step": 5315
    },
    {
      "entropy": 5.72483344078064,
      "epoch": 0.4139272515074888,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004987275161322697,
      "loss": 5.56,
      "mean_token_accuracy": 0.16279975324869156,
      "num_tokens": 9233785.0,
      "step": 5320
    },
    {
      "entropy": 5.665732669830322,
      "epoch": 0.41431628087920636,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004987245709686337,
      "loss": 5.5195,
      "mean_token_accuracy": 0.16579267680644988,
      "num_tokens": 9242168.0,
      "step": 5325
    },
    {
      "entropy": 5.641189622879028,
      "epoch": 0.41470531025092394,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004987216224103334,
      "loss": 5.529,
      "mean_token_accuracy": 0.1659076526761055,
      "num_tokens": 9250862.0,
      "step": 5330
    },
    {
      "entropy": 5.74923677444458,
      "epoch": 0.41509433962264153,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004987186704574136,
      "loss": 5.5401,
      "mean_token_accuracy": 0.16061441153287886,
      "num_tokens": 9259957.0,
      "step": 5335
    },
    {
      "entropy": 5.631114196777344,
      "epoch": 0.41548336899435906,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004987157151099189,
      "loss": 5.4134,
      "mean_token_accuracy": 0.1633456379175186,
      "num_tokens": 9268590.0,
      "step": 5340
    },
    {
      "entropy": 5.67240538597107,
      "epoch": 0.41587239836607665,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004987127563678945,
      "loss": 5.4773,
      "mean_token_accuracy": 0.16453050673007966,
      "num_tokens": 9276872.0,
      "step": 5345
    },
    {
      "entropy": 5.608943080902099,
      "epoch": 0.4162614277377942,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004987097942313852,
      "loss": 5.3563,
      "mean_token_accuracy": 0.17066091150045395,
      "num_tokens": 9284556.0,
      "step": 5350
    },
    {
      "entropy": 5.595340871810913,
      "epoch": 0.41665045710951176,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004987068287004355,
      "loss": 5.5191,
      "mean_token_accuracy": 0.16076190620660782,
      "num_tokens": 9293315.0,
      "step": 5355
    },
    {
      "entropy": 5.683136224746704,
      "epoch": 0.41703948648122935,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004987038597750909,
      "loss": 5.4955,
      "mean_token_accuracy": 0.1678859144449234,
      "num_tokens": 9302923.0,
      "step": 5360
    },
    {
      "entropy": 5.685339021682739,
      "epoch": 0.4174285158529469,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004987008874553964,
      "loss": 5.4286,
      "mean_token_accuracy": 0.16690774857997895,
      "num_tokens": 9311388.0,
      "step": 5365
    },
    {
      "entropy": 5.621911287307739,
      "epoch": 0.41781754522466447,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004986979117413969,
      "loss": 5.3816,
      "mean_token_accuracy": 0.16745009273290634,
      "num_tokens": 9319880.0,
      "step": 5370
    },
    {
      "entropy": 5.619531869888306,
      "epoch": 0.41820657459638205,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004986949326331376,
      "loss": 5.5176,
      "mean_token_accuracy": 0.1664719372987747,
      "num_tokens": 9328374.0,
      "step": 5375
    },
    {
      "entropy": 5.740171337127686,
      "epoch": 0.4185956039680996,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004986919501306638,
      "loss": 5.4799,
      "mean_token_accuracy": 0.17262143194675444,
      "num_tokens": 9336428.0,
      "step": 5380
    },
    {
      "entropy": 5.612084150314331,
      "epoch": 0.41898463333981717,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004986889642340207,
      "loss": 5.574,
      "mean_token_accuracy": 0.16087047159671783,
      "num_tokens": 9345794.0,
      "step": 5385
    },
    {
      "entropy": 5.740064382553101,
      "epoch": 0.4193736627115347,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004986859749432536,
      "loss": 5.539,
      "mean_token_accuracy": 0.16427928507328032,
      "num_tokens": 9354724.0,
      "step": 5390
    },
    {
      "entropy": 5.720560503005982,
      "epoch": 0.4197626920832523,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986829822584078,
      "loss": 5.5483,
      "mean_token_accuracy": 0.15803782045841216,
      "num_tokens": 9364008.0,
      "step": 5395
    },
    {
      "entropy": 5.6430220127105715,
      "epoch": 0.4201517214549699,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986799861795289,
      "loss": 5.5,
      "mean_token_accuracy": 0.17185320258140563,
      "num_tokens": 9372409.0,
      "step": 5400
    },
    {
      "entropy": 5.666001272201538,
      "epoch": 0.4205407508266874,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004986769867066622,
      "loss": 5.4575,
      "mean_token_accuracy": 0.16317709684371948,
      "num_tokens": 9381172.0,
      "step": 5405
    },
    {
      "entropy": 5.644277095794678,
      "epoch": 0.420929780198405,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004986739838398532,
      "loss": 5.4765,
      "mean_token_accuracy": 0.16965698897838594,
      "num_tokens": 9389245.0,
      "step": 5410
    },
    {
      "entropy": 5.645105075836182,
      "epoch": 0.4213188095701225,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004986709775791475,
      "loss": 5.5055,
      "mean_token_accuracy": 0.16165761649608612,
      "num_tokens": 9397954.0,
      "step": 5415
    },
    {
      "entropy": 5.673592185974121,
      "epoch": 0.4217078389418401,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004986679679245907,
      "loss": 5.6166,
      "mean_token_accuracy": 0.15879604816436768,
      "num_tokens": 9406986.0,
      "step": 5420
    },
    {
      "entropy": 5.686118459701538,
      "epoch": 0.4220968683135577,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004986649548762286,
      "loss": 5.5508,
      "mean_token_accuracy": 0.1635198250412941,
      "num_tokens": 9415163.0,
      "step": 5425
    },
    {
      "entropy": 5.636553859710693,
      "epoch": 0.4224858976852752,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004986619384341066,
      "loss": 5.3748,
      "mean_token_accuracy": 0.18211421966552735,
      "num_tokens": 9423422.0,
      "step": 5430
    },
    {
      "entropy": 5.578596305847168,
      "epoch": 0.4228749270569928,
      "grad_norm": 1.0,
      "learning_rate": 0.0004986589185982708,
      "loss": 5.4068,
      "mean_token_accuracy": 0.17945402711629868,
      "num_tokens": 9431349.0,
      "step": 5435
    },
    {
      "entropy": 5.67623233795166,
      "epoch": 0.42326395642871034,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004986558953687671,
      "loss": 5.5172,
      "mean_token_accuracy": 0.1618877172470093,
      "num_tokens": 9439715.0,
      "step": 5440
    },
    {
      "entropy": 5.683085584640503,
      "epoch": 0.42365298580042793,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004986528687456409,
      "loss": 5.5154,
      "mean_token_accuracy": 0.1649433985352516,
      "num_tokens": 9448216.0,
      "step": 5445
    },
    {
      "entropy": 5.703312730789184,
      "epoch": 0.4240420151721455,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004986498387289384,
      "loss": 5.456,
      "mean_token_accuracy": 0.16703744381666183,
      "num_tokens": 9456346.0,
      "step": 5450
    },
    {
      "entropy": 5.654667711257934,
      "epoch": 0.42443104454386305,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004986468053187058,
      "loss": 5.5211,
      "mean_token_accuracy": 0.16369810104370117,
      "num_tokens": 9465476.0,
      "step": 5455
    },
    {
      "entropy": 5.685988807678223,
      "epoch": 0.42482007391558063,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004986437685149887,
      "loss": 5.5787,
      "mean_token_accuracy": 0.1575814224779606,
      "num_tokens": 9474582.0,
      "step": 5460
    },
    {
      "entropy": 5.694247817993164,
      "epoch": 0.4252091032872982,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986407283178334,
      "loss": 5.4598,
      "mean_token_accuracy": 0.16146390438079833,
      "num_tokens": 9482801.0,
      "step": 5465
    },
    {
      "entropy": 5.597379589080811,
      "epoch": 0.42559813265901575,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004986376847272861,
      "loss": 5.4963,
      "mean_token_accuracy": 0.16186577677726746,
      "num_tokens": 9490894.0,
      "step": 5470
    },
    {
      "entropy": 5.625309896469116,
      "epoch": 0.42598716203073334,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004986346377433929,
      "loss": 5.4518,
      "mean_token_accuracy": 0.16440507918596267,
      "num_tokens": 9499414.0,
      "step": 5475
    },
    {
      "entropy": 5.6179710865020756,
      "epoch": 0.42637619140245087,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004986315873661999,
      "loss": 5.4118,
      "mean_token_accuracy": 0.1651856407523155,
      "num_tokens": 9508945.0,
      "step": 5480
    },
    {
      "entropy": 5.6642967700958256,
      "epoch": 0.42676522077416845,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004986285335957536,
      "loss": 5.4819,
      "mean_token_accuracy": 0.1687757909297943,
      "num_tokens": 9517833.0,
      "step": 5485
    },
    {
      "entropy": 5.678032922744751,
      "epoch": 0.42715425014588604,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004986254764321002,
      "loss": 5.6181,
      "mean_token_accuracy": 0.1571633294224739,
      "num_tokens": 9526551.0,
      "step": 5490
    },
    {
      "entropy": 5.620652914047241,
      "epoch": 0.42754327951760357,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004986224158752861,
      "loss": 5.3924,
      "mean_token_accuracy": 0.1691167712211609,
      "num_tokens": 9535224.0,
      "step": 5495
    },
    {
      "entropy": 5.6598100662231445,
      "epoch": 0.42793230888932116,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004986193519253578,
      "loss": 5.5084,
      "mean_token_accuracy": 0.1580831378698349,
      "num_tokens": 9544466.0,
      "step": 5500
    },
    {
      "entropy": 5.615001058578491,
      "epoch": 0.4283213382610387,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004986162845823618,
      "loss": 5.5103,
      "mean_token_accuracy": 0.1643732085824013,
      "num_tokens": 9553520.0,
      "step": 5505
    },
    {
      "entropy": 5.632220363616943,
      "epoch": 0.4287103676327563,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004986132138463446,
      "loss": 5.4095,
      "mean_token_accuracy": 0.17255716621875763,
      "num_tokens": 9562053.0,
      "step": 5510
    },
    {
      "entropy": 5.6332581520080565,
      "epoch": 0.42909939700447386,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004986101397173528,
      "loss": 5.3992,
      "mean_token_accuracy": 0.17297570407390594,
      "num_tokens": 9570632.0,
      "step": 5515
    },
    {
      "entropy": 5.639089155197143,
      "epoch": 0.4294884263761914,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000498607062195433,
      "loss": 5.4017,
      "mean_token_accuracy": 0.1708621010184288,
      "num_tokens": 9578739.0,
      "step": 5520
    },
    {
      "entropy": 5.579420709609986,
      "epoch": 0.429877455747909,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000498603981280632,
      "loss": 5.4379,
      "mean_token_accuracy": 0.1644877687096596,
      "num_tokens": 9587314.0,
      "step": 5525
    },
    {
      "entropy": 5.653287601470947,
      "epoch": 0.4302664851196265,
      "grad_norm": 1.0,
      "learning_rate": 0.0004986008969729964,
      "loss": 5.5102,
      "mean_token_accuracy": 0.16773233115673064,
      "num_tokens": 9595954.0,
      "step": 5530
    },
    {
      "entropy": 5.689511823654175,
      "epoch": 0.4306555144913441,
      "grad_norm": 1.0,
      "learning_rate": 0.0004985978092725731,
      "loss": 5.5158,
      "mean_token_accuracy": 0.15608498603105544,
      "num_tokens": 9604767.0,
      "step": 5535
    },
    {
      "entropy": 5.598193216323852,
      "epoch": 0.4310445438630617,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000498594718179409,
      "loss": 5.38,
      "mean_token_accuracy": 0.16736597567796707,
      "num_tokens": 9613173.0,
      "step": 5540
    },
    {
      "entropy": 5.671756887435913,
      "epoch": 0.4314335732347792,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004985916236935508,
      "loss": 5.4646,
      "mean_token_accuracy": 0.16545387953519822,
      "num_tokens": 9622777.0,
      "step": 5545
    },
    {
      "entropy": 5.616490173339844,
      "epoch": 0.4318226026064968,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004985885258150458,
      "loss": 5.4274,
      "mean_token_accuracy": 0.16021243780851363,
      "num_tokens": 9631019.0,
      "step": 5550
    },
    {
      "entropy": 5.621186923980713,
      "epoch": 0.43221163197821433,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004985854245439408,
      "loss": 5.4839,
      "mean_token_accuracy": 0.159197573363781,
      "num_tokens": 9639797.0,
      "step": 5555
    },
    {
      "entropy": 5.6225427150726315,
      "epoch": 0.4326006613499319,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004985823198802827,
      "loss": 5.4752,
      "mean_token_accuracy": 0.1676049068570137,
      "num_tokens": 9648393.0,
      "step": 5560
    },
    {
      "entropy": 5.6570898532867435,
      "epoch": 0.4329896907216495,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004985792118241188,
      "loss": 5.4623,
      "mean_token_accuracy": 0.16745230555534363,
      "num_tokens": 9657260.0,
      "step": 5565
    },
    {
      "entropy": 5.624749183654785,
      "epoch": 0.43337872009336703,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004985761003754961,
      "loss": 5.5217,
      "mean_token_accuracy": 0.16828898340463638,
      "num_tokens": 9665883.0,
      "step": 5570
    },
    {
      "entropy": 5.706521701812744,
      "epoch": 0.4337677494650846,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004985729855344622,
      "loss": 5.5316,
      "mean_token_accuracy": 0.16746796816587448,
      "num_tokens": 9674100.0,
      "step": 5575
    },
    {
      "entropy": 5.671209716796875,
      "epoch": 0.4341567788368022,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000498569867301064,
      "loss": 5.4968,
      "mean_token_accuracy": 0.16826301962137222,
      "num_tokens": 9683236.0,
      "step": 5580
    },
    {
      "entropy": 5.602039051055908,
      "epoch": 0.43454580820851973,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004985667456753489,
      "loss": 5.4773,
      "mean_token_accuracy": 0.16572456955909728,
      "num_tokens": 9692076.0,
      "step": 5585
    },
    {
      "entropy": 5.674462890625,
      "epoch": 0.4349348375802373,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004985636206573642,
      "loss": 5.5142,
      "mean_token_accuracy": 0.16270209848880768,
      "num_tokens": 9700266.0,
      "step": 5590
    },
    {
      "entropy": 5.63533205986023,
      "epoch": 0.43532386695195485,
      "grad_norm": 1.0,
      "learning_rate": 0.0004985604922471575,
      "loss": 5.456,
      "mean_token_accuracy": 0.16878321170806884,
      "num_tokens": 9708643.0,
      "step": 5595
    },
    {
      "entropy": 5.650230789184571,
      "epoch": 0.43571289632367244,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004985573604447761,
      "loss": 5.5071,
      "mean_token_accuracy": 0.16387132704257965,
      "num_tokens": 9717459.0,
      "step": 5600
    },
    {
      "entropy": 5.608091926574707,
      "epoch": 0.43610192569539,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004985542252502676,
      "loss": 5.4057,
      "mean_token_accuracy": 0.1740733951330185,
      "num_tokens": 9726251.0,
      "step": 5605
    },
    {
      "entropy": 5.539948225021362,
      "epoch": 0.43649095506710756,
      "grad_norm": 1.0,
      "learning_rate": 0.0004985510866636796,
      "loss": 5.4086,
      "mean_token_accuracy": 0.16527086347341538,
      "num_tokens": 9735327.0,
      "step": 5610
    },
    {
      "entropy": 5.63954119682312,
      "epoch": 0.43687998443882514,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004985479446850596,
      "loss": 5.5749,
      "mean_token_accuracy": 0.15904381573200227,
      "num_tokens": 9743642.0,
      "step": 5615
    },
    {
      "entropy": 5.677852296829224,
      "epoch": 0.4372690138105427,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004985447993144554,
      "loss": 5.4357,
      "mean_token_accuracy": 0.1692972019314766,
      "num_tokens": 9752235.0,
      "step": 5620
    },
    {
      "entropy": 5.688651418685913,
      "epoch": 0.43765804318226026,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004985416505519147,
      "loss": 5.5148,
      "mean_token_accuracy": 0.15892294347286223,
      "num_tokens": 9761012.0,
      "step": 5625
    },
    {
      "entropy": 5.623238372802734,
      "epoch": 0.43804707255397785,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004985384983974853,
      "loss": 5.471,
      "mean_token_accuracy": 0.17131368666887284,
      "num_tokens": 9769810.0,
      "step": 5630
    },
    {
      "entropy": 5.633467721939087,
      "epoch": 0.4384361019256954,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004985353428512148,
      "loss": 5.4165,
      "mean_token_accuracy": 0.16389521211385727,
      "num_tokens": 9778186.0,
      "step": 5635
    },
    {
      "entropy": 5.591325139999389,
      "epoch": 0.43882513129741296,
      "grad_norm": 1.0,
      "learning_rate": 0.0004985321839131514,
      "loss": 5.4767,
      "mean_token_accuracy": 0.16914807856082917,
      "num_tokens": 9786456.0,
      "step": 5640
    },
    {
      "entropy": 5.629720783233642,
      "epoch": 0.4392141606691305,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004985290215833428,
      "loss": 5.4796,
      "mean_token_accuracy": 0.16214304715394973,
      "num_tokens": 9794613.0,
      "step": 5645
    },
    {
      "entropy": 5.7324135303497314,
      "epoch": 0.4396031900408481,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004985258558618372,
      "loss": 5.3918,
      "mean_token_accuracy": 0.17065405696630478,
      "num_tokens": 9803388.0,
      "step": 5650
    },
    {
      "entropy": 5.626570796966552,
      "epoch": 0.43999221941256567,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004985226867486825,
      "loss": 5.4745,
      "mean_token_accuracy": 0.16745347231626512,
      "num_tokens": 9812450.0,
      "step": 5655
    },
    {
      "entropy": 5.62529616355896,
      "epoch": 0.4403812487842832,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004985195142439267,
      "loss": 5.5514,
      "mean_token_accuracy": 0.1591256856918335,
      "num_tokens": 9821377.0,
      "step": 5660
    },
    {
      "entropy": 5.661211919784546,
      "epoch": 0.4407702781560008,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004985163383476181,
      "loss": 5.455,
      "mean_token_accuracy": 0.16048079878091812,
      "num_tokens": 9829761.0,
      "step": 5665
    },
    {
      "entropy": 5.668930578231811,
      "epoch": 0.44115930752771837,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004985131590598048,
      "loss": 5.5249,
      "mean_token_accuracy": 0.15734247416257857,
      "num_tokens": 9838921.0,
      "step": 5670
    },
    {
      "entropy": 5.592563962936401,
      "epoch": 0.4415483368994359,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004985099763805352,
      "loss": 5.3575,
      "mean_token_accuracy": 0.17540861517190934,
      "num_tokens": 9846822.0,
      "step": 5675
    },
    {
      "entropy": 5.686899566650391,
      "epoch": 0.4419373662711535,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004985067903098574,
      "loss": 5.5683,
      "mean_token_accuracy": 0.1694743275642395,
      "num_tokens": 9855951.0,
      "step": 5680
    },
    {
      "entropy": 5.6965892791748045,
      "epoch": 0.442326395642871,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004985036008478197,
      "loss": 5.5217,
      "mean_token_accuracy": 0.16461412459611893,
      "num_tokens": 9864875.0,
      "step": 5685
    },
    {
      "entropy": 5.680837154388428,
      "epoch": 0.4427154250145886,
      "grad_norm": 1.125,
      "learning_rate": 0.0004985004079944707,
      "loss": 5.4781,
      "mean_token_accuracy": 0.1686537191271782,
      "num_tokens": 9873785.0,
      "step": 5690
    },
    {
      "entropy": 5.659858894348145,
      "epoch": 0.4431044543863062,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004984972117498587,
      "loss": 5.495,
      "mean_token_accuracy": 0.16421714872121812,
      "num_tokens": 9883214.0,
      "step": 5695
    },
    {
      "entropy": 5.608301591873169,
      "epoch": 0.4434934837580237,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004984940121140323,
      "loss": 5.4572,
      "mean_token_accuracy": 0.16669983863830568,
      "num_tokens": 9891346.0,
      "step": 5700
    },
    {
      "entropy": 5.5851539134979244,
      "epoch": 0.4438825131297413,
      "grad_norm": 1.0625,
      "learning_rate": 0.00049849080908704,
      "loss": 5.3759,
      "mean_token_accuracy": 0.16997243613004684,
      "num_tokens": 9900646.0,
      "step": 5705
    },
    {
      "entropy": 5.682978773117066,
      "epoch": 0.44427154250145884,
      "grad_norm": 1.0,
      "learning_rate": 0.0004984876026689302,
      "loss": 5.5274,
      "mean_token_accuracy": 0.16418917775154113,
      "num_tokens": 9910851.0,
      "step": 5710
    },
    {
      "entropy": 5.695684194564819,
      "epoch": 0.4446605718731764,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000498484392859752,
      "loss": 5.533,
      "mean_token_accuracy": 0.16463363021612168,
      "num_tokens": 9919919.0,
      "step": 5715
    },
    {
      "entropy": 5.650063467025757,
      "epoch": 0.445049601244894,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004984811796595538,
      "loss": 5.4689,
      "mean_token_accuracy": 0.16352860778570175,
      "num_tokens": 9928831.0,
      "step": 5720
    },
    {
      "entropy": 5.606766128540039,
      "epoch": 0.44543863061661154,
      "grad_norm": 1.0,
      "learning_rate": 0.0004984779630683843,
      "loss": 5.4252,
      "mean_token_accuracy": 0.17202684581279754,
      "num_tokens": 9936840.0,
      "step": 5725
    },
    {
      "entropy": 5.628744173049927,
      "epoch": 0.44582765998832913,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004984747430862924,
      "loss": 5.5005,
      "mean_token_accuracy": 0.1661605268716812,
      "num_tokens": 9945845.0,
      "step": 5730
    },
    {
      "entropy": 5.688571310043335,
      "epoch": 0.44621668936004666,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004984715197133271,
      "loss": 5.4884,
      "mean_token_accuracy": 0.1676606684923172,
      "num_tokens": 9955176.0,
      "step": 5735
    },
    {
      "entropy": 5.541508436203003,
      "epoch": 0.44660571873176425,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004984682929495371,
      "loss": 5.3768,
      "mean_token_accuracy": 0.1689297780394554,
      "num_tokens": 9963182.0,
      "step": 5740
    },
    {
      "entropy": 5.640884494781494,
      "epoch": 0.44699474810348183,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004984650627949715,
      "loss": 5.5316,
      "mean_token_accuracy": 0.1640244409441948,
      "num_tokens": 9971239.0,
      "step": 5745
    },
    {
      "entropy": 5.669719123840332,
      "epoch": 0.44738377747519936,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004984618292496792,
      "loss": 5.4698,
      "mean_token_accuracy": 0.1678232803940773,
      "num_tokens": 9980285.0,
      "step": 5750
    },
    {
      "entropy": 5.658495330810547,
      "epoch": 0.44777280684691695,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004984585923137093,
      "loss": 5.4804,
      "mean_token_accuracy": 0.16718869507312775,
      "num_tokens": 9988922.0,
      "step": 5755
    },
    {
      "entropy": 5.6383864402771,
      "epoch": 0.4481618362186345,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000498455351987111,
      "loss": 5.5701,
      "mean_token_accuracy": 0.16479225903749467,
      "num_tokens": 9998538.0,
      "step": 5760
    },
    {
      "entropy": 5.753770446777343,
      "epoch": 0.44855086559035207,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004984521082699333,
      "loss": 5.4682,
      "mean_token_accuracy": 0.16479181349277497,
      "num_tokens": 10006646.0,
      "step": 5765
    },
    {
      "entropy": 5.636715221405029,
      "epoch": 0.44893989496206965,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004984488611622256,
      "loss": 5.4514,
      "mean_token_accuracy": 0.16885284036397935,
      "num_tokens": 10016511.0,
      "step": 5770
    },
    {
      "entropy": 5.5878339290618895,
      "epoch": 0.4493289243337872,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000498445610664037,
      "loss": 5.3602,
      "mean_token_accuracy": 0.1687435194849968,
      "num_tokens": 10025223.0,
      "step": 5775
    },
    {
      "entropy": 5.569753217697143,
      "epoch": 0.44971795370550477,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000498442356775417,
      "loss": 5.4556,
      "mean_token_accuracy": 0.16437877714633942,
      "num_tokens": 10033678.0,
      "step": 5780
    },
    {
      "entropy": 5.620095252990723,
      "epoch": 0.45010698307722236,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004984390994964148,
      "loss": 5.3746,
      "mean_token_accuracy": 0.17267125844955444,
      "num_tokens": 10042028.0,
      "step": 5785
    },
    {
      "entropy": 5.655372524261475,
      "epoch": 0.4504960124489399,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00049843583882708,
      "loss": 5.5167,
      "mean_token_accuracy": 0.16230278611183166,
      "num_tokens": 10050758.0,
      "step": 5790
    },
    {
      "entropy": 5.563958215713501,
      "epoch": 0.4508850418206575,
      "grad_norm": 1.0625,
      "learning_rate": 0.000498432574767462,
      "loss": 5.3073,
      "mean_token_accuracy": 0.18065344244241716,
      "num_tokens": 10058688.0,
      "step": 5795
    },
    {
      "entropy": 5.552063608169556,
      "epoch": 0.451274071192375,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004984293073176103,
      "loss": 5.3894,
      "mean_token_accuracy": 0.1747409075498581,
      "num_tokens": 10067302.0,
      "step": 5800
    },
    {
      "entropy": 5.559607934951782,
      "epoch": 0.4516631005640926,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004984260364775744,
      "loss": 5.3891,
      "mean_token_accuracy": 0.17128931134939193,
      "num_tokens": 10076424.0,
      "step": 5805
    },
    {
      "entropy": 5.66033034324646,
      "epoch": 0.4520521299358102,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000498422762247404,
      "loss": 5.5288,
      "mean_token_accuracy": 0.16378835141658782,
      "num_tokens": 10086307.0,
      "step": 5810
    },
    {
      "entropy": 5.618446731567383,
      "epoch": 0.4524411593075277,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004984194846271489,
      "loss": 5.3937,
      "mean_token_accuracy": 0.16565483212471008,
      "num_tokens": 10094954.0,
      "step": 5815
    },
    {
      "entropy": 5.598624515533447,
      "epoch": 0.4528301886792453,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004984162036168586,
      "loss": 5.4779,
      "mean_token_accuracy": 0.1666440784931183,
      "num_tokens": 10104401.0,
      "step": 5820
    },
    {
      "entropy": 5.575947904586792,
      "epoch": 0.4532192180509628,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004984129192165831,
      "loss": 5.4902,
      "mean_token_accuracy": 0.16404329389333724,
      "num_tokens": 10112596.0,
      "step": 5825
    },
    {
      "entropy": 5.663758945465088,
      "epoch": 0.4536082474226804,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004984096314263722,
      "loss": 5.5278,
      "mean_token_accuracy": 0.1639996126294136,
      "num_tokens": 10121311.0,
      "step": 5830
    },
    {
      "entropy": 5.596688652038575,
      "epoch": 0.453997276794398,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004984063402462757,
      "loss": 5.3286,
      "mean_token_accuracy": 0.17286605685949324,
      "num_tokens": 10130010.0,
      "step": 5835
    },
    {
      "entropy": 5.582390308380127,
      "epoch": 0.45438630616611553,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004984030456763435,
      "loss": 5.4316,
      "mean_token_accuracy": 0.17585542798042297,
      "num_tokens": 10138473.0,
      "step": 5840
    },
    {
      "entropy": 5.72431173324585,
      "epoch": 0.4547753355378331,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004983997477166257,
      "loss": 5.5141,
      "mean_token_accuracy": 0.1647747352719307,
      "num_tokens": 10146843.0,
      "step": 5845
    },
    {
      "entropy": 5.543897390365601,
      "epoch": 0.45516436490955064,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004983964463671723,
      "loss": 5.3779,
      "mean_token_accuracy": 0.17528259009122849,
      "num_tokens": 10155319.0,
      "step": 5850
    },
    {
      "entropy": 5.615078401565552,
      "epoch": 0.45555339428126823,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004983931416280334,
      "loss": 5.5535,
      "mean_token_accuracy": 0.1545005477964878,
      "num_tokens": 10164308.0,
      "step": 5855
    },
    {
      "entropy": 5.696380519866944,
      "epoch": 0.4559424236529858,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004983898334992591,
      "loss": 5.5508,
      "mean_token_accuracy": 0.15858444571495056,
      "num_tokens": 10173173.0,
      "step": 5860
    },
    {
      "entropy": 5.656243467330933,
      "epoch": 0.45633145302470335,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004983865219808998,
      "loss": 5.4625,
      "mean_token_accuracy": 0.16409115344285966,
      "num_tokens": 10181376.0,
      "step": 5865
    },
    {
      "entropy": 5.71471905708313,
      "epoch": 0.45672048239642093,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004983832070730055,
      "loss": 5.5752,
      "mean_token_accuracy": 0.16178102046251297,
      "num_tokens": 10189701.0,
      "step": 5870
    },
    {
      "entropy": 5.6418993949890135,
      "epoch": 0.4571095117681385,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004983798887756265,
      "loss": 5.4219,
      "mean_token_accuracy": 0.16949813663959504,
      "num_tokens": 10198721.0,
      "step": 5875
    },
    {
      "entropy": 5.538687229156494,
      "epoch": 0.45749854113985605,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004983765670888133,
      "loss": 5.3188,
      "mean_token_accuracy": 0.172306090593338,
      "num_tokens": 10206428.0,
      "step": 5880
    },
    {
      "entropy": 5.504544305801391,
      "epoch": 0.45788757051157364,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004983732420126163,
      "loss": 5.4705,
      "mean_token_accuracy": 0.16727178543806076,
      "num_tokens": 10214941.0,
      "step": 5885
    },
    {
      "entropy": 5.648996162414551,
      "epoch": 0.45827659988329117,
      "grad_norm": 1.0,
      "learning_rate": 0.0004983699135470858,
      "loss": 5.4617,
      "mean_token_accuracy": 0.16921409964561462,
      "num_tokens": 10223491.0,
      "step": 5890
    },
    {
      "entropy": 5.669284391403198,
      "epoch": 0.45866562925500876,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004983665816922723,
      "loss": 5.3963,
      "mean_token_accuracy": 0.171782548725605,
      "num_tokens": 10231820.0,
      "step": 5895
    },
    {
      "entropy": 5.577320098876953,
      "epoch": 0.45905465862672634,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004983632464482267,
      "loss": 5.3941,
      "mean_token_accuracy": 0.17510838955640792,
      "num_tokens": 10239781.0,
      "step": 5900
    },
    {
      "entropy": 5.594802141189575,
      "epoch": 0.4594436879984439,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004983599078149991,
      "loss": 5.4106,
      "mean_token_accuracy": 0.1815456435084343,
      "num_tokens": 10248093.0,
      "step": 5905
    },
    {
      "entropy": 5.65915994644165,
      "epoch": 0.45983271737016146,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004983565657926405,
      "loss": 5.4403,
      "mean_token_accuracy": 0.1665576696395874,
      "num_tokens": 10256370.0,
      "step": 5910
    },
    {
      "entropy": 5.555393409729004,
      "epoch": 0.460221746741879,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004983532203812017,
      "loss": 5.4115,
      "mean_token_accuracy": 0.16040660440921783,
      "num_tokens": 10264924.0,
      "step": 5915
    },
    {
      "entropy": 5.59886212348938,
      "epoch": 0.4606107761135966,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004983498715807331,
      "loss": 5.3484,
      "mean_token_accuracy": 0.16863510608673096,
      "num_tokens": 10273488.0,
      "step": 5920
    },
    {
      "entropy": 5.64832444190979,
      "epoch": 0.46099980548531416,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004983465193912857,
      "loss": 5.4405,
      "mean_token_accuracy": 0.16802628189325333,
      "num_tokens": 10281534.0,
      "step": 5925
    },
    {
      "entropy": 5.662678289413452,
      "epoch": 0.4613888348570317,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004983431638129104,
      "loss": 5.4783,
      "mean_token_accuracy": 0.16641798838973046,
      "num_tokens": 10289613.0,
      "step": 5930
    },
    {
      "entropy": 5.655657577514648,
      "epoch": 0.4617778642287493,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004983398048456581,
      "loss": 5.5218,
      "mean_token_accuracy": 0.17005051374435426,
      "num_tokens": 10299027.0,
      "step": 5935
    },
    {
      "entropy": 5.671428251266479,
      "epoch": 0.4621668936004668,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004983364424895796,
      "loss": 5.4903,
      "mean_token_accuracy": 0.1599871724843979,
      "num_tokens": 10308727.0,
      "step": 5940
    },
    {
      "entropy": 5.694900131225586,
      "epoch": 0.4625559229721844,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004983330767447262,
      "loss": 5.5063,
      "mean_token_accuracy": 0.1694519266486168,
      "num_tokens": 10318129.0,
      "step": 5945
    },
    {
      "entropy": 5.56030125617981,
      "epoch": 0.462944952343902,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004983297076111489,
      "loss": 5.3458,
      "mean_token_accuracy": 0.17852959036827087,
      "num_tokens": 10326504.0,
      "step": 5950
    },
    {
      "entropy": 5.574175643920898,
      "epoch": 0.4633339817156195,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004983263350888987,
      "loss": 5.4123,
      "mean_token_accuracy": 0.17486206889152528,
      "num_tokens": 10334585.0,
      "step": 5955
    },
    {
      "entropy": 5.61399130821228,
      "epoch": 0.4637230110873371,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004983229591780268,
      "loss": 5.4187,
      "mean_token_accuracy": 0.1757299780845642,
      "num_tokens": 10343328.0,
      "step": 5960
    },
    {
      "entropy": 5.5717075824737545,
      "epoch": 0.46411204045905463,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004983195798785844,
      "loss": 5.4523,
      "mean_token_accuracy": 0.1647307336330414,
      "num_tokens": 10352289.0,
      "step": 5965
    },
    {
      "entropy": 5.681837654113769,
      "epoch": 0.4645010698307722,
      "grad_norm": 1.0,
      "learning_rate": 0.0004983161971906228,
      "loss": 5.5663,
      "mean_token_accuracy": 0.15927473157644273,
      "num_tokens": 10360847.0,
      "step": 5970
    },
    {
      "entropy": 5.655223321914673,
      "epoch": 0.4648900992024898,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004983128111141935,
      "loss": 5.4756,
      "mean_token_accuracy": 0.1694716840982437,
      "num_tokens": 10368911.0,
      "step": 5975
    },
    {
      "entropy": 5.585180425643921,
      "epoch": 0.46527912857420733,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004983094216493475,
      "loss": 5.4119,
      "mean_token_accuracy": 0.17236740589141847,
      "num_tokens": 10377257.0,
      "step": 5980
    },
    {
      "entropy": 5.617059707641602,
      "epoch": 0.4656681579459249,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004983060287961367,
      "loss": 5.4788,
      "mean_token_accuracy": 0.16893793195486068,
      "num_tokens": 10386167.0,
      "step": 5985
    },
    {
      "entropy": 5.617970180511475,
      "epoch": 0.4660571873176425,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004983026325546123,
      "loss": 5.3693,
      "mean_token_accuracy": 0.16516103446483613,
      "num_tokens": 10395802.0,
      "step": 5990
    },
    {
      "entropy": 5.608391046524048,
      "epoch": 0.46644621668936004,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004982992329248257,
      "loss": 5.5062,
      "mean_token_accuracy": 0.16500955075025558,
      "num_tokens": 10404836.0,
      "step": 5995
    },
    {
      "entropy": 5.703184175491333,
      "epoch": 0.4668352460610776,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004982958299068289,
      "loss": 5.4583,
      "mean_token_accuracy": 0.16742442846298217,
      "num_tokens": 10413394.0,
      "step": 6000
    },
    {
      "epoch": 0.4668352460610776,
      "eval_entropy": 5.455045991866375,
      "eval_loss": 5.484153747558594,
      "eval_mean_token_accuracy": 0.17452393070049763,
      "eval_num_tokens": 10413394.0,
      "eval_runtime": 21.7324,
      "eval_samples_per_second": 1912.264,
      "eval_steps_per_second": 239.044,
      "step": 6000
    },
    {
      "entropy": 5.57760214805603,
      "epoch": 0.46722427543279516,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000498292423500673,
      "loss": 5.3674,
      "mean_token_accuracy": 0.1734383910894394,
      "num_tokens": 10421916.0,
      "step": 6005
    },
    {
      "entropy": 5.5814759731292725,
      "epoch": 0.46761330480451274,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004982890137064102,
      "loss": 5.446,
      "mean_token_accuracy": 0.1695609897375107,
      "num_tokens": 10430703.0,
      "step": 6010
    },
    {
      "entropy": 5.64775276184082,
      "epoch": 0.46800233417623033,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000498285600524092,
      "loss": 5.4861,
      "mean_token_accuracy": 0.16832491159439086,
      "num_tokens": 10439523.0,
      "step": 6015
    },
    {
      "entropy": 5.618974447250366,
      "epoch": 0.46839136354794786,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004982821839537701,
      "loss": 5.3928,
      "mean_token_accuracy": 0.17448728680610656,
      "num_tokens": 10447777.0,
      "step": 6020
    },
    {
      "entropy": 5.59127516746521,
      "epoch": 0.46878039291966545,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004982787639954966,
      "loss": 5.3757,
      "mean_token_accuracy": 0.17489583045244217,
      "num_tokens": 10456105.0,
      "step": 6025
    },
    {
      "entropy": 5.6200073719024655,
      "epoch": 0.469169422291383,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004982753406493232,
      "loss": 5.4247,
      "mean_token_accuracy": 0.16724000573158265,
      "num_tokens": 10464590.0,
      "step": 6030
    },
    {
      "entropy": 5.65100588798523,
      "epoch": 0.46955845166310056,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004982719139153018,
      "loss": 5.4373,
      "mean_token_accuracy": 0.164061538875103,
      "num_tokens": 10472702.0,
      "step": 6035
    },
    {
      "entropy": 5.537286376953125,
      "epoch": 0.46994748103481815,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004982684837934845,
      "loss": 5.3764,
      "mean_token_accuracy": 0.17117306143045424,
      "num_tokens": 10480682.0,
      "step": 6040
    },
    {
      "entropy": 5.6783849716186525,
      "epoch": 0.4703365104065357,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004982650502839234,
      "loss": 5.4674,
      "mean_token_accuracy": 0.16657331436872483,
      "num_tokens": 10488873.0,
      "step": 6045
    },
    {
      "entropy": 5.630449104309082,
      "epoch": 0.47072553977825327,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004982616133866705,
      "loss": 5.4436,
      "mean_token_accuracy": 0.17212734669446944,
      "num_tokens": 10497077.0,
      "step": 6050
    },
    {
      "entropy": 5.621734952926635,
      "epoch": 0.4711145691499708,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004982581731017779,
      "loss": 5.4275,
      "mean_token_accuracy": 0.16766708344221115,
      "num_tokens": 10506072.0,
      "step": 6055
    },
    {
      "entropy": 5.598079490661621,
      "epoch": 0.4715035985216884,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004982547294292979,
      "loss": 5.4853,
      "mean_token_accuracy": 0.16255218386650086,
      "num_tokens": 10515175.0,
      "step": 6060
    },
    {
      "entropy": 5.5633021831512455,
      "epoch": 0.47189262789340597,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004982512823692828,
      "loss": 5.3765,
      "mean_token_accuracy": 0.16666678339242935,
      "num_tokens": 10524800.0,
      "step": 6065
    },
    {
      "entropy": 5.610130167007446,
      "epoch": 0.4722816572651235,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004982478319217848,
      "loss": 5.5596,
      "mean_token_accuracy": 0.15958869457244873,
      "num_tokens": 10534608.0,
      "step": 6070
    },
    {
      "entropy": 5.708355045318603,
      "epoch": 0.4726706866368411,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004982443780868562,
      "loss": 5.4561,
      "mean_token_accuracy": 0.16599004119634628,
      "num_tokens": 10542992.0,
      "step": 6075
    },
    {
      "entropy": 5.656585454940796,
      "epoch": 0.4730597160085587,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004982409208645496,
      "loss": 5.451,
      "mean_token_accuracy": 0.1671524837613106,
      "num_tokens": 10551682.0,
      "step": 6080
    },
    {
      "entropy": 5.571417236328125,
      "epoch": 0.4734487453802762,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004982374602549173,
      "loss": 5.4114,
      "mean_token_accuracy": 0.17300263792276382,
      "num_tokens": 10560556.0,
      "step": 6085
    },
    {
      "entropy": 5.651502799987793,
      "epoch": 0.4738377747519938,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004982339962580119,
      "loss": 5.588,
      "mean_token_accuracy": 0.1589594915509224,
      "num_tokens": 10569361.0,
      "step": 6090
    },
    {
      "entropy": 5.575188159942627,
      "epoch": 0.4742268041237113,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004982305288738859,
      "loss": 5.4054,
      "mean_token_accuracy": 0.1707441434264183,
      "num_tokens": 10577997.0,
      "step": 6095
    },
    {
      "entropy": 5.6310670375823975,
      "epoch": 0.4746158334954289,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004982270581025919,
      "loss": 5.4956,
      "mean_token_accuracy": 0.1723932906985283,
      "num_tokens": 10587005.0,
      "step": 6100
    },
    {
      "entropy": 5.609803056716919,
      "epoch": 0.4750048628671465,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004982235839441826,
      "loss": 5.3391,
      "mean_token_accuracy": 0.18069400787353515,
      "num_tokens": 10595196.0,
      "step": 6105
    },
    {
      "entropy": 5.6674364566802975,
      "epoch": 0.475393892238864,
      "grad_norm": 1.0,
      "learning_rate": 0.0004982201063987108,
      "loss": 5.5135,
      "mean_token_accuracy": 0.16270238161087036,
      "num_tokens": 10603796.0,
      "step": 6110
    },
    {
      "entropy": 5.647106885910034,
      "epoch": 0.4757829216105816,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004982166254662292,
      "loss": 5.4786,
      "mean_token_accuracy": 0.1692984938621521,
      "num_tokens": 10612895.0,
      "step": 6115
    },
    {
      "entropy": 5.635853862762451,
      "epoch": 0.47617195098229914,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004982131411467904,
      "loss": 5.5493,
      "mean_token_accuracy": 0.1670703336596489,
      "num_tokens": 10621310.0,
      "step": 6120
    },
    {
      "entropy": 5.5323131561279295,
      "epoch": 0.4765609803540167,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004982096534404476,
      "loss": 5.3757,
      "mean_token_accuracy": 0.16931189149618148,
      "num_tokens": 10629598.0,
      "step": 6125
    },
    {
      "entropy": 5.6776985168457035,
      "epoch": 0.4769500097257343,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004982061623472535,
      "loss": 5.3825,
      "mean_token_accuracy": 0.17378225326538085,
      "num_tokens": 10638218.0,
      "step": 6130
    },
    {
      "entropy": 5.665579319000244,
      "epoch": 0.47733903909745184,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004982026678672612,
      "loss": 5.492,
      "mean_token_accuracy": 0.15960929840803145,
      "num_tokens": 10647215.0,
      "step": 6135
    },
    {
      "entropy": 5.638845300674438,
      "epoch": 0.47772806846916943,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004981991700005238,
      "loss": 5.5906,
      "mean_token_accuracy": 0.1585724726319313,
      "num_tokens": 10655755.0,
      "step": 6140
    },
    {
      "entropy": 5.696963310241699,
      "epoch": 0.47811709784088696,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000498195668747094,
      "loss": 5.4835,
      "mean_token_accuracy": 0.16718413829803466,
      "num_tokens": 10663969.0,
      "step": 6145
    },
    {
      "entropy": 5.655819749832153,
      "epoch": 0.47850612721260455,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004981921641070254,
      "loss": 5.3483,
      "mean_token_accuracy": 0.17026633620262147,
      "num_tokens": 10672048.0,
      "step": 6150
    },
    {
      "entropy": 5.579485988616943,
      "epoch": 0.47889515658432213,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004981886560803708,
      "loss": 5.4555,
      "mean_token_accuracy": 0.16290266662836075,
      "num_tokens": 10681802.0,
      "step": 6155
    },
    {
      "entropy": 5.651374053955078,
      "epoch": 0.47928418595603967,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004981851446671838,
      "loss": 5.5001,
      "mean_token_accuracy": 0.16652705818414687,
      "num_tokens": 10690618.0,
      "step": 6160
    },
    {
      "entropy": 5.613935375213623,
      "epoch": 0.47967321532775725,
      "grad_norm": 1.0,
      "learning_rate": 0.0004981816298675173,
      "loss": 5.4175,
      "mean_token_accuracy": 0.1716702923178673,
      "num_tokens": 10699080.0,
      "step": 6165
    },
    {
      "entropy": 5.624830150604248,
      "epoch": 0.4800622446994748,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004981781116814248,
      "loss": 5.4291,
      "mean_token_accuracy": 0.16670003682374954,
      "num_tokens": 10708939.0,
      "step": 6170
    },
    {
      "entropy": 5.654634475708008,
      "epoch": 0.48045127407119237,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004981745901089596,
      "loss": 5.3385,
      "mean_token_accuracy": 0.1697847366333008,
      "num_tokens": 10717356.0,
      "step": 6175
    },
    {
      "entropy": 5.626035737991333,
      "epoch": 0.48084030344290996,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004981710651501753,
      "loss": 5.4281,
      "mean_token_accuracy": 0.16301917284727097,
      "num_tokens": 10725616.0,
      "step": 6180
    },
    {
      "entropy": 5.588677644729614,
      "epoch": 0.4812293328146275,
      "grad_norm": 1.0,
      "learning_rate": 0.0004981675368051254,
      "loss": 5.4264,
      "mean_token_accuracy": 0.17776992470026015,
      "num_tokens": 10734337.0,
      "step": 6185
    },
    {
      "entropy": 5.621305656433106,
      "epoch": 0.4816183621863451,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004981640050738633,
      "loss": 5.4645,
      "mean_token_accuracy": 0.17033440619707108,
      "num_tokens": 10742720.0,
      "step": 6190
    },
    {
      "entropy": 5.677616167068481,
      "epoch": 0.48200739155806266,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004981604699564426,
      "loss": 5.4883,
      "mean_token_accuracy": 0.1639797195792198,
      "num_tokens": 10751809.0,
      "step": 6195
    },
    {
      "entropy": 5.61591329574585,
      "epoch": 0.4823964209297802,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004981569314529169,
      "loss": 5.3474,
      "mean_token_accuracy": 0.17736923545598984,
      "num_tokens": 10759879.0,
      "step": 6200
    },
    {
      "entropy": 5.5417656898498535,
      "epoch": 0.4827854503014978,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00049815338956334,
      "loss": 5.4154,
      "mean_token_accuracy": 0.17056161612272264,
      "num_tokens": 10769309.0,
      "step": 6205
    },
    {
      "entropy": 5.609933757781983,
      "epoch": 0.4831744796732153,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004981498442877656,
      "loss": 5.3623,
      "mean_token_accuracy": 0.17132142633199693,
      "num_tokens": 10777676.0,
      "step": 6210
    },
    {
      "entropy": 5.610863924026489,
      "epoch": 0.4835635090449329,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004981462956262474,
      "loss": 5.5553,
      "mean_token_accuracy": 0.16069165468215943,
      "num_tokens": 10786278.0,
      "step": 6215
    },
    {
      "entropy": 5.667156028747558,
      "epoch": 0.4839525384166505,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004981427435788396,
      "loss": 5.4765,
      "mean_token_accuracy": 0.1695808067917824,
      "num_tokens": 10795323.0,
      "step": 6220
    },
    {
      "entropy": 5.662508821487426,
      "epoch": 0.484341567788368,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004981391881455957,
      "loss": 5.3717,
      "mean_token_accuracy": 0.17230603396892546,
      "num_tokens": 10803978.0,
      "step": 6225
    },
    {
      "entropy": 5.626446294784546,
      "epoch": 0.4847305971600856,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004981356293265696,
      "loss": 5.4771,
      "mean_token_accuracy": 0.16738393157720566,
      "num_tokens": 10812488.0,
      "step": 6230
    },
    {
      "entropy": 5.646540594100952,
      "epoch": 0.4851196265318031,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004981320671218157,
      "loss": 5.4601,
      "mean_token_accuracy": 0.16784393042325974,
      "num_tokens": 10821076.0,
      "step": 6235
    },
    {
      "entropy": 5.567465448379517,
      "epoch": 0.4855086559035207,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004981285015313877,
      "loss": 5.2854,
      "mean_token_accuracy": 0.1801194965839386,
      "num_tokens": 10828634.0,
      "step": 6240
    },
    {
      "entropy": 5.601449966430664,
      "epoch": 0.4858976852752383,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004981249325553399,
      "loss": 5.4265,
      "mean_token_accuracy": 0.17164730727672578,
      "num_tokens": 10837481.0,
      "step": 6245
    },
    {
      "entropy": 5.58616452217102,
      "epoch": 0.48628671464695583,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004981213601937264,
      "loss": 5.4858,
      "mean_token_accuracy": 0.15897511690855026,
      "num_tokens": 10846729.0,
      "step": 6250
    },
    {
      "entropy": 5.643878412246704,
      "epoch": 0.4866757440186734,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004981177844466013,
      "loss": 5.4644,
      "mean_token_accuracy": 0.1645166888833046,
      "num_tokens": 10855344.0,
      "step": 6255
    },
    {
      "entropy": 5.684425592422485,
      "epoch": 0.48706477339039095,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004981142053140192,
      "loss": 5.497,
      "mean_token_accuracy": 0.1651305839419365,
      "num_tokens": 10863934.0,
      "step": 6260
    },
    {
      "entropy": 5.614658498764038,
      "epoch": 0.48745380276210853,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004981106227960339,
      "loss": 5.4849,
      "mean_token_accuracy": 0.16493846774101256,
      "num_tokens": 10872031.0,
      "step": 6265
    },
    {
      "entropy": 5.666933012008667,
      "epoch": 0.4878428321338261,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004981070368927001,
      "loss": 5.5386,
      "mean_token_accuracy": 0.16845548897981644,
      "num_tokens": 10881385.0,
      "step": 6270
    },
    {
      "entropy": 5.6381768703460695,
      "epoch": 0.48823186150554365,
      "grad_norm": 0.96875,
      "learning_rate": 0.000498103447604072,
      "loss": 5.4623,
      "mean_token_accuracy": 0.1679671362042427,
      "num_tokens": 10890483.0,
      "step": 6275
    },
    {
      "entropy": 5.617257308959961,
      "epoch": 0.48862089087726124,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004980998549302044,
      "loss": 5.4774,
      "mean_token_accuracy": 0.17509936690330505,
      "num_tokens": 10898338.0,
      "step": 6280
    },
    {
      "entropy": 5.552199363708496,
      "epoch": 0.4890099202489788,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004980962588711516,
      "loss": 5.4584,
      "mean_token_accuracy": 0.1705517053604126,
      "num_tokens": 10908087.0,
      "step": 6285
    },
    {
      "entropy": 5.682706069946289,
      "epoch": 0.48939894962069636,
      "grad_norm": 1.046875,
      "learning_rate": 0.000498092659426968,
      "loss": 5.4463,
      "mean_token_accuracy": 0.16522486060857772,
      "num_tokens": 10916704.0,
      "step": 6290
    },
    {
      "entropy": 5.617280960083008,
      "epoch": 0.48978797899241394,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004980890565977085,
      "loss": 5.4108,
      "mean_token_accuracy": 0.17075762003660203,
      "num_tokens": 10925186.0,
      "step": 6295
    },
    {
      "entropy": 5.597516345977783,
      "epoch": 0.4901770083641315,
      "grad_norm": 1.0,
      "learning_rate": 0.0004980854503834276,
      "loss": 5.4592,
      "mean_token_accuracy": 0.1715566873550415,
      "num_tokens": 10934247.0,
      "step": 6300
    },
    {
      "entropy": 5.691966485977173,
      "epoch": 0.49056603773584906,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004980818407841802,
      "loss": 5.489,
      "mean_token_accuracy": 0.16149261146783828,
      "num_tokens": 10944260.0,
      "step": 6305
    },
    {
      "entropy": 5.624559926986694,
      "epoch": 0.49095506710756665,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004980782278000207,
      "loss": 5.4486,
      "mean_token_accuracy": 0.1665085107088089,
      "num_tokens": 10954065.0,
      "step": 6310
    },
    {
      "entropy": 5.601347780227661,
      "epoch": 0.4913440964792842,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004980746114310043,
      "loss": 5.5157,
      "mean_token_accuracy": 0.16822022050619126,
      "num_tokens": 10962918.0,
      "step": 6315
    },
    {
      "entropy": 5.550816249847412,
      "epoch": 0.49173312585100176,
      "grad_norm": 0.875,
      "learning_rate": 0.0004980709916771858,
      "loss": 5.3865,
      "mean_token_accuracy": 0.16956816762685775,
      "num_tokens": 10972356.0,
      "step": 6320
    },
    {
      "entropy": 5.565039730072021,
      "epoch": 0.4921221552227193,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004980673685386198,
      "loss": 5.4411,
      "mean_token_accuracy": 0.16410028636455537,
      "num_tokens": 10980984.0,
      "step": 6325
    },
    {
      "entropy": 5.583667898178101,
      "epoch": 0.4925111845944369,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004980637420153618,
      "loss": 5.3821,
      "mean_token_accuracy": 0.17071073800325393,
      "num_tokens": 10989809.0,
      "step": 6330
    },
    {
      "entropy": 5.593800401687622,
      "epoch": 0.49290021396615447,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004980601121074664,
      "loss": 5.4213,
      "mean_token_accuracy": 0.16924805641174318,
      "num_tokens": 10998824.0,
      "step": 6335
    },
    {
      "entropy": 5.684945869445801,
      "epoch": 0.493289243337872,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004980564788149889,
      "loss": 5.5004,
      "mean_token_accuracy": 0.17203586399555207,
      "num_tokens": 11006755.0,
      "step": 6340
    },
    {
      "entropy": 5.593352174758911,
      "epoch": 0.4936782727095896,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004980528421379844,
      "loss": 5.3922,
      "mean_token_accuracy": 0.17423760145902634,
      "num_tokens": 11015337.0,
      "step": 6345
    },
    {
      "entropy": 5.541840124130249,
      "epoch": 0.4940673020813071,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000498049202076508,
      "loss": 5.3058,
      "mean_token_accuracy": 0.17895494997501374,
      "num_tokens": 11024186.0,
      "step": 6350
    },
    {
      "entropy": 5.613464212417602,
      "epoch": 0.4944563314530247,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004980455586306149,
      "loss": 5.4296,
      "mean_token_accuracy": 0.17041152119636535,
      "num_tokens": 11032504.0,
      "step": 6355
    },
    {
      "entropy": 5.663481903076172,
      "epoch": 0.4948453608247423,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004980419118003605,
      "loss": 5.3696,
      "mean_token_accuracy": 0.1725605770945549,
      "num_tokens": 11042252.0,
      "step": 6360
    },
    {
      "entropy": 5.619900846481324,
      "epoch": 0.4952343901964598,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004980382615858001,
      "loss": 5.458,
      "mean_token_accuracy": 0.16668125987052917,
      "num_tokens": 11051361.0,
      "step": 6365
    },
    {
      "entropy": 5.56270694732666,
      "epoch": 0.4956234195681774,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004980346079869892,
      "loss": 5.4267,
      "mean_token_accuracy": 0.1661438524723053,
      "num_tokens": 11060822.0,
      "step": 6370
    },
    {
      "entropy": 5.620711994171143,
      "epoch": 0.49601244893989493,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000498030951003983,
      "loss": 5.4892,
      "mean_token_accuracy": 0.16607844531536103,
      "num_tokens": 11069528.0,
      "step": 6375
    },
    {
      "entropy": 5.591744136810303,
      "epoch": 0.4964014783116125,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004980272906368371,
      "loss": 5.3406,
      "mean_token_accuracy": 0.1787100166082382,
      "num_tokens": 11077917.0,
      "step": 6380
    },
    {
      "entropy": 5.55047116279602,
      "epoch": 0.4967905076833301,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004980236268856071,
      "loss": 5.4104,
      "mean_token_accuracy": 0.16398482620716096,
      "num_tokens": 11085938.0,
      "step": 6385
    },
    {
      "entropy": 5.583984661102295,
      "epoch": 0.49717953705504764,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004980199597503487,
      "loss": 5.468,
      "mean_token_accuracy": 0.1609756290912628,
      "num_tokens": 11094898.0,
      "step": 6390
    },
    {
      "entropy": 5.590396404266357,
      "epoch": 0.4975685664267652,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004980162892311171,
      "loss": 5.3292,
      "mean_token_accuracy": 0.17555033713579177,
      "num_tokens": 11103038.0,
      "step": 6395
    },
    {
      "entropy": 5.613178873062134,
      "epoch": 0.4979575957984828,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004980126153279684,
      "loss": 5.4196,
      "mean_token_accuracy": 0.1670178636908531,
      "num_tokens": 11111965.0,
      "step": 6400
    },
    {
      "entropy": 5.615345287322998,
      "epoch": 0.49834662517020034,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004980089380409583,
      "loss": 5.4328,
      "mean_token_accuracy": 0.16547031104564666,
      "num_tokens": 11121482.0,
      "step": 6405
    },
    {
      "entropy": 5.596158838272094,
      "epoch": 0.4987356545419179,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004980052573701424,
      "loss": 5.428,
      "mean_token_accuracy": 0.16882085353136062,
      "num_tokens": 11129742.0,
      "step": 6410
    },
    {
      "entropy": 5.5777904987335205,
      "epoch": 0.49912468391363546,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004980015733155767,
      "loss": 5.4408,
      "mean_token_accuracy": 0.16935502588748932,
      "num_tokens": 11138388.0,
      "step": 6415
    },
    {
      "entropy": 5.447121810913086,
      "epoch": 0.49951371328535304,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004979978858773171,
      "loss": 5.1733,
      "mean_token_accuracy": 0.18193307518959045,
      "num_tokens": 11146967.0,
      "step": 6420
    },
    {
      "entropy": 5.607067060470581,
      "epoch": 0.49990274265707063,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004979941950554195,
      "loss": 5.5309,
      "mean_token_accuracy": 0.16334948539733887,
      "num_tokens": 11155888.0,
      "step": 6425
    },
    {
      "entropy": 5.667019271850586,
      "epoch": 0.5002917720287882,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004979905008499399,
      "loss": 5.3792,
      "mean_token_accuracy": 0.1710866242647171,
      "num_tokens": 11164358.0,
      "step": 6430
    },
    {
      "entropy": 5.5354045867919925,
      "epoch": 0.5006808014005057,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004979868032609344,
      "loss": 5.2391,
      "mean_token_accuracy": 0.18528240025043488,
      "num_tokens": 11172416.0,
      "step": 6435
    },
    {
      "entropy": 5.609890556335449,
      "epoch": 0.5010698307722233,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004979831022884591,
      "loss": 5.4384,
      "mean_token_accuracy": 0.16946105062961578,
      "num_tokens": 11181159.0,
      "step": 6440
    },
    {
      "entropy": 5.649543380737304,
      "epoch": 0.5014588601439409,
      "grad_norm": 1.0,
      "learning_rate": 0.0004979793979325701,
      "loss": 5.4731,
      "mean_token_accuracy": 0.16498799622058868,
      "num_tokens": 11189725.0,
      "step": 6445
    },
    {
      "entropy": 5.627615880966187,
      "epoch": 0.5018478895156584,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004979756901933236,
      "loss": 5.4467,
      "mean_token_accuracy": 0.16980287730693816,
      "num_tokens": 11197795.0,
      "step": 6450
    },
    {
      "entropy": 5.615576887130738,
      "epoch": 0.502236918887376,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000497971979070776,
      "loss": 5.4077,
      "mean_token_accuracy": 0.1722092866897583,
      "num_tokens": 11205352.0,
      "step": 6455
    },
    {
      "entropy": 5.568438148498535,
      "epoch": 0.5026259482590936,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004979682645649834,
      "loss": 5.4368,
      "mean_token_accuracy": 0.17089465260505676,
      "num_tokens": 11214142.0,
      "step": 6460
    },
    {
      "entropy": 5.579258871078491,
      "epoch": 0.5030149776308112,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004979645466760025,
      "loss": 5.3316,
      "mean_token_accuracy": 0.17273433059453963,
      "num_tokens": 11222565.0,
      "step": 6465
    },
    {
      "entropy": 5.582017612457276,
      "epoch": 0.5034040070025287,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004979608254038892,
      "loss": 5.4462,
      "mean_token_accuracy": 0.1604880064725876,
      "num_tokens": 11231829.0,
      "step": 6470
    },
    {
      "entropy": 5.582902669906616,
      "epoch": 0.5037930363742462,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004979571007487003,
      "loss": 5.441,
      "mean_token_accuracy": 0.16832706481218337,
      "num_tokens": 11240302.0,
      "step": 6475
    },
    {
      "entropy": 5.694792461395264,
      "epoch": 0.5041820657459638,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004979533727104924,
      "loss": 5.5423,
      "mean_token_accuracy": 0.1603501871228218,
      "num_tokens": 11248610.0,
      "step": 6480
    },
    {
      "entropy": 5.598382377624512,
      "epoch": 0.5045710951176814,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000497949641289322,
      "loss": 5.444,
      "mean_token_accuracy": 0.17378710061311722,
      "num_tokens": 11257441.0,
      "step": 6485
    },
    {
      "entropy": 5.523042249679565,
      "epoch": 0.504960124489399,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004979459064852456,
      "loss": 5.3234,
      "mean_token_accuracy": 0.16616344153881074,
      "num_tokens": 11265915.0,
      "step": 6490
    },
    {
      "entropy": 5.611534786224365,
      "epoch": 0.5053491538611166,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004979421682983197,
      "loss": 5.4414,
      "mean_token_accuracy": 0.16530156061053275,
      "num_tokens": 11274165.0,
      "step": 6495
    },
    {
      "entropy": 5.618980073928833,
      "epoch": 0.505738183232834,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004979384267286015,
      "loss": 5.391,
      "mean_token_accuracy": 0.17459912300109864,
      "num_tokens": 11283316.0,
      "step": 6500
    },
    {
      "entropy": 5.612117004394531,
      "epoch": 0.5061272126045516,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004979346817761475,
      "loss": 5.5109,
      "mean_token_accuracy": 0.1655980721116066,
      "num_tokens": 11292289.0,
      "step": 6505
    },
    {
      "entropy": 5.577218294143677,
      "epoch": 0.5065162419762692,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004979309334410144,
      "loss": 5.4357,
      "mean_token_accuracy": 0.1642925187945366,
      "num_tokens": 11301192.0,
      "step": 6510
    },
    {
      "entropy": 5.654868507385254,
      "epoch": 0.5069052713479868,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004979271817232594,
      "loss": 5.4526,
      "mean_token_accuracy": 0.17635779529809953,
      "num_tokens": 11309906.0,
      "step": 6515
    },
    {
      "entropy": 5.648916149139405,
      "epoch": 0.5072943007197044,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004979234266229391,
      "loss": 5.4318,
      "mean_token_accuracy": 0.16642682403326034,
      "num_tokens": 11318227.0,
      "step": 6520
    },
    {
      "entropy": 5.530883407592773,
      "epoch": 0.5076833300914219,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004979196681401106,
      "loss": 5.4639,
      "mean_token_accuracy": 0.16614597737789155,
      "num_tokens": 11327524.0,
      "step": 6525
    },
    {
      "entropy": 5.56320858001709,
      "epoch": 0.5080723594631394,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000497915906274831,
      "loss": 5.3661,
      "mean_token_accuracy": 0.1760654181241989,
      "num_tokens": 11336214.0,
      "step": 6530
    },
    {
      "entropy": 5.615737056732177,
      "epoch": 0.508461388834857,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004979121410271574,
      "loss": 5.2423,
      "mean_token_accuracy": 0.18362002521753312,
      "num_tokens": 11344551.0,
      "step": 6535
    },
    {
      "entropy": 5.481628704071045,
      "epoch": 0.5088504182065746,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004979083723971468,
      "loss": 5.3072,
      "mean_token_accuracy": 0.1763471022248268,
      "num_tokens": 11353180.0,
      "step": 6540
    },
    {
      "entropy": 5.419930839538575,
      "epoch": 0.5092394475782922,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004979046003848564,
      "loss": 5.2875,
      "mean_token_accuracy": 0.18146508634090425,
      "num_tokens": 11361343.0,
      "step": 6545
    },
    {
      "entropy": 5.569938135147095,
      "epoch": 0.5096284769500097,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004979008249903435,
      "loss": 5.3775,
      "mean_token_accuracy": 0.1741035982966423,
      "num_tokens": 11369915.0,
      "step": 6550
    },
    {
      "entropy": 5.606975793838501,
      "epoch": 0.5100175063217273,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004978970462136655,
      "loss": 5.426,
      "mean_token_accuracy": 0.1679391846060753,
      "num_tokens": 11379813.0,
      "step": 6555
    },
    {
      "entropy": 5.629864501953125,
      "epoch": 0.5104065356934449,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004978932640548794,
      "loss": 5.3618,
      "mean_token_accuracy": 0.17083254903554917,
      "num_tokens": 11388029.0,
      "step": 6560
    },
    {
      "entropy": 5.550869989395141,
      "epoch": 0.5107955650651624,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004978894785140429,
      "loss": 5.2906,
      "mean_token_accuracy": 0.17675724625587463,
      "num_tokens": 11395852.0,
      "step": 6565
    },
    {
      "entropy": 5.556102228164673,
      "epoch": 0.51118459443688,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004978856895912132,
      "loss": 5.3893,
      "mean_token_accuracy": 0.1705313131213188,
      "num_tokens": 11404756.0,
      "step": 6570
    },
    {
      "entropy": 5.71532096862793,
      "epoch": 0.5115736238085975,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004978818972864481,
      "loss": 5.5504,
      "mean_token_accuracy": 0.16391270309686662,
      "num_tokens": 11413301.0,
      "step": 6575
    },
    {
      "entropy": 5.64161319732666,
      "epoch": 0.5119626531803151,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004978781015998048,
      "loss": 5.3304,
      "mean_token_accuracy": 0.17462185472249986,
      "num_tokens": 11421034.0,
      "step": 6580
    },
    {
      "entropy": 5.564785623550415,
      "epoch": 0.5123516825520327,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004978743025313413,
      "loss": 5.3704,
      "mean_token_accuracy": 0.1695050537586212,
      "num_tokens": 11429066.0,
      "step": 6585
    },
    {
      "entropy": 5.463819169998169,
      "epoch": 0.5127407119237503,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004978705000811148,
      "loss": 5.3844,
      "mean_token_accuracy": 0.17122116088867187,
      "num_tokens": 11438109.0,
      "step": 6590
    },
    {
      "entropy": 5.608554220199585,
      "epoch": 0.5131297412954678,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004978666942491832,
      "loss": 5.4056,
      "mean_token_accuracy": 0.16849517226219177,
      "num_tokens": 11446787.0,
      "step": 6595
    },
    {
      "entropy": 5.540587091445923,
      "epoch": 0.5135187706671853,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004978628850356043,
      "loss": 5.3669,
      "mean_token_accuracy": 0.1661093682050705,
      "num_tokens": 11456262.0,
      "step": 6600
    },
    {
      "entropy": 5.619626092910766,
      "epoch": 0.5139078000389029,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004978590724404358,
      "loss": 5.3631,
      "mean_token_accuracy": 0.17528488487005234,
      "num_tokens": 11464297.0,
      "step": 6605
    },
    {
      "entropy": 5.595329666137696,
      "epoch": 0.5142968294106205,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004978552564637356,
      "loss": 5.4353,
      "mean_token_accuracy": 0.1698346272110939,
      "num_tokens": 11473519.0,
      "step": 6610
    },
    {
      "entropy": 5.611803293228149,
      "epoch": 0.5146858587823381,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004978514371055616,
      "loss": 5.4251,
      "mean_token_accuracy": 0.16267327815294266,
      "num_tokens": 11482249.0,
      "step": 6615
    },
    {
      "entropy": 5.541417646408081,
      "epoch": 0.5150748881540557,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004978476143659718,
      "loss": 5.3848,
      "mean_token_accuracy": 0.1657371535897255,
      "num_tokens": 11490687.0,
      "step": 6620
    },
    {
      "entropy": 5.518591451644897,
      "epoch": 0.5154639175257731,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004978437882450241,
      "loss": 5.3911,
      "mean_token_accuracy": 0.16727733761072158,
      "num_tokens": 11500225.0,
      "step": 6625
    },
    {
      "entropy": 5.589768934249878,
      "epoch": 0.5158529468974907,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004978399587427766,
      "loss": 5.3287,
      "mean_token_accuracy": 0.172044076025486,
      "num_tokens": 11508116.0,
      "step": 6630
    },
    {
      "entropy": 5.536455869674683,
      "epoch": 0.5162419762692083,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004978361258592874,
      "loss": 5.4177,
      "mean_token_accuracy": 0.164792999625206,
      "num_tokens": 11516525.0,
      "step": 6635
    },
    {
      "entropy": 5.564665985107422,
      "epoch": 0.5166310056409259,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004978322895946147,
      "loss": 5.3791,
      "mean_token_accuracy": 0.17165726274251938,
      "num_tokens": 11525434.0,
      "step": 6640
    },
    {
      "entropy": 5.631292772293091,
      "epoch": 0.5170200350126435,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004978284499488167,
      "loss": 5.4412,
      "mean_token_accuracy": 0.1706455558538437,
      "num_tokens": 11534318.0,
      "step": 6645
    },
    {
      "entropy": 5.574923372268676,
      "epoch": 0.5174090643843611,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004978246069219516,
      "loss": 5.3597,
      "mean_token_accuracy": 0.18113650977611542,
      "num_tokens": 11542777.0,
      "step": 6650
    },
    {
      "entropy": 5.5791538715362545,
      "epoch": 0.5177980937560785,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004978207605140777,
      "loss": 5.3498,
      "mean_token_accuracy": 0.17651850134134292,
      "num_tokens": 11550888.0,
      "step": 6655
    },
    {
      "entropy": 5.592405462265015,
      "epoch": 0.5181871231277961,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004978169107252534,
      "loss": 5.4109,
      "mean_token_accuracy": 0.165172179043293,
      "num_tokens": 11560047.0,
      "step": 6660
    },
    {
      "entropy": 5.6135838508605955,
      "epoch": 0.5185761524995137,
      "grad_norm": 1.0,
      "learning_rate": 0.0004978130575555373,
      "loss": 5.3923,
      "mean_token_accuracy": 0.17554190754890442,
      "num_tokens": 11568862.0,
      "step": 6665
    },
    {
      "entropy": 5.488176393508911,
      "epoch": 0.5189651818712313,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004978092010049877,
      "loss": 5.3601,
      "mean_token_accuracy": 0.1723656326532364,
      "num_tokens": 11577509.0,
      "step": 6670
    },
    {
      "entropy": 5.712526178359985,
      "epoch": 0.5193542112429489,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000497805341073663,
      "loss": 5.4713,
      "mean_token_accuracy": 0.1692461058497429,
      "num_tokens": 11585902.0,
      "step": 6675
    },
    {
      "entropy": 5.646485376358032,
      "epoch": 0.5197432406146664,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004978014777616219,
      "loss": 5.4625,
      "mean_token_accuracy": 0.16923384368419647,
      "num_tokens": 11594895.0,
      "step": 6680
    },
    {
      "entropy": 5.5356780052185055,
      "epoch": 0.520132269986384,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004977976110689229,
      "loss": 5.2809,
      "mean_token_accuracy": 0.17121015340089799,
      "num_tokens": 11604080.0,
      "step": 6685
    },
    {
      "entropy": 5.624505233764649,
      "epoch": 0.5205212993581015,
      "grad_norm": 0.890625,
      "learning_rate": 0.000497793740995625,
      "loss": 5.4099,
      "mean_token_accuracy": 0.165664704144001,
      "num_tokens": 11613370.0,
      "step": 6690
    },
    {
      "entropy": 5.542125701904297,
      "epoch": 0.5209103287298191,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004977898675417866,
      "loss": 5.3951,
      "mean_token_accuracy": 0.1671002522110939,
      "num_tokens": 11621913.0,
      "step": 6695
    },
    {
      "entropy": 5.596344661712647,
      "epoch": 0.5212993581015367,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004977859907074664,
      "loss": 5.5293,
      "mean_token_accuracy": 0.1685265377163887,
      "num_tokens": 11630393.0,
      "step": 6700
    },
    {
      "entropy": 5.673050022125244,
      "epoch": 0.5216883874732542,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004977821104927236,
      "loss": 5.4621,
      "mean_token_accuracy": 0.16705554723739624,
      "num_tokens": 11639038.0,
      "step": 6705
    },
    {
      "entropy": 5.625570058822632,
      "epoch": 0.5220774168449718,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004977782268976167,
      "loss": 5.379,
      "mean_token_accuracy": 0.16754479259252547,
      "num_tokens": 11647355.0,
      "step": 6710
    },
    {
      "entropy": 5.532090282440185,
      "epoch": 0.5224664462166894,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004977743399222048,
      "loss": 5.3564,
      "mean_token_accuracy": 0.17099345177412034,
      "num_tokens": 11655899.0,
      "step": 6715
    },
    {
      "entropy": 5.657911539077759,
      "epoch": 0.522855475588407,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004977704495665471,
      "loss": 5.5552,
      "mean_token_accuracy": 0.15934895277023314,
      "num_tokens": 11664602.0,
      "step": 6720
    },
    {
      "entropy": 5.642425584793091,
      "epoch": 0.5232445049601245,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004977665558307023,
      "loss": 5.3746,
      "mean_token_accuracy": 0.17409038692712783,
      "num_tokens": 11673011.0,
      "step": 6725
    },
    {
      "entropy": 5.697009134292602,
      "epoch": 0.523633534331842,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004977626587147295,
      "loss": 5.5345,
      "mean_token_accuracy": 0.16368037313222886,
      "num_tokens": 11681989.0,
      "step": 6730
    },
    {
      "entropy": 5.637260293960571,
      "epoch": 0.5240225637035596,
      "grad_norm": 1.0,
      "learning_rate": 0.000497758758218688,
      "loss": 5.4739,
      "mean_token_accuracy": 0.1594838172197342,
      "num_tokens": 11690502.0,
      "step": 6735
    },
    {
      "entropy": 5.575308609008789,
      "epoch": 0.5244115930752772,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004977548543426368,
      "loss": 5.4671,
      "mean_token_accuracy": 0.16562397554516792,
      "num_tokens": 11699594.0,
      "step": 6740
    },
    {
      "entropy": 5.537110805511475,
      "epoch": 0.5248006224469948,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004977509470866353,
      "loss": 5.3447,
      "mean_token_accuracy": 0.17038798630237578,
      "num_tokens": 11708009.0,
      "step": 6745
    },
    {
      "entropy": 5.663848733901977,
      "epoch": 0.5251896518187124,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004977470364507427,
      "loss": 5.5046,
      "mean_token_accuracy": 0.16341960281133652,
      "num_tokens": 11716592.0,
      "step": 6750
    },
    {
      "entropy": 5.618674945831299,
      "epoch": 0.5255786811904298,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004977431224350184,
      "loss": 5.5235,
      "mean_token_accuracy": 0.16120998114347457,
      "num_tokens": 11725894.0,
      "step": 6755
    },
    {
      "entropy": 5.6809728145599365,
      "epoch": 0.5259677105621474,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004977392050395217,
      "loss": 5.3888,
      "mean_token_accuracy": 0.1697473183274269,
      "num_tokens": 11734141.0,
      "step": 6760
    },
    {
      "entropy": 5.621731805801391,
      "epoch": 0.526356739933865,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004977352842643121,
      "loss": 5.4757,
      "mean_token_accuracy": 0.16483270525932311,
      "num_tokens": 11742539.0,
      "step": 6765
    },
    {
      "entropy": 5.612393999099732,
      "epoch": 0.5267457693055826,
      "grad_norm": 1.015625,
      "learning_rate": 0.000497731360109449,
      "loss": 5.4475,
      "mean_token_accuracy": 0.1666994646191597,
      "num_tokens": 11751649.0,
      "step": 6770
    },
    {
      "entropy": 5.646832132339478,
      "epoch": 0.5271347986773002,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004977274325749922,
      "loss": 5.4477,
      "mean_token_accuracy": 0.16829011142253875,
      "num_tokens": 11760376.0,
      "step": 6775
    },
    {
      "entropy": 5.643862009048462,
      "epoch": 0.5275238280490177,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004977235016610011,
      "loss": 5.421,
      "mean_token_accuracy": 0.17095642983913423,
      "num_tokens": 11768644.0,
      "step": 6780
    },
    {
      "entropy": 5.682761001586914,
      "epoch": 0.5279128574207352,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004977195673675353,
      "loss": 5.4918,
      "mean_token_accuracy": 0.1657439336180687,
      "num_tokens": 11777282.0,
      "step": 6785
    },
    {
      "entropy": 5.595294666290283,
      "epoch": 0.5283018867924528,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004977156296946545,
      "loss": 5.4046,
      "mean_token_accuracy": 0.16914087831974028,
      "num_tokens": 11786108.0,
      "step": 6790
    },
    {
      "entropy": 5.5805693626403805,
      "epoch": 0.5286909161641704,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004977116886424187,
      "loss": 5.4384,
      "mean_token_accuracy": 0.16978208869695663,
      "num_tokens": 11795151.0,
      "step": 6795
    },
    {
      "entropy": 5.715737247467041,
      "epoch": 0.529079945535888,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004977077442108873,
      "loss": 5.3878,
      "mean_token_accuracy": 0.17581014186143876,
      "num_tokens": 11803971.0,
      "step": 6800
    },
    {
      "entropy": 5.574142837524414,
      "epoch": 0.5294689749076055,
      "grad_norm": 1.0,
      "learning_rate": 0.0004977037964001204,
      "loss": 5.4478,
      "mean_token_accuracy": 0.17486259937286378,
      "num_tokens": 11812483.0,
      "step": 6805
    },
    {
      "entropy": 5.524339008331299,
      "epoch": 0.5298580042793231,
      "grad_norm": 1.015625,
      "learning_rate": 0.000497699845210178,
      "loss": 5.3246,
      "mean_token_accuracy": 0.1793253093957901,
      "num_tokens": 11821144.0,
      "step": 6810
    },
    {
      "entropy": 5.584891271591187,
      "epoch": 0.5302470336510406,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004976958906411197,
      "loss": 5.2943,
      "mean_token_accuracy": 0.181274676322937,
      "num_tokens": 11829298.0,
      "step": 6815
    },
    {
      "entropy": 5.635812282562256,
      "epoch": 0.5306360630227582,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004976919326930059,
      "loss": 5.388,
      "mean_token_accuracy": 0.17337559312582015,
      "num_tokens": 11837558.0,
      "step": 6820
    },
    {
      "entropy": 5.477765369415283,
      "epoch": 0.5310250923944758,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004976879713658964,
      "loss": 5.2802,
      "mean_token_accuracy": 0.17405419796705246,
      "num_tokens": 11845770.0,
      "step": 6825
    },
    {
      "entropy": 5.638754606246948,
      "epoch": 0.5314141217661933,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004976840066598514,
      "loss": 5.4689,
      "mean_token_accuracy": 0.16189807653427124,
      "num_tokens": 11854351.0,
      "step": 6830
    },
    {
      "entropy": 5.7618780612945555,
      "epoch": 0.5318031511379109,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004976800385749309,
      "loss": 5.474,
      "mean_token_accuracy": 0.17206030040979386,
      "num_tokens": 11862610.0,
      "step": 6835
    },
    {
      "entropy": 5.608887481689453,
      "epoch": 0.5321921805096285,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004976760671111954,
      "loss": 5.444,
      "mean_token_accuracy": 0.17438511103391646,
      "num_tokens": 11871078.0,
      "step": 6840
    },
    {
      "entropy": 5.530498409271241,
      "epoch": 0.532581209881346,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000497672092268705,
      "loss": 5.3171,
      "mean_token_accuracy": 0.1797177627682686,
      "num_tokens": 11879303.0,
      "step": 6845
    },
    {
      "entropy": 5.5963081359863285,
      "epoch": 0.5329702392530636,
      "grad_norm": 1.0625,
      "learning_rate": 0.00049766811404752,
      "loss": 5.3707,
      "mean_token_accuracy": 0.17965137511491774,
      "num_tokens": 11887855.0,
      "step": 6850
    },
    {
      "entropy": 5.646993207931518,
      "epoch": 0.5333592686247812,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004976641324477008,
      "loss": 5.3286,
      "mean_token_accuracy": 0.17595505118370056,
      "num_tokens": 11896262.0,
      "step": 6855
    },
    {
      "entropy": 5.467989587783814,
      "epoch": 0.5337482979964987,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004976601474693077,
      "loss": 5.247,
      "mean_token_accuracy": 0.17795947194099426,
      "num_tokens": 11905140.0,
      "step": 6860
    },
    {
      "entropy": 5.514421796798706,
      "epoch": 0.5341373273682163,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004976561591124014,
      "loss": 5.2982,
      "mean_token_accuracy": 0.1761595129966736,
      "num_tokens": 11913177.0,
      "step": 6865
    },
    {
      "entropy": 5.440929698944092,
      "epoch": 0.5345263567399339,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004976521673770421,
      "loss": 5.3339,
      "mean_token_accuracy": 0.18069458454847337,
      "num_tokens": 11921955.0,
      "step": 6870
    },
    {
      "entropy": 5.563420343399048,
      "epoch": 0.5349153861116515,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004976481722632907,
      "loss": 5.3231,
      "mean_token_accuracy": 0.17072801738977433,
      "num_tokens": 11930976.0,
      "step": 6875
    },
    {
      "entropy": 5.577029085159301,
      "epoch": 0.535304415483369,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004976441737712076,
      "loss": 5.259,
      "mean_token_accuracy": 0.17814161032438278,
      "num_tokens": 11939708.0,
      "step": 6880
    },
    {
      "entropy": 5.560328722000122,
      "epoch": 0.5356934448550865,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004976401719008536,
      "loss": 5.443,
      "mean_token_accuracy": 0.16814709156751634,
      "num_tokens": 11949176.0,
      "step": 6885
    },
    {
      "entropy": 5.58608045578003,
      "epoch": 0.5360824742268041,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004976361666522893,
      "loss": 5.3123,
      "mean_token_accuracy": 0.17927956730127334,
      "num_tokens": 11958089.0,
      "step": 6890
    },
    {
      "entropy": 5.614568758010864,
      "epoch": 0.5364715035985217,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004976321580255755,
      "loss": 5.388,
      "mean_token_accuracy": 0.17972830086946487,
      "num_tokens": 11966719.0,
      "step": 6895
    },
    {
      "entropy": 5.617801094055176,
      "epoch": 0.5368605329702393,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004976281460207731,
      "loss": 5.4669,
      "mean_token_accuracy": 0.16239609718322753,
      "num_tokens": 11975959.0,
      "step": 6900
    },
    {
      "entropy": 5.615549850463867,
      "epoch": 0.5372495623419569,
      "grad_norm": 0.96875,
      "learning_rate": 0.000497624130637943,
      "loss": 5.4099,
      "mean_token_accuracy": 0.16989550590515137,
      "num_tokens": 11984282.0,
      "step": 6905
    },
    {
      "entropy": 5.605066251754761,
      "epoch": 0.5376385917136743,
      "grad_norm": 1.0625,
      "learning_rate": 0.000497620111877146,
      "loss": 5.398,
      "mean_token_accuracy": 0.16931922733783722,
      "num_tokens": 11993289.0,
      "step": 6910
    },
    {
      "entropy": 5.548697805404663,
      "epoch": 0.5380276210853919,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004976160897384431,
      "loss": 5.3456,
      "mean_token_accuracy": 0.17435020804405213,
      "num_tokens": 12002425.0,
      "step": 6915
    },
    {
      "entropy": 5.591514587402344,
      "epoch": 0.5384166504571095,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004976120642218955,
      "loss": 5.3564,
      "mean_token_accuracy": 0.176065856218338,
      "num_tokens": 12010435.0,
      "step": 6920
    },
    {
      "entropy": 5.539251756668091,
      "epoch": 0.5388056798288271,
      "grad_norm": 0.984375,
      "learning_rate": 0.000497608035327564,
      "loss": 5.4008,
      "mean_token_accuracy": 0.16208278834819795,
      "num_tokens": 12018768.0,
      "step": 6925
    },
    {
      "entropy": 5.625278520584106,
      "epoch": 0.5391947092005447,
      "grad_norm": 0.984375,
      "learning_rate": 0.00049760400305551,
      "loss": 5.4074,
      "mean_token_accuracy": 0.16875990480184555,
      "num_tokens": 12027617.0,
      "step": 6930
    },
    {
      "entropy": 5.607542943954468,
      "epoch": 0.5395837385722622,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004975999674057944,
      "loss": 5.382,
      "mean_token_accuracy": 0.17100832462310792,
      "num_tokens": 12036024.0,
      "step": 6935
    },
    {
      "entropy": 5.621946001052857,
      "epoch": 0.5399727679439797,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004975959283784787,
      "loss": 5.437,
      "mean_token_accuracy": 0.17009972184896469,
      "num_tokens": 12044206.0,
      "step": 6940
    },
    {
      "entropy": 5.556149482727051,
      "epoch": 0.5403617973156973,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004975918859736241,
      "loss": 5.3135,
      "mean_token_accuracy": 0.17374830096960067,
      "num_tokens": 12052294.0,
      "step": 6945
    },
    {
      "entropy": 5.5647095203399655,
      "epoch": 0.5407508266874149,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004975878401912919,
      "loss": 5.4326,
      "mean_token_accuracy": 0.1707412749528885,
      "num_tokens": 12060446.0,
      "step": 6950
    },
    {
      "entropy": 5.616201305389405,
      "epoch": 0.5411398560591325,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004975837910315436,
      "loss": 5.5087,
      "mean_token_accuracy": 0.16847269237041473,
      "num_tokens": 12069267.0,
      "step": 6955
    },
    {
      "entropy": 5.642180013656616,
      "epoch": 0.54152888543085,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004975797384944405,
      "loss": 5.4386,
      "mean_token_accuracy": 0.17013390660285949,
      "num_tokens": 12077808.0,
      "step": 6960
    },
    {
      "entropy": 5.583532905578613,
      "epoch": 0.5419179148025676,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000497575682580044,
      "loss": 5.3888,
      "mean_token_accuracy": 0.1725155904889107,
      "num_tokens": 12086893.0,
      "step": 6965
    },
    {
      "entropy": 5.56446270942688,
      "epoch": 0.5423069441742852,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000497571623288416,
      "loss": 5.2745,
      "mean_token_accuracy": 0.17880961447954177,
      "num_tokens": 12095462.0,
      "step": 6970
    },
    {
      "entropy": 5.591189336776734,
      "epoch": 0.5426959735460027,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004975675606196177,
      "loss": 5.3784,
      "mean_token_accuracy": 0.17177537828683853,
      "num_tokens": 12103477.0,
      "step": 6975
    },
    {
      "entropy": 5.688836717605591,
      "epoch": 0.5430850029177203,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000497563494573711,
      "loss": 5.4961,
      "mean_token_accuracy": 0.1686207875609398,
      "num_tokens": 12112752.0,
      "step": 6980
    },
    {
      "entropy": 5.634454154968262,
      "epoch": 0.5434740322894378,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004975594251507575,
      "loss": 5.4356,
      "mean_token_accuracy": 0.16421550512313843,
      "num_tokens": 12121883.0,
      "step": 6985
    },
    {
      "entropy": 5.599533939361573,
      "epoch": 0.5438630616611554,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000497555352350819,
      "loss": 5.3698,
      "mean_token_accuracy": 0.1720726728439331,
      "num_tokens": 12130316.0,
      "step": 6990
    },
    {
      "entropy": 5.563831472396851,
      "epoch": 0.544252091032873,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004975512761739572,
      "loss": 5.3472,
      "mean_token_accuracy": 0.17171506583690643,
      "num_tokens": 12138856.0,
      "step": 6995
    },
    {
      "entropy": 5.540815782546997,
      "epoch": 0.5446411204045906,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000497547196620234,
      "loss": 5.4185,
      "mean_token_accuracy": 0.16540385335683822,
      "num_tokens": 12147933.0,
      "step": 7000
    },
    {
      "entropy": 5.535080766677856,
      "epoch": 0.5450301497763081,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004975431136897114,
      "loss": 5.3128,
      "mean_token_accuracy": 0.17722246497869493,
      "num_tokens": 12156529.0,
      "step": 7005
    },
    {
      "entropy": 5.554189872741699,
      "epoch": 0.5454191791480256,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004975390273824512,
      "loss": 5.2682,
      "mean_token_accuracy": 0.1779352605342865,
      "num_tokens": 12165233.0,
      "step": 7010
    },
    {
      "entropy": 5.632707118988037,
      "epoch": 0.5458082085197432,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004975349376985155,
      "loss": 5.4603,
      "mean_token_accuracy": 0.162058125436306,
      "num_tokens": 12173260.0,
      "step": 7015
    },
    {
      "entropy": 5.618736648559571,
      "epoch": 0.5461972378914608,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004975308446379663,
      "loss": 5.4678,
      "mean_token_accuracy": 0.16470558494329451,
      "num_tokens": 12181676.0,
      "step": 7020
    },
    {
      "entropy": 5.622576475143433,
      "epoch": 0.5465862672631784,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004975267482008657,
      "loss": 5.3808,
      "mean_token_accuracy": 0.16731790751218795,
      "num_tokens": 12189485.0,
      "step": 7025
    },
    {
      "entropy": 5.550467824935913,
      "epoch": 0.546975296634896,
      "grad_norm": 0.921875,
      "learning_rate": 0.000497522648387276,
      "loss": 5.3053,
      "mean_token_accuracy": 0.16856795102357863,
      "num_tokens": 12198429.0,
      "step": 7030
    },
    {
      "entropy": 5.581407833099365,
      "epoch": 0.5473643260066134,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004975185451972592,
      "loss": 5.4248,
      "mean_token_accuracy": 0.17068366408348085,
      "num_tokens": 12207404.0,
      "step": 7035
    },
    {
      "entropy": 5.656442451477051,
      "epoch": 0.547753355378331,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004975144386308776,
      "loss": 5.3961,
      "mean_token_accuracy": 0.16845954358577728,
      "num_tokens": 12216301.0,
      "step": 7040
    },
    {
      "entropy": 5.553708410263061,
      "epoch": 0.5481423847500486,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004975103286881936,
      "loss": 5.3653,
      "mean_token_accuracy": 0.17893509417772294,
      "num_tokens": 12224873.0,
      "step": 7045
    },
    {
      "entropy": 5.572200345993042,
      "epoch": 0.5485314141217662,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004975062153692696,
      "loss": 5.35,
      "mean_token_accuracy": 0.17169779539108276,
      "num_tokens": 12233439.0,
      "step": 7050
    },
    {
      "entropy": 5.623873567581176,
      "epoch": 0.5489204434934838,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004975020986741678,
      "loss": 5.4384,
      "mean_token_accuracy": 0.16889616549015046,
      "num_tokens": 12242235.0,
      "step": 7055
    },
    {
      "entropy": 5.548147678375244,
      "epoch": 0.5493094728652014,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004974979786029509,
      "loss": 5.3659,
      "mean_token_accuracy": 0.17649002820253373,
      "num_tokens": 12251295.0,
      "step": 7060
    },
    {
      "entropy": 5.584689331054688,
      "epoch": 0.5496985022369189,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004974938551556814,
      "loss": 5.29,
      "mean_token_accuracy": 0.17674800306558608,
      "num_tokens": 12260130.0,
      "step": 7065
    },
    {
      "entropy": 5.5405723571777346,
      "epoch": 0.5500875316086364,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004974897283324217,
      "loss": 5.3746,
      "mean_token_accuracy": 0.16957095116376877,
      "num_tokens": 12269542.0,
      "step": 7070
    },
    {
      "entropy": 5.5729736328125,
      "epoch": 0.550476560980354,
      "grad_norm": 1.0,
      "learning_rate": 0.0004974855981332343,
      "loss": 5.4366,
      "mean_token_accuracy": 0.17217087149620056,
      "num_tokens": 12278174.0,
      "step": 7075
    },
    {
      "entropy": 5.60262622833252,
      "epoch": 0.5508655903520716,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004974814645581823,
      "loss": 5.5248,
      "mean_token_accuracy": 0.16110899150371552,
      "num_tokens": 12287651.0,
      "step": 7080
    },
    {
      "entropy": 5.7456566333770756,
      "epoch": 0.5512546197237892,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004974773276073282,
      "loss": 5.5026,
      "mean_token_accuracy": 0.17000897228717804,
      "num_tokens": 12296009.0,
      "step": 7085
    },
    {
      "entropy": 5.58961238861084,
      "epoch": 0.5516436490955067,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004974731872807347,
      "loss": 5.382,
      "mean_token_accuracy": 0.1686262458562851,
      "num_tokens": 12305406.0,
      "step": 7090
    },
    {
      "entropy": 5.602985811233521,
      "epoch": 0.5520326784672243,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004974690435784647,
      "loss": 5.4292,
      "mean_token_accuracy": 0.17008743435144424,
      "num_tokens": 12314169.0,
      "step": 7095
    },
    {
      "entropy": 5.511575031280517,
      "epoch": 0.5524217078389418,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004974648965005811,
      "loss": 5.3067,
      "mean_token_accuracy": 0.17620205134153366,
      "num_tokens": 12323019.0,
      "step": 7100
    },
    {
      "entropy": 5.565052938461304,
      "epoch": 0.5528107372106594,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004974607460471466,
      "loss": 5.4161,
      "mean_token_accuracy": 0.1709034577012062,
      "num_tokens": 12331491.0,
      "step": 7105
    },
    {
      "entropy": 5.630934715270996,
      "epoch": 0.553199766582377,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004974565922182246,
      "loss": 5.37,
      "mean_token_accuracy": 0.1731913149356842,
      "num_tokens": 12339647.0,
      "step": 7110
    },
    {
      "entropy": 5.599842071533203,
      "epoch": 0.5535887959540945,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004974524350138777,
      "loss": 5.3681,
      "mean_token_accuracy": 0.17500039935112,
      "num_tokens": 12348469.0,
      "step": 7115
    },
    {
      "entropy": 5.621744680404663,
      "epoch": 0.5539778253258121,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004974482744341693,
      "loss": 5.3558,
      "mean_token_accuracy": 0.17349795550107955,
      "num_tokens": 12357168.0,
      "step": 7120
    },
    {
      "entropy": 5.516009521484375,
      "epoch": 0.5543668546975297,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004974441104791624,
      "loss": 5.3509,
      "mean_token_accuracy": 0.17069997638463974,
      "num_tokens": 12365750.0,
      "step": 7125
    },
    {
      "entropy": 5.574317216873169,
      "epoch": 0.5547558840692473,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004974399431489201,
      "loss": 5.3804,
      "mean_token_accuracy": 0.1693907469511032,
      "num_tokens": 12374086.0,
      "step": 7130
    },
    {
      "entropy": 5.614137268066406,
      "epoch": 0.5551449134409648,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004974357724435057,
      "loss": 5.3291,
      "mean_token_accuracy": 0.17812907248735427,
      "num_tokens": 12381997.0,
      "step": 7135
    },
    {
      "entropy": 5.610338354110718,
      "epoch": 0.5555339428126823,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004974315983629825,
      "loss": 5.43,
      "mean_token_accuracy": 0.17456079125404358,
      "num_tokens": 12391083.0,
      "step": 7140
    },
    {
      "entropy": 5.585184764862061,
      "epoch": 0.5559229721843999,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004974274209074139,
      "loss": 5.3485,
      "mean_token_accuracy": 0.17291898876428605,
      "num_tokens": 12400016.0,
      "step": 7145
    },
    {
      "entropy": 5.617244863510132,
      "epoch": 0.5563120015561175,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004974232400768632,
      "loss": 5.4466,
      "mean_token_accuracy": 0.17553191184997557,
      "num_tokens": 12409255.0,
      "step": 7150
    },
    {
      "entropy": 5.61101803779602,
      "epoch": 0.5567010309278351,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004974190558713939,
      "loss": 5.3605,
      "mean_token_accuracy": 0.17722705900669097,
      "num_tokens": 12417765.0,
      "step": 7155
    },
    {
      "entropy": 5.62712082862854,
      "epoch": 0.5570900602995527,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004974148682910694,
      "loss": 5.5898,
      "mean_token_accuracy": 0.15361103117465974,
      "num_tokens": 12427588.0,
      "step": 7160
    },
    {
      "entropy": 5.552091979980469,
      "epoch": 0.5574790896712701,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004974106773359533,
      "loss": 5.2687,
      "mean_token_accuracy": 0.17982016503810883,
      "num_tokens": 12435773.0,
      "step": 7165
    },
    {
      "entropy": 5.584134578704834,
      "epoch": 0.5578681190429877,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004974064830061091,
      "loss": 5.3916,
      "mean_token_accuracy": 0.17531634122133255,
      "num_tokens": 12443837.0,
      "step": 7170
    },
    {
      "entropy": 5.531421375274658,
      "epoch": 0.5582571484147053,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004974022853016006,
      "loss": 5.3846,
      "mean_token_accuracy": 0.1778782993555069,
      "num_tokens": 12452945.0,
      "step": 7175
    },
    {
      "entropy": 5.565284490585327,
      "epoch": 0.5586461777864229,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004973980842224913,
      "loss": 5.2949,
      "mean_token_accuracy": 0.1806413859128952,
      "num_tokens": 12461008.0,
      "step": 7180
    },
    {
      "entropy": 5.514020681381226,
      "epoch": 0.5590352071581405,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004973938797688452,
      "loss": 5.3115,
      "mean_token_accuracy": 0.17192093282938004,
      "num_tokens": 12469314.0,
      "step": 7185
    },
    {
      "entropy": 5.555478811264038,
      "epoch": 0.559424236529858,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004973896719407259,
      "loss": 5.3318,
      "mean_token_accuracy": 0.17091093063354493,
      "num_tokens": 12477540.0,
      "step": 7190
    },
    {
      "entropy": 5.5698850631713865,
      "epoch": 0.5598132659015755,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004973854607381972,
      "loss": 5.3996,
      "mean_token_accuracy": 0.17508567422628402,
      "num_tokens": 12486894.0,
      "step": 7195
    },
    {
      "entropy": 5.629411077499389,
      "epoch": 0.5602022952732931,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004973812461613232,
      "loss": 5.4008,
      "mean_token_accuracy": 0.1661403402686119,
      "num_tokens": 12496213.0,
      "step": 7200
    },
    {
      "entropy": 5.57621808052063,
      "epoch": 0.5605913246450107,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004973770282101677,
      "loss": 5.3757,
      "mean_token_accuracy": 0.1691313549876213,
      "num_tokens": 12504160.0,
      "step": 7205
    },
    {
      "entropy": 5.473678302764893,
      "epoch": 0.5609803540167283,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004973728068847947,
      "loss": 5.315,
      "mean_token_accuracy": 0.17716611474752425,
      "num_tokens": 12511915.0,
      "step": 7210
    },
    {
      "entropy": 5.58756537437439,
      "epoch": 0.5613693833884458,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004973685821852684,
      "loss": 5.3553,
      "mean_token_accuracy": 0.17398306727409363,
      "num_tokens": 12520400.0,
      "step": 7215
    },
    {
      "entropy": 5.537965536117554,
      "epoch": 0.5617584127601634,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004973643541116528,
      "loss": 5.2707,
      "mean_token_accuracy": 0.1798969566822052,
      "num_tokens": 12528724.0,
      "step": 7220
    },
    {
      "entropy": 5.598689317703247,
      "epoch": 0.562147442131881,
      "grad_norm": 0.984375,
      "learning_rate": 0.000497360122664012,
      "loss": 5.4587,
      "mean_token_accuracy": 0.17003613114356994,
      "num_tokens": 12537378.0,
      "step": 7225
    },
    {
      "entropy": 5.587683868408203,
      "epoch": 0.5625364715035985,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004973558878424104,
      "loss": 5.3778,
      "mean_token_accuracy": 0.17496745586395263,
      "num_tokens": 12546080.0,
      "step": 7230
    },
    {
      "entropy": 5.629578447341919,
      "epoch": 0.5629255008753161,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004973516496469119,
      "loss": 5.3956,
      "mean_token_accuracy": 0.17400780469179153,
      "num_tokens": 12554526.0,
      "step": 7235
    },
    {
      "entropy": 5.561227893829345,
      "epoch": 0.5633145302470336,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004973474080775811,
      "loss": 5.3575,
      "mean_token_accuracy": 0.17506588250398636,
      "num_tokens": 12563018.0,
      "step": 7240
    },
    {
      "entropy": 5.534848070144653,
      "epoch": 0.5637035596187512,
      "grad_norm": 1.0,
      "learning_rate": 0.0004973431631344824,
      "loss": 5.4094,
      "mean_token_accuracy": 0.17164857685565948,
      "num_tokens": 12571611.0,
      "step": 7245
    },
    {
      "entropy": 5.643917846679687,
      "epoch": 0.5640925889904688,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00049733891481768,
      "loss": 5.4894,
      "mean_token_accuracy": 0.16730506122112274,
      "num_tokens": 12580168.0,
      "step": 7250
    },
    {
      "entropy": 5.554508209228516,
      "epoch": 0.5644816183621864,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004973346631272384,
      "loss": 5.3791,
      "mean_token_accuracy": 0.17357684224843978,
      "num_tokens": 12589843.0,
      "step": 7255
    },
    {
      "entropy": 5.651784038543701,
      "epoch": 0.5648706477339039,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004973304080632222,
      "loss": 5.4276,
      "mean_token_accuracy": 0.16982462406158447,
      "num_tokens": 12598219.0,
      "step": 7260
    },
    {
      "entropy": 5.574033308029175,
      "epoch": 0.5652596771056215,
      "grad_norm": 0.984375,
      "learning_rate": 0.000497326149625696,
      "loss": 5.4189,
      "mean_token_accuracy": 0.17014081329107283,
      "num_tokens": 12606665.0,
      "step": 7265
    },
    {
      "entropy": 5.556108713150024,
      "epoch": 0.565648706477339,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004973218878147244,
      "loss": 5.3435,
      "mean_token_accuracy": 0.17241014540195465,
      "num_tokens": 12615062.0,
      "step": 7270
    },
    {
      "entropy": 5.652842807769775,
      "epoch": 0.5660377358490566,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004973176226303719,
      "loss": 5.4257,
      "mean_token_accuracy": 0.17043905407190324,
      "num_tokens": 12623923.0,
      "step": 7275
    },
    {
      "entropy": 5.581016731262207,
      "epoch": 0.5664267652207742,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004973133540727033,
      "loss": 5.352,
      "mean_token_accuracy": 0.1738426610827446,
      "num_tokens": 12632579.0,
      "step": 7280
    },
    {
      "entropy": 5.52339358329773,
      "epoch": 0.5668157945924918,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004973090821417835,
      "loss": 5.3128,
      "mean_token_accuracy": 0.17193036377429963,
      "num_tokens": 12640847.0,
      "step": 7285
    },
    {
      "entropy": 5.632245826721191,
      "epoch": 0.5672048239642093,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004973048068376772,
      "loss": 5.2943,
      "mean_token_accuracy": 0.1762019619345665,
      "num_tokens": 12649020.0,
      "step": 7290
    },
    {
      "entropy": 5.499836397171021,
      "epoch": 0.5675938533359268,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004973005281604492,
      "loss": 5.2367,
      "mean_token_accuracy": 0.18058727085590362,
      "num_tokens": 12657569.0,
      "step": 7295
    },
    {
      "entropy": 5.656816291809082,
      "epoch": 0.5679828827076444,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004972962461101646,
      "loss": 5.4805,
      "mean_token_accuracy": 0.16379775926470758,
      "num_tokens": 12666681.0,
      "step": 7300
    },
    {
      "entropy": 5.581071615219116,
      "epoch": 0.568371912079362,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004972919606868883,
      "loss": 5.3022,
      "mean_token_accuracy": 0.17431648224592208,
      "num_tokens": 12675210.0,
      "step": 7305
    },
    {
      "entropy": 5.603531980514527,
      "epoch": 0.5687609414510796,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004972876718906852,
      "loss": 5.3581,
      "mean_token_accuracy": 0.17259342074394227,
      "num_tokens": 12684019.0,
      "step": 7310
    },
    {
      "entropy": 5.547083616256714,
      "epoch": 0.5691499708227972,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004972833797216206,
      "loss": 5.3968,
      "mean_token_accuracy": 0.1641467198729515,
      "num_tokens": 12692891.0,
      "step": 7315
    },
    {
      "entropy": 5.558692789077758,
      "epoch": 0.5695390001945146,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004972790841797595,
      "loss": 5.3082,
      "mean_token_accuracy": 0.172683946788311,
      "num_tokens": 12700606.0,
      "step": 7320
    },
    {
      "entropy": 5.554307365417481,
      "epoch": 0.5699280295662322,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000497274785265167,
      "loss": 5.4055,
      "mean_token_accuracy": 0.1702592447400093,
      "num_tokens": 12709116.0,
      "step": 7325
    },
    {
      "entropy": 5.570007562637329,
      "epoch": 0.5703170589379498,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004972704829779086,
      "loss": 5.3764,
      "mean_token_accuracy": 0.172584468126297,
      "num_tokens": 12717484.0,
      "step": 7330
    },
    {
      "entropy": 5.5243353843688965,
      "epoch": 0.5707060883096674,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004972661773180492,
      "loss": 5.2927,
      "mean_token_accuracy": 0.17464651316404342,
      "num_tokens": 12726216.0,
      "step": 7335
    },
    {
      "entropy": 5.566395998001099,
      "epoch": 0.571095117681385,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004972618682856545,
      "loss": 5.2781,
      "mean_token_accuracy": 0.18160381466150283,
      "num_tokens": 12735312.0,
      "step": 7340
    },
    {
      "entropy": 5.552312850952148,
      "epoch": 0.5714841470531025,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004972575558807896,
      "loss": 5.41,
      "mean_token_accuracy": 0.16768458634614944,
      "num_tokens": 12744641.0,
      "step": 7345
    },
    {
      "entropy": 5.585736989974976,
      "epoch": 0.57187317642482,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004972532401035201,
      "loss": 5.387,
      "mean_token_accuracy": 0.1750989004969597,
      "num_tokens": 12753951.0,
      "step": 7350
    },
    {
      "entropy": 5.6440749168396,
      "epoch": 0.5722622057965376,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004972489209539114,
      "loss": 5.3903,
      "mean_token_accuracy": 0.1683877557516098,
      "num_tokens": 12762571.0,
      "step": 7355
    },
    {
      "entropy": 5.644950723648071,
      "epoch": 0.5726512351682552,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004972445984320292,
      "loss": 5.4502,
      "mean_token_accuracy": 0.1704435646533966,
      "num_tokens": 12771275.0,
      "step": 7360
    },
    {
      "entropy": 5.5334583759307865,
      "epoch": 0.5730402645399728,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004972402725379388,
      "loss": 5.3633,
      "mean_token_accuracy": 0.17209473252296448,
      "num_tokens": 12780488.0,
      "step": 7365
    },
    {
      "entropy": 5.601672744750976,
      "epoch": 0.5734292939116903,
      "grad_norm": 1.09375,
      "learning_rate": 0.000497235943271706,
      "loss": 5.3753,
      "mean_token_accuracy": 0.17395491898059845,
      "num_tokens": 12788588.0,
      "step": 7370
    },
    {
      "entropy": 5.556140327453614,
      "epoch": 0.5738183232834079,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004972316106333966,
      "loss": 5.3261,
      "mean_token_accuracy": 0.17215914875268937,
      "num_tokens": 12797010.0,
      "step": 7375
    },
    {
      "entropy": 5.592095041275025,
      "epoch": 0.5742073526551255,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004972272746230761,
      "loss": 5.4174,
      "mean_token_accuracy": 0.1703107550740242,
      "num_tokens": 12805430.0,
      "step": 7380
    },
    {
      "entropy": 5.661344909667969,
      "epoch": 0.574596382026843,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004972229352408104,
      "loss": 5.389,
      "mean_token_accuracy": 0.1679370880126953,
      "num_tokens": 12813912.0,
      "step": 7385
    },
    {
      "entropy": 5.65712251663208,
      "epoch": 0.5749854113985606,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004972185924866655,
      "loss": 5.4024,
      "mean_token_accuracy": 0.1678002208471298,
      "num_tokens": 12822421.0,
      "step": 7390
    },
    {
      "entropy": 5.513266229629517,
      "epoch": 0.5753744407702781,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004972142463607071,
      "loss": 5.2814,
      "mean_token_accuracy": 0.17622633278369904,
      "num_tokens": 12830670.0,
      "step": 7395
    },
    {
      "entropy": 5.593383073806763,
      "epoch": 0.5757634701419957,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004972098968630012,
      "loss": 5.4153,
      "mean_token_accuracy": 0.1742699220776558,
      "num_tokens": 12839887.0,
      "step": 7400
    },
    {
      "entropy": 5.501554298400879,
      "epoch": 0.5761524995137133,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004972055439936137,
      "loss": 5.3132,
      "mean_token_accuracy": 0.1788620099425316,
      "num_tokens": 12848028.0,
      "step": 7405
    },
    {
      "entropy": 5.467774200439453,
      "epoch": 0.5765415288854309,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004972011877526109,
      "loss": 5.3136,
      "mean_token_accuracy": 0.17193938791751862,
      "num_tokens": 12857248.0,
      "step": 7410
    },
    {
      "entropy": 5.593406009674072,
      "epoch": 0.5769305582571485,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004971968281400587,
      "loss": 5.3163,
      "mean_token_accuracy": 0.1750161498785019,
      "num_tokens": 12866352.0,
      "step": 7415
    },
    {
      "entropy": 5.6203666687011715,
      "epoch": 0.5773195876288659,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004971924651560233,
      "loss": 5.4137,
      "mean_token_accuracy": 0.16567861288785934,
      "num_tokens": 12874548.0,
      "step": 7420
    },
    {
      "entropy": 5.545903587341309,
      "epoch": 0.5777086170005835,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004971880988005709,
      "loss": 5.2666,
      "mean_token_accuracy": 0.17062354534864427,
      "num_tokens": 12883172.0,
      "step": 7425
    },
    {
      "entropy": 5.480578327178955,
      "epoch": 0.5780976463723011,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004971837290737677,
      "loss": 5.3033,
      "mean_token_accuracy": 0.18337597101926803,
      "num_tokens": 12891776.0,
      "step": 7430
    },
    {
      "entropy": 5.525628614425659,
      "epoch": 0.5784866757440187,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004971793559756801,
      "loss": 5.2732,
      "mean_token_accuracy": 0.18172142803668975,
      "num_tokens": 12900530.0,
      "step": 7435
    },
    {
      "entropy": 5.497667694091797,
      "epoch": 0.5788757051157363,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004971749795063744,
      "loss": 5.3071,
      "mean_token_accuracy": 0.17193772196769713,
      "num_tokens": 12909260.0,
      "step": 7440
    },
    {
      "entropy": 5.526607894897461,
      "epoch": 0.5792647344874537,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000497170599665917,
      "loss": 5.3047,
      "mean_token_accuracy": 0.17156598716974258,
      "num_tokens": 12918143.0,
      "step": 7445
    },
    {
      "entropy": 5.60771598815918,
      "epoch": 0.5796537638591713,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004971662164543743,
      "loss": 5.3756,
      "mean_token_accuracy": 0.17291728258132935,
      "num_tokens": 12926361.0,
      "step": 7450
    },
    {
      "entropy": 5.524640893936157,
      "epoch": 0.5800427932308889,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004971618298718131,
      "loss": 5.3106,
      "mean_token_accuracy": 0.17592185884714126,
      "num_tokens": 12935080.0,
      "step": 7455
    },
    {
      "entropy": 5.616127300262451,
      "epoch": 0.5804318226026065,
      "grad_norm": 1.0,
      "learning_rate": 0.0004971574399182995,
      "loss": 5.4667,
      "mean_token_accuracy": 0.1685623988509178,
      "num_tokens": 12944078.0,
      "step": 7460
    },
    {
      "entropy": 5.599328470230103,
      "epoch": 0.5808208519743241,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004971530465939005,
      "loss": 5.3752,
      "mean_token_accuracy": 0.1741485044360161,
      "num_tokens": 12952800.0,
      "step": 7465
    },
    {
      "entropy": 5.6505719184875485,
      "epoch": 0.5812098813460417,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004971486498986825,
      "loss": 5.4673,
      "mean_token_accuracy": 0.16936276257038116,
      "num_tokens": 12962402.0,
      "step": 7470
    },
    {
      "entropy": 5.5517088890075685,
      "epoch": 0.5815989107177592,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004971442498327124,
      "loss": 5.2465,
      "mean_token_accuracy": 0.18356094360351563,
      "num_tokens": 12971116.0,
      "step": 7475
    },
    {
      "entropy": 5.464197635650635,
      "epoch": 0.5819879400894767,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004971398463960568,
      "loss": 5.343,
      "mean_token_accuracy": 0.16875373274087907,
      "num_tokens": 12979570.0,
      "step": 7480
    },
    {
      "entropy": 5.592851352691651,
      "epoch": 0.5823769694611943,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004971354395887827,
      "loss": 5.367,
      "mean_token_accuracy": 0.17882476449012757,
      "num_tokens": 12987942.0,
      "step": 7485
    },
    {
      "entropy": 5.575359201431274,
      "epoch": 0.5827659988329119,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004971310294109569,
      "loss": 5.395,
      "mean_token_accuracy": 0.1713891863822937,
      "num_tokens": 12996499.0,
      "step": 7490
    },
    {
      "entropy": 5.506797552108765,
      "epoch": 0.5831550282046295,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004971266158626461,
      "loss": 5.3314,
      "mean_token_accuracy": 0.17794760316610336,
      "num_tokens": 13005622.0,
      "step": 7495
    },
    {
      "entropy": 5.555897760391235,
      "epoch": 0.583544057576347,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004971221989439177,
      "loss": 5.3846,
      "mean_token_accuracy": 0.16468334197998047,
      "num_tokens": 13014417.0,
      "step": 7500
    },
    {
      "entropy": 5.545801782608033,
      "epoch": 0.5839330869480646,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004971177786548384,
      "loss": 5.2885,
      "mean_token_accuracy": 0.17062005549669265,
      "num_tokens": 13022887.0,
      "step": 7505
    },
    {
      "entropy": 5.531611633300781,
      "epoch": 0.5843221163197821,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004971133549954753,
      "loss": 5.3884,
      "mean_token_accuracy": 0.17030563056468964,
      "num_tokens": 13031676.0,
      "step": 7510
    },
    {
      "entropy": 5.575408124923706,
      "epoch": 0.5847111456914997,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004971089279658954,
      "loss": 5.3555,
      "mean_token_accuracy": 0.17267583310604095,
      "num_tokens": 13040489.0,
      "step": 7515
    },
    {
      "entropy": 5.615947246551514,
      "epoch": 0.5851001750632173,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004971044975661663,
      "loss": 5.277,
      "mean_token_accuracy": 0.17653228640556334,
      "num_tokens": 13048280.0,
      "step": 7520
    },
    {
      "entropy": 5.470430040359497,
      "epoch": 0.5854892044349348,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004971000637963549,
      "loss": 5.2568,
      "mean_token_accuracy": 0.18650708198547364,
      "num_tokens": 13056471.0,
      "step": 7525
    },
    {
      "entropy": 5.560556650161743,
      "epoch": 0.5858782338066524,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004970956266565285,
      "loss": 5.444,
      "mean_token_accuracy": 0.16605811268091203,
      "num_tokens": 13065289.0,
      "step": 7530
    },
    {
      "entropy": 5.640527200698853,
      "epoch": 0.58626726317837,
      "grad_norm": 1.125,
      "learning_rate": 0.0004970911861467544,
      "loss": 5.4562,
      "mean_token_accuracy": 0.1743879571557045,
      "num_tokens": 13073648.0,
      "step": 7535
    },
    {
      "entropy": 5.626132869720459,
      "epoch": 0.5866562925500876,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004970867422671002,
      "loss": 5.4657,
      "mean_token_accuracy": 0.17313232123851777,
      "num_tokens": 13082205.0,
      "step": 7540
    },
    {
      "entropy": 5.591260528564453,
      "epoch": 0.5870453219218051,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000497082295017633,
      "loss": 5.3345,
      "mean_token_accuracy": 0.17170399129390718,
      "num_tokens": 13090970.0,
      "step": 7545
    },
    {
      "entropy": 5.584366416931152,
      "epoch": 0.5874343512935226,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004970778443984206,
      "loss": 5.2623,
      "mean_token_accuracy": 0.1804880440235138,
      "num_tokens": 13098290.0,
      "step": 7550
    },
    {
      "entropy": 5.4597996234893795,
      "epoch": 0.5878233806652402,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004970733904095303,
      "loss": 5.2084,
      "mean_token_accuracy": 0.18538361936807632,
      "num_tokens": 13106971.0,
      "step": 7555
    },
    {
      "entropy": 5.632787656784058,
      "epoch": 0.5882124100369578,
      "grad_norm": 1.0,
      "learning_rate": 0.0004970689330510298,
      "loss": 5.4741,
      "mean_token_accuracy": 0.169888673722744,
      "num_tokens": 13115742.0,
      "step": 7560
    },
    {
      "entropy": 5.617022943496704,
      "epoch": 0.5886014394086754,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004970644723229868,
      "loss": 5.4583,
      "mean_token_accuracy": 0.17384405434131622,
      "num_tokens": 13124192.0,
      "step": 7565
    },
    {
      "entropy": 5.529296731948852,
      "epoch": 0.588990468780393,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004970600082254687,
      "loss": 5.2828,
      "mean_token_accuracy": 0.17330437451601027,
      "num_tokens": 13133510.0,
      "step": 7570
    },
    {
      "entropy": 5.528960418701172,
      "epoch": 0.5893794981521104,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004970555407585436,
      "loss": 5.3909,
      "mean_token_accuracy": 0.1745387554168701,
      "num_tokens": 13141704.0,
      "step": 7575
    },
    {
      "entropy": 5.524944829940796,
      "epoch": 0.589768527523828,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000497051069922279,
      "loss": 5.2617,
      "mean_token_accuracy": 0.17319603860378266,
      "num_tokens": 13150559.0,
      "step": 7580
    },
    {
      "entropy": 5.5150316715240475,
      "epoch": 0.5901575568955456,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004970465957167429,
      "loss": 5.3414,
      "mean_token_accuracy": 0.1729081988334656,
      "num_tokens": 13159428.0,
      "step": 7585
    },
    {
      "entropy": 5.554745483398437,
      "epoch": 0.5905465862672632,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000497042118142003,
      "loss": 5.4786,
      "mean_token_accuracy": 0.16202798634767532,
      "num_tokens": 13168299.0,
      "step": 7590
    },
    {
      "entropy": 5.585139608383178,
      "epoch": 0.5909356156389808,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004970376371981275,
      "loss": 5.3315,
      "mean_token_accuracy": 0.17860794961452484,
      "num_tokens": 13176906.0,
      "step": 7595
    },
    {
      "entropy": 5.622454357147217,
      "epoch": 0.5913246450106983,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004970331528851842,
      "loss": 5.4245,
      "mean_token_accuracy": 0.17163319438695906,
      "num_tokens": 13186180.0,
      "step": 7600
    },
    {
      "entropy": 5.4992204189300535,
      "epoch": 0.5917136743824158,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004970286652032412,
      "loss": 5.2581,
      "mean_token_accuracy": 0.18248205333948136,
      "num_tokens": 13195111.0,
      "step": 7605
    },
    {
      "entropy": 5.573828792572021,
      "epoch": 0.5921027037541334,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004970241741523667,
      "loss": 5.3293,
      "mean_token_accuracy": 0.17229389101266862,
      "num_tokens": 13204170.0,
      "step": 7610
    },
    {
      "entropy": 5.5608381748199465,
      "epoch": 0.592491733125851,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004970196797326286,
      "loss": 5.3371,
      "mean_token_accuracy": 0.17119126170873641,
      "num_tokens": 13213479.0,
      "step": 7615
    },
    {
      "entropy": 5.557682085037231,
      "epoch": 0.5928807624975686,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004970151819440955,
      "loss": 5.3634,
      "mean_token_accuracy": 0.17168747633695602,
      "num_tokens": 13222182.0,
      "step": 7620
    },
    {
      "entropy": 5.498529291152954,
      "epoch": 0.5932697918692861,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004970106807868351,
      "loss": 5.238,
      "mean_token_accuracy": 0.180169041454792,
      "num_tokens": 13230681.0,
      "step": 7625
    },
    {
      "entropy": 5.65567421913147,
      "epoch": 0.5936588212410037,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000497006176260916,
      "loss": 5.4252,
      "mean_token_accuracy": 0.1727098271250725,
      "num_tokens": 13239254.0,
      "step": 7630
    },
    {
      "entropy": 5.519259929656982,
      "epoch": 0.5940478506127213,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004970016683664068,
      "loss": 5.2946,
      "mean_token_accuracy": 0.1789883628487587,
      "num_tokens": 13248317.0,
      "step": 7635
    },
    {
      "entropy": 5.641921758651733,
      "epoch": 0.5944368799844388,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004969971571033754,
      "loss": 5.3517,
      "mean_token_accuracy": 0.17641935795545577,
      "num_tokens": 13256977.0,
      "step": 7640
    },
    {
      "entropy": 5.57003345489502,
      "epoch": 0.5948259093561564,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004969926424718906,
      "loss": 5.4162,
      "mean_token_accuracy": 0.17275196313858032,
      "num_tokens": 13266012.0,
      "step": 7645
    },
    {
      "entropy": 5.5508636951446535,
      "epoch": 0.5952149387278739,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004969881244720207,
      "loss": 5.4352,
      "mean_token_accuracy": 0.174187308549881,
      "num_tokens": 13275248.0,
      "step": 7650
    },
    {
      "entropy": 5.568617630004883,
      "epoch": 0.5956039680995915,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004969836031038343,
      "loss": 5.2945,
      "mean_token_accuracy": 0.16968248039484024,
      "num_tokens": 13284235.0,
      "step": 7655
    },
    {
      "entropy": 5.593735265731811,
      "epoch": 0.5959929974713091,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004969790783674002,
      "loss": 5.3737,
      "mean_token_accuracy": 0.16619754135608672,
      "num_tokens": 13292423.0,
      "step": 7660
    },
    {
      "entropy": 5.565545845031738,
      "epoch": 0.5963820268430267,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004969745502627868,
      "loss": 5.4047,
      "mean_token_accuracy": 0.165803362429142,
      "num_tokens": 13300936.0,
      "step": 7665
    },
    {
      "entropy": 5.525141096115112,
      "epoch": 0.5967710562147442,
      "grad_norm": 1.0,
      "learning_rate": 0.0004969700187900629,
      "loss": 5.1976,
      "mean_token_accuracy": 0.18360133320093155,
      "num_tokens": 13308881.0,
      "step": 7670
    },
    {
      "entropy": 5.580707693099976,
      "epoch": 0.5971600855864618,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004969654839492973,
      "loss": 5.3862,
      "mean_token_accuracy": 0.17280948162078857,
      "num_tokens": 13317864.0,
      "step": 7675
    },
    {
      "entropy": 5.577540302276612,
      "epoch": 0.5975491149581793,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004969609457405588,
      "loss": 5.356,
      "mean_token_accuracy": 0.17092010527849197,
      "num_tokens": 13326990.0,
      "step": 7680
    },
    {
      "entropy": 5.615563678741455,
      "epoch": 0.5979381443298969,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004969564041639163,
      "loss": 5.5341,
      "mean_token_accuracy": 0.16558944284915925,
      "num_tokens": 13336437.0,
      "step": 7685
    },
    {
      "entropy": 5.5491780757904055,
      "epoch": 0.5983271737016145,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004969518592194386,
      "loss": 5.3585,
      "mean_token_accuracy": 0.17135425359010698,
      "num_tokens": 13345368.0,
      "step": 7690
    },
    {
      "entropy": 5.52543158531189,
      "epoch": 0.5987162030733321,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004969473109071946,
      "loss": 5.3921,
      "mean_token_accuracy": 0.17189282327890396,
      "num_tokens": 13354171.0,
      "step": 7695
    },
    {
      "entropy": 5.620519161224365,
      "epoch": 0.5991052324450497,
      "grad_norm": 1.0,
      "learning_rate": 0.0004969427592272535,
      "loss": 5.3866,
      "mean_token_accuracy": 0.16920358836650848,
      "num_tokens": 13362912.0,
      "step": 7700
    },
    {
      "entropy": 5.61610221862793,
      "epoch": 0.5994942618167671,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004969382041796843,
      "loss": 5.284,
      "mean_token_accuracy": 0.1749517321586609,
      "num_tokens": 13370913.0,
      "step": 7705
    },
    {
      "entropy": 5.5636087417602536,
      "epoch": 0.5998832911884847,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000496933645764556,
      "loss": 5.4512,
      "mean_token_accuracy": 0.17191523611545562,
      "num_tokens": 13379557.0,
      "step": 7710
    },
    {
      "entropy": 5.511366653442383,
      "epoch": 0.6002723205602023,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004969290839819381,
      "loss": 5.3387,
      "mean_token_accuracy": 0.17561249881982804,
      "num_tokens": 13388129.0,
      "step": 7715
    },
    {
      "entropy": 5.601404523849487,
      "epoch": 0.6006613499319199,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004969245188318994,
      "loss": 5.4723,
      "mean_token_accuracy": 0.16843156069517135,
      "num_tokens": 13396713.0,
      "step": 7720
    },
    {
      "entropy": 5.617475509643555,
      "epoch": 0.6010503793036375,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004969199503145093,
      "loss": 5.4052,
      "mean_token_accuracy": 0.17606594711542128,
      "num_tokens": 13406171.0,
      "step": 7725
    },
    {
      "entropy": 5.523739337921143,
      "epoch": 0.601439408675355,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004969153784298374,
      "loss": 5.3184,
      "mean_token_accuracy": 0.1757847398519516,
      "num_tokens": 13415107.0,
      "step": 7730
    },
    {
      "entropy": 5.537905883789063,
      "epoch": 0.6018284380470725,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004969108031779527,
      "loss": 5.4256,
      "mean_token_accuracy": 0.17127961814403533,
      "num_tokens": 13423533.0,
      "step": 7735
    },
    {
      "entropy": 5.638061428070069,
      "epoch": 0.6022174674187901,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004969062245589247,
      "loss": 5.4084,
      "mean_token_accuracy": 0.16514907032251358,
      "num_tokens": 13432798.0,
      "step": 7740
    },
    {
      "entropy": 5.627174711227417,
      "epoch": 0.6026064967905077,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004969016425728231,
      "loss": 5.3531,
      "mean_token_accuracy": 0.17923811078071594,
      "num_tokens": 13441123.0,
      "step": 7745
    },
    {
      "entropy": 5.5383693218231205,
      "epoch": 0.6029955261622253,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004968970572197172,
      "loss": 5.28,
      "mean_token_accuracy": 0.17662344723939896,
      "num_tokens": 13449548.0,
      "step": 7750
    },
    {
      "entropy": 5.557244777679443,
      "epoch": 0.6033845555339428,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004968924684996765,
      "loss": 5.3605,
      "mean_token_accuracy": 0.17326037883758544,
      "num_tokens": 13458022.0,
      "step": 7755
    },
    {
      "entropy": 5.554460048675537,
      "epoch": 0.6037735849056604,
      "grad_norm": 0.953125,
      "learning_rate": 0.000496887876412771,
      "loss": 5.3267,
      "mean_token_accuracy": 0.17779916226863862,
      "num_tokens": 13466871.0,
      "step": 7760
    },
    {
      "entropy": 5.563388442993164,
      "epoch": 0.6041626142773779,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004968832809590701,
      "loss": 5.3464,
      "mean_token_accuracy": 0.17357995510101318,
      "num_tokens": 13475659.0,
      "step": 7765
    },
    {
      "entropy": 5.582509183883667,
      "epoch": 0.6045516436490955,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004968786821386435,
      "loss": 5.3704,
      "mean_token_accuracy": 0.17079875618219376,
      "num_tokens": 13484146.0,
      "step": 7770
    },
    {
      "entropy": 5.546985483169555,
      "epoch": 0.6049406730208131,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004968740799515611,
      "loss": 5.3697,
      "mean_token_accuracy": 0.17335487455129622,
      "num_tokens": 13492831.0,
      "step": 7775
    },
    {
      "entropy": 5.542489051818848,
      "epoch": 0.6053297023925306,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004968694743978926,
      "loss": 5.3335,
      "mean_token_accuracy": 0.1755594164133072,
      "num_tokens": 13501640.0,
      "step": 7780
    },
    {
      "entropy": 5.6146242141723635,
      "epoch": 0.6057187317642482,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004968648654777081,
      "loss": 5.3546,
      "mean_token_accuracy": 0.17896517217159272,
      "num_tokens": 13509816.0,
      "step": 7785
    },
    {
      "entropy": 5.619046640396118,
      "epoch": 0.6061077611359658,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004968602531910773,
      "loss": 5.4292,
      "mean_token_accuracy": 0.1723052129149437,
      "num_tokens": 13518384.0,
      "step": 7790
    },
    {
      "entropy": 5.5350583553314205,
      "epoch": 0.6064967905076833,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004968556375380704,
      "loss": 5.2788,
      "mean_token_accuracy": 0.17632309943437577,
      "num_tokens": 13527115.0,
      "step": 7795
    },
    {
      "entropy": 5.583134269714355,
      "epoch": 0.6068858198794009,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004968510185187571,
      "loss": 5.5069,
      "mean_token_accuracy": 0.17107459753751755,
      "num_tokens": 13535743.0,
      "step": 7800
    },
    {
      "entropy": 5.560557794570923,
      "epoch": 0.6072748492511184,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000496846396133208,
      "loss": 5.3208,
      "mean_token_accuracy": 0.17716536968946456,
      "num_tokens": 13544167.0,
      "step": 7805
    },
    {
      "entropy": 5.574625539779663,
      "epoch": 0.607663878622836,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004968417703814928,
      "loss": 5.3684,
      "mean_token_accuracy": 0.17458470314741134,
      "num_tokens": 13552291.0,
      "step": 7810
    },
    {
      "entropy": 5.602114009857178,
      "epoch": 0.6080529079945536,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004968371412636818,
      "loss": 5.4179,
      "mean_token_accuracy": 0.16921926140785218,
      "num_tokens": 13561185.0,
      "step": 7815
    },
    {
      "entropy": 5.629643154144287,
      "epoch": 0.6084419373662712,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004968325087798453,
      "loss": 5.3057,
      "mean_token_accuracy": 0.1815168261528015,
      "num_tokens": 13569815.0,
      "step": 7820
    },
    {
      "entropy": 5.555459213256836,
      "epoch": 0.6088309667379888,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004968278729300536,
      "loss": 5.4164,
      "mean_token_accuracy": 0.17127194106578827,
      "num_tokens": 13578882.0,
      "step": 7825
    },
    {
      "entropy": 5.547662878036499,
      "epoch": 0.6092199961097062,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004968232337143769,
      "loss": 5.3037,
      "mean_token_accuracy": 0.18051619529724122,
      "num_tokens": 13587479.0,
      "step": 7830
    },
    {
      "entropy": 5.46766095161438,
      "epoch": 0.6096090254814238,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004968185911328858,
      "loss": 5.1894,
      "mean_token_accuracy": 0.17267035692930222,
      "num_tokens": 13596049.0,
      "step": 7835
    },
    {
      "entropy": 5.596208238601685,
      "epoch": 0.6099980548531414,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004968139451856506,
      "loss": 5.4672,
      "mean_token_accuracy": 0.16732316911220552,
      "num_tokens": 13604766.0,
      "step": 7840
    },
    {
      "entropy": 5.608496761322021,
      "epoch": 0.610387084224859,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004968092958727419,
      "loss": 5.4506,
      "mean_token_accuracy": 0.1715890273451805,
      "num_tokens": 13614051.0,
      "step": 7845
    },
    {
      "entropy": 5.591432857513428,
      "epoch": 0.6107761135965766,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00049680464319423,
      "loss": 5.3735,
      "mean_token_accuracy": 0.17343621253967284,
      "num_tokens": 13623100.0,
      "step": 7850
    },
    {
      "entropy": 5.609377717971801,
      "epoch": 0.6111651429682942,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004967999871501858,
      "loss": 5.3911,
      "mean_token_accuracy": 0.17528704851865767,
      "num_tokens": 13631732.0,
      "step": 7855
    },
    {
      "entropy": 5.5660487651824955,
      "epoch": 0.6115541723400116,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004967953277406798,
      "loss": 5.3686,
      "mean_token_accuracy": 0.1777533009648323,
      "num_tokens": 13640355.0,
      "step": 7860
    },
    {
      "entropy": 5.595680522918701,
      "epoch": 0.6119432017117292,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004967906649657828,
      "loss": 5.4436,
      "mean_token_accuracy": 0.17478672862052919,
      "num_tokens": 13649108.0,
      "step": 7865
    },
    {
      "entropy": 5.560012006759644,
      "epoch": 0.6123322310834468,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004967859988255655,
      "loss": 5.2909,
      "mean_token_accuracy": 0.18835069984197617,
      "num_tokens": 13656765.0,
      "step": 7870
    },
    {
      "entropy": 5.564635944366455,
      "epoch": 0.6127212604551644,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004967813293200985,
      "loss": 5.3627,
      "mean_token_accuracy": 0.17726267576217652,
      "num_tokens": 13664692.0,
      "step": 7875
    },
    {
      "entropy": 5.440789461135864,
      "epoch": 0.613110289826882,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004967766564494529,
      "loss": 5.2333,
      "mean_token_accuracy": 0.18237595409154891,
      "num_tokens": 13672913.0,
      "step": 7880
    },
    {
      "entropy": 5.497929620742798,
      "epoch": 0.6134993191985995,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004967719802136996,
      "loss": 5.2306,
      "mean_token_accuracy": 0.18572678714990615,
      "num_tokens": 13682427.0,
      "step": 7885
    },
    {
      "entropy": 5.557631158828736,
      "epoch": 0.613888348570317,
      "grad_norm": 1.0,
      "learning_rate": 0.0004967673006129094,
      "loss": 5.3286,
      "mean_token_accuracy": 0.1765195220708847,
      "num_tokens": 13690815.0,
      "step": 7890
    },
    {
      "entropy": 5.642125272750855,
      "epoch": 0.6142773779420346,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004967626176471536,
      "loss": 5.4432,
      "mean_token_accuracy": 0.16881301403045654,
      "num_tokens": 13699522.0,
      "step": 7895
    },
    {
      "entropy": 5.659100437164307,
      "epoch": 0.6146664073137522,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004967579313165028,
      "loss": 5.4548,
      "mean_token_accuracy": 0.16711433976888657,
      "num_tokens": 13708075.0,
      "step": 7900
    },
    {
      "entropy": 5.626817226409912,
      "epoch": 0.6150554366854698,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004967532416210284,
      "loss": 5.3373,
      "mean_token_accuracy": 0.1766982078552246,
      "num_tokens": 13716397.0,
      "step": 7905
    },
    {
      "entropy": 5.535777807235718,
      "epoch": 0.6154444660571873,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004967485485608016,
      "loss": 5.3347,
      "mean_token_accuracy": 0.17953308224678038,
      "num_tokens": 13724724.0,
      "step": 7910
    },
    {
      "entropy": 5.516162967681884,
      "epoch": 0.6158334954289049,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004967438521358934,
      "loss": 5.3018,
      "mean_token_accuracy": 0.18072032779455185,
      "num_tokens": 13733389.0,
      "step": 7915
    },
    {
      "entropy": 5.548145055770874,
      "epoch": 0.6162225248006225,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004967391523463754,
      "loss": 5.2714,
      "mean_token_accuracy": 0.1799410566687584,
      "num_tokens": 13742569.0,
      "step": 7920
    },
    {
      "entropy": 5.500254726409912,
      "epoch": 0.61661155417234,
      "grad_norm": 1.0,
      "learning_rate": 0.0004967344491923185,
      "loss": 5.247,
      "mean_token_accuracy": 0.17956051081418992,
      "num_tokens": 13750975.0,
      "step": 7925
    },
    {
      "entropy": 5.5295875549316404,
      "epoch": 0.6170005835440576,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004967297426737943,
      "loss": 5.3039,
      "mean_token_accuracy": 0.18055013567209244,
      "num_tokens": 13758782.0,
      "step": 7930
    },
    {
      "entropy": 5.599636268615723,
      "epoch": 0.6173896129157751,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004967250327908742,
      "loss": 5.4332,
      "mean_token_accuracy": 0.16596893668174745,
      "num_tokens": 13767321.0,
      "step": 7935
    },
    {
      "entropy": 5.62333116531372,
      "epoch": 0.6177786422874927,
      "grad_norm": 1.125,
      "learning_rate": 0.0004967203195436297,
      "loss": 5.4226,
      "mean_token_accuracy": 0.17486030757427215,
      "num_tokens": 13775992.0,
      "step": 7940
    },
    {
      "entropy": 5.552783727645874,
      "epoch": 0.6181676716592103,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004967156029321322,
      "loss": 5.259,
      "mean_token_accuracy": 0.18195210695266723,
      "num_tokens": 13784358.0,
      "step": 7945
    },
    {
      "entropy": 5.472700834274292,
      "epoch": 0.6185567010309279,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004967108829564532,
      "loss": 5.2574,
      "mean_token_accuracy": 0.18075720816850663,
      "num_tokens": 13792834.0,
      "step": 7950
    },
    {
      "entropy": 5.555425357818604,
      "epoch": 0.6189457304026454,
      "grad_norm": 1.0,
      "learning_rate": 0.0004967061596166646,
      "loss": 5.3525,
      "mean_token_accuracy": 0.16884554475545882,
      "num_tokens": 13801414.0,
      "step": 7955
    },
    {
      "entropy": 5.581367826461792,
      "epoch": 0.6193347597743629,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004967014329128378,
      "loss": 5.3831,
      "mean_token_accuracy": 0.1757428005337715,
      "num_tokens": 13810459.0,
      "step": 7960
    },
    {
      "entropy": 5.54815902709961,
      "epoch": 0.6197237891460805,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004966967028450448,
      "loss": 5.2158,
      "mean_token_accuracy": 0.18533872663974763,
      "num_tokens": 13818460.0,
      "step": 7965
    },
    {
      "entropy": 5.526509523391724,
      "epoch": 0.6201128185177981,
      "grad_norm": 1.015625,
      "learning_rate": 0.000496691969413357,
      "loss": 5.3546,
      "mean_token_accuracy": 0.17237673848867416,
      "num_tokens": 13827693.0,
      "step": 7970
    },
    {
      "entropy": 5.559570932388306,
      "epoch": 0.6205018478895157,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004966872326178466,
      "loss": 5.3507,
      "mean_token_accuracy": 0.16821354031562805,
      "num_tokens": 13836506.0,
      "step": 7975
    },
    {
      "entropy": 5.547670364379883,
      "epoch": 0.6208908772612333,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004966824924585851,
      "loss": 5.3778,
      "mean_token_accuracy": 0.17321672886610032,
      "num_tokens": 13844977.0,
      "step": 7980
    },
    {
      "entropy": 5.565658664703369,
      "epoch": 0.6212799066329507,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004966777489356448,
      "loss": 5.3463,
      "mean_token_accuracy": 0.1782146066427231,
      "num_tokens": 13854514.0,
      "step": 7985
    },
    {
      "entropy": 5.56265811920166,
      "epoch": 0.6216689360046683,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004966730020490975,
      "loss": 5.3346,
      "mean_token_accuracy": 0.17269507944583892,
      "num_tokens": 13863037.0,
      "step": 7990
    },
    {
      "entropy": 5.479045581817627,
      "epoch": 0.6220579653763859,
      "grad_norm": 1.0,
      "learning_rate": 0.000496668251799015,
      "loss": 5.322,
      "mean_token_accuracy": 0.1842358112335205,
      "num_tokens": 13871720.0,
      "step": 7995
    },
    {
      "entropy": 5.5524838924407955,
      "epoch": 0.6224469947481035,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004966634981854699,
      "loss": 5.3547,
      "mean_token_accuracy": 0.1798065960407257,
      "num_tokens": 13880214.0,
      "step": 8000
    },
    {
      "entropy": 5.567934513092041,
      "epoch": 0.6228360241198211,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004966587412085339,
      "loss": 5.449,
      "mean_token_accuracy": 0.1677377462387085,
      "num_tokens": 13888784.0,
      "step": 8005
    },
    {
      "entropy": 5.577069473266602,
      "epoch": 0.6232250534915386,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004966539808682793,
      "loss": 5.3299,
      "mean_token_accuracy": 0.17567553371191025,
      "num_tokens": 13897466.0,
      "step": 8010
    },
    {
      "entropy": 5.511372900009155,
      "epoch": 0.6236140828632561,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004966492171647783,
      "loss": 5.3632,
      "mean_token_accuracy": 0.17059484124183655,
      "num_tokens": 13906556.0,
      "step": 8015
    },
    {
      "entropy": 5.516188478469848,
      "epoch": 0.6240031122349737,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004966444500981032,
      "loss": 5.3019,
      "mean_token_accuracy": 0.1749770686030388,
      "num_tokens": 13915473.0,
      "step": 8020
    },
    {
      "entropy": 5.551458358764648,
      "epoch": 0.6243921416066913,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004966396796683265,
      "loss": 5.4636,
      "mean_token_accuracy": 0.16586939841508866,
      "num_tokens": 13924775.0,
      "step": 8025
    },
    {
      "entropy": 5.578491830825806,
      "epoch": 0.6247811709784089,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004966349058755204,
      "loss": 5.3463,
      "mean_token_accuracy": 0.16579750031232834,
      "num_tokens": 13933992.0,
      "step": 8030
    },
    {
      "entropy": 5.67163314819336,
      "epoch": 0.6251702003501264,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004966301287197573,
      "loss": 5.488,
      "mean_token_accuracy": 0.17156226187944412,
      "num_tokens": 13942877.0,
      "step": 8035
    },
    {
      "entropy": 5.563678741455078,
      "epoch": 0.625559229721844,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004966253482011098,
      "loss": 5.3606,
      "mean_token_accuracy": 0.17174550592899324,
      "num_tokens": 13952003.0,
      "step": 8040
    },
    {
      "entropy": 5.622415685653687,
      "epoch": 0.6259482590935616,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004966205643196503,
      "loss": 5.3035,
      "mean_token_accuracy": 0.18188226073980332,
      "num_tokens": 13959801.0,
      "step": 8045
    },
    {
      "entropy": 5.521593475341797,
      "epoch": 0.6263372884652791,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004966157770754516,
      "loss": 5.2971,
      "mean_token_accuracy": 0.17040681689977646,
      "num_tokens": 13968535.0,
      "step": 8050
    },
    {
      "entropy": 5.550281286239624,
      "epoch": 0.6267263178369967,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000496610986468586,
      "loss": 5.2778,
      "mean_token_accuracy": 0.17464786022901535,
      "num_tokens": 13977079.0,
      "step": 8055
    },
    {
      "entropy": 5.520701742172241,
      "epoch": 0.6271153472087143,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004966061924991267,
      "loss": 5.2777,
      "mean_token_accuracy": 0.18244906961917878,
      "num_tokens": 13985118.0,
      "step": 8060
    },
    {
      "entropy": 5.539714193344116,
      "epoch": 0.6275043765804318,
      "grad_norm": 0.96875,
      "learning_rate": 0.000496601395167146,
      "loss": 5.2478,
      "mean_token_accuracy": 0.17663197070360184,
      "num_tokens": 13994061.0,
      "step": 8065
    },
    {
      "entropy": 5.512888669967651,
      "epoch": 0.6278934059521494,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004965965944727168,
      "loss": 5.4147,
      "mean_token_accuracy": 0.16780133545398712,
      "num_tokens": 14002879.0,
      "step": 8070
    },
    {
      "entropy": 5.535863399505615,
      "epoch": 0.628282435323867,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004965917904159121,
      "loss": 5.3582,
      "mean_token_accuracy": 0.1754468560218811,
      "num_tokens": 14011570.0,
      "step": 8075
    },
    {
      "entropy": 5.471121549606323,
      "epoch": 0.6286714646955845,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004965869829968046,
      "loss": 5.2141,
      "mean_token_accuracy": 0.18506875038146972,
      "num_tokens": 14019282.0,
      "step": 8080
    },
    {
      "entropy": 5.475094175338745,
      "epoch": 0.6290604940673021,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004965821722154674,
      "loss": 5.2337,
      "mean_token_accuracy": 0.18745357394218445,
      "num_tokens": 14027774.0,
      "step": 8085
    },
    {
      "entropy": 5.517023706436158,
      "epoch": 0.6294495234390196,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004965773580719734,
      "loss": 5.2169,
      "mean_token_accuracy": 0.18093859851360322,
      "num_tokens": 14036754.0,
      "step": 8090
    },
    {
      "entropy": 5.584210062026978,
      "epoch": 0.6298385528107372,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004965725405663957,
      "loss": 5.3998,
      "mean_token_accuracy": 0.17579556703567506,
      "num_tokens": 14045714.0,
      "step": 8095
    },
    {
      "entropy": 5.612393712997436,
      "epoch": 0.6302275821824548,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004965677196988073,
      "loss": 5.4307,
      "mean_token_accuracy": 0.17204743176698684,
      "num_tokens": 14054973.0,
      "step": 8100
    },
    {
      "entropy": 5.573684597015381,
      "epoch": 0.6306166115541724,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004965628954692815,
      "loss": 5.3433,
      "mean_token_accuracy": 0.17390788197517396,
      "num_tokens": 14064037.0,
      "step": 8105
    },
    {
      "entropy": 5.588153982162476,
      "epoch": 0.63100564092589,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004965580678778913,
      "loss": 5.3624,
      "mean_token_accuracy": 0.1763614758849144,
      "num_tokens": 14072269.0,
      "step": 8110
    },
    {
      "entropy": 5.539870548248291,
      "epoch": 0.6313946702976074,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004965532369247102,
      "loss": 5.3049,
      "mean_token_accuracy": 0.1714685872197151,
      "num_tokens": 14080595.0,
      "step": 8115
    },
    {
      "entropy": 5.541055202484131,
      "epoch": 0.631783699669325,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004965484026098111,
      "loss": 5.2763,
      "mean_token_accuracy": 0.17949116080999375,
      "num_tokens": 14089159.0,
      "step": 8120
    },
    {
      "entropy": 5.6018811702728275,
      "epoch": 0.6321727290410426,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004965435649332679,
      "loss": 5.3888,
      "mean_token_accuracy": 0.17253922373056413,
      "num_tokens": 14098192.0,
      "step": 8125
    },
    {
      "entropy": 5.504631805419922,
      "epoch": 0.6325617584127602,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004965387238951536,
      "loss": 5.2133,
      "mean_token_accuracy": 0.1832298681139946,
      "num_tokens": 14106552.0,
      "step": 8130
    },
    {
      "entropy": 5.507626390457153,
      "epoch": 0.6329507877844778,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004965338794955418,
      "loss": 5.2718,
      "mean_token_accuracy": 0.18028005212545395,
      "num_tokens": 14114783.0,
      "step": 8135
    },
    {
      "entropy": 5.4900048732757565,
      "epoch": 0.6333398171561953,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004965290317345061,
      "loss": 5.3348,
      "mean_token_accuracy": 0.1706690102815628,
      "num_tokens": 14124182.0,
      "step": 8140
    },
    {
      "entropy": 5.6119460105896,
      "epoch": 0.6337288465279128,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004965241806121198,
      "loss": 5.3569,
      "mean_token_accuracy": 0.17082157880067825,
      "num_tokens": 14133037.0,
      "step": 8145
    },
    {
      "entropy": 5.577513885498047,
      "epoch": 0.6341178758996304,
      "grad_norm": 1.0,
      "learning_rate": 0.0004965193261284567,
      "loss": 5.3332,
      "mean_token_accuracy": 0.176353295147419,
      "num_tokens": 14141677.0,
      "step": 8150
    },
    {
      "entropy": 5.5060628890991214,
      "epoch": 0.634506905271348,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004965144682835904,
      "loss": 5.2828,
      "mean_token_accuracy": 0.18138475567102433,
      "num_tokens": 14149980.0,
      "step": 8155
    },
    {
      "entropy": 5.534550285339355,
      "epoch": 0.6348959346430656,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004965096070775946,
      "loss": 5.3267,
      "mean_token_accuracy": 0.18476232290267944,
      "num_tokens": 14158298.0,
      "step": 8160
    },
    {
      "entropy": 5.546457004547119,
      "epoch": 0.6352849640147831,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004965047425105431,
      "loss": 5.2807,
      "mean_token_accuracy": 0.18288637697696686,
      "num_tokens": 14166321.0,
      "step": 8165
    },
    {
      "entropy": 5.439369249343872,
      "epoch": 0.6356739933865007,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004964998745825097,
      "loss": 5.2712,
      "mean_token_accuracy": 0.18472588509321214,
      "num_tokens": 14174619.0,
      "step": 8170
    },
    {
      "entropy": 5.526070976257325,
      "epoch": 0.6360630227582182,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004964950032935682,
      "loss": 5.2799,
      "mean_token_accuracy": 0.17525465339422225,
      "num_tokens": 14183548.0,
      "step": 8175
    },
    {
      "entropy": 5.661414289474488,
      "epoch": 0.6364520521299358,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004964901286437927,
      "loss": 5.468,
      "mean_token_accuracy": 0.17575431913137435,
      "num_tokens": 14192306.0,
      "step": 8180
    },
    {
      "entropy": 5.454582262039184,
      "epoch": 0.6368410815016534,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004964852506332568,
      "loss": 5.1592,
      "mean_token_accuracy": 0.18672306090593338,
      "num_tokens": 14200280.0,
      "step": 8185
    },
    {
      "entropy": 5.51651029586792,
      "epoch": 0.6372301108733709,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000496480369262035,
      "loss": 5.3979,
      "mean_token_accuracy": 0.16993670761585236,
      "num_tokens": 14208484.0,
      "step": 8190
    },
    {
      "entropy": 5.549438095092773,
      "epoch": 0.6376191402450885,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000496475484530201,
      "loss": 5.3809,
      "mean_token_accuracy": 0.17450919449329377,
      "num_tokens": 14217368.0,
      "step": 8195
    },
    {
      "entropy": 5.642722940444946,
      "epoch": 0.6380081696168061,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004964705964378292,
      "loss": 5.4062,
      "mean_token_accuracy": 0.17003117948770524,
      "num_tokens": 14226663.0,
      "step": 8200
    },
    {
      "entropy": 5.542659282684326,
      "epoch": 0.6383971989885237,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004964657049849934,
      "loss": 5.2502,
      "mean_token_accuracy": 0.1807483196258545,
      "num_tokens": 14235153.0,
      "step": 8205
    },
    {
      "entropy": 5.513643455505371,
      "epoch": 0.6387862283602412,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004964608101717681,
      "loss": 5.3918,
      "mean_token_accuracy": 0.17519544661045075,
      "num_tokens": 14244245.0,
      "step": 8210
    },
    {
      "entropy": 5.566900682449341,
      "epoch": 0.6391752577319587,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004964559119982275,
      "loss": 5.3092,
      "mean_token_accuracy": 0.17790648192167283,
      "num_tokens": 14252666.0,
      "step": 8215
    },
    {
      "entropy": 5.589597845077515,
      "epoch": 0.6395642871036763,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000496451010464446,
      "loss": 5.3757,
      "mean_token_accuracy": 0.17137881517410278,
      "num_tokens": 14262232.0,
      "step": 8220
    },
    {
      "entropy": 5.5738012313842775,
      "epoch": 0.6399533164753939,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004964461055704978,
      "loss": 5.318,
      "mean_token_accuracy": 0.1772593691945076,
      "num_tokens": 14270496.0,
      "step": 8225
    },
    {
      "entropy": 5.479415369033814,
      "epoch": 0.6403423458471115,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004964411973164574,
      "loss": 5.2345,
      "mean_token_accuracy": 0.18309761285781861,
      "num_tokens": 14278597.0,
      "step": 8230
    },
    {
      "entropy": 5.4940520286560055,
      "epoch": 0.6407313752188291,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004964362857023993,
      "loss": 5.3129,
      "mean_token_accuracy": 0.1732625499367714,
      "num_tokens": 14287444.0,
      "step": 8235
    },
    {
      "entropy": 5.496807956695557,
      "epoch": 0.6411204045905465,
      "grad_norm": 1.015625,
      "learning_rate": 0.000496431370728398,
      "loss": 5.2298,
      "mean_token_accuracy": 0.1787653997540474,
      "num_tokens": 14296041.0,
      "step": 8240
    },
    {
      "entropy": 5.52669734954834,
      "epoch": 0.6415094339622641,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004964264523945281,
      "loss": 5.2445,
      "mean_token_accuracy": 0.17653031051158904,
      "num_tokens": 14304700.0,
      "step": 8245
    },
    {
      "entropy": 5.5509796142578125,
      "epoch": 0.6418984633339817,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004964215307008643,
      "loss": 5.3706,
      "mean_token_accuracy": 0.17451013773679733,
      "num_tokens": 14312873.0,
      "step": 8250
    },
    {
      "entropy": 5.55800085067749,
      "epoch": 0.6422874927056993,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004964166056474811,
      "loss": 5.3732,
      "mean_token_accuracy": 0.16890601366758345,
      "num_tokens": 14321697.0,
      "step": 8255
    },
    {
      "entropy": 5.52422833442688,
      "epoch": 0.6426765220774169,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004964116772344534,
      "loss": 5.3501,
      "mean_token_accuracy": 0.16809773743152617,
      "num_tokens": 14330074.0,
      "step": 8260
    },
    {
      "entropy": 5.616354608535767,
      "epoch": 0.6430655514491345,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004964067454618559,
      "loss": 5.4086,
      "mean_token_accuracy": 0.16859800815582277,
      "num_tokens": 14339485.0,
      "step": 8265
    },
    {
      "entropy": 5.624785995483398,
      "epoch": 0.6434545808208519,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004964018103297634,
      "loss": 5.3792,
      "mean_token_accuracy": 0.17220971435308458,
      "num_tokens": 14348182.0,
      "step": 8270
    },
    {
      "entropy": 5.6248821258544925,
      "epoch": 0.6438436101925695,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004963968718382509,
      "loss": 5.44,
      "mean_token_accuracy": 0.16919139474630357,
      "num_tokens": 14357202.0,
      "step": 8275
    },
    {
      "entropy": 5.563972473144531,
      "epoch": 0.6442326395642871,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004963919299873931,
      "loss": 5.3607,
      "mean_token_accuracy": 0.16897559016942978,
      "num_tokens": 14366001.0,
      "step": 8280
    },
    {
      "entropy": 5.613673686981201,
      "epoch": 0.6446216689360047,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004963869847772653,
      "loss": 5.3924,
      "mean_token_accuracy": 0.16792071908712386,
      "num_tokens": 14374894.0,
      "step": 8285
    },
    {
      "entropy": 5.520499515533447,
      "epoch": 0.6450106983077223,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004963820362079423,
      "loss": 5.35,
      "mean_token_accuracy": 0.1725335881114006,
      "num_tokens": 14383396.0,
      "step": 8290
    },
    {
      "entropy": 5.535679769515991,
      "epoch": 0.6453997276794398,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004963770842794994,
      "loss": 5.2822,
      "mean_token_accuracy": 0.18232600390911102,
      "num_tokens": 14392514.0,
      "step": 8295
    },
    {
      "entropy": 5.569995975494384,
      "epoch": 0.6457887570511573,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004963721289920115,
      "loss": 5.3131,
      "mean_token_accuracy": 0.17455365359783173,
      "num_tokens": 14401916.0,
      "step": 8300
    },
    {
      "entropy": 5.524501991271973,
      "epoch": 0.6461777864228749,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004963671703455538,
      "loss": 5.3774,
      "mean_token_accuracy": 0.17223564833402633,
      "num_tokens": 14410476.0,
      "step": 8305
    },
    {
      "entropy": 5.546697950363159,
      "epoch": 0.6465668157945925,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004963622083402018,
      "loss": 5.3695,
      "mean_token_accuracy": 0.1732681781053543,
      "num_tokens": 14418437.0,
      "step": 8310
    },
    {
      "entropy": 5.5088029384613035,
      "epoch": 0.6469558451663101,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004963572429760305,
      "loss": 5.3092,
      "mean_token_accuracy": 0.17601361125707626,
      "num_tokens": 14426978.0,
      "step": 8315
    },
    {
      "entropy": 5.589383268356324,
      "epoch": 0.6473448745380276,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004963522742531155,
      "loss": 5.3255,
      "mean_token_accuracy": 0.17566727995872497,
      "num_tokens": 14435043.0,
      "step": 8320
    },
    {
      "entropy": 5.534536409378052,
      "epoch": 0.6477339039097452,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000496347302171532,
      "loss": 5.2453,
      "mean_token_accuracy": 0.1858903259038925,
      "num_tokens": 14443397.0,
      "step": 8325
    },
    {
      "entropy": 5.533512353897095,
      "epoch": 0.6481229332814628,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004963423267313554,
      "loss": 5.3537,
      "mean_token_accuracy": 0.17729001194238664,
      "num_tokens": 14452536.0,
      "step": 8330
    },
    {
      "entropy": 5.563993167877197,
      "epoch": 0.6485119626531803,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004963373479326614,
      "loss": 5.4202,
      "mean_token_accuracy": 0.16432647705078124,
      "num_tokens": 14461901.0,
      "step": 8335
    },
    {
      "entropy": 5.522555923461914,
      "epoch": 0.6489009920248979,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004963323657755255,
      "loss": 5.3253,
      "mean_token_accuracy": 0.1801208570599556,
      "num_tokens": 14470547.0,
      "step": 8340
    },
    {
      "entropy": 5.486880540847778,
      "epoch": 0.6492900213966154,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004963273802600231,
      "loss": 5.2023,
      "mean_token_accuracy": 0.17749185115098953,
      "num_tokens": 14478439.0,
      "step": 8345
    },
    {
      "entropy": 5.585784292221069,
      "epoch": 0.649679050768333,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00049632239138623,
      "loss": 5.2649,
      "mean_token_accuracy": 0.18034340888261796,
      "num_tokens": 14486621.0,
      "step": 8350
    },
    {
      "entropy": 5.5146787643432615,
      "epoch": 0.6500680801400506,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004963173991542219,
      "loss": 5.2593,
      "mean_token_accuracy": 0.17521694600582122,
      "num_tokens": 14495313.0,
      "step": 8355
    },
    {
      "entropy": 5.568447208404541,
      "epoch": 0.6504571095117682,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004963124035640746,
      "loss": 5.3301,
      "mean_token_accuracy": 0.1780635118484497,
      "num_tokens": 14503647.0,
      "step": 8360
    },
    {
      "entropy": 5.555171251296997,
      "epoch": 0.6508461388834857,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004963074046158637,
      "loss": 5.2613,
      "mean_token_accuracy": 0.17885815352201462,
      "num_tokens": 14512350.0,
      "step": 8365
    },
    {
      "entropy": 5.589003276824951,
      "epoch": 0.6512351682552032,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004963024023096652,
      "loss": 5.4138,
      "mean_token_accuracy": 0.1642599418759346,
      "num_tokens": 14521801.0,
      "step": 8370
    },
    {
      "entropy": 5.488497734069824,
      "epoch": 0.6516241976269208,
      "grad_norm": 1.0,
      "learning_rate": 0.0004962973966455551,
      "loss": 5.2293,
      "mean_token_accuracy": 0.1776614621281624,
      "num_tokens": 14530045.0,
      "step": 8375
    },
    {
      "entropy": 5.551271867752075,
      "epoch": 0.6520132269986384,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004962923876236092,
      "loss": 5.3531,
      "mean_token_accuracy": 0.17800607681274414,
      "num_tokens": 14538944.0,
      "step": 8380
    },
    {
      "entropy": 5.645231294631958,
      "epoch": 0.652402256370356,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004962873752439035,
      "loss": 5.3667,
      "mean_token_accuracy": 0.17218226194381714,
      "num_tokens": 14548180.0,
      "step": 8385
    },
    {
      "entropy": 5.598902034759521,
      "epoch": 0.6527912857420736,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004962823595065141,
      "loss": 5.3937,
      "mean_token_accuracy": 0.17164658457040788,
      "num_tokens": 14556907.0,
      "step": 8390
    },
    {
      "entropy": 5.546781921386719,
      "epoch": 0.653180315113791,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004962773404115172,
      "loss": 5.3622,
      "mean_token_accuracy": 0.17036627531051635,
      "num_tokens": 14565317.0,
      "step": 8395
    },
    {
      "entropy": 5.577868413925171,
      "epoch": 0.6535693444855086,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004962723179589886,
      "loss": 5.4085,
      "mean_token_accuracy": 0.1759839326143265,
      "num_tokens": 14574485.0,
      "step": 8400
    },
    {
      "entropy": 5.591546821594238,
      "epoch": 0.6539583738572262,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000496267292149005,
      "loss": 5.3006,
      "mean_token_accuracy": 0.17521705478429794,
      "num_tokens": 14584002.0,
      "step": 8405
    },
    {
      "entropy": 5.548676252365112,
      "epoch": 0.6543474032289438,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004962622629816423,
      "loss": 5.2514,
      "mean_token_accuracy": 0.17681823223829268,
      "num_tokens": 14592033.0,
      "step": 8410
    },
    {
      "entropy": 5.5112104415893555,
      "epoch": 0.6547364326006614,
      "grad_norm": 1.015625,
      "learning_rate": 0.000496257230456977,
      "loss": 5.4078,
      "mean_token_accuracy": 0.16894861608743666,
      "num_tokens": 14600084.0,
      "step": 8415
    },
    {
      "entropy": 5.498833274841308,
      "epoch": 0.6551254619723789,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004962521945750855,
      "loss": 5.3102,
      "mean_token_accuracy": 0.183327354490757,
      "num_tokens": 14608380.0,
      "step": 8420
    },
    {
      "entropy": 5.571792650222778,
      "epoch": 0.6555144913440965,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000496247155336044,
      "loss": 5.376,
      "mean_token_accuracy": 0.17865682095289231,
      "num_tokens": 14616864.0,
      "step": 8425
    },
    {
      "entropy": 5.572298336029053,
      "epoch": 0.655903520715814,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004962421127399291,
      "loss": 5.4099,
      "mean_token_accuracy": 0.1754586726427078,
      "num_tokens": 14627054.0,
      "step": 8430
    },
    {
      "entropy": 5.58860330581665,
      "epoch": 0.6562925500875316,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004962370667868172,
      "loss": 5.3056,
      "mean_token_accuracy": 0.17600467056035995,
      "num_tokens": 14634985.0,
      "step": 8435
    },
    {
      "entropy": 5.6015636920928955,
      "epoch": 0.6566815794592492,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004962320174767851,
      "loss": 5.3684,
      "mean_token_accuracy": 0.1794813171029091,
      "num_tokens": 14644135.0,
      "step": 8440
    },
    {
      "entropy": 5.5272680759429935,
      "epoch": 0.6570706088309667,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000496226964809909,
      "loss": 5.3331,
      "mean_token_accuracy": 0.17841904014348983,
      "num_tokens": 14652997.0,
      "step": 8445
    },
    {
      "entropy": 5.522934532165527,
      "epoch": 0.6574596382026843,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000496221908786266,
      "loss": 5.397,
      "mean_token_accuracy": 0.17456745654344558,
      "num_tokens": 14661961.0,
      "step": 8450
    },
    {
      "entropy": 5.646849918365478,
      "epoch": 0.6578486675744019,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004962168494059327,
      "loss": 5.5142,
      "mean_token_accuracy": 0.16666198074817656,
      "num_tokens": 14671407.0,
      "step": 8455
    },
    {
      "entropy": 5.593811273574829,
      "epoch": 0.6582376969461194,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004962117866689857,
      "loss": 5.3605,
      "mean_token_accuracy": 0.17761484831571578,
      "num_tokens": 14680015.0,
      "step": 8460
    },
    {
      "entropy": 5.461378383636474,
      "epoch": 0.658626726317837,
      "grad_norm": 1.078125,
      "learning_rate": 0.000496206720575502,
      "loss": 5.2089,
      "mean_token_accuracy": 0.17984002977609634,
      "num_tokens": 14688444.0,
      "step": 8465
    },
    {
      "entropy": 5.48771858215332,
      "epoch": 0.6590157556895546,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004962016511255584,
      "loss": 5.3461,
      "mean_token_accuracy": 0.17608871161937714,
      "num_tokens": 14697246.0,
      "step": 8470
    },
    {
      "entropy": 5.512195730209351,
      "epoch": 0.6594047850612721,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004961965783192318,
      "loss": 5.2215,
      "mean_token_accuracy": 0.18361225873231887,
      "num_tokens": 14706066.0,
      "step": 8475
    },
    {
      "entropy": 5.566125392913818,
      "epoch": 0.6597938144329897,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004961915021565992,
      "loss": 5.3449,
      "mean_token_accuracy": 0.17158096730709077,
      "num_tokens": 14714327.0,
      "step": 8480
    },
    {
      "entropy": 5.514739656448365,
      "epoch": 0.6601828438047073,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004961864226377377,
      "loss": 5.2204,
      "mean_token_accuracy": 0.17796500772237778,
      "num_tokens": 14723053.0,
      "step": 8485
    },
    {
      "entropy": 5.477275419235229,
      "epoch": 0.6605718731764249,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004961813397627241,
      "loss": 5.2097,
      "mean_token_accuracy": 0.1825549766421318,
      "num_tokens": 14731825.0,
      "step": 8490
    },
    {
      "entropy": 5.547224283218384,
      "epoch": 0.6609609025481424,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004961762535316358,
      "loss": 5.4071,
      "mean_token_accuracy": 0.16715431362390518,
      "num_tokens": 14740751.0,
      "step": 8495
    },
    {
      "entropy": 5.566222238540649,
      "epoch": 0.6613499319198599,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004961711639445499,
      "loss": 5.2749,
      "mean_token_accuracy": 0.1821089878678322,
      "num_tokens": 14748994.0,
      "step": 8500
    },
    {
      "entropy": 5.548926830291748,
      "epoch": 0.6617389612915775,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004961660710015435,
      "loss": 5.2596,
      "mean_token_accuracy": 0.18302846550941468,
      "num_tokens": 14757388.0,
      "step": 8505
    },
    {
      "entropy": 5.622605180740356,
      "epoch": 0.6621279906632951,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004961609747026943,
      "loss": 5.4292,
      "mean_token_accuracy": 0.16988320499658585,
      "num_tokens": 14766654.0,
      "step": 8510
    },
    {
      "entropy": 5.557560443878174,
      "epoch": 0.6625170200350127,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004961558750480791,
      "loss": 5.3383,
      "mean_token_accuracy": 0.17223811894655228,
      "num_tokens": 14775561.0,
      "step": 8515
    },
    {
      "entropy": 5.547590255737305,
      "epoch": 0.6629060494067303,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004961507720377756,
      "loss": 5.3998,
      "mean_token_accuracy": 0.17088692337274553,
      "num_tokens": 14783596.0,
      "step": 8520
    },
    {
      "entropy": 5.571688461303711,
      "epoch": 0.6632950787784477,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004961456656718611,
      "loss": 5.4016,
      "mean_token_accuracy": 0.1744227558374405,
      "num_tokens": 14791932.0,
      "step": 8525
    },
    {
      "entropy": 5.599920225143433,
      "epoch": 0.6636841081501653,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000496140555950413,
      "loss": 5.4367,
      "mean_token_accuracy": 0.16860510408878326,
      "num_tokens": 14801261.0,
      "step": 8530
    },
    {
      "entropy": 5.551525974273682,
      "epoch": 0.6640731375218829,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004961354428735091,
      "loss": 5.3048,
      "mean_token_accuracy": 0.17448654621839524,
      "num_tokens": 14809788.0,
      "step": 8535
    },
    {
      "entropy": 5.4737877368927,
      "epoch": 0.6644621668936005,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004961303264412267,
      "loss": 5.294,
      "mean_token_accuracy": 0.1741286888718605,
      "num_tokens": 14818324.0,
      "step": 8540
    },
    {
      "entropy": 5.470654439926148,
      "epoch": 0.6648511962653181,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004961252066536437,
      "loss": 5.2568,
      "mean_token_accuracy": 0.17392080575227736,
      "num_tokens": 14827020.0,
      "step": 8545
    },
    {
      "entropy": 5.534530735015869,
      "epoch": 0.6652402256370356,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004961200835108375,
      "loss": 5.3676,
      "mean_token_accuracy": 0.17271433770656586,
      "num_tokens": 14835522.0,
      "step": 8550
    },
    {
      "entropy": 5.565613842010498,
      "epoch": 0.6656292550087531,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000496114957012886,
      "loss": 5.326,
      "mean_token_accuracy": 0.177302485704422,
      "num_tokens": 14844676.0,
      "step": 8555
    },
    {
      "entropy": 5.634394311904908,
      "epoch": 0.6660182843804707,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004961098271598671,
      "loss": 5.3589,
      "mean_token_accuracy": 0.17336404025554658,
      "num_tokens": 14853996.0,
      "step": 8560
    },
    {
      "entropy": 5.571876478195191,
      "epoch": 0.6664073137521883,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004961046939518585,
      "loss": 5.37,
      "mean_token_accuracy": 0.1714632362127304,
      "num_tokens": 14862981.0,
      "step": 8565
    },
    {
      "entropy": 5.57582950592041,
      "epoch": 0.6667963431239059,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004960995573889379,
      "loss": 5.3944,
      "mean_token_accuracy": 0.16898008435964584,
      "num_tokens": 14872378.0,
      "step": 8570
    },
    {
      "entropy": 5.506377506256103,
      "epoch": 0.6671853724956234,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004960944174711836,
      "loss": 5.3356,
      "mean_token_accuracy": 0.169660747051239,
      "num_tokens": 14881662.0,
      "step": 8575
    },
    {
      "entropy": 5.56787075996399,
      "epoch": 0.667574401867341,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004960892741986734,
      "loss": 5.213,
      "mean_token_accuracy": 0.17883460819721222,
      "num_tokens": 14890211.0,
      "step": 8580
    },
    {
      "entropy": 5.621744394302368,
      "epoch": 0.6679634312390585,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004960841275714853,
      "loss": 5.4422,
      "mean_token_accuracy": 0.16384433805942536,
      "num_tokens": 14899865.0,
      "step": 8585
    },
    {
      "entropy": 5.5386497497558596,
      "epoch": 0.6683524606107761,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004960789775896975,
      "loss": 5.3548,
      "mean_token_accuracy": 0.17040929198265076,
      "num_tokens": 14908334.0,
      "step": 8590
    },
    {
      "entropy": 5.490477991104126,
      "epoch": 0.6687414899824937,
      "grad_norm": 1.0,
      "learning_rate": 0.0004960738242533881,
      "loss": 5.2994,
      "mean_token_accuracy": 0.17860868126153945,
      "num_tokens": 14916477.0,
      "step": 8595
    },
    {
      "entropy": 5.509631633758545,
      "epoch": 0.6691305193542112,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004960686675626353,
      "loss": 5.2311,
      "mean_token_accuracy": 0.18317468017339705,
      "num_tokens": 14925223.0,
      "step": 8600
    },
    {
      "entropy": 5.604669713973999,
      "epoch": 0.6695195487259288,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004960635075175173,
      "loss": 5.3892,
      "mean_token_accuracy": 0.17834445387125014,
      "num_tokens": 14934427.0,
      "step": 8605
    },
    {
      "entropy": 5.571853494644165,
      "epoch": 0.6699085780976464,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004960583441181124,
      "loss": 5.3009,
      "mean_token_accuracy": 0.17614462077617646,
      "num_tokens": 14943175.0,
      "step": 8610
    },
    {
      "entropy": 5.576683807373047,
      "epoch": 0.670297607469364,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004960531773644992,
      "loss": 5.337,
      "mean_token_accuracy": 0.17879977971315383,
      "num_tokens": 14951933.0,
      "step": 8615
    },
    {
      "entropy": 5.51773624420166,
      "epoch": 0.6706866368410815,
      "grad_norm": 1.0,
      "learning_rate": 0.0004960480072567557,
      "loss": 5.3151,
      "mean_token_accuracy": 0.17952475547790528,
      "num_tokens": 14960735.0,
      "step": 8620
    },
    {
      "entropy": 5.551405668258667,
      "epoch": 0.671075666212799,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004960428337949604,
      "loss": 5.4102,
      "mean_token_accuracy": 0.16797017753124238,
      "num_tokens": 14970326.0,
      "step": 8625
    },
    {
      "entropy": 5.5701700210571286,
      "epoch": 0.6714646955845166,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004960376569791921,
      "loss": 5.2656,
      "mean_token_accuracy": 0.18205383270978928,
      "num_tokens": 14979822.0,
      "step": 8630
    },
    {
      "entropy": 5.6438360691070555,
      "epoch": 0.6718537249562342,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004960324768095291,
      "loss": 5.3818,
      "mean_token_accuracy": 0.1707278609275818,
      "num_tokens": 14989350.0,
      "step": 8635
    },
    {
      "entropy": 5.569951868057251,
      "epoch": 0.6722427543279518,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004960272932860502,
      "loss": 5.3872,
      "mean_token_accuracy": 0.16884440332651138,
      "num_tokens": 14998763.0,
      "step": 8640
    },
    {
      "entropy": 5.548202753067017,
      "epoch": 0.6726317836996694,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004960221064088338,
      "loss": 5.2572,
      "mean_token_accuracy": 0.17987560778856276,
      "num_tokens": 15006497.0,
      "step": 8645
    },
    {
      "entropy": 5.477806901931762,
      "epoch": 0.6730208130713868,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004960169161779588,
      "loss": 5.228,
      "mean_token_accuracy": 0.18003067076206208,
      "num_tokens": 15014397.0,
      "step": 8650
    },
    {
      "entropy": 5.507954406738281,
      "epoch": 0.6734098424431044,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004960117225935038,
      "loss": 5.256,
      "mean_token_accuracy": 0.17942583560943604,
      "num_tokens": 15022726.0,
      "step": 8655
    },
    {
      "entropy": 5.655488348007202,
      "epoch": 0.673798871814822,
      "grad_norm": 1.0,
      "learning_rate": 0.0004960065256555477,
      "loss": 5.4745,
      "mean_token_accuracy": 0.16527340114116668,
      "num_tokens": 15032317.0,
      "step": 8660
    },
    {
      "entropy": 5.545851612091065,
      "epoch": 0.6741879011865396,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004960013253641695,
      "loss": 5.3231,
      "mean_token_accuracy": 0.17964551001787185,
      "num_tokens": 15041091.0,
      "step": 8665
    },
    {
      "entropy": 5.580912733078003,
      "epoch": 0.6745769305582572,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004959961217194477,
      "loss": 5.4719,
      "mean_token_accuracy": 0.16498138159513473,
      "num_tokens": 15049797.0,
      "step": 8670
    },
    {
      "entropy": 5.499323558807373,
      "epoch": 0.6749659599299748,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004959909147214615,
      "loss": 5.1675,
      "mean_token_accuracy": 0.19018810391426086,
      "num_tokens": 15057736.0,
      "step": 8675
    },
    {
      "entropy": 5.510775804519653,
      "epoch": 0.6753549893016922,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004959857043702902,
      "loss": 5.3369,
      "mean_token_accuracy": 0.1799772009253502,
      "num_tokens": 15066072.0,
      "step": 8680
    },
    {
      "entropy": 5.480511665344238,
      "epoch": 0.6757440186734098,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004959804906660124,
      "loss": 5.2477,
      "mean_token_accuracy": 0.18066450357437133,
      "num_tokens": 15073918.0,
      "step": 8685
    },
    {
      "entropy": 5.568206119537353,
      "epoch": 0.6761330480451274,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004959752736087073,
      "loss": 5.3406,
      "mean_token_accuracy": 0.1788862630724907,
      "num_tokens": 15082937.0,
      "step": 8690
    },
    {
      "entropy": 5.552415657043457,
      "epoch": 0.676522077416845,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004959700531984543,
      "loss": 5.3121,
      "mean_token_accuracy": 0.18420778959989548,
      "num_tokens": 15091218.0,
      "step": 8695
    },
    {
      "entropy": 5.522793579101562,
      "epoch": 0.6769111067885626,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004959648294353324,
      "loss": 5.2271,
      "mean_token_accuracy": 0.18417756855487824,
      "num_tokens": 15099640.0,
      "step": 8700
    },
    {
      "entropy": 5.5514482975006105,
      "epoch": 0.6773001361602801,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004959596023194208,
      "loss": 5.3016,
      "mean_token_accuracy": 0.17648228108882905,
      "num_tokens": 15108710.0,
      "step": 8705
    },
    {
      "entropy": 5.530297803878784,
      "epoch": 0.6776891655319977,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000495954371850799,
      "loss": 5.2661,
      "mean_token_accuracy": 0.17515871822834014,
      "num_tokens": 15117074.0,
      "step": 8710
    },
    {
      "entropy": 5.563857698440552,
      "epoch": 0.6780781949037152,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004959491380295463,
      "loss": 5.2511,
      "mean_token_accuracy": 0.179531230032444,
      "num_tokens": 15124714.0,
      "step": 8715
    },
    {
      "entropy": 5.49883623123169,
      "epoch": 0.6784672242754328,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004959439008557422,
      "loss": 5.2814,
      "mean_token_accuracy": 0.17466178089380263,
      "num_tokens": 15134177.0,
      "step": 8720
    },
    {
      "entropy": 5.565686225891113,
      "epoch": 0.6788562536471504,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000495938660329466,
      "loss": 5.2904,
      "mean_token_accuracy": 0.1761787474155426,
      "num_tokens": 15142659.0,
      "step": 8725
    },
    {
      "entropy": 5.533785057067871,
      "epoch": 0.6792452830188679,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004959334164507973,
      "loss": 5.4468,
      "mean_token_accuracy": 0.17017551362514496,
      "num_tokens": 15151423.0,
      "step": 8730
    },
    {
      "entropy": 5.516269111633301,
      "epoch": 0.6796343123905855,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004959281692198155,
      "loss": 5.2165,
      "mean_token_accuracy": 0.18090467751026154,
      "num_tokens": 15159986.0,
      "step": 8735
    },
    {
      "entropy": 5.577383804321289,
      "epoch": 0.6800233417623031,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004959229186366007,
      "loss": 5.3127,
      "mean_token_accuracy": 0.17540758550167085,
      "num_tokens": 15168183.0,
      "step": 8740
    },
    {
      "entropy": 5.487510919570923,
      "epoch": 0.6804123711340206,
      "grad_norm": 1.03125,
      "learning_rate": 0.000495917664701232,
      "loss": 5.2246,
      "mean_token_accuracy": 0.17727212756872177,
      "num_tokens": 15176476.0,
      "step": 8745
    },
    {
      "entropy": 5.6223084926605225,
      "epoch": 0.6808014005057382,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004959124074137894,
      "loss": 5.5002,
      "mean_token_accuracy": 0.16629669964313507,
      "num_tokens": 15186873.0,
      "step": 8750
    },
    {
      "entropy": 5.584752416610717,
      "epoch": 0.6811904298774557,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004959071467743526,
      "loss": 5.288,
      "mean_token_accuracy": 0.17138179689645766,
      "num_tokens": 15194965.0,
      "step": 8755
    },
    {
      "entropy": 5.506965398788452,
      "epoch": 0.6815794592491733,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004959018827830016,
      "loss": 5.1428,
      "mean_token_accuracy": 0.18453218191862106,
      "num_tokens": 15204028.0,
      "step": 8760
    },
    {
      "entropy": 5.416647291183471,
      "epoch": 0.6819684886208909,
      "grad_norm": 1.0625,
      "learning_rate": 0.000495896615439816,
      "loss": 5.1703,
      "mean_token_accuracy": 0.18949619084596633,
      "num_tokens": 15212011.0,
      "step": 8765
    },
    {
      "entropy": 5.459946537017823,
      "epoch": 0.6823575179926085,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004958913447448759,
      "loss": 5.2655,
      "mean_token_accuracy": 0.1803118497133255,
      "num_tokens": 15220088.0,
      "step": 8770
    },
    {
      "entropy": 5.566954708099365,
      "epoch": 0.682746547364326,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004958860706982613,
      "loss": 5.2162,
      "mean_token_accuracy": 0.1857928141951561,
      "num_tokens": 15228676.0,
      "step": 8775
    },
    {
      "entropy": 5.484989833831787,
      "epoch": 0.6831355767360435,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000495880793300052,
      "loss": 5.2574,
      "mean_token_accuracy": 0.17824719846248627,
      "num_tokens": 15237392.0,
      "step": 8780
    },
    {
      "entropy": 5.490963888168335,
      "epoch": 0.6835246061077611,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004958755125503284,
      "loss": 5.3815,
      "mean_token_accuracy": 0.17574606388807296,
      "num_tokens": 15246738.0,
      "step": 8785
    },
    {
      "entropy": 5.595108985900879,
      "epoch": 0.6839136354794787,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004958702284491703,
      "loss": 5.368,
      "mean_token_accuracy": 0.17641041427850723,
      "num_tokens": 15255154.0,
      "step": 8790
    },
    {
      "entropy": 5.508785390853882,
      "epoch": 0.6843026648511963,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004958649409966581,
      "loss": 5.2593,
      "mean_token_accuracy": 0.1813506916165352,
      "num_tokens": 15264178.0,
      "step": 8795
    },
    {
      "entropy": 5.5767475128173825,
      "epoch": 0.6846916942229139,
      "grad_norm": 1.0625,
      "learning_rate": 0.000495859650192872,
      "loss": 5.3604,
      "mean_token_accuracy": 0.1751423180103302,
      "num_tokens": 15272591.0,
      "step": 8800
    },
    {
      "entropy": 5.571512746810913,
      "epoch": 0.6850807235946313,
      "grad_norm": 1.0,
      "learning_rate": 0.0004958543560378922,
      "loss": 5.2703,
      "mean_token_accuracy": 0.17355967313051224,
      "num_tokens": 15281280.0,
      "step": 8805
    },
    {
      "entropy": 5.527570390701294,
      "epoch": 0.6854697529663489,
      "grad_norm": 1.0,
      "learning_rate": 0.0004958490585317991,
      "loss": 5.2057,
      "mean_token_accuracy": 0.18195094764232636,
      "num_tokens": 15290639.0,
      "step": 8810
    },
    {
      "entropy": 5.48770432472229,
      "epoch": 0.6858587823380665,
      "grad_norm": 1.046875,
      "learning_rate": 0.000495843757674673,
      "loss": 5.2,
      "mean_token_accuracy": 0.18026164025068284,
      "num_tokens": 15298877.0,
      "step": 8815
    },
    {
      "entropy": 5.57325963973999,
      "epoch": 0.6862478117097841,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004958384534665945,
      "loss": 5.3518,
      "mean_token_accuracy": 0.17648355811834335,
      "num_tokens": 15307710.0,
      "step": 8820
    },
    {
      "entropy": 5.571392679214478,
      "epoch": 0.6866368410815017,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004958331459076438,
      "loss": 5.372,
      "mean_token_accuracy": 0.17579764127731323,
      "num_tokens": 15315749.0,
      "step": 8825
    },
    {
      "entropy": 5.610786294937133,
      "epoch": 0.6870258704532192,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004958278349979018,
      "loss": 5.3273,
      "mean_token_accuracy": 0.1696029394865036,
      "num_tokens": 15324161.0,
      "step": 8830
    },
    {
      "entropy": 5.415595006942749,
      "epoch": 0.6874148998249368,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004958225207374488,
      "loss": 5.2138,
      "mean_token_accuracy": 0.17722057551145554,
      "num_tokens": 15333414.0,
      "step": 8835
    },
    {
      "entropy": 5.4682440757751465,
      "epoch": 0.6878039291966543,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004958172031263655,
      "loss": 5.2847,
      "mean_token_accuracy": 0.17830086201429368,
      "num_tokens": 15342369.0,
      "step": 8840
    },
    {
      "entropy": 5.562797260284424,
      "epoch": 0.6881929585683719,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004958118821647326,
      "loss": 5.3634,
      "mean_token_accuracy": 0.17160274982452392,
      "num_tokens": 15351109.0,
      "step": 8845
    },
    {
      "entropy": 5.534557819366455,
      "epoch": 0.6885819879400895,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004958065578526308,
      "loss": 5.2542,
      "mean_token_accuracy": 0.1786276564002037,
      "num_tokens": 15359837.0,
      "step": 8850
    },
    {
      "entropy": 5.475378322601318,
      "epoch": 0.688971017311807,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000495801230190141,
      "loss": 5.2874,
      "mean_token_accuracy": 0.18135666996240615,
      "num_tokens": 15368776.0,
      "step": 8855
    },
    {
      "entropy": 5.496303844451904,
      "epoch": 0.6893600466835246,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004957958991773441,
      "loss": 5.2796,
      "mean_token_accuracy": 0.17713318914175033,
      "num_tokens": 15378127.0,
      "step": 8860
    },
    {
      "entropy": 5.524332714080811,
      "epoch": 0.6897490760552422,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004957905648143206,
      "loss": 5.251,
      "mean_token_accuracy": 0.18107418417930604,
      "num_tokens": 15386580.0,
      "step": 8865
    },
    {
      "entropy": 5.541257715225219,
      "epoch": 0.6901381054269597,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004957852271011519,
      "loss": 5.377,
      "mean_token_accuracy": 0.17538450062274932,
      "num_tokens": 15395973.0,
      "step": 8870
    },
    {
      "entropy": 5.539264726638794,
      "epoch": 0.6905271347986773,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004957798860379187,
      "loss": 5.2783,
      "mean_token_accuracy": 0.1800198435783386,
      "num_tokens": 15404738.0,
      "step": 8875
    },
    {
      "entropy": 5.587938404083252,
      "epoch": 0.6909161641703949,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004957745416247022,
      "loss": 5.4206,
      "mean_token_accuracy": 0.17570649683475495,
      "num_tokens": 15413530.0,
      "step": 8880
    },
    {
      "entropy": 5.570610094070434,
      "epoch": 0.6913051935421124,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004957691938615833,
      "loss": 5.3946,
      "mean_token_accuracy": 0.17589055001735687,
      "num_tokens": 15423268.0,
      "step": 8885
    },
    {
      "entropy": 5.50956540107727,
      "epoch": 0.69169422291383,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004957638427486434,
      "loss": 5.2223,
      "mean_token_accuracy": 0.18491893410682678,
      "num_tokens": 15431690.0,
      "step": 8890
    },
    {
      "entropy": 5.545279932022095,
      "epoch": 0.6920832522855476,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004957584882859636,
      "loss": 5.3765,
      "mean_token_accuracy": 0.17465498000383378,
      "num_tokens": 15440110.0,
      "step": 8895
    },
    {
      "entropy": 5.598361921310425,
      "epoch": 0.6924722816572652,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004957531304736251,
      "loss": 5.389,
      "mean_token_accuracy": 0.17417429983615876,
      "num_tokens": 15449022.0,
      "step": 8900
    },
    {
      "entropy": 5.620433664321899,
      "epoch": 0.6928613110289827,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004957477693117091,
      "loss": 5.4095,
      "mean_token_accuracy": 0.1689018189907074,
      "num_tokens": 15457826.0,
      "step": 8905
    },
    {
      "entropy": 5.526278829574585,
      "epoch": 0.6932503404007002,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004957424048002971,
      "loss": 5.3298,
      "mean_token_accuracy": 0.17603361159563063,
      "num_tokens": 15466631.0,
      "step": 8910
    },
    {
      "entropy": 5.463741159439087,
      "epoch": 0.6936393697724178,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004957370369394706,
      "loss": 5.2291,
      "mean_token_accuracy": 0.17747206389904022,
      "num_tokens": 15474666.0,
      "step": 8915
    },
    {
      "entropy": 5.471134662628174,
      "epoch": 0.6940283991441354,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004957316657293107,
      "loss": 5.2119,
      "mean_token_accuracy": 0.18169043064117432,
      "num_tokens": 15483576.0,
      "step": 8920
    },
    {
      "entropy": 5.684396839141845,
      "epoch": 0.694417428515853,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004957262911698992,
      "loss": 5.3614,
      "mean_token_accuracy": 0.17332629412412642,
      "num_tokens": 15491313.0,
      "step": 8925
    },
    {
      "entropy": 5.561741590499878,
      "epoch": 0.6948064578875706,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004957209132613175,
      "loss": 5.3101,
      "mean_token_accuracy": 0.17961831539869308,
      "num_tokens": 15500638.0,
      "step": 8930
    },
    {
      "entropy": 5.5182092666625975,
      "epoch": 0.695195487259288,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004957155320036473,
      "loss": 5.3074,
      "mean_token_accuracy": 0.18057447224855422,
      "num_tokens": 15509170.0,
      "step": 8935
    },
    {
      "entropy": 5.633941411972046,
      "epoch": 0.6955845166310056,
      "grad_norm": 1.0,
      "learning_rate": 0.0004957101473969702,
      "loss": 5.4,
      "mean_token_accuracy": 0.1688307300209999,
      "num_tokens": 15518108.0,
      "step": 8940
    },
    {
      "entropy": 5.507960414886474,
      "epoch": 0.6959735460027232,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000495704759441368,
      "loss": 5.258,
      "mean_token_accuracy": 0.18316535651683807,
      "num_tokens": 15526558.0,
      "step": 8945
    },
    {
      "entropy": 5.517344236373901,
      "epoch": 0.6963625753744408,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004956993681369221,
      "loss": 5.2259,
      "mean_token_accuracy": 0.18120407164096833,
      "num_tokens": 15535685.0,
      "step": 8950
    },
    {
      "entropy": 5.496462726593018,
      "epoch": 0.6967516047461584,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004956939734837148,
      "loss": 5.2902,
      "mean_token_accuracy": 0.17986082136631013,
      "num_tokens": 15544072.0,
      "step": 8955
    },
    {
      "entropy": 5.5415102481842045,
      "epoch": 0.6971406341178759,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004956885754818277,
      "loss": 5.38,
      "mean_token_accuracy": 0.16789984107017517,
      "num_tokens": 15553303.0,
      "step": 8960
    },
    {
      "entropy": 5.589702939987182,
      "epoch": 0.6975296634895934,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004956831741313427,
      "loss": 5.3154,
      "mean_token_accuracy": 0.1755547121167183,
      "num_tokens": 15561942.0,
      "step": 8965
    },
    {
      "entropy": 5.514275455474854,
      "epoch": 0.697918692861311,
      "grad_norm": 1.0,
      "learning_rate": 0.0004956777694323418,
      "loss": 5.1674,
      "mean_token_accuracy": 0.18837740272283554,
      "num_tokens": 15569988.0,
      "step": 8970
    },
    {
      "entropy": 5.508825063705444,
      "epoch": 0.6983077222330286,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000495672361384907,
      "loss": 5.3641,
      "mean_token_accuracy": 0.17950625866651534,
      "num_tokens": 15579816.0,
      "step": 8975
    },
    {
      "entropy": 5.539467811584473,
      "epoch": 0.6986967516047462,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004956669499891202,
      "loss": 5.2544,
      "mean_token_accuracy": 0.18300477415323257,
      "num_tokens": 15587843.0,
      "step": 8980
    },
    {
      "entropy": 5.510665798187256,
      "epoch": 0.6990857809764637,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004956615352450639,
      "loss": 5.2986,
      "mean_token_accuracy": 0.18106939047574996,
      "num_tokens": 15596798.0,
      "step": 8985
    },
    {
      "entropy": 5.556807661056519,
      "epoch": 0.6994748103481813,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004956561171528198,
      "loss": 5.3904,
      "mean_token_accuracy": 0.1831585004925728,
      "num_tokens": 15605360.0,
      "step": 8990
    },
    {
      "entropy": 5.513580703735352,
      "epoch": 0.6998638397198989,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004956506957124704,
      "loss": 5.2418,
      "mean_token_accuracy": 0.1830413043498993,
      "num_tokens": 15613557.0,
      "step": 8995
    },
    {
      "entropy": 5.475227975845337,
      "epoch": 0.7002528690916164,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000495645270924098,
      "loss": 5.3214,
      "mean_token_accuracy": 0.17621372193098067,
      "num_tokens": 15622144.0,
      "step": 9000
    },
    {
      "epoch": 0.7002528690916164,
      "eval_entropy": 5.314674048933189,
      "eval_loss": 5.330162525177002,
      "eval_mean_token_accuracy": 0.1845590501665151,
      "eval_num_tokens": 15622144.0,
      "eval_runtime": 21.7032,
      "eval_samples_per_second": 1914.831,
      "eval_steps_per_second": 239.365,
      "step": 9000
    },
    {
      "entropy": 5.480877304077149,
      "epoch": 0.700641898463334,
      "grad_norm": 1.0,
      "learning_rate": 0.0004956398427877847,
      "loss": 5.2613,
      "mean_token_accuracy": 0.18023111075162887,
      "num_tokens": 15631119.0,
      "step": 9005
    },
    {
      "entropy": 5.46811809539795,
      "epoch": 0.7010309278350515,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000495634411303613,
      "loss": 5.1783,
      "mean_token_accuracy": 0.18572062104940415,
      "num_tokens": 15639969.0,
      "step": 9010
    },
    {
      "entropy": 5.508111906051636,
      "epoch": 0.7014199572067691,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004956289764716654,
      "loss": 5.3223,
      "mean_token_accuracy": 0.180294868350029,
      "num_tokens": 15648522.0,
      "step": 9015
    },
    {
      "entropy": 5.495264863967895,
      "epoch": 0.7018089865784867,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004956235382920241,
      "loss": 5.2524,
      "mean_token_accuracy": 0.1724682107567787,
      "num_tokens": 15656834.0,
      "step": 9020
    },
    {
      "entropy": 5.60712833404541,
      "epoch": 0.7021980159502043,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004956180967647717,
      "loss": 5.4342,
      "mean_token_accuracy": 0.16941430568695068,
      "num_tokens": 15665066.0,
      "step": 9025
    },
    {
      "entropy": 5.5672718524932865,
      "epoch": 0.7025870453219218,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004956126518899911,
      "loss": 5.3023,
      "mean_token_accuracy": 0.17963931560516358,
      "num_tokens": 15673733.0,
      "step": 9030
    },
    {
      "entropy": 5.642737150192261,
      "epoch": 0.7029760746936393,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004956072036677644,
      "loss": 5.3427,
      "mean_token_accuracy": 0.17978107333183288,
      "num_tokens": 15682212.0,
      "step": 9035
    },
    {
      "entropy": 5.5662132740020756,
      "epoch": 0.7033651040653569,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004956017520981746,
      "loss": 5.3495,
      "mean_token_accuracy": 0.16661105155944825,
      "num_tokens": 15690518.0,
      "step": 9040
    },
    {
      "entropy": 5.602406930923462,
      "epoch": 0.7037541334370745,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004955962971813044,
      "loss": 5.375,
      "mean_token_accuracy": 0.16700301468372344,
      "num_tokens": 15699474.0,
      "step": 9045
    },
    {
      "entropy": 5.621792936325074,
      "epoch": 0.7041431628087921,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004955908389172364,
      "loss": 5.3996,
      "mean_token_accuracy": 0.17017440497875214,
      "num_tokens": 15708571.0,
      "step": 9050
    },
    {
      "entropy": 5.640527248382568,
      "epoch": 0.7045321921805097,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004955853773060536,
      "loss": 5.3797,
      "mean_token_accuracy": 0.1751769095659256,
      "num_tokens": 15717307.0,
      "step": 9055
    },
    {
      "entropy": 5.5418007373809814,
      "epoch": 0.7049212215522271,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004955799123478388,
      "loss": 5.1794,
      "mean_token_accuracy": 0.18809955418109894,
      "num_tokens": 15725502.0,
      "step": 9060
    },
    {
      "entropy": 5.504517269134522,
      "epoch": 0.7053102509239447,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004955744440426748,
      "loss": 5.3856,
      "mean_token_accuracy": 0.17095099687576293,
      "num_tokens": 15734805.0,
      "step": 9065
    },
    {
      "entropy": 5.5177594184875485,
      "epoch": 0.7056992802956623,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004955689723906448,
      "loss": 5.2336,
      "mean_token_accuracy": 0.17607430070638658,
      "num_tokens": 15743313.0,
      "step": 9070
    },
    {
      "entropy": 5.489732646942139,
      "epoch": 0.7060883096673799,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004955634973918318,
      "loss": 5.1957,
      "mean_token_accuracy": 0.18229084461927414,
      "num_tokens": 15752552.0,
      "step": 9075
    },
    {
      "entropy": 5.557054805755615,
      "epoch": 0.7064773390390975,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004955580190463186,
      "loss": 5.2963,
      "mean_token_accuracy": 0.18903578966856002,
      "num_tokens": 15761404.0,
      "step": 9080
    },
    {
      "entropy": 5.565775060653687,
      "epoch": 0.7068663684108151,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004955525373541886,
      "loss": 5.3588,
      "mean_token_accuracy": 0.17453867495059966,
      "num_tokens": 15770605.0,
      "step": 9085
    },
    {
      "entropy": 5.501002073287964,
      "epoch": 0.7072553977825325,
      "grad_norm": 1.125,
      "learning_rate": 0.0004955470523155249,
      "loss": 5.2385,
      "mean_token_accuracy": 0.18286987543106079,
      "num_tokens": 15778047.0,
      "step": 9090
    },
    {
      "entropy": 5.482538461685181,
      "epoch": 0.7076444271542501,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004955415639304107,
      "loss": 5.3183,
      "mean_token_accuracy": 0.1798519253730774,
      "num_tokens": 15788304.0,
      "step": 9095
    },
    {
      "entropy": 5.56830062866211,
      "epoch": 0.7080334565259677,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004955360721989292,
      "loss": 5.2989,
      "mean_token_accuracy": 0.1756359949707985,
      "num_tokens": 15796448.0,
      "step": 9100
    },
    {
      "entropy": 5.515803956985474,
      "epoch": 0.7084224858976853,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004955305771211641,
      "loss": 5.2718,
      "mean_token_accuracy": 0.17800428569316865,
      "num_tokens": 15805072.0,
      "step": 9105
    },
    {
      "entropy": 5.527458620071411,
      "epoch": 0.7088115152694029,
      "grad_norm": 1.0,
      "learning_rate": 0.0004955250786971982,
      "loss": 5.3193,
      "mean_token_accuracy": 0.177976393699646,
      "num_tokens": 15813918.0,
      "step": 9110
    },
    {
      "entropy": 5.642554998397827,
      "epoch": 0.7092005446411204,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004955195769271154,
      "loss": 5.4498,
      "mean_token_accuracy": 0.16878994554281235,
      "num_tokens": 15822675.0,
      "step": 9115
    },
    {
      "entropy": 5.555654811859131,
      "epoch": 0.709589574012838,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004955140718109991,
      "loss": 5.3172,
      "mean_token_accuracy": 0.17535509318113326,
      "num_tokens": 15831056.0,
      "step": 9120
    },
    {
      "entropy": 5.54572868347168,
      "epoch": 0.7099786033845555,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004955085633489326,
      "loss": 5.3851,
      "mean_token_accuracy": 0.1702793061733246,
      "num_tokens": 15839948.0,
      "step": 9125
    },
    {
      "entropy": 5.566775369644165,
      "epoch": 0.7103676327562731,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004955030515409997,
      "loss": 5.2903,
      "mean_token_accuracy": 0.17616434544324874,
      "num_tokens": 15848738.0,
      "step": 9130
    },
    {
      "entropy": 5.545501852035523,
      "epoch": 0.7107566621279907,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004954975363872838,
      "loss": 5.3509,
      "mean_token_accuracy": 0.17266394942998886,
      "num_tokens": 15859073.0,
      "step": 9135
    },
    {
      "entropy": 5.529724979400635,
      "epoch": 0.7111456914997082,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004954920178878689,
      "loss": 5.2481,
      "mean_token_accuracy": 0.18114888072013854,
      "num_tokens": 15868146.0,
      "step": 9140
    },
    {
      "entropy": 5.471456861495971,
      "epoch": 0.7115347208714258,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004954864960428385,
      "loss": 5.1966,
      "mean_token_accuracy": 0.17927344739437104,
      "num_tokens": 15876915.0,
      "step": 9145
    },
    {
      "entropy": 5.514035701751709,
      "epoch": 0.7119237502431434,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004954809708522765,
      "loss": 5.2848,
      "mean_token_accuracy": 0.17866091281175614,
      "num_tokens": 15884947.0,
      "step": 9150
    },
    {
      "entropy": 5.55972375869751,
      "epoch": 0.712312779614861,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004954754423162667,
      "loss": 5.2826,
      "mean_token_accuracy": 0.1738219976425171,
      "num_tokens": 15894595.0,
      "step": 9155
    },
    {
      "entropy": 5.549484443664551,
      "epoch": 0.7127018089865785,
      "grad_norm": 1.0,
      "learning_rate": 0.000495469910434893,
      "loss": 5.2744,
      "mean_token_accuracy": 0.18147733360528945,
      "num_tokens": 15902920.0,
      "step": 9160
    },
    {
      "entropy": 5.470484256744385,
      "epoch": 0.713090838358296,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004954643752082392,
      "loss": 5.2685,
      "mean_token_accuracy": 0.18070472925901412,
      "num_tokens": 15911414.0,
      "step": 9165
    },
    {
      "entropy": 5.575806760787964,
      "epoch": 0.7134798677300136,
      "grad_norm": 1.125,
      "learning_rate": 0.0004954588366363896,
      "loss": 5.3622,
      "mean_token_accuracy": 0.16838099658489228,
      "num_tokens": 15919855.0,
      "step": 9170
    },
    {
      "entropy": 5.537714338302612,
      "epoch": 0.7138688971017312,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004954532947194279,
      "loss": 5.2818,
      "mean_token_accuracy": 0.17650104314088821,
      "num_tokens": 15928932.0,
      "step": 9175
    },
    {
      "entropy": 5.5572038173675535,
      "epoch": 0.7142579264734488,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004954477494574384,
      "loss": 5.3467,
      "mean_token_accuracy": 0.17612511962652205,
      "num_tokens": 15938266.0,
      "step": 9180
    },
    {
      "entropy": 5.661629247665405,
      "epoch": 0.7146469558451664,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004954422008505052,
      "loss": 5.3616,
      "mean_token_accuracy": 0.17050657123327256,
      "num_tokens": 15947100.0,
      "step": 9185
    },
    {
      "entropy": 5.556237745285034,
      "epoch": 0.7150359852168838,
      "grad_norm": 1.125,
      "learning_rate": 0.0004954366488987125,
      "loss": 5.3531,
      "mean_token_accuracy": 0.171370729804039,
      "num_tokens": 15955544.0,
      "step": 9190
    },
    {
      "entropy": 5.559494495391846,
      "epoch": 0.7154250145886014,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004954310936021444,
      "loss": 5.3772,
      "mean_token_accuracy": 0.17268841564655305,
      "num_tokens": 15963372.0,
      "step": 9195
    },
    {
      "entropy": 5.581085157394409,
      "epoch": 0.715814043960319,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004954255349608853,
      "loss": 5.3491,
      "mean_token_accuracy": 0.17209953665733338,
      "num_tokens": 15972940.0,
      "step": 9200
    },
    {
      "entropy": 5.600600862503052,
      "epoch": 0.7162030733320366,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004954199729750198,
      "loss": 5.3116,
      "mean_token_accuracy": 0.17283549010753632,
      "num_tokens": 15982238.0,
      "step": 9205
    },
    {
      "entropy": 5.575990676879883,
      "epoch": 0.7165921027037542,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004954144076446318,
      "loss": 5.3415,
      "mean_token_accuracy": 0.1794824108481407,
      "num_tokens": 15990835.0,
      "step": 9210
    },
    {
      "entropy": 5.5823462963104244,
      "epoch": 0.7169811320754716,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004954088389698061,
      "loss": 5.4313,
      "mean_token_accuracy": 0.17295270413160324,
      "num_tokens": 16000649.0,
      "step": 9215
    },
    {
      "entropy": 5.5323731899261475,
      "epoch": 0.7173701614471892,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000495403266950627,
      "loss": 5.3321,
      "mean_token_accuracy": 0.17856037318706514,
      "num_tokens": 16009214.0,
      "step": 9220
    },
    {
      "entropy": 5.591493225097656,
      "epoch": 0.7177591908189068,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004953976915871792,
      "loss": 5.319,
      "mean_token_accuracy": 0.1777602568268776,
      "num_tokens": 16018173.0,
      "step": 9225
    },
    {
      "entropy": 5.486226892471313,
      "epoch": 0.7181482201906244,
      "grad_norm": 1.0,
      "learning_rate": 0.0004953921128795472,
      "loss": 5.2035,
      "mean_token_accuracy": 0.1935838297009468,
      "num_tokens": 16026857.0,
      "step": 9230
    },
    {
      "entropy": 5.509092330932617,
      "epoch": 0.718537249562342,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004953865308278156,
      "loss": 5.3806,
      "mean_token_accuracy": 0.17093168646097184,
      "num_tokens": 16035792.0,
      "step": 9235
    },
    {
      "entropy": 5.5780352592468265,
      "epoch": 0.7189262789340595,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004953809454320693,
      "loss": 5.3054,
      "mean_token_accuracy": 0.18395649492740632,
      "num_tokens": 16044874.0,
      "step": 9240
    },
    {
      "entropy": 5.58904390335083,
      "epoch": 0.7193153083057771,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004953753566923929,
      "loss": 5.2817,
      "mean_token_accuracy": 0.17687446027994155,
      "num_tokens": 16053705.0,
      "step": 9245
    },
    {
      "entropy": 5.504145622253418,
      "epoch": 0.7197043376774946,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004953697646088712,
      "loss": 5.2897,
      "mean_token_accuracy": 0.1790979638695717,
      "num_tokens": 16062987.0,
      "step": 9250
    },
    {
      "entropy": 5.498252058029175,
      "epoch": 0.7200933670492122,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004953641691815891,
      "loss": 5.2638,
      "mean_token_accuracy": 0.18136994540691376,
      "num_tokens": 16071247.0,
      "step": 9255
    },
    {
      "entropy": 5.479151868820191,
      "epoch": 0.7204823964209298,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004953585704106315,
      "loss": 5.1936,
      "mean_token_accuracy": 0.18132317215204238,
      "num_tokens": 16080190.0,
      "step": 9260
    },
    {
      "entropy": 5.542968225479126,
      "epoch": 0.7208714257926473,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004953529682960832,
      "loss": 5.2863,
      "mean_token_accuracy": 0.1796716883778572,
      "num_tokens": 16088885.0,
      "step": 9265
    },
    {
      "entropy": 5.498762035369873,
      "epoch": 0.7212604551643649,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004953473628380295,
      "loss": 5.2576,
      "mean_token_accuracy": 0.17897747009992598,
      "num_tokens": 16097202.0,
      "step": 9270
    },
    {
      "entropy": 5.526819849014283,
      "epoch": 0.7216494845360825,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004953417540365553,
      "loss": 5.2134,
      "mean_token_accuracy": 0.1786024898290634,
      "num_tokens": 16105468.0,
      "step": 9275
    },
    {
      "entropy": 5.59045443534851,
      "epoch": 0.7220385139078,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004953361418917455,
      "loss": 5.3669,
      "mean_token_accuracy": 0.1773991122841835,
      "num_tokens": 16114624.0,
      "step": 9280
    },
    {
      "entropy": 5.468513679504395,
      "epoch": 0.7224275432795176,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004953305264036856,
      "loss": 5.2078,
      "mean_token_accuracy": 0.17511059045791627,
      "num_tokens": 16123983.0,
      "step": 9285
    },
    {
      "entropy": 5.58538408279419,
      "epoch": 0.7228165726512352,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004953249075724607,
      "loss": 5.3165,
      "mean_token_accuracy": 0.17857147455215455,
      "num_tokens": 16132886.0,
      "step": 9290
    },
    {
      "entropy": 5.483609437942505,
      "epoch": 0.7232056020229527,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004953192853981559,
      "loss": 5.1569,
      "mean_token_accuracy": 0.18487202376127243,
      "num_tokens": 16141526.0,
      "step": 9295
    },
    {
      "entropy": 5.534828615188599,
      "epoch": 0.7235946313946703,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004953136598808566,
      "loss": 5.3493,
      "mean_token_accuracy": 0.18662586361169814,
      "num_tokens": 16149949.0,
      "step": 9300
    },
    {
      "entropy": 5.5371300220489506,
      "epoch": 0.7239836607663879,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004953080310206481,
      "loss": 5.3288,
      "mean_token_accuracy": 0.17804008573293686,
      "num_tokens": 16158131.0,
      "step": 9305
    },
    {
      "entropy": 5.4698076248168945,
      "epoch": 0.7243726901381055,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004953023988176162,
      "loss": 5.2284,
      "mean_token_accuracy": 0.18422405272722245,
      "num_tokens": 16167131.0,
      "step": 9310
    },
    {
      "entropy": 5.596793699264526,
      "epoch": 0.724761719509823,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004952967632718458,
      "loss": 5.436,
      "mean_token_accuracy": 0.1720057561993599,
      "num_tokens": 16176198.0,
      "step": 9315
    },
    {
      "entropy": 5.52623233795166,
      "epoch": 0.7251507488815405,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004952911243834227,
      "loss": 5.314,
      "mean_token_accuracy": 0.1763841137290001,
      "num_tokens": 16184860.0,
      "step": 9320
    },
    {
      "entropy": 5.547797393798828,
      "epoch": 0.7255397782532581,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004952854821524324,
      "loss": 5.3055,
      "mean_token_accuracy": 0.1784798800945282,
      "num_tokens": 16193126.0,
      "step": 9325
    },
    {
      "entropy": 5.509217882156372,
      "epoch": 0.7259288076249757,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004952798365789606,
      "loss": 5.2711,
      "mean_token_accuracy": 0.17647185027599335,
      "num_tokens": 16201763.0,
      "step": 9330
    },
    {
      "entropy": 5.514590501785278,
      "epoch": 0.7263178369966933,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004952741876630928,
      "loss": 5.2264,
      "mean_token_accuracy": 0.17956349849700928,
      "num_tokens": 16210947.0,
      "step": 9335
    },
    {
      "entropy": 5.590847110748291,
      "epoch": 0.7267068663684109,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004952685354049148,
      "loss": 5.3133,
      "mean_token_accuracy": 0.1742142543196678,
      "num_tokens": 16219772.0,
      "step": 9340
    },
    {
      "entropy": 5.576527833938599,
      "epoch": 0.7270958957401283,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004952628798045124,
      "loss": 5.2979,
      "mean_token_accuracy": 0.18574303686618804,
      "num_tokens": 16227759.0,
      "step": 9345
    },
    {
      "entropy": 5.485273456573486,
      "epoch": 0.7274849251118459,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004952572208619714,
      "loss": 5.1884,
      "mean_token_accuracy": 0.18349898010492324,
      "num_tokens": 16236913.0,
      "step": 9350
    },
    {
      "entropy": 5.5475630283355715,
      "epoch": 0.7278739544835635,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004952515585773778,
      "loss": 5.3199,
      "mean_token_accuracy": 0.17370689213275908,
      "num_tokens": 16246190.0,
      "step": 9355
    },
    {
      "entropy": 5.541135311126709,
      "epoch": 0.7282629838552811,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004952458929508171,
      "loss": 5.2695,
      "mean_token_accuracy": 0.1770156666636467,
      "num_tokens": 16255476.0,
      "step": 9360
    },
    {
      "entropy": 5.539259386062622,
      "epoch": 0.7286520132269987,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004952402239823757,
      "loss": 5.3044,
      "mean_token_accuracy": 0.17563280016183852,
      "num_tokens": 16264701.0,
      "step": 9365
    },
    {
      "entropy": 5.367294836044311,
      "epoch": 0.7290410425987162,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004952345516721393,
      "loss": 5.1397,
      "mean_token_accuracy": 0.18621472418308258,
      "num_tokens": 16272999.0,
      "step": 9370
    },
    {
      "entropy": 5.599175119400025,
      "epoch": 0.7294300719704337,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004952288760201942,
      "loss": 5.4244,
      "mean_token_accuracy": 0.1757317990064621,
      "num_tokens": 16282537.0,
      "step": 9375
    },
    {
      "entropy": 5.578078174591065,
      "epoch": 0.7298191013421513,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004952231970266265,
      "loss": 5.2725,
      "mean_token_accuracy": 0.18148419708013536,
      "num_tokens": 16291155.0,
      "step": 9380
    },
    {
      "entropy": 5.60799469947815,
      "epoch": 0.7302081307138689,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004952175146915223,
      "loss": 5.3461,
      "mean_token_accuracy": 0.17183467298746108,
      "num_tokens": 16300466.0,
      "step": 9385
    },
    {
      "entropy": 5.532757520675659,
      "epoch": 0.7305971600855865,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004952118290149679,
      "loss": 5.25,
      "mean_token_accuracy": 0.17785014659166337,
      "num_tokens": 16309250.0,
      "step": 9390
    },
    {
      "entropy": 5.578254699707031,
      "epoch": 0.730986189457304,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004952061399970492,
      "loss": 5.36,
      "mean_token_accuracy": 0.17332724630832672,
      "num_tokens": 16318056.0,
      "step": 9395
    },
    {
      "entropy": 5.488435220718384,
      "epoch": 0.7313752188290216,
      "grad_norm": 1.078125,
      "learning_rate": 0.000495200447637853,
      "loss": 5.2165,
      "mean_token_accuracy": 0.17693322747945786,
      "num_tokens": 16326455.0,
      "step": 9400
    },
    {
      "entropy": 5.487587785720825,
      "epoch": 0.7317642482007392,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004951947519374655,
      "loss": 5.1977,
      "mean_token_accuracy": 0.17901090234518052,
      "num_tokens": 16334584.0,
      "step": 9405
    },
    {
      "entropy": 5.478155517578125,
      "epoch": 0.7321532775724567,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004951890528959731,
      "loss": 5.2233,
      "mean_token_accuracy": 0.18859104961156845,
      "num_tokens": 16342361.0,
      "step": 9410
    },
    {
      "entropy": 5.57849702835083,
      "epoch": 0.7325423069441743,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004951833505134623,
      "loss": 5.3699,
      "mean_token_accuracy": 0.17101291418075562,
      "num_tokens": 16350414.0,
      "step": 9415
    },
    {
      "entropy": 5.4418848037719725,
      "epoch": 0.7329313363158918,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004951776447900196,
      "loss": 5.2144,
      "mean_token_accuracy": 0.1863088607788086,
      "num_tokens": 16359266.0,
      "step": 9420
    },
    {
      "entropy": 5.441287088394165,
      "epoch": 0.7333203656876094,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004951719357257317,
      "loss": 5.181,
      "mean_token_accuracy": 0.18190018385648726,
      "num_tokens": 16367558.0,
      "step": 9425
    },
    {
      "entropy": 5.408336400985718,
      "epoch": 0.733709395059327,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000495166223320685,
      "loss": 5.1723,
      "mean_token_accuracy": 0.1882142335176468,
      "num_tokens": 16375690.0,
      "step": 9430
    },
    {
      "entropy": 5.554342651367188,
      "epoch": 0.7340984244310446,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004951605075749662,
      "loss": 5.2841,
      "mean_token_accuracy": 0.17990492284297943,
      "num_tokens": 16384157.0,
      "step": 9435
    },
    {
      "entropy": 5.539931726455689,
      "epoch": 0.7344874538027621,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004951547884886623,
      "loss": 5.3478,
      "mean_token_accuracy": 0.18536732345819473,
      "num_tokens": 16392153.0,
      "step": 9440
    },
    {
      "entropy": 5.410438871383667,
      "epoch": 0.7348764831744796,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004951490660618598,
      "loss": 5.0892,
      "mean_token_accuracy": 0.19295663386583328,
      "num_tokens": 16400132.0,
      "step": 9445
    },
    {
      "entropy": 5.5202525615692135,
      "epoch": 0.7352655125461972,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004951433402946457,
      "loss": 5.2611,
      "mean_token_accuracy": 0.18991921842098236,
      "num_tokens": 16408297.0,
      "step": 9450
    },
    {
      "entropy": 5.523857450485229,
      "epoch": 0.7356545419179148,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004951376111871068,
      "loss": 5.3218,
      "mean_token_accuracy": 0.17884534895420073,
      "num_tokens": 16417977.0,
      "step": 9455
    },
    {
      "entropy": 5.52310266494751,
      "epoch": 0.7360435712896324,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004951318787393299,
      "loss": 5.2181,
      "mean_token_accuracy": 0.1823731929063797,
      "num_tokens": 16427156.0,
      "step": 9460
    },
    {
      "entropy": 5.592959260940551,
      "epoch": 0.73643260066135,
      "grad_norm": 1.0,
      "learning_rate": 0.0004951261429514023,
      "loss": 5.4103,
      "mean_token_accuracy": 0.16966954618692398,
      "num_tokens": 16435845.0,
      "step": 9465
    },
    {
      "entropy": 5.577749538421631,
      "epoch": 0.7368216300330674,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004951204038234106,
      "loss": 5.3666,
      "mean_token_accuracy": 0.17865575850009918,
      "num_tokens": 16445876.0,
      "step": 9470
    },
    {
      "entropy": 5.55617036819458,
      "epoch": 0.737210659404785,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004951146613554424,
      "loss": 5.314,
      "mean_token_accuracy": 0.18426597714424134,
      "num_tokens": 16454688.0,
      "step": 9475
    },
    {
      "entropy": 5.575452566146851,
      "epoch": 0.7375996887765026,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004951089155475843,
      "loss": 5.4023,
      "mean_token_accuracy": 0.17011319547891618,
      "num_tokens": 16463932.0,
      "step": 9480
    },
    {
      "entropy": 5.550261783599853,
      "epoch": 0.7379887181482202,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000495103166399924,
      "loss": 5.3056,
      "mean_token_accuracy": 0.17290811985731125,
      "num_tokens": 16472627.0,
      "step": 9485
    },
    {
      "entropy": 5.6118803977966305,
      "epoch": 0.7383777475199378,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004950974139125484,
      "loss": 5.3445,
      "mean_token_accuracy": 0.17864836603403092,
      "num_tokens": 16480819.0,
      "step": 9490
    },
    {
      "entropy": 5.56328854560852,
      "epoch": 0.7387667768916554,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004950916580855448,
      "loss": 5.3079,
      "mean_token_accuracy": 0.176665960252285,
      "num_tokens": 16489055.0,
      "step": 9495
    },
    {
      "entropy": 5.503185844421386,
      "epoch": 0.7391558062633728,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004950858989190007,
      "loss": 5.3133,
      "mean_token_accuracy": 0.1837766781449318,
      "num_tokens": 16497375.0,
      "step": 9500
    },
    {
      "entropy": 5.437498569488525,
      "epoch": 0.7395448356350904,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004950801364130032,
      "loss": 5.1383,
      "mean_token_accuracy": 0.1859965667128563,
      "num_tokens": 16505183.0,
      "step": 9505
    },
    {
      "entropy": 5.552580451965332,
      "epoch": 0.739933865006808,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004950743705676401,
      "loss": 5.4181,
      "mean_token_accuracy": 0.16870557591319085,
      "num_tokens": 16514018.0,
      "step": 9510
    },
    {
      "entropy": 5.597504186630249,
      "epoch": 0.7403228943785256,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004950686013829987,
      "loss": 5.291,
      "mean_token_accuracy": 0.18029272109270095,
      "num_tokens": 16522954.0,
      "step": 9515
    },
    {
      "entropy": 5.652072906494141,
      "epoch": 0.7407119237502432,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004950628288591665,
      "loss": 5.3297,
      "mean_token_accuracy": 0.17712741941213608,
      "num_tokens": 16530845.0,
      "step": 9520
    },
    {
      "entropy": 5.507139778137207,
      "epoch": 0.7411009531219607,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004950570529962311,
      "loss": 5.3137,
      "mean_token_accuracy": 0.17810894697904586,
      "num_tokens": 16540464.0,
      "step": 9525
    },
    {
      "entropy": 5.415792036056518,
      "epoch": 0.7414899824936783,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004950512737942803,
      "loss": 5.1962,
      "mean_token_accuracy": 0.1890740156173706,
      "num_tokens": 16549485.0,
      "step": 9530
    },
    {
      "entropy": 5.599626255035401,
      "epoch": 0.7418790118653958,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004950454912534015,
      "loss": 5.3536,
      "mean_token_accuracy": 0.17996269911527635,
      "num_tokens": 16557854.0,
      "step": 9535
    },
    {
      "entropy": 5.588563013076782,
      "epoch": 0.7422680412371134,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004950397053736827,
      "loss": 5.3825,
      "mean_token_accuracy": 0.17422034591436386,
      "num_tokens": 16565975.0,
      "step": 9540
    },
    {
      "entropy": 5.558382797241211,
      "epoch": 0.742657070608831,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004950339161552117,
      "loss": 5.332,
      "mean_token_accuracy": 0.17990780919790267,
      "num_tokens": 16575038.0,
      "step": 9545
    },
    {
      "entropy": 5.514080667495728,
      "epoch": 0.7430460999805485,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004950281235980761,
      "loss": 5.2034,
      "mean_token_accuracy": 0.18488625884056092,
      "num_tokens": 16584292.0,
      "step": 9550
    },
    {
      "entropy": 5.532594013214111,
      "epoch": 0.7434351293522661,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000495022327702364,
      "loss": 5.3043,
      "mean_token_accuracy": 0.17876578569412233,
      "num_tokens": 16592331.0,
      "step": 9555
    },
    {
      "entropy": 5.497908210754394,
      "epoch": 0.7438241587239837,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004950165284681631,
      "loss": 5.2363,
      "mean_token_accuracy": 0.17994905412197112,
      "num_tokens": 16601845.0,
      "step": 9560
    },
    {
      "entropy": 5.543626499176026,
      "epoch": 0.7442131880957013,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004950107258955618,
      "loss": 5.3367,
      "mean_token_accuracy": 0.18117440193891526,
      "num_tokens": 16610438.0,
      "step": 9565
    },
    {
      "entropy": 5.498148488998413,
      "epoch": 0.7446022174674188,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004950049199846479,
      "loss": 5.2462,
      "mean_token_accuracy": 0.1805313155055046,
      "num_tokens": 16618892.0,
      "step": 9570
    },
    {
      "entropy": 5.572566413879395,
      "epoch": 0.7449912468391363,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004949991107355095,
      "loss": 5.2908,
      "mean_token_accuracy": 0.17819033265113832,
      "num_tokens": 16627702.0,
      "step": 9575
    },
    {
      "entropy": 5.466162872314453,
      "epoch": 0.7453802762108539,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004949932981482347,
      "loss": 5.2019,
      "mean_token_accuracy": 0.19009317308664322,
      "num_tokens": 16635653.0,
      "step": 9580
    },
    {
      "entropy": 5.533832740783692,
      "epoch": 0.7457693055825715,
      "grad_norm": 1.0,
      "learning_rate": 0.0004949874822229118,
      "loss": 5.3352,
      "mean_token_accuracy": 0.18005185723304748,
      "num_tokens": 16643898.0,
      "step": 9585
    },
    {
      "entropy": 5.530915832519531,
      "epoch": 0.7461583349542891,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004949816629596288,
      "loss": 5.1988,
      "mean_token_accuracy": 0.181541246175766,
      "num_tokens": 16653126.0,
      "step": 9590
    },
    {
      "entropy": 5.581034755706787,
      "epoch": 0.7465473643260067,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004949758403584746,
      "loss": 5.3189,
      "mean_token_accuracy": 0.17657433748245238,
      "num_tokens": 16661561.0,
      "step": 9595
    },
    {
      "entropy": 5.49182825088501,
      "epoch": 0.7469363936977241,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004949700144195368,
      "loss": 5.2766,
      "mean_token_accuracy": 0.18048131614923477,
      "num_tokens": 16669657.0,
      "step": 9600
    },
    {
      "entropy": 5.481636905670166,
      "epoch": 0.7473254230694417,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004949641851429043,
      "loss": 5.268,
      "mean_token_accuracy": 0.18197850584983827,
      "num_tokens": 16678486.0,
      "step": 9605
    },
    {
      "entropy": 5.559024095535278,
      "epoch": 0.7477144524411593,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004949583525286654,
      "loss": 5.2585,
      "mean_token_accuracy": 0.18317112773656846,
      "num_tokens": 16686638.0,
      "step": 9610
    },
    {
      "entropy": 5.522567462921143,
      "epoch": 0.7481034818128769,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004949525165769085,
      "loss": 5.2011,
      "mean_token_accuracy": 0.18153988122940062,
      "num_tokens": 16694393.0,
      "step": 9615
    },
    {
      "entropy": 5.504567670822143,
      "epoch": 0.7484925111845945,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004949466772877224,
      "loss": 5.2434,
      "mean_token_accuracy": 0.18640972673892975,
      "num_tokens": 16702613.0,
      "step": 9620
    },
    {
      "entropy": 5.529562139511109,
      "epoch": 0.748881540556312,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004949408346611955,
      "loss": 5.2186,
      "mean_token_accuracy": 0.182411490380764,
      "num_tokens": 16711210.0,
      "step": 9625
    },
    {
      "entropy": 5.528042507171631,
      "epoch": 0.7492705699280295,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004949349886974165,
      "loss": 5.2634,
      "mean_token_accuracy": 0.18221026360988618,
      "num_tokens": 16719124.0,
      "step": 9630
    },
    {
      "entropy": 5.569600439071655,
      "epoch": 0.7496595992997471,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004949291393964741,
      "loss": 5.2584,
      "mean_token_accuracy": 0.18103076219558717,
      "num_tokens": 16727240.0,
      "step": 9635
    },
    {
      "entropy": 5.534894704818726,
      "epoch": 0.7500486286714647,
      "grad_norm": 1.03125,
      "learning_rate": 0.000494923286758457,
      "loss": 5.3542,
      "mean_token_accuracy": 0.18086203634738923,
      "num_tokens": 16736524.0,
      "step": 9640
    },
    {
      "entropy": 5.64549994468689,
      "epoch": 0.7504376580431823,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004949174307834541,
      "loss": 5.4704,
      "mean_token_accuracy": 0.1627942994236946,
      "num_tokens": 16745798.0,
      "step": 9645
    },
    {
      "entropy": 5.516387987136841,
      "epoch": 0.7508266874148998,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004949115714715543,
      "loss": 5.2499,
      "mean_token_accuracy": 0.18076571077108383,
      "num_tokens": 16754263.0,
      "step": 9650
    },
    {
      "entropy": 5.4704972267150875,
      "epoch": 0.7512157167866174,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004949057088228465,
      "loss": 5.1385,
      "mean_token_accuracy": 0.18984874337911606,
      "num_tokens": 16762610.0,
      "step": 9655
    },
    {
      "entropy": 5.567148923873901,
      "epoch": 0.751604746158335,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004948998428374194,
      "loss": 5.3414,
      "mean_token_accuracy": 0.17255682349205018,
      "num_tokens": 16771436.0,
      "step": 9660
    },
    {
      "entropy": 5.585309839248657,
      "epoch": 0.7519937755300525,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004948939735153622,
      "loss": 5.3799,
      "mean_token_accuracy": 0.17757310569286347,
      "num_tokens": 16779647.0,
      "step": 9665
    },
    {
      "entropy": 5.614080619812012,
      "epoch": 0.7523828049017701,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004948881008567641,
      "loss": 5.333,
      "mean_token_accuracy": 0.17550697177648544,
      "num_tokens": 16788116.0,
      "step": 9670
    },
    {
      "entropy": 5.495042705535889,
      "epoch": 0.7527718342734876,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004948822248617139,
      "loss": 5.2703,
      "mean_token_accuracy": 0.1816123068332672,
      "num_tokens": 16796518.0,
      "step": 9675
    },
    {
      "entropy": 5.49826717376709,
      "epoch": 0.7531608636452052,
      "grad_norm": 1.015625,
      "learning_rate": 0.000494876345530301,
      "loss": 5.2444,
      "mean_token_accuracy": 0.1791821837425232,
      "num_tokens": 16805511.0,
      "step": 9680
    },
    {
      "entropy": 5.540137815475464,
      "epoch": 0.7535498930169228,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004948704628626145,
      "loss": 5.261,
      "mean_token_accuracy": 0.17889161556959152,
      "num_tokens": 16813173.0,
      "step": 9685
    },
    {
      "entropy": 5.580785131454467,
      "epoch": 0.7539389223886404,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004948645768587437,
      "loss": 5.3507,
      "mean_token_accuracy": 0.16703595072031022,
      "num_tokens": 16822193.0,
      "step": 9690
    },
    {
      "entropy": 5.465354681015015,
      "epoch": 0.7543279517603579,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004948586875187778,
      "loss": 5.0764,
      "mean_token_accuracy": 0.19261002391576768,
      "num_tokens": 16830421.0,
      "step": 9695
    },
    {
      "entropy": 5.512388849258423,
      "epoch": 0.7547169811320755,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004948527948428064,
      "loss": 5.2036,
      "mean_token_accuracy": 0.17943453192710876,
      "num_tokens": 16838988.0,
      "step": 9700
    },
    {
      "entropy": 5.484957075119018,
      "epoch": 0.755106010503793,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004948468988309187,
      "loss": 5.3333,
      "mean_token_accuracy": 0.1767300397157669,
      "num_tokens": 16847980.0,
      "step": 9705
    },
    {
      "entropy": 5.452110004425049,
      "epoch": 0.7554950398755106,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004948409994832043,
      "loss": 5.2113,
      "mean_token_accuracy": 0.18523395657539368,
      "num_tokens": 16857263.0,
      "step": 9710
    },
    {
      "entropy": 5.5448160648345945,
      "epoch": 0.7558840692472282,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004948350967997526,
      "loss": 5.2158,
      "mean_token_accuracy": 0.18251669704914092,
      "num_tokens": 16865314.0,
      "step": 9715
    },
    {
      "entropy": 5.534396171569824,
      "epoch": 0.7562730986189458,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004948291907806532,
      "loss": 5.304,
      "mean_token_accuracy": 0.18537853807210922,
      "num_tokens": 16875359.0,
      "step": 9720
    },
    {
      "entropy": 5.566549253463745,
      "epoch": 0.7566621279906633,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004948232814259957,
      "loss": 5.4211,
      "mean_token_accuracy": 0.16794775128364564,
      "num_tokens": 16885279.0,
      "step": 9725
    },
    {
      "entropy": 5.602229070663452,
      "epoch": 0.7570511573623808,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004948173687358699,
      "loss": 5.3507,
      "mean_token_accuracy": 0.17869512736797333,
      "num_tokens": 16894463.0,
      "step": 9730
    },
    {
      "entropy": 5.538092756271363,
      "epoch": 0.7574401867340984,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004948114527103652,
      "loss": 5.2818,
      "mean_token_accuracy": 0.17454845756292342,
      "num_tokens": 16903378.0,
      "step": 9735
    },
    {
      "entropy": 5.518202066421509,
      "epoch": 0.757829216105816,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004948055333495717,
      "loss": 5.3252,
      "mean_token_accuracy": 0.17525799125432967,
      "num_tokens": 16912039.0,
      "step": 9740
    },
    {
      "entropy": 5.539395523071289,
      "epoch": 0.7582182454775336,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004947996106535791,
      "loss": 5.2808,
      "mean_token_accuracy": 0.17099659740924836,
      "num_tokens": 16920516.0,
      "step": 9745
    },
    {
      "entropy": 5.622849845886231,
      "epoch": 0.7586072748492512,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004947936846224773,
      "loss": 5.284,
      "mean_token_accuracy": 0.17632976323366165,
      "num_tokens": 16929131.0,
      "step": 9750
    },
    {
      "entropy": 5.504240465164185,
      "epoch": 0.7589963042209686,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004947877552563562,
      "loss": 5.3125,
      "mean_token_accuracy": 0.1821568250656128,
      "num_tokens": 16937352.0,
      "step": 9755
    },
    {
      "entropy": 5.546800994873047,
      "epoch": 0.7593853335926862,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004947818225553056,
      "loss": 5.3992,
      "mean_token_accuracy": 0.17383539080619811,
      "num_tokens": 16946439.0,
      "step": 9760
    },
    {
      "entropy": 5.604864501953125,
      "epoch": 0.7597743629644038,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004947758865194156,
      "loss": 5.2725,
      "mean_token_accuracy": 0.18054522275924684,
      "num_tokens": 16955402.0,
      "step": 9765
    },
    {
      "entropy": 5.453227996826172,
      "epoch": 0.7601633923361214,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004947699471487766,
      "loss": 5.2024,
      "mean_token_accuracy": 0.18543512225151063,
      "num_tokens": 16964299.0,
      "step": 9770
    },
    {
      "entropy": 5.459178352355957,
      "epoch": 0.760552421707839,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004947640044434782,
      "loss": 5.2578,
      "mean_token_accuracy": 0.17591900676488875,
      "num_tokens": 16973483.0,
      "step": 9775
    },
    {
      "entropy": 5.535987854003906,
      "epoch": 0.7609414510795565,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004947580584036109,
      "loss": 5.2883,
      "mean_token_accuracy": 0.18448403477668762,
      "num_tokens": 16982234.0,
      "step": 9780
    },
    {
      "entropy": 5.533037519454956,
      "epoch": 0.761330480451274,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004947521090292649,
      "loss": 5.265,
      "mean_token_accuracy": 0.1792624056339264,
      "num_tokens": 16990669.0,
      "step": 9785
    },
    {
      "entropy": 5.561952829360962,
      "epoch": 0.7617195098229916,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004947461563205304,
      "loss": 5.3137,
      "mean_token_accuracy": 0.18060989528894425,
      "num_tokens": 16999998.0,
      "step": 9790
    },
    {
      "entropy": 5.5878520011901855,
      "epoch": 0.7621085391947092,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004947402002774977,
      "loss": 5.3279,
      "mean_token_accuracy": 0.17748137563467026,
      "num_tokens": 17008518.0,
      "step": 9795
    },
    {
      "entropy": 5.503102159500122,
      "epoch": 0.7624975685664268,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004947342409002572,
      "loss": 5.2534,
      "mean_token_accuracy": 0.17506785690784454,
      "num_tokens": 17017182.0,
      "step": 9800
    },
    {
      "entropy": 5.488873767852783,
      "epoch": 0.7628865979381443,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004947282781888994,
      "loss": 5.2721,
      "mean_token_accuracy": 0.17734420150518418,
      "num_tokens": 17026186.0,
      "step": 9805
    },
    {
      "entropy": 5.450332832336426,
      "epoch": 0.7632756273098619,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004947223121435147,
      "loss": 5.159,
      "mean_token_accuracy": 0.1892400339245796,
      "num_tokens": 17034295.0,
      "step": 9810
    },
    {
      "entropy": 5.440847682952881,
      "epoch": 0.7636646566815795,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004947163427641936,
      "loss": 5.2303,
      "mean_token_accuracy": 0.1812972143292427,
      "num_tokens": 17043195.0,
      "step": 9815
    },
    {
      "entropy": 5.4579917907714846,
      "epoch": 0.764053686053297,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004947103700510268,
      "loss": 5.1796,
      "mean_token_accuracy": 0.18050909489393235,
      "num_tokens": 17051903.0,
      "step": 9820
    },
    {
      "entropy": 5.465172958374024,
      "epoch": 0.7644427154250146,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004947043940041047,
      "loss": 5.2377,
      "mean_token_accuracy": 0.18154272735118865,
      "num_tokens": 17060495.0,
      "step": 9825
    },
    {
      "entropy": 5.5413196086883545,
      "epoch": 0.7648317447967321,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004946984146235183,
      "loss": 5.2918,
      "mean_token_accuracy": 0.17810165584087373,
      "num_tokens": 17069254.0,
      "step": 9830
    },
    {
      "entropy": 5.59796724319458,
      "epoch": 0.7652207741684497,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004946924319093579,
      "loss": 5.2677,
      "mean_token_accuracy": 0.17897392958402633,
      "num_tokens": 17077323.0,
      "step": 9835
    },
    {
      "entropy": 5.4760113716125485,
      "epoch": 0.7656098035401673,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004946864458617148,
      "loss": 5.1319,
      "mean_token_accuracy": 0.18981318324804305,
      "num_tokens": 17085377.0,
      "step": 9840
    },
    {
      "entropy": 5.4714634895324705,
      "epoch": 0.7659988329118849,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004946804564806793,
      "loss": 5.2258,
      "mean_token_accuracy": 0.18119042813777925,
      "num_tokens": 17094288.0,
      "step": 9845
    },
    {
      "entropy": 5.519982385635376,
      "epoch": 0.7663878622836025,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004946744637663427,
      "loss": 5.1533,
      "mean_token_accuracy": 0.18477026522159576,
      "num_tokens": 17101572.0,
      "step": 9850
    },
    {
      "entropy": 5.506848096847534,
      "epoch": 0.7667768916553199,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004946684677187956,
      "loss": 5.289,
      "mean_token_accuracy": 0.17754042744636536,
      "num_tokens": 17110151.0,
      "step": 9855
    },
    {
      "entropy": 5.5674072265625,
      "epoch": 0.7671659210270375,
      "grad_norm": 1.125,
      "learning_rate": 0.0004946624683381292,
      "loss": 5.2846,
      "mean_token_accuracy": 0.18072846978902818,
      "num_tokens": 17118172.0,
      "step": 9860
    },
    {
      "entropy": 5.476708507537841,
      "epoch": 0.7675549503987551,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004946564656244345,
      "loss": 5.1486,
      "mean_token_accuracy": 0.18821113854646682,
      "num_tokens": 17126855.0,
      "step": 9865
    },
    {
      "entropy": 5.535451030731201,
      "epoch": 0.7679439797704727,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004946504595778026,
      "loss": 5.3531,
      "mean_token_accuracy": 0.1740744024515152,
      "num_tokens": 17135737.0,
      "step": 9870
    },
    {
      "entropy": 5.6018438816070555,
      "epoch": 0.7683330091421903,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004946444501983246,
      "loss": 5.2791,
      "mean_token_accuracy": 0.17825541496276856,
      "num_tokens": 17144521.0,
      "step": 9875
    },
    {
      "entropy": 5.597659730911255,
      "epoch": 0.7687220385139077,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004946384374860916,
      "loss": 5.3698,
      "mean_token_accuracy": 0.18095106333494188,
      "num_tokens": 17153534.0,
      "step": 9880
    },
    {
      "entropy": 5.435767364501953,
      "epoch": 0.7691110678856253,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004946324214411949,
      "loss": 5.2109,
      "mean_token_accuracy": 0.1844596818089485,
      "num_tokens": 17162600.0,
      "step": 9885
    },
    {
      "entropy": 5.507652425765992,
      "epoch": 0.7695000972573429,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004946264020637259,
      "loss": 5.2974,
      "mean_token_accuracy": 0.17652562260627747,
      "num_tokens": 17171373.0,
      "step": 9890
    },
    {
      "entropy": 5.573658990859985,
      "epoch": 0.7698891266290605,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004946203793537757,
      "loss": 5.319,
      "mean_token_accuracy": 0.17605869472026825,
      "num_tokens": 17180090.0,
      "step": 9895
    },
    {
      "entropy": 5.5605254650115965,
      "epoch": 0.7702781560007781,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004946143533114358,
      "loss": 5.2308,
      "mean_token_accuracy": 0.18466345667839051,
      "num_tokens": 17189592.0,
      "step": 9900
    },
    {
      "entropy": 5.461719942092896,
      "epoch": 0.7706671853724957,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004946083239367976,
      "loss": 5.1833,
      "mean_token_accuracy": 0.19113331586122512,
      "num_tokens": 17198287.0,
      "step": 9905
    },
    {
      "entropy": 5.5281964302062985,
      "epoch": 0.7710562147442132,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004946022912299527,
      "loss": 5.2959,
      "mean_token_accuracy": 0.17453667372465134,
      "num_tokens": 17206349.0,
      "step": 9910
    },
    {
      "entropy": 5.54696741104126,
      "epoch": 0.7714452441159307,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004945962551909926,
      "loss": 5.3897,
      "mean_token_accuracy": 0.1722372964024544,
      "num_tokens": 17215408.0,
      "step": 9915
    },
    {
      "entropy": 5.617879199981689,
      "epoch": 0.7718342734876483,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004945902158200089,
      "loss": 5.3325,
      "mean_token_accuracy": 0.17207663804292678,
      "num_tokens": 17224342.0,
      "step": 9920
    },
    {
      "entropy": 5.5664297580719,
      "epoch": 0.7722233028593659,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004945841731170931,
      "loss": 5.2001,
      "mean_token_accuracy": 0.17807358354330063,
      "num_tokens": 17232532.0,
      "step": 9925
    },
    {
      "entropy": 5.462725687026977,
      "epoch": 0.7726123322310835,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004945781270823369,
      "loss": 5.2169,
      "mean_token_accuracy": 0.18606288135051727,
      "num_tokens": 17240786.0,
      "step": 9930
    },
    {
      "entropy": 5.432136535644531,
      "epoch": 0.773001361602801,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004945720777158323,
      "loss": 5.2066,
      "mean_token_accuracy": 0.18757349550724028,
      "num_tokens": 17248515.0,
      "step": 9935
    },
    {
      "entropy": 5.525846767425537,
      "epoch": 0.7733903909745186,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004945660250176708,
      "loss": 5.2626,
      "mean_token_accuracy": 0.1868579715490341,
      "num_tokens": 17257098.0,
      "step": 9940
    },
    {
      "entropy": 5.517870712280273,
      "epoch": 0.7737794203462361,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004945599689879443,
      "loss": 5.2596,
      "mean_token_accuracy": 0.17976729124784468,
      "num_tokens": 17266851.0,
      "step": 9945
    },
    {
      "entropy": 5.509455251693725,
      "epoch": 0.7741684497179537,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004945539096267448,
      "loss": 5.2049,
      "mean_token_accuracy": 0.1737503170967102,
      "num_tokens": 17275568.0,
      "step": 9950
    },
    {
      "entropy": 5.449204635620117,
      "epoch": 0.7745574790896713,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004945478469341642,
      "loss": 5.1682,
      "mean_token_accuracy": 0.18405324965715408,
      "num_tokens": 17284528.0,
      "step": 9955
    },
    {
      "entropy": 5.553600120544433,
      "epoch": 0.7749465084613888,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004945417809102944,
      "loss": 5.3154,
      "mean_token_accuracy": 0.17621127963066102,
      "num_tokens": 17293973.0,
      "step": 9960
    },
    {
      "entropy": 5.490083122253418,
      "epoch": 0.7753355378331064,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004945357115552275,
      "loss": 5.2114,
      "mean_token_accuracy": 0.1820702999830246,
      "num_tokens": 17302321.0,
      "step": 9965
    },
    {
      "entropy": 5.538445663452149,
      "epoch": 0.775724567204824,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004945296388690557,
      "loss": 5.3322,
      "mean_token_accuracy": 0.17512760311365128,
      "num_tokens": 17311226.0,
      "step": 9970
    },
    {
      "entropy": 5.534921550750733,
      "epoch": 0.7761135965765416,
      "grad_norm": 1.015625,
      "learning_rate": 0.000494523562851871,
      "loss": 5.3037,
      "mean_token_accuracy": 0.1739043802022934,
      "num_tokens": 17319818.0,
      "step": 9975
    },
    {
      "entropy": 5.574634170532226,
      "epoch": 0.7765026259482591,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004945174835037655,
      "loss": 5.3285,
      "mean_token_accuracy": 0.17049792557954788,
      "num_tokens": 17329195.0,
      "step": 9980
    },
    {
      "entropy": 5.60396032333374,
      "epoch": 0.7768916553199766,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004945114008248319,
      "loss": 5.4076,
      "mean_token_accuracy": 0.1777074322104454,
      "num_tokens": 17337689.0,
      "step": 9985
    },
    {
      "entropy": 5.500341939926147,
      "epoch": 0.7772806846916942,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004945053148151619,
      "loss": 5.1745,
      "mean_token_accuracy": 0.18221574276685715,
      "num_tokens": 17345467.0,
      "step": 9990
    },
    {
      "entropy": 5.516344976425171,
      "epoch": 0.7776697140634118,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004944992254748482,
      "loss": 5.3069,
      "mean_token_accuracy": 0.18074913024902345,
      "num_tokens": 17353651.0,
      "step": 9995
    },
    {
      "entropy": 5.572245693206787,
      "epoch": 0.7780587434351294,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004944931328039832,
      "loss": 5.3591,
      "mean_token_accuracy": 0.17541846483945847,
      "num_tokens": 17362777.0,
      "step": 10000
    },
    {
      "entropy": 5.618011474609375,
      "epoch": 0.778447772806847,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004944870368026592,
      "loss": 5.2966,
      "mean_token_accuracy": 0.1844174712896347,
      "num_tokens": 17371562.0,
      "step": 10005
    },
    {
      "entropy": 5.573336505889893,
      "epoch": 0.7788368021785644,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004944809374709689,
      "loss": 5.2968,
      "mean_token_accuracy": 0.1739783212542534,
      "num_tokens": 17381185.0,
      "step": 10010
    },
    {
      "entropy": 5.526984977722168,
      "epoch": 0.779225831550282,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004944748348090045,
      "loss": 5.2588,
      "mean_token_accuracy": 0.18662239909172057,
      "num_tokens": 17390658.0,
      "step": 10015
    },
    {
      "entropy": 5.488504362106323,
      "epoch": 0.7796148609219996,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004944687288168589,
      "loss": 5.2047,
      "mean_token_accuracy": 0.18196306079626084,
      "num_tokens": 17399968.0,
      "step": 10020
    },
    {
      "entropy": 5.522232389450073,
      "epoch": 0.7800038902937172,
      "grad_norm": 1.0,
      "learning_rate": 0.0004944626194946246,
      "loss": 5.3204,
      "mean_token_accuracy": 0.18025138229131699,
      "num_tokens": 17408727.0,
      "step": 10025
    },
    {
      "entropy": 5.570869874954224,
      "epoch": 0.7803929196654348,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004944565068423945,
      "loss": 5.3448,
      "mean_token_accuracy": 0.1805488184094429,
      "num_tokens": 17417517.0,
      "step": 10030
    },
    {
      "entropy": 5.616558599472046,
      "epoch": 0.7807819490371523,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004944503908602612,
      "loss": 5.3253,
      "mean_token_accuracy": 0.17853676080703734,
      "num_tokens": 17427035.0,
      "step": 10035
    },
    {
      "entropy": 5.564046573638916,
      "epoch": 0.7811709784088698,
      "grad_norm": 1.0,
      "learning_rate": 0.0004944442715483174,
      "loss": 5.231,
      "mean_token_accuracy": 0.18222047686576842,
      "num_tokens": 17435585.0,
      "step": 10040
    },
    {
      "entropy": 5.506319952011109,
      "epoch": 0.7815600077805874,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004944381489066561,
      "loss": 5.2416,
      "mean_token_accuracy": 0.1860206753015518,
      "num_tokens": 17444317.0,
      "step": 10045
    },
    {
      "entropy": 5.58783745765686,
      "epoch": 0.781949037152305,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004944320229353702,
      "loss": 5.4476,
      "mean_token_accuracy": 0.1736738547682762,
      "num_tokens": 17453585.0,
      "step": 10050
    },
    {
      "entropy": 5.547242116928101,
      "epoch": 0.7823380665240226,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004944258936345526,
      "loss": 5.2327,
      "mean_token_accuracy": 0.18293302953243257,
      "num_tokens": 17462233.0,
      "step": 10055
    },
    {
      "entropy": 5.514138555526733,
      "epoch": 0.7827270958957401,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004944197610042963,
      "loss": 5.1458,
      "mean_token_accuracy": 0.19179946631193162,
      "num_tokens": 17470566.0,
      "step": 10060
    },
    {
      "entropy": 5.492088794708252,
      "epoch": 0.7831161252674577,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004944136250446944,
      "loss": 5.2631,
      "mean_token_accuracy": 0.17675694227218627,
      "num_tokens": 17479267.0,
      "step": 10065
    },
    {
      "entropy": 5.409222269058228,
      "epoch": 0.7835051546391752,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004944074857558399,
      "loss": 5.1497,
      "mean_token_accuracy": 0.1882484272122383,
      "num_tokens": 17488060.0,
      "step": 10070
    },
    {
      "entropy": 5.612513160705566,
      "epoch": 0.7838941840108928,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004944013431378261,
      "loss": 5.3965,
      "mean_token_accuracy": 0.1748053565621376,
      "num_tokens": 17497240.0,
      "step": 10075
    },
    {
      "entropy": 5.58105058670044,
      "epoch": 0.7842832133826104,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004943951971907461,
      "loss": 5.3469,
      "mean_token_accuracy": 0.18082948327064513,
      "num_tokens": 17506017.0,
      "step": 10080
    },
    {
      "entropy": 5.59242091178894,
      "epoch": 0.7846722427543279,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004943890479146933,
      "loss": 5.2865,
      "mean_token_accuracy": 0.17967495024204255,
      "num_tokens": 17514764.0,
      "step": 10085
    },
    {
      "entropy": 5.498639869689941,
      "epoch": 0.7850612721260455,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004943828953097607,
      "loss": 5.2671,
      "mean_token_accuracy": 0.18300705999135972,
      "num_tokens": 17524183.0,
      "step": 10090
    },
    {
      "entropy": 5.503949403762817,
      "epoch": 0.7854503014977631,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000494376739376042,
      "loss": 5.2222,
      "mean_token_accuracy": 0.1848127454519272,
      "num_tokens": 17533426.0,
      "step": 10095
    },
    {
      "entropy": 5.5370659828186035,
      "epoch": 0.7858393308694807,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004943705801136304,
      "loss": 5.1687,
      "mean_token_accuracy": 0.18645709604024888,
      "num_tokens": 17541966.0,
      "step": 10100
    },
    {
      "entropy": 5.533256196975708,
      "epoch": 0.7862283602411982,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004943644175226193,
      "loss": 5.2652,
      "mean_token_accuracy": 0.18316768556833268,
      "num_tokens": 17551157.0,
      "step": 10105
    },
    {
      "entropy": 5.458074617385864,
      "epoch": 0.7866173896129158,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004943582516031025,
      "loss": 5.1811,
      "mean_token_accuracy": 0.189007468521595,
      "num_tokens": 17559999.0,
      "step": 10110
    },
    {
      "entropy": 5.513240242004395,
      "epoch": 0.7870064189846333,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004943520823551732,
      "loss": 5.2271,
      "mean_token_accuracy": 0.18528834134340286,
      "num_tokens": 17568873.0,
      "step": 10115
    },
    {
      "entropy": 5.5409516334533695,
      "epoch": 0.7873954483563509,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004943459097789252,
      "loss": 5.3041,
      "mean_token_accuracy": 0.18329188078641892,
      "num_tokens": 17577807.0,
      "step": 10120
    },
    {
      "entropy": 5.560406827926636,
      "epoch": 0.7877844777280685,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004943397338744523,
      "loss": 5.3238,
      "mean_token_accuracy": 0.17690241634845733,
      "num_tokens": 17586646.0,
      "step": 10125
    },
    {
      "entropy": 5.589386796951294,
      "epoch": 0.7881735070997861,
      "grad_norm": 1.0,
      "learning_rate": 0.000494333554641848,
      "loss": 5.395,
      "mean_token_accuracy": 0.16847542375326158,
      "num_tokens": 17595458.0,
      "step": 10130
    },
    {
      "entropy": 5.563756370544434,
      "epoch": 0.7885625364715037,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000494327372081206,
      "loss": 5.2995,
      "mean_token_accuracy": 0.18116143494844436,
      "num_tokens": 17603997.0,
      "step": 10135
    },
    {
      "entropy": 5.601966714859008,
      "epoch": 0.7889515658432211,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004943211861926204,
      "loss": 5.3202,
      "mean_token_accuracy": 0.17464739382266997,
      "num_tokens": 17612347.0,
      "step": 10140
    },
    {
      "entropy": 5.494529962539673,
      "epoch": 0.7893405952149387,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004943149969761848,
      "loss": 5.1632,
      "mean_token_accuracy": 0.191545370221138,
      "num_tokens": 17620618.0,
      "step": 10145
    },
    {
      "entropy": 5.448789024353028,
      "epoch": 0.7897296245866563,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004943088044319932,
      "loss": 5.1907,
      "mean_token_accuracy": 0.18437443375587464,
      "num_tokens": 17629164.0,
      "step": 10150
    },
    {
      "entropy": 5.618792629241943,
      "epoch": 0.7901186539583739,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004943026085601395,
      "loss": 5.3589,
      "mean_token_accuracy": 0.17427270859479904,
      "num_tokens": 17638454.0,
      "step": 10155
    },
    {
      "entropy": 5.549109601974488,
      "epoch": 0.7905076833300915,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004942964093607178,
      "loss": 5.2874,
      "mean_token_accuracy": 0.17659604996442796,
      "num_tokens": 17647099.0,
      "step": 10160
    },
    {
      "entropy": 5.564661741256714,
      "epoch": 0.790896712701809,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004942902068338222,
      "loss": 5.2452,
      "mean_token_accuracy": 0.18648452907800675,
      "num_tokens": 17655214.0,
      "step": 10165
    },
    {
      "entropy": 5.491228485107422,
      "epoch": 0.7912857420735265,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004942840009795466,
      "loss": 5.2063,
      "mean_token_accuracy": 0.17931612581014633,
      "num_tokens": 17663342.0,
      "step": 10170
    },
    {
      "entropy": 5.658391427993775,
      "epoch": 0.7916747714452441,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004942777917979855,
      "loss": 5.4204,
      "mean_token_accuracy": 0.16952992528676986,
      "num_tokens": 17671725.0,
      "step": 10175
    },
    {
      "entropy": 5.525932121276855,
      "epoch": 0.7920638008169617,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004942715792892328,
      "loss": 5.3288,
      "mean_token_accuracy": 0.17597885876893998,
      "num_tokens": 17680627.0,
      "step": 10180
    },
    {
      "entropy": 5.561020803451538,
      "epoch": 0.7924528301886793,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000494265363453383,
      "loss": 5.2875,
      "mean_token_accuracy": 0.18264877051115036,
      "num_tokens": 17688392.0,
      "step": 10185
    },
    {
      "entropy": 5.448076772689819,
      "epoch": 0.7928418595603968,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004942591442905302,
      "loss": 5.2227,
      "mean_token_accuracy": 0.18574137538671492,
      "num_tokens": 17697546.0,
      "step": 10190
    },
    {
      "entropy": 5.449612188339233,
      "epoch": 0.7932308889321144,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004942529218007689,
      "loss": 5.1929,
      "mean_token_accuracy": 0.18285220116376877,
      "num_tokens": 17705555.0,
      "step": 10195
    },
    {
      "entropy": 5.5286705493927,
      "epoch": 0.7936199183038319,
      "grad_norm": 1.0,
      "learning_rate": 0.0004942466959841936,
      "loss": 5.3656,
      "mean_token_accuracy": 0.17250725924968718,
      "num_tokens": 17715652.0,
      "step": 10200
    },
    {
      "entropy": 5.464176750183105,
      "epoch": 0.7940089476755495,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004942404668408985,
      "loss": 5.1144,
      "mean_token_accuracy": 0.1782376617193222,
      "num_tokens": 17724111.0,
      "step": 10205
    },
    {
      "entropy": 5.426705694198608,
      "epoch": 0.7943979770472671,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004942342343709783,
      "loss": 5.1159,
      "mean_token_accuracy": 0.18732880204916,
      "num_tokens": 17732732.0,
      "step": 10210
    },
    {
      "entropy": 5.491913557052612,
      "epoch": 0.7947870064189846,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004942279985745275,
      "loss": 5.2139,
      "mean_token_accuracy": 0.18290556520223616,
      "num_tokens": 17741338.0,
      "step": 10215
    },
    {
      "entropy": 5.592301082611084,
      "epoch": 0.7951760357907022,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004942217594516409,
      "loss": 5.4275,
      "mean_token_accuracy": 0.17138956785202025,
      "num_tokens": 17750651.0,
      "step": 10220
    },
    {
      "entropy": 5.554042720794678,
      "epoch": 0.7955650651624198,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000494215517002413,
      "loss": 5.242,
      "mean_token_accuracy": 0.1854332521557808,
      "num_tokens": 17759052.0,
      "step": 10225
    },
    {
      "entropy": 5.577816438674927,
      "epoch": 0.7959540945341373,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004942092712269384,
      "loss": 5.3031,
      "mean_token_accuracy": 0.1735773429274559,
      "num_tokens": 17767246.0,
      "step": 10230
    },
    {
      "entropy": 5.596259737014771,
      "epoch": 0.7963431239058549,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004942030221253122,
      "loss": 5.3244,
      "mean_token_accuracy": 0.18097350895404815,
      "num_tokens": 17777022.0,
      "step": 10235
    },
    {
      "entropy": 5.54957423210144,
      "epoch": 0.7967321532775724,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004941967696976289,
      "loss": 5.2868,
      "mean_token_accuracy": 0.1756106436252594,
      "num_tokens": 17785586.0,
      "step": 10240
    },
    {
      "entropy": 5.510403966903686,
      "epoch": 0.79712118264929,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004941905139439835,
      "loss": 5.2843,
      "mean_token_accuracy": 0.1757577121257782,
      "num_tokens": 17794320.0,
      "step": 10245
    },
    {
      "entropy": 5.568962144851684,
      "epoch": 0.7975102120210076,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004941842548644712,
      "loss": 5.2921,
      "mean_token_accuracy": 0.18555116951465606,
      "num_tokens": 17803384.0,
      "step": 10250
    },
    {
      "entropy": 5.568940734863281,
      "epoch": 0.7978992413927252,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004941779924591864,
      "loss": 5.266,
      "mean_token_accuracy": 0.17868881672620773,
      "num_tokens": 17811955.0,
      "step": 10255
    },
    {
      "entropy": 5.529417085647583,
      "epoch": 0.7982882707644428,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004941717267282244,
      "loss": 5.3099,
      "mean_token_accuracy": 0.17377544045448304,
      "num_tokens": 17820582.0,
      "step": 10260
    },
    {
      "entropy": 5.5706902027130125,
      "epoch": 0.7986773001361602,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004941654576716806,
      "loss": 5.2177,
      "mean_token_accuracy": 0.18322414755821229,
      "num_tokens": 17828465.0,
      "step": 10265
    },
    {
      "entropy": 5.516387939453125,
      "epoch": 0.7990663295078778,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004941591852896495,
      "loss": 5.2206,
      "mean_token_accuracy": 0.18754516094923018,
      "num_tokens": 17836931.0,
      "step": 10270
    },
    {
      "entropy": 5.539124536514282,
      "epoch": 0.7994553588795954,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004941529095822268,
      "loss": 5.3174,
      "mean_token_accuracy": 0.1779179736971855,
      "num_tokens": 17845192.0,
      "step": 10275
    },
    {
      "entropy": 5.532001733779907,
      "epoch": 0.799844388251313,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004941466305495074,
      "loss": 5.2916,
      "mean_token_accuracy": 0.1780277818441391,
      "num_tokens": 17854121.0,
      "step": 10280
    },
    {
      "entropy": 5.585430145263672,
      "epoch": 0.8002334176230306,
      "grad_norm": 1.0,
      "learning_rate": 0.0004941403481915867,
      "loss": 5.2761,
      "mean_token_accuracy": 0.18030497431755066,
      "num_tokens": 17862931.0,
      "step": 10285
    },
    {
      "entropy": 5.594873476028442,
      "epoch": 0.800622446994748,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004941340625085601,
      "loss": 5.2962,
      "mean_token_accuracy": 0.17580545395612718,
      "num_tokens": 17871909.0,
      "step": 10290
    },
    {
      "entropy": 5.489644289016724,
      "epoch": 0.8010114763664656,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004941277735005228,
      "loss": 5.2001,
      "mean_token_accuracy": 0.18702718764543533,
      "num_tokens": 17880487.0,
      "step": 10295
    },
    {
      "entropy": 5.514432144165039,
      "epoch": 0.8014005057381832,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004941214811675702,
      "loss": 5.2713,
      "mean_token_accuracy": 0.17735567837953567,
      "num_tokens": 17888712.0,
      "step": 10300
    },
    {
      "entropy": 5.581064176559448,
      "epoch": 0.8017895351099008,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004941151855097982,
      "loss": 5.3256,
      "mean_token_accuracy": 0.17269032448530197,
      "num_tokens": 17898163.0,
      "step": 10305
    },
    {
      "entropy": 5.517892456054687,
      "epoch": 0.8021785644816184,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004941088865273018,
      "loss": 5.2071,
      "mean_token_accuracy": 0.18535101115703584,
      "num_tokens": 17906849.0,
      "step": 10310
    },
    {
      "entropy": 5.538178634643555,
      "epoch": 0.802567593853336,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004941025842201769,
      "loss": 5.2435,
      "mean_token_accuracy": 0.17921622097492218,
      "num_tokens": 17915695.0,
      "step": 10315
    },
    {
      "entropy": 5.5844494819641115,
      "epoch": 0.8029566232250535,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004940962785885189,
      "loss": 5.3321,
      "mean_token_accuracy": 0.1785748451948166,
      "num_tokens": 17924706.0,
      "step": 10320
    },
    {
      "entropy": 5.5333826541900635,
      "epoch": 0.803345652596771,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004940899696324237,
      "loss": 5.3037,
      "mean_token_accuracy": 0.18385335057973862,
      "num_tokens": 17934061.0,
      "step": 10325
    },
    {
      "entropy": 5.504366207122803,
      "epoch": 0.8037346819684886,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004940836573519868,
      "loss": 5.183,
      "mean_token_accuracy": 0.1820183962583542,
      "num_tokens": 17942580.0,
      "step": 10330
    },
    {
      "entropy": 5.448630666732788,
      "epoch": 0.8041237113402062,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004940773417473043,
      "loss": 5.0817,
      "mean_token_accuracy": 0.18724218308925628,
      "num_tokens": 17951171.0,
      "step": 10335
    },
    {
      "entropy": 5.432222557067871,
      "epoch": 0.8045127407119238,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004940710228184717,
      "loss": 5.1643,
      "mean_token_accuracy": 0.1955781549215317,
      "num_tokens": 17959674.0,
      "step": 10340
    },
    {
      "entropy": 5.503968715667725,
      "epoch": 0.8049017700836413,
      "grad_norm": 0.96875,
      "learning_rate": 0.000494064700565585,
      "loss": 5.225,
      "mean_token_accuracy": 0.18283116370439528,
      "num_tokens": 17969036.0,
      "step": 10345
    },
    {
      "entropy": 5.507812595367431,
      "epoch": 0.8052907994553589,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004940583749887403,
      "loss": 5.2338,
      "mean_token_accuracy": 0.18076378256082534,
      "num_tokens": 17977406.0,
      "step": 10350
    },
    {
      "entropy": 5.493749332427979,
      "epoch": 0.8056798288270764,
      "grad_norm": 1.0,
      "learning_rate": 0.0004940520460880333,
      "loss": 5.2093,
      "mean_token_accuracy": 0.18296364247798919,
      "num_tokens": 17985981.0,
      "step": 10355
    },
    {
      "entropy": 5.566988134384156,
      "epoch": 0.806068858198794,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004940457138635601,
      "loss": 5.3981,
      "mean_token_accuracy": 0.17437873482704164,
      "num_tokens": 17994246.0,
      "step": 10360
    },
    {
      "entropy": 5.5437664031982425,
      "epoch": 0.8064578875705116,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004940393783154169,
      "loss": 5.2342,
      "mean_token_accuracy": 0.1805388644337654,
      "num_tokens": 18002444.0,
      "step": 10365
    },
    {
      "entropy": 5.502651309967041,
      "epoch": 0.8068469169422291,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004940330394436999,
      "loss": 5.1554,
      "mean_token_accuracy": 0.1860331952571869,
      "num_tokens": 18011076.0,
      "step": 10370
    },
    {
      "entropy": 5.549663925170899,
      "epoch": 0.8072359463139467,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000494026697248505,
      "loss": 5.4519,
      "mean_token_accuracy": 0.16915490329265595,
      "num_tokens": 18020232.0,
      "step": 10375
    },
    {
      "entropy": 5.585958099365234,
      "epoch": 0.8076249756856643,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004940203517299287,
      "loss": 5.2521,
      "mean_token_accuracy": 0.17841964066028596,
      "num_tokens": 18029055.0,
      "step": 10380
    },
    {
      "entropy": 5.617270517349243,
      "epoch": 0.8080140050573819,
      "grad_norm": 1.234375,
      "learning_rate": 0.000494014002888067,
      "loss": 5.3926,
      "mean_token_accuracy": 0.16873783469200135,
      "num_tokens": 18037827.0,
      "step": 10385
    },
    {
      "entropy": 5.46048493385315,
      "epoch": 0.8084030344290994,
      "grad_norm": 1.0,
      "learning_rate": 0.0004940076507230166,
      "loss": 5.2248,
      "mean_token_accuracy": 0.17724504470825195,
      "num_tokens": 18047017.0,
      "step": 10390
    },
    {
      "entropy": 5.539479303359985,
      "epoch": 0.8087920638008169,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004940012952348735,
      "loss": 5.3205,
      "mean_token_accuracy": 0.17949073314666747,
      "num_tokens": 18056256.0,
      "step": 10395
    },
    {
      "entropy": 5.51496729850769,
      "epoch": 0.8091810931725345,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004939949364237345,
      "loss": 5.0904,
      "mean_token_accuracy": 0.20243958830833436,
      "num_tokens": 18064127.0,
      "step": 10400
    },
    {
      "entropy": 5.475569438934326,
      "epoch": 0.8095701225442521,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004939885742896958,
      "loss": 5.245,
      "mean_token_accuracy": 0.18268114626407622,
      "num_tokens": 18072766.0,
      "step": 10405
    },
    {
      "entropy": 5.470303344726562,
      "epoch": 0.8099591519159697,
      "grad_norm": 1.03125,
      "learning_rate": 0.000493982208832854,
      "loss": 5.2202,
      "mean_token_accuracy": 0.18196416944265364,
      "num_tokens": 18080685.0,
      "step": 10410
    },
    {
      "entropy": 5.535878324508667,
      "epoch": 0.8103481812876873,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004939758400533058,
      "loss": 5.2894,
      "mean_token_accuracy": 0.17753487676382065,
      "num_tokens": 18090026.0,
      "step": 10415
    },
    {
      "entropy": 5.528467798233033,
      "epoch": 0.8107372106594047,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004939694679511478,
      "loss": 5.203,
      "mean_token_accuracy": 0.18221452385187148,
      "num_tokens": 18098638.0,
      "step": 10420
    },
    {
      "entropy": 5.50969500541687,
      "epoch": 0.8111262400311223,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004939630925264765,
      "loss": 5.2425,
      "mean_token_accuracy": 0.18338799327611924,
      "num_tokens": 18106844.0,
      "step": 10425
    },
    {
      "entropy": 5.5410887718200685,
      "epoch": 0.8115152694028399,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000493956713779389,
      "loss": 5.2232,
      "mean_token_accuracy": 0.1803697630763054,
      "num_tokens": 18114758.0,
      "step": 10430
    },
    {
      "entropy": 5.4937553882598875,
      "epoch": 0.8119042987745575,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004939503317099817,
      "loss": 5.2037,
      "mean_token_accuracy": 0.18593893200159073,
      "num_tokens": 18122839.0,
      "step": 10435
    },
    {
      "entropy": 5.519395208358764,
      "epoch": 0.8122933281462751,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004939439463183517,
      "loss": 5.2883,
      "mean_token_accuracy": 0.18012109100818635,
      "num_tokens": 18131709.0,
      "step": 10440
    },
    {
      "entropy": 5.525179719924926,
      "epoch": 0.8126823575179926,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004939375576045958,
      "loss": 5.265,
      "mean_token_accuracy": 0.18175515234470369,
      "num_tokens": 18140334.0,
      "step": 10445
    },
    {
      "entropy": 5.473891019821167,
      "epoch": 0.8130713868897101,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004939311655688109,
      "loss": 5.2583,
      "mean_token_accuracy": 0.18083884567022324,
      "num_tokens": 18148648.0,
      "step": 10450
    },
    {
      "entropy": 5.5419450283050535,
      "epoch": 0.8134604162614277,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004939247702110941,
      "loss": 5.2754,
      "mean_token_accuracy": 0.17854345887899398,
      "num_tokens": 18156891.0,
      "step": 10455
    },
    {
      "entropy": 5.51627926826477,
      "epoch": 0.8138494456331453,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004939183715315423,
      "loss": 5.228,
      "mean_token_accuracy": 0.18572111874818803,
      "num_tokens": 18166512.0,
      "step": 10460
    },
    {
      "entropy": 5.550294589996338,
      "epoch": 0.8142384750048629,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004939119695302526,
      "loss": 5.1859,
      "mean_token_accuracy": 0.1846105068922043,
      "num_tokens": 18175064.0,
      "step": 10465
    },
    {
      "entropy": 5.531307315826416,
      "epoch": 0.8146275043765804,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004939055642073224,
      "loss": 5.3142,
      "mean_token_accuracy": 0.17835467159748078,
      "num_tokens": 18184251.0,
      "step": 10470
    },
    {
      "entropy": 5.4800835132598875,
      "epoch": 0.815016533748298,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004938991555628484,
      "loss": 5.1502,
      "mean_token_accuracy": 0.1907275840640068,
      "num_tokens": 18193274.0,
      "step": 10475
    },
    {
      "entropy": 5.505615901947022,
      "epoch": 0.8154055631200156,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004938927435969283,
      "loss": 5.2064,
      "mean_token_accuracy": 0.1893477365374565,
      "num_tokens": 18201878.0,
      "step": 10480
    },
    {
      "entropy": 5.4706063747406,
      "epoch": 0.8157945924917331,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004938863283096592,
      "loss": 5.1703,
      "mean_token_accuracy": 0.1867490217089653,
      "num_tokens": 18210343.0,
      "step": 10485
    },
    {
      "entropy": 5.578630018234253,
      "epoch": 0.8161836218634507,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004938799097011384,
      "loss": 5.3058,
      "mean_token_accuracy": 0.18627608716487884,
      "num_tokens": 18218599.0,
      "step": 10490
    },
    {
      "entropy": 5.50228066444397,
      "epoch": 0.8165726512351682,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004938734877714634,
      "loss": 5.1883,
      "mean_token_accuracy": 0.193308225274086,
      "num_tokens": 18226384.0,
      "step": 10495
    },
    {
      "entropy": 5.4807662010192875,
      "epoch": 0.8169616806068858,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004938670625207317,
      "loss": 5.1814,
      "mean_token_accuracy": 0.185356043279171,
      "num_tokens": 18234894.0,
      "step": 10500
    },
    {
      "entropy": 5.494987106323242,
      "epoch": 0.8173507099786034,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004938606339490406,
      "loss": 5.2284,
      "mean_token_accuracy": 0.18029185831546785,
      "num_tokens": 18243768.0,
      "step": 10505
    },
    {
      "entropy": 5.5527674674987795,
      "epoch": 0.817739739350321,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004938542020564877,
      "loss": 5.2481,
      "mean_token_accuracy": 0.18838313817977906,
      "num_tokens": 18252364.0,
      "step": 10510
    },
    {
      "entropy": 5.506614875793457,
      "epoch": 0.8181287687220385,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004938477668431706,
      "loss": 5.1996,
      "mean_token_accuracy": 0.18282681107521057,
      "num_tokens": 18260997.0,
      "step": 10515
    },
    {
      "entropy": 5.645081901550293,
      "epoch": 0.8185177980937561,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004938413283091871,
      "loss": 5.3844,
      "mean_token_accuracy": 0.17370312213897704,
      "num_tokens": 18270171.0,
      "step": 10520
    },
    {
      "entropy": 5.601317119598389,
      "epoch": 0.8189068274654736,
      "grad_norm": 1.0,
      "learning_rate": 0.0004938348864546347,
      "loss": 5.2493,
      "mean_token_accuracy": 0.18100894391536712,
      "num_tokens": 18278772.0,
      "step": 10525
    },
    {
      "entropy": 5.4411101818084715,
      "epoch": 0.8192958568371912,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004938284412796111,
      "loss": 5.077,
      "mean_token_accuracy": 0.1939646065235138,
      "num_tokens": 18287305.0,
      "step": 10530
    },
    {
      "entropy": 5.442954015731812,
      "epoch": 0.8196848862089088,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004938219927842144,
      "loss": 5.315,
      "mean_token_accuracy": 0.17508912235498428,
      "num_tokens": 18296532.0,
      "step": 10535
    },
    {
      "entropy": 5.5125280857086185,
      "epoch": 0.8200739155806264,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004938155409685422,
      "loss": 5.1683,
      "mean_token_accuracy": 0.19022439122200013,
      "num_tokens": 18304612.0,
      "step": 10540
    },
    {
      "entropy": 5.530983352661133,
      "epoch": 0.820462944952344,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004938090858326923,
      "loss": 5.2214,
      "mean_token_accuracy": 0.18764003366231918,
      "num_tokens": 18312970.0,
      "step": 10545
    },
    {
      "entropy": 5.546092939376831,
      "epoch": 0.8208519743240614,
      "grad_norm": 1.078125,
      "learning_rate": 0.000493802627376763,
      "loss": 5.254,
      "mean_token_accuracy": 0.17960648536682128,
      "num_tokens": 18321073.0,
      "step": 10550
    },
    {
      "entropy": 5.52388072013855,
      "epoch": 0.821241003695779,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004937961656008518,
      "loss": 5.2408,
      "mean_token_accuracy": 0.18145296424627305,
      "num_tokens": 18329960.0,
      "step": 10555
    },
    {
      "entropy": 5.472028017044067,
      "epoch": 0.8216300330674966,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004937897005050573,
      "loss": 5.2062,
      "mean_token_accuracy": 0.18540115654468536,
      "num_tokens": 18337935.0,
      "step": 10560
    },
    {
      "entropy": 5.4654826641082765,
      "epoch": 0.8220190624392142,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004937832320894772,
      "loss": 5.202,
      "mean_token_accuracy": 0.18395290225744249,
      "num_tokens": 18347445.0,
      "step": 10565
    },
    {
      "entropy": 5.495949459075928,
      "epoch": 0.8224080918109318,
      "grad_norm": 1.0,
      "learning_rate": 0.0004937767603542098,
      "loss": 5.2212,
      "mean_token_accuracy": 0.1867908790707588,
      "num_tokens": 18355817.0,
      "step": 10570
    },
    {
      "entropy": 5.529239082336426,
      "epoch": 0.8227971211826492,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004937702852993534,
      "loss": 5.2105,
      "mean_token_accuracy": 0.18544757813215257,
      "num_tokens": 18363478.0,
      "step": 10575
    },
    {
      "entropy": 5.487400341033935,
      "epoch": 0.8231861505543668,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000493763806925006,
      "loss": 5.1745,
      "mean_token_accuracy": 0.19082464426755905,
      "num_tokens": 18371788.0,
      "step": 10580
    },
    {
      "entropy": 5.47348108291626,
      "epoch": 0.8235751799260844,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004937573252312661,
      "loss": 5.2144,
      "mean_token_accuracy": 0.18611689656972885,
      "num_tokens": 18380361.0,
      "step": 10585
    },
    {
      "entropy": 5.547488355636597,
      "epoch": 0.823964209297802,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000493750840218232,
      "loss": 5.2959,
      "mean_token_accuracy": 0.18320767879486083,
      "num_tokens": 18389396.0,
      "step": 10590
    },
    {
      "entropy": 5.516589164733887,
      "epoch": 0.8243532386695196,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004937443518860021,
      "loss": 5.338,
      "mean_token_accuracy": 0.17054078578948975,
      "num_tokens": 18398517.0,
      "step": 10595
    },
    {
      "entropy": 5.566125202178955,
      "epoch": 0.8247422680412371,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004937378602346747,
      "loss": 5.2637,
      "mean_token_accuracy": 0.18514072448015212,
      "num_tokens": 18406620.0,
      "step": 10600
    },
    {
      "entropy": 5.5272440910339355,
      "epoch": 0.8251312974129547,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004937313652643485,
      "loss": 5.2954,
      "mean_token_accuracy": 0.1755000039935112,
      "num_tokens": 18414849.0,
      "step": 10605
    },
    {
      "entropy": 5.477118921279907,
      "epoch": 0.8255203267846722,
      "grad_norm": 0.953125,
      "learning_rate": 0.000493724866975122,
      "loss": 5.2523,
      "mean_token_accuracy": 0.17994569540023803,
      "num_tokens": 18423675.0,
      "step": 10610
    },
    {
      "entropy": 5.596554327011108,
      "epoch": 0.8259093561563898,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004937183653670937,
      "loss": 5.3086,
      "mean_token_accuracy": 0.18348225951194763,
      "num_tokens": 18432848.0,
      "step": 10615
    },
    {
      "entropy": 5.5981764793396,
      "epoch": 0.8262983855281074,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004937118604403623,
      "loss": 5.3491,
      "mean_token_accuracy": 0.17721238136291503,
      "num_tokens": 18441210.0,
      "step": 10620
    },
    {
      "entropy": 5.546292304992676,
      "epoch": 0.8266874148998249,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004937053521950266,
      "loss": 5.2601,
      "mean_token_accuracy": 0.18121777325868607,
      "num_tokens": 18450084.0,
      "step": 10625
    },
    {
      "entropy": 5.445037317276001,
      "epoch": 0.8270764442715425,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004936988406311854,
      "loss": 5.1437,
      "mean_token_accuracy": 0.18346010893583298,
      "num_tokens": 18458329.0,
      "step": 10630
    },
    {
      "entropy": 5.491110372543335,
      "epoch": 0.8274654736432601,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004936923257489373,
      "loss": 5.2489,
      "mean_token_accuracy": 0.17862151563167572,
      "num_tokens": 18467518.0,
      "step": 10635
    },
    {
      "entropy": 5.496564865112305,
      "epoch": 0.8278545030149776,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004936858075483811,
      "loss": 5.1444,
      "mean_token_accuracy": 0.20085552483797073,
      "num_tokens": 18476051.0,
      "step": 10640
    },
    {
      "entropy": 5.522183990478515,
      "epoch": 0.8282435323866952,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000493679286029616,
      "loss": 5.2401,
      "mean_token_accuracy": 0.18506155610084535,
      "num_tokens": 18484492.0,
      "step": 10645
    },
    {
      "entropy": 5.509636116027832,
      "epoch": 0.8286325617584127,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004936727611927407,
      "loss": 5.2953,
      "mean_token_accuracy": 0.17371538430452346,
      "num_tokens": 18493759.0,
      "step": 10650
    },
    {
      "entropy": 5.535672807693482,
      "epoch": 0.8290215911301303,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004936662330378546,
      "loss": 5.157,
      "mean_token_accuracy": 0.18664028197526933,
      "num_tokens": 18501750.0,
      "step": 10655
    },
    {
      "entropy": 5.44293737411499,
      "epoch": 0.8294106205018479,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004936597015650561,
      "loss": 5.1654,
      "mean_token_accuracy": 0.18426862359046936,
      "num_tokens": 18511003.0,
      "step": 10660
    },
    {
      "entropy": 5.467996454238891,
      "epoch": 0.8297996498735655,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004936531667744448,
      "loss": 5.2194,
      "mean_token_accuracy": 0.1902633860707283,
      "num_tokens": 18519544.0,
      "step": 10665
    },
    {
      "entropy": 5.618622493743897,
      "epoch": 0.8301886792452831,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004936466286661197,
      "loss": 5.3078,
      "mean_token_accuracy": 0.17962168753147126,
      "num_tokens": 18528766.0,
      "step": 10670
    },
    {
      "entropy": 5.542390441894531,
      "epoch": 0.8305777086170005,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00049364008724018,
      "loss": 5.3098,
      "mean_token_accuracy": 0.17553483247756957,
      "num_tokens": 18538243.0,
      "step": 10675
    },
    {
      "entropy": 5.533434534072876,
      "epoch": 0.8309667379887181,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000493633542496725,
      "loss": 5.3203,
      "mean_token_accuracy": 0.1801365315914154,
      "num_tokens": 18547028.0,
      "step": 10680
    },
    {
      "entropy": 5.57064642906189,
      "epoch": 0.8313557673604357,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004936269944358539,
      "loss": 5.3339,
      "mean_token_accuracy": 0.1779732421040535,
      "num_tokens": 18555821.0,
      "step": 10685
    },
    {
      "entropy": 5.508390140533447,
      "epoch": 0.8317447967321533,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004936204430576664,
      "loss": 5.2912,
      "mean_token_accuracy": 0.1796719953417778,
      "num_tokens": 18564213.0,
      "step": 10690
    },
    {
      "entropy": 5.540386009216308,
      "epoch": 0.8321338261038709,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004936138883622614,
      "loss": 5.2677,
      "mean_token_accuracy": 0.17724219411611558,
      "num_tokens": 18573132.0,
      "step": 10695
    },
    {
      "entropy": 5.536636114120483,
      "epoch": 0.8325228554755884,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004936073303497387,
      "loss": 5.1756,
      "mean_token_accuracy": 0.18300071507692336,
      "num_tokens": 18581131.0,
      "step": 10700
    },
    {
      "entropy": 5.5129159450531,
      "epoch": 0.8329118848473059,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004936007690201976,
      "loss": 5.2655,
      "mean_token_accuracy": 0.18008433431386947,
      "num_tokens": 18589602.0,
      "step": 10705
    },
    {
      "entropy": 5.493002462387085,
      "epoch": 0.8333009142190235,
      "grad_norm": 1.0,
      "learning_rate": 0.000493594204373738,
      "loss": 5.251,
      "mean_token_accuracy": 0.17954349666833877,
      "num_tokens": 18598263.0,
      "step": 10710
    },
    {
      "entropy": 5.523498249053955,
      "epoch": 0.8336899435907411,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004935876364104591,
      "loss": 5.1447,
      "mean_token_accuracy": 0.19066095352172852,
      "num_tokens": 18606432.0,
      "step": 10715
    },
    {
      "entropy": 5.529076766967774,
      "epoch": 0.8340789729624587,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004935810651304608,
      "loss": 5.2798,
      "mean_token_accuracy": 0.18001518547534942,
      "num_tokens": 18614769.0,
      "step": 10720
    },
    {
      "entropy": 5.5900355815887455,
      "epoch": 0.8344680023341763,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004935744905338427,
      "loss": 5.2507,
      "mean_token_accuracy": 0.18012447506189347,
      "num_tokens": 18623505.0,
      "step": 10725
    },
    {
      "entropy": 5.5634377002716064,
      "epoch": 0.8348570317058938,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004935679126207046,
      "loss": 5.2219,
      "mean_token_accuracy": 0.19116541892290115,
      "num_tokens": 18632072.0,
      "step": 10730
    },
    {
      "entropy": 5.572908544540406,
      "epoch": 0.8352460610776113,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004935613313911463,
      "loss": 5.2942,
      "mean_token_accuracy": 0.17705805897712706,
      "num_tokens": 18640439.0,
      "step": 10735
    },
    {
      "entropy": 5.502774429321289,
      "epoch": 0.8356350904493289,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004935547468452678,
      "loss": 5.288,
      "mean_token_accuracy": 0.170834019780159,
      "num_tokens": 18649554.0,
      "step": 10740
    },
    {
      "entropy": 5.564027976989746,
      "epoch": 0.8360241198210465,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004935481589831688,
      "loss": 5.1641,
      "mean_token_accuracy": 0.185481259226799,
      "num_tokens": 18657918.0,
      "step": 10745
    },
    {
      "entropy": 5.596664237976074,
      "epoch": 0.8364131491927641,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004935415678049492,
      "loss": 5.2685,
      "mean_token_accuracy": 0.175859534740448,
      "num_tokens": 18666735.0,
      "step": 10750
    },
    {
      "entropy": 5.491724109649658,
      "epoch": 0.8368021785644816,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004935349733107094,
      "loss": 5.1587,
      "mean_token_accuracy": 0.18574313968420028,
      "num_tokens": 18674977.0,
      "step": 10755
    },
    {
      "entropy": 5.493093824386596,
      "epoch": 0.8371912079361992,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000493528375500549,
      "loss": 5.3008,
      "mean_token_accuracy": 0.1820133298635483,
      "num_tokens": 18683624.0,
      "step": 10760
    },
    {
      "entropy": 5.4717155456542965,
      "epoch": 0.8375802373079168,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004935217743745685,
      "loss": 5.2006,
      "mean_token_accuracy": 0.18414881825447083,
      "num_tokens": 18692183.0,
      "step": 10765
    },
    {
      "entropy": 5.578032493591309,
      "epoch": 0.8379692666796343,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004935151699328677,
      "loss": 5.2885,
      "mean_token_accuracy": 0.17705097943544387,
      "num_tokens": 18700989.0,
      "step": 10770
    },
    {
      "entropy": 5.522564506530761,
      "epoch": 0.8383582960513519,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004935085621755471,
      "loss": 5.1524,
      "mean_token_accuracy": 0.18496323525905609,
      "num_tokens": 18709126.0,
      "step": 10775
    },
    {
      "entropy": 5.550716781616211,
      "epoch": 0.8387473254230694,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000493501951102707,
      "loss": 5.2424,
      "mean_token_accuracy": 0.1835919812321663,
      "num_tokens": 18717378.0,
      "step": 10780
    },
    {
      "entropy": 5.585535144805908,
      "epoch": 0.839136354794787,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004934953367144474,
      "loss": 5.3144,
      "mean_token_accuracy": 0.17845938354730606,
      "num_tokens": 18725694.0,
      "step": 10785
    },
    {
      "entropy": 5.499988985061646,
      "epoch": 0.8395253841665046,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004934887190108688,
      "loss": 5.2263,
      "mean_token_accuracy": 0.1819104865193367,
      "num_tokens": 18734737.0,
      "step": 10790
    },
    {
      "entropy": 5.662831974029541,
      "epoch": 0.8399144135382222,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004934820979920719,
      "loss": 5.4445,
      "mean_token_accuracy": 0.17079488784074784,
      "num_tokens": 18743895.0,
      "step": 10795
    },
    {
      "entropy": 5.5817817687988285,
      "epoch": 0.8403034429099397,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004934754736581567,
      "loss": 5.301,
      "mean_token_accuracy": 0.18126434534788133,
      "num_tokens": 18753105.0,
      "step": 10800
    },
    {
      "entropy": 5.4522534847259525,
      "epoch": 0.8406924722816572,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000493468846009224,
      "loss": 5.1705,
      "mean_token_accuracy": 0.17899207174777984,
      "num_tokens": 18761593.0,
      "step": 10805
    },
    {
      "entropy": 5.564798879623413,
      "epoch": 0.8410815016533748,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004934622150453742,
      "loss": 5.2167,
      "mean_token_accuracy": 0.18947377502918245,
      "num_tokens": 18770105.0,
      "step": 10810
    },
    {
      "entropy": 5.557617330551148,
      "epoch": 0.8414705310250924,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000493455580766708,
      "loss": 5.2719,
      "mean_token_accuracy": 0.1804944857954979,
      "num_tokens": 18778429.0,
      "step": 10815
    },
    {
      "entropy": 5.620841217041016,
      "epoch": 0.84185956039681,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004934489431733262,
      "loss": 5.2903,
      "mean_token_accuracy": 0.1760850504040718,
      "num_tokens": 18787072.0,
      "step": 10820
    },
    {
      "entropy": 5.515456581115723,
      "epoch": 0.8422485897685276,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004934423022653293,
      "loss": 5.1842,
      "mean_token_accuracy": 0.18739484548568724,
      "num_tokens": 18795628.0,
      "step": 10825
    },
    {
      "entropy": 5.538226652145386,
      "epoch": 0.842637619140245,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004934356580428182,
      "loss": 5.2647,
      "mean_token_accuracy": 0.1825709968805313,
      "num_tokens": 18804313.0,
      "step": 10830
    },
    {
      "entropy": 5.565060710906982,
      "epoch": 0.8430266485119626,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004934290105058937,
      "loss": 5.3069,
      "mean_token_accuracy": 0.1762733578681946,
      "num_tokens": 18813170.0,
      "step": 10835
    },
    {
      "entropy": 5.507975435256958,
      "epoch": 0.8434156778836802,
      "grad_norm": 1.0,
      "learning_rate": 0.0004934223596546565,
      "loss": 5.2387,
      "mean_token_accuracy": 0.18738016188144685,
      "num_tokens": 18823009.0,
      "step": 10840
    },
    {
      "entropy": 5.527263593673706,
      "epoch": 0.8438047072553978,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004934157054892077,
      "loss": 5.2818,
      "mean_token_accuracy": 0.1757340595126152,
      "num_tokens": 18831272.0,
      "step": 10845
    },
    {
      "entropy": 5.491315269470215,
      "epoch": 0.8441937366271154,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004934090480096482,
      "loss": 5.2828,
      "mean_token_accuracy": 0.17550265192985534,
      "num_tokens": 18839481.0,
      "step": 10850
    },
    {
      "entropy": 5.471037864685059,
      "epoch": 0.8445827659988329,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004934023872160791,
      "loss": 5.2412,
      "mean_token_accuracy": 0.1860124483704567,
      "num_tokens": 18848586.0,
      "step": 10855
    },
    {
      "entropy": 5.543521690368652,
      "epoch": 0.8449717953705504,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004933957231086014,
      "loss": 5.1445,
      "mean_token_accuracy": 0.1888479009270668,
      "num_tokens": 18856948.0,
      "step": 10860
    },
    {
      "entropy": 5.568339395523071,
      "epoch": 0.845360824742268,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004933890556873161,
      "loss": 5.2086,
      "mean_token_accuracy": 0.18441223353147507,
      "num_tokens": 18864741.0,
      "step": 10865
    },
    {
      "entropy": 5.515665769577026,
      "epoch": 0.8457498541139856,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004933823849523246,
      "loss": 5.2568,
      "mean_token_accuracy": 0.17967778146266938,
      "num_tokens": 18873373.0,
      "step": 10870
    },
    {
      "entropy": 5.546140813827515,
      "epoch": 0.8461388834857032,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004933757109037279,
      "loss": 5.1418,
      "mean_token_accuracy": 0.18498702794313432,
      "num_tokens": 18881590.0,
      "step": 10875
    },
    {
      "entropy": 5.533450603485107,
      "epoch": 0.8465279128574207,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004933690335416274,
      "loss": 5.2555,
      "mean_token_accuracy": 0.181799778342247,
      "num_tokens": 18889838.0,
      "step": 10880
    },
    {
      "entropy": 5.473286485671997,
      "epoch": 0.8469169422291383,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004933623528661245,
      "loss": 5.1495,
      "mean_token_accuracy": 0.19025446474552155,
      "num_tokens": 18898577.0,
      "step": 10885
    },
    {
      "entropy": 5.575493097305298,
      "epoch": 0.8473059716008559,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004933556688773203,
      "loss": 5.3099,
      "mean_token_accuracy": 0.18210299760103227,
      "num_tokens": 18907309.0,
      "step": 10890
    },
    {
      "entropy": 5.533792543411255,
      "epoch": 0.8476950009725734,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004933489815753165,
      "loss": 5.2443,
      "mean_token_accuracy": 0.17948850691318513,
      "num_tokens": 18915956.0,
      "step": 10895
    },
    {
      "entropy": 5.5209626197814945,
      "epoch": 0.848084030344291,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004933422909602143,
      "loss": 5.3516,
      "mean_token_accuracy": 0.17663799971342087,
      "num_tokens": 18924603.0,
      "step": 10900
    },
    {
      "entropy": 5.538794231414795,
      "epoch": 0.8484730597160085,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004933355970321155,
      "loss": 5.2057,
      "mean_token_accuracy": 0.1915137141942978,
      "num_tokens": 18933231.0,
      "step": 10905
    },
    {
      "entropy": 5.585106420516968,
      "epoch": 0.8488620890877261,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004933288997911215,
      "loss": 5.3243,
      "mean_token_accuracy": 0.17297962605953215,
      "num_tokens": 18942172.0,
      "step": 10910
    },
    {
      "entropy": 5.548247623443603,
      "epoch": 0.8492511184594437,
      "grad_norm": 1.0,
      "learning_rate": 0.0004933221992373338,
      "loss": 5.2869,
      "mean_token_accuracy": 0.18840184360742568,
      "num_tokens": 18951440.0,
      "step": 10915
    },
    {
      "entropy": 5.5250507354736325,
      "epoch": 0.8496401478311613,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004933154953708544,
      "loss": 5.2341,
      "mean_token_accuracy": 0.1767898440361023,
      "num_tokens": 18959797.0,
      "step": 10920
    },
    {
      "entropy": 5.515337753295898,
      "epoch": 0.8500291772028788,
      "grad_norm": 1.0,
      "learning_rate": 0.0004933087881917848,
      "loss": 5.2387,
      "mean_token_accuracy": 0.18071039766073227,
      "num_tokens": 18968811.0,
      "step": 10925
    },
    {
      "entropy": 5.593693017959595,
      "epoch": 0.8504182065745964,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004933020777002268,
      "loss": 5.3818,
      "mean_token_accuracy": 0.17162843644618989,
      "num_tokens": 18977467.0,
      "step": 10930
    },
    {
      "entropy": 5.594732856750488,
      "epoch": 0.8508072359463139,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004932953638962823,
      "loss": 5.2563,
      "mean_token_accuracy": 0.17883533090353013,
      "num_tokens": 18985760.0,
      "step": 10935
    },
    {
      "entropy": 5.656961154937744,
      "epoch": 0.8511962653180315,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000493288646780053,
      "loss": 5.3041,
      "mean_token_accuracy": 0.17105769217014313,
      "num_tokens": 18993812.0,
      "step": 10940
    },
    {
      "entropy": 5.539013528823853,
      "epoch": 0.8515852946897491,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004932819263516409,
      "loss": 5.2987,
      "mean_token_accuracy": 0.17625754177570344,
      "num_tokens": 19003056.0,
      "step": 10945
    },
    {
      "entropy": 5.589330673217773,
      "epoch": 0.8519743240614667,
      "grad_norm": 1.0,
      "learning_rate": 0.0004932752026111481,
      "loss": 5.3051,
      "mean_token_accuracy": 0.18107609450817108,
      "num_tokens": 19011533.0,
      "step": 10950
    },
    {
      "entropy": 5.53038182258606,
      "epoch": 0.8523633534331843,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004932684755586765,
      "loss": 5.2282,
      "mean_token_accuracy": 0.18666124641895293,
      "num_tokens": 19020773.0,
      "step": 10955
    },
    {
      "entropy": 5.485508680343628,
      "epoch": 0.8527523828049017,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004932617451943282,
      "loss": 5.2395,
      "mean_token_accuracy": 0.17614906430244445,
      "num_tokens": 19029809.0,
      "step": 10960
    },
    {
      "entropy": 5.512433767318726,
      "epoch": 0.8531414121766193,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004932550115182053,
      "loss": 5.1599,
      "mean_token_accuracy": 0.18353773057460784,
      "num_tokens": 19038875.0,
      "step": 10965
    },
    {
      "entropy": 5.589402961730957,
      "epoch": 0.8535304415483369,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00049324827453041,
      "loss": 5.2753,
      "mean_token_accuracy": 0.18687373548746108,
      "num_tokens": 19047273.0,
      "step": 10970
    },
    {
      "entropy": 5.432324361801148,
      "epoch": 0.8539194709200545,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004932415342310446,
      "loss": 5.2033,
      "mean_token_accuracy": 0.18860424757003785,
      "num_tokens": 19056309.0,
      "step": 10975
    },
    {
      "entropy": 5.496869325637817,
      "epoch": 0.8543085002917721,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004932347906202111,
      "loss": 5.1825,
      "mean_token_accuracy": 0.18541284650564194,
      "num_tokens": 19064487.0,
      "step": 10980
    },
    {
      "entropy": 5.57986364364624,
      "epoch": 0.8546975296634896,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000493228043698012,
      "loss": 5.4094,
      "mean_token_accuracy": 0.1805096909403801,
      "num_tokens": 19073431.0,
      "step": 10985
    },
    {
      "entropy": 5.507370710372925,
      "epoch": 0.8550865590352071,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004932212934645498,
      "loss": 5.2654,
      "mean_token_accuracy": 0.17887515276670457,
      "num_tokens": 19081648.0,
      "step": 10990
    },
    {
      "entropy": 5.454296445846557,
      "epoch": 0.8554755884069247,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004932145399199268,
      "loss": 5.1947,
      "mean_token_accuracy": 0.18084492087364196,
      "num_tokens": 19090581.0,
      "step": 10995
    },
    {
      "entropy": 5.616622686386108,
      "epoch": 0.8558646177786423,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004932077830642455,
      "loss": 5.2726,
      "mean_token_accuracy": 0.17999816983938216,
      "num_tokens": 19098924.0,
      "step": 11000
    },
    {
      "entropy": 5.570590496063232,
      "epoch": 0.8562536471503599,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004932010228976084,
      "loss": 5.2707,
      "mean_token_accuracy": 0.1806684672832489,
      "num_tokens": 19108267.0,
      "step": 11005
    },
    {
      "entropy": 5.55898003578186,
      "epoch": 0.8566426765220774,
      "grad_norm": 0.984375,
      "learning_rate": 0.000493194259420118,
      "loss": 5.2997,
      "mean_token_accuracy": 0.17755764573812485,
      "num_tokens": 19117158.0,
      "step": 11010
    },
    {
      "entropy": 5.51257290840149,
      "epoch": 0.857031705893795,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000493187492631877,
      "loss": 5.2667,
      "mean_token_accuracy": 0.18469720780849458,
      "num_tokens": 19126013.0,
      "step": 11015
    },
    {
      "entropy": 5.5020078182220455,
      "epoch": 0.8574207352655125,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004931807225329882,
      "loss": 5.1997,
      "mean_token_accuracy": 0.18354545086622237,
      "num_tokens": 19134655.0,
      "step": 11020
    },
    {
      "entropy": 5.490265417098999,
      "epoch": 0.8578097646372301,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004931739491235541,
      "loss": 5.1058,
      "mean_token_accuracy": 0.18068982511758805,
      "num_tokens": 19143311.0,
      "step": 11025
    },
    {
      "entropy": 5.537204837799072,
      "epoch": 0.8581987940089477,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004931671724036777,
      "loss": 5.2357,
      "mean_token_accuracy": 0.18675446808338164,
      "num_tokens": 19152422.0,
      "step": 11030
    },
    {
      "entropy": 5.5695281505584715,
      "epoch": 0.8585878233806652,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004931603923734616,
      "loss": 5.3506,
      "mean_token_accuracy": 0.17644894868135452,
      "num_tokens": 19161510.0,
      "step": 11035
    },
    {
      "entropy": 5.494791269302368,
      "epoch": 0.8589768527523828,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004931536090330088,
      "loss": 5.1639,
      "mean_token_accuracy": 0.18735045790672303,
      "num_tokens": 19170097.0,
      "step": 11040
    },
    {
      "entropy": 5.532437753677368,
      "epoch": 0.8593658821241004,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004931468223824222,
      "loss": 5.2145,
      "mean_token_accuracy": 0.1859789341688156,
      "num_tokens": 19179131.0,
      "step": 11045
    },
    {
      "entropy": 5.446110439300537,
      "epoch": 0.859754911495818,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004931400324218048,
      "loss": 5.1187,
      "mean_token_accuracy": 0.18858676105737687,
      "num_tokens": 19187289.0,
      "step": 11050
    },
    {
      "entropy": 5.454470157623291,
      "epoch": 0.8601439408675355,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004931332391512597,
      "loss": 5.2414,
      "mean_token_accuracy": 0.18397646844387056,
      "num_tokens": 19195746.0,
      "step": 11055
    },
    {
      "entropy": 5.605712175369263,
      "epoch": 0.860532970239253,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004931264425708897,
      "loss": 5.2743,
      "mean_token_accuracy": 0.17812152951955795,
      "num_tokens": 19204683.0,
      "step": 11060
    },
    {
      "entropy": 5.513314199447632,
      "epoch": 0.8609219996109706,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004931196426807983,
      "loss": 5.174,
      "mean_token_accuracy": 0.17793773263692855,
      "num_tokens": 19213405.0,
      "step": 11065
    },
    {
      "entropy": 5.51433916091919,
      "epoch": 0.8613110289826882,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004931128394810884,
      "loss": 5.273,
      "mean_token_accuracy": 0.18461571037769317,
      "num_tokens": 19222717.0,
      "step": 11070
    },
    {
      "entropy": 5.579733848571777,
      "epoch": 0.8617000583544058,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004931060329718634,
      "loss": 5.2759,
      "mean_token_accuracy": 0.1818013533949852,
      "num_tokens": 19230688.0,
      "step": 11075
    },
    {
      "entropy": 5.585348081588745,
      "epoch": 0.8620890877261234,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004930992231532265,
      "loss": 5.3348,
      "mean_token_accuracy": 0.17010061889886857,
      "num_tokens": 19239192.0,
      "step": 11080
    },
    {
      "entropy": 5.5678917407989506,
      "epoch": 0.8624781170978408,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000493092410025281,
      "loss": 5.3115,
      "mean_token_accuracy": 0.1736423820257187,
      "num_tokens": 19247903.0,
      "step": 11085
    },
    {
      "entropy": 5.51790452003479,
      "epoch": 0.8628671464695584,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004930855935881302,
      "loss": 5.2599,
      "mean_token_accuracy": 0.18231950253248214,
      "num_tokens": 19256127.0,
      "step": 11090
    },
    {
      "entropy": 5.572448444366455,
      "epoch": 0.863256175841276,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004930787738418777,
      "loss": 5.2341,
      "mean_token_accuracy": 0.1836707577109337,
      "num_tokens": 19264742.0,
      "step": 11095
    },
    {
      "entropy": 5.520915460586548,
      "epoch": 0.8636452052129936,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004930719507866269,
      "loss": 5.1915,
      "mean_token_accuracy": 0.18350018411874772,
      "num_tokens": 19273618.0,
      "step": 11100
    },
    {
      "entropy": 5.5309460163116455,
      "epoch": 0.8640342345847112,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004930651244224814,
      "loss": 5.249,
      "mean_token_accuracy": 0.1856066107749939,
      "num_tokens": 19281792.0,
      "step": 11105
    },
    {
      "entropy": 5.456627464294433,
      "epoch": 0.8644232639564287,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004930582947495448,
      "loss": 5.2133,
      "mean_token_accuracy": 0.18572230637073517,
      "num_tokens": 19289738.0,
      "step": 11110
    },
    {
      "entropy": 5.546131134033203,
      "epoch": 0.8648122933281462,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004930514617679206,
      "loss": 5.1973,
      "mean_token_accuracy": 0.18326290994882583,
      "num_tokens": 19298711.0,
      "step": 11115
    },
    {
      "entropy": 5.607199335098267,
      "epoch": 0.8652013226998638,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004930446254777125,
      "loss": 5.3056,
      "mean_token_accuracy": 0.18197255283594133,
      "num_tokens": 19307435.0,
      "step": 11120
    },
    {
      "entropy": 5.496180534362793,
      "epoch": 0.8655903520715814,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004930377858790242,
      "loss": 5.1779,
      "mean_token_accuracy": 0.18431518375873565,
      "num_tokens": 19315327.0,
      "step": 11125
    },
    {
      "entropy": 5.487762784957885,
      "epoch": 0.865979381443299,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004930309429719595,
      "loss": 5.1428,
      "mean_token_accuracy": 0.18945733606815338,
      "num_tokens": 19323261.0,
      "step": 11130
    },
    {
      "entropy": 5.422307300567627,
      "epoch": 0.8663684108150166,
      "grad_norm": 1.0,
      "learning_rate": 0.0004930240967566224,
      "loss": 5.1031,
      "mean_token_accuracy": 0.18810771852731706,
      "num_tokens": 19332297.0,
      "step": 11135
    },
    {
      "entropy": 5.477681684494018,
      "epoch": 0.8667574401867341,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004930172472331167,
      "loss": 5.1431,
      "mean_token_accuracy": 0.1872953414916992,
      "num_tokens": 19340848.0,
      "step": 11140
    },
    {
      "entropy": 5.453296566009522,
      "epoch": 0.8671464695584516,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004930103944015463,
      "loss": 5.1875,
      "mean_token_accuracy": 0.18128651678562163,
      "num_tokens": 19349072.0,
      "step": 11145
    },
    {
      "entropy": 5.544670248031617,
      "epoch": 0.8675354989301692,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004930035382620149,
      "loss": 5.2429,
      "mean_token_accuracy": 0.18343978226184846,
      "num_tokens": 19357723.0,
      "step": 11150
    },
    {
      "entropy": 5.475110769271851,
      "epoch": 0.8679245283018868,
      "grad_norm": 1.046875,
      "learning_rate": 0.000492996678814627,
      "loss": 5.1915,
      "mean_token_accuracy": 0.18486727327108382,
      "num_tokens": 19365925.0,
      "step": 11155
    },
    {
      "entropy": 5.489379644393921,
      "epoch": 0.8683135576736044,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004929898160594865,
      "loss": 5.2367,
      "mean_token_accuracy": 0.18154717832803727,
      "num_tokens": 19374190.0,
      "step": 11160
    },
    {
      "entropy": 5.48993148803711,
      "epoch": 0.8687025870453219,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004929829499966974,
      "loss": 5.2267,
      "mean_token_accuracy": 0.182435542345047,
      "num_tokens": 19382375.0,
      "step": 11165
    },
    {
      "entropy": 5.602983617782593,
      "epoch": 0.8690916164170395,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004929760806263641,
      "loss": 5.2678,
      "mean_token_accuracy": 0.18531643301248552,
      "num_tokens": 19390191.0,
      "step": 11170
    },
    {
      "entropy": 5.4475806713104244,
      "epoch": 0.8694806457887571,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004929692079485906,
      "loss": 5.0713,
      "mean_token_accuracy": 0.19421324729919434,
      "num_tokens": 19399401.0,
      "step": 11175
    },
    {
      "entropy": 5.430244255065918,
      "epoch": 0.8698696751604746,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004929623319634814,
      "loss": 5.1759,
      "mean_token_accuracy": 0.19058845788240433,
      "num_tokens": 19407956.0,
      "step": 11180
    },
    {
      "entropy": 5.526352882385254,
      "epoch": 0.8702587045321922,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004929554526711407,
      "loss": 5.3368,
      "mean_token_accuracy": 0.17986630499362946,
      "num_tokens": 19416432.0,
      "step": 11185
    },
    {
      "entropy": 5.511863040924072,
      "epoch": 0.8706477339039097,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004929485700716729,
      "loss": 5.2108,
      "mean_token_accuracy": 0.1863880857825279,
      "num_tokens": 19425208.0,
      "step": 11190
    },
    {
      "entropy": 5.574793672561645,
      "epoch": 0.8710367632756273,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004929416841651824,
      "loss": 5.3227,
      "mean_token_accuracy": 0.18036556839942933,
      "num_tokens": 19434169.0,
      "step": 11195
    },
    {
      "entropy": 5.486089563369751,
      "epoch": 0.8714257926473449,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004929347949517739,
      "loss": 5.0737,
      "mean_token_accuracy": 0.19685696959495544,
      "num_tokens": 19443484.0,
      "step": 11200
    },
    {
      "entropy": 5.472464084625244,
      "epoch": 0.8718148220190625,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004929279024315516,
      "loss": 5.2381,
      "mean_token_accuracy": 0.18787862807512284,
      "num_tokens": 19452193.0,
      "step": 11205
    },
    {
      "entropy": 5.601831340789795,
      "epoch": 0.87220385139078,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004929210066046204,
      "loss": 5.3652,
      "mean_token_accuracy": 0.1726883515715599,
      "num_tokens": 19460684.0,
      "step": 11210
    },
    {
      "entropy": 5.581402015686035,
      "epoch": 0.8725928807624975,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004929141074710847,
      "loss": 5.212,
      "mean_token_accuracy": 0.18295029252767564,
      "num_tokens": 19469206.0,
      "step": 11215
    },
    {
      "entropy": 5.519418382644654,
      "epoch": 0.8729819101342151,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004929072050310491,
      "loss": 5.254,
      "mean_token_accuracy": 0.18337550908327102,
      "num_tokens": 19477514.0,
      "step": 11220
    },
    {
      "entropy": 5.501196718215942,
      "epoch": 0.8733709395059327,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004929002992846188,
      "loss": 5.3069,
      "mean_token_accuracy": 0.17899565994739533,
      "num_tokens": 19486792.0,
      "step": 11225
    },
    {
      "entropy": 5.5645153522491455,
      "epoch": 0.8737599688776503,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004928933902318981,
      "loss": 5.2461,
      "mean_token_accuracy": 0.1856403633952141,
      "num_tokens": 19495708.0,
      "step": 11230
    },
    {
      "entropy": 5.615293264389038,
      "epoch": 0.8741489982493679,
      "grad_norm": 1.015625,
      "learning_rate": 0.000492886477872992,
      "loss": 5.3992,
      "mean_token_accuracy": 0.16661400347948074,
      "num_tokens": 19505063.0,
      "step": 11235
    },
    {
      "entropy": 5.614978265762329,
      "epoch": 0.8745380276210853,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004928795622080054,
      "loss": 5.3218,
      "mean_token_accuracy": 0.1738759309053421,
      "num_tokens": 19514641.0,
      "step": 11240
    },
    {
      "entropy": 5.523322868347168,
      "epoch": 0.8749270569928029,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004928726432370434,
      "loss": 5.1938,
      "mean_token_accuracy": 0.18731012642383577,
      "num_tokens": 19523136.0,
      "step": 11245
    },
    {
      "entropy": 5.451340961456299,
      "epoch": 0.8753160863645205,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004928657209602107,
      "loss": 5.0873,
      "mean_token_accuracy": 0.1941683381795883,
      "num_tokens": 19532258.0,
      "step": 11250
    },
    {
      "entropy": 5.503048276901245,
      "epoch": 0.8757051157362381,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004928587953776125,
      "loss": 5.3444,
      "mean_token_accuracy": 0.1800684452056885,
      "num_tokens": 19541235.0,
      "step": 11255
    },
    {
      "entropy": 5.548913192749024,
      "epoch": 0.8760941451079557,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004928518664893538,
      "loss": 5.2774,
      "mean_token_accuracy": 0.17916169613599778,
      "num_tokens": 19550259.0,
      "step": 11260
    },
    {
      "entropy": 5.585548830032349,
      "epoch": 0.8764831744796732,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004928449342955397,
      "loss": 5.2431,
      "mean_token_accuracy": 0.18700739741325378,
      "num_tokens": 19558500.0,
      "step": 11265
    },
    {
      "entropy": 5.512920761108399,
      "epoch": 0.8768722038513908,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004928379987962756,
      "loss": 5.1967,
      "mean_token_accuracy": 0.18584613054990767,
      "num_tokens": 19566562.0,
      "step": 11270
    },
    {
      "entropy": 5.526313972473145,
      "epoch": 0.8772612332231083,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004928310599916666,
      "loss": 5.2133,
      "mean_token_accuracy": 0.17974434047937393,
      "num_tokens": 19575219.0,
      "step": 11275
    },
    {
      "entropy": 5.464342451095581,
      "epoch": 0.8776502625948259,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004928241178818178,
      "loss": 5.1414,
      "mean_token_accuracy": 0.18692018687725068,
      "num_tokens": 19584311.0,
      "step": 11280
    },
    {
      "entropy": 5.4995660305023195,
      "epoch": 0.8780392919665435,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004928171724668349,
      "loss": 5.2838,
      "mean_token_accuracy": 0.17654700726270675,
      "num_tokens": 19593802.0,
      "step": 11285
    },
    {
      "entropy": 5.532325315475464,
      "epoch": 0.878428321338261,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004928102237468229,
      "loss": 5.2466,
      "mean_token_accuracy": 0.17566501945257187,
      "num_tokens": 19603046.0,
      "step": 11290
    },
    {
      "entropy": 5.461694622039795,
      "epoch": 0.8788173507099786,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004928032717218876,
      "loss": 5.132,
      "mean_token_accuracy": 0.18887731730937957,
      "num_tokens": 19611609.0,
      "step": 11295
    },
    {
      "entropy": 5.5138427734375,
      "epoch": 0.8792063800816962,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004927963163921343,
      "loss": 5.1547,
      "mean_token_accuracy": 0.18305641114711763,
      "num_tokens": 19619380.0,
      "step": 11300
    },
    {
      "entropy": 5.6087480068206785,
      "epoch": 0.8795954094534137,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004927893577576685,
      "loss": 5.3188,
      "mean_token_accuracy": 0.17588026374578475,
      "num_tokens": 19628336.0,
      "step": 11305
    },
    {
      "entropy": 5.510103845596314,
      "epoch": 0.8799844388251313,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004927823958185959,
      "loss": 5.299,
      "mean_token_accuracy": 0.18678108155727385,
      "num_tokens": 19637888.0,
      "step": 11310
    },
    {
      "entropy": 5.470029306411743,
      "epoch": 0.8803734681968488,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000492775430575022,
      "loss": 5.2031,
      "mean_token_accuracy": 0.18080607205629348,
      "num_tokens": 19646123.0,
      "step": 11315
    },
    {
      "entropy": 5.489652442932129,
      "epoch": 0.8807624975685664,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004927684620270527,
      "loss": 5.1907,
      "mean_token_accuracy": 0.19049484133720399,
      "num_tokens": 19655264.0,
      "step": 11320
    },
    {
      "entropy": 5.640956163406372,
      "epoch": 0.881151526940284,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004927614901747935,
      "loss": 5.3193,
      "mean_token_accuracy": 0.16887407898902893,
      "num_tokens": 19663594.0,
      "step": 11325
    },
    {
      "entropy": 5.586167192459106,
      "epoch": 0.8815405563120016,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004927545150183503,
      "loss": 5.2657,
      "mean_token_accuracy": 0.18213797956705094,
      "num_tokens": 19671515.0,
      "step": 11330
    },
    {
      "entropy": 5.4725054740905765,
      "epoch": 0.8819295856837192,
      "grad_norm": 0.953125,
      "learning_rate": 0.000492747536557829,
      "loss": 5.1311,
      "mean_token_accuracy": 0.18904923498630524,
      "num_tokens": 19680288.0,
      "step": 11335
    },
    {
      "entropy": 5.496434450149536,
      "epoch": 0.8823186150554367,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004927405547933353,
      "loss": 5.1564,
      "mean_token_accuracy": 0.18999323695898057,
      "num_tokens": 19688717.0,
      "step": 11340
    },
    {
      "entropy": 5.459884691238403,
      "epoch": 0.8827076444271542,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004927335697249753,
      "loss": 5.1051,
      "mean_token_accuracy": 0.1930097758769989,
      "num_tokens": 19697412.0,
      "step": 11345
    },
    {
      "entropy": 5.471190738677978,
      "epoch": 0.8830966737988718,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004927265813528549,
      "loss": 5.2494,
      "mean_token_accuracy": 0.18212933987379074,
      "num_tokens": 19706127.0,
      "step": 11350
    },
    {
      "entropy": 5.584218072891235,
      "epoch": 0.8834857031705894,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004927195896770804,
      "loss": 5.3177,
      "mean_token_accuracy": 0.17738562673330308,
      "num_tokens": 19715178.0,
      "step": 11355
    },
    {
      "entropy": 5.558844709396363,
      "epoch": 0.883874732542307,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004927125946977574,
      "loss": 5.2021,
      "mean_token_accuracy": 0.18528930693864823,
      "num_tokens": 19723814.0,
      "step": 11360
    },
    {
      "entropy": 5.561185836791992,
      "epoch": 0.8842637619140246,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004927055964149923,
      "loss": 5.3357,
      "mean_token_accuracy": 0.17682393044233322,
      "num_tokens": 19732153.0,
      "step": 11365
    },
    {
      "entropy": 5.539278507232666,
      "epoch": 0.884652791285742,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004926985948288914,
      "loss": 5.2552,
      "mean_token_accuracy": 0.18145778477191926,
      "num_tokens": 19740251.0,
      "step": 11370
    },
    {
      "entropy": 5.464604711532592,
      "epoch": 0.8850418206574596,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004926915899395608,
      "loss": 5.1615,
      "mean_token_accuracy": 0.19519666880369185,
      "num_tokens": 19748695.0,
      "step": 11375
    },
    {
      "entropy": 5.50255765914917,
      "epoch": 0.8854308500291772,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004926845817471068,
      "loss": 5.1838,
      "mean_token_accuracy": 0.17977945059537886,
      "num_tokens": 19757674.0,
      "step": 11380
    },
    {
      "entropy": 5.475516223907471,
      "epoch": 0.8858198794008948,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004926775702516358,
      "loss": 5.0696,
      "mean_token_accuracy": 0.19355673789978028,
      "num_tokens": 19765555.0,
      "step": 11385
    },
    {
      "entropy": 5.4411208152771,
      "epoch": 0.8862089087726124,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004926705554532541,
      "loss": 5.2219,
      "mean_token_accuracy": 0.18339011520147325,
      "num_tokens": 19773986.0,
      "step": 11390
    },
    {
      "entropy": 5.542670011520386,
      "epoch": 0.8865979381443299,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004926635373520682,
      "loss": 5.2387,
      "mean_token_accuracy": 0.18688921332359315,
      "num_tokens": 19782308.0,
      "step": 11395
    },
    {
      "entropy": 5.5365972995758055,
      "epoch": 0.8869869675160474,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004926565159481845,
      "loss": 5.2023,
      "mean_token_accuracy": 0.1816079929471016,
      "num_tokens": 19790447.0,
      "step": 11400
    },
    {
      "entropy": 5.527568054199219,
      "epoch": 0.887375996887765,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004926494912417097,
      "loss": 5.3175,
      "mean_token_accuracy": 0.18069407492876052,
      "num_tokens": 19798889.0,
      "step": 11405
    },
    {
      "entropy": 5.479697895050049,
      "epoch": 0.8877650262594826,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004926424632327503,
      "loss": 5.1311,
      "mean_token_accuracy": 0.1875539094209671,
      "num_tokens": 19807192.0,
      "step": 11410
    },
    {
      "entropy": 5.549045372009277,
      "epoch": 0.8881540556312002,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004926354319214129,
      "loss": 5.2673,
      "mean_token_accuracy": 0.1824059545993805,
      "num_tokens": 19815864.0,
      "step": 11415
    },
    {
      "entropy": 5.562523555755615,
      "epoch": 0.8885430850029177,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004926283973078041,
      "loss": 5.1861,
      "mean_token_accuracy": 0.18301030099391938,
      "num_tokens": 19823218.0,
      "step": 11420
    },
    {
      "entropy": 5.504983377456665,
      "epoch": 0.8889321143746353,
      "grad_norm": 1.0,
      "learning_rate": 0.0004926213593920309,
      "loss": 5.2383,
      "mean_token_accuracy": 0.18590783178806305,
      "num_tokens": 19832098.0,
      "step": 11425
    },
    {
      "entropy": 5.4997772693634035,
      "epoch": 0.8893211437463528,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004926143181742,
      "loss": 5.1764,
      "mean_token_accuracy": 0.1834106460213661,
      "num_tokens": 19840180.0,
      "step": 11430
    },
    {
      "entropy": 5.545825576782226,
      "epoch": 0.8897101731180704,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004926072736544181,
      "loss": 5.2287,
      "mean_token_accuracy": 0.18168114572763444,
      "num_tokens": 19849016.0,
      "step": 11435
    },
    {
      "entropy": 5.435892772674561,
      "epoch": 0.890099202489788,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004926002258327922,
      "loss": 5.1593,
      "mean_token_accuracy": 0.18454376459121705,
      "num_tokens": 19858044.0,
      "step": 11440
    },
    {
      "entropy": 5.459627485275268,
      "epoch": 0.8904882318615055,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004925931747094292,
      "loss": 5.0998,
      "mean_token_accuracy": 0.19237664043903352,
      "num_tokens": 19867557.0,
      "step": 11445
    },
    {
      "entropy": 5.5084045886993405,
      "epoch": 0.8908772612332231,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004925861202844361,
      "loss": 5.208,
      "mean_token_accuracy": 0.18517891466617584,
      "num_tokens": 19876009.0,
      "step": 11450
    },
    {
      "entropy": 5.5580192565917965,
      "epoch": 0.8912662906049407,
      "grad_norm": 1.046875,
      "learning_rate": 0.00049257906255792,
      "loss": 5.3027,
      "mean_token_accuracy": 0.17847901135683059,
      "num_tokens": 19885085.0,
      "step": 11455
    },
    {
      "entropy": 5.589003038406372,
      "epoch": 0.8916553199766583,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000492572001529988,
      "loss": 5.2485,
      "mean_token_accuracy": 0.18414873331785203,
      "num_tokens": 19893877.0,
      "step": 11460
    },
    {
      "entropy": 5.601091909408569,
      "epoch": 0.8920443493483758,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004925649372007469,
      "loss": 5.3628,
      "mean_token_accuracy": 0.18146915286779403,
      "num_tokens": 19902480.0,
      "step": 11465
    },
    {
      "entropy": 5.490126419067383,
      "epoch": 0.8924333787200933,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004925578695703045,
      "loss": 5.1391,
      "mean_token_accuracy": 0.18876268416643144,
      "num_tokens": 19910981.0,
      "step": 11470
    },
    {
      "entropy": 5.529132652282715,
      "epoch": 0.8928224080918109,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004925507986387676,
      "loss": 5.2731,
      "mean_token_accuracy": 0.177990685403347,
      "num_tokens": 19920055.0,
      "step": 11475
    },
    {
      "entropy": 5.5890837669372555,
      "epoch": 0.8932114374635285,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004925437244062436,
      "loss": 5.3176,
      "mean_token_accuracy": 0.18110155463218688,
      "num_tokens": 19928010.0,
      "step": 11480
    },
    {
      "entropy": 5.574079847335815,
      "epoch": 0.8936004668352461,
      "grad_norm": 1.125,
      "learning_rate": 0.0004925366468728397,
      "loss": 5.209,
      "mean_token_accuracy": 0.18246833384037017,
      "num_tokens": 19935992.0,
      "step": 11485
    },
    {
      "entropy": 5.542336320877075,
      "epoch": 0.8939894962069637,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004925295660386635,
      "loss": 5.3221,
      "mean_token_accuracy": 0.1750795751810074,
      "num_tokens": 19944522.0,
      "step": 11490
    },
    {
      "entropy": 5.630357646942139,
      "epoch": 0.8943785255786811,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004925224819038224,
      "loss": 5.2607,
      "mean_token_accuracy": 0.18660108745098114,
      "num_tokens": 19952991.0,
      "step": 11495
    },
    {
      "entropy": 5.505500078201294,
      "epoch": 0.8947675549503987,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004925153944684239,
      "loss": 5.2802,
      "mean_token_accuracy": 0.18006787747144698,
      "num_tokens": 19962301.0,
      "step": 11500
    },
    {
      "entropy": 5.548112249374389,
      "epoch": 0.8951565843221163,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004925083037325754,
      "loss": 5.3431,
      "mean_token_accuracy": 0.17341893911361694,
      "num_tokens": 19971159.0,
      "step": 11505
    },
    {
      "entropy": 5.573211765289306,
      "epoch": 0.8955456136938339,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004925012096963847,
      "loss": 5.2219,
      "mean_token_accuracy": 0.17810489237308502,
      "num_tokens": 19979608.0,
      "step": 11510
    },
    {
      "entropy": 5.538905382156372,
      "epoch": 0.8959346430655515,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004924941123599593,
      "loss": 5.1897,
      "mean_token_accuracy": 0.18554722964763642,
      "num_tokens": 19988348.0,
      "step": 11515
    },
    {
      "entropy": 5.490309619903565,
      "epoch": 0.896323672437269,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004924870117234069,
      "loss": 5.2335,
      "mean_token_accuracy": 0.18003566861152648,
      "num_tokens": 19996970.0,
      "step": 11520
    },
    {
      "entropy": 5.613439416885376,
      "epoch": 0.8967127018089865,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004924799077868353,
      "loss": 5.3539,
      "mean_token_accuracy": 0.17875537872314454,
      "num_tokens": 20005297.0,
      "step": 11525
    },
    {
      "entropy": 5.603316068649292,
      "epoch": 0.8971017311807041,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004924728005503522,
      "loss": 5.2734,
      "mean_token_accuracy": 0.17533954828977585,
      "num_tokens": 20013971.0,
      "step": 11530
    },
    {
      "entropy": 5.526136970520019,
      "epoch": 0.8974907605524217,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004924656900140655,
      "loss": 5.2304,
      "mean_token_accuracy": 0.18262754529714584,
      "num_tokens": 20022536.0,
      "step": 11535
    },
    {
      "entropy": 5.469740772247315,
      "epoch": 0.8978797899241393,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004924585761780831,
      "loss": 5.169,
      "mean_token_accuracy": 0.18438226878643035,
      "num_tokens": 20031676.0,
      "step": 11540
    },
    {
      "entropy": 5.54424467086792,
      "epoch": 0.8982688192958569,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000492451459042513,
      "loss": 5.2478,
      "mean_token_accuracy": 0.18332523703575135,
      "num_tokens": 20039783.0,
      "step": 11545
    },
    {
      "entropy": 5.508765506744385,
      "epoch": 0.8986578486675744,
      "grad_norm": 1.03125,
      "learning_rate": 0.000492444338607463,
      "loss": 5.2919,
      "mean_token_accuracy": 0.17934702336788177,
      "num_tokens": 20047859.0,
      "step": 11550
    },
    {
      "entropy": 5.561261701583862,
      "epoch": 0.899046878039292,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004924372148730413,
      "loss": 5.3268,
      "mean_token_accuracy": 0.17403075397014617,
      "num_tokens": 20056711.0,
      "step": 11555
    },
    {
      "entropy": 5.522530364990234,
      "epoch": 0.8994359074110095,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000492430087839356,
      "loss": 5.1575,
      "mean_token_accuracy": 0.18537795692682266,
      "num_tokens": 20065715.0,
      "step": 11560
    },
    {
      "entropy": 5.502246713638305,
      "epoch": 0.8998249367827271,
      "grad_norm": 1.0,
      "learning_rate": 0.0004924229575065152,
      "loss": 5.2603,
      "mean_token_accuracy": 0.18284232765436173,
      "num_tokens": 20073815.0,
      "step": 11565
    },
    {
      "entropy": 5.558191728591919,
      "epoch": 0.9002139661544447,
      "grad_norm": 1.03125,
      "learning_rate": 0.000492415823874627,
      "loss": 5.2424,
      "mean_token_accuracy": 0.18106332719326018,
      "num_tokens": 20082331.0,
      "step": 11570
    },
    {
      "entropy": 5.5434516906738285,
      "epoch": 0.9006029955261622,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004924086869437998,
      "loss": 5.3132,
      "mean_token_accuracy": 0.1804625168442726,
      "num_tokens": 20091095.0,
      "step": 11575
    },
    {
      "entropy": 5.543539762496948,
      "epoch": 0.9009920248978798,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004924015467141417,
      "loss": 5.2199,
      "mean_token_accuracy": 0.18811064809560776,
      "num_tokens": 20100111.0,
      "step": 11580
    },
    {
      "entropy": 5.60541877746582,
      "epoch": 0.9013810542695974,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004923944031857613,
      "loss": 5.2883,
      "mean_token_accuracy": 0.18593319207429887,
      "num_tokens": 20108542.0,
      "step": 11585
    },
    {
      "entropy": 5.508515310287476,
      "epoch": 0.901770083641315,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004923872563587668,
      "loss": 5.2122,
      "mean_token_accuracy": 0.1845849335193634,
      "num_tokens": 20117831.0,
      "step": 11590
    },
    {
      "entropy": 5.485803604125977,
      "epoch": 0.9021591130130325,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004923801062332666,
      "loss": 5.1701,
      "mean_token_accuracy": 0.18338950872421264,
      "num_tokens": 20127322.0,
      "step": 11595
    },
    {
      "entropy": 5.607252502441407,
      "epoch": 0.90254814238475,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004923729528093694,
      "loss": 5.2254,
      "mean_token_accuracy": 0.18323505222797393,
      "num_tokens": 20135772.0,
      "step": 11600
    },
    {
      "entropy": 5.579242277145386,
      "epoch": 0.9029371717564676,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004923657960871836,
      "loss": 5.2441,
      "mean_token_accuracy": 0.18849527835845947,
      "num_tokens": 20144110.0,
      "step": 11605
    },
    {
      "entropy": 5.546326112747193,
      "epoch": 0.9033262011281852,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004923586360668178,
      "loss": 5.3461,
      "mean_token_accuracy": 0.17696415185928344,
      "num_tokens": 20153391.0,
      "step": 11610
    },
    {
      "entropy": 5.476033687591553,
      "epoch": 0.9037152304999028,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004923514727483807,
      "loss": 5.1654,
      "mean_token_accuracy": 0.18849604278802873,
      "num_tokens": 20161877.0,
      "step": 11615
    },
    {
      "entropy": 5.4428685188293455,
      "epoch": 0.9041042598716204,
      "grad_norm": 1.0625,
      "learning_rate": 0.000492344306131981,
      "loss": 5.136,
      "mean_token_accuracy": 0.19015321731567383,
      "num_tokens": 20169922.0,
      "step": 11620
    },
    {
      "entropy": 5.455940103530883,
      "epoch": 0.9044932892433378,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004923371362177272,
      "loss": 5.183,
      "mean_token_accuracy": 0.188796167075634,
      "num_tokens": 20177780.0,
      "step": 11625
    },
    {
      "entropy": 5.5696611404418945,
      "epoch": 0.9048823186150554,
      "grad_norm": 1.0,
      "learning_rate": 0.0004923299630057284,
      "loss": 5.1724,
      "mean_token_accuracy": 0.1908013865351677,
      "num_tokens": 20186303.0,
      "step": 11630
    },
    {
      "entropy": 5.481515169143677,
      "epoch": 0.905271347986773,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004923227864960934,
      "loss": 5.1675,
      "mean_token_accuracy": 0.18642303645610808,
      "num_tokens": 20194042.0,
      "step": 11635
    },
    {
      "entropy": 5.410983562469482,
      "epoch": 0.9056603773584906,
      "grad_norm": 1.046875,
      "learning_rate": 0.000492315606688931,
      "loss": 5.171,
      "mean_token_accuracy": 0.18848992139101028,
      "num_tokens": 20202929.0,
      "step": 11640
    },
    {
      "entropy": 5.503472805023193,
      "epoch": 0.9060494067302082,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004923084235843501,
      "loss": 5.219,
      "mean_token_accuracy": 0.17720972150564193,
      "num_tokens": 20210864.0,
      "step": 11645
    },
    {
      "entropy": 5.569815921783447,
      "epoch": 0.9064384361019256,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004923012371824598,
      "loss": 5.2364,
      "mean_token_accuracy": 0.18099671006202697,
      "num_tokens": 20219390.0,
      "step": 11650
    },
    {
      "entropy": 5.486623287200928,
      "epoch": 0.9068274654736432,
      "grad_norm": 1.046875,
      "learning_rate": 0.000492294047483369,
      "loss": 5.1826,
      "mean_token_accuracy": 0.18897834867239,
      "num_tokens": 20227241.0,
      "step": 11655
    },
    {
      "entropy": 5.484668636322022,
      "epoch": 0.9072164948453608,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004922868544871869,
      "loss": 5.3309,
      "mean_token_accuracy": 0.17811907231807708,
      "num_tokens": 20236001.0,
      "step": 11660
    },
    {
      "entropy": 5.5983078956604,
      "epoch": 0.9076055242170784,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004922796581940227,
      "loss": 5.2153,
      "mean_token_accuracy": 0.18252134770154954,
      "num_tokens": 20244316.0,
      "step": 11665
    },
    {
      "entropy": 5.469456815719605,
      "epoch": 0.907994553588796,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004922724586039855,
      "loss": 5.1886,
      "mean_token_accuracy": 0.1855294108390808,
      "num_tokens": 20252693.0,
      "step": 11670
    },
    {
      "entropy": 5.453940296173096,
      "epoch": 0.9083835829605135,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004922652557171846,
      "loss": 5.2398,
      "mean_token_accuracy": 0.18441332131624222,
      "num_tokens": 20262252.0,
      "step": 11675
    },
    {
      "entropy": 5.493917942047119,
      "epoch": 0.9087726123322311,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004922580495337292,
      "loss": 5.0808,
      "mean_token_accuracy": 0.19598100036382676,
      "num_tokens": 20270755.0,
      "step": 11680
    },
    {
      "entropy": 5.513514804840088,
      "epoch": 0.9091616417039486,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004922508400537288,
      "loss": 5.1569,
      "mean_token_accuracy": 0.18096383512020112,
      "num_tokens": 20278739.0,
      "step": 11685
    },
    {
      "entropy": 5.49625449180603,
      "epoch": 0.9095506710756662,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004922436272772926,
      "loss": 5.2993,
      "mean_token_accuracy": 0.17530277669429778,
      "num_tokens": 20288160.0,
      "step": 11690
    },
    {
      "entropy": 5.505362462997437,
      "epoch": 0.9099397004473838,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004922364112045301,
      "loss": 5.2402,
      "mean_token_accuracy": 0.18546832650899886,
      "num_tokens": 20296730.0,
      "step": 11695
    },
    {
      "entropy": 5.531610202789307,
      "epoch": 0.9103287298191013,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000492229191835551,
      "loss": 5.1906,
      "mean_token_accuracy": 0.18129133880138398,
      "num_tokens": 20305941.0,
      "step": 11700
    },
    {
      "entropy": 5.502279901504517,
      "epoch": 0.9107177591908189,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004922219691704645,
      "loss": 5.2073,
      "mean_token_accuracy": 0.186127969622612,
      "num_tokens": 20314081.0,
      "step": 11705
    },
    {
      "entropy": 5.536750602722168,
      "epoch": 0.9111067885625365,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004922147432093805,
      "loss": 5.2688,
      "mean_token_accuracy": 0.17636242657899856,
      "num_tokens": 20323004.0,
      "step": 11710
    },
    {
      "entropy": 5.517223501205445,
      "epoch": 0.911495817934254,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004922075139524083,
      "loss": 5.1687,
      "mean_token_accuracy": 0.1817627564072609,
      "num_tokens": 20331122.0,
      "step": 11715
    },
    {
      "entropy": 5.50209903717041,
      "epoch": 0.9118848473059716,
      "grad_norm": 1.078125,
      "learning_rate": 0.000492200281399658,
      "loss": 5.255,
      "mean_token_accuracy": 0.1775050863623619,
      "num_tokens": 20339602.0,
      "step": 11720
    },
    {
      "entropy": 5.505890512466431,
      "epoch": 0.9122738766776891,
      "grad_norm": 1.046875,
      "learning_rate": 0.000492193045551239,
      "loss": 5.1106,
      "mean_token_accuracy": 0.18178810477256774,
      "num_tokens": 20348431.0,
      "step": 11725
    },
    {
      "entropy": 5.485821962356567,
      "epoch": 0.9126629060494067,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004921858064072612,
      "loss": 5.1736,
      "mean_token_accuracy": 0.1830815330147743,
      "num_tokens": 20357235.0,
      "step": 11730
    },
    {
      "entropy": 5.514485788345337,
      "epoch": 0.9130519354211243,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004921785639678347,
      "loss": 5.1292,
      "mean_token_accuracy": 0.18329256027936935,
      "num_tokens": 20366169.0,
      "step": 11735
    },
    {
      "entropy": 5.5283403396606445,
      "epoch": 0.9134409647928419,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000492171318233069,
      "loss": 5.2257,
      "mean_token_accuracy": 0.1810527488589287,
      "num_tokens": 20375213.0,
      "step": 11740
    },
    {
      "entropy": 5.483912611007691,
      "epoch": 0.9138299941645595,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004921640692030742,
      "loss": 5.2209,
      "mean_token_accuracy": 0.18104212880134582,
      "num_tokens": 20383992.0,
      "step": 11745
    },
    {
      "entropy": 5.52657265663147,
      "epoch": 0.914219023536277,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004921568168779603,
      "loss": 5.1808,
      "mean_token_accuracy": 0.19195883572101594,
      "num_tokens": 20392318.0,
      "step": 11750
    },
    {
      "entropy": 5.572586870193481,
      "epoch": 0.9146080529079945,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004921495612578374,
      "loss": 5.3113,
      "mean_token_accuracy": 0.18139930814504623,
      "num_tokens": 20401556.0,
      "step": 11755
    },
    {
      "entropy": 5.4868388175964355,
      "epoch": 0.9149970822797121,
      "grad_norm": 1.0,
      "learning_rate": 0.0004921423023428156,
      "loss": 5.2058,
      "mean_token_accuracy": 0.18366467952728271,
      "num_tokens": 20410480.0,
      "step": 11760
    },
    {
      "entropy": 5.482418537139893,
      "epoch": 0.9153861116514297,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004921350401330049,
      "loss": 5.1562,
      "mean_token_accuracy": 0.18877200037240982,
      "num_tokens": 20419748.0,
      "step": 11765
    },
    {
      "entropy": 5.539521026611328,
      "epoch": 0.9157751410231473,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004921277746285155,
      "loss": 5.225,
      "mean_token_accuracy": 0.18052946478128434,
      "num_tokens": 20429552.0,
      "step": 11770
    },
    {
      "entropy": 5.671813058853149,
      "epoch": 0.9161641703948649,
      "grad_norm": 1.0,
      "learning_rate": 0.0004921205058294579,
      "loss": 5.2842,
      "mean_token_accuracy": 0.18181693702936172,
      "num_tokens": 20438476.0,
      "step": 11775
    },
    {
      "entropy": 5.487366819381714,
      "epoch": 0.9165531997665823,
      "grad_norm": 1.046875,
      "learning_rate": 0.000492113233735942,
      "loss": 5.0968,
      "mean_token_accuracy": 0.18619766980409622,
      "num_tokens": 20447035.0,
      "step": 11780
    },
    {
      "entropy": 5.49984655380249,
      "epoch": 0.9169422291382999,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004921059583480785,
      "loss": 5.2452,
      "mean_token_accuracy": 0.1809439867734909,
      "num_tokens": 20456200.0,
      "step": 11785
    },
    {
      "entropy": 5.5475890159606935,
      "epoch": 0.9173312585100175,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004920986796659775,
      "loss": 5.2385,
      "mean_token_accuracy": 0.18756065666675567,
      "num_tokens": 20465444.0,
      "step": 11790
    },
    {
      "entropy": 5.483853769302368,
      "epoch": 0.9177202878817351,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004920913976897498,
      "loss": 5.1283,
      "mean_token_accuracy": 0.18811491131782532,
      "num_tokens": 20473851.0,
      "step": 11795
    },
    {
      "entropy": 5.482855272293091,
      "epoch": 0.9181093172534527,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004920841124195055,
      "loss": 5.2268,
      "mean_token_accuracy": 0.18887475430965422,
      "num_tokens": 20482232.0,
      "step": 11800
    },
    {
      "entropy": 5.434363794326782,
      "epoch": 0.9184983466251702,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004920768238553554,
      "loss": 5.1028,
      "mean_token_accuracy": 0.19732008576393129,
      "num_tokens": 20490498.0,
      "step": 11805
    },
    {
      "entropy": 5.4704948425292965,
      "epoch": 0.9188873759968877,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004920695319974099,
      "loss": 5.1884,
      "mean_token_accuracy": 0.18140217065811157,
      "num_tokens": 20499125.0,
      "step": 11810
    },
    {
      "entropy": 5.542912292480469,
      "epoch": 0.9192764053686053,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004920622368457798,
      "loss": 5.3256,
      "mean_token_accuracy": 0.1799159273505211,
      "num_tokens": 20507665.0,
      "step": 11815
    },
    {
      "entropy": 5.523641300201416,
      "epoch": 0.9196654347403229,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004920549384005759,
      "loss": 5.2588,
      "mean_token_accuracy": 0.18213907033205032,
      "num_tokens": 20516509.0,
      "step": 11820
    },
    {
      "entropy": 5.539924144744873,
      "epoch": 0.9200544641120405,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004920476366619086,
      "loss": 5.2877,
      "mean_token_accuracy": 0.18220312595367433,
      "num_tokens": 20524830.0,
      "step": 11825
    },
    {
      "entropy": 5.5222255229949955,
      "epoch": 0.920443493483758,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000492040331629889,
      "loss": 5.1729,
      "mean_token_accuracy": 0.18410732746124267,
      "num_tokens": 20533746.0,
      "step": 11830
    },
    {
      "entropy": 5.509216785430908,
      "epoch": 0.9208325228554756,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004920330233046277,
      "loss": 5.154,
      "mean_token_accuracy": 0.19174797385931014,
      "num_tokens": 20542350.0,
      "step": 11835
    },
    {
      "entropy": 5.486442279815674,
      "epoch": 0.9212215522271932,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004920257116862359,
      "loss": 5.2827,
      "mean_token_accuracy": 0.18453929871320723,
      "num_tokens": 20551670.0,
      "step": 11840
    },
    {
      "entropy": 5.529729509353638,
      "epoch": 0.9216105815989107,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004920183967748241,
      "loss": 5.2371,
      "mean_token_accuracy": 0.18128502666950225,
      "num_tokens": 20560692.0,
      "step": 11845
    },
    {
      "entropy": 5.5278490543365475,
      "epoch": 0.9219996109706283,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004920110785705037,
      "loss": 5.1648,
      "mean_token_accuracy": 0.1820360690355301,
      "num_tokens": 20569262.0,
      "step": 11850
    },
    {
      "entropy": 5.496212911605835,
      "epoch": 0.9223886403423458,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004920037570733857,
      "loss": 5.1866,
      "mean_token_accuracy": 0.18535787761211395,
      "num_tokens": 20577850.0,
      "step": 11855
    },
    {
      "entropy": 5.520932531356811,
      "epoch": 0.9227776697140634,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004919964322835809,
      "loss": 5.2586,
      "mean_token_accuracy": 0.1754717618227005,
      "num_tokens": 20587021.0,
      "step": 11860
    },
    {
      "entropy": 5.551356315612793,
      "epoch": 0.923166699085781,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004919891042012006,
      "loss": 5.1973,
      "mean_token_accuracy": 0.1846883177757263,
      "num_tokens": 20594631.0,
      "step": 11865
    },
    {
      "entropy": 5.494245147705078,
      "epoch": 0.9235557284574986,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004919817728263562,
      "loss": 5.1031,
      "mean_token_accuracy": 0.18781720697879792,
      "num_tokens": 20602765.0,
      "step": 11870
    },
    {
      "entropy": 5.561323738098144,
      "epoch": 0.9239447578292161,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004919744381591586,
      "loss": 5.2521,
      "mean_token_accuracy": 0.1918638288974762,
      "num_tokens": 20611350.0,
      "step": 11875
    },
    {
      "entropy": 5.50352373123169,
      "epoch": 0.9243337872009336,
      "grad_norm": 1.0,
      "learning_rate": 0.0004919671001997193,
      "loss": 5.1749,
      "mean_token_accuracy": 0.18756291419267654,
      "num_tokens": 20620201.0,
      "step": 11880
    },
    {
      "entropy": 5.55888180732727,
      "epoch": 0.9247228165726512,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004919597589481497,
      "loss": 5.2335,
      "mean_token_accuracy": 0.18659303039312364,
      "num_tokens": 20627870.0,
      "step": 11885
    },
    {
      "entropy": 5.513376903533936,
      "epoch": 0.9251118459443688,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000491952414404561,
      "loss": 5.2147,
      "mean_token_accuracy": 0.18615731000900268,
      "num_tokens": 20635894.0,
      "step": 11890
    },
    {
      "entropy": 5.5336854457855225,
      "epoch": 0.9255008753160864,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004919450665690646,
      "loss": 5.2604,
      "mean_token_accuracy": 0.18203404247760774,
      "num_tokens": 20644091.0,
      "step": 11895
    },
    {
      "entropy": 5.534296703338623,
      "epoch": 0.925889904687804,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004919377154417724,
      "loss": 5.1313,
      "mean_token_accuracy": 0.18835406452417375,
      "num_tokens": 20651994.0,
      "step": 11900
    },
    {
      "entropy": 5.501727294921875,
      "epoch": 0.9262789340595214,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004919303610227954,
      "loss": 5.1621,
      "mean_token_accuracy": 0.18716610819101334,
      "num_tokens": 20660192.0,
      "step": 11905
    },
    {
      "entropy": 5.62515606880188,
      "epoch": 0.926667963431239,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004919230033122457,
      "loss": 5.2327,
      "mean_token_accuracy": 0.1798260271549225,
      "num_tokens": 20668275.0,
      "step": 11910
    },
    {
      "entropy": 5.456781387329102,
      "epoch": 0.9270569928029566,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004919156423102345,
      "loss": 5.2211,
      "mean_token_accuracy": 0.18448174893856048,
      "num_tokens": 20676936.0,
      "step": 11915
    },
    {
      "entropy": 5.551150131225586,
      "epoch": 0.9274460221746742,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004919082780168736,
      "loss": 5.3299,
      "mean_token_accuracy": 0.17673248946666717,
      "num_tokens": 20685600.0,
      "step": 11920
    },
    {
      "entropy": 5.554179668426514,
      "epoch": 0.9278350515463918,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004919009104322751,
      "loss": 5.2633,
      "mean_token_accuracy": 0.177079938352108,
      "num_tokens": 20693714.0,
      "step": 11925
    },
    {
      "entropy": 5.50539722442627,
      "epoch": 0.9282240809181093,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004918935395565503,
      "loss": 5.2183,
      "mean_token_accuracy": 0.17974751591682434,
      "num_tokens": 20702529.0,
      "step": 11930
    },
    {
      "entropy": 5.517735862731934,
      "epoch": 0.9286131102898268,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004918861653898113,
      "loss": 5.1445,
      "mean_token_accuracy": 0.18577435612678528,
      "num_tokens": 20711775.0,
      "step": 11935
    },
    {
      "entropy": 5.557012128829956,
      "epoch": 0.9290021396615444,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004918787879321701,
      "loss": 5.2046,
      "mean_token_accuracy": 0.18271944522857667,
      "num_tokens": 20720653.0,
      "step": 11940
    },
    {
      "entropy": 5.536213207244873,
      "epoch": 0.929391169033262,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004918714071837384,
      "loss": 5.2236,
      "mean_token_accuracy": 0.18275836408138274,
      "num_tokens": 20729208.0,
      "step": 11945
    },
    {
      "entropy": 5.53324408531189,
      "epoch": 0.9297801984049796,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004918640231446282,
      "loss": 5.2771,
      "mean_token_accuracy": 0.18584540039300917,
      "num_tokens": 20738670.0,
      "step": 11950
    },
    {
      "entropy": 5.534483146667481,
      "epoch": 0.9301692277766972,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004918566358149517,
      "loss": 5.3176,
      "mean_token_accuracy": 0.17935009747743608,
      "num_tokens": 20748315.0,
      "step": 11955
    },
    {
      "entropy": 5.603895759582519,
      "epoch": 0.9305582571484147,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000491849245194821,
      "loss": 5.3762,
      "mean_token_accuracy": 0.17650924623012543,
      "num_tokens": 20757623.0,
      "step": 11960
    },
    {
      "entropy": 5.551540517807007,
      "epoch": 0.9309472865201323,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004918418512843479,
      "loss": 5.199,
      "mean_token_accuracy": 0.1850524827837944,
      "num_tokens": 20765817.0,
      "step": 11965
    },
    {
      "entropy": 5.4734776496887205,
      "epoch": 0.9313363158918498,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004918344540836451,
      "loss": 5.0873,
      "mean_token_accuracy": 0.19647066295146942,
      "num_tokens": 20773931.0,
      "step": 11970
    },
    {
      "entropy": 5.5662108898162845,
      "epoch": 0.9317253452635674,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004918270535928244,
      "loss": 5.2148,
      "mean_token_accuracy": 0.17945789098739623,
      "num_tokens": 20782652.0,
      "step": 11975
    },
    {
      "entropy": 5.512031841278076,
      "epoch": 0.932114374635285,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004918196498119984,
      "loss": 5.1196,
      "mean_token_accuracy": 0.19094449281692505,
      "num_tokens": 20791293.0,
      "step": 11980
    },
    {
      "entropy": 5.48109073638916,
      "epoch": 0.9325034040070025,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004918122427412793,
      "loss": 5.2237,
      "mean_token_accuracy": 0.17959160208702088,
      "num_tokens": 20800317.0,
      "step": 11985
    },
    {
      "entropy": 5.493942403793335,
      "epoch": 0.9328924333787201,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004918048323807795,
      "loss": 5.1973,
      "mean_token_accuracy": 0.19213191866874696,
      "num_tokens": 20809231.0,
      "step": 11990
    },
    {
      "entropy": 5.535474443435669,
      "epoch": 0.9332814627504377,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004917974187306114,
      "loss": 5.1347,
      "mean_token_accuracy": 0.18500607460737228,
      "num_tokens": 20818327.0,
      "step": 11995
    },
    {
      "entropy": 5.580713891983033,
      "epoch": 0.9336704921221552,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004917900017908875,
      "loss": 5.2948,
      "mean_token_accuracy": 0.17334431111812593,
      "num_tokens": 20826995.0,
      "step": 12000
    },
    {
      "epoch": 0.9336704921221552,
      "eval_entropy": 5.300555547965732,
      "eval_loss": 5.243749618530273,
      "eval_mean_token_accuracy": 0.19068797228856083,
      "eval_num_tokens": 20826995.0,
      "eval_runtime": 21.7014,
      "eval_samples_per_second": 1914.988,
      "eval_steps_per_second": 239.385,
      "step": 12000
    },
    {
      "entropy": 5.394803857803344,
      "epoch": 0.9340595214938728,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004917825815617205,
      "loss": 5.0103,
      "mean_token_accuracy": 0.19607135355472566,
      "num_tokens": 20835641.0,
      "step": 12005
    },
    {
      "entropy": 5.537971925735474,
      "epoch": 0.9344485508655903,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004917751580432229,
      "loss": 5.1574,
      "mean_token_accuracy": 0.1878841832280159,
      "num_tokens": 20844728.0,
      "step": 12010
    },
    {
      "entropy": 5.490711736679077,
      "epoch": 0.9348375802373079,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004917677312355072,
      "loss": 5.1432,
      "mean_token_accuracy": 0.19067799299955368,
      "num_tokens": 20853013.0,
      "step": 12015
    },
    {
      "entropy": 5.560920906066895,
      "epoch": 0.9352266096090255,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004917603011386863,
      "loss": 5.225,
      "mean_token_accuracy": 0.181848905980587,
      "num_tokens": 20861592.0,
      "step": 12020
    },
    {
      "entropy": 5.509307336807251,
      "epoch": 0.9356156389807431,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004917528677528727,
      "loss": 5.1752,
      "mean_token_accuracy": 0.18179640918970108,
      "num_tokens": 20869645.0,
      "step": 12025
    },
    {
      "entropy": 5.511589384078979,
      "epoch": 0.9360046683524607,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004917454310781795,
      "loss": 5.2361,
      "mean_token_accuracy": 0.18627407848834993,
      "num_tokens": 20878154.0,
      "step": 12030
    },
    {
      "entropy": 5.540191650390625,
      "epoch": 0.9363936977241781,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004917379911147193,
      "loss": 5.167,
      "mean_token_accuracy": 0.19877610057592393,
      "num_tokens": 20886260.0,
      "step": 12035
    },
    {
      "entropy": 5.478352975845337,
      "epoch": 0.9367827270958957,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004917305478626049,
      "loss": 5.1446,
      "mean_token_accuracy": 0.18928674906492232,
      "num_tokens": 20895688.0,
      "step": 12040
    },
    {
      "entropy": 5.502701711654663,
      "epoch": 0.9371717564676133,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004917231013219495,
      "loss": 5.2527,
      "mean_token_accuracy": 0.1833703711628914,
      "num_tokens": 20903904.0,
      "step": 12045
    },
    {
      "entropy": 5.551312208175659,
      "epoch": 0.9375607858393309,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000491715651492866,
      "loss": 5.2441,
      "mean_token_accuracy": 0.18637545108795167,
      "num_tokens": 20913411.0,
      "step": 12050
    },
    {
      "entropy": 5.546638870239258,
      "epoch": 0.9379498152110485,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004917081983754675,
      "loss": 5.22,
      "mean_token_accuracy": 0.17622820287942886,
      "num_tokens": 20922088.0,
      "step": 12055
    },
    {
      "entropy": 5.430462789535523,
      "epoch": 0.938338844582766,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004917007419698669,
      "loss": 5.2162,
      "mean_token_accuracy": 0.1845819279551506,
      "num_tokens": 20930879.0,
      "step": 12060
    },
    {
      "entropy": 5.467606687545777,
      "epoch": 0.9387278739544835,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004916932822761776,
      "loss": 5.1833,
      "mean_token_accuracy": 0.18965914696455002,
      "num_tokens": 20939092.0,
      "step": 12065
    },
    {
      "entropy": 5.516851902008057,
      "epoch": 0.9391169033262011,
      "grad_norm": 1.0,
      "learning_rate": 0.0004916858192945126,
      "loss": 5.2278,
      "mean_token_accuracy": 0.18450311571359634,
      "num_tokens": 20947599.0,
      "step": 12070
    },
    {
      "entropy": 5.549443626403809,
      "epoch": 0.9395059326979187,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004916783530249852,
      "loss": 5.2649,
      "mean_token_accuracy": 0.18046874552965164,
      "num_tokens": 20955725.0,
      "step": 12075
    },
    {
      "entropy": 5.425970268249512,
      "epoch": 0.9398949620696363,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004916708834677086,
      "loss": 5.0842,
      "mean_token_accuracy": 0.19225522130727768,
      "num_tokens": 20965040.0,
      "step": 12080
    },
    {
      "entropy": 5.508653593063355,
      "epoch": 0.9402839914413538,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004916634106227962,
      "loss": 5.2486,
      "mean_token_accuracy": 0.17833081483840943,
      "num_tokens": 20974204.0,
      "step": 12085
    },
    {
      "entropy": 5.601053142547608,
      "epoch": 0.9406730208130714,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004916559344903614,
      "loss": 5.238,
      "mean_token_accuracy": 0.1862572729587555,
      "num_tokens": 20982109.0,
      "step": 12090
    },
    {
      "entropy": 5.539611864089966,
      "epoch": 0.941062050184789,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004916484550705176,
      "loss": 5.2554,
      "mean_token_accuracy": 0.17573074847459794,
      "num_tokens": 20990375.0,
      "step": 12095
    },
    {
      "entropy": 5.506649112701416,
      "epoch": 0.9414510795565065,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004916409723633785,
      "loss": 5.2683,
      "mean_token_accuracy": 0.18533024936914444,
      "num_tokens": 20998535.0,
      "step": 12100
    },
    {
      "entropy": 5.454801273345947,
      "epoch": 0.9418401089282241,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004916334863690573,
      "loss": 5.1212,
      "mean_token_accuracy": 0.18566579967737198,
      "num_tokens": 21006597.0,
      "step": 12105
    },
    {
      "entropy": 5.497125720977783,
      "epoch": 0.9422291382999416,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004916259970876678,
      "loss": 5.1912,
      "mean_token_accuracy": 0.18758206218481063,
      "num_tokens": 21015498.0,
      "step": 12110
    },
    {
      "entropy": 5.568619346618652,
      "epoch": 0.9426181676716592,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004916185045193237,
      "loss": 5.2209,
      "mean_token_accuracy": 0.1797187075018883,
      "num_tokens": 21024090.0,
      "step": 12115
    },
    {
      "entropy": 5.567369890213013,
      "epoch": 0.9430071970433768,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004916110086641384,
      "loss": 5.2571,
      "mean_token_accuracy": 0.1779808446764946,
      "num_tokens": 21032928.0,
      "step": 12120
    },
    {
      "entropy": 5.547671127319336,
      "epoch": 0.9433962264150944,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004916035095222259,
      "loss": 5.3211,
      "mean_token_accuracy": 0.18082233369350434,
      "num_tokens": 21041467.0,
      "step": 12125
    },
    {
      "entropy": 5.533851957321167,
      "epoch": 0.9437852557868119,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004915960070936999,
      "loss": 5.2315,
      "mean_token_accuracy": 0.19182967394590378,
      "num_tokens": 21049325.0,
      "step": 12130
    },
    {
      "entropy": 5.510747671127319,
      "epoch": 0.9441742851585294,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004915885013786742,
      "loss": 5.2851,
      "mean_token_accuracy": 0.1774992197751999,
      "num_tokens": 21057911.0,
      "step": 12135
    },
    {
      "entropy": 5.554391813278198,
      "epoch": 0.944563314530247,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004915809923772628,
      "loss": 5.2815,
      "mean_token_accuracy": 0.17487628757953644,
      "num_tokens": 21066974.0,
      "step": 12140
    },
    {
      "entropy": 5.5443826675415036,
      "epoch": 0.9449523439019646,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004915734800895796,
      "loss": 5.2083,
      "mean_token_accuracy": 0.18278666585683823,
      "num_tokens": 21075591.0,
      "step": 12145
    },
    {
      "entropy": 5.516664648056031,
      "epoch": 0.9453413732736822,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004915659645157383,
      "loss": 5.2596,
      "mean_token_accuracy": 0.1780250772833824,
      "num_tokens": 21084496.0,
      "step": 12150
    },
    {
      "entropy": 5.56315336227417,
      "epoch": 0.9457304026453998,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004915584456558535,
      "loss": 5.1474,
      "mean_token_accuracy": 0.18724167197942734,
      "num_tokens": 21092811.0,
      "step": 12155
    },
    {
      "entropy": 5.551301717758179,
      "epoch": 0.9461194320171173,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004915509235100388,
      "loss": 5.2002,
      "mean_token_accuracy": 0.18451928496360778,
      "num_tokens": 21101449.0,
      "step": 12160
    },
    {
      "entropy": 5.635024166107177,
      "epoch": 0.9465084613888348,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004915433980784086,
      "loss": 5.3164,
      "mean_token_accuracy": 0.17514045983552934,
      "num_tokens": 21109476.0,
      "step": 12165
    },
    {
      "entropy": 5.507270669937133,
      "epoch": 0.9468974907605524,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004915358693610769,
      "loss": 5.2044,
      "mean_token_accuracy": 0.1891069769859314,
      "num_tokens": 21118431.0,
      "step": 12170
    },
    {
      "entropy": 5.490405321121216,
      "epoch": 0.94728652013227,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004915283373581581,
      "loss": 5.1794,
      "mean_token_accuracy": 0.18598416894674302,
      "num_tokens": 21127390.0,
      "step": 12175
    },
    {
      "entropy": 5.596442794799804,
      "epoch": 0.9476755495039876,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004915208020697664,
      "loss": 5.2947,
      "mean_token_accuracy": 0.17948462963104247,
      "num_tokens": 21136323.0,
      "step": 12180
    },
    {
      "entropy": 5.513361978530884,
      "epoch": 0.9480645788757052,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004915132634960162,
      "loss": 5.2424,
      "mean_token_accuracy": 0.18058507591485978,
      "num_tokens": 21145228.0,
      "step": 12185
    },
    {
      "entropy": 5.512512111663819,
      "epoch": 0.9484536082474226,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004915057216370218,
      "loss": 5.1387,
      "mean_token_accuracy": 0.18589064925909043,
      "num_tokens": 21153288.0,
      "step": 12190
    },
    {
      "entropy": 5.5034928798675535,
      "epoch": 0.9488426376191402,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004914981764928977,
      "loss": 5.1983,
      "mean_token_accuracy": 0.1897186130285263,
      "num_tokens": 21161802.0,
      "step": 12195
    },
    {
      "entropy": 5.45917501449585,
      "epoch": 0.9492316669908578,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004914906280637584,
      "loss": 5.1545,
      "mean_token_accuracy": 0.18774567544460297,
      "num_tokens": 21170018.0,
      "step": 12200
    },
    {
      "entropy": 5.584926557540894,
      "epoch": 0.9496206963625754,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004914830763497183,
      "loss": 5.346,
      "mean_token_accuracy": 0.17234289050102233,
      "num_tokens": 21179128.0,
      "step": 12205
    },
    {
      "entropy": 5.584822368621826,
      "epoch": 0.950009725734293,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004914755213508922,
      "loss": 5.193,
      "mean_token_accuracy": 0.18860131204128266,
      "num_tokens": 21188138.0,
      "step": 12210
    },
    {
      "entropy": 5.545753097534179,
      "epoch": 0.9503987551060105,
      "grad_norm": 1.125,
      "learning_rate": 0.0004914679630673945,
      "loss": 5.2547,
      "mean_token_accuracy": 0.18758994787931443,
      "num_tokens": 21196465.0,
      "step": 12215
    },
    {
      "entropy": 5.4517231464385985,
      "epoch": 0.950787784477728,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004914604014993401,
      "loss": 5.1684,
      "mean_token_accuracy": 0.1835176721215248,
      "num_tokens": 21205309.0,
      "step": 12220
    },
    {
      "entropy": 5.474662733078003,
      "epoch": 0.9511768138494456,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004914528366468436,
      "loss": 5.1389,
      "mean_token_accuracy": 0.1898630291223526,
      "num_tokens": 21214216.0,
      "step": 12225
    },
    {
      "entropy": 5.494830989837647,
      "epoch": 0.9515658432211632,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004914452685100199,
      "loss": 5.148,
      "mean_token_accuracy": 0.18989835530519486,
      "num_tokens": 21222246.0,
      "step": 12230
    },
    {
      "entropy": 5.504146242141724,
      "epoch": 0.9519548725928808,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004914376970889836,
      "loss": 5.2068,
      "mean_token_accuracy": 0.18950028121471404,
      "num_tokens": 21230696.0,
      "step": 12235
    },
    {
      "entropy": 5.551062870025635,
      "epoch": 0.9523439019645983,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004914301223838497,
      "loss": 5.3245,
      "mean_token_accuracy": 0.18047645539045334,
      "num_tokens": 21239909.0,
      "step": 12240
    },
    {
      "entropy": 5.497389221191407,
      "epoch": 0.9527329313363159,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004914225443947334,
      "loss": 5.1747,
      "mean_token_accuracy": 0.18310207724571229,
      "num_tokens": 21248124.0,
      "step": 12245
    },
    {
      "entropy": 5.642432689666748,
      "epoch": 0.9531219607080335,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004914149631217494,
      "loss": 5.3646,
      "mean_token_accuracy": 0.17044292241334916,
      "num_tokens": 21257042.0,
      "step": 12250
    },
    {
      "entropy": 5.610968780517578,
      "epoch": 0.953510990079751,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004914073785650127,
      "loss": 5.303,
      "mean_token_accuracy": 0.18259962052106857,
      "num_tokens": 21265627.0,
      "step": 12255
    },
    {
      "entropy": 5.572254085540772,
      "epoch": 0.9539000194514686,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004913997907246385,
      "loss": 5.218,
      "mean_token_accuracy": 0.17659999281167985,
      "num_tokens": 21273688.0,
      "step": 12260
    },
    {
      "entropy": 5.512134981155396,
      "epoch": 0.9542890488231861,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000491392199600742,
      "loss": 5.2014,
      "mean_token_accuracy": 0.18070801496505737,
      "num_tokens": 21283432.0,
      "step": 12265
    },
    {
      "entropy": 5.545650672912598,
      "epoch": 0.9546780781949037,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004913846051934382,
      "loss": 5.2373,
      "mean_token_accuracy": 0.1790466770529747,
      "num_tokens": 21292003.0,
      "step": 12270
    },
    {
      "entropy": 5.509623622894287,
      "epoch": 0.9550671075666213,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004913770075028425,
      "loss": 5.1517,
      "mean_token_accuracy": 0.1911682292819023,
      "num_tokens": 21299914.0,
      "step": 12275
    },
    {
      "entropy": 5.54285683631897,
      "epoch": 0.9554561369383389,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004913694065290701,
      "loss": 5.3598,
      "mean_token_accuracy": 0.17218934446573259,
      "num_tokens": 21309398.0,
      "step": 12280
    },
    {
      "entropy": 5.528409337997436,
      "epoch": 0.9558451663100564,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004913618022722363,
      "loss": 5.1968,
      "mean_token_accuracy": 0.1825581192970276,
      "num_tokens": 21317828.0,
      "step": 12285
    },
    {
      "entropy": 5.578054332733155,
      "epoch": 0.9562341956817739,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004913541947324566,
      "loss": 5.3038,
      "mean_token_accuracy": 0.17761738151311873,
      "num_tokens": 21326266.0,
      "step": 12290
    },
    {
      "entropy": 5.585971641540527,
      "epoch": 0.9566232250534915,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004913465839098463,
      "loss": 5.2371,
      "mean_token_accuracy": 0.1779303014278412,
      "num_tokens": 21335476.0,
      "step": 12295
    },
    {
      "entropy": 5.55401382446289,
      "epoch": 0.9570122544252091,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004913389698045209,
      "loss": 5.1641,
      "mean_token_accuracy": 0.18507529497146608,
      "num_tokens": 21344045.0,
      "step": 12300
    },
    {
      "entropy": 5.513487434387207,
      "epoch": 0.9574012837969267,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000491331352416596,
      "loss": 5.2079,
      "mean_token_accuracy": 0.18327265828847886,
      "num_tokens": 21352309.0,
      "step": 12305
    },
    {
      "entropy": 5.465046024322509,
      "epoch": 0.9577903131686443,
      "grad_norm": 1.0,
      "learning_rate": 0.0004913237317461872,
      "loss": 5.2424,
      "mean_token_accuracy": 0.1823181539773941,
      "num_tokens": 21362164.0,
      "step": 12310
    },
    {
      "entropy": 5.62496771812439,
      "epoch": 0.9581793425403617,
      "grad_norm": 1.046875,
      "learning_rate": 0.00049131610779341,
      "loss": 5.2036,
      "mean_token_accuracy": 0.18495702892541885,
      "num_tokens": 21372817.0,
      "step": 12315
    },
    {
      "entropy": 5.56168007850647,
      "epoch": 0.9585683719120793,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004913084805583803,
      "loss": 5.1866,
      "mean_token_accuracy": 0.18252000510692595,
      "num_tokens": 21381035.0,
      "step": 12320
    },
    {
      "entropy": 5.470486307144165,
      "epoch": 0.9589574012837969,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004913008500412136,
      "loss": 5.1567,
      "mean_token_accuracy": 0.19161093682050706,
      "num_tokens": 21390074.0,
      "step": 12325
    },
    {
      "entropy": 5.510234451293945,
      "epoch": 0.9593464306555145,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004912932162420259,
      "loss": 5.2111,
      "mean_token_accuracy": 0.18297782987356187,
      "num_tokens": 21398726.0,
      "step": 12330
    },
    {
      "entropy": 5.539515542984009,
      "epoch": 0.9597354600272321,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004912855791609327,
      "loss": 5.1965,
      "mean_token_accuracy": 0.1834713861346245,
      "num_tokens": 21407225.0,
      "step": 12335
    },
    {
      "entropy": 5.5185376644134525,
      "epoch": 0.9601244893989496,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004912779387980502,
      "loss": 5.1816,
      "mean_token_accuracy": 0.1831630140542984,
      "num_tokens": 21416746.0,
      "step": 12340
    },
    {
      "entropy": 5.539494562149048,
      "epoch": 0.9605135187706672,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004912702951534942,
      "loss": 5.2596,
      "mean_token_accuracy": 0.17447047531604767,
      "num_tokens": 21425621.0,
      "step": 12345
    },
    {
      "entropy": 5.543867492675782,
      "epoch": 0.9609025481423847,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004912626482273809,
      "loss": 5.1403,
      "mean_token_accuracy": 0.19669148623943328,
      "num_tokens": 21433674.0,
      "step": 12350
    },
    {
      "entropy": 5.558842039108276,
      "epoch": 0.9612915775141023,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004912549980198258,
      "loss": 5.2378,
      "mean_token_accuracy": 0.18305605202913283,
      "num_tokens": 21442285.0,
      "step": 12355
    },
    {
      "entropy": 5.53687801361084,
      "epoch": 0.9616806068858199,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004912473445309455,
      "loss": 5.2087,
      "mean_token_accuracy": 0.1866432324051857,
      "num_tokens": 21450345.0,
      "step": 12360
    },
    {
      "entropy": 5.4617656707763675,
      "epoch": 0.9620696362575375,
      "grad_norm": 1.109375,
      "learning_rate": 0.000491239687760856,
      "loss": 5.1125,
      "mean_token_accuracy": 0.1905210241675377,
      "num_tokens": 21458618.0,
      "step": 12365
    },
    {
      "entropy": 5.4627772808074955,
      "epoch": 0.962458665629255,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004912320277096732,
      "loss": 5.1528,
      "mean_token_accuracy": 0.1852654904127121,
      "num_tokens": 21467410.0,
      "step": 12370
    },
    {
      "entropy": 5.491646814346313,
      "epoch": 0.9628476950009726,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004912243643775137,
      "loss": 5.2046,
      "mean_token_accuracy": 0.18066623955965042,
      "num_tokens": 21475454.0,
      "step": 12375
    },
    {
      "entropy": 5.479880142211914,
      "epoch": 0.9632367243726901,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004912166977644936,
      "loss": 5.1216,
      "mean_token_accuracy": 0.18577821254730226,
      "num_tokens": 21483431.0,
      "step": 12380
    },
    {
      "entropy": 5.534336805343628,
      "epoch": 0.9636257537444077,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004912090278707293,
      "loss": 5.1959,
      "mean_token_accuracy": 0.18276784867048262,
      "num_tokens": 21491199.0,
      "step": 12385
    },
    {
      "entropy": 5.526461839675903,
      "epoch": 0.9640147831161253,
      "grad_norm": 1.0,
      "learning_rate": 0.000491201354696337,
      "loss": 5.2774,
      "mean_token_accuracy": 0.1810939073562622,
      "num_tokens": 21500339.0,
      "step": 12390
    },
    {
      "entropy": 5.52966628074646,
      "epoch": 0.9644038124878428,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004911936782414334,
      "loss": 5.2317,
      "mean_token_accuracy": 0.18086676448583602,
      "num_tokens": 21508945.0,
      "step": 12395
    },
    {
      "entropy": 5.56132869720459,
      "epoch": 0.9647928418595604,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004911859985061348,
      "loss": 5.1755,
      "mean_token_accuracy": 0.18211557418107988,
      "num_tokens": 21517649.0,
      "step": 12400
    },
    {
      "entropy": 5.546835660934448,
      "epoch": 0.965181871231278,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004911783154905577,
      "loss": 5.1929,
      "mean_token_accuracy": 0.18279263973236085,
      "num_tokens": 21526650.0,
      "step": 12405
    },
    {
      "entropy": 5.5483191967010494,
      "epoch": 0.9655709006029956,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004911706291948188,
      "loss": 5.2087,
      "mean_token_accuracy": 0.18675265163183213,
      "num_tokens": 21535401.0,
      "step": 12410
    },
    {
      "entropy": 5.588740444183349,
      "epoch": 0.9659599299747131,
      "grad_norm": 1.125,
      "learning_rate": 0.0004911629396190348,
      "loss": 5.239,
      "mean_token_accuracy": 0.18004606664180756,
      "num_tokens": 21543973.0,
      "step": 12415
    },
    {
      "entropy": 5.514178609848022,
      "epoch": 0.9663489593464306,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004911552467633221,
      "loss": 5.1748,
      "mean_token_accuracy": 0.18599266111850737,
      "num_tokens": 21552875.0,
      "step": 12420
    },
    {
      "entropy": 5.43506269454956,
      "epoch": 0.9667379887181482,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004911475506277976,
      "loss": 5.097,
      "mean_token_accuracy": 0.19554020315408707,
      "num_tokens": 21561469.0,
      "step": 12425
    },
    {
      "entropy": 5.512064027786255,
      "epoch": 0.9671270180898658,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000491139851212578,
      "loss": 5.1776,
      "mean_token_accuracy": 0.19110138714313507,
      "num_tokens": 21569523.0,
      "step": 12430
    },
    {
      "entropy": 5.463532543182373,
      "epoch": 0.9675160474615834,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004911321485177802,
      "loss": 5.1357,
      "mean_token_accuracy": 0.19027219265699385,
      "num_tokens": 21577700.0,
      "step": 12435
    },
    {
      "entropy": 5.555771207809448,
      "epoch": 0.967905076833301,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004911244425435212,
      "loss": 5.1839,
      "mean_token_accuracy": 0.18572254478931427,
      "num_tokens": 21586233.0,
      "step": 12440
    },
    {
      "entropy": 5.552957010269165,
      "epoch": 0.9682941062050184,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004911167332899176,
      "loss": 5.2306,
      "mean_token_accuracy": 0.1883271664381027,
      "num_tokens": 21595330.0,
      "step": 12445
    },
    {
      "entropy": 5.483322381973267,
      "epoch": 0.968683135576736,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004911090207570867,
      "loss": 5.1657,
      "mean_token_accuracy": 0.19029588699340821,
      "num_tokens": 21604522.0,
      "step": 12450
    },
    {
      "entropy": 5.564072036743164,
      "epoch": 0.9690721649484536,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004911013049451453,
      "loss": 5.139,
      "mean_token_accuracy": 0.18658262640237808,
      "num_tokens": 21613599.0,
      "step": 12455
    },
    {
      "entropy": 5.498353052139282,
      "epoch": 0.9694611943201712,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004910935858542106,
      "loss": 5.2214,
      "mean_token_accuracy": 0.17912377566099166,
      "num_tokens": 21622587.0,
      "step": 12460
    },
    {
      "entropy": 5.51771035194397,
      "epoch": 0.9698502236918888,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004910858634843997,
      "loss": 5.2132,
      "mean_token_accuracy": 0.17963476181030275,
      "num_tokens": 21630782.0,
      "step": 12465
    },
    {
      "entropy": 5.536013221740722,
      "epoch": 0.9702392530636063,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004910781378358297,
      "loss": 5.0895,
      "mean_token_accuracy": 0.19495368152856826,
      "num_tokens": 21639354.0,
      "step": 12470
    },
    {
      "entropy": 5.505559396743775,
      "epoch": 0.9706282824353238,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004910704089086178,
      "loss": 5.2707,
      "mean_token_accuracy": 0.1790216311812401,
      "num_tokens": 21648292.0,
      "step": 12475
    },
    {
      "entropy": 5.5507360935211185,
      "epoch": 0.9710173118070414,
      "grad_norm": 1.0,
      "learning_rate": 0.0004910626767028815,
      "loss": 5.1991,
      "mean_token_accuracy": 0.192071832716465,
      "num_tokens": 21657382.0,
      "step": 12480
    },
    {
      "entropy": 5.57020583152771,
      "epoch": 0.971406341178759,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004910549412187379,
      "loss": 5.199,
      "mean_token_accuracy": 0.18157916218042375,
      "num_tokens": 21666327.0,
      "step": 12485
    },
    {
      "entropy": 5.541747140884399,
      "epoch": 0.9717953705504766,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004910472024563045,
      "loss": 5.2172,
      "mean_token_accuracy": 0.18841089010238649,
      "num_tokens": 21674882.0,
      "step": 12490
    },
    {
      "entropy": 5.635183429718017,
      "epoch": 0.9721843999221941,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004910394604156987,
      "loss": 5.3012,
      "mean_token_accuracy": 0.17672445327043534,
      "num_tokens": 21683255.0,
      "step": 12495
    },
    {
      "entropy": 5.566134452819824,
      "epoch": 0.9725734292939117,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004910317150970379,
      "loss": 5.1721,
      "mean_token_accuracy": 0.1840107634663582,
      "num_tokens": 21692274.0,
      "step": 12500
    },
    {
      "entropy": 5.433190059661865,
      "epoch": 0.9729624586656292,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004910239665004397,
      "loss": 5.0927,
      "mean_token_accuracy": 0.1979975715279579,
      "num_tokens": 21700767.0,
      "step": 12505
    },
    {
      "entropy": 5.501289033889771,
      "epoch": 0.9733514880373468,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004910162146260216,
      "loss": 5.2378,
      "mean_token_accuracy": 0.18531136512756347,
      "num_tokens": 21709430.0,
      "step": 12510
    },
    {
      "entropy": 5.472946834564209,
      "epoch": 0.9737405174090644,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004910084594739013,
      "loss": 5.1406,
      "mean_token_accuracy": 0.1866540253162384,
      "num_tokens": 21717978.0,
      "step": 12515
    },
    {
      "entropy": 5.402873086929321,
      "epoch": 0.9741295467807819,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004910007010441964,
      "loss": 5.0066,
      "mean_token_accuracy": 0.20045141577720643,
      "num_tokens": 21726608.0,
      "step": 12520
    },
    {
      "entropy": 5.4793988227844235,
      "epoch": 0.9745185761524995,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004909929393370248,
      "loss": 5.1821,
      "mean_token_accuracy": 0.18620161712169647,
      "num_tokens": 21735402.0,
      "step": 12525
    },
    {
      "entropy": 5.477156686782837,
      "epoch": 0.9749076055242171,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004909851743525041,
      "loss": 5.1497,
      "mean_token_accuracy": 0.18945762813091277,
      "num_tokens": 21744620.0,
      "step": 12530
    },
    {
      "entropy": 5.525365495681763,
      "epoch": 0.9752966348959347,
      "grad_norm": 1.078125,
      "learning_rate": 0.000490977406090752,
      "loss": 5.2108,
      "mean_token_accuracy": 0.18496988862752914,
      "num_tokens": 21753270.0,
      "step": 12535
    },
    {
      "entropy": 5.587452125549317,
      "epoch": 0.9756856642676522,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004909696345518867,
      "loss": 5.2263,
      "mean_token_accuracy": 0.1811433404684067,
      "num_tokens": 21761477.0,
      "step": 12540
    },
    {
      "entropy": 5.43634443283081,
      "epoch": 0.9760746936393697,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004909618597360258,
      "loss": 5.1693,
      "mean_token_accuracy": 0.18882392048835756,
      "num_tokens": 21769925.0,
      "step": 12545
    },
    {
      "entropy": 5.533937740325928,
      "epoch": 0.9764637230110873,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004909540816432876,
      "loss": 5.219,
      "mean_token_accuracy": 0.19265445023775102,
      "num_tokens": 21778850.0,
      "step": 12550
    },
    {
      "entropy": 5.555978584289551,
      "epoch": 0.9768527523828049,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004909463002737897,
      "loss": 5.27,
      "mean_token_accuracy": 0.1795575201511383,
      "num_tokens": 21787408.0,
      "step": 12555
    },
    {
      "entropy": 5.57997407913208,
      "epoch": 0.9772417817545225,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004909385156276506,
      "loss": 5.2266,
      "mean_token_accuracy": 0.18589951992034912,
      "num_tokens": 21795967.0,
      "step": 12560
    },
    {
      "entropy": 5.522610282897949,
      "epoch": 0.9776308111262401,
      "grad_norm": 1.0,
      "learning_rate": 0.0004909307277049881,
      "loss": 5.1816,
      "mean_token_accuracy": 0.1849852353334427,
      "num_tokens": 21804472.0,
      "step": 12565
    },
    {
      "entropy": 5.462722682952881,
      "epoch": 0.9780198404979576,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004909229365059205,
      "loss": 5.1846,
      "mean_token_accuracy": 0.18201228529214858,
      "num_tokens": 21813233.0,
      "step": 12570
    },
    {
      "entropy": 5.462178087234497,
      "epoch": 0.9784088698696751,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004909151420305661,
      "loss": 5.0724,
      "mean_token_accuracy": 0.19895288050174714,
      "num_tokens": 21821413.0,
      "step": 12575
    },
    {
      "entropy": 5.520402288436889,
      "epoch": 0.9787978992413927,
      "grad_norm": 1.109375,
      "learning_rate": 0.000490907344279043,
      "loss": 5.1544,
      "mean_token_accuracy": 0.18895181715488435,
      "num_tokens": 21829222.0,
      "step": 12580
    },
    {
      "entropy": 5.4774010181427,
      "epoch": 0.9791869286131103,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004908995432514698,
      "loss": 5.1731,
      "mean_token_accuracy": 0.18411139249801636,
      "num_tokens": 21837033.0,
      "step": 12585
    },
    {
      "entropy": 5.573005867004395,
      "epoch": 0.9795759579848279,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004908917389479645,
      "loss": 5.2211,
      "mean_token_accuracy": 0.18206905722618102,
      "num_tokens": 21845464.0,
      "step": 12590
    },
    {
      "entropy": 5.4723039150238035,
      "epoch": 0.9799649873565455,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004908839313686456,
      "loss": 5.1767,
      "mean_token_accuracy": 0.18662499487400055,
      "num_tokens": 21853497.0,
      "step": 12595
    },
    {
      "entropy": 5.552912759780884,
      "epoch": 0.980354016728263,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004908761205136319,
      "loss": 5.2583,
      "mean_token_accuracy": 0.1832936555147171,
      "num_tokens": 21861710.0,
      "step": 12600
    },
    {
      "entropy": 5.58861608505249,
      "epoch": 0.9807430460999805,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004908683063830414,
      "loss": 5.1874,
      "mean_token_accuracy": 0.17955945581197738,
      "num_tokens": 21870472.0,
      "step": 12605
    },
    {
      "entropy": 5.586430597305298,
      "epoch": 0.9811320754716981,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004908604889769932,
      "loss": 5.2551,
      "mean_token_accuracy": 0.17836642563343047,
      "num_tokens": 21878287.0,
      "step": 12610
    },
    {
      "entropy": 5.585499858856201,
      "epoch": 0.9815211048434157,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004908526682956054,
      "loss": 5.2951,
      "mean_token_accuracy": 0.17450341433286667,
      "num_tokens": 21887587.0,
      "step": 12615
    },
    {
      "entropy": 5.5218127250671385,
      "epoch": 0.9819101342151333,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004908448443389972,
      "loss": 5.1317,
      "mean_token_accuracy": 0.19177911877632142,
      "num_tokens": 21895923.0,
      "step": 12620
    },
    {
      "entropy": 5.548227596282959,
      "epoch": 0.9822991635868508,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004908370171072869,
      "loss": 5.1644,
      "mean_token_accuracy": 0.19415276646614074,
      "num_tokens": 21904439.0,
      "step": 12625
    },
    {
      "entropy": 5.497436952590943,
      "epoch": 0.9826881929585684,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004908291866005933,
      "loss": 5.1406,
      "mean_token_accuracy": 0.18685363680124284,
      "num_tokens": 21912473.0,
      "step": 12630
    },
    {
      "entropy": 5.449019718170166,
      "epoch": 0.9830772223302859,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004908213528190355,
      "loss": 5.2062,
      "mean_token_accuracy": 0.18050660341978073,
      "num_tokens": 21921631.0,
      "step": 12635
    },
    {
      "entropy": 5.454296016693116,
      "epoch": 0.9834662517020035,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004908135157627321,
      "loss": 5.1113,
      "mean_token_accuracy": 0.19113940447568895,
      "num_tokens": 21929984.0,
      "step": 12640
    },
    {
      "entropy": 5.45877480506897,
      "epoch": 0.9838552810737211,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004908056754318022,
      "loss": 5.0537,
      "mean_token_accuracy": 0.19162175059318542,
      "num_tokens": 21937349.0,
      "step": 12645
    },
    {
      "entropy": 5.4563123226165775,
      "epoch": 0.9842443104454386,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004907978318263646,
      "loss": 5.1849,
      "mean_token_accuracy": 0.18215082287788392,
      "num_tokens": 21946176.0,
      "step": 12650
    },
    {
      "entropy": 5.520415449142456,
      "epoch": 0.9846333398171562,
      "grad_norm": 1.125,
      "learning_rate": 0.0004907899849465383,
      "loss": 5.1639,
      "mean_token_accuracy": 0.18894150108098984,
      "num_tokens": 21954645.0,
      "step": 12655
    },
    {
      "entropy": 5.542278242111206,
      "epoch": 0.9850223691888738,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004907821347924424,
      "loss": 5.2091,
      "mean_token_accuracy": 0.17826470732688904,
      "num_tokens": 21963290.0,
      "step": 12660
    },
    {
      "entropy": 5.509417390823364,
      "epoch": 0.9854113985605913,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004907742813641963,
      "loss": 5.2145,
      "mean_token_accuracy": 0.18907163441181182,
      "num_tokens": 21971005.0,
      "step": 12665
    },
    {
      "entropy": 5.4789299964904785,
      "epoch": 0.9858004279323089,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004907664246619187,
      "loss": 5.1281,
      "mean_token_accuracy": 0.19063290655612947,
      "num_tokens": 21980067.0,
      "step": 12670
    },
    {
      "entropy": 5.5096711158752445,
      "epoch": 0.9861894573040264,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004907585646857293,
      "loss": 5.1699,
      "mean_token_accuracy": 0.18219104558229446,
      "num_tokens": 21988511.0,
      "step": 12675
    },
    {
      "entropy": 5.500848007202149,
      "epoch": 0.986578486675744,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004907507014357467,
      "loss": 5.1478,
      "mean_token_accuracy": 0.18494538962841034,
      "num_tokens": 21997074.0,
      "step": 12680
    },
    {
      "entropy": 5.5642900466918945,
      "epoch": 0.9869675160474616,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004907428349120909,
      "loss": 5.2705,
      "mean_token_accuracy": 0.18180770725011824,
      "num_tokens": 22005412.0,
      "step": 12685
    },
    {
      "entropy": 5.4829648494720455,
      "epoch": 0.9873565454191792,
      "grad_norm": 1.0,
      "learning_rate": 0.0004907349651148809,
      "loss": 5.0766,
      "mean_token_accuracy": 0.195713546872139,
      "num_tokens": 22013588.0,
      "step": 12690
    },
    {
      "entropy": 5.516116333007813,
      "epoch": 0.9877455747908968,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004907270920442361,
      "loss": 5.2364,
      "mean_token_accuracy": 0.18529512733221054,
      "num_tokens": 22021669.0,
      "step": 12695
    },
    {
      "entropy": 5.468686294555664,
      "epoch": 0.9881346041626142,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004907192157002762,
      "loss": 5.0836,
      "mean_token_accuracy": 0.19446644484996795,
      "num_tokens": 22030295.0,
      "step": 12700
    },
    {
      "entropy": 5.543958616256714,
      "epoch": 0.9885236335343318,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004907113360831204,
      "loss": 5.2244,
      "mean_token_accuracy": 0.18010995984077455,
      "num_tokens": 22038924.0,
      "step": 12705
    },
    {
      "entropy": 5.516066074371338,
      "epoch": 0.9889126629060494,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004907034531928886,
      "loss": 5.1545,
      "mean_token_accuracy": 0.1901146113872528,
      "num_tokens": 22047586.0,
      "step": 12710
    },
    {
      "entropy": 5.488874101638794,
      "epoch": 0.989301692277767,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004906955670297001,
      "loss": 5.1409,
      "mean_token_accuracy": 0.18591995388269425,
      "num_tokens": 22055695.0,
      "step": 12715
    },
    {
      "entropy": 5.584533929824829,
      "epoch": 0.9896907216494846,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004906876775936746,
      "loss": 5.2954,
      "mean_token_accuracy": 0.17977903485298158,
      "num_tokens": 22064472.0,
      "step": 12720
    },
    {
      "entropy": 5.6011803150177,
      "epoch": 0.990079751021202,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004906797848849321,
      "loss": 5.2155,
      "mean_token_accuracy": 0.17661443650722503,
      "num_tokens": 22072726.0,
      "step": 12725
    },
    {
      "entropy": 5.541741275787354,
      "epoch": 0.9904687803929196,
      "grad_norm": 1.015625,
      "learning_rate": 0.000490671888903592,
      "loss": 5.1207,
      "mean_token_accuracy": 0.1934003710746765,
      "num_tokens": 22081074.0,
      "step": 12730
    },
    {
      "entropy": 5.4531578540802,
      "epoch": 0.9908578097646372,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004906639896497744,
      "loss": 5.1058,
      "mean_token_accuracy": 0.19296412467956542,
      "num_tokens": 22089924.0,
      "step": 12735
    },
    {
      "entropy": 5.518669176101684,
      "epoch": 0.9912468391363548,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004906560871235988,
      "loss": 5.1284,
      "mean_token_accuracy": 0.18981075584888457,
      "num_tokens": 22099088.0,
      "step": 12740
    },
    {
      "entropy": 5.6007359504699705,
      "epoch": 0.9916358685080724,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004906481813251854,
      "loss": 5.2724,
      "mean_token_accuracy": 0.18061267137527465,
      "num_tokens": 22108344.0,
      "step": 12745
    },
    {
      "entropy": 5.524589347839355,
      "epoch": 0.9920248978797899,
      "grad_norm": 1.125,
      "learning_rate": 0.0004906402722546542,
      "loss": 5.1715,
      "mean_token_accuracy": 0.18323837220668793,
      "num_tokens": 22116476.0,
      "step": 12750
    },
    {
      "entropy": 5.559820699691772,
      "epoch": 0.9924139272515075,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000490632359912125,
      "loss": 5.2822,
      "mean_token_accuracy": 0.17923307418823242,
      "num_tokens": 22125445.0,
      "step": 12755
    },
    {
      "entropy": 5.567450857162475,
      "epoch": 0.992802956623225,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000490624444297718,
      "loss": 5.3073,
      "mean_token_accuracy": 0.1826224982738495,
      "num_tokens": 22134886.0,
      "step": 12760
    },
    {
      "entropy": 5.490061807632446,
      "epoch": 0.9931919859949426,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004906165254115533,
      "loss": 5.1574,
      "mean_token_accuracy": 0.1883865162730217,
      "num_tokens": 22144521.0,
      "step": 12765
    },
    {
      "entropy": 5.479331541061401,
      "epoch": 0.9935810153666602,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000490608603253751,
      "loss": 5.1735,
      "mean_token_accuracy": 0.1924226775765419,
      "num_tokens": 22152754.0,
      "step": 12770
    },
    {
      "entropy": 5.534777879714966,
      "epoch": 0.9939700447383778,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004906006778244312,
      "loss": 5.1514,
      "mean_token_accuracy": 0.18945085257291794,
      "num_tokens": 22161711.0,
      "step": 12775
    },
    {
      "entropy": 5.566807126998901,
      "epoch": 0.9943590741100953,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004905927491237145,
      "loss": 5.281,
      "mean_token_accuracy": 0.18048788905143737,
      "num_tokens": 22169762.0,
      "step": 12780
    },
    {
      "entropy": 5.505855417251587,
      "epoch": 0.9947481034818129,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004905848171517208,
      "loss": 5.1272,
      "mean_token_accuracy": 0.19252506792545318,
      "num_tokens": 22178415.0,
      "step": 12785
    },
    {
      "entropy": 5.439107131958008,
      "epoch": 0.9951371328535304,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004905768819085706,
      "loss": 5.078,
      "mean_token_accuracy": 0.19238829463720322,
      "num_tokens": 22187643.0,
      "step": 12790
    },
    {
      "entropy": 5.408346033096313,
      "epoch": 0.995526162225248,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004905689433943846,
      "loss": 5.1147,
      "mean_token_accuracy": 0.1897441789507866,
      "num_tokens": 22196355.0,
      "step": 12795
    },
    {
      "entropy": 5.543659925460815,
      "epoch": 0.9959151915969656,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004905610016092828,
      "loss": 5.2246,
      "mean_token_accuracy": 0.1837788075208664,
      "num_tokens": 22204648.0,
      "step": 12800
    },
    {
      "entropy": 5.515322589874268,
      "epoch": 0.9963042209686831,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004905530565533859,
      "loss": 5.2061,
      "mean_token_accuracy": 0.18639307022094725,
      "num_tokens": 22212811.0,
      "step": 12805
    },
    {
      "entropy": 5.443220853805542,
      "epoch": 0.9966932503404007,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004905451082268146,
      "loss": 5.1686,
      "mean_token_accuracy": 0.18368617296218873,
      "num_tokens": 22220611.0,
      "step": 12810
    },
    {
      "entropy": 5.559359312057495,
      "epoch": 0.9970822797121183,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004905371566296891,
      "loss": 5.1888,
      "mean_token_accuracy": 0.1797492101788521,
      "num_tokens": 22229606.0,
      "step": 12815
    },
    {
      "entropy": 5.588633060455322,
      "epoch": 0.9974713090838359,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004905292017621305,
      "loss": 5.1649,
      "mean_token_accuracy": 0.18955003768205642,
      "num_tokens": 22238157.0,
      "step": 12820
    },
    {
      "entropy": 5.5039488792419435,
      "epoch": 0.9978603384555534,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004905212436242593,
      "loss": 5.1954,
      "mean_token_accuracy": 0.18170868456363679,
      "num_tokens": 22246696.0,
      "step": 12825
    },
    {
      "entropy": 5.5117762088775635,
      "epoch": 0.9982493678272709,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004905132822161962,
      "loss": 5.2252,
      "mean_token_accuracy": 0.18366746008396148,
      "num_tokens": 22255028.0,
      "step": 12830
    },
    {
      "entropy": 5.627627658843994,
      "epoch": 0.9986383971989885,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004905053175380621,
      "loss": 5.2792,
      "mean_token_accuracy": 0.1858209788799286,
      "num_tokens": 22263198.0,
      "step": 12835
    },
    {
      "entropy": 5.600813293457032,
      "epoch": 0.9990274265707061,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004904973495899778,
      "loss": 5.2751,
      "mean_token_accuracy": 0.18394966721534728,
      "num_tokens": 22273780.0,
      "step": 12840
    },
    {
      "entropy": 5.620848321914673,
      "epoch": 0.9994164559424237,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004904893783720642,
      "loss": 5.3109,
      "mean_token_accuracy": 0.1782144635915756,
      "num_tokens": 22281545.0,
      "step": 12845
    },
    {
      "entropy": 5.51906795501709,
      "epoch": 0.9998054853141413,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004904814038844424,
      "loss": 5.1915,
      "mean_token_accuracy": 0.18182659298181533,
      "num_tokens": 22290426.0,
      "step": 12850
    },
    {
      "entropy": 5.56677950753106,
      "epoch": 1.000155611748687,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004904734261272332,
      "loss": 5.2413,
      "mean_token_accuracy": 0.1858718345562617,
      "num_tokens": 22297475.0,
      "step": 12855
    },
    {
      "entropy": 5.466022634506226,
      "epoch": 1.0005446411204046,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004904654451005576,
      "loss": 4.9536,
      "mean_token_accuracy": 0.20117281824350358,
      "num_tokens": 22305784.0,
      "step": 12860
    },
    {
      "entropy": 5.477919054031372,
      "epoch": 1.0009336704921221,
      "grad_norm": 1.125,
      "learning_rate": 0.0004904574608045369,
      "loss": 5.0756,
      "mean_token_accuracy": 0.19339680671691895,
      "num_tokens": 22313951.0,
      "step": 12865
    },
    {
      "entropy": 5.612641286849976,
      "epoch": 1.0013226998638398,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004904494732392923,
      "loss": 5.1643,
      "mean_token_accuracy": 0.18693701922893524,
      "num_tokens": 22322509.0,
      "step": 12870
    },
    {
      "entropy": 5.525688934326172,
      "epoch": 1.0017117292355573,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004904414824049448,
      "loss": 5.1314,
      "mean_token_accuracy": 0.18645160943269729,
      "num_tokens": 22331475.0,
      "step": 12875
    },
    {
      "entropy": 5.485478448867798,
      "epoch": 1.0021007586072748,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004904334883016156,
      "loss": 5.0572,
      "mean_token_accuracy": 0.1911776155233383,
      "num_tokens": 22341114.0,
      "step": 12880
    },
    {
      "entropy": 5.53808741569519,
      "epoch": 1.0024897879789925,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004904254909294261,
      "loss": 5.1178,
      "mean_token_accuracy": 0.18769189715385437,
      "num_tokens": 22349199.0,
      "step": 12885
    },
    {
      "entropy": 5.53015456199646,
      "epoch": 1.00287881735071,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004904174902884978,
      "loss": 5.1046,
      "mean_token_accuracy": 0.18914662301540375,
      "num_tokens": 22358485.0,
      "step": 12890
    },
    {
      "entropy": 5.578337621688843,
      "epoch": 1.0032678467224276,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004904094863789519,
      "loss": 5.1485,
      "mean_token_accuracy": 0.18097074031829835,
      "num_tokens": 22366936.0,
      "step": 12895
    },
    {
      "entropy": 5.612206602096558,
      "epoch": 1.0036568760941451,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00049040147920091,
      "loss": 5.1267,
      "mean_token_accuracy": 0.18217714130878448,
      "num_tokens": 22376007.0,
      "step": 12900
    },
    {
      "entropy": 5.493472385406494,
      "epoch": 1.0040459054658626,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004903934687544933,
      "loss": 5.0469,
      "mean_token_accuracy": 0.202043779194355,
      "num_tokens": 22384747.0,
      "step": 12905
    },
    {
      "entropy": 5.508139944076538,
      "epoch": 1.0044349348375803,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004903854550398237,
      "loss": 5.0717,
      "mean_token_accuracy": 0.19737527519464493,
      "num_tokens": 22393730.0,
      "step": 12910
    },
    {
      "entropy": 5.569031858444214,
      "epoch": 1.0048239642092978,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004903774380570226,
      "loss": 5.0967,
      "mean_token_accuracy": 0.1873924732208252,
      "num_tokens": 22402121.0,
      "step": 12915
    },
    {
      "entropy": 5.5301741600036625,
      "epoch": 1.0052129935810155,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004903694178062117,
      "loss": 5.1042,
      "mean_token_accuracy": 0.18720748275518417,
      "num_tokens": 22410943.0,
      "step": 12920
    },
    {
      "entropy": 5.570095777511597,
      "epoch": 1.005602022952733,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004903613942875126,
      "loss": 5.1782,
      "mean_token_accuracy": 0.18708937615156174,
      "num_tokens": 22420137.0,
      "step": 12925
    },
    {
      "entropy": 5.549974632263184,
      "epoch": 1.0059910523244504,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004903533675010472,
      "loss": 5.115,
      "mean_token_accuracy": 0.18424922823905945,
      "num_tokens": 22429031.0,
      "step": 12930
    },
    {
      "entropy": 5.502824687957764,
      "epoch": 1.006380081696168,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004903453374469373,
      "loss": 5.1655,
      "mean_token_accuracy": 0.1848785996437073,
      "num_tokens": 22438735.0,
      "step": 12935
    },
    {
      "entropy": 5.571599340438842,
      "epoch": 1.0067691110678856,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004903373041253045,
      "loss": 5.151,
      "mean_token_accuracy": 0.1839245527982712,
      "num_tokens": 22447374.0,
      "step": 12940
    },
    {
      "entropy": 5.565502643585205,
      "epoch": 1.0071581404396033,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004903292675362709,
      "loss": 5.0864,
      "mean_token_accuracy": 0.19257624596357345,
      "num_tokens": 22455542.0,
      "step": 12945
    },
    {
      "entropy": 5.452770519256592,
      "epoch": 1.0075471698113208,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004903212276799584,
      "loss": 5.1313,
      "mean_token_accuracy": 0.18552372902631759,
      "num_tokens": 22464591.0,
      "step": 12950
    },
    {
      "entropy": 5.5261882781982425,
      "epoch": 1.0079361991830382,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004903131845564889,
      "loss": 5.0656,
      "mean_token_accuracy": 0.20058392733335495,
      "num_tokens": 22472315.0,
      "step": 12955
    },
    {
      "entropy": 5.550468683242798,
      "epoch": 1.008325228554756,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004903051381659847,
      "loss": 5.2082,
      "mean_token_accuracy": 0.18271043598651887,
      "num_tokens": 22480662.0,
      "step": 12960
    },
    {
      "entropy": 5.438960266113281,
      "epoch": 1.0087142579264734,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004902970885085677,
      "loss": 5.1083,
      "mean_token_accuracy": 0.18753167539834975,
      "num_tokens": 22489600.0,
      "step": 12965
    },
    {
      "entropy": 5.567250156402588,
      "epoch": 1.009103287298191,
      "grad_norm": 1.03125,
      "learning_rate": 0.00049028903558436,
      "loss": 5.1463,
      "mean_token_accuracy": 0.1862206295132637,
      "num_tokens": 22498114.0,
      "step": 12970
    },
    {
      "entropy": 5.499183320999146,
      "epoch": 1.0094923166699086,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004902809793934838,
      "loss": 4.9905,
      "mean_token_accuracy": 0.2024974122643471,
      "num_tokens": 22506349.0,
      "step": 12975
    },
    {
      "entropy": 5.520967245101929,
      "epoch": 1.009881346041626,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004902729199360615,
      "loss": 5.0746,
      "mean_token_accuracy": 0.1941869005560875,
      "num_tokens": 22514878.0,
      "step": 12980
    },
    {
      "entropy": 5.549259996414184,
      "epoch": 1.0102703754133437,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004902648572122153,
      "loss": 5.1022,
      "mean_token_accuracy": 0.17777488082647325,
      "num_tokens": 22522992.0,
      "step": 12985
    },
    {
      "entropy": 5.5899733543396,
      "epoch": 1.0106594047850612,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004902567912220674,
      "loss": 5.1627,
      "mean_token_accuracy": 0.18187704384326936,
      "num_tokens": 22531407.0,
      "step": 12990
    },
    {
      "entropy": 5.468643569946289,
      "epoch": 1.011048434156779,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004902487219657404,
      "loss": 5.0549,
      "mean_token_accuracy": 0.18903764486312866,
      "num_tokens": 22540255.0,
      "step": 12995
    },
    {
      "entropy": 5.498075294494629,
      "epoch": 1.0114374635284964,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004902406494433566,
      "loss": 5.0765,
      "mean_token_accuracy": 0.1950945094227791,
      "num_tokens": 22549139.0,
      "step": 13000
    },
    {
      "entropy": 5.545360517501831,
      "epoch": 1.0118264929002139,
      "grad_norm": 1.0,
      "learning_rate": 0.0004902325736550386,
      "loss": 5.095,
      "mean_token_accuracy": 0.19284993261098862,
      "num_tokens": 22557746.0,
      "step": 13005
    },
    {
      "entropy": 5.522401285171509,
      "epoch": 1.0122155222719316,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004902244946009088,
      "loss": 5.094,
      "mean_token_accuracy": 0.1892332538962364,
      "num_tokens": 22566695.0,
      "step": 13010
    },
    {
      "entropy": 5.48822021484375,
      "epoch": 1.012604551643649,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004902164122810899,
      "loss": 5.0919,
      "mean_token_accuracy": 0.19051875174045563,
      "num_tokens": 22575196.0,
      "step": 13015
    },
    {
      "entropy": 5.531784105300903,
      "epoch": 1.0129935810153667,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004902083266957045,
      "loss": 5.0828,
      "mean_token_accuracy": 0.19156216830015182,
      "num_tokens": 22583520.0,
      "step": 13020
    },
    {
      "entropy": 5.53649754524231,
      "epoch": 1.0133826103870842,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004902002378448753,
      "loss": 5.1522,
      "mean_token_accuracy": 0.18567444384098053,
      "num_tokens": 22591896.0,
      "step": 13025
    },
    {
      "entropy": 5.489823341369629,
      "epoch": 1.0137716397588017,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000490192145728725,
      "loss": 5.0658,
      "mean_token_accuracy": 0.19599737524986266,
      "num_tokens": 22600097.0,
      "step": 13030
    },
    {
      "entropy": 5.508410835266114,
      "epoch": 1.0141606691305194,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004901840503473764,
      "loss": 5.0157,
      "mean_token_accuracy": 0.19004160016775132,
      "num_tokens": 22607956.0,
      "step": 13035
    },
    {
      "entropy": 5.501716089248657,
      "epoch": 1.0145496985022369,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004901759517009522,
      "loss": 5.1006,
      "mean_token_accuracy": 0.18657678365707397,
      "num_tokens": 22616549.0,
      "step": 13040
    },
    {
      "entropy": 5.474237537384033,
      "epoch": 1.0149387278739546,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004901678497895755,
      "loss": 5.0296,
      "mean_token_accuracy": 0.18746725618839263,
      "num_tokens": 22625138.0,
      "step": 13045
    },
    {
      "entropy": 5.5160315990447994,
      "epoch": 1.015327757245672,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004901597446133692,
      "loss": 5.1063,
      "mean_token_accuracy": 0.18791033029556276,
      "num_tokens": 22633808.0,
      "step": 13050
    },
    {
      "entropy": 5.520609140396118,
      "epoch": 1.0157167866173895,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004901516361724564,
      "loss": 5.0802,
      "mean_token_accuracy": 0.19141926914453505,
      "num_tokens": 22642517.0,
      "step": 13055
    },
    {
      "entropy": 5.559398365020752,
      "epoch": 1.0161058159891072,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004901435244669597,
      "loss": 5.1351,
      "mean_token_accuracy": 0.18367298543453217,
      "num_tokens": 22651822.0,
      "step": 13060
    },
    {
      "entropy": 5.515314960479737,
      "epoch": 1.0164948453608247,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004901354094970025,
      "loss": 5.1082,
      "mean_token_accuracy": 0.19164054542779924,
      "num_tokens": 22660763.0,
      "step": 13065
    },
    {
      "entropy": 5.471757411956787,
      "epoch": 1.0168838747325424,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004901272912627081,
      "loss": 5.0853,
      "mean_token_accuracy": 0.18645355105400085,
      "num_tokens": 22669449.0,
      "step": 13070
    },
    {
      "entropy": 5.543807172775269,
      "epoch": 1.0172729041042599,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004901191697641992,
      "loss": 5.129,
      "mean_token_accuracy": 0.18621139377355575,
      "num_tokens": 22678212.0,
      "step": 13075
    },
    {
      "entropy": 5.490407848358155,
      "epoch": 1.0176619334759773,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004901110450015994,
      "loss": 5.0543,
      "mean_token_accuracy": 0.19825416952371597,
      "num_tokens": 22686509.0,
      "step": 13080
    },
    {
      "entropy": 5.460367679595947,
      "epoch": 1.018050962847695,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004901029169750319,
      "loss": 5.0973,
      "mean_token_accuracy": 0.1891048491001129,
      "num_tokens": 22695473.0,
      "step": 13085
    },
    {
      "entropy": 5.538697862625122,
      "epoch": 1.0184399922194125,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004900947856846201,
      "loss": 5.1557,
      "mean_token_accuracy": 0.18486980199813843,
      "num_tokens": 22704567.0,
      "step": 13090
    },
    {
      "entropy": 5.473818635940551,
      "epoch": 1.0188290215911302,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000490086651130487,
      "loss": 5.0369,
      "mean_token_accuracy": 0.19180958569049836,
      "num_tokens": 22713981.0,
      "step": 13095
    },
    {
      "entropy": 5.424191522598266,
      "epoch": 1.0192180509628477,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004900785133127566,
      "loss": 4.9708,
      "mean_token_accuracy": 0.20190931260585784,
      "num_tokens": 22722855.0,
      "step": 13100
    },
    {
      "entropy": 5.461099338531494,
      "epoch": 1.0196070803345652,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000490070372231552,
      "loss": 5.0901,
      "mean_token_accuracy": 0.1867565393447876,
      "num_tokens": 22732454.0,
      "step": 13105
    },
    {
      "entropy": 5.465971183776856,
      "epoch": 1.0199961097062828,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004900622278869968,
      "loss": 4.9585,
      "mean_token_accuracy": 0.20626488626003264,
      "num_tokens": 22741331.0,
      "step": 13110
    },
    {
      "entropy": 5.540648317337036,
      "epoch": 1.0203851390780003,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004900540802792146,
      "loss": 5.1485,
      "mean_token_accuracy": 0.18770444095134736,
      "num_tokens": 22749983.0,
      "step": 13115
    },
    {
      "entropy": 5.454984045028686,
      "epoch": 1.020774168449718,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000490045929408329,
      "loss": 5.1031,
      "mean_token_accuracy": 0.18563680350780487,
      "num_tokens": 22759112.0,
      "step": 13120
    },
    {
      "entropy": 5.503142499923706,
      "epoch": 1.0211631978214355,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004900377752744637,
      "loss": 5.1595,
      "mean_token_accuracy": 0.18962114155292512,
      "num_tokens": 22768456.0,
      "step": 13125
    },
    {
      "entropy": 5.514393949508667,
      "epoch": 1.021552227193153,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004900296178777426,
      "loss": 5.0451,
      "mean_token_accuracy": 0.19471228122711182,
      "num_tokens": 22777450.0,
      "step": 13130
    },
    {
      "entropy": 5.532179069519043,
      "epoch": 1.0219412565648707,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000490021457218289,
      "loss": 5.1085,
      "mean_token_accuracy": 0.19628282338380815,
      "num_tokens": 22785704.0,
      "step": 13135
    },
    {
      "entropy": 5.443779039382934,
      "epoch": 1.0223302859365881,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004900132932962272,
      "loss": 5.0923,
      "mean_token_accuracy": 0.18789590001106263,
      "num_tokens": 22794649.0,
      "step": 13140
    },
    {
      "entropy": 5.557936668395996,
      "epoch": 1.0227193153083058,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000490005126111681,
      "loss": 5.1911,
      "mean_token_accuracy": 0.1876332700252533,
      "num_tokens": 22803501.0,
      "step": 13145
    },
    {
      "entropy": 5.602274417877197,
      "epoch": 1.0231083446800233,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000489996955664774,
      "loss": 5.1245,
      "mean_token_accuracy": 0.18635753244161607,
      "num_tokens": 22811779.0,
      "step": 13150
    },
    {
      "entropy": 5.473807621002197,
      "epoch": 1.023497374051741,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004899887819556306,
      "loss": 5.1255,
      "mean_token_accuracy": 0.18699083626270294,
      "num_tokens": 22821108.0,
      "step": 13155
    },
    {
      "entropy": 5.546422863006592,
      "epoch": 1.0238864034234585,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004899806049843745,
      "loss": 5.1518,
      "mean_token_accuracy": 0.18772016763687133,
      "num_tokens": 22829369.0,
      "step": 13160
    },
    {
      "entropy": 5.524056148529053,
      "epoch": 1.024275432795176,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004899724247511299,
      "loss": 5.1001,
      "mean_token_accuracy": 0.18888365030288695,
      "num_tokens": 22838441.0,
      "step": 13165
    },
    {
      "entropy": 5.563871574401856,
      "epoch": 1.0246644621668937,
      "grad_norm": 0.984375,
      "learning_rate": 0.000489964241256021,
      "loss": 5.0687,
      "mean_token_accuracy": 0.1884496405720711,
      "num_tokens": 22847587.0,
      "step": 13170
    },
    {
      "entropy": 5.4519572257995605,
      "epoch": 1.0250534915386111,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004899560544991718,
      "loss": 5.041,
      "mean_token_accuracy": 0.19919597655534743,
      "num_tokens": 22856051.0,
      "step": 13175
    },
    {
      "entropy": 5.491789436340332,
      "epoch": 1.0254425209103288,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004899478644807068,
      "loss": 5.1257,
      "mean_token_accuracy": 0.18814466297626495,
      "num_tokens": 22864822.0,
      "step": 13180
    },
    {
      "entropy": 5.51660566329956,
      "epoch": 1.0258315502820463,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004899396712007498,
      "loss": 5.0863,
      "mean_token_accuracy": 0.19246355146169664,
      "num_tokens": 22873199.0,
      "step": 13185
    },
    {
      "entropy": 5.512306642532349,
      "epoch": 1.0262205796537638,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004899314746594256,
      "loss": 5.1485,
      "mean_token_accuracy": 0.1833132117986679,
      "num_tokens": 22881184.0,
      "step": 13190
    },
    {
      "entropy": 5.517817687988281,
      "epoch": 1.0266096090254815,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004899232748568584,
      "loss": 5.0621,
      "mean_token_accuracy": 0.19179992228746415,
      "num_tokens": 22890087.0,
      "step": 13195
    },
    {
      "entropy": 5.4983762264251705,
      "epoch": 1.026998638397199,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004899150717931724,
      "loss": 5.0791,
      "mean_token_accuracy": 0.1896684929728508,
      "num_tokens": 22899013.0,
      "step": 13200
    },
    {
      "entropy": 5.48650050163269,
      "epoch": 1.0273876677689167,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004899068654684924,
      "loss": 5.2336,
      "mean_token_accuracy": 0.18164822459220886,
      "num_tokens": 22908740.0,
      "step": 13205
    },
    {
      "entropy": 5.5651360034942625,
      "epoch": 1.0277766971406341,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004898986558829428,
      "loss": 5.1588,
      "mean_token_accuracy": 0.19015513509511947,
      "num_tokens": 22917502.0,
      "step": 13210
    },
    {
      "entropy": 5.502773380279541,
      "epoch": 1.0281657265123516,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004898904430366479,
      "loss": 5.0604,
      "mean_token_accuracy": 0.1925795704126358,
      "num_tokens": 22927076.0,
      "step": 13215
    },
    {
      "entropy": 5.571869993209839,
      "epoch": 1.0285547558840693,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004898822269297328,
      "loss": 5.1425,
      "mean_token_accuracy": 0.181087888777256,
      "num_tokens": 22936111.0,
      "step": 13220
    },
    {
      "entropy": 5.464747524261474,
      "epoch": 1.0289437852557868,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004898740075623218,
      "loss": 4.9933,
      "mean_token_accuracy": 0.2003425821661949,
      "num_tokens": 22944752.0,
      "step": 13225
    },
    {
      "entropy": 5.537498807907104,
      "epoch": 1.0293328146275045,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004898657849345399,
      "loss": 5.2204,
      "mean_token_accuracy": 0.18457093089818954,
      "num_tokens": 22952971.0,
      "step": 13230
    },
    {
      "entropy": 5.558270597457886,
      "epoch": 1.029721843999222,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004898575590465116,
      "loss": 5.1535,
      "mean_token_accuracy": 0.1928550645709038,
      "num_tokens": 22961257.0,
      "step": 13235
    },
    {
      "entropy": 5.5241330623626705,
      "epoch": 1.0301108733709394,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004898493298983619,
      "loss": 5.1898,
      "mean_token_accuracy": 0.18619605749845505,
      "num_tokens": 22970659.0,
      "step": 13240
    },
    {
      "entropy": 5.509057092666626,
      "epoch": 1.0304999027426571,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004898410974902155,
      "loss": 5.124,
      "mean_token_accuracy": 0.19114654809236525,
      "num_tokens": 22978830.0,
      "step": 13245
    },
    {
      "entropy": 5.5587382316589355,
      "epoch": 1.0308889321143746,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004898328618221974,
      "loss": 5.085,
      "mean_token_accuracy": 0.19281753599643708,
      "num_tokens": 22987100.0,
      "step": 13250
    },
    {
      "entropy": 5.524179553985595,
      "epoch": 1.0312779614860923,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004898246228944327,
      "loss": 5.1363,
      "mean_token_accuracy": 0.18367309421300887,
      "num_tokens": 22995749.0,
      "step": 13255
    },
    {
      "entropy": 5.592971897125244,
      "epoch": 1.0316669908578098,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004898163807070461,
      "loss": 5.1875,
      "mean_token_accuracy": 0.19096410125494004,
      "num_tokens": 23004410.0,
      "step": 13260
    },
    {
      "entropy": 5.548664712905884,
      "epoch": 1.0320560202295272,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004898081352601629,
      "loss": 5.0474,
      "mean_token_accuracy": 0.19274751991033554,
      "num_tokens": 23012935.0,
      "step": 13265
    },
    {
      "entropy": 5.493165016174316,
      "epoch": 1.032445049601245,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004897998865539082,
      "loss": 5.0323,
      "mean_token_accuracy": 0.19124120622873306,
      "num_tokens": 23021769.0,
      "step": 13270
    },
    {
      "entropy": 5.479048109054565,
      "epoch": 1.0328340789729624,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004897916345884069,
      "loss": 5.044,
      "mean_token_accuracy": 0.19254784882068635,
      "num_tokens": 23030915.0,
      "step": 13275
    },
    {
      "entropy": 5.502058506011963,
      "epoch": 1.0332231083446801,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004897833793637847,
      "loss": 5.168,
      "mean_token_accuracy": 0.18352843970060348,
      "num_tokens": 23039422.0,
      "step": 13280
    },
    {
      "entropy": 5.5198094844818115,
      "epoch": 1.0336121377163976,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004897751208801665,
      "loss": 5.2122,
      "mean_token_accuracy": 0.18021536469459534,
      "num_tokens": 23048485.0,
      "step": 13285
    },
    {
      "entropy": 5.511947679519653,
      "epoch": 1.034001167088115,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004897668591376777,
      "loss": 5.0777,
      "mean_token_accuracy": 0.2034519776701927,
      "num_tokens": 23057506.0,
      "step": 13290
    },
    {
      "entropy": 5.493970155715942,
      "epoch": 1.0343901964598328,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004897585941364436,
      "loss": 4.9459,
      "mean_token_accuracy": 0.2041737452149391,
      "num_tokens": 23065430.0,
      "step": 13295
    },
    {
      "entropy": 5.425747871398926,
      "epoch": 1.0347792258315502,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004897503258765896,
      "loss": 5.0636,
      "mean_token_accuracy": 0.18638816177845002,
      "num_tokens": 23074533.0,
      "step": 13300
    },
    {
      "entropy": 5.505717754364014,
      "epoch": 1.035168255203268,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004897420543582413,
      "loss": 5.0626,
      "mean_token_accuracy": 0.18895838409662247,
      "num_tokens": 23083020.0,
      "step": 13305
    },
    {
      "entropy": 5.48665075302124,
      "epoch": 1.0355572845749854,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000489733779581524,
      "loss": 5.1033,
      "mean_token_accuracy": 0.19441334754228592,
      "num_tokens": 23091992.0,
      "step": 13310
    },
    {
      "entropy": 5.427390813827515,
      "epoch": 1.035946313946703,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004897255015465635,
      "loss": 5.0865,
      "mean_token_accuracy": 0.1928173378109932,
      "num_tokens": 23100411.0,
      "step": 13315
    },
    {
      "entropy": 5.494736099243164,
      "epoch": 1.0363353433184206,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004897172202534853,
      "loss": 5.1727,
      "mean_token_accuracy": 0.18805998265743257,
      "num_tokens": 23108867.0,
      "step": 13320
    },
    {
      "entropy": 5.515126895904541,
      "epoch": 1.036724372690138,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000489708935702415,
      "loss": 5.0151,
      "mean_token_accuracy": 0.1937367007136345,
      "num_tokens": 23116692.0,
      "step": 13325
    },
    {
      "entropy": 5.513948106765747,
      "epoch": 1.0371134020618558,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004897006478934784,
      "loss": 5.0907,
      "mean_token_accuracy": 0.19035588949918747,
      "num_tokens": 23125138.0,
      "step": 13330
    },
    {
      "entropy": 5.459046840667725,
      "epoch": 1.0375024314335732,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004896923568268011,
      "loss": 5.0969,
      "mean_token_accuracy": 0.185422320663929,
      "num_tokens": 23133263.0,
      "step": 13335
    },
    {
      "entropy": 5.550583362579346,
      "epoch": 1.0378914608052907,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000489684062502509,
      "loss": 5.0865,
      "mean_token_accuracy": 0.1903218924999237,
      "num_tokens": 23141984.0,
      "step": 13340
    },
    {
      "entropy": 5.508817577362061,
      "epoch": 1.0382804901770084,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004896757649207281,
      "loss": 5.0624,
      "mean_token_accuracy": 0.19766537994146346,
      "num_tokens": 23151598.0,
      "step": 13345
    },
    {
      "entropy": 5.531688833236695,
      "epoch": 1.0386695195487259,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000489667464081584,
      "loss": 5.2037,
      "mean_token_accuracy": 0.1841859132051468,
      "num_tokens": 23159737.0,
      "step": 13350
    },
    {
      "entropy": 5.484114646911621,
      "epoch": 1.0390585489204436,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004896591599852029,
      "loss": 5.0498,
      "mean_token_accuracy": 0.19665172100067138,
      "num_tokens": 23168852.0,
      "step": 13355
    },
    {
      "entropy": 5.560707855224609,
      "epoch": 1.039447578292161,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004896508526317107,
      "loss": 5.1241,
      "mean_token_accuracy": 0.1893393650650978,
      "num_tokens": 23178227.0,
      "step": 13360
    },
    {
      "entropy": 5.488317632675171,
      "epoch": 1.0398366076638785,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004896425420212334,
      "loss": 5.0434,
      "mean_token_accuracy": 0.19881423562765121,
      "num_tokens": 23187286.0,
      "step": 13365
    },
    {
      "entropy": 5.531023168563843,
      "epoch": 1.0402256370355962,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004896342281538973,
      "loss": 5.0992,
      "mean_token_accuracy": 0.1970721498131752,
      "num_tokens": 23196163.0,
      "step": 13370
    },
    {
      "entropy": 5.533740091323852,
      "epoch": 1.0406146664073137,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004896259110298283,
      "loss": 5.0634,
      "mean_token_accuracy": 0.19692256897687913,
      "num_tokens": 23203742.0,
      "step": 13375
    },
    {
      "entropy": 5.567911958694458,
      "epoch": 1.0410036957790314,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004896175906491528,
      "loss": 5.218,
      "mean_token_accuracy": 0.1864864781498909,
      "num_tokens": 23212860.0,
      "step": 13380
    },
    {
      "entropy": 5.559319067001343,
      "epoch": 1.0413927251507489,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004896092670119968,
      "loss": 5.0921,
      "mean_token_accuracy": 0.18651237934827805,
      "num_tokens": 23222558.0,
      "step": 13385
    },
    {
      "entropy": 5.534350395202637,
      "epoch": 1.0417817545224664,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004896009401184869,
      "loss": 5.1351,
      "mean_token_accuracy": 0.1875055328011513,
      "num_tokens": 23231413.0,
      "step": 13390
    },
    {
      "entropy": 5.499296569824219,
      "epoch": 1.042170783894184,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004895926099687493,
      "loss": 5.085,
      "mean_token_accuracy": 0.19679968506097795,
      "num_tokens": 23240783.0,
      "step": 13395
    },
    {
      "entropy": 5.498710060119629,
      "epoch": 1.0425598132659015,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004895842765629104,
      "loss": 5.1857,
      "mean_token_accuracy": 0.1896827057003975,
      "num_tokens": 23249125.0,
      "step": 13400
    },
    {
      "entropy": 5.522385931015014,
      "epoch": 1.0429488426376192,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004895759399010966,
      "loss": 5.0431,
      "mean_token_accuracy": 0.1913505420088768,
      "num_tokens": 23257793.0,
      "step": 13405
    },
    {
      "entropy": 5.439147233963013,
      "epoch": 1.0433378720093367,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004895675999834345,
      "loss": 5.0299,
      "mean_token_accuracy": 0.1957654282450676,
      "num_tokens": 23266361.0,
      "step": 13410
    },
    {
      "entropy": 5.4606969356536865,
      "epoch": 1.0437269013810542,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004895592568100505,
      "loss": 5.0011,
      "mean_token_accuracy": 0.19829651713371277,
      "num_tokens": 23274914.0,
      "step": 13415
    },
    {
      "entropy": 5.482345724105835,
      "epoch": 1.0441159307527719,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004895509103810714,
      "loss": 5.082,
      "mean_token_accuracy": 0.19065336883068085,
      "num_tokens": 23283067.0,
      "step": 13420
    },
    {
      "entropy": 5.521182489395142,
      "epoch": 1.0445049601244893,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004895425606966237,
      "loss": 5.0989,
      "mean_token_accuracy": 0.18766561150550842,
      "num_tokens": 23291114.0,
      "step": 13425
    },
    {
      "entropy": 5.4168150424957275,
      "epoch": 1.044893989496207,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000489534207756834,
      "loss": 4.9732,
      "mean_token_accuracy": 0.20178361237049103,
      "num_tokens": 23300525.0,
      "step": 13430
    },
    {
      "entropy": 5.474925184249878,
      "epoch": 1.0452830188679245,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004895258515618293,
      "loss": 5.0576,
      "mean_token_accuracy": 0.1925063982605934,
      "num_tokens": 23308984.0,
      "step": 13435
    },
    {
      "entropy": 5.526336145401001,
      "epoch": 1.045672048239642,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004895174921117362,
      "loss": 5.1704,
      "mean_token_accuracy": 0.1820365607738495,
      "num_tokens": 23318137.0,
      "step": 13440
    },
    {
      "entropy": 5.5889527797698975,
      "epoch": 1.0460610776113597,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004895091294066816,
      "loss": 5.2392,
      "mean_token_accuracy": 0.18011942356824875,
      "num_tokens": 23326276.0,
      "step": 13445
    },
    {
      "entropy": 5.515794181823731,
      "epoch": 1.0464501069830772,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004895007634467924,
      "loss": 5.1056,
      "mean_token_accuracy": 0.1882613018155098,
      "num_tokens": 23334578.0,
      "step": 13450
    },
    {
      "entropy": 5.520166873931885,
      "epoch": 1.0468391363547949,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004894923942321955,
      "loss": 5.1299,
      "mean_token_accuracy": 0.18824388831853867,
      "num_tokens": 23344173.0,
      "step": 13455
    },
    {
      "entropy": 5.482270193099976,
      "epoch": 1.0472281657265123,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004894840217630179,
      "loss": 4.9793,
      "mean_token_accuracy": 0.19356620013713838,
      "num_tokens": 23353098.0,
      "step": 13460
    },
    {
      "entropy": 5.557977151870728,
      "epoch": 1.0476171950982298,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004894756460393868,
      "loss": 5.1257,
      "mean_token_accuracy": 0.19226764738559723,
      "num_tokens": 23362364.0,
      "step": 13465
    },
    {
      "entropy": 5.550920391082764,
      "epoch": 1.0480062244699475,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000489467267061429,
      "loss": 5.1416,
      "mean_token_accuracy": 0.18445586711168288,
      "num_tokens": 23371903.0,
      "step": 13470
    },
    {
      "entropy": 5.505354976654052,
      "epoch": 1.048395253841665,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004894588848292718,
      "loss": 5.0763,
      "mean_token_accuracy": 0.1984793171286583,
      "num_tokens": 23381711.0,
      "step": 13475
    },
    {
      "entropy": 5.490509796142578,
      "epoch": 1.0487842832133827,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004894504993430425,
      "loss": 5.0218,
      "mean_token_accuracy": 0.1924188494682312,
      "num_tokens": 23389812.0,
      "step": 13480
    },
    {
      "entropy": 5.517077541351318,
      "epoch": 1.0491733125851002,
      "grad_norm": 1.0625,
      "learning_rate": 0.000489442110602868,
      "loss": 5.0654,
      "mean_token_accuracy": 0.2000178202986717,
      "num_tokens": 23399231.0,
      "step": 13485
    },
    {
      "entropy": 5.454672956466675,
      "epoch": 1.0495623419568176,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004894337186088759,
      "loss": 5.0567,
      "mean_token_accuracy": 0.1873787373304367,
      "num_tokens": 23407764.0,
      "step": 13490
    },
    {
      "entropy": 5.476378631591797,
      "epoch": 1.0499513713285353,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004894253233611934,
      "loss": 5.0048,
      "mean_token_accuracy": 0.19392048418521882,
      "num_tokens": 23416368.0,
      "step": 13495
    },
    {
      "entropy": 5.606404256820679,
      "epoch": 1.0503404007002528,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000489416924859948,
      "loss": 5.2472,
      "mean_token_accuracy": 0.18424750566482545,
      "num_tokens": 23425415.0,
      "step": 13500
    },
    {
      "entropy": 5.506924819946289,
      "epoch": 1.0507294300719705,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000489408523105267,
      "loss": 5.054,
      "mean_token_accuracy": 0.19067147970199586,
      "num_tokens": 23433435.0,
      "step": 13505
    },
    {
      "entropy": 5.532734060287476,
      "epoch": 1.051118459443688,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004894001180972779,
      "loss": 5.1172,
      "mean_token_accuracy": 0.18772961497306823,
      "num_tokens": 23442002.0,
      "step": 13510
    },
    {
      "entropy": 5.436568737030029,
      "epoch": 1.0515074888154055,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004893917098361082,
      "loss": 4.9125,
      "mean_token_accuracy": 0.19587039947509766,
      "num_tokens": 23450437.0,
      "step": 13515
    },
    {
      "entropy": 5.415784120559692,
      "epoch": 1.0518965181871232,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004893832983218856,
      "loss": 5.0112,
      "mean_token_accuracy": 0.19686825275421144,
      "num_tokens": 23458755.0,
      "step": 13520
    },
    {
      "entropy": 5.50698299407959,
      "epoch": 1.0522855475588406,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004893748835547377,
      "loss": 5.137,
      "mean_token_accuracy": 0.18671559989452363,
      "num_tokens": 23467138.0,
      "step": 13525
    },
    {
      "entropy": 5.480596733093262,
      "epoch": 1.0526745769305583,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004893664655347921,
      "loss": 5.0047,
      "mean_token_accuracy": 0.19173648506402968,
      "num_tokens": 23475494.0,
      "step": 13530
    },
    {
      "entropy": 5.440220308303833,
      "epoch": 1.0530636063022758,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004893580442621767,
      "loss": 4.9536,
      "mean_token_accuracy": 0.2004094675183296,
      "num_tokens": 23484215.0,
      "step": 13535
    },
    {
      "entropy": 5.4320900440216064,
      "epoch": 1.0534526356739935,
      "grad_norm": 1.015625,
      "learning_rate": 0.000489349619737019,
      "loss": 5.0417,
      "mean_token_accuracy": 0.19129815548658372,
      "num_tokens": 23492604.0,
      "step": 13540
    },
    {
      "entropy": 5.597353792190551,
      "epoch": 1.053841665045711,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000489341191959447,
      "loss": 5.2524,
      "mean_token_accuracy": 0.1829664409160614,
      "num_tokens": 23501817.0,
      "step": 13545
    },
    {
      "entropy": 5.59983081817627,
      "epoch": 1.0542306944174284,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004893327609295887,
      "loss": 5.1551,
      "mean_token_accuracy": 0.18885855972766877,
      "num_tokens": 23509996.0,
      "step": 13550
    },
    {
      "entropy": 5.437367296218872,
      "epoch": 1.0546197237891461,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004893243266475719,
      "loss": 4.906,
      "mean_token_accuracy": 0.21002590507268906,
      "num_tokens": 23518316.0,
      "step": 13555
    },
    {
      "entropy": 5.440684366226196,
      "epoch": 1.0550087531608636,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004893158891135245,
      "loss": 5.1197,
      "mean_token_accuracy": 0.18702333569526672,
      "num_tokens": 23526851.0,
      "step": 13560
    },
    {
      "entropy": 5.521849632263184,
      "epoch": 1.055397782532581,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004893074483275746,
      "loss": 5.0385,
      "mean_token_accuracy": 0.19291110932826996,
      "num_tokens": 23534757.0,
      "step": 13565
    },
    {
      "entropy": 5.487427759170532,
      "epoch": 1.0557868119042988,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004892990042898503,
      "loss": 5.0986,
      "mean_token_accuracy": 0.18691545426845552,
      "num_tokens": 23544102.0,
      "step": 13570
    },
    {
      "entropy": 5.4398791790008545,
      "epoch": 1.0561758412760163,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004892905570004798,
      "loss": 5.1372,
      "mean_token_accuracy": 0.18297408372163773,
      "num_tokens": 23552663.0,
      "step": 13575
    },
    {
      "entropy": 5.488240575790405,
      "epoch": 1.056564870647734,
      "grad_norm": 1.0625,
      "learning_rate": 0.000489282106459591,
      "loss": 5.0162,
      "mean_token_accuracy": 0.19762213081121444,
      "num_tokens": 23561144.0,
      "step": 13580
    },
    {
      "entropy": 5.472207403182983,
      "epoch": 1.0569539000194514,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004892736526673124,
      "loss": 5.0116,
      "mean_token_accuracy": 0.19642287194728852,
      "num_tokens": 23569833.0,
      "step": 13585
    },
    {
      "entropy": 5.503046989440918,
      "epoch": 1.0573429293911691,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004892651956237721,
      "loss": 5.079,
      "mean_token_accuracy": 0.18773457407951355,
      "num_tokens": 23578643.0,
      "step": 13590
    },
    {
      "entropy": 5.530203819274902,
      "epoch": 1.0577319587628866,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004892567353290986,
      "loss": 5.1573,
      "mean_token_accuracy": 0.18732548505067825,
      "num_tokens": 23587324.0,
      "step": 13595
    },
    {
      "entropy": 5.463409662246704,
      "epoch": 1.058120988134604,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004892482717834201,
      "loss": 5.0953,
      "mean_token_accuracy": 0.18745317459106445,
      "num_tokens": 23595670.0,
      "step": 13600
    },
    {
      "entropy": 5.474406385421753,
      "epoch": 1.0585100175063218,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004892398049868651,
      "loss": 4.9836,
      "mean_token_accuracy": 0.20680466294288635,
      "num_tokens": 23603833.0,
      "step": 13605
    },
    {
      "entropy": 5.4800210952758786,
      "epoch": 1.0588990468780393,
      "grad_norm": 1.03125,
      "learning_rate": 0.000489231334939562,
      "loss": 5.0924,
      "mean_token_accuracy": 0.1919196754693985,
      "num_tokens": 23612815.0,
      "step": 13610
    },
    {
      "entropy": 5.531673955917358,
      "epoch": 1.059288076249757,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004892228616416395,
      "loss": 5.0934,
      "mean_token_accuracy": 0.18803882151842116,
      "num_tokens": 23621178.0,
      "step": 13615
    },
    {
      "entropy": 5.51334023475647,
      "epoch": 1.0596771056214744,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004892143850932259,
      "loss": 5.2852,
      "mean_token_accuracy": 0.18177369832992554,
      "num_tokens": 23630681.0,
      "step": 13620
    },
    {
      "entropy": 5.521235799789428,
      "epoch": 1.060066134993192,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048920590529445,
      "loss": 5.0758,
      "mean_token_accuracy": 0.1929500237107277,
      "num_tokens": 23639374.0,
      "step": 13625
    },
    {
      "entropy": 5.494675731658935,
      "epoch": 1.0604551643649096,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004891974222454406,
      "loss": 5.0764,
      "mean_token_accuracy": 0.1986416220664978,
      "num_tokens": 23648217.0,
      "step": 13630
    },
    {
      "entropy": 5.5171043395996096,
      "epoch": 1.060844193736627,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004891889359463261,
      "loss": 5.0549,
      "mean_token_accuracy": 0.19728301018476485,
      "num_tokens": 23656688.0,
      "step": 13635
    },
    {
      "entropy": 5.511386680603027,
      "epoch": 1.0612332231083448,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004891804463972354,
      "loss": 5.1402,
      "mean_token_accuracy": 0.1892149731516838,
      "num_tokens": 23666646.0,
      "step": 13640
    },
    {
      "entropy": 5.4855513095855715,
      "epoch": 1.0616222524800623,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004891719535982974,
      "loss": 5.0439,
      "mean_token_accuracy": 0.1967901408672333,
      "num_tokens": 23675278.0,
      "step": 13645
    },
    {
      "entropy": 5.5024559020996096,
      "epoch": 1.0620112818517797,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004891634575496408,
      "loss": 5.1473,
      "mean_token_accuracy": 0.1896624118089676,
      "num_tokens": 23684263.0,
      "step": 13650
    },
    {
      "entropy": 5.479881477355957,
      "epoch": 1.0624003112234974,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004891549582513946,
      "loss": 4.9724,
      "mean_token_accuracy": 0.1970535397529602,
      "num_tokens": 23692249.0,
      "step": 13655
    },
    {
      "entropy": 5.552881669998169,
      "epoch": 1.062789340595215,
      "grad_norm": 1.046875,
      "learning_rate": 0.000489146455703688,
      "loss": 5.1951,
      "mean_token_accuracy": 0.18986672013998032,
      "num_tokens": 23701082.0,
      "step": 13660
    },
    {
      "entropy": 5.5221329689025875,
      "epoch": 1.0631783699669326,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004891379499066495,
      "loss": 5.1424,
      "mean_token_accuracy": 0.18091514855623245,
      "num_tokens": 23710277.0,
      "step": 13665
    },
    {
      "entropy": 5.510159111022949,
      "epoch": 1.06356739933865,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004891294408604087,
      "loss": 5.1115,
      "mean_token_accuracy": 0.18644489198923112,
      "num_tokens": 23719131.0,
      "step": 13670
    },
    {
      "entropy": 5.471171808242798,
      "epoch": 1.0639564287103676,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004891209285650942,
      "loss": 5.0479,
      "mean_token_accuracy": 0.18626420199871063,
      "num_tokens": 23727674.0,
      "step": 13675
    },
    {
      "entropy": 5.525305461883545,
      "epoch": 1.0643454580820852,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004891124130208355,
      "loss": 5.1744,
      "mean_token_accuracy": 0.1840501219034195,
      "num_tokens": 23736181.0,
      "step": 13680
    },
    {
      "entropy": 5.4968729496002195,
      "epoch": 1.0647344874538027,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004891038942277618,
      "loss": 5.0578,
      "mean_token_accuracy": 0.19397028535604477,
      "num_tokens": 23744552.0,
      "step": 13685
    },
    {
      "entropy": 5.592499303817749,
      "epoch": 1.0651235168255204,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004890953721860022,
      "loss": 5.1889,
      "mean_token_accuracy": 0.1824129894375801,
      "num_tokens": 23753755.0,
      "step": 13690
    },
    {
      "entropy": 5.536994409561157,
      "epoch": 1.065512546197238,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004890868468956862,
      "loss": 5.0808,
      "mean_token_accuracy": 0.1964101731777191,
      "num_tokens": 23761759.0,
      "step": 13695
    },
    {
      "entropy": 5.477020263671875,
      "epoch": 1.0659015755689554,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000489078318356943,
      "loss": 5.082,
      "mean_token_accuracy": 0.18774908035993576,
      "num_tokens": 23769952.0,
      "step": 13700
    },
    {
      "entropy": 5.451444292068482,
      "epoch": 1.066290604940673,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004890697865699021,
      "loss": 5.1177,
      "mean_token_accuracy": 0.1944337710738182,
      "num_tokens": 23778045.0,
      "step": 13705
    },
    {
      "entropy": 5.545817518234253,
      "epoch": 1.0666796343123905,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004890612515346929,
      "loss": 5.0632,
      "mean_token_accuracy": 0.1906737670302391,
      "num_tokens": 23787477.0,
      "step": 13710
    },
    {
      "entropy": 5.506264114379883,
      "epoch": 1.0670686636841082,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004890527132514448,
      "loss": 5.0553,
      "mean_token_accuracy": 0.2003220483660698,
      "num_tokens": 23795905.0,
      "step": 13715
    },
    {
      "entropy": 5.569752883911133,
      "epoch": 1.0674576930558257,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004890441717202876,
      "loss": 5.1334,
      "mean_token_accuracy": 0.1918783515691757,
      "num_tokens": 23804339.0,
      "step": 13720
    },
    {
      "entropy": 5.521624755859375,
      "epoch": 1.0678467224275432,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004890356269413507,
      "loss": 5.1465,
      "mean_token_accuracy": 0.1847417265176773,
      "num_tokens": 23813032.0,
      "step": 13725
    },
    {
      "entropy": 5.490877294540406,
      "epoch": 1.068235751799261,
      "grad_norm": 1.078125,
      "learning_rate": 0.000489027078914764,
      "loss": 5.0029,
      "mean_token_accuracy": 0.20671516060829162,
      "num_tokens": 23822047.0,
      "step": 13730
    },
    {
      "entropy": 5.414293670654297,
      "epoch": 1.0686247811709784,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004890185276406569,
      "loss": 4.9937,
      "mean_token_accuracy": 0.19433670192956926,
      "num_tokens": 23831220.0,
      "step": 13735
    },
    {
      "entropy": 5.48832859992981,
      "epoch": 1.069013810542696,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004890099731191592,
      "loss": 5.1464,
      "mean_token_accuracy": 0.18784899562597274,
      "num_tokens": 23840118.0,
      "step": 13740
    },
    {
      "entropy": 5.535661888122559,
      "epoch": 1.0694028399144135,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000489001415350401,
      "loss": 5.111,
      "mean_token_accuracy": 0.19171647876501083,
      "num_tokens": 23848826.0,
      "step": 13745
    },
    {
      "entropy": 5.589157009124756,
      "epoch": 1.069791869286131,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004889928543345118,
      "loss": 5.2251,
      "mean_token_accuracy": 0.1858771726489067,
      "num_tokens": 23857462.0,
      "step": 13750
    },
    {
      "entropy": 5.483942461013794,
      "epoch": 1.0701808986578487,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004889842900716217,
      "loss": 5.0924,
      "mean_token_accuracy": 0.1896417871117592,
      "num_tokens": 23866842.0,
      "step": 13755
    },
    {
      "entropy": 5.514560699462891,
      "epoch": 1.0705699280295662,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004889757225618606,
      "loss": 5.1152,
      "mean_token_accuracy": 0.18408040404319764,
      "num_tokens": 23875079.0,
      "step": 13760
    },
    {
      "entropy": 5.556627035140991,
      "epoch": 1.0709589574012839,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004889671518053584,
      "loss": 5.0906,
      "mean_token_accuracy": 0.1864870995283127,
      "num_tokens": 23883963.0,
      "step": 13765
    },
    {
      "entropy": 5.517716646194458,
      "epoch": 1.0713479867730014,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004889585778022453,
      "loss": 5.0614,
      "mean_token_accuracy": 0.19662969559431076,
      "num_tokens": 23892709.0,
      "step": 13770
    },
    {
      "entropy": 5.515289831161499,
      "epoch": 1.0717370161447188,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004889500005526514,
      "loss": 5.1955,
      "mean_token_accuracy": 0.18402273803949357,
      "num_tokens": 23901124.0,
      "step": 13775
    },
    {
      "entropy": 5.528458452224731,
      "epoch": 1.0721260455164365,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004889414200567067,
      "loss": 5.1371,
      "mean_token_accuracy": 0.1809419646859169,
      "num_tokens": 23909442.0,
      "step": 13780
    },
    {
      "entropy": 5.517196416854858,
      "epoch": 1.072515074888154,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004889328363145415,
      "loss": 5.1056,
      "mean_token_accuracy": 0.19546132981777192,
      "num_tokens": 23918143.0,
      "step": 13785
    },
    {
      "entropy": 5.535054016113281,
      "epoch": 1.0729041042598717,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004889242493262859,
      "loss": 5.0515,
      "mean_token_accuracy": 0.19446122646331787,
      "num_tokens": 23926432.0,
      "step": 13790
    },
    {
      "entropy": 5.536283254623413,
      "epoch": 1.0732931336315892,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004889156590920704,
      "loss": 5.1986,
      "mean_token_accuracy": 0.17785345166921615,
      "num_tokens": 23935362.0,
      "step": 13795
    },
    {
      "entropy": 5.433492136001587,
      "epoch": 1.0736821630033067,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004889070656120253,
      "loss": 4.9464,
      "mean_token_accuracy": 0.19931708723306657,
      "num_tokens": 23943926.0,
      "step": 13800
    },
    {
      "entropy": 5.5216656684875485,
      "epoch": 1.0740711923750244,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004888984688862809,
      "loss": 5.1075,
      "mean_token_accuracy": 0.18864704370498658,
      "num_tokens": 23952513.0,
      "step": 13805
    },
    {
      "entropy": 5.473466777801514,
      "epoch": 1.0744602217467418,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004888898689149677,
      "loss": 5.1567,
      "mean_token_accuracy": 0.18419950902462007,
      "num_tokens": 23961529.0,
      "step": 13810
    },
    {
      "entropy": 5.599845933914184,
      "epoch": 1.0748492511184595,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004888812656982162,
      "loss": 5.1678,
      "mean_token_accuracy": 0.1867597982287407,
      "num_tokens": 23971156.0,
      "step": 13815
    },
    {
      "entropy": 5.536244201660156,
      "epoch": 1.075238280490177,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004888726592361569,
      "loss": 5.087,
      "mean_token_accuracy": 0.18730096518993378,
      "num_tokens": 23979676.0,
      "step": 13820
    },
    {
      "entropy": 5.502763032913208,
      "epoch": 1.0756273098618945,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004888640495289204,
      "loss": 5.0442,
      "mean_token_accuracy": 0.19743094295263292,
      "num_tokens": 23987764.0,
      "step": 13825
    },
    {
      "entropy": 5.516732168197632,
      "epoch": 1.0760163392336122,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004888554365766374,
      "loss": 5.0889,
      "mean_token_accuracy": 0.18741167187690735,
      "num_tokens": 23995844.0,
      "step": 13830
    },
    {
      "entropy": 5.582467460632325,
      "epoch": 1.0764053686053296,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004888468203794385,
      "loss": 5.1551,
      "mean_token_accuracy": 0.18705841451883315,
      "num_tokens": 24003824.0,
      "step": 13835
    },
    {
      "entropy": 5.520799207687378,
      "epoch": 1.0767943979770473,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004888382009374545,
      "loss": 5.1104,
      "mean_token_accuracy": 0.19018265902996062,
      "num_tokens": 24012471.0,
      "step": 13840
    },
    {
      "entropy": 5.489942026138306,
      "epoch": 1.0771834273487648,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004888295782508161,
      "loss": 5.0774,
      "mean_token_accuracy": 0.18765824437141418,
      "num_tokens": 24021344.0,
      "step": 13845
    },
    {
      "entropy": 5.486586713790894,
      "epoch": 1.0775724567204823,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004888209523196542,
      "loss": 5.0585,
      "mean_token_accuracy": 0.191952583193779,
      "num_tokens": 24029598.0,
      "step": 13850
    },
    {
      "entropy": 5.478898525238037,
      "epoch": 1.0779614860922,
      "grad_norm": 1.0,
      "learning_rate": 0.0004888123231440997,
      "loss": 5.0937,
      "mean_token_accuracy": 0.19664800763130189,
      "num_tokens": 24038097.0,
      "step": 13855
    },
    {
      "entropy": 5.507804441452026,
      "epoch": 1.0783505154639175,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004888036907242834,
      "loss": 5.1394,
      "mean_token_accuracy": 0.1909420445561409,
      "num_tokens": 24046812.0,
      "step": 13860
    },
    {
      "entropy": 5.532537126541138,
      "epoch": 1.0787395448356352,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004887950550603366,
      "loss": 5.1034,
      "mean_token_accuracy": 0.18934880644083024,
      "num_tokens": 24054908.0,
      "step": 13865
    },
    {
      "entropy": 5.445264434814453,
      "epoch": 1.0791285742073526,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004887864161523901,
      "loss": 5.0881,
      "mean_token_accuracy": 0.19156306385993957,
      "num_tokens": 24063695.0,
      "step": 13870
    },
    {
      "entropy": 5.471846675872802,
      "epoch": 1.0795176035790701,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004887777740005749,
      "loss": 5.0944,
      "mean_token_accuracy": 0.19618339389562606,
      "num_tokens": 24072481.0,
      "step": 13875
    },
    {
      "entropy": 5.449295997619629,
      "epoch": 1.0799066329507878,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004887691286050224,
      "loss": 5.1033,
      "mean_token_accuracy": 0.19354427009820938,
      "num_tokens": 24080751.0,
      "step": 13880
    },
    {
      "entropy": 5.498619031906128,
      "epoch": 1.0802956623225053,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004887604799658635,
      "loss": 5.0627,
      "mean_token_accuracy": 0.19357450008392335,
      "num_tokens": 24088618.0,
      "step": 13885
    },
    {
      "entropy": 5.552611064910889,
      "epoch": 1.080684691694223,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004887518280832295,
      "loss": 5.1612,
      "mean_token_accuracy": 0.1876908838748932,
      "num_tokens": 24096754.0,
      "step": 13890
    },
    {
      "entropy": 5.509295797348022,
      "epoch": 1.0810737210659405,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004887431729572519,
      "loss": 5.1046,
      "mean_token_accuracy": 0.19504912346601486,
      "num_tokens": 24104793.0,
      "step": 13895
    },
    {
      "entropy": 5.645768070220948,
      "epoch": 1.081462750437658,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004887345145880617,
      "loss": 5.2399,
      "mean_token_accuracy": 0.17526261061429976,
      "num_tokens": 24113816.0,
      "step": 13900
    },
    {
      "entropy": 5.4748670101165775,
      "epoch": 1.0818517798093756,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004887258529757904,
      "loss": 5.0151,
      "mean_token_accuracy": 0.19832040667533873,
      "num_tokens": 24123002.0,
      "step": 13905
    },
    {
      "entropy": 5.527581691741943,
      "epoch": 1.082240809181093,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004887171881205696,
      "loss": 5.0827,
      "mean_token_accuracy": 0.1928435355424881,
      "num_tokens": 24130841.0,
      "step": 13910
    },
    {
      "entropy": 5.529430389404297,
      "epoch": 1.0826298385528108,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004887085200225306,
      "loss": 5.1369,
      "mean_token_accuracy": 0.1842429146170616,
      "num_tokens": 24139236.0,
      "step": 13915
    },
    {
      "entropy": 5.4993288040161135,
      "epoch": 1.0830188679245283,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004886998486818049,
      "loss": 5.0007,
      "mean_token_accuracy": 0.19117014557123185,
      "num_tokens": 24147824.0,
      "step": 13920
    },
    {
      "entropy": 5.4894407272338865,
      "epoch": 1.083407897296246,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004886911740985242,
      "loss": 5.1378,
      "mean_token_accuracy": 0.18899588137865067,
      "num_tokens": 24156555.0,
      "step": 13925
    },
    {
      "entropy": 5.572859048843384,
      "epoch": 1.0837969266679635,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048868249627282,
      "loss": 5.067,
      "mean_token_accuracy": 0.1919061616063118,
      "num_tokens": 24164424.0,
      "step": 13930
    },
    {
      "entropy": 5.554365062713623,
      "epoch": 1.084185956039681,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000488673815204824,
      "loss": 5.0665,
      "mean_token_accuracy": 0.19214770644903184,
      "num_tokens": 24173325.0,
      "step": 13935
    },
    {
      "entropy": 5.566837120056152,
      "epoch": 1.0845749854113986,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004886651308946681,
      "loss": 5.2351,
      "mean_token_accuracy": 0.18521015346050262,
      "num_tokens": 24181781.0,
      "step": 13940
    },
    {
      "entropy": 5.535018825531006,
      "epoch": 1.084964014783116,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004886564433424839,
      "loss": 5.0446,
      "mean_token_accuracy": 0.1946351185441017,
      "num_tokens": 24190065.0,
      "step": 13945
    },
    {
      "entropy": 5.552212905883789,
      "epoch": 1.0853530441548336,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004886477525484032,
      "loss": 5.1895,
      "mean_token_accuracy": 0.17955034971237183,
      "num_tokens": 24198814.0,
      "step": 13950
    },
    {
      "entropy": 5.50428352355957,
      "epoch": 1.0857420735265513,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004886390585125579,
      "loss": 5.0703,
      "mean_token_accuracy": 0.18745607882738113,
      "num_tokens": 24208008.0,
      "step": 13955
    },
    {
      "entropy": 5.567502403259278,
      "epoch": 1.0861311028982688,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004886303612350799,
      "loss": 5.0599,
      "mean_token_accuracy": 0.1963200479745865,
      "num_tokens": 24216098.0,
      "step": 13960
    },
    {
      "entropy": 5.44388165473938,
      "epoch": 1.0865201322699864,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004886216607161013,
      "loss": 5.0052,
      "mean_token_accuracy": 0.19607273936271669,
      "num_tokens": 24224943.0,
      "step": 13965
    },
    {
      "entropy": 5.492292070388794,
      "epoch": 1.086909161641704,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004886129569557538,
      "loss": 5.0602,
      "mean_token_accuracy": 0.18439961224794388,
      "num_tokens": 24232950.0,
      "step": 13970
    },
    {
      "entropy": 5.5669811248779295,
      "epoch": 1.0872981910134216,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004886042499541699,
      "loss": 5.2174,
      "mean_token_accuracy": 0.1809267058968544,
      "num_tokens": 24242355.0,
      "step": 13975
    },
    {
      "entropy": 5.5212867736816404,
      "epoch": 1.087687220385139,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004885955397114813,
      "loss": 5.0799,
      "mean_token_accuracy": 0.19551039189100267,
      "num_tokens": 24251247.0,
      "step": 13980
    },
    {
      "entropy": 5.516960620880127,
      "epoch": 1.0880762497568566,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004885868262278205,
      "loss": 5.1354,
      "mean_token_accuracy": 0.1920214042067528,
      "num_tokens": 24260281.0,
      "step": 13985
    },
    {
      "entropy": 5.584014463424682,
      "epoch": 1.0884652791285743,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004885781095033195,
      "loss": 5.1458,
      "mean_token_accuracy": 0.1842520847916603,
      "num_tokens": 24268474.0,
      "step": 13990
    },
    {
      "entropy": 5.576725864410401,
      "epoch": 1.0888543085002917,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004885693895381108,
      "loss": 5.1152,
      "mean_token_accuracy": 0.19198710322380066,
      "num_tokens": 24276897.0,
      "step": 13995
    },
    {
      "entropy": 5.508225297927856,
      "epoch": 1.0892433378720092,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004885606663323263,
      "loss": 5.1037,
      "mean_token_accuracy": 0.1859440565109253,
      "num_tokens": 24285892.0,
      "step": 14000
    },
    {
      "entropy": 5.492849445343017,
      "epoch": 1.089632367243727,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004885519398860987,
      "loss": 5.1479,
      "mean_token_accuracy": 0.18897592425346374,
      "num_tokens": 24294141.0,
      "step": 14005
    },
    {
      "entropy": 5.484719085693359,
      "epoch": 1.0900213966154444,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004885432101995604,
      "loss": 5.0419,
      "mean_token_accuracy": 0.19183039367198945,
      "num_tokens": 24302803.0,
      "step": 14010
    },
    {
      "entropy": 5.612570571899414,
      "epoch": 1.090410425987162,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004885344772728436,
      "loss": 5.2109,
      "mean_token_accuracy": 0.1866873636841774,
      "num_tokens": 24311755.0,
      "step": 14015
    },
    {
      "entropy": 5.512559032440185,
      "epoch": 1.0907994553588796,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004885257411060812,
      "loss": 5.0389,
      "mean_token_accuracy": 0.19991969168186188,
      "num_tokens": 24320913.0,
      "step": 14020
    },
    {
      "entropy": 5.563611316680908,
      "epoch": 1.0911884847305973,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004885170016994053,
      "loss": 5.2466,
      "mean_token_accuracy": 0.1805800497531891,
      "num_tokens": 24331112.0,
      "step": 14025
    },
    {
      "entropy": 5.493542051315307,
      "epoch": 1.0915775141023147,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004885082590529489,
      "loss": 5.0313,
      "mean_token_accuracy": 0.18927630931138992,
      "num_tokens": 24339286.0,
      "step": 14030
    },
    {
      "entropy": 5.474504184722901,
      "epoch": 1.0919665434740322,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004884995131668445,
      "loss": 5.0481,
      "mean_token_accuracy": 0.19487200826406478,
      "num_tokens": 24347787.0,
      "step": 14035
    },
    {
      "entropy": 5.478480672836303,
      "epoch": 1.09235557284575,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004884907640412248,
      "loss": 5.0171,
      "mean_token_accuracy": 0.19764152616262437,
      "num_tokens": 24356340.0,
      "step": 14040
    },
    {
      "entropy": 5.499757242202759,
      "epoch": 1.0927446022174674,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004884820116762225,
      "loss": 5.1043,
      "mean_token_accuracy": 0.1859050989151001,
      "num_tokens": 24365182.0,
      "step": 14045
    },
    {
      "entropy": 5.59054708480835,
      "epoch": 1.093133631589185,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004884732560719706,
      "loss": 5.1251,
      "mean_token_accuracy": 0.18418388366699218,
      "num_tokens": 24373455.0,
      "step": 14050
    },
    {
      "entropy": 5.541578340530395,
      "epoch": 1.0935226609609026,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004884644972286017,
      "loss": 5.2097,
      "mean_token_accuracy": 0.18187592774629593,
      "num_tokens": 24382433.0,
      "step": 14055
    },
    {
      "entropy": 5.459346628189087,
      "epoch": 1.09391169033262,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004884557351462488,
      "loss": 5.037,
      "mean_token_accuracy": 0.18847588896751405,
      "num_tokens": 24390971.0,
      "step": 14060
    },
    {
      "entropy": 5.567386817932129,
      "epoch": 1.0943007197043377,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004884469698250449,
      "loss": 5.0809,
      "mean_token_accuracy": 0.18430597633123397,
      "num_tokens": 24399508.0,
      "step": 14065
    },
    {
      "entropy": 5.503924751281739,
      "epoch": 1.0946897490760552,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000488438201265123,
      "loss": 5.0427,
      "mean_token_accuracy": 0.19206362068653107,
      "num_tokens": 24408424.0,
      "step": 14070
    },
    {
      "entropy": 5.451383209228515,
      "epoch": 1.095078778447773,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004884294294666161,
      "loss": 5.1517,
      "mean_token_accuracy": 0.18028387874364854,
      "num_tokens": 24416918.0,
      "step": 14075
    },
    {
      "entropy": 5.46699743270874,
      "epoch": 1.0954678078194904,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004884206544296573,
      "loss": 5.0063,
      "mean_token_accuracy": 0.1981320023536682,
      "num_tokens": 24425779.0,
      "step": 14080
    },
    {
      "entropy": 5.436846208572388,
      "epoch": 1.0958568371912079,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004884118761543797,
      "loss": 5.0115,
      "mean_token_accuracy": 0.19454351514577867,
      "num_tokens": 24434567.0,
      "step": 14085
    },
    {
      "entropy": 5.4949877738952635,
      "epoch": 1.0962458665629256,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004884030946409167,
      "loss": 5.1149,
      "mean_token_accuracy": 0.1891827717423439,
      "num_tokens": 24442980.0,
      "step": 14090
    },
    {
      "entropy": 5.504739618301391,
      "epoch": 1.096634895934643,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004883943098894013,
      "loss": 5.0369,
      "mean_token_accuracy": 0.19354359954595565,
      "num_tokens": 24452273.0,
      "step": 14095
    },
    {
      "entropy": 5.558314418792724,
      "epoch": 1.0970239253063607,
      "grad_norm": 1.046875,
      "learning_rate": 0.000488385521899967,
      "loss": 5.1346,
      "mean_token_accuracy": 0.19268351793289185,
      "num_tokens": 24461854.0,
      "step": 14100
    },
    {
      "entropy": 5.55611400604248,
      "epoch": 1.0974129546780782,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000488376730672747,
      "loss": 5.181,
      "mean_token_accuracy": 0.18543296083807945,
      "num_tokens": 24470468.0,
      "step": 14105
    },
    {
      "entropy": 5.472955942153931,
      "epoch": 1.0978019840497957,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004883679362078746,
      "loss": 5.0612,
      "mean_token_accuracy": 0.1938111111521721,
      "num_tokens": 24479282.0,
      "step": 14110
    },
    {
      "entropy": 5.569868040084839,
      "epoch": 1.0981910134215134,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004883591385054836,
      "loss": 5.1203,
      "mean_token_accuracy": 0.18583229482173919,
      "num_tokens": 24487331.0,
      "step": 14115
    },
    {
      "entropy": 5.532917070388794,
      "epoch": 1.0985800427932308,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004883503375657072,
      "loss": 5.0333,
      "mean_token_accuracy": 0.19619555920362472,
      "num_tokens": 24496101.0,
      "step": 14120
    },
    {
      "entropy": 5.56544041633606,
      "epoch": 1.0989690721649485,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004883415333886789,
      "loss": 5.1313,
      "mean_token_accuracy": 0.18872231543064116,
      "num_tokens": 24504321.0,
      "step": 14125
    },
    {
      "entropy": 5.523133325576782,
      "epoch": 1.099358101536666,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004883327259745325,
      "loss": 5.0343,
      "mean_token_accuracy": 0.2005436897277832,
      "num_tokens": 24511887.0,
      "step": 14130
    },
    {
      "entropy": 5.467382097244263,
      "epoch": 1.0997471309083835,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004883239153234015,
      "loss": 5.0408,
      "mean_token_accuracy": 0.19444339126348495,
      "num_tokens": 24521188.0,
      "step": 14135
    },
    {
      "entropy": 5.502029752731323,
      "epoch": 1.1001361602801012,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004883151014354197,
      "loss": 5.042,
      "mean_token_accuracy": 0.19536309987306594,
      "num_tokens": 24530656.0,
      "step": 14140
    },
    {
      "entropy": 5.47239580154419,
      "epoch": 1.1005251896518187,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004883062843107207,
      "loss": 5.036,
      "mean_token_accuracy": 0.18762117326259614,
      "num_tokens": 24539438.0,
      "step": 14145
    },
    {
      "entropy": 5.524498224258423,
      "epoch": 1.1009142190235364,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004882974639494383,
      "loss": 5.1131,
      "mean_token_accuracy": 0.19000244736671448,
      "num_tokens": 24548005.0,
      "step": 14150
    },
    {
      "entropy": 5.5723482131958,
      "epoch": 1.1013032483952538,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004882886403517065,
      "loss": 5.1766,
      "mean_token_accuracy": 0.1834762141108513,
      "num_tokens": 24556507.0,
      "step": 14155
    },
    {
      "entropy": 5.512179756164551,
      "epoch": 1.1016922777669713,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004882798135176589,
      "loss": 5.0981,
      "mean_token_accuracy": 0.18978823721408844,
      "num_tokens": 24565228.0,
      "step": 14160
    },
    {
      "entropy": 5.492662191390991,
      "epoch": 1.102081307138689,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048827098344742973,
      "loss": 5.0612,
      "mean_token_accuracy": 0.19266658574342727,
      "num_tokens": 24573977.0,
      "step": 14165
    },
    {
      "entropy": 5.47968544960022,
      "epoch": 1.1024703365104065,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004882621501411528,
      "loss": 5.0084,
      "mean_token_accuracy": 0.19525153189897537,
      "num_tokens": 24582897.0,
      "step": 14170
    },
    {
      "entropy": 5.500520706176758,
      "epoch": 1.1028593658821242,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004882533135989622,
      "loss": 5.0083,
      "mean_token_accuracy": 0.19499415755271912,
      "num_tokens": 24591132.0,
      "step": 14175
    },
    {
      "entropy": 5.44187741279602,
      "epoch": 1.1032483952538417,
      "grad_norm": 1.09375,
      "learning_rate": 0.000488244473820992,
      "loss": 5.0474,
      "mean_token_accuracy": 0.19462545961141586,
      "num_tokens": 24600111.0,
      "step": 14180
    },
    {
      "entropy": 5.468484830856323,
      "epoch": 1.1036374246255591,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048823563080737634,
      "loss": 5.0821,
      "mean_token_accuracy": 0.18705769032239913,
      "num_tokens": 24609229.0,
      "step": 14185
    },
    {
      "entropy": 5.513787841796875,
      "epoch": 1.1040264539972768,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004882267845582493,
      "loss": 5.0174,
      "mean_token_accuracy": 0.19238975793123245,
      "num_tokens": 24618012.0,
      "step": 14190
    },
    {
      "entropy": 5.499605274200439,
      "epoch": 1.1044154833689943,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048821793507374526,
      "loss": 5.0828,
      "mean_token_accuracy": 0.1874783769249916,
      "num_tokens": 24626801.0,
      "step": 14195
    },
    {
      "entropy": 5.582557535171508,
      "epoch": 1.104804512740712,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004882090823539984,
      "loss": 5.1441,
      "mean_token_accuracy": 0.19009732455015182,
      "num_tokens": 24635365.0,
      "step": 14200
    },
    {
      "entropy": 5.528085708618164,
      "epoch": 1.1051935421124295,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004882002263991431,
      "loss": 5.0614,
      "mean_token_accuracy": 0.19266999065876006,
      "num_tokens": 24643745.0,
      "step": 14205
    },
    {
      "entropy": 5.451296997070313,
      "epoch": 1.105582571484147,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048819136720931364,
      "loss": 4.9862,
      "mean_token_accuracy": 0.19295700192451476,
      "num_tokens": 24652760.0,
      "step": 14210
    },
    {
      "entropy": 5.468032741546631,
      "epoch": 1.1059716008558647,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048818250478464457,
      "loss": 5.0917,
      "mean_token_accuracy": 0.19449607282876968,
      "num_tokens": 24661326.0,
      "step": 14215
    },
    {
      "entropy": 5.53998441696167,
      "epoch": 1.1063606302275821,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004881736391252703,
      "loss": 4.9831,
      "mean_token_accuracy": 0.20357064604759217,
      "num_tokens": 24669611.0,
      "step": 14220
    },
    {
      "entropy": 5.5587647438049315,
      "epoch": 1.1067496595992998,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004881647702313253,
      "loss": 5.1331,
      "mean_token_accuracy": 0.18797731697559356,
      "num_tokens": 24677780.0,
      "step": 14225
    },
    {
      "entropy": 5.449176788330078,
      "epoch": 1.1071386889710173,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048815589810294427,
      "loss": 4.992,
      "mean_token_accuracy": 0.19787534475326538,
      "num_tokens": 24685807.0,
      "step": 14230
    },
    {
      "entropy": 5.505717706680298,
      "epoch": 1.1075277183427348,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048814702274026174,
      "loss": 5.0674,
      "mean_token_accuracy": 0.19014875292778016,
      "num_tokens": 24694383.0,
      "step": 14235
    },
    {
      "entropy": 5.4938231945037845,
      "epoch": 1.1079167477144525,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004881381441434124,
      "loss": 5.0332,
      "mean_token_accuracy": 0.19944778978824615,
      "num_tokens": 24702325.0,
      "step": 14240
    },
    {
      "entropy": 5.562407922744751,
      "epoch": 1.10830577708617,
      "grad_norm": 1.03125,
      "learning_rate": 0.000488129262312531,
      "loss": 5.047,
      "mean_token_accuracy": 0.1880396455526352,
      "num_tokens": 24710725.0,
      "step": 14245
    },
    {
      "entropy": 5.5140186786651615,
      "epoch": 1.1086948064578876,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048812037724775217,
      "loss": 5.089,
      "mean_token_accuracy": 0.18963437527418137,
      "num_tokens": 24720230.0,
      "step": 14250
    },
    {
      "entropy": 5.514590835571289,
      "epoch": 1.1090838358296051,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004881114889492109,
      "loss": 5.0333,
      "mean_token_accuracy": 0.19527243375778197,
      "num_tokens": 24728896.0,
      "step": 14255
    },
    {
      "entropy": 5.552097272872925,
      "epoch": 1.1094728652013226,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048810259741704197,
      "loss": 5.1255,
      "mean_token_accuracy": 0.1868831530213356,
      "num_tokens": 24737731.0,
      "step": 14260
    },
    {
      "entropy": 5.490778255462646,
      "epoch": 1.1098618945730403,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004880937026513803,
      "loss": 4.9972,
      "mean_token_accuracy": 0.20356862992048264,
      "num_tokens": 24746335.0,
      "step": 14265
    },
    {
      "entropy": 5.468053483963013,
      "epoch": 1.1102509239447578,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004880848046523609,
      "loss": 5.011,
      "mean_token_accuracy": 0.19655592292547225,
      "num_tokens": 24755917.0,
      "step": 14270
    },
    {
      "entropy": 5.56942458152771,
      "epoch": 1.1106399533164755,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048807590342011875,
      "loss": 5.1565,
      "mean_token_accuracy": 0.18422688394784928,
      "num_tokens": 24764713.0,
      "step": 14275
    },
    {
      "entropy": 5.624075222015381,
      "epoch": 1.111028982688193,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048806699895478895,
      "loss": 5.1736,
      "mean_token_accuracy": 0.18587102890014648,
      "num_tokens": 24773464.0,
      "step": 14280
    },
    {
      "entropy": 5.536322355270386,
      "epoch": 1.1114180120599104,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004880580912565065,
      "loss": 5.0805,
      "mean_token_accuracy": 0.19150267243385316,
      "num_tokens": 24781437.0,
      "step": 14285
    },
    {
      "entropy": 5.424597883224488,
      "epoch": 1.1118070414316281,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004880491803254067,
      "loss": 4.9369,
      "mean_token_accuracy": 0.2050044596195221,
      "num_tokens": 24789505.0,
      "step": 14290
    },
    {
      "entropy": 5.566405391693115,
      "epoch": 1.1121960708033456,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004880402661616246,
      "loss": 5.1591,
      "mean_token_accuracy": 0.18626979291439055,
      "num_tokens": 24798492.0,
      "step": 14295
    },
    {
      "entropy": 5.560792398452759,
      "epoch": 1.1125851001750633,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004880313487652956,
      "loss": 5.1253,
      "mean_token_accuracy": 0.18197024911642073,
      "num_tokens": 24807472.0,
      "step": 14300
    },
    {
      "entropy": 5.555822801589966,
      "epoch": 1.1129741295467808,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048802242813655503,
      "loss": 5.1201,
      "mean_token_accuracy": 0.19133339524269105,
      "num_tokens": 24815838.0,
      "step": 14305
    },
    {
      "entropy": 5.47281665802002,
      "epoch": 1.1133631589184985,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000488013504275538,
      "loss": 5.0191,
      "mean_token_accuracy": 0.20027527660131456,
      "num_tokens": 24823953.0,
      "step": 14310
    },
    {
      "entropy": 5.478111028671265,
      "epoch": 1.113752188290216,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004880045771823802,
      "loss": 4.9503,
      "mean_token_accuracy": 0.2032728061079979,
      "num_tokens": 24832414.0,
      "step": 14315
    },
    {
      "entropy": 5.470767974853516,
      "epoch": 1.1141412176619334,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048799564685721695,
      "loss": 5.0831,
      "mean_token_accuracy": 0.19745009392499924,
      "num_tokens": 24840503.0,
      "step": 14320
    },
    {
      "entropy": 5.546571922302246,
      "epoch": 1.1145302470336511,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004879867133001837,
      "loss": 5.2095,
      "mean_token_accuracy": 0.18024653792381287,
      "num_tokens": 24849272.0,
      "step": 14325
    },
    {
      "entropy": 5.601848077774048,
      "epoch": 1.1149192764053686,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004879777765114162,
      "loss": 5.1524,
      "mean_token_accuracy": 0.1855832278728485,
      "num_tokens": 24858565.0,
      "step": 14330
    },
    {
      "entropy": 5.559348487854004,
      "epoch": 1.115308305777086,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048796883649104996,
      "loss": 5.0995,
      "mean_token_accuracy": 0.19043328016996383,
      "num_tokens": 24867811.0,
      "step": 14335
    },
    {
      "entropy": 5.5027812957763675,
      "epoch": 1.1156973351488038,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004879598932392206,
      "loss": 5.141,
      "mean_token_accuracy": 0.18531684279441835,
      "num_tokens": 24876776.0,
      "step": 14340
    },
    {
      "entropy": 5.515747594833374,
      "epoch": 1.1160863645205212,
      "grad_norm": 1.0,
      "learning_rate": 0.00048795094675606364,
      "loss": 5.0432,
      "mean_token_accuracy": 0.18900105804204942,
      "num_tokens": 24885030.0,
      "step": 14345
    },
    {
      "entropy": 5.618272304534912,
      "epoch": 1.116475393892239,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004879419970417152,
      "loss": 5.2094,
      "mean_token_accuracy": 0.1811125546693802,
      "num_tokens": 24894336.0,
      "step": 14350
    },
    {
      "entropy": 5.567889404296875,
      "epoch": 1.1168644232639564,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048793304409631077,
      "loss": 5.1523,
      "mean_token_accuracy": 0.19123206585645675,
      "num_tokens": 24903107.0,
      "step": 14355
    },
    {
      "entropy": 5.5856328964233395,
      "epoch": 1.117253452635674,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004879240879199863,
      "loss": 5.2228,
      "mean_token_accuracy": 0.18544175028800963,
      "num_tokens": 24912727.0,
      "step": 14360
    },
    {
      "entropy": 5.506678438186645,
      "epoch": 1.1176424820073916,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048791512851287786,
      "loss": 5.011,
      "mean_token_accuracy": 0.19116442650556564,
      "num_tokens": 24921534.0,
      "step": 14365
    },
    {
      "entropy": 5.542968273162842,
      "epoch": 1.118031511379109,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048790616587512107,
      "loss": 5.0759,
      "mean_token_accuracy": 0.1946049377322197,
      "num_tokens": 24930093.0,
      "step": 14370
    },
    {
      "entropy": 5.553399705886841,
      "epoch": 1.1184205407508268,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004878972000068521,
      "loss": 5.1571,
      "mean_token_accuracy": 0.1915786013007164,
      "num_tokens": 24939230.0,
      "step": 14375
    },
    {
      "entropy": 5.581459665298462,
      "epoch": 1.1188095701225442,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048788823090820707,
      "loss": 5.2005,
      "mean_token_accuracy": 0.1927233710885048,
      "num_tokens": 24947595.0,
      "step": 14380
    },
    {
      "entropy": 5.617218542098999,
      "epoch": 1.1191985994942617,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048787925857932194,
      "loss": 5.1641,
      "mean_token_accuracy": 0.1869514748454094,
      "num_tokens": 24957099.0,
      "step": 14385
    },
    {
      "entropy": 5.499294185638428,
      "epoch": 1.1195876288659794,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048787028302033287,
      "loss": 5.0508,
      "mean_token_accuracy": 0.19672874361276627,
      "num_tokens": 24965701.0,
      "step": 14390
    },
    {
      "entropy": 5.585096836090088,
      "epoch": 1.1199766582376969,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048786130423137606,
      "loss": 5.1678,
      "mean_token_accuracy": 0.17971692532300948,
      "num_tokens": 24974025.0,
      "step": 14395
    },
    {
      "entropy": 5.605995559692383,
      "epoch": 1.1203656876094146,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004878523222125879,
      "loss": 5.212,
      "mean_token_accuracy": 0.1861012578010559,
      "num_tokens": 24982930.0,
      "step": 14400
    },
    {
      "entropy": 5.5747809410095215,
      "epoch": 1.120754716981132,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004878433369641044,
      "loss": 5.0839,
      "mean_token_accuracy": 0.1896425664424896,
      "num_tokens": 24991332.0,
      "step": 14405
    },
    {
      "entropy": 5.533490514755249,
      "epoch": 1.1211437463528497,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004878343484860621,
      "loss": 5.073,
      "mean_token_accuracy": 0.19489451348781586,
      "num_tokens": 25000130.0,
      "step": 14410
    },
    {
      "entropy": 5.498442029953003,
      "epoch": 1.1215327757245672,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004878253567785972,
      "loss": 5.0862,
      "mean_token_accuracy": 0.19168929159641265,
      "num_tokens": 25008423.0,
      "step": 14415
    },
    {
      "entropy": 5.5474930763244625,
      "epoch": 1.1219218050962847,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048781636184184644,
      "loss": 5.1535,
      "mean_token_accuracy": 0.18940316438674926,
      "num_tokens": 25017399.0,
      "step": 14420
    },
    {
      "entropy": 5.600519275665283,
      "epoch": 1.1223108344680024,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000487807363675946,
      "loss": 5.1323,
      "mean_token_accuracy": 0.19237236380577089,
      "num_tokens": 25025957.0,
      "step": 14425
    },
    {
      "entropy": 5.5836108207702635,
      "epoch": 1.1226998638397199,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004877983622810326,
      "loss": 5.0517,
      "mean_token_accuracy": 0.19297924786806106,
      "num_tokens": 25034246.0,
      "step": 14430
    },
    {
      "entropy": 5.473474454879761,
      "epoch": 1.1230888932114376,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004877893576572427,
      "loss": 5.0679,
      "mean_token_accuracy": 0.1966800644993782,
      "num_tokens": 25042461.0,
      "step": 14435
    },
    {
      "entropy": 5.499518251419067,
      "epoch": 1.123477922583155,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000487780349804713,
      "loss": 5.0166,
      "mean_token_accuracy": 0.198726686835289,
      "num_tokens": 25051291.0,
      "step": 14440
    },
    {
      "entropy": 5.56500825881958,
      "epoch": 1.1238669519548725,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004877713387235802,
      "loss": 5.1769,
      "mean_token_accuracy": 0.1876739665865898,
      "num_tokens": 25060538.0,
      "step": 14445
    },
    {
      "entropy": 5.56879072189331,
      "epoch": 1.1242559813265902,
      "grad_norm": 1.1875,
      "learning_rate": 0.000487762324413981,
      "loss": 5.1383,
      "mean_token_accuracy": 0.1828738421201706,
      "num_tokens": 25068965.0,
      "step": 14450
    },
    {
      "entropy": 5.551497316360473,
      "epoch": 1.1246450106983077,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048775330687605213,
      "loss": 5.1172,
      "mean_token_accuracy": 0.19131905734539031,
      "num_tokens": 25077021.0,
      "step": 14455
    },
    {
      "entropy": 5.521827030181885,
      "epoch": 1.1250340400700254,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004877442861099305,
      "loss": 5.1248,
      "mean_token_accuracy": 0.19296469837427138,
      "num_tokens": 25086009.0,
      "step": 14460
    },
    {
      "entropy": 5.485211563110352,
      "epoch": 1.1254230694417429,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004877352621157529,
      "loss": 5.0394,
      "mean_token_accuracy": 0.19775542616844177,
      "num_tokens": 25094745.0,
      "step": 14465
    },
    {
      "entropy": 5.541313123703003,
      "epoch": 1.1258120988134603,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048772623489365635,
      "loss": 5.0533,
      "mean_token_accuracy": 0.18670387268066407,
      "num_tokens": 25103087.0,
      "step": 14470
    },
    {
      "entropy": 5.580843877792359,
      "epoch": 1.126201128185178,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004877172044437777,
      "loss": 5.0909,
      "mean_token_accuracy": 0.18991893231868745,
      "num_tokens": 25111822.0,
      "step": 14475
    },
    {
      "entropy": 5.534433841705322,
      "epoch": 1.1265901575568955,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048770817076625416,
      "loss": 5.0777,
      "mean_token_accuracy": 0.18970127701759337,
      "num_tokens": 25119975.0,
      "step": 14480
    },
    {
      "entropy": 5.498084688186646,
      "epoch": 1.126979186928613,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048769913386122253,
      "loss": 5.1352,
      "mean_token_accuracy": 0.18403317928314208,
      "num_tokens": 25129031.0,
      "step": 14485
    },
    {
      "entropy": 5.559974622726441,
      "epoch": 1.1273682163003307,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004876900937288202,
      "loss": 5.0603,
      "mean_token_accuracy": 0.19354215115308762,
      "num_tokens": 25138230.0,
      "step": 14490
    },
    {
      "entropy": 5.594967031478882,
      "epoch": 1.1277572456720482,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048768105036918426,
      "loss": 5.1066,
      "mean_token_accuracy": 0.18996895998716354,
      "num_tokens": 25147120.0,
      "step": 14495
    },
    {
      "entropy": 5.508925104141236,
      "epoch": 1.1281462750437659,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048767200378245187,
      "loss": 5.0949,
      "mean_token_accuracy": 0.1873033821582794,
      "num_tokens": 25155623.0,
      "step": 14500
    },
    {
      "entropy": 5.416905736923217,
      "epoch": 1.1285353044154833,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048766295396876025,
      "loss": 4.9894,
      "mean_token_accuracy": 0.2006245657801628,
      "num_tokens": 25163867.0,
      "step": 14505
    },
    {
      "entropy": 5.492851257324219,
      "epoch": 1.128924333787201,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048765390092824683,
      "loss": 5.1112,
      "mean_token_accuracy": 0.19573555290699005,
      "num_tokens": 25172838.0,
      "step": 14510
    },
    {
      "entropy": 5.517008304595947,
      "epoch": 1.1293133631589185,
      "grad_norm": 1.046875,
      "learning_rate": 0.000487644844661049,
      "loss": 5.0425,
      "mean_token_accuracy": 0.19584952890872956,
      "num_tokens": 25181168.0,
      "step": 14515
    },
    {
      "entropy": 5.529186773300171,
      "epoch": 1.129702392530636,
      "grad_norm": 1.09375,
      "learning_rate": 0.000487635785167304,
      "loss": 5.1322,
      "mean_token_accuracy": 0.19279411882162095,
      "num_tokens": 25190115.0,
      "step": 14520
    },
    {
      "entropy": 5.513104200363159,
      "epoch": 1.1300914219023537,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048762672244714953,
      "loss": 5.1152,
      "mean_token_accuracy": 0.19561296403408052,
      "num_tokens": 25199594.0,
      "step": 14525
    },
    {
      "entropy": 5.545372438430786,
      "epoch": 1.1304804512740712,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004876176565007229,
      "loss": 5.0942,
      "mean_token_accuracy": 0.18271679878234864,
      "num_tokens": 25208060.0,
      "step": 14530
    },
    {
      "entropy": 5.522700405120849,
      "epoch": 1.1308694806457888,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048760858732816174,
      "loss": 5.0332,
      "mean_token_accuracy": 0.19625958800315857,
      "num_tokens": 25216257.0,
      "step": 14535
    },
    {
      "entropy": 5.503209972381592,
      "epoch": 1.1312585100175063,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004875995149296037,
      "loss": 5.0665,
      "mean_token_accuracy": 0.18915267139673234,
      "num_tokens": 25225032.0,
      "step": 14540
    },
    {
      "entropy": 5.527342796325684,
      "epoch": 1.1316475393892238,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004875904393051864,
      "loss": 5.0992,
      "mean_token_accuracy": 0.19268207550048827,
      "num_tokens": 25233624.0,
      "step": 14545
    },
    {
      "entropy": 5.482742547988892,
      "epoch": 1.1320365687609415,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004875813604550476,
      "loss": 5.0742,
      "mean_token_accuracy": 0.1890130117535591,
      "num_tokens": 25242132.0,
      "step": 14550
    },
    {
      "entropy": 5.5358116149902346,
      "epoch": 1.132425598132659,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048757227837932494,
      "loss": 5.0665,
      "mean_token_accuracy": 0.19654684513807297,
      "num_tokens": 25250923.0,
      "step": 14555
    },
    {
      "entropy": 5.50623927116394,
      "epoch": 1.1328146275043767,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004875631930781563,
      "loss": 5.0797,
      "mean_token_accuracy": 0.1868762791156769,
      "num_tokens": 25259783.0,
      "step": 14560
    },
    {
      "entropy": 5.608754968643188,
      "epoch": 1.1332036568760941,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048755410455167967,
      "loss": 5.2545,
      "mean_token_accuracy": 0.1812473177909851,
      "num_tokens": 25268801.0,
      "step": 14565
    },
    {
      "entropy": 5.5011719226837155,
      "epoch": 1.1335926862478116,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004875450128000326,
      "loss": 5.0795,
      "mean_token_accuracy": 0.20246496200561523,
      "num_tokens": 25277226.0,
      "step": 14570
    },
    {
      "entropy": 5.547874164581299,
      "epoch": 1.1339817156195293,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004875359178233534,
      "loss": 5.1334,
      "mean_token_accuracy": 0.184638549387455,
      "num_tokens": 25285887.0,
      "step": 14575
    },
    {
      "entropy": 5.466899013519287,
      "epoch": 1.1343707449912468,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048752681962177985,
      "loss": 4.989,
      "mean_token_accuracy": 0.19612828642129898,
      "num_tokens": 25294025.0,
      "step": 14580
    },
    {
      "entropy": 5.540697622299194,
      "epoch": 1.1347597743629645,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048751771819545013,
      "loss": 5.1841,
      "mean_token_accuracy": 0.1805033266544342,
      "num_tokens": 25303016.0,
      "step": 14585
    },
    {
      "entropy": 5.4904622554779055,
      "epoch": 1.135148803734682,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048750861354450236,
      "loss": 4.9969,
      "mean_token_accuracy": 0.19832673221826552,
      "num_tokens": 25311193.0,
      "step": 14590
    },
    {
      "entropy": 5.502160882949829,
      "epoch": 1.1355378331063994,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004874995056690746,
      "loss": 5.0413,
      "mean_token_accuracy": 0.19251364320516587,
      "num_tokens": 25319380.0,
      "step": 14595
    },
    {
      "entropy": 5.479501867294312,
      "epoch": 1.1359268624781171,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000487490394569305,
      "loss": 5.0738,
      "mean_token_accuracy": 0.19011088460683823,
      "num_tokens": 25327854.0,
      "step": 14600
    },
    {
      "entropy": 5.538140773773193,
      "epoch": 1.1363158918498346,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004874812802453319,
      "loss": 5.1382,
      "mean_token_accuracy": 0.1864347591996193,
      "num_tokens": 25335912.0,
      "step": 14605
    },
    {
      "entropy": 5.540270757675171,
      "epoch": 1.1367049212215523,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004874721626972936,
      "loss": 5.1312,
      "mean_token_accuracy": 0.18981873542070388,
      "num_tokens": 25344104.0,
      "step": 14610
    },
    {
      "entropy": 5.40794186592102,
      "epoch": 1.1370939505932698,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004874630419253284,
      "loss": 4.9002,
      "mean_token_accuracy": 0.2072306215763092,
      "num_tokens": 25352583.0,
      "step": 14615
    },
    {
      "entropy": 5.489667749404907,
      "epoch": 1.1374829799649873,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048745391792957474,
      "loss": 5.0237,
      "mean_token_accuracy": 0.19759504050016402,
      "num_tokens": 25361035.0,
      "step": 14620
    },
    {
      "entropy": 5.511130857467651,
      "epoch": 1.137872009336705,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048744479071017097,
      "loss": 5.0757,
      "mean_token_accuracy": 0.18722357451915742,
      "num_tokens": 25369511.0,
      "step": 14625
    },
    {
      "entropy": 5.537259340286255,
      "epoch": 1.1382610387084224,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004874356602672556,
      "loss": 5.0593,
      "mean_token_accuracy": 0.1942879229784012,
      "num_tokens": 25378023.0,
      "step": 14630
    },
    {
      "entropy": 5.515888595581055,
      "epoch": 1.1386500680801401,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004874265266009673,
      "loss": 5.0876,
      "mean_token_accuracy": 0.19389840960502625,
      "num_tokens": 25386791.0,
      "step": 14635
    },
    {
      "entropy": 5.541774559020996,
      "epoch": 1.1390390974518576,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004874173897114445,
      "loss": 5.1233,
      "mean_token_accuracy": 0.18887362033128738,
      "num_tokens": 25395401.0,
      "step": 14640
    },
    {
      "entropy": 5.517326307296753,
      "epoch": 1.1394281268235753,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048740824959882603,
      "loss": 5.0922,
      "mean_token_accuracy": 0.18885716944932937,
      "num_tokens": 25404193.0,
      "step": 14645
    },
    {
      "entropy": 5.589709711074829,
      "epoch": 1.1398171561952928,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004873991062632504,
      "loss": 5.1486,
      "mean_token_accuracy": 0.18541148602962493,
      "num_tokens": 25413073.0,
      "step": 14650
    },
    {
      "entropy": 5.518786954879761,
      "epoch": 1.1402061855670103,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048738995970485635,
      "loss": 5.0516,
      "mean_token_accuracy": 0.19625996202230453,
      "num_tokens": 25422064.0,
      "step": 14655
    },
    {
      "entropy": 5.4679289817810055,
      "epoch": 1.140595214938728,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004873808099237827,
      "loss": 5.0442,
      "mean_token_accuracy": 0.19268875271081926,
      "num_tokens": 25430711.0,
      "step": 14660
    },
    {
      "entropy": 5.533324909210205,
      "epoch": 1.1409842443104454,
      "grad_norm": 1.125,
      "learning_rate": 0.0004873716569201684,
      "loss": 5.1484,
      "mean_token_accuracy": 0.18824423998594284,
      "num_tokens": 25439527.0,
      "step": 14665
    },
    {
      "entropy": 5.57870397567749,
      "epoch": 1.141373273682163,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048736250069415213,
      "loss": 5.1338,
      "mean_token_accuracy": 0.18862760066986084,
      "num_tokens": 25448633.0,
      "step": 14670
    },
    {
      "entropy": 5.526706075668335,
      "epoch": 1.1417623030538806,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004873533412458729,
      "loss": 5.0826,
      "mean_token_accuracy": 0.19548732787370682,
      "num_tokens": 25457657.0,
      "step": 14675
    },
    {
      "entropy": 5.538207101821899,
      "epoch": 1.142151332425598,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048734417857546986,
      "loss": 5.1426,
      "mean_token_accuracy": 0.1832851141691208,
      "num_tokens": 25466698.0,
      "step": 14680
    },
    {
      "entropy": 5.53098430633545,
      "epoch": 1.1425403617973158,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004873350126830818,
      "loss": 5.0953,
      "mean_token_accuracy": 0.19191282391548156,
      "num_tokens": 25474921.0,
      "step": 14685
    },
    {
      "entropy": 5.4922059059143065,
      "epoch": 1.1429293911690332,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004873258435688479,
      "loss": 5.007,
      "mean_token_accuracy": 0.19682122468948365,
      "num_tokens": 25483553.0,
      "step": 14690
    },
    {
      "entropy": 5.509087467193604,
      "epoch": 1.143318420540751,
      "grad_norm": 1.0,
      "learning_rate": 0.0004873166712329073,
      "loss": 5.0841,
      "mean_token_accuracy": 0.1958193376660347,
      "num_tokens": 25492408.0,
      "step": 14695
    },
    {
      "entropy": 5.48152232170105,
      "epoch": 1.1437074499124684,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048730749567539914,
      "loss": 5.1171,
      "mean_token_accuracy": 0.19028908610343934,
      "num_tokens": 25501404.0,
      "step": 14700
    },
    {
      "entropy": 5.62200231552124,
      "epoch": 1.144096479284186,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048729831689646257,
      "loss": 5.087,
      "mean_token_accuracy": 0.19033405780792237,
      "num_tokens": 25510176.0,
      "step": 14705
    },
    {
      "entropy": 5.48921914100647,
      "epoch": 1.1444855086559036,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048728913489623705,
      "loss": 5.018,
      "mean_token_accuracy": 0.19810061156749725,
      "num_tokens": 25518809.0,
      "step": 14710
    },
    {
      "entropy": 5.509998178482055,
      "epoch": 1.144874538027621,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004872799496748618,
      "loss": 5.1353,
      "mean_token_accuracy": 0.18691265881061553,
      "num_tokens": 25527415.0,
      "step": 14715
    },
    {
      "entropy": 5.506222486495972,
      "epoch": 1.1452635673993385,
      "grad_norm": 1.125,
      "learning_rate": 0.0004872707612324761,
      "loss": 5.0665,
      "mean_token_accuracy": 0.19148872196674346,
      "num_tokens": 25535618.0,
      "step": 14720
    },
    {
      "entropy": 5.5276792526245115,
      "epoch": 1.1456525967710562,
      "grad_norm": 1.125,
      "learning_rate": 0.0004872615695692196,
      "loss": 5.0556,
      "mean_token_accuracy": 0.19384928792715073,
      "num_tokens": 25544228.0,
      "step": 14725
    },
    {
      "entropy": 5.560665607452393,
      "epoch": 1.1460416261427737,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004872523746852315,
      "loss": 5.1217,
      "mean_token_accuracy": 0.1918061375617981,
      "num_tokens": 25553557.0,
      "step": 14730
    },
    {
      "entropy": 5.524463653564453,
      "epoch": 1.1464306555144914,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048724317658065146,
      "loss": 5.0597,
      "mean_token_accuracy": 0.19602214992046357,
      "num_tokens": 25562018.0,
      "step": 14735
    },
    {
      "entropy": 5.542082023620606,
      "epoch": 1.146819684886209,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048723397525561916,
      "loss": 5.077,
      "mean_token_accuracy": 0.19392478913068772,
      "num_tokens": 25569382.0,
      "step": 14740
    },
    {
      "entropy": 5.446687841415406,
      "epoch": 1.1472087142579266,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048722477071027394,
      "loss": 5.0488,
      "mean_token_accuracy": 0.19076944291591644,
      "num_tokens": 25578172.0,
      "step": 14745
    },
    {
      "entropy": 5.483876276016235,
      "epoch": 1.147597743629644,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048721556294475574,
      "loss": 5.0811,
      "mean_token_accuracy": 0.18641868233680725,
      "num_tokens": 25587572.0,
      "step": 14750
    },
    {
      "entropy": 5.5010094165802,
      "epoch": 1.1479867730013615,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048720635195920403,
      "loss": 5.1224,
      "mean_token_accuracy": 0.18839685022830963,
      "num_tokens": 25596609.0,
      "step": 14755
    },
    {
      "entropy": 5.492743921279907,
      "epoch": 1.1483758023730792,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004871971377537588,
      "loss": 5.1172,
      "mean_token_accuracy": 0.18480881452560424,
      "num_tokens": 25605707.0,
      "step": 14760
    },
    {
      "entropy": 5.500701475143432,
      "epoch": 1.1487648317447967,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004871879203285597,
      "loss": 5.0974,
      "mean_token_accuracy": 0.19200366139411926,
      "num_tokens": 25614218.0,
      "step": 14765
    },
    {
      "entropy": 5.479060983657837,
      "epoch": 1.1491538611165142,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048717869968374655,
      "loss": 5.015,
      "mean_token_accuracy": 0.19801199436187744,
      "num_tokens": 25622613.0,
      "step": 14770
    },
    {
      "entropy": 5.596790790557861,
      "epoch": 1.1495428904882319,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048716947581945954,
      "loss": 5.1651,
      "mean_token_accuracy": 0.19067128002643585,
      "num_tokens": 25631396.0,
      "step": 14775
    },
    {
      "entropy": 5.512734079360962,
      "epoch": 1.1499319198599494,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004871602487358383,
      "loss": 5.0624,
      "mean_token_accuracy": 0.1906217947602272,
      "num_tokens": 25640870.0,
      "step": 14780
    },
    {
      "entropy": 5.512539386749268,
      "epoch": 1.150320949231667,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000487151018433023,
      "loss": 5.1023,
      "mean_token_accuracy": 0.19320798367261888,
      "num_tokens": 25649395.0,
      "step": 14785
    },
    {
      "entropy": 5.552593517303467,
      "epoch": 1.1507099786033845,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048714178491115367,
      "loss": 5.2303,
      "mean_token_accuracy": 0.17922023981809615,
      "num_tokens": 25659087.0,
      "step": 14790
    },
    {
      "entropy": 5.601691436767578,
      "epoch": 1.1510990079751022,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048713254817037046,
      "loss": 5.1332,
      "mean_token_accuracy": 0.19814841598272323,
      "num_tokens": 25667847.0,
      "step": 14795
    },
    {
      "entropy": 5.560252857208252,
      "epoch": 1.1514880373468197,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004871233082108135,
      "loss": 5.0973,
      "mean_token_accuracy": 0.19683469384908675,
      "num_tokens": 25676786.0,
      "step": 14800
    },
    {
      "entropy": 5.559467315673828,
      "epoch": 1.1518770667185372,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004871140650326228,
      "loss": 5.122,
      "mean_token_accuracy": 0.1866964191198349,
      "num_tokens": 25685223.0,
      "step": 14805
    },
    {
      "entropy": 5.505224990844726,
      "epoch": 1.1522660960902549,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004871048186359389,
      "loss": 5.081,
      "mean_token_accuracy": 0.18967041671276091,
      "num_tokens": 25694243.0,
      "step": 14810
    },
    {
      "entropy": 5.4880866527557375,
      "epoch": 1.1526551254619724,
      "grad_norm": 1.0625,
      "learning_rate": 0.000487095569020902,
      "loss": 4.9706,
      "mean_token_accuracy": 0.1987641155719757,
      "num_tokens": 25702735.0,
      "step": 14815
    },
    {
      "entropy": 5.495584774017334,
      "epoch": 1.1530441548336898,
      "grad_norm": 1.125,
      "learning_rate": 0.0004870863161876524,
      "loss": 5.0398,
      "mean_token_accuracy": 0.1969444677233696,
      "num_tokens": 25711577.0,
      "step": 14820
    },
    {
      "entropy": 5.442644119262695,
      "epoch": 1.1534331842054075,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004870770601363305,
      "loss": 4.9708,
      "mean_token_accuracy": 0.20398910939693451,
      "num_tokens": 25719927.0,
      "step": 14825
    },
    {
      "entropy": 5.475719594955445,
      "epoch": 1.153822213577125,
      "grad_norm": 1.125,
      "learning_rate": 0.00048706780086707675,
      "loss": 5.0374,
      "mean_token_accuracy": 0.19568579345941545,
      "num_tokens": 25728144.0,
      "step": 14830
    },
    {
      "entropy": 5.509170818328857,
      "epoch": 1.1542112429488427,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004870585383800317,
      "loss": 5.0145,
      "mean_token_accuracy": 0.2013348937034607,
      "num_tokens": 25736736.0,
      "step": 14835
    },
    {
      "entropy": 5.561881351470947,
      "epoch": 1.1546002723205602,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048704927267533585,
      "loss": 5.1741,
      "mean_token_accuracy": 0.19281785637140275,
      "num_tokens": 25745031.0,
      "step": 14840
    },
    {
      "entropy": 5.59914984703064,
      "epoch": 1.1549893016922779,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048704000375312975,
      "loss": 5.1744,
      "mean_token_accuracy": 0.19195466339588166,
      "num_tokens": 25754265.0,
      "step": 14845
    },
    {
      "entropy": 5.60050368309021,
      "epoch": 1.1553783310639953,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048703073161355414,
      "loss": 5.1569,
      "mean_token_accuracy": 0.1798067256808281,
      "num_tokens": 25763084.0,
      "step": 14850
    },
    {
      "entropy": 5.585230445861816,
      "epoch": 1.1557673604357128,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004870214562567497,
      "loss": 5.1278,
      "mean_token_accuracy": 0.1833072006702423,
      "num_tokens": 25771626.0,
      "step": 14855
    },
    {
      "entropy": 5.500636243820191,
      "epoch": 1.1561563898074305,
      "grad_norm": 1.109375,
      "learning_rate": 0.000487012177682857,
      "loss": 5.1048,
      "mean_token_accuracy": 0.18746039718389512,
      "num_tokens": 25780326.0,
      "step": 14860
    },
    {
      "entropy": 5.6091166019439695,
      "epoch": 1.156545419179148,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004870028958920171,
      "loss": 5.2143,
      "mean_token_accuracy": 0.18252416253089904,
      "num_tokens": 25789318.0,
      "step": 14865
    },
    {
      "entropy": 5.560399055480957,
      "epoch": 1.1569344485508655,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004869936108843706,
      "loss": 5.1714,
      "mean_token_accuracy": 0.18942274749279023,
      "num_tokens": 25798025.0,
      "step": 14870
    },
    {
      "entropy": 5.513164520263672,
      "epoch": 1.1573234779225832,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048698432266005854,
      "loss": 5.0801,
      "mean_token_accuracy": 0.18999859243631362,
      "num_tokens": 25806669.0,
      "step": 14875
    },
    {
      "entropy": 5.545013809204102,
      "epoch": 1.1577125072943006,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004869750312192217,
      "loss": 5.0613,
      "mean_token_accuracy": 0.19719772338867186,
      "num_tokens": 25815938.0,
      "step": 14880
    },
    {
      "entropy": 5.524427890777588,
      "epoch": 1.1581015366660183,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048696573656200123,
      "loss": 5.025,
      "mean_token_accuracy": 0.19465565979480742,
      "num_tokens": 25824293.0,
      "step": 14885
    },
    {
      "entropy": 5.573456859588623,
      "epoch": 1.1584905660377358,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048695643868853806,
      "loss": 5.1394,
      "mean_token_accuracy": 0.19229037314653397,
      "num_tokens": 25833081.0,
      "step": 14890
    },
    {
      "entropy": 5.5616230964660645,
      "epoch": 1.1588795954094535,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004869471375989733,
      "loss": 5.1509,
      "mean_token_accuracy": 0.18271852433681487,
      "num_tokens": 25841344.0,
      "step": 14895
    },
    {
      "entropy": 5.542437887191772,
      "epoch": 1.159268624781171,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004869378332934481,
      "loss": 5.1355,
      "mean_token_accuracy": 0.1918633058667183,
      "num_tokens": 25850617.0,
      "step": 14900
    },
    {
      "entropy": 5.503354930877686,
      "epoch": 1.1596576541528885,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048692852577210356,
      "loss": 5.1073,
      "mean_token_accuracy": 0.1922723948955536,
      "num_tokens": 25858569.0,
      "step": 14905
    },
    {
      "entropy": 5.480866861343384,
      "epoch": 1.1600466835246062,
      "grad_norm": 1.015625,
      "learning_rate": 0.000486919215035081,
      "loss": 5.0969,
      "mean_token_accuracy": 0.19177360832691193,
      "num_tokens": 25866892.0,
      "step": 14910
    },
    {
      "entropy": 5.572716188430786,
      "epoch": 1.1604357128963236,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004869099010825217,
      "loss": 5.169,
      "mean_token_accuracy": 0.18786127269268035,
      "num_tokens": 25875661.0,
      "step": 14915
    },
    {
      "entropy": 5.4618480682373045,
      "epoch": 1.1608247422680413,
      "grad_norm": 1.0,
      "learning_rate": 0.00048690058391456684,
      "loss": 4.9913,
      "mean_token_accuracy": 0.20505907982587815,
      "num_tokens": 25884951.0,
      "step": 14920
    },
    {
      "entropy": 5.580494832992554,
      "epoch": 1.1612137716397588,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004868912635313579,
      "loss": 5.2153,
      "mean_token_accuracy": 0.18756548762321473,
      "num_tokens": 25893652.0,
      "step": 14925
    },
    {
      "entropy": 5.556605863571167,
      "epoch": 1.1616028010114763,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004868819399330364,
      "loss": 5.0353,
      "mean_token_accuracy": 0.19741895496845246,
      "num_tokens": 25902375.0,
      "step": 14930
    },
    {
      "entropy": 5.541285228729248,
      "epoch": 1.161991830383194,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004868726131197436,
      "loss": 5.0305,
      "mean_token_accuracy": 0.19814417958259584,
      "num_tokens": 25911322.0,
      "step": 14935
    },
    {
      "entropy": 5.54860553741455,
      "epoch": 1.1623808597549115,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004868632830916212,
      "loss": 5.1246,
      "mean_token_accuracy": 0.19302973449230193,
      "num_tokens": 25920444.0,
      "step": 14940
    },
    {
      "entropy": 5.520397472381592,
      "epoch": 1.1627698891266292,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004868539498488106,
      "loss": 5.1589,
      "mean_token_accuracy": 0.18462863713502883,
      "num_tokens": 25928986.0,
      "step": 14945
    },
    {
      "entropy": 5.543773031234741,
      "epoch": 1.1631589184983466,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004868446133914536,
      "loss": 5.1473,
      "mean_token_accuracy": 0.1835136741399765,
      "num_tokens": 25938209.0,
      "step": 14950
    },
    {
      "entropy": 5.561061859130859,
      "epoch": 1.163547947870064,
      "grad_norm": 1.0,
      "learning_rate": 0.00048683527371969185,
      "loss": 5.1618,
      "mean_token_accuracy": 0.1805805891752243,
      "num_tokens": 25946909.0,
      "step": 14955
    },
    {
      "entropy": 5.549157571792603,
      "epoch": 1.1639369772417818,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048682593083366685,
      "loss": 5.1932,
      "mean_token_accuracy": 0.1887127012014389,
      "num_tokens": 25956154.0,
      "step": 14960
    },
    {
      "entropy": 5.544176769256592,
      "epoch": 1.1643260066134993,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048681658473352055,
      "loss": 5.078,
      "mean_token_accuracy": 0.1977374732494354,
      "num_tokens": 25964478.0,
      "step": 14965
    },
    {
      "entropy": 5.496403646469116,
      "epoch": 1.164715035985217,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004868072354193947,
      "loss": 5.0817,
      "mean_token_accuracy": 0.18507562577724457,
      "num_tokens": 25973162.0,
      "step": 14970
    },
    {
      "entropy": 5.479502582550049,
      "epoch": 1.1651040653569344,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004867978828914312,
      "loss": 5.0375,
      "mean_token_accuracy": 0.19386684894561768,
      "num_tokens": 25982790.0,
      "step": 14975
    },
    {
      "entropy": 5.515274572372436,
      "epoch": 1.165493094728652,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004867885271497719,
      "loss": 4.994,
      "mean_token_accuracy": 0.19537770748138428,
      "num_tokens": 25992111.0,
      "step": 14980
    },
    {
      "entropy": 5.579084300994873,
      "epoch": 1.1658821241003696,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004867791681945588,
      "loss": 5.1369,
      "mean_token_accuracy": 0.1857917472720146,
      "num_tokens": 26000803.0,
      "step": 14985
    },
    {
      "entropy": 5.593853616714478,
      "epoch": 1.166271153472087,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048676980602593395,
      "loss": 5.2397,
      "mean_token_accuracy": 0.18015141040086746,
      "num_tokens": 26010152.0,
      "step": 14990
    },
    {
      "entropy": 5.556007719039917,
      "epoch": 1.1666601828438048,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048676044064403936,
      "loss": 5.1268,
      "mean_token_accuracy": 0.17995952516794206,
      "num_tokens": 26019183.0,
      "step": 14995
    },
    {
      "entropy": 5.546884346008301,
      "epoch": 1.1670492122155223,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004867510720490171,
      "loss": 5.0711,
      "mean_token_accuracy": 0.19535953104496,
      "num_tokens": 26027459.0,
      "step": 15000
    },
    {
      "epoch": 1.1670492122155223,
      "eval_entropy": 5.285074955827348,
      "eval_loss": 5.185548782348633,
      "eval_mean_token_accuracy": 0.19627875544855064,
      "eval_num_tokens": 26027459.0,
      "eval_runtime": 21.6246,
      "eval_samples_per_second": 1921.795,
      "eval_steps_per_second": 240.236,
      "step": 15000
    },
    {
      "entropy": 5.5424901962280275,
      "epoch": 1.1674382415872397,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004867417002410094,
      "loss": 5.1287,
      "mean_token_accuracy": 0.18980713486671447,
      "num_tokens": 26036421.0,
      "step": 15005
    },
    {
      "entropy": 5.521946001052856,
      "epoch": 1.1678272709589574,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048673232522015845,
      "loss": 5.1377,
      "mean_token_accuracy": 0.18729576468467712,
      "num_tokens": 26044971.0,
      "step": 15010
    },
    {
      "entropy": 5.51538782119751,
      "epoch": 1.168216300330675,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004867229469866064,
      "loss": 5.0752,
      "mean_token_accuracy": 0.19356657564640045,
      "num_tokens": 26053597.0,
      "step": 15015
    },
    {
      "entropy": 5.553880119323731,
      "epoch": 1.1686053297023926,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004867135655404956,
      "loss": 5.1408,
      "mean_token_accuracy": 0.1860286369919777,
      "num_tokens": 26061928.0,
      "step": 15020
    },
    {
      "entropy": 5.518475675582886,
      "epoch": 1.16899435907411,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048670418088196844,
      "loss": 5.0999,
      "mean_token_accuracy": 0.19340391010046004,
      "num_tokens": 26070917.0,
      "step": 15025
    },
    {
      "entropy": 5.477444791793824,
      "epoch": 1.1693833884458276,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004866947930111674,
      "loss": 5.0182,
      "mean_token_accuracy": 0.19597697108983994,
      "num_tokens": 26079556.0,
      "step": 15030
    },
    {
      "entropy": 5.497043561935425,
      "epoch": 1.1697724178175453,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048668540192823467,
      "loss": 5.1242,
      "mean_token_accuracy": 0.18889948427677156,
      "num_tokens": 26087963.0,
      "step": 15035
    },
    {
      "entropy": 5.530980587005615,
      "epoch": 1.1701614471892627,
      "grad_norm": 1.0625,
      "learning_rate": 0.000486676007633313,
      "loss": 5.0681,
      "mean_token_accuracy": 0.19246715754270555,
      "num_tokens": 26096208.0,
      "step": 15040
    },
    {
      "entropy": 5.530849170684815,
      "epoch": 1.1705504765609804,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004866666101265448,
      "loss": 5.0799,
      "mean_token_accuracy": 0.19375794380903244,
      "num_tokens": 26104961.0,
      "step": 15045
    },
    {
      "entropy": 5.482483911514282,
      "epoch": 1.170939505932698,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048665720940807274,
      "loss": 5.1178,
      "mean_token_accuracy": 0.1854879677295685,
      "num_tokens": 26113889.0,
      "step": 15050
    },
    {
      "entropy": 5.461640930175781,
      "epoch": 1.1713285353044154,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048664780547803935,
      "loss": 5.0602,
      "mean_token_accuracy": 0.19064800888299943,
      "num_tokens": 26122139.0,
      "step": 15055
    },
    {
      "entropy": 5.619145631790161,
      "epoch": 1.171717564676133,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004866383983365873,
      "loss": 5.117,
      "mean_token_accuracy": 0.1966429755091667,
      "num_tokens": 26130817.0,
      "step": 15060
    },
    {
      "entropy": 5.566153287887573,
      "epoch": 1.1721065940478506,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048662898798385956,
      "loss": 5.1402,
      "mean_token_accuracy": 0.184890952706337,
      "num_tokens": 26140035.0,
      "step": 15065
    },
    {
      "entropy": 5.5695007801055905,
      "epoch": 1.1724956234195683,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048661957441999875,
      "loss": 5.1876,
      "mean_token_accuracy": 0.1897725760936737,
      "num_tokens": 26149575.0,
      "step": 15070
    },
    {
      "entropy": 5.559959554672242,
      "epoch": 1.1728846527912857,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048661015764514765,
      "loss": 5.1153,
      "mean_token_accuracy": 0.19433170109987258,
      "num_tokens": 26158743.0,
      "step": 15075
    },
    {
      "entropy": 5.551405143737793,
      "epoch": 1.1732736821630034,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048660073765944927,
      "loss": 5.1355,
      "mean_token_accuracy": 0.1897335857152939,
      "num_tokens": 26167380.0,
      "step": 15080
    },
    {
      "entropy": 5.39802565574646,
      "epoch": 1.173662711534721,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004865913144630465,
      "loss": 5.0486,
      "mean_token_accuracy": 0.19642680436372756,
      "num_tokens": 26175613.0,
      "step": 15085
    },
    {
      "entropy": 5.557881546020508,
      "epoch": 1.1740517409064384,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048658188805608227,
      "loss": 5.1614,
      "mean_token_accuracy": 0.18580930531024933,
      "num_tokens": 26184347.0,
      "step": 15090
    },
    {
      "entropy": 5.552322196960449,
      "epoch": 1.174440770278156,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004865724584386998,
      "loss": 5.1077,
      "mean_token_accuracy": 0.19181185364723205,
      "num_tokens": 26193513.0,
      "step": 15095
    },
    {
      "entropy": 5.548204183578491,
      "epoch": 1.1748297996498736,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004865630256110418,
      "loss": 5.0918,
      "mean_token_accuracy": 0.18959540575742723,
      "num_tokens": 26202790.0,
      "step": 15100
    },
    {
      "entropy": 5.498344230651855,
      "epoch": 1.175218829021591,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048655358957325174,
      "loss": 5.0563,
      "mean_token_accuracy": 0.19345860183238983,
      "num_tokens": 26211032.0,
      "step": 15105
    },
    {
      "entropy": 5.441040182113648,
      "epoch": 1.1756078583933087,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048654415032547277,
      "loss": 4.9986,
      "mean_token_accuracy": 0.19915515780448914,
      "num_tokens": 26219963.0,
      "step": 15110
    },
    {
      "entropy": 5.564775276184082,
      "epoch": 1.1759968877650262,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048653470786784794,
      "loss": 5.0384,
      "mean_token_accuracy": 0.1919045016169548,
      "num_tokens": 26228164.0,
      "step": 15115
    },
    {
      "entropy": 5.527921867370606,
      "epoch": 1.176385917136744,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004865252622005206,
      "loss": 5.0644,
      "mean_token_accuracy": 0.1919855535030365,
      "num_tokens": 26236294.0,
      "step": 15120
    },
    {
      "entropy": 5.444288349151611,
      "epoch": 1.1767749465084614,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048651581332363405,
      "loss": 5.0276,
      "mean_token_accuracy": 0.19645013213157653,
      "num_tokens": 26244955.0,
      "step": 15125
    },
    {
      "entropy": 5.545812940597534,
      "epoch": 1.177163975880179,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048650636123733176,
      "loss": 5.1076,
      "mean_token_accuracy": 0.18812841325998306,
      "num_tokens": 26253221.0,
      "step": 15130
    },
    {
      "entropy": 5.599039888381958,
      "epoch": 1.1775530052518965,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004864969059417571,
      "loss": 5.2016,
      "mean_token_accuracy": 0.17640066295862197,
      "num_tokens": 26262544.0,
      "step": 15135
    },
    {
      "entropy": 5.569678878784179,
      "epoch": 1.177942034623614,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004864874474370534,
      "loss": 5.0873,
      "mean_token_accuracy": 0.19341803193092347,
      "num_tokens": 26271864.0,
      "step": 15140
    },
    {
      "entropy": 5.617960405349732,
      "epoch": 1.1783310639953317,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048647798572336445,
      "loss": 5.1308,
      "mean_token_accuracy": 0.1932969272136688,
      "num_tokens": 26279713.0,
      "step": 15145
    },
    {
      "entropy": 5.4831994533538815,
      "epoch": 1.1787200933670492,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048646852080083355,
      "loss": 4.9686,
      "mean_token_accuracy": 0.20108863562345505,
      "num_tokens": 26288654.0,
      "step": 15150
    },
    {
      "entropy": 5.536346769332885,
      "epoch": 1.1791091227387667,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004864590526696045,
      "loss": 5.1566,
      "mean_token_accuracy": 0.1809620037674904,
      "num_tokens": 26297007.0,
      "step": 15155
    },
    {
      "entropy": 5.548009061813355,
      "epoch": 1.1794981521104844,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048644958132982094,
      "loss": 5.0957,
      "mean_token_accuracy": 0.1899002641439438,
      "num_tokens": 26305601.0,
      "step": 15160
    },
    {
      "entropy": 5.572567462921143,
      "epoch": 1.1798871814822018,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048644010678162644,
      "loss": 5.1289,
      "mean_token_accuracy": 0.19584194719791412,
      "num_tokens": 26313455.0,
      "step": 15165
    },
    {
      "entropy": 5.550379085540771,
      "epoch": 1.1802762108539195,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004864306290251649,
      "loss": 5.1316,
      "mean_token_accuracy": 0.18646952658891677,
      "num_tokens": 26322450.0,
      "step": 15170
    },
    {
      "entropy": 5.583476257324219,
      "epoch": 1.180665240225637,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048642114806058014,
      "loss": 5.1381,
      "mean_token_accuracy": 0.18496544510126114,
      "num_tokens": 26331301.0,
      "step": 15175
    },
    {
      "entropy": 5.661446952819825,
      "epoch": 1.1810542695973547,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004864116638880159,
      "loss": 5.2145,
      "mean_token_accuracy": 0.18682073503732682,
      "num_tokens": 26339937.0,
      "step": 15180
    },
    {
      "entropy": 5.561805629730225,
      "epoch": 1.1814432989690722,
      "grad_norm": 1.0,
      "learning_rate": 0.0004864021765076162,
      "loss": 5.1033,
      "mean_token_accuracy": 0.1903789073228836,
      "num_tokens": 26348541.0,
      "step": 15185
    },
    {
      "entropy": 5.515233659744263,
      "epoch": 1.1818323283407897,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048639268591952494,
      "loss": 5.0482,
      "mean_token_accuracy": 0.1971786856651306,
      "num_tokens": 26356621.0,
      "step": 15190
    },
    {
      "entropy": 5.578024005889892,
      "epoch": 1.1822213577125074,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048638319212388614,
      "loss": 5.1572,
      "mean_token_accuracy": 0.1811002105474472,
      "num_tokens": 26366112.0,
      "step": 15195
    },
    {
      "entropy": 5.615432214736939,
      "epoch": 1.1826103870842248,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004863736951208438,
      "loss": 5.1243,
      "mean_token_accuracy": 0.18959402590990065,
      "num_tokens": 26374768.0,
      "step": 15200
    },
    {
      "entropy": 5.559296798706055,
      "epoch": 1.1829994164559423,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048636419491054217,
      "loss": 5.1067,
      "mean_token_accuracy": 0.19149987846612931,
      "num_tokens": 26383097.0,
      "step": 15205
    },
    {
      "entropy": 5.501048040390015,
      "epoch": 1.18338844582766,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004863546914931252,
      "loss": 5.0911,
      "mean_token_accuracy": 0.19010667502880096,
      "num_tokens": 26391361.0,
      "step": 15210
    },
    {
      "entropy": 5.5398722171783445,
      "epoch": 1.1837774751993775,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004863451848687373,
      "loss": 5.0409,
      "mean_token_accuracy": 0.19561050832271576,
      "num_tokens": 26400818.0,
      "step": 15215
    },
    {
      "entropy": 5.567616415023804,
      "epoch": 1.1841665045710952,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048633567503752253,
      "loss": 5.1131,
      "mean_token_accuracy": 0.19112410098314286,
      "num_tokens": 26409650.0,
      "step": 15220
    },
    {
      "entropy": 5.471266365051269,
      "epoch": 1.1845555339428127,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004863261619996253,
      "loss": 4.9954,
      "mean_token_accuracy": 0.19494962841272354,
      "num_tokens": 26418237.0,
      "step": 15225
    },
    {
      "entropy": 5.411331224441528,
      "epoch": 1.1849445633145304,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004863166457551899,
      "loss": 4.9619,
      "mean_token_accuracy": 0.2030158281326294,
      "num_tokens": 26426366.0,
      "step": 15230
    },
    {
      "entropy": 5.455976295471191,
      "epoch": 1.1853335926862478,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004863071263043608,
      "loss": 5.0554,
      "mean_token_accuracy": 0.18722727447748183,
      "num_tokens": 26435603.0,
      "step": 15235
    },
    {
      "entropy": 5.566494417190552,
      "epoch": 1.1857226220579653,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004862976036472823,
      "loss": 5.1664,
      "mean_token_accuracy": 0.1888579845428467,
      "num_tokens": 26444230.0,
      "step": 15240
    },
    {
      "entropy": 5.523755073547363,
      "epoch": 1.186111651429683,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048628807778409907,
      "loss": 5.033,
      "mean_token_accuracy": 0.1854676440358162,
      "num_tokens": 26452779.0,
      "step": 15245
    },
    {
      "entropy": 5.536189842224121,
      "epoch": 1.1865006808014005,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004862785487149555,
      "loss": 5.1866,
      "mean_token_accuracy": 0.18533898293972015,
      "num_tokens": 26461977.0,
      "step": 15250
    },
    {
      "entropy": 5.509300565719604,
      "epoch": 1.186889710173118,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004862690164399963,
      "loss": 4.9821,
      "mean_token_accuracy": 0.19534797519445418,
      "num_tokens": 26470182.0,
      "step": 15255
    },
    {
      "entropy": 5.586657476425171,
      "epoch": 1.1872787395448356,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048625948095936593,
      "loss": 5.1864,
      "mean_token_accuracy": 0.18496657460927962,
      "num_tokens": 26479445.0,
      "step": 15260
    },
    {
      "entropy": 5.534335994720459,
      "epoch": 1.1876677689165531,
      "grad_norm": 1.0,
      "learning_rate": 0.00048624994227320923,
      "loss": 5.0613,
      "mean_token_accuracy": 0.19475616067647933,
      "num_tokens": 26488085.0,
      "step": 15265
    },
    {
      "entropy": 5.6325846195220945,
      "epoch": 1.1880567982882708,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048624040038167094,
      "loss": 5.2003,
      "mean_token_accuracy": 0.18820146322250367,
      "num_tokens": 26496849.0,
      "step": 15270
    },
    {
      "entropy": 5.522543907165527,
      "epoch": 1.1884458276599883,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048623085528489584,
      "loss": 5.0741,
      "mean_token_accuracy": 0.19307767301797868,
      "num_tokens": 26505412.0,
      "step": 15275
    },
    {
      "entropy": 5.447043704986572,
      "epoch": 1.188834857031706,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048622130698302863,
      "loss": 5.0463,
      "mean_token_accuracy": 0.19248784929513932,
      "num_tokens": 26513564.0,
      "step": 15280
    },
    {
      "entropy": 5.4453691959381105,
      "epoch": 1.1892238864034235,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048621175547621426,
      "loss": 5.0726,
      "mean_token_accuracy": 0.18962300419807435,
      "num_tokens": 26522230.0,
      "step": 15285
    },
    {
      "entropy": 5.506038331985474,
      "epoch": 1.189612915775141,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004862022007645978,
      "loss": 5.0757,
      "mean_token_accuracy": 0.18838860392570494,
      "num_tokens": 26531719.0,
      "step": 15290
    },
    {
      "entropy": 5.473969268798828,
      "epoch": 1.1900019451468586,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048619264284832403,
      "loss": 5.0119,
      "mean_token_accuracy": 0.20891416519880296,
      "num_tokens": 26540352.0,
      "step": 15295
    },
    {
      "entropy": 5.541357040405273,
      "epoch": 1.1903909745185761,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048618308172753804,
      "loss": 4.997,
      "mean_token_accuracy": 0.20202008485794068,
      "num_tokens": 26548252.0,
      "step": 15300
    },
    {
      "entropy": 5.547775554656982,
      "epoch": 1.1907800038902936,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004861735174023849,
      "loss": 5.0896,
      "mean_token_accuracy": 0.1903584599494934,
      "num_tokens": 26557267.0,
      "step": 15305
    },
    {
      "entropy": 5.555643367767334,
      "epoch": 1.1911690332620113,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048616394987300985,
      "loss": 5.1799,
      "mean_token_accuracy": 0.18931990265846252,
      "num_tokens": 26566221.0,
      "step": 15310
    },
    {
      "entropy": 5.515669536590576,
      "epoch": 1.1915580626337288,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004861543791395579,
      "loss": 5.1904,
      "mean_token_accuracy": 0.18758565783500672,
      "num_tokens": 26574826.0,
      "step": 15315
    },
    {
      "entropy": 5.56443943977356,
      "epoch": 1.1919470920054465,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004861448052021743,
      "loss": 5.1682,
      "mean_token_accuracy": 0.18198258876800538,
      "num_tokens": 26583442.0,
      "step": 15320
    },
    {
      "entropy": 5.599044322967529,
      "epoch": 1.192336121377164,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004861352280610044,
      "loss": 5.1053,
      "mean_token_accuracy": 0.19430507570505143,
      "num_tokens": 26592106.0,
      "step": 15325
    },
    {
      "entropy": 5.546983766555786,
      "epoch": 1.1927251507488816,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048612564771619346,
      "loss": 4.9891,
      "mean_token_accuracy": 0.2027801126241684,
      "num_tokens": 26601275.0,
      "step": 15330
    },
    {
      "entropy": 5.413982772827149,
      "epoch": 1.193114180120599,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048611606416788686,
      "loss": 4.9488,
      "mean_token_accuracy": 0.19674730449914932,
      "num_tokens": 26610179.0,
      "step": 15335
    },
    {
      "entropy": 5.491106843948364,
      "epoch": 1.1935032094923166,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048610647741622996,
      "loss": 5.1338,
      "mean_token_accuracy": 0.1892850250005722,
      "num_tokens": 26619463.0,
      "step": 15340
    },
    {
      "entropy": 5.492243814468384,
      "epoch": 1.1938922388640343,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004860968874613683,
      "loss": 4.9691,
      "mean_token_accuracy": 0.20187818855047227,
      "num_tokens": 26627597.0,
      "step": 15345
    },
    {
      "entropy": 5.521858978271484,
      "epoch": 1.1942812682357518,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004860872943034474,
      "loss": 5.0682,
      "mean_token_accuracy": 0.19797768145799638,
      "num_tokens": 26636180.0,
      "step": 15350
    },
    {
      "entropy": 5.493476438522339,
      "epoch": 1.1946702976074695,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004860776979426128,
      "loss": 4.9951,
      "mean_token_accuracy": 0.19941410571336746,
      "num_tokens": 26644803.0,
      "step": 15355
    },
    {
      "entropy": 5.461283159255982,
      "epoch": 1.195059326979187,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004860680983790101,
      "loss": 5.06,
      "mean_token_accuracy": 0.1821353554725647,
      "num_tokens": 26653619.0,
      "step": 15360
    },
    {
      "entropy": 5.566916847229004,
      "epoch": 1.1954483563509044,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004860584956127849,
      "loss": 5.0803,
      "mean_token_accuracy": 0.18993414044380189,
      "num_tokens": 26662291.0,
      "step": 15365
    },
    {
      "entropy": 5.547433519363404,
      "epoch": 1.195837385722622,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048604888964408306,
      "loss": 5.0186,
      "mean_token_accuracy": 0.20256989002227782,
      "num_tokens": 26670592.0,
      "step": 15370
    },
    {
      "entropy": 5.522049474716186,
      "epoch": 1.1962264150943396,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004860392804730502,
      "loss": 5.1049,
      "mean_token_accuracy": 0.19238650053739548,
      "num_tokens": 26679816.0,
      "step": 15375
    },
    {
      "entropy": 5.54670262336731,
      "epoch": 1.1966154444660573,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048602966809983204,
      "loss": 5.0638,
      "mean_token_accuracy": 0.19726591408252717,
      "num_tokens": 26688185.0,
      "step": 15380
    },
    {
      "entropy": 5.574678564071656,
      "epoch": 1.1970044738377748,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004860200525245746,
      "loss": 5.1174,
      "mean_token_accuracy": 0.19449135661125183,
      "num_tokens": 26697157.0,
      "step": 15385
    },
    {
      "entropy": 5.530929136276245,
      "epoch": 1.1973935032094922,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004860104337474239,
      "loss": 5.0759,
      "mean_token_accuracy": 0.18623018264770508,
      "num_tokens": 26705868.0,
      "step": 15390
    },
    {
      "entropy": 5.544787168502808,
      "epoch": 1.19778253258121,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048600081176852555,
      "loss": 5.1225,
      "mean_token_accuracy": 0.18925859183073043,
      "num_tokens": 26714364.0,
      "step": 15395
    },
    {
      "entropy": 5.540149354934693,
      "epoch": 1.1981715619529274,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048599118658802575,
      "loss": 5.104,
      "mean_token_accuracy": 0.1901453360915184,
      "num_tokens": 26722866.0,
      "step": 15400
    },
    {
      "entropy": 5.569733333587647,
      "epoch": 1.198560591324645,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004859815582060706,
      "loss": 5.0678,
      "mean_token_accuracy": 0.1933328241109848,
      "num_tokens": 26731305.0,
      "step": 15405
    },
    {
      "entropy": 5.556345272064209,
      "epoch": 1.1989496206963626,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004859719266228061,
      "loss": 5.0725,
      "mean_token_accuracy": 0.18924600630998611,
      "num_tokens": 26740807.0,
      "step": 15410
    },
    {
      "entropy": 5.530645036697388,
      "epoch": 1.19933865006808,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004859622918383784,
      "loss": 5.0769,
      "mean_token_accuracy": 0.19913332611322404,
      "num_tokens": 26749748.0,
      "step": 15415
    },
    {
      "entropy": 5.560524368286133,
      "epoch": 1.1997276794397977,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004859526538529337,
      "loss": 5.1085,
      "mean_token_accuracy": 0.1903930962085724,
      "num_tokens": 26758597.0,
      "step": 15420
    },
    {
      "entropy": 5.520481014251709,
      "epoch": 1.2001167088115152,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004859430126666182,
      "loss": 5.1716,
      "mean_token_accuracy": 0.1873624250292778,
      "num_tokens": 26766772.0,
      "step": 15425
    },
    {
      "entropy": 5.434782695770264,
      "epoch": 1.200505738183233,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048593336827957824,
      "loss": 4.908,
      "mean_token_accuracy": 0.2124838188290596,
      "num_tokens": 26775229.0,
      "step": 15430
    },
    {
      "entropy": 5.492747783660889,
      "epoch": 1.2008947675549504,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048592372069196024,
      "loss": 5.0657,
      "mean_token_accuracy": 0.18502474278211595,
      "num_tokens": 26784188.0,
      "step": 15435
    },
    {
      "entropy": 5.391122198104858,
      "epoch": 1.2012837969266679,
      "grad_norm": 1.125,
      "learning_rate": 0.0004859140699039104,
      "loss": 5.0033,
      "mean_token_accuracy": 0.1958404377102852,
      "num_tokens": 26792687.0,
      "step": 15440
    },
    {
      "entropy": 5.4335612773895265,
      "epoch": 1.2016728262983856,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048590441591557526,
      "loss": 5.0696,
      "mean_token_accuracy": 0.19458646327257156,
      "num_tokens": 26801931.0,
      "step": 15445
    },
    {
      "entropy": 5.541347408294678,
      "epoch": 1.202061855670103,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048589475872710134,
      "loss": 5.2837,
      "mean_token_accuracy": 0.1805791325867176,
      "num_tokens": 26810689.0,
      "step": 15450
    },
    {
      "entropy": 5.555945014953613,
      "epoch": 1.2024508850418207,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004858850983386351,
      "loss": 5.0892,
      "mean_token_accuracy": 0.1876358225941658,
      "num_tokens": 26819038.0,
      "step": 15455
    },
    {
      "entropy": 5.512599468231201,
      "epoch": 1.2028399144135382,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004858754347503231,
      "loss": 5.0038,
      "mean_token_accuracy": 0.20283834785223007,
      "num_tokens": 26827443.0,
      "step": 15460
    },
    {
      "entropy": 5.489543724060058,
      "epoch": 1.203228943785256,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048586576796231216,
      "loss": 4.9887,
      "mean_token_accuracy": 0.20017201453447342,
      "num_tokens": 26836306.0,
      "step": 15465
    },
    {
      "entropy": 5.527681589126587,
      "epoch": 1.2036179731569734,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004858560979747487,
      "loss": 5.0941,
      "mean_token_accuracy": 0.19172093123197556,
      "num_tokens": 26845043.0,
      "step": 15470
    },
    {
      "entropy": 5.546275997161866,
      "epoch": 1.2040070025286909,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048584642478777967,
      "loss": 5.0305,
      "mean_token_accuracy": 0.19456254094839096,
      "num_tokens": 26853978.0,
      "step": 15475
    },
    {
      "entropy": 5.492612361907959,
      "epoch": 1.2043960319004086,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004858367484015517,
      "loss": 4.9623,
      "mean_token_accuracy": 0.1972953349351883,
      "num_tokens": 26862487.0,
      "step": 15480
    },
    {
      "entropy": 5.491152715682984,
      "epoch": 1.204785061272126,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048582706881621166,
      "loss": 5.1217,
      "mean_token_accuracy": 0.195041660964489,
      "num_tokens": 26871137.0,
      "step": 15485
    },
    {
      "entropy": 5.498955202102661,
      "epoch": 1.2051740906438435,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004858173860319063,
      "loss": 5.0204,
      "mean_token_accuracy": 0.19208043217658996,
      "num_tokens": 26880091.0,
      "step": 15490
    },
    {
      "entropy": 5.501911067962647,
      "epoch": 1.2055631200155612,
      "grad_norm": 1.125,
      "learning_rate": 0.00048580770004878277,
      "loss": 5.0741,
      "mean_token_accuracy": 0.19491726756095887,
      "num_tokens": 26888777.0,
      "step": 15495
    },
    {
      "entropy": 5.4712714672088625,
      "epoch": 1.2059521493872787,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004857980108669879,
      "loss": 5.0192,
      "mean_token_accuracy": 0.18907157480716705,
      "num_tokens": 26897689.0,
      "step": 15500
    },
    {
      "entropy": 5.566845369338989,
      "epoch": 1.2063411787589964,
      "grad_norm": 1.125,
      "learning_rate": 0.00048578831848666875,
      "loss": 5.0747,
      "mean_token_accuracy": 0.19245895445346833,
      "num_tokens": 26906209.0,
      "step": 15505
    },
    {
      "entropy": 5.609748697280883,
      "epoch": 1.2067302081307139,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048577862290797235,
      "loss": 5.2193,
      "mean_token_accuracy": 0.18533357828855515,
      "num_tokens": 26916340.0,
      "step": 15510
    },
    {
      "entropy": 5.564051675796509,
      "epoch": 1.2071192375024316,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048576892413104586,
      "loss": 5.0993,
      "mean_token_accuracy": 0.18928911983966829,
      "num_tokens": 26924482.0,
      "step": 15515
    },
    {
      "entropy": 5.507539129257202,
      "epoch": 1.207508266874149,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004857592221560364,
      "loss": 5.0485,
      "mean_token_accuracy": 0.19933302253484725,
      "num_tokens": 26932844.0,
      "step": 15520
    },
    {
      "entropy": 5.39739842414856,
      "epoch": 1.2078972962458665,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004857495169830912,
      "loss": 4.9461,
      "mean_token_accuracy": 0.19588245302438737,
      "num_tokens": 26940916.0,
      "step": 15525
    },
    {
      "entropy": 5.532335901260376,
      "epoch": 1.2082863256175842,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004857398086123575,
      "loss": 5.0752,
      "mean_token_accuracy": 0.18882836103439332,
      "num_tokens": 26949182.0,
      "step": 15530
    },
    {
      "entropy": 5.542239761352539,
      "epoch": 1.2086753549893017,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004857300970439827,
      "loss": 5.0477,
      "mean_token_accuracy": 0.19428498893976212,
      "num_tokens": 26957531.0,
      "step": 15535
    },
    {
      "entropy": 5.539743709564209,
      "epoch": 1.2090643843610192,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048572038227811404,
      "loss": 5.1167,
      "mean_token_accuracy": 0.19446775317192078,
      "num_tokens": 26965430.0,
      "step": 15540
    },
    {
      "entropy": 5.570826148986816,
      "epoch": 1.2094534137327368,
      "grad_norm": 1.09375,
      "learning_rate": 0.000485710664314899,
      "loss": 5.1531,
      "mean_token_accuracy": 0.18896981924772263,
      "num_tokens": 26974132.0,
      "step": 15545
    },
    {
      "entropy": 5.525284194946289,
      "epoch": 1.2098424431044543,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000485700943154485,
      "loss": 5.0402,
      "mean_token_accuracy": 0.19996629804372787,
      "num_tokens": 26982277.0,
      "step": 15550
    },
    {
      "entropy": 5.4517481327056885,
      "epoch": 1.210231472476172,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004856912187970195,
      "loss": 5.0079,
      "mean_token_accuracy": 0.1890098884701729,
      "num_tokens": 26991827.0,
      "step": 15555
    },
    {
      "entropy": 5.6049519062042235,
      "epoch": 1.2106205018478895,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004856814912426502,
      "loss": 5.16,
      "mean_token_accuracy": 0.18736978024244308,
      "num_tokens": 27000697.0,
      "step": 15560
    },
    {
      "entropy": 5.523280000686645,
      "epoch": 1.2110095312196072,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004856717604915245,
      "loss": 5.1039,
      "mean_token_accuracy": 0.19244175255298615,
      "num_tokens": 27009461.0,
      "step": 15565
    },
    {
      "entropy": 5.443089818954467,
      "epoch": 1.2113985605913247,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004856620265437902,
      "loss": 5.0266,
      "mean_token_accuracy": 0.19833761900663377,
      "num_tokens": 27017970.0,
      "step": 15570
    },
    {
      "entropy": 5.527230310440063,
      "epoch": 1.2117875899630421,
      "grad_norm": 1.046875,
      "learning_rate": 0.000485652289399595,
      "loss": 5.1068,
      "mean_token_accuracy": 0.19574206620454787,
      "num_tokens": 27026439.0,
      "step": 15575
    },
    {
      "entropy": 5.554113960266113,
      "epoch": 1.2121766193347598,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048564254905908655,
      "loss": 5.0589,
      "mean_token_accuracy": 0.1886896923184395,
      "num_tokens": 27034985.0,
      "step": 15580
    },
    {
      "entropy": 5.609583139419556,
      "epoch": 1.2125656487064773,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048563280552241266,
      "loss": 5.1869,
      "mean_token_accuracy": 0.18620618134737016,
      "num_tokens": 27044171.0,
      "step": 15585
    },
    {
      "entropy": 5.587800550460815,
      "epoch": 1.2129546780781948,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004856230587897212,
      "loss": 5.149,
      "mean_token_accuracy": 0.18904296308755875,
      "num_tokens": 27053219.0,
      "step": 15590
    },
    {
      "entropy": 5.486422920227051,
      "epoch": 1.2133437074499125,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004856133088611602,
      "loss": 5.0047,
      "mean_token_accuracy": 0.20015834718942643,
      "num_tokens": 27062012.0,
      "step": 15595
    },
    {
      "entropy": 5.51766939163208,
      "epoch": 1.21373273682163,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004856035557368773,
      "loss": 5.0936,
      "mean_token_accuracy": 0.19238274693489074,
      "num_tokens": 27070854.0,
      "step": 15600
    },
    {
      "entropy": 5.59365758895874,
      "epoch": 1.2141217661933477,
      "grad_norm": 1.0,
      "learning_rate": 0.00048559379941702074,
      "loss": 5.1466,
      "mean_token_accuracy": 0.19042540788650514,
      "num_tokens": 27079295.0,
      "step": 15605
    },
    {
      "entropy": 5.539794921875,
      "epoch": 1.2145107955650651,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048558403990173844,
      "loss": 5.0324,
      "mean_token_accuracy": 0.19549874514341353,
      "num_tokens": 27087909.0,
      "step": 15610
    },
    {
      "entropy": 5.467324590682983,
      "epoch": 1.2148998249367828,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048557427719117855,
      "loss": 5.0638,
      "mean_token_accuracy": 0.1950097918510437,
      "num_tokens": 27096461.0,
      "step": 15615
    },
    {
      "entropy": 5.554795408248902,
      "epoch": 1.2152888543085003,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004855645112854891,
      "loss": 5.1983,
      "mean_token_accuracy": 0.1869576409459114,
      "num_tokens": 27104846.0,
      "step": 15620
    },
    {
      "entropy": 5.564477157592774,
      "epoch": 1.2156778836802178,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004855547421848184,
      "loss": 4.9857,
      "mean_token_accuracy": 0.20039092600345612,
      "num_tokens": 27112623.0,
      "step": 15625
    },
    {
      "entropy": 5.547893333435058,
      "epoch": 1.2160669130519355,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048554496988931456,
      "loss": 5.0386,
      "mean_token_accuracy": 0.19176285564899445,
      "num_tokens": 27120752.0,
      "step": 15630
    },
    {
      "entropy": 5.595613813400268,
      "epoch": 1.216455942423653,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048553519439912597,
      "loss": 5.1936,
      "mean_token_accuracy": 0.1799352988600731,
      "num_tokens": 27129583.0,
      "step": 15635
    },
    {
      "entropy": 5.566268396377564,
      "epoch": 1.2168449717953704,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004855254157144009,
      "loss": 5.0967,
      "mean_token_accuracy": 0.19540041089057922,
      "num_tokens": 27137807.0,
      "step": 15640
    },
    {
      "entropy": 5.4816748142242435,
      "epoch": 1.2172340011670881,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004855156338352877,
      "loss": 4.963,
      "mean_token_accuracy": 0.2021173506975174,
      "num_tokens": 27146246.0,
      "step": 15645
    },
    {
      "entropy": 5.524583435058593,
      "epoch": 1.2176230305388056,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048550584876193493,
      "loss": 5.0493,
      "mean_token_accuracy": 0.19537570029497148,
      "num_tokens": 27154811.0,
      "step": 15650
    },
    {
      "entropy": 5.567844438552856,
      "epoch": 1.2180120599105233,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048549606049449085,
      "loss": 5.0601,
      "mean_token_accuracy": 0.20395116955041886,
      "num_tokens": 27162849.0,
      "step": 15655
    },
    {
      "entropy": 5.509085273742675,
      "epoch": 1.2184010892822408,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004854862690331041,
      "loss": 5.1266,
      "mean_token_accuracy": 0.18835447877645492,
      "num_tokens": 27171268.0,
      "step": 15660
    },
    {
      "entropy": 5.535284328460693,
      "epoch": 1.2187901186539585,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048547647437792327,
      "loss": 5.082,
      "mean_token_accuracy": 0.1900589644908905,
      "num_tokens": 27179571.0,
      "step": 15665
    },
    {
      "entropy": 5.515299367904663,
      "epoch": 1.219179148025676,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000485466676529097,
      "loss": 5.0555,
      "mean_token_accuracy": 0.19890601187944412,
      "num_tokens": 27188441.0,
      "step": 15670
    },
    {
      "entropy": 5.540939474105835,
      "epoch": 1.2195681773973934,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048545687548677373,
      "loss": 5.0849,
      "mean_token_accuracy": 0.19345983117818832,
      "num_tokens": 27196960.0,
      "step": 15675
    },
    {
      "entropy": 5.505633354187012,
      "epoch": 1.2199572067691111,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004854470712511025,
      "loss": 5.0875,
      "mean_token_accuracy": 0.19290500432252883,
      "num_tokens": 27205129.0,
      "step": 15680
    },
    {
      "entropy": 5.502199935913086,
      "epoch": 1.2203462361408286,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048543726382223193,
      "loss": 5.0306,
      "mean_token_accuracy": 0.19404651671648027,
      "num_tokens": 27214213.0,
      "step": 15685
    },
    {
      "entropy": 5.494686651229858,
      "epoch": 1.220735265512546,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048542745320031075,
      "loss": 5.0828,
      "mean_token_accuracy": 0.18573139607906342,
      "num_tokens": 27222447.0,
      "step": 15690
    },
    {
      "entropy": 5.549720287322998,
      "epoch": 1.2211242948842638,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048541763938548795,
      "loss": 5.0617,
      "mean_token_accuracy": 0.186423659324646,
      "num_tokens": 27230872.0,
      "step": 15695
    },
    {
      "entropy": 5.513267374038696,
      "epoch": 1.2215133242559812,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048540782237791244,
      "loss": 5.1097,
      "mean_token_accuracy": 0.190444315969944,
      "num_tokens": 27239171.0,
      "step": 15700
    },
    {
      "entropy": 5.475875663757324,
      "epoch": 1.221902353627699,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048539800217773293,
      "loss": 5.033,
      "mean_token_accuracy": 0.19151302874088288,
      "num_tokens": 27247530.0,
      "step": 15705
    },
    {
      "entropy": 5.5478339195251465,
      "epoch": 1.2222913829994164,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004853881787850988,
      "loss": 5.124,
      "mean_token_accuracy": 0.1902747079730034,
      "num_tokens": 27255579.0,
      "step": 15710
    },
    {
      "entropy": 5.4685959815979,
      "epoch": 1.2226804123711341,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048537835220015886,
      "loss": 5.0556,
      "mean_token_accuracy": 0.1953658401966095,
      "num_tokens": 27264209.0,
      "step": 15715
    },
    {
      "entropy": 5.581566905975341,
      "epoch": 1.2230694417428516,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004853685224230623,
      "loss": 5.1285,
      "mean_token_accuracy": 0.18381873220205308,
      "num_tokens": 27272424.0,
      "step": 15720
    },
    {
      "entropy": 5.476224946975708,
      "epoch": 1.223458471114569,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048535868945395825,
      "loss": 4.9385,
      "mean_token_accuracy": 0.19939552247524261,
      "num_tokens": 27281265.0,
      "step": 15725
    },
    {
      "entropy": 5.5620765209198,
      "epoch": 1.2238475004862868,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048534885329299586,
      "loss": 5.1966,
      "mean_token_accuracy": 0.19736030101776122,
      "num_tokens": 27290026.0,
      "step": 15730
    },
    {
      "entropy": 5.5719279766082765,
      "epoch": 1.2242365298580042,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004853390139403245,
      "loss": 5.1585,
      "mean_token_accuracy": 0.1858789414167404,
      "num_tokens": 27298597.0,
      "step": 15735
    },
    {
      "entropy": 5.5864521026611325,
      "epoch": 1.224625559229722,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048532917139609334,
      "loss": 5.1247,
      "mean_token_accuracy": 0.1893510937690735,
      "num_tokens": 27306763.0,
      "step": 15740
    },
    {
      "entropy": 5.53582353591919,
      "epoch": 1.2250145886014394,
      "grad_norm": 1.125,
      "learning_rate": 0.0004853193256604518,
      "loss": 5.12,
      "mean_token_accuracy": 0.19078795611858368,
      "num_tokens": 27315782.0,
      "step": 15745
    },
    {
      "entropy": 5.636805009841919,
      "epoch": 1.2254036179731569,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048530947673354924,
      "loss": 5.1667,
      "mean_token_accuracy": 0.1861603632569313,
      "num_tokens": 27324487.0,
      "step": 15750
    },
    {
      "entropy": 5.551959753036499,
      "epoch": 1.2257926473448746,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004852996246155351,
      "loss": 5.1009,
      "mean_token_accuracy": 0.1873983085155487,
      "num_tokens": 27333099.0,
      "step": 15755
    },
    {
      "entropy": 5.536805868148804,
      "epoch": 1.226181676716592,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048528976930655896,
      "loss": 5.1094,
      "mean_token_accuracy": 0.19018182158470154,
      "num_tokens": 27341697.0,
      "step": 15760
    },
    {
      "entropy": 5.549471998214722,
      "epoch": 1.2265707060883098,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048527991080677015,
      "loss": 5.0786,
      "mean_token_accuracy": 0.1906165897846222,
      "num_tokens": 27349811.0,
      "step": 15765
    },
    {
      "entropy": 5.50189175605774,
      "epoch": 1.2269597354600272,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048527004911631843,
      "loss": 5.0838,
      "mean_token_accuracy": 0.1898494318127632,
      "num_tokens": 27358257.0,
      "step": 15770
    },
    {
      "entropy": 5.5471290111541744,
      "epoch": 1.2273487648317447,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004852601842353534,
      "loss": 5.0501,
      "mean_token_accuracy": 0.19749913662672042,
      "num_tokens": 27366993.0,
      "step": 15775
    },
    {
      "entropy": 5.552340745925903,
      "epoch": 1.2277377942034624,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048525031616402476,
      "loss": 5.1013,
      "mean_token_accuracy": 0.19684531837701796,
      "num_tokens": 27375780.0,
      "step": 15780
    },
    {
      "entropy": 5.541801595687867,
      "epoch": 1.2281268235751799,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048524044490248227,
      "loss": 5.1347,
      "mean_token_accuracy": 0.19540255516767502,
      "num_tokens": 27384641.0,
      "step": 15785
    },
    {
      "entropy": 5.596535301208496,
      "epoch": 1.2285158529468976,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048523057045087557,
      "loss": 5.1093,
      "mean_token_accuracy": 0.19576272517442703,
      "num_tokens": 27392990.0,
      "step": 15790
    },
    {
      "entropy": 5.604401063919068,
      "epoch": 1.228904882318615,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048522069280935466,
      "loss": 5.2634,
      "mean_token_accuracy": 0.18289327919483184,
      "num_tokens": 27401012.0,
      "step": 15795
    },
    {
      "entropy": 5.482778787612915,
      "epoch": 1.2292939116903325,
      "grad_norm": 1.0,
      "learning_rate": 0.0004852108119780693,
      "loss": 5.0627,
      "mean_token_accuracy": 0.19397327303886414,
      "num_tokens": 27409758.0,
      "step": 15800
    },
    {
      "entropy": 5.570701932907104,
      "epoch": 1.2296829410620502,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004852009279571694,
      "loss": 5.0987,
      "mean_token_accuracy": 0.19275259524583815,
      "num_tokens": 27417592.0,
      "step": 15805
    },
    {
      "entropy": 5.532994794845581,
      "epoch": 1.2300719704337677,
      "grad_norm": 1.046875,
      "learning_rate": 0.000485191040746805,
      "loss": 5.0367,
      "mean_token_accuracy": 0.19200771898031235,
      "num_tokens": 27426822.0,
      "step": 15810
    },
    {
      "entropy": 5.472552156448364,
      "epoch": 1.2304609998054854,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004851811503471262,
      "loss": 5.0271,
      "mean_token_accuracy": 0.20359096378087999,
      "num_tokens": 27435446.0,
      "step": 15815
    },
    {
      "entropy": 5.551928281784058,
      "epoch": 1.2308500291772029,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048517125675828294,
      "loss": 5.0251,
      "mean_token_accuracy": 0.19706559777259827,
      "num_tokens": 27443156.0,
      "step": 15820
    },
    {
      "entropy": 5.618428182601929,
      "epoch": 1.2312390585489204,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048516135998042536,
      "loss": 5.1565,
      "mean_token_accuracy": 0.18697489947080612,
      "num_tokens": 27451500.0,
      "step": 15825
    },
    {
      "entropy": 5.578810691833496,
      "epoch": 1.231628087920638,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004851514600137037,
      "loss": 5.1138,
      "mean_token_accuracy": 0.19595395028591156,
      "num_tokens": 27460237.0,
      "step": 15830
    },
    {
      "entropy": 5.535499382019043,
      "epoch": 1.2320171172923555,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048514155685826807,
      "loss": 5.0631,
      "mean_token_accuracy": 0.19833202362060548,
      "num_tokens": 27468434.0,
      "step": 15835
    },
    {
      "entropy": 5.570291042327881,
      "epoch": 1.2324061466640732,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004851316505142688,
      "loss": 5.0974,
      "mean_token_accuracy": 0.19613568484783173,
      "num_tokens": 27477667.0,
      "step": 15840
    },
    {
      "entropy": 5.5753316402435305,
      "epoch": 1.2327951760357907,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048512174098185615,
      "loss": 5.0522,
      "mean_token_accuracy": 0.18449437767267227,
      "num_tokens": 27486365.0,
      "step": 15845
    },
    {
      "entropy": 5.50152735710144,
      "epoch": 1.2331842054075082,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048511182826118055,
      "loss": 5.0553,
      "mean_token_accuracy": 0.20079087167978288,
      "num_tokens": 27495310.0,
      "step": 15850
    },
    {
      "entropy": 5.442192125320434,
      "epoch": 1.2335732347792259,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048510191235239246,
      "loss": 5.0214,
      "mean_token_accuracy": 0.19757838547229767,
      "num_tokens": 27504175.0,
      "step": 15855
    },
    {
      "entropy": 5.600802373886109,
      "epoch": 1.2339622641509433,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048509199325564217,
      "loss": 5.0658,
      "mean_token_accuracy": 0.18818131685256959,
      "num_tokens": 27512042.0,
      "step": 15860
    },
    {
      "entropy": 5.594395732879638,
      "epoch": 1.234351293522661,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004850820709710803,
      "loss": 5.1685,
      "mean_token_accuracy": 0.1940791204571724,
      "num_tokens": 27520721.0,
      "step": 15865
    },
    {
      "entropy": 5.456285667419434,
      "epoch": 1.2347403228943785,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004850721454988574,
      "loss": 5.0383,
      "mean_token_accuracy": 0.18705368489027024,
      "num_tokens": 27529431.0,
      "step": 15870
    },
    {
      "entropy": 5.45005578994751,
      "epoch": 1.235129352266096,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048506221683912405,
      "loss": 4.9814,
      "mean_token_accuracy": 0.2023360788822174,
      "num_tokens": 27537380.0,
      "step": 15875
    },
    {
      "entropy": 5.468703317642212,
      "epoch": 1.2355183816378137,
      "grad_norm": 0.875,
      "learning_rate": 0.0004850522849920309,
      "loss": 5.0055,
      "mean_token_accuracy": 0.1920049324631691,
      "num_tokens": 27547487.0,
      "step": 15880
    },
    {
      "entropy": 5.599043941497802,
      "epoch": 1.2359074110095312,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048504234995772863,
      "loss": 5.0747,
      "mean_token_accuracy": 0.19443098157644273,
      "num_tokens": 27556813.0,
      "step": 15885
    },
    {
      "entropy": 5.5700986862182615,
      "epoch": 1.2362964403812489,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004850324117363681,
      "loss": 5.1899,
      "mean_token_accuracy": 0.18336285799741744,
      "num_tokens": 27566602.0,
      "step": 15890
    },
    {
      "entropy": 5.537034893035889,
      "epoch": 1.2366854697529663,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048502247032809997,
      "loss": 5.092,
      "mean_token_accuracy": 0.19569528698921204,
      "num_tokens": 27574577.0,
      "step": 15895
    },
    {
      "entropy": 5.657501077651977,
      "epoch": 1.237074499124684,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004850125257330751,
      "loss": 5.2706,
      "mean_token_accuracy": 0.1752875804901123,
      "num_tokens": 27582702.0,
      "step": 15900
    },
    {
      "entropy": 5.570566558837891,
      "epoch": 1.2374635284964015,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048500257795144446,
      "loss": 5.1372,
      "mean_token_accuracy": 0.18710373044013978,
      "num_tokens": 27591457.0,
      "step": 15905
    },
    {
      "entropy": 5.537832117080688,
      "epoch": 1.237852557868119,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000484992626983359,
      "loss": 5.0081,
      "mean_token_accuracy": 0.1988607555627823,
      "num_tokens": 27600643.0,
      "step": 15910
    },
    {
      "entropy": 5.5123209953308105,
      "epoch": 1.2382415872398367,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004849826728289696,
      "loss": 5.0033,
      "mean_token_accuracy": 0.19717362970113755,
      "num_tokens": 27608438.0,
      "step": 15915
    },
    {
      "entropy": 5.5448394298553465,
      "epoch": 1.2386306166115542,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048497271548842737,
      "loss": 5.1143,
      "mean_token_accuracy": 0.1912641257047653,
      "num_tokens": 27617149.0,
      "step": 15920
    },
    {
      "entropy": 5.499539947509765,
      "epoch": 1.2390196459832716,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004849627549618834,
      "loss": 5.0648,
      "mean_token_accuracy": 0.19102533906698227,
      "num_tokens": 27625188.0,
      "step": 15925
    },
    {
      "entropy": 5.4939899921417235,
      "epoch": 1.2394086753549893,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048495279124948886,
      "loss": 5.138,
      "mean_token_accuracy": 0.19139318466186522,
      "num_tokens": 27633865.0,
      "step": 15930
    },
    {
      "entropy": 5.5067455768585205,
      "epoch": 1.2397977047267068,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004849428243513948,
      "loss": 5.0805,
      "mean_token_accuracy": 0.19231319576501846,
      "num_tokens": 27642528.0,
      "step": 15935
    },
    {
      "entropy": 5.568923759460449,
      "epoch": 1.2401867340984245,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048493285426775255,
      "loss": 5.1669,
      "mean_token_accuracy": 0.18037957847118377,
      "num_tokens": 27651291.0,
      "step": 15940
    },
    {
      "entropy": 5.616667890548706,
      "epoch": 1.240575763470142,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004849228809987135,
      "loss": 5.1745,
      "mean_token_accuracy": 0.18740049451589585,
      "num_tokens": 27660725.0,
      "step": 15945
    },
    {
      "entropy": 5.583749723434448,
      "epoch": 1.2409647928418597,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004849129045444288,
      "loss": 5.0118,
      "mean_token_accuracy": 0.19693499952554702,
      "num_tokens": 27669282.0,
      "step": 15950
    },
    {
      "entropy": 5.501932954788208,
      "epoch": 1.2413538222135772,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004849029249050498,
      "loss": 5.0233,
      "mean_token_accuracy": 0.20107685774564743,
      "num_tokens": 27677987.0,
      "step": 15955
    },
    {
      "entropy": 5.444202280044555,
      "epoch": 1.2417428515852946,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048489294208072805,
      "loss": 5.0276,
      "mean_token_accuracy": 0.19272937178611754,
      "num_tokens": 27686409.0,
      "step": 15960
    },
    {
      "entropy": 5.506213855743408,
      "epoch": 1.2421318809570123,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048488295607161495,
      "loss": 5.0305,
      "mean_token_accuracy": 0.19374141991138458,
      "num_tokens": 27695485.0,
      "step": 15965
    },
    {
      "entropy": 5.50422043800354,
      "epoch": 1.2425209103287298,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004848729668778621,
      "loss": 4.9234,
      "mean_token_accuracy": 0.2057070791721344,
      "num_tokens": 27703962.0,
      "step": 15970
    },
    {
      "entropy": 5.542482137680054,
      "epoch": 1.2429099397004473,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004848629744996211,
      "loss": 5.0692,
      "mean_token_accuracy": 0.19228322207927703,
      "num_tokens": 27713632.0,
      "step": 15975
    },
    {
      "entropy": 5.505822467803955,
      "epoch": 1.243298969072165,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004848529789370434,
      "loss": 5.1339,
      "mean_token_accuracy": 0.1912308156490326,
      "num_tokens": 27722890.0,
      "step": 15980
    },
    {
      "entropy": 5.517408275604248,
      "epoch": 1.2436879984438824,
      "grad_norm": 1.125,
      "learning_rate": 0.0004848429801902808,
      "loss": 5.0471,
      "mean_token_accuracy": 0.19769877046346665,
      "num_tokens": 27731325.0,
      "step": 15985
    },
    {
      "entropy": 5.554013204574585,
      "epoch": 1.2440770278156001,
      "grad_norm": 1.125,
      "learning_rate": 0.00048483297825948485,
      "loss": 5.15,
      "mean_token_accuracy": 0.1872444897890091,
      "num_tokens": 27739600.0,
      "step": 15990
    },
    {
      "entropy": 5.550027799606323,
      "epoch": 1.2444660571873176,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048482297314480756,
      "loss": 5.1338,
      "mean_token_accuracy": 0.190052430331707,
      "num_tokens": 27747860.0,
      "step": 15995
    },
    {
      "entropy": 5.49091854095459,
      "epoch": 1.2448550865590353,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004848129648464006,
      "loss": 5.0387,
      "mean_token_accuracy": 0.19647593796253204,
      "num_tokens": 27757141.0,
      "step": 16000
    },
    {
      "entropy": 5.5767535209655765,
      "epoch": 1.2452441159307528,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004848029533644159,
      "loss": 5.2023,
      "mean_token_accuracy": 0.1846934124827385,
      "num_tokens": 27766832.0,
      "step": 16005
    },
    {
      "entropy": 5.5448877811431885,
      "epoch": 1.2456331453024703,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004847929386990052,
      "loss": 5.0856,
      "mean_token_accuracy": 0.18803768008947372,
      "num_tokens": 27775487.0,
      "step": 16010
    },
    {
      "entropy": 5.527425050735474,
      "epoch": 1.246022174674188,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048478292085032065,
      "loss": 5.0212,
      "mean_token_accuracy": 0.19579944312572478,
      "num_tokens": 27783601.0,
      "step": 16015
    },
    {
      "entropy": 5.565778970718384,
      "epoch": 1.2464112040459054,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004847728998185142,
      "loss": 5.1382,
      "mean_token_accuracy": 0.18673194348812103,
      "num_tokens": 27792931.0,
      "step": 16020
    },
    {
      "entropy": 5.617130088806152,
      "epoch": 1.246800233417623,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048476287560373786,
      "loss": 5.0782,
      "mean_token_accuracy": 0.19357449412345887,
      "num_tokens": 27801388.0,
      "step": 16025
    },
    {
      "entropy": 5.510725593566894,
      "epoch": 1.2471892627893406,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048475284820614375,
      "loss": 5.0593,
      "mean_token_accuracy": 0.19279939085245132,
      "num_tokens": 27810047.0,
      "step": 16030
    },
    {
      "entropy": 5.504441165924073,
      "epoch": 1.247578292161058,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048474281762588407,
      "loss": 5.0583,
      "mean_token_accuracy": 0.19652713984251022,
      "num_tokens": 27818800.0,
      "step": 16035
    },
    {
      "entropy": 5.473054218292236,
      "epoch": 1.2479673215327758,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004847327838631109,
      "loss": 5.0337,
      "mean_token_accuracy": 0.18869485706090927,
      "num_tokens": 27827623.0,
      "step": 16040
    },
    {
      "entropy": 5.455634498596192,
      "epoch": 1.2483563509044933,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004847227469179766,
      "loss": 5.0172,
      "mean_token_accuracy": 0.20616044998168945,
      "num_tokens": 27836491.0,
      "step": 16045
    },
    {
      "entropy": 5.55037031173706,
      "epoch": 1.248745380276211,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004847127067906334,
      "loss": 5.0647,
      "mean_token_accuracy": 0.19604194313287734,
      "num_tokens": 27845563.0,
      "step": 16050
    },
    {
      "entropy": 5.508697891235352,
      "epoch": 1.2491344096479284,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048470266348123365,
      "loss": 5.0165,
      "mean_token_accuracy": 0.1985260009765625,
      "num_tokens": 27854263.0,
      "step": 16055
    },
    {
      "entropy": 5.508735704421997,
      "epoch": 1.249523439019646,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004846926169899298,
      "loss": 5.0478,
      "mean_token_accuracy": 0.19574496001005173,
      "num_tokens": 27862868.0,
      "step": 16060
    },
    {
      "entropy": 5.56677656173706,
      "epoch": 1.2499124683913636,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048468256731687426,
      "loss": 5.1473,
      "mean_token_accuracy": 0.19273220747709274,
      "num_tokens": 27872053.0,
      "step": 16065
    },
    {
      "entropy": 5.505532360076904,
      "epoch": 1.250301497763081,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004846725144622194,
      "loss": 4.9724,
      "mean_token_accuracy": 0.19897133857011795,
      "num_tokens": 27880953.0,
      "step": 16070
    },
    {
      "entropy": 5.513091611862182,
      "epoch": 1.2506905271347986,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000484662458426118,
      "loss": 5.0641,
      "mean_token_accuracy": 0.19407759010791778,
      "num_tokens": 27890102.0,
      "step": 16075
    },
    {
      "entropy": 5.562303781509399,
      "epoch": 1.2510795565065163,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048465239920872247,
      "loss": 5.1647,
      "mean_token_accuracy": 0.18874465078115463,
      "num_tokens": 27899305.0,
      "step": 16080
    },
    {
      "entropy": 5.616161298751831,
      "epoch": 1.2514685858782337,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048464233681018545,
      "loss": 5.0876,
      "mean_token_accuracy": 0.19358059614896775,
      "num_tokens": 27907476.0,
      "step": 16085
    },
    {
      "entropy": 5.568417358398437,
      "epoch": 1.2518576152499514,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004846322712306596,
      "loss": 5.1609,
      "mean_token_accuracy": 0.19405679553747177,
      "num_tokens": 27915353.0,
      "step": 16090
    },
    {
      "entropy": 5.470607089996338,
      "epoch": 1.252246644621669,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048462220247029783,
      "loss": 5.0789,
      "mean_token_accuracy": 0.19470122903585435,
      "num_tokens": 27923980.0,
      "step": 16095
    },
    {
      "entropy": 5.520428609848023,
      "epoch": 1.2526356739933866,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048461213052925265,
      "loss": 5.1516,
      "mean_token_accuracy": 0.19147412478923798,
      "num_tokens": 27932760.0,
      "step": 16100
    },
    {
      "entropy": 5.53990330696106,
      "epoch": 1.253024703365104,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004846020554076771,
      "loss": 5.0484,
      "mean_token_accuracy": 0.19671657383441926,
      "num_tokens": 27941328.0,
      "step": 16105
    },
    {
      "entropy": 5.605533838272095,
      "epoch": 1.2534137327368216,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004845919771057239,
      "loss": 5.2211,
      "mean_token_accuracy": 0.1857961595058441,
      "num_tokens": 27949597.0,
      "step": 16110
    },
    {
      "entropy": 5.57187852859497,
      "epoch": 1.2538027621085392,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004845818956235462,
      "loss": 5.0994,
      "mean_token_accuracy": 0.19588743150234222,
      "num_tokens": 27958798.0,
      "step": 16115
    },
    {
      "entropy": 5.581532001495361,
      "epoch": 1.2541917914802567,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048457181096129664,
      "loss": 5.037,
      "mean_token_accuracy": 0.19875202625989913,
      "num_tokens": 27967582.0,
      "step": 16120
    },
    {
      "entropy": 5.521014785766601,
      "epoch": 1.2545808208519742,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004845617231191285,
      "loss": 5.0191,
      "mean_token_accuracy": 0.19087808430194855,
      "num_tokens": 27975959.0,
      "step": 16125
    },
    {
      "entropy": 5.492209243774414,
      "epoch": 1.254969850223692,
      "grad_norm": 1.125,
      "learning_rate": 0.0004845516320971948,
      "loss": 5.0496,
      "mean_token_accuracy": 0.19797075986862184,
      "num_tokens": 27983980.0,
      "step": 16130
    },
    {
      "entropy": 5.497616147994995,
      "epoch": 1.2553588795954094,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004845415378956486,
      "loss": 4.9526,
      "mean_token_accuracy": 0.19749454408884048,
      "num_tokens": 27992563.0,
      "step": 16135
    },
    {
      "entropy": 5.464643430709839,
      "epoch": 1.255747908967127,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000484531440514643,
      "loss": 4.956,
      "mean_token_accuracy": 0.20556332170963287,
      "num_tokens": 28001457.0,
      "step": 16140
    },
    {
      "entropy": 5.58230504989624,
      "epoch": 1.2561369383388445,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048452133995433136,
      "loss": 5.1135,
      "mean_token_accuracy": 0.19272407442331313,
      "num_tokens": 28009507.0,
      "step": 16145
    },
    {
      "entropy": 5.463147306442261,
      "epoch": 1.2565259677105622,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004845112362148668,
      "loss": 5.0729,
      "mean_token_accuracy": 0.18405213952064514,
      "num_tokens": 28017858.0,
      "step": 16150
    },
    {
      "entropy": 5.565900707244873,
      "epoch": 1.2569149970822797,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004845011292964028,
      "loss": 5.1336,
      "mean_token_accuracy": 0.18732133656740188,
      "num_tokens": 28026832.0,
      "step": 16155
    },
    {
      "entropy": 5.565466356277466,
      "epoch": 1.2573040264539972,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048449101919909265,
      "loss": 5.0058,
      "mean_token_accuracy": 0.19498329609632492,
      "num_tokens": 28034546.0,
      "step": 16160
    },
    {
      "entropy": 5.492333984375,
      "epoch": 1.257693055825715,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048448090592308955,
      "loss": 5.0886,
      "mean_token_accuracy": 0.1896803081035614,
      "num_tokens": 28043400.0,
      "step": 16165
    },
    {
      "entropy": 5.462855005264283,
      "epoch": 1.2580820851974324,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004844707894685473,
      "loss": 5.0482,
      "mean_token_accuracy": 0.19748084545135497,
      "num_tokens": 28052764.0,
      "step": 16170
    },
    {
      "entropy": 5.562974691390991,
      "epoch": 1.2584711145691498,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004844606698356192,
      "loss": 5.129,
      "mean_token_accuracy": 0.18233443349599837,
      "num_tokens": 28061859.0,
      "step": 16175
    },
    {
      "entropy": 5.514362001419068,
      "epoch": 1.2588601439408675,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004844505470244588,
      "loss": 5.0213,
      "mean_token_accuracy": 0.19929979592561722,
      "num_tokens": 28070935.0,
      "step": 16180
    },
    {
      "entropy": 5.561143207550049,
      "epoch": 1.2592491733125852,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004844404210352197,
      "loss": 5.0785,
      "mean_token_accuracy": 0.18881667107343675,
      "num_tokens": 28079864.0,
      "step": 16185
    },
    {
      "entropy": 5.458213949203492,
      "epoch": 1.2596382026843027,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048443029186805554,
      "loss": 4.9516,
      "mean_token_accuracy": 0.1965406596660614,
      "num_tokens": 28088938.0,
      "step": 16190
    },
    {
      "entropy": 5.522599506378174,
      "epoch": 1.2600272320560202,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004844201595231201,
      "loss": 5.0541,
      "mean_token_accuracy": 0.193293234705925,
      "num_tokens": 28097213.0,
      "step": 16195
    },
    {
      "entropy": 5.509239721298218,
      "epoch": 1.2604162614277379,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048441002400056706,
      "loss": 5.0952,
      "mean_token_accuracy": 0.19133833348751067,
      "num_tokens": 28105187.0,
      "step": 16200
    },
    {
      "entropy": 5.513419342041016,
      "epoch": 1.2608052907994554,
      "grad_norm": 1.125,
      "learning_rate": 0.0004843998853005502,
      "loss": 5.0403,
      "mean_token_accuracy": 0.19761250168085098,
      "num_tokens": 28113238.0,
      "step": 16205
    },
    {
      "entropy": 5.474771738052368,
      "epoch": 1.2611943201711728,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004843897434232233,
      "loss": 4.9697,
      "mean_token_accuracy": 0.19867065846920012,
      "num_tokens": 28121502.0,
      "step": 16210
    },
    {
      "entropy": 5.519422817230224,
      "epoch": 1.2615833495428905,
      "grad_norm": 1.125,
      "learning_rate": 0.00048437959836874047,
      "loss": 5.0678,
      "mean_token_accuracy": 0.19417955577373505,
      "num_tokens": 28130269.0,
      "step": 16215
    },
    {
      "entropy": 5.572734785079956,
      "epoch": 1.261972378914608,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048436945013725544,
      "loss": 5.2184,
      "mean_token_accuracy": 0.1829194575548172,
      "num_tokens": 28139482.0,
      "step": 16220
    },
    {
      "entropy": 5.551571798324585,
      "epoch": 1.2623614082863255,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048435929872892226,
      "loss": 5.1505,
      "mean_token_accuracy": 0.18855098485946656,
      "num_tokens": 28148443.0,
      "step": 16225
    },
    {
      "entropy": 5.563281917572022,
      "epoch": 1.2627504376580432,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004843491441438949,
      "loss": 5.0011,
      "mean_token_accuracy": 0.2016681104898453,
      "num_tokens": 28156807.0,
      "step": 16230
    },
    {
      "entropy": 5.53257040977478,
      "epoch": 1.2631394670297609,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048433898638232755,
      "loss": 5.0728,
      "mean_token_accuracy": 0.1981820896267891,
      "num_tokens": 28165019.0,
      "step": 16235
    },
    {
      "entropy": 5.522033977508545,
      "epoch": 1.2635284964014784,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004843288254443742,
      "loss": 5.1131,
      "mean_token_accuracy": 0.189405857026577,
      "num_tokens": 28173505.0,
      "step": 16240
    },
    {
      "entropy": 5.575769424438477,
      "epoch": 1.2639175257731958,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004843186613301892,
      "loss": 5.1428,
      "mean_token_accuracy": 0.1881885603070259,
      "num_tokens": 28182524.0,
      "step": 16245
    },
    {
      "entropy": 5.606953716278076,
      "epoch": 1.2643065551449135,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004843084940399266,
      "loss": 5.1878,
      "mean_token_accuracy": 0.1839436560869217,
      "num_tokens": 28191093.0,
      "step": 16250
    },
    {
      "entropy": 5.569440460205078,
      "epoch": 1.264695584516631,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004842983235737408,
      "loss": 5.0548,
      "mean_token_accuracy": 0.19338791817426682,
      "num_tokens": 28199092.0,
      "step": 16255
    },
    {
      "entropy": 5.618682241439819,
      "epoch": 1.2650846138883485,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004842881499317861,
      "loss": 5.1549,
      "mean_token_accuracy": 0.19013646245002747,
      "num_tokens": 28207341.0,
      "step": 16260
    },
    {
      "entropy": 5.5635782241821286,
      "epoch": 1.2654736432600662,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004842779731142168,
      "loss": 5.0651,
      "mean_token_accuracy": 0.1950272426009178,
      "num_tokens": 28216370.0,
      "step": 16265
    },
    {
      "entropy": 5.577887392044067,
      "epoch": 1.2658626726317836,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048426779312118735,
      "loss": 5.1155,
      "mean_token_accuracy": 0.19069255739450455,
      "num_tokens": 28224554.0,
      "step": 16270
    },
    {
      "entropy": 5.606639003753662,
      "epoch": 1.2662517020035013,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004842576099528522,
      "loss": 5.1281,
      "mean_token_accuracy": 0.18990982323884964,
      "num_tokens": 28232672.0,
      "step": 16275
    },
    {
      "entropy": 5.471338891983033,
      "epoch": 1.2666407313752188,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004842474236093659,
      "loss": 4.9518,
      "mean_token_accuracy": 0.19516842812299728,
      "num_tokens": 28240730.0,
      "step": 16280
    },
    {
      "entropy": 5.499918651580811,
      "epoch": 1.2670297607469365,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048423723409088306,
      "loss": 5.0683,
      "mean_token_accuracy": 0.19348925352096558,
      "num_tokens": 28248989.0,
      "step": 16285
    },
    {
      "entropy": 5.552557325363159,
      "epoch": 1.267418790118654,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004842270413975582,
      "loss": 5.159,
      "mean_token_accuracy": 0.19127894192934036,
      "num_tokens": 28256900.0,
      "step": 16290
    },
    {
      "entropy": 5.577032136917114,
      "epoch": 1.2678078194903715,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000484216845529546,
      "loss": 5.1036,
      "mean_token_accuracy": 0.18557945638895035,
      "num_tokens": 28265832.0,
      "step": 16295
    },
    {
      "entropy": 5.497867584228516,
      "epoch": 1.2681968488620892,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048420664648700113,
      "loss": 4.9845,
      "mean_token_accuracy": 0.19498049169778825,
      "num_tokens": 28274661.0,
      "step": 16300
    },
    {
      "entropy": 5.578892612457276,
      "epoch": 1.2685858782338066,
      "grad_norm": 1.125,
      "learning_rate": 0.0004841964442700784,
      "loss": 5.058,
      "mean_token_accuracy": 0.19404691457748413,
      "num_tokens": 28283303.0,
      "step": 16305
    },
    {
      "entropy": 5.535975551605224,
      "epoch": 1.2689749076055241,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048418623887893264,
      "loss": 4.945,
      "mean_token_accuracy": 0.19633139222860335,
      "num_tokens": 28291660.0,
      "step": 16310
    },
    {
      "entropy": 5.548477458953857,
      "epoch": 1.2693639369772418,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048417603031371867,
      "loss": 5.1173,
      "mean_token_accuracy": 0.18814481496810914,
      "num_tokens": 28301314.0,
      "step": 16315
    },
    {
      "entropy": 5.553690099716187,
      "epoch": 1.2697529663489593,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004841658185745913,
      "loss": 5.0718,
      "mean_token_accuracy": 0.18898946791887283,
      "num_tokens": 28310044.0,
      "step": 16320
    },
    {
      "entropy": 5.691943979263305,
      "epoch": 1.270141995720677,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048415560366170564,
      "loss": 5.1708,
      "mean_token_accuracy": 0.19910440742969512,
      "num_tokens": 28319773.0,
      "step": 16325
    },
    {
      "entropy": 5.5080399990081785,
      "epoch": 1.2705310250923945,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004841453855752165,
      "loss": 4.9575,
      "mean_token_accuracy": 0.19963991940021514,
      "num_tokens": 28328070.0,
      "step": 16330
    },
    {
      "entropy": 5.53001856803894,
      "epoch": 1.2709200544641122,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004841351643152791,
      "loss": 5.0622,
      "mean_token_accuracy": 0.20209717750549316,
      "num_tokens": 28337636.0,
      "step": 16335
    },
    {
      "entropy": 5.455282497406006,
      "epoch": 1.2713090838358296,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004841249398820485,
      "loss": 5.062,
      "mean_token_accuracy": 0.19543709605932236,
      "num_tokens": 28346609.0,
      "step": 16340
    },
    {
      "entropy": 5.536366033554077,
      "epoch": 1.271698113207547,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004841147122756797,
      "loss": 5.0862,
      "mean_token_accuracy": 0.19476913362741471,
      "num_tokens": 28355395.0,
      "step": 16345
    },
    {
      "entropy": 5.528757095336914,
      "epoch": 1.2720871425792648,
      "grad_norm": 1.078125,
      "learning_rate": 0.000484104481496328,
      "loss": 5.1679,
      "mean_token_accuracy": 0.1870650976896286,
      "num_tokens": 28364447.0,
      "step": 16350
    },
    {
      "entropy": 5.583292388916016,
      "epoch": 1.2724761719509823,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004840942475441486,
      "loss": 5.1024,
      "mean_token_accuracy": 0.19182195514440536,
      "num_tokens": 28372872.0,
      "step": 16355
    },
    {
      "entropy": 5.542284059524536,
      "epoch": 1.2728652013226998,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004840840104192969,
      "loss": 4.9513,
      "mean_token_accuracy": 0.1970968648791313,
      "num_tokens": 28380840.0,
      "step": 16360
    },
    {
      "entropy": 5.4953296184539795,
      "epoch": 1.2732542306944175,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000484073770121928,
      "loss": 5.015,
      "mean_token_accuracy": 0.19180364459753035,
      "num_tokens": 28390417.0,
      "step": 16365
    },
    {
      "entropy": 5.53775954246521,
      "epoch": 1.273643260066135,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004840635266521975,
      "loss": 5.0687,
      "mean_token_accuracy": 0.1881555885076523,
      "num_tokens": 28400101.0,
      "step": 16370
    },
    {
      "entropy": 5.488388299942017,
      "epoch": 1.2740322894378526,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004840532800102607,
      "loss": 5.0128,
      "mean_token_accuracy": 0.19763373881578444,
      "num_tokens": 28408336.0,
      "step": 16375
    },
    {
      "entropy": 5.500676107406616,
      "epoch": 1.27442131880957,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048404303019627314,
      "loss": 5.0259,
      "mean_token_accuracy": 0.19480105340480805,
      "num_tokens": 28416985.0,
      "step": 16380
    },
    {
      "entropy": 5.526996374130249,
      "epoch": 1.2748103481812878,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048403277721039036,
      "loss": 5.1793,
      "mean_token_accuracy": 0.18110944628715514,
      "num_tokens": 28425543.0,
      "step": 16385
    },
    {
      "entropy": 5.578274250030518,
      "epoch": 1.2751993775530053,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004840225210527679,
      "loss": 5.1186,
      "mean_token_accuracy": 0.1899320051074028,
      "num_tokens": 28435151.0,
      "step": 16390
    },
    {
      "entropy": 5.567580318450927,
      "epoch": 1.2755884069247228,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004840122617235613,
      "loss": 5.114,
      "mean_token_accuracy": 0.18739798814058303,
      "num_tokens": 28444535.0,
      "step": 16395
    },
    {
      "entropy": 5.573686122894287,
      "epoch": 1.2759774362964404,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004840019992229263,
      "loss": 5.0597,
      "mean_token_accuracy": 0.19457401782274247,
      "num_tokens": 28452852.0,
      "step": 16400
    },
    {
      "entropy": 5.532476091384888,
      "epoch": 1.276366465668158,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048399173355101867,
      "loss": 5.0991,
      "mean_token_accuracy": 0.18772910237312318,
      "num_tokens": 28462573.0,
      "step": 16405
    },
    {
      "entropy": 5.593065690994263,
      "epoch": 1.2767554950398754,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048398146470799417,
      "loss": 5.1393,
      "mean_token_accuracy": 0.1841592386364937,
      "num_tokens": 28471714.0,
      "step": 16410
    },
    {
      "entropy": 5.541199827194214,
      "epoch": 1.277144524411593,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048397119269400846,
      "loss": 5.0328,
      "mean_token_accuracy": 0.19867901206016542,
      "num_tokens": 28479989.0,
      "step": 16415
    },
    {
      "entropy": 5.534481573104858,
      "epoch": 1.2775335537833106,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048396091750921766,
      "loss": 5.0864,
      "mean_token_accuracy": 0.18902281373739244,
      "num_tokens": 28488707.0,
      "step": 16420
    },
    {
      "entropy": 5.591275930404663,
      "epoch": 1.2779225831550283,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004839506391537774,
      "loss": 5.2183,
      "mean_token_accuracy": 0.18998702466487885,
      "num_tokens": 28497228.0,
      "step": 16425
    },
    {
      "entropy": 5.63020544052124,
      "epoch": 1.2783116125267457,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004839403576278438,
      "loss": 5.0686,
      "mean_token_accuracy": 0.1960077852010727,
      "num_tokens": 28505604.0,
      "step": 16430
    },
    {
      "entropy": 5.511933994293213,
      "epoch": 1.2787006418984634,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004839300729315729,
      "loss": 5.1281,
      "mean_token_accuracy": 0.1891779735684395,
      "num_tokens": 28514798.0,
      "step": 16435
    },
    {
      "entropy": 5.593146705627442,
      "epoch": 1.279089671270181,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004839197850651206,
      "loss": 5.1186,
      "mean_token_accuracy": 0.19475816041231156,
      "num_tokens": 28523883.0,
      "step": 16440
    },
    {
      "entropy": 5.578013324737549,
      "epoch": 1.2794787006418984,
      "grad_norm": 1.125,
      "learning_rate": 0.00048390949402864317,
      "loss": 5.1076,
      "mean_token_accuracy": 0.18868862837553024,
      "num_tokens": 28532290.0,
      "step": 16445
    },
    {
      "entropy": 5.508526039123535,
      "epoch": 1.279867730013616,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048389919982229666,
      "loss": 5.0226,
      "mean_token_accuracy": 0.19841886758804322,
      "num_tokens": 28540775.0,
      "step": 16450
    },
    {
      "entropy": 5.5292247295379635,
      "epoch": 1.2802567593853336,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048388890244623723,
      "loss": 5.1331,
      "mean_token_accuracy": 0.187634414434433,
      "num_tokens": 28549517.0,
      "step": 16455
    },
    {
      "entropy": 5.549252271652222,
      "epoch": 1.280645788757051,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004838786019006213,
      "loss": 5.0815,
      "mean_token_accuracy": 0.19291755855083464,
      "num_tokens": 28557708.0,
      "step": 16460
    },
    {
      "entropy": 5.596685409545898,
      "epoch": 1.2810348181287687,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048386829818560493,
      "loss": 5.0962,
      "mean_token_accuracy": 0.19002194255590438,
      "num_tokens": 28566912.0,
      "step": 16465
    },
    {
      "entropy": 5.523095035552979,
      "epoch": 1.2814238475004862,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048385799130134464,
      "loss": 5.1,
      "mean_token_accuracy": 0.19242098927497864,
      "num_tokens": 28577104.0,
      "step": 16470
    },
    {
      "entropy": 5.517349720001221,
      "epoch": 1.281812876872204,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004838476812479968,
      "loss": 5.095,
      "mean_token_accuracy": 0.19091260880231858,
      "num_tokens": 28585984.0,
      "step": 16475
    },
    {
      "entropy": 5.587490749359131,
      "epoch": 1.2822019062439214,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048383736802571774,
      "loss": 5.1146,
      "mean_token_accuracy": 0.189076167345047,
      "num_tokens": 28594945.0,
      "step": 16480
    },
    {
      "entropy": 5.652865934371948,
      "epoch": 1.282590935615639,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000483827051634664,
      "loss": 5.1442,
      "mean_token_accuracy": 0.18616907447576522,
      "num_tokens": 28603546.0,
      "step": 16485
    },
    {
      "entropy": 5.576054477691651,
      "epoch": 1.2829799649873566,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048381673207499224,
      "loss": 5.0866,
      "mean_token_accuracy": 0.19676136821508408,
      "num_tokens": 28612057.0,
      "step": 16490
    },
    {
      "entropy": 5.484951782226562,
      "epoch": 1.283368994359074,
      "grad_norm": 1.125,
      "learning_rate": 0.0004838064093468588,
      "loss": 5.0424,
      "mean_token_accuracy": 0.19007549583911895,
      "num_tokens": 28620945.0,
      "step": 16495
    },
    {
      "entropy": 5.45639934539795,
      "epoch": 1.2837580237307917,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004837960834504206,
      "loss": 4.9543,
      "mean_token_accuracy": 0.2014954537153244,
      "num_tokens": 28629028.0,
      "step": 16500
    },
    {
      "entropy": 5.480666828155518,
      "epoch": 1.2841470531025092,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004837857543858341,
      "loss": 5.0174,
      "mean_token_accuracy": 0.19504469335079194,
      "num_tokens": 28637533.0,
      "step": 16505
    },
    {
      "entropy": 5.570009374618531,
      "epoch": 1.2845360824742267,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004837754221532561,
      "loss": 5.1485,
      "mean_token_accuracy": 0.19065336585044862,
      "num_tokens": 28646906.0,
      "step": 16510
    },
    {
      "entropy": 5.549227476119995,
      "epoch": 1.2849251118459444,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048376508675284334,
      "loss": 5.0196,
      "mean_token_accuracy": 0.2002625435590744,
      "num_tokens": 28655597.0,
      "step": 16515
    },
    {
      "entropy": 5.5836098194122314,
      "epoch": 1.2853141412176619,
      "grad_norm": 1.125,
      "learning_rate": 0.00048375474818475274,
      "loss": 5.1723,
      "mean_token_accuracy": 0.1845028892159462,
      "num_tokens": 28664806.0,
      "step": 16520
    },
    {
      "entropy": 5.490910148620605,
      "epoch": 1.2857031705893796,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048374440644914104,
      "loss": 5.002,
      "mean_token_accuracy": 0.2011221617460251,
      "num_tokens": 28673610.0,
      "step": 16525
    },
    {
      "entropy": 5.560319662094116,
      "epoch": 1.286092199961097,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048373406154616523,
      "loss": 5.0577,
      "mean_token_accuracy": 0.1932368442416191,
      "num_tokens": 28681726.0,
      "step": 16530
    },
    {
      "entropy": 5.61279525756836,
      "epoch": 1.2864812293328147,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048372371347598226,
      "loss": 5.2267,
      "mean_token_accuracy": 0.17939860820770265,
      "num_tokens": 28691014.0,
      "step": 16535
    },
    {
      "entropy": 5.566826486587525,
      "epoch": 1.2868702587045322,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048371336223874923,
      "loss": 5.0934,
      "mean_token_accuracy": 0.1858469918370247,
      "num_tokens": 28699304.0,
      "step": 16540
    },
    {
      "entropy": 5.596908235549927,
      "epoch": 1.2872592880762497,
      "grad_norm": 1.09375,
      "learning_rate": 0.000483703007834623,
      "loss": 5.0186,
      "mean_token_accuracy": 0.2008773237466812,
      "num_tokens": 28707950.0,
      "step": 16545
    },
    {
      "entropy": 5.502289152145385,
      "epoch": 1.2876483174479674,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004836926502637609,
      "loss": 5.0319,
      "mean_token_accuracy": 0.20044110715389252,
      "num_tokens": 28716678.0,
      "step": 16550
    },
    {
      "entropy": 5.4846220970153805,
      "epoch": 1.2880373468196848,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048368228952632005,
      "loss": 5.0597,
      "mean_token_accuracy": 0.20138204991817474,
      "num_tokens": 28725658.0,
      "step": 16555
    },
    {
      "entropy": 5.578532743453979,
      "epoch": 1.2884263761914023,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048367192562245756,
      "loss": 5.0942,
      "mean_token_accuracy": 0.19462292045354843,
      "num_tokens": 28733859.0,
      "step": 16560
    },
    {
      "entropy": 5.481110858917236,
      "epoch": 1.28881540556312,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048366155855233067,
      "loss": 4.9912,
      "mean_token_accuracy": 0.19842422753572464,
      "num_tokens": 28741941.0,
      "step": 16565
    },
    {
      "entropy": 5.494843626022339,
      "epoch": 1.2892044349348377,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004836511883160968,
      "loss": 4.9263,
      "mean_token_accuracy": 0.20172116607427598,
      "num_tokens": 28750296.0,
      "step": 16570
    },
    {
      "entropy": 5.427780246734619,
      "epoch": 1.2895934643065552,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004836408149139133,
      "loss": 4.9174,
      "mean_token_accuracy": 0.2067351758480072,
      "num_tokens": 28758359.0,
      "step": 16575
    },
    {
      "entropy": 5.56018590927124,
      "epoch": 1.2899824936782727,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048363043834593743,
      "loss": 5.1139,
      "mean_token_accuracy": 0.19459905922412873,
      "num_tokens": 28766944.0,
      "step": 16580
    },
    {
      "entropy": 5.56161413192749,
      "epoch": 1.2903715230499904,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004836200586123268,
      "loss": 5.1306,
      "mean_token_accuracy": 0.18650305718183519,
      "num_tokens": 28775897.0,
      "step": 16585
    },
    {
      "entropy": 5.526115798950196,
      "epoch": 1.2907605524217078,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048360967571323875,
      "loss": 5.0453,
      "mean_token_accuracy": 0.19613178968429565,
      "num_tokens": 28784509.0,
      "step": 16590
    },
    {
      "entropy": 5.468918418884277,
      "epoch": 1.2911495817934253,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048359928964883094,
      "loss": 4.9234,
      "mean_token_accuracy": 0.2010936215519905,
      "num_tokens": 28793347.0,
      "step": 16595
    },
    {
      "entropy": 5.491485834121704,
      "epoch": 1.291538611165143,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000483588900419261,
      "loss": 5.0222,
      "mean_token_accuracy": 0.2012392044067383,
      "num_tokens": 28801100.0,
      "step": 16600
    },
    {
      "entropy": 5.45463752746582,
      "epoch": 1.2919276405368605,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004835785080246864,
      "loss": 4.992,
      "mean_token_accuracy": 0.20365075767040253,
      "num_tokens": 28809508.0,
      "step": 16605
    },
    {
      "entropy": 5.504297828674316,
      "epoch": 1.292316669908578,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000483568112465265,
      "loss": 5.0794,
      "mean_token_accuracy": 0.19034943729639053,
      "num_tokens": 28819263.0,
      "step": 16610
    },
    {
      "entropy": 5.638720369338989,
      "epoch": 1.2927056992802957,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048355771374115436,
      "loss": 5.1598,
      "mean_token_accuracy": 0.17853228002786636,
      "num_tokens": 28827316.0,
      "step": 16615
    },
    {
      "entropy": 5.609198760986328,
      "epoch": 1.2930947286520134,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004835473118525125,
      "loss": 5.1773,
      "mean_token_accuracy": 0.19016696363687516,
      "num_tokens": 28836272.0,
      "step": 16620
    },
    {
      "entropy": 5.54950795173645,
      "epoch": 1.2934837580237308,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004835369067994971,
      "loss": 5.0624,
      "mean_token_accuracy": 0.19191914945840835,
      "num_tokens": 28845180.0,
      "step": 16625
    },
    {
      "entropy": 5.490755653381347,
      "epoch": 1.2938727873954483,
      "grad_norm": 1.109375,
      "learning_rate": 0.000483526498582266,
      "loss": 5.0105,
      "mean_token_accuracy": 0.18844596296548843,
      "num_tokens": 28853903.0,
      "step": 16630
    },
    {
      "entropy": 5.484728288650513,
      "epoch": 1.294261816767166,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048351608720097715,
      "loss": 5.0215,
      "mean_token_accuracy": 0.19810349941253663,
      "num_tokens": 28863071.0,
      "step": 16635
    },
    {
      "entropy": 5.594375228881836,
      "epoch": 1.2946508461388835,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048350567265578867,
      "loss": 5.1146,
      "mean_token_accuracy": 0.1831539422273636,
      "num_tokens": 28871620.0,
      "step": 16640
    },
    {
      "entropy": 5.625013828277588,
      "epoch": 1.295039875510601,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004834952549468584,
      "loss": 5.1708,
      "mean_token_accuracy": 0.18477946668863296,
      "num_tokens": 28879940.0,
      "step": 16645
    },
    {
      "entropy": 5.598568916320801,
      "epoch": 1.2954289048823187,
      "grad_norm": 1.0,
      "learning_rate": 0.0004834848340743446,
      "loss": 5.1812,
      "mean_token_accuracy": 0.18652515709400178,
      "num_tokens": 28889012.0,
      "step": 16650
    },
    {
      "entropy": 5.5727862358093265,
      "epoch": 1.2958179342540361,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004834744100384052,
      "loss": 5.0541,
      "mean_token_accuracy": 0.1956882208585739,
      "num_tokens": 28897312.0,
      "step": 16655
    },
    {
      "entropy": 5.559055662155151,
      "epoch": 1.2962069636257538,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004834639828391984,
      "loss": 5.013,
      "mean_token_accuracy": 0.19725292325019836,
      "num_tokens": 28905977.0,
      "step": 16660
    },
    {
      "entropy": 5.47237195968628,
      "epoch": 1.2965959929974713,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048345355247688256,
      "loss": 5.0015,
      "mean_token_accuracy": 0.1998746797442436,
      "num_tokens": 28913979.0,
      "step": 16665
    },
    {
      "entropy": 5.498237276077271,
      "epoch": 1.296985022369189,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004834431189516158,
      "loss": 5.0455,
      "mean_token_accuracy": 0.19905532151460648,
      "num_tokens": 28922090.0,
      "step": 16670
    },
    {
      "entropy": 5.600349092483521,
      "epoch": 1.2973740517409065,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004834326822635565,
      "loss": 5.18,
      "mean_token_accuracy": 0.1842682808637619,
      "num_tokens": 28930816.0,
      "step": 16675
    },
    {
      "entropy": 5.59349570274353,
      "epoch": 1.297763081112624,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048342224241286296,
      "loss": 5.139,
      "mean_token_accuracy": 0.18491749316453934,
      "num_tokens": 28939125.0,
      "step": 16680
    },
    {
      "entropy": 5.4965204238891605,
      "epoch": 1.2981521104843416,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004834117993996937,
      "loss": 5.0339,
      "mean_token_accuracy": 0.20044483840465546,
      "num_tokens": 28947783.0,
      "step": 16685
    },
    {
      "entropy": 5.636521339416504,
      "epoch": 1.2985411398560591,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004834013532242071,
      "loss": 5.1641,
      "mean_token_accuracy": 0.19029086381196975,
      "num_tokens": 28957305.0,
      "step": 16690
    },
    {
      "entropy": 5.587459707260132,
      "epoch": 1.2989301692277766,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004833909038865616,
      "loss": 5.1096,
      "mean_token_accuracy": 0.1861346408724785,
      "num_tokens": 28965757.0,
      "step": 16695
    },
    {
      "entropy": 5.570179510116577,
      "epoch": 1.2993191985994943,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004833804513869159,
      "loss": 5.1033,
      "mean_token_accuracy": 0.19049333781003952,
      "num_tokens": 28974846.0,
      "step": 16700
    },
    {
      "entropy": 5.45418267250061,
      "epoch": 1.2997082279712118,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004833699957254284,
      "loss": 4.9614,
      "mean_token_accuracy": 0.19327454566955565,
      "num_tokens": 28984060.0,
      "step": 16705
    },
    {
      "entropy": 5.532863664627075,
      "epoch": 1.3000972573429295,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000483359536902258,
      "loss": 5.0034,
      "mean_token_accuracy": 0.1919676885008812,
      "num_tokens": 28992705.0,
      "step": 16710
    },
    {
      "entropy": 5.561108541488648,
      "epoch": 1.300486286714647,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004833490749175632,
      "loss": 5.1151,
      "mean_token_accuracy": 0.19143617451190947,
      "num_tokens": 29001306.0,
      "step": 16715
    },
    {
      "entropy": 5.519209957122802,
      "epoch": 1.3008753160863646,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048333860977150286,
      "loss": 5.139,
      "mean_token_accuracy": 0.18987470716238022,
      "num_tokens": 29010371.0,
      "step": 16720
    },
    {
      "entropy": 5.57491250038147,
      "epoch": 1.3012643454580821,
      "grad_norm": 1.0,
      "learning_rate": 0.00048332814146423566,
      "loss": 5.0838,
      "mean_token_accuracy": 0.195457261800766,
      "num_tokens": 29019940.0,
      "step": 16725
    },
    {
      "entropy": 5.600268363952637,
      "epoch": 1.3016533748297996,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004833176699959206,
      "loss": 5.1074,
      "mean_token_accuracy": 0.1974085733294487,
      "num_tokens": 29027644.0,
      "step": 16730
    },
    {
      "entropy": 5.5469756603240965,
      "epoch": 1.3020424042015173,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048330719536671633,
      "loss": 5.0417,
      "mean_token_accuracy": 0.19703566730022432,
      "num_tokens": 29035166.0,
      "step": 16735
    },
    {
      "entropy": 5.487120914459228,
      "epoch": 1.3024314335732348,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000483296717576782,
      "loss": 4.9948,
      "mean_token_accuracy": 0.19976236224174498,
      "num_tokens": 29044539.0,
      "step": 16740
    },
    {
      "entropy": 5.545634412765503,
      "epoch": 1.3028204629449522,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004832862366262765,
      "loss": 5.0631,
      "mean_token_accuracy": 0.20691450834274291,
      "num_tokens": 29052849.0,
      "step": 16745
    },
    {
      "entropy": 5.49363145828247,
      "epoch": 1.30320949231667,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048327575251535886,
      "loss": 5.0613,
      "mean_token_accuracy": 0.1911639615893364,
      "num_tokens": 29061881.0,
      "step": 16750
    },
    {
      "entropy": 5.647519636154175,
      "epoch": 1.3035985216883874,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004832652652441883,
      "loss": 5.1399,
      "mean_token_accuracy": 0.18321709781885148,
      "num_tokens": 29070460.0,
      "step": 16755
    },
    {
      "entropy": 5.62577633857727,
      "epoch": 1.303987551060105,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048325477481292375,
      "loss": 5.141,
      "mean_token_accuracy": 0.18978486508131026,
      "num_tokens": 29079061.0,
      "step": 16760
    },
    {
      "entropy": 5.497308778762817,
      "epoch": 1.3043765804318226,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004832442812217244,
      "loss": 5.0328,
      "mean_token_accuracy": 0.19651973247528076,
      "num_tokens": 29087525.0,
      "step": 16765
    },
    {
      "entropy": 5.510684585571289,
      "epoch": 1.3047656098035403,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004832337844707496,
      "loss": 5.0084,
      "mean_token_accuracy": 0.19796415865421296,
      "num_tokens": 29096206.0,
      "step": 16770
    },
    {
      "entropy": 5.567849922180176,
      "epoch": 1.3051546391752578,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048322328456015855,
      "loss": 5.2014,
      "mean_token_accuracy": 0.19181587994098664,
      "num_tokens": 29105025.0,
      "step": 16775
    },
    {
      "entropy": 5.567648935317993,
      "epoch": 1.3055436685469752,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048321278149011053,
      "loss": 5.1706,
      "mean_token_accuracy": 0.1886436492204666,
      "num_tokens": 29113468.0,
      "step": 16780
    },
    {
      "entropy": 5.558434534072876,
      "epoch": 1.305932697918693,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048320227526076504,
      "loss": 5.0532,
      "mean_token_accuracy": 0.19704548120498658,
      "num_tokens": 29122904.0,
      "step": 16785
    },
    {
      "entropy": 5.527788782119751,
      "epoch": 1.3063217272904104,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004831917658722814,
      "loss": 5.0422,
      "mean_token_accuracy": 0.20200203657150267,
      "num_tokens": 29132408.0,
      "step": 16790
    },
    {
      "entropy": 5.573469877243042,
      "epoch": 1.3067107566621279,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048318125332481903,
      "loss": 5.1845,
      "mean_token_accuracy": 0.18612269759178163,
      "num_tokens": 29141840.0,
      "step": 16795
    },
    {
      "entropy": 5.646148729324341,
      "epoch": 1.3070997860338456,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048317073761853764,
      "loss": 5.1594,
      "mean_token_accuracy": 0.18959160149097443,
      "num_tokens": 29150759.0,
      "step": 16800
    },
    {
      "entropy": 5.527711725234985,
      "epoch": 1.307488815405563,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004831602187535965,
      "loss": 4.9732,
      "mean_token_accuracy": 0.20588632076978683,
      "num_tokens": 29159489.0,
      "step": 16805
    },
    {
      "entropy": 5.529496955871582,
      "epoch": 1.3078778447772808,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004831496967301554,
      "loss": 5.0007,
      "mean_token_accuracy": 0.19174186885356903,
      "num_tokens": 29168160.0,
      "step": 16810
    },
    {
      "entropy": 5.570239734649658,
      "epoch": 1.3082668741489982,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048313917154837386,
      "loss": 5.0874,
      "mean_token_accuracy": 0.19374872148036956,
      "num_tokens": 29176534.0,
      "step": 16815
    },
    {
      "entropy": 5.473195648193359,
      "epoch": 1.308655903520716,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004831286432084118,
      "loss": 4.9826,
      "mean_token_accuracy": 0.20440248399972916,
      "num_tokens": 29185461.0,
      "step": 16820
    },
    {
      "entropy": 5.561170339584351,
      "epoch": 1.3090449328924334,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004831181117104289,
      "loss": 5.1314,
      "mean_token_accuracy": 0.18485222309827803,
      "num_tokens": 29194218.0,
      "step": 16825
    },
    {
      "entropy": 5.508718109130859,
      "epoch": 1.3094339622641509,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048310757705458476,
      "loss": 5.0936,
      "mean_token_accuracy": 0.19062842130661012,
      "num_tokens": 29203720.0,
      "step": 16830
    },
    {
      "entropy": 5.58176212310791,
      "epoch": 1.3098229916358686,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048309703924103944,
      "loss": 5.1596,
      "mean_token_accuracy": 0.18810677230358125,
      "num_tokens": 29212588.0,
      "step": 16835
    },
    {
      "entropy": 5.600096845626831,
      "epoch": 1.310212021007586,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048308649826995283,
      "loss": 5.1585,
      "mean_token_accuracy": 0.1872943624854088,
      "num_tokens": 29220718.0,
      "step": 16840
    },
    {
      "entropy": 5.525977945327758,
      "epoch": 1.3106010503793035,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048307595414148475,
      "loss": 5.0115,
      "mean_token_accuracy": 0.1965139016509056,
      "num_tokens": 29229386.0,
      "step": 16845
    },
    {
      "entropy": 5.531512784957886,
      "epoch": 1.3109900797510212,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004830654068557952,
      "loss": 5.0839,
      "mean_token_accuracy": 0.1964070737361908,
      "num_tokens": 29238390.0,
      "step": 16850
    },
    {
      "entropy": 5.602889108657837,
      "epoch": 1.3113791091227387,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004830548564130444,
      "loss": 5.1713,
      "mean_token_accuracy": 0.19223704636096955,
      "num_tokens": 29246502.0,
      "step": 16855
    },
    {
      "entropy": 5.5605803489685055,
      "epoch": 1.3117681384944564,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048304430281339216,
      "loss": 5.1068,
      "mean_token_accuracy": 0.18467968553304673,
      "num_tokens": 29254954.0,
      "step": 16860
    },
    {
      "entropy": 5.547167348861694,
      "epoch": 1.3121571678661739,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004830337460569989,
      "loss": 5.0554,
      "mean_token_accuracy": 0.2003497913479805,
      "num_tokens": 29263294.0,
      "step": 16865
    },
    {
      "entropy": 5.5047790050506595,
      "epoch": 1.3125461972378916,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004830231861440246,
      "loss": 5.0428,
      "mean_token_accuracy": 0.19816186875104905,
      "num_tokens": 29271737.0,
      "step": 16870
    },
    {
      "entropy": 5.611327743530273,
      "epoch": 1.312935226609609,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004830126230746295,
      "loss": 5.1673,
      "mean_token_accuracy": 0.18788998275995256,
      "num_tokens": 29279473.0,
      "step": 16875
    },
    {
      "entropy": 5.556150293350219,
      "epoch": 1.3133242559813265,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048300205684897405,
      "loss": 5.0766,
      "mean_token_accuracy": 0.20586208254098892,
      "num_tokens": 29287650.0,
      "step": 16880
    },
    {
      "entropy": 5.5243556022644045,
      "epoch": 1.3137132853530442,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004829914874672184,
      "loss": 5.1138,
      "mean_token_accuracy": 0.19034238457679747,
      "num_tokens": 29296914.0,
      "step": 16885
    },
    {
      "entropy": 5.59660792350769,
      "epoch": 1.3141023147247617,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048298091492952297,
      "loss": 5.0955,
      "mean_token_accuracy": 0.1925625279545784,
      "num_tokens": 29305409.0,
      "step": 16890
    },
    {
      "entropy": 5.654264307022094,
      "epoch": 1.3144913440964792,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004829703392360482,
      "loss": 5.1276,
      "mean_token_accuracy": 0.1888740301132202,
      "num_tokens": 29314333.0,
      "step": 16895
    },
    {
      "entropy": 5.466478586196899,
      "epoch": 1.3148803734681969,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048295976038695455,
      "loss": 4.9639,
      "mean_token_accuracy": 0.19954014122486113,
      "num_tokens": 29323764.0,
      "step": 16900
    },
    {
      "entropy": 5.495169353485108,
      "epoch": 1.3152694028399143,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004829491783824025,
      "loss": 5.0429,
      "mean_token_accuracy": 0.19919885396957399,
      "num_tokens": 29331734.0,
      "step": 16905
    },
    {
      "entropy": 5.50154595375061,
      "epoch": 1.315658432211632,
      "grad_norm": 1.125,
      "learning_rate": 0.00048293859322255276,
      "loss": 4.9656,
      "mean_token_accuracy": 0.20241139680147172,
      "num_tokens": 29339791.0,
      "step": 16910
    },
    {
      "entropy": 5.4954526901245115,
      "epoch": 1.3160474615833495,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004829280049075657,
      "loss": 5.0882,
      "mean_token_accuracy": 0.18745041489601136,
      "num_tokens": 29348548.0,
      "step": 16915
    },
    {
      "entropy": 5.49891619682312,
      "epoch": 1.3164364909550672,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048291741343760216,
      "loss": 5.0177,
      "mean_token_accuracy": 0.19515870809555053,
      "num_tokens": 29357151.0,
      "step": 16920
    },
    {
      "entropy": 5.591091346740723,
      "epoch": 1.3168255203267847,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004829068188128229,
      "loss": 5.0362,
      "mean_token_accuracy": 0.19670390635728835,
      "num_tokens": 29365643.0,
      "step": 16925
    },
    {
      "entropy": 5.521182060241699,
      "epoch": 1.3172145496985022,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004828962210333885,
      "loss": 5.1053,
      "mean_token_accuracy": 0.19022891968488692,
      "num_tokens": 29374215.0,
      "step": 16930
    },
    {
      "entropy": 5.496845245361328,
      "epoch": 1.3176035790702199,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004828856200994598,
      "loss": 5.0185,
      "mean_token_accuracy": 0.19011422097682953,
      "num_tokens": 29382586.0,
      "step": 16935
    },
    {
      "entropy": 5.465261554718017,
      "epoch": 1.3179926084419373,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004828750160111978,
      "loss": 5.0729,
      "mean_token_accuracy": 0.18973015248775482,
      "num_tokens": 29391922.0,
      "step": 16940
    },
    {
      "entropy": 5.591983222961426,
      "epoch": 1.3183816378136548,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048286440876876325,
      "loss": 5.041,
      "mean_token_accuracy": 0.19132227152585984,
      "num_tokens": 29400408.0,
      "step": 16945
    },
    {
      "entropy": 5.520861387252808,
      "epoch": 1.3187706671853725,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048285379837231714,
      "loss": 4.98,
      "mean_token_accuracy": 0.2023572012782097,
      "num_tokens": 29408130.0,
      "step": 16950
    },
    {
      "entropy": 5.519441604614258,
      "epoch": 1.31915969655709,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004828431848220205,
      "loss": 5.0615,
      "mean_token_accuracy": 0.19052825570106507,
      "num_tokens": 29417446.0,
      "step": 16955
    },
    {
      "entropy": 5.588409852981568,
      "epoch": 1.3195487259288077,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004828325681180343,
      "loss": 5.1332,
      "mean_token_accuracy": 0.1937786802649498,
      "num_tokens": 29426023.0,
      "step": 16960
    },
    {
      "entropy": 5.578017950057983,
      "epoch": 1.3199377553005252,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004828219482605197,
      "loss": 5.0635,
      "mean_token_accuracy": 0.1946910873055458,
      "num_tokens": 29434737.0,
      "step": 16965
    },
    {
      "entropy": 5.529328060150147,
      "epoch": 1.3203267846722428,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048281132524963786,
      "loss": 5.0745,
      "mean_token_accuracy": 0.19062233716249466,
      "num_tokens": 29443926.0,
      "step": 16970
    },
    {
      "entropy": 5.5510965347290036,
      "epoch": 1.3207158140439603,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004828006990855499,
      "loss": 5.0653,
      "mean_token_accuracy": 0.1963367447257042,
      "num_tokens": 29452495.0,
      "step": 16975
    },
    {
      "entropy": 5.47893500328064,
      "epoch": 1.3211048434156778,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048279006976841704,
      "loss": 5.0178,
      "mean_token_accuracy": 0.1971873253583908,
      "num_tokens": 29460661.0,
      "step": 16980
    },
    {
      "entropy": 5.506717252731323,
      "epoch": 1.3214938727873955,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004827794372984007,
      "loss": 5.0128,
      "mean_token_accuracy": 0.20216816663742065,
      "num_tokens": 29469155.0,
      "step": 16985
    },
    {
      "entropy": 5.566567707061767,
      "epoch": 1.321882902159113,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000482768801675662,
      "loss": 5.1332,
      "mean_token_accuracy": 0.1964087590575218,
      "num_tokens": 29478220.0,
      "step": 16990
    },
    {
      "entropy": 5.526828861236572,
      "epoch": 1.3222719315308304,
      "grad_norm": 1.125,
      "learning_rate": 0.0004827581629003625,
      "loss": 5.0804,
      "mean_token_accuracy": 0.1918095290660858,
      "num_tokens": 29486336.0,
      "step": 16995
    },
    {
      "entropy": 5.5904313087463375,
      "epoch": 1.3226609609025481,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048274752097266356,
      "loss": 5.1875,
      "mean_token_accuracy": 0.18848473578691483,
      "num_tokens": 29495150.0,
      "step": 17000
    },
    {
      "entropy": 5.576676082611084,
      "epoch": 1.3230499902742658,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004827368758927266,
      "loss": 5.0684,
      "mean_token_accuracy": 0.19018929600715637,
      "num_tokens": 29503508.0,
      "step": 17005
    },
    {
      "entropy": 5.5547338962554935,
      "epoch": 1.3234390196459833,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004827262276607132,
      "loss": 5.0582,
      "mean_token_accuracy": 0.191013802587986,
      "num_tokens": 29513038.0,
      "step": 17010
    },
    {
      "entropy": 5.488562536239624,
      "epoch": 1.3238280490177008,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000482715576276785,
      "loss": 5.0143,
      "mean_token_accuracy": 0.19450548142194748,
      "num_tokens": 29521344.0,
      "step": 17015
    },
    {
      "entropy": 5.510985517501831,
      "epoch": 1.3242170783894185,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004827049217411035,
      "loss": 5.1419,
      "mean_token_accuracy": 0.18839769810438156,
      "num_tokens": 29530109.0,
      "step": 17020
    },
    {
      "entropy": 5.571167993545532,
      "epoch": 1.324606107761136,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048269426405383043,
      "loss": 5.047,
      "mean_token_accuracy": 0.1843200519680977,
      "num_tokens": 29539114.0,
      "step": 17025
    },
    {
      "entropy": 5.564735460281372,
      "epoch": 1.3249951371328534,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048268360321512745,
      "loss": 5.1009,
      "mean_token_accuracy": 0.19079017043113708,
      "num_tokens": 29548746.0,
      "step": 17030
    },
    {
      "entropy": 5.554321384429931,
      "epoch": 1.3253841665045711,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004826729392251564,
      "loss": 5.1527,
      "mean_token_accuracy": 0.18578368574380874,
      "num_tokens": 29557681.0,
      "step": 17035
    },
    {
      "entropy": 5.577318811416626,
      "epoch": 1.3257731958762886,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000482662272084079,
      "loss": 5.1231,
      "mean_token_accuracy": 0.1926843836903572,
      "num_tokens": 29566151.0,
      "step": 17040
    },
    {
      "entropy": 5.5151914119720455,
      "epoch": 1.326162225248006,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004826516017920571,
      "loss": 4.9243,
      "mean_token_accuracy": 0.21577063649892808,
      "num_tokens": 29574853.0,
      "step": 17045
    },
    {
      "entropy": 5.506992292404175,
      "epoch": 1.3265512546197238,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004826409283492528,
      "loss": 5.0471,
      "mean_token_accuracy": 0.19436998367309571,
      "num_tokens": 29583319.0,
      "step": 17050
    },
    {
      "entropy": 5.53065767288208,
      "epoch": 1.3269402839914415,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048263025175582783,
      "loss": 5.1326,
      "mean_token_accuracy": 0.1900242805480957,
      "num_tokens": 29591949.0,
      "step": 17055
    },
    {
      "entropy": 5.602450513839722,
      "epoch": 1.327329313363159,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004826195720119442,
      "loss": 5.1346,
      "mean_token_accuracy": 0.18691054582595826,
      "num_tokens": 29600293.0,
      "step": 17060
    },
    {
      "entropy": 5.612691068649292,
      "epoch": 1.3277183427348764,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004826088891177641,
      "loss": 5.1501,
      "mean_token_accuracy": 0.19679902493953705,
      "num_tokens": 29608602.0,
      "step": 17065
    },
    {
      "entropy": 5.543917989730835,
      "epoch": 1.3281073721065941,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048259820307344954,
      "loss": 5.1208,
      "mean_token_accuracy": 0.1930314287543297,
      "num_tokens": 29617464.0,
      "step": 17070
    },
    {
      "entropy": 5.558330917358399,
      "epoch": 1.3284964014783116,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048258751387916267,
      "loss": 5.0193,
      "mean_token_accuracy": 0.20459766685962677,
      "num_tokens": 29626297.0,
      "step": 17075
    },
    {
      "entropy": 5.518637990951538,
      "epoch": 1.328885430850029,
      "grad_norm": 1.125,
      "learning_rate": 0.00048257682153506564,
      "loss": 5.0471,
      "mean_token_accuracy": 0.19119945019483567,
      "num_tokens": 29635085.0,
      "step": 17080
    },
    {
      "entropy": 5.506569814682007,
      "epoch": 1.3292744602217468,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004825661260413208,
      "loss": 5.0345,
      "mean_token_accuracy": 0.20060601234436035,
      "num_tokens": 29643528.0,
      "step": 17085
    },
    {
      "entropy": 5.510303115844726,
      "epoch": 1.3296634895934643,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048255542739809035,
      "loss": 5.0847,
      "mean_token_accuracy": 0.19258638322353364,
      "num_tokens": 29651883.0,
      "step": 17090
    },
    {
      "entropy": 5.497211122512818,
      "epoch": 1.330052518965182,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048254472560553665,
      "loss": 5.0832,
      "mean_token_accuracy": 0.18981143832206726,
      "num_tokens": 29661092.0,
      "step": 17095
    },
    {
      "entropy": 5.58232274055481,
      "epoch": 1.3304415483368994,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048253402066382207,
      "loss": 5.1077,
      "mean_token_accuracy": 0.1956985354423523,
      "num_tokens": 29670285.0,
      "step": 17100
    },
    {
      "entropy": 5.620194292068481,
      "epoch": 1.3308305777086171,
      "grad_norm": 1.125,
      "learning_rate": 0.0004825233125731091,
      "loss": 5.1525,
      "mean_token_accuracy": 0.19757503867149354,
      "num_tokens": 29679249.0,
      "step": 17105
    },
    {
      "entropy": 5.5793907165527346,
      "epoch": 1.3312196070803346,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048251260133356014,
      "loss": 5.1268,
      "mean_token_accuracy": 0.18529922962188722,
      "num_tokens": 29687978.0,
      "step": 17110
    },
    {
      "entropy": 5.560362768173218,
      "epoch": 1.331608636452052,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048250188694533774,
      "loss": 5.1432,
      "mean_token_accuracy": 0.18937211930751802,
      "num_tokens": 29695972.0,
      "step": 17115
    },
    {
      "entropy": 5.5562005043029785,
      "epoch": 1.3319976658237698,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004824911694086044,
      "loss": 4.9803,
      "mean_token_accuracy": 0.1954292818903923,
      "num_tokens": 29704997.0,
      "step": 17120
    },
    {
      "entropy": 5.649397993087769,
      "epoch": 1.3323866951954872,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000482480448723523,
      "loss": 5.2033,
      "mean_token_accuracy": 0.19126739501953124,
      "num_tokens": 29713957.0,
      "step": 17125
    },
    {
      "entropy": 5.579473781585693,
      "epoch": 1.3327757245672047,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048246972489025586,
      "loss": 5.0906,
      "mean_token_accuracy": 0.19456228911876677,
      "num_tokens": 29721989.0,
      "step": 17130
    },
    {
      "entropy": 5.5350266456604,
      "epoch": 1.3331647539389224,
      "grad_norm": 1.125,
      "learning_rate": 0.00048245899790896597,
      "loss": 5.0299,
      "mean_token_accuracy": 0.195353926718235,
      "num_tokens": 29730032.0,
      "step": 17135
    },
    {
      "entropy": 5.559976434707641,
      "epoch": 1.33355378331064,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004824482677798159,
      "loss": 5.0457,
      "mean_token_accuracy": 0.19699425250291824,
      "num_tokens": 29738545.0,
      "step": 17140
    },
    {
      "entropy": 5.568692541122436,
      "epoch": 1.3339428126823576,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048243753450296863,
      "loss": 5.057,
      "mean_token_accuracy": 0.1962989017367363,
      "num_tokens": 29747371.0,
      "step": 17145
    },
    {
      "entropy": 5.515608453750611,
      "epoch": 1.334331842054075,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048242679807858693,
      "loss": 5.0071,
      "mean_token_accuracy": 0.19719647765159606,
      "num_tokens": 29756565.0,
      "step": 17150
    },
    {
      "entropy": 5.518465566635132,
      "epoch": 1.3347208714257928,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048241605850683365,
      "loss": 5.0228,
      "mean_token_accuracy": 0.19539978355169296,
      "num_tokens": 29765846.0,
      "step": 17155
    },
    {
      "entropy": 5.549522781372071,
      "epoch": 1.3351099007975102,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048240531578787197,
      "loss": 5.0613,
      "mean_token_accuracy": 0.19682093411684037,
      "num_tokens": 29774222.0,
      "step": 17160
    },
    {
      "entropy": 5.498159790039063,
      "epoch": 1.3354989301692277,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004823945699218647,
      "loss": 4.9686,
      "mean_token_accuracy": 0.2013892263174057,
      "num_tokens": 29782366.0,
      "step": 17165
    },
    {
      "entropy": 5.566667461395264,
      "epoch": 1.3358879595409454,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004823838209089749,
      "loss": 5.1288,
      "mean_token_accuracy": 0.18865095376968383,
      "num_tokens": 29792147.0,
      "step": 17170
    },
    {
      "entropy": 5.582582759857178,
      "epoch": 1.3362769889126629,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048237306874936565,
      "loss": 5.0695,
      "mean_token_accuracy": 0.1969226270914078,
      "num_tokens": 29801325.0,
      "step": 17175
    },
    {
      "entropy": 5.522589588165284,
      "epoch": 1.3366660182843804,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004823623134432001,
      "loss": 5.0262,
      "mean_token_accuracy": 0.19460276812314986,
      "num_tokens": 29809353.0,
      "step": 17180
    },
    {
      "entropy": 5.535323190689087,
      "epoch": 1.337055047656098,
      "grad_norm": 1.125,
      "learning_rate": 0.00048235155499064166,
      "loss": 5.0868,
      "mean_token_accuracy": 0.1935117393732071,
      "num_tokens": 29817746.0,
      "step": 17185
    },
    {
      "entropy": 5.543033504486084,
      "epoch": 1.3374440770278155,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048234079339185335,
      "loss": 5.1061,
      "mean_token_accuracy": 0.1909686118364334,
      "num_tokens": 29826468.0,
      "step": 17190
    },
    {
      "entropy": 5.641568565368653,
      "epoch": 1.3378331063995332,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048233002864699856,
      "loss": 5.1618,
      "mean_token_accuracy": 0.18000449389219284,
      "num_tokens": 29835051.0,
      "step": 17195
    },
    {
      "entropy": 5.572253608703614,
      "epoch": 1.3382221357712507,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004823192607562405,
      "loss": 5.0571,
      "mean_token_accuracy": 0.19564492106437684,
      "num_tokens": 29843514.0,
      "step": 17200
    },
    {
      "entropy": 5.530295515060425,
      "epoch": 1.3386111651429684,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048230848971974265,
      "loss": 4.9906,
      "mean_token_accuracy": 0.1937766432762146,
      "num_tokens": 29851861.0,
      "step": 17205
    },
    {
      "entropy": 5.511221027374267,
      "epoch": 1.3390001945146859,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004822977155376684,
      "loss": 4.9895,
      "mean_token_accuracy": 0.19781352281570436,
      "num_tokens": 29860639.0,
      "step": 17210
    },
    {
      "entropy": 5.509042692184448,
      "epoch": 1.3393892238864034,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004822869382101814,
      "loss": 5.1931,
      "mean_token_accuracy": 0.18604647517204284,
      "num_tokens": 29869931.0,
      "step": 17215
    },
    {
      "entropy": 5.587409257888794,
      "epoch": 1.339778253258121,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004822761577374449,
      "loss": 5.0801,
      "mean_token_accuracy": 0.1908223435282707,
      "num_tokens": 29878933.0,
      "step": 17220
    },
    {
      "entropy": 5.57335352897644,
      "epoch": 1.3401672826298385,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004822653741196227,
      "loss": 5.0674,
      "mean_token_accuracy": 0.19485208243131638,
      "num_tokens": 29888031.0,
      "step": 17225
    },
    {
      "entropy": 5.525623369216919,
      "epoch": 1.340556312001556,
      "grad_norm": 1.0,
      "learning_rate": 0.0004822545873568783,
      "loss": 5.0359,
      "mean_token_accuracy": 0.19167845100164413,
      "num_tokens": 29896792.0,
      "step": 17230
    },
    {
      "entropy": 5.491623973846435,
      "epoch": 1.3409453413732737,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048224379744937545,
      "loss": 5.0309,
      "mean_token_accuracy": 0.1934276506304741,
      "num_tokens": 29905185.0,
      "step": 17235
    },
    {
      "entropy": 5.576599597930908,
      "epoch": 1.3413343707449912,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048223300439727783,
      "loss": 5.1213,
      "mean_token_accuracy": 0.1889025241136551,
      "num_tokens": 29913289.0,
      "step": 17240
    },
    {
      "entropy": 5.535838842391968,
      "epoch": 1.3417234001167089,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048222220820074926,
      "loss": 5.0871,
      "mean_token_accuracy": 0.19821808338165284,
      "num_tokens": 29922112.0,
      "step": 17245
    },
    {
      "entropy": 5.566669225692749,
      "epoch": 1.3421124294884264,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048221140885995346,
      "loss": 5.057,
      "mean_token_accuracy": 0.19136505126953124,
      "num_tokens": 29930944.0,
      "step": 17250
    },
    {
      "entropy": 5.6259430885314945,
      "epoch": 1.342501458860144,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004822006063750543,
      "loss": 5.1834,
      "mean_token_accuracy": 0.18399889022111893,
      "num_tokens": 29938576.0,
      "step": 17255
    },
    {
      "entropy": 5.559614276885986,
      "epoch": 1.3428904882318615,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048218980074621575,
      "loss": 5.0538,
      "mean_token_accuracy": 0.19086371660232543,
      "num_tokens": 29946771.0,
      "step": 17260
    },
    {
      "entropy": 5.610192489624024,
      "epoch": 1.343279517603579,
      "grad_norm": 1.125,
      "learning_rate": 0.00048217899197360183,
      "loss": 5.0721,
      "mean_token_accuracy": 0.19174696803092955,
      "num_tokens": 29955098.0,
      "step": 17265
    },
    {
      "entropy": 5.51542477607727,
      "epoch": 1.3436685469752967,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004821681800573764,
      "loss": 5.107,
      "mean_token_accuracy": 0.18705251812934875,
      "num_tokens": 29964538.0,
      "step": 17270
    },
    {
      "entropy": 5.532406234741211,
      "epoch": 1.3440575763470142,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004821573649977036,
      "loss": 5.0329,
      "mean_token_accuracy": 0.19478993713855744,
      "num_tokens": 29973965.0,
      "step": 17275
    },
    {
      "entropy": 5.470836877822876,
      "epoch": 1.3444466057187316,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048214654679474753,
      "loss": 4.9554,
      "mean_token_accuracy": 0.20025464445352553,
      "num_tokens": 29982742.0,
      "step": 17280
    },
    {
      "entropy": 5.465939521789551,
      "epoch": 1.3448356350904493,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048213572544867224,
      "loss": 5.0282,
      "mean_token_accuracy": 0.18938456922769548,
      "num_tokens": 29991846.0,
      "step": 17285
    },
    {
      "entropy": 5.590778398513794,
      "epoch": 1.3452246644621668,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048212490095964213,
      "loss": 4.9807,
      "mean_token_accuracy": 0.20843166708946229,
      "num_tokens": 29999784.0,
      "step": 17290
    },
    {
      "entropy": 5.571971607208252,
      "epoch": 1.3456136938338845,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004821140733278212,
      "loss": 5.1102,
      "mean_token_accuracy": 0.1912888690829277,
      "num_tokens": 30008139.0,
      "step": 17295
    },
    {
      "entropy": 5.599435520172119,
      "epoch": 1.346002723205602,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004821032425533739,
      "loss": 5.1071,
      "mean_token_accuracy": 0.19594431221485137,
      "num_tokens": 30016891.0,
      "step": 17300
    },
    {
      "entropy": 5.514035940170288,
      "epoch": 1.3463917525773197,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004820924086364646,
      "loss": 5.0793,
      "mean_token_accuracy": 0.19504499286413193,
      "num_tokens": 30025472.0,
      "step": 17305
    },
    {
      "entropy": 5.597099637985229,
      "epoch": 1.3467807819490372,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048208157157725756,
      "loss": 5.0747,
      "mean_token_accuracy": 0.1960415229201317,
      "num_tokens": 30034088.0,
      "step": 17310
    },
    {
      "entropy": 5.51554274559021,
      "epoch": 1.3471698113207546,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048207073137591726,
      "loss": 5.0282,
      "mean_token_accuracy": 0.19251953363418578,
      "num_tokens": 30042651.0,
      "step": 17315
    },
    {
      "entropy": 5.553288173675537,
      "epoch": 1.3475588406924723,
      "grad_norm": 1.125,
      "learning_rate": 0.00048205988803260824,
      "loss": 4.9781,
      "mean_token_accuracy": 0.2018101617693901,
      "num_tokens": 30050898.0,
      "step": 17320
    },
    {
      "entropy": 5.62773494720459,
      "epoch": 1.3479478700641898,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048204904154749496,
      "loss": 5.1512,
      "mean_token_accuracy": 0.19082681983709335,
      "num_tokens": 30059357.0,
      "step": 17325
    },
    {
      "entropy": 5.569904565811157,
      "epoch": 1.3483368994359073,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048203819192074206,
      "loss": 5.0711,
      "mean_token_accuracy": 0.19212352484464645,
      "num_tokens": 30068137.0,
      "step": 17330
    },
    {
      "entropy": 5.57463321685791,
      "epoch": 1.348725928807625,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048202733915251407,
      "loss": 5.1139,
      "mean_token_accuracy": 0.19380989968776702,
      "num_tokens": 30077139.0,
      "step": 17335
    },
    {
      "entropy": 5.570014429092407,
      "epoch": 1.3491149581793425,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048201648324297573,
      "loss": 5.1136,
      "mean_token_accuracy": 0.1996541753411293,
      "num_tokens": 30085267.0,
      "step": 17340
    },
    {
      "entropy": 5.530363416671753,
      "epoch": 1.3495039875510602,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048200562419229185,
      "loss": 5.0223,
      "mean_token_accuracy": 0.20284795314073562,
      "num_tokens": 30093804.0,
      "step": 17345
    },
    {
      "entropy": 5.56676607131958,
      "epoch": 1.3498930169227776,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000481994762000627,
      "loss": 5.0272,
      "mean_token_accuracy": 0.20327912718057634,
      "num_tokens": 30102653.0,
      "step": 17350
    },
    {
      "entropy": 5.449421405792236,
      "epoch": 1.3502820462944953,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048198389666814616,
      "loss": 4.9841,
      "mean_token_accuracy": 0.1984077513217926,
      "num_tokens": 30111978.0,
      "step": 17355
    },
    {
      "entropy": 5.577654695510864,
      "epoch": 1.3506710756662128,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048197302819501416,
      "loss": 5.2313,
      "mean_token_accuracy": 0.18188676461577416,
      "num_tokens": 30120387.0,
      "step": 17360
    },
    {
      "entropy": 5.559333848953247,
      "epoch": 1.3510601050379303,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048196215658139586,
      "loss": 5.0775,
      "mean_token_accuracy": 0.19406591951847077,
      "num_tokens": 30129069.0,
      "step": 17365
    },
    {
      "entropy": 5.549539375305176,
      "epoch": 1.351449134409648,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004819512818274562,
      "loss": 5.0444,
      "mean_token_accuracy": 0.19188395589590074,
      "num_tokens": 30138077.0,
      "step": 17370
    },
    {
      "entropy": 5.52253155708313,
      "epoch": 1.3518381637813655,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004819404039333603,
      "loss": 4.9883,
      "mean_token_accuracy": 0.1963801547884941,
      "num_tokens": 30146189.0,
      "step": 17375
    },
    {
      "entropy": 5.539373397827148,
      "epoch": 1.352227193153083,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004819295228992731,
      "loss": 5.1082,
      "mean_token_accuracy": 0.1909661039710045,
      "num_tokens": 30155664.0,
      "step": 17380
    },
    {
      "entropy": 5.5493323802948,
      "epoch": 1.3526162225248006,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048191863872535984,
      "loss": 5.0746,
      "mean_token_accuracy": 0.19623038321733474,
      "num_tokens": 30164701.0,
      "step": 17385
    },
    {
      "entropy": 5.572610998153687,
      "epoch": 1.3530052518965183,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004819077514117855,
      "loss": 5.105,
      "mean_token_accuracy": 0.18941740691661835,
      "num_tokens": 30173916.0,
      "step": 17390
    },
    {
      "entropy": 5.547691869735718,
      "epoch": 1.3533942812682358,
      "grad_norm": 1.125,
      "learning_rate": 0.00048189686095871545,
      "loss": 5.051,
      "mean_token_accuracy": 0.1960417613387108,
      "num_tokens": 30182803.0,
      "step": 17395
    },
    {
      "entropy": 5.525691556930542,
      "epoch": 1.3537833106399533,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048188596736631475,
      "loss": 5.0649,
      "mean_token_accuracy": 0.19962005466222763,
      "num_tokens": 30191589.0,
      "step": 17400
    },
    {
      "entropy": 5.5625330924987795,
      "epoch": 1.354172340011671,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048187507063474885,
      "loss": 5.1403,
      "mean_token_accuracy": 0.19279083013534545,
      "num_tokens": 30200170.0,
      "step": 17405
    },
    {
      "entropy": 5.568058872222901,
      "epoch": 1.3545613693833884,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048186417076418294,
      "loss": 5.0301,
      "mean_token_accuracy": 0.194656640291214,
      "num_tokens": 30208683.0,
      "step": 17410
    },
    {
      "entropy": 5.517553329467773,
      "epoch": 1.354950398755106,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048185326775478256,
      "loss": 5.0831,
      "mean_token_accuracy": 0.1854734793305397,
      "num_tokens": 30217141.0,
      "step": 17415
    },
    {
      "entropy": 5.534998178482056,
      "epoch": 1.3553394281268236,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048184236160671306,
      "loss": 5.0686,
      "mean_token_accuracy": 0.1888451471924782,
      "num_tokens": 30226759.0,
      "step": 17420
    },
    {
      "entropy": 5.585585737228394,
      "epoch": 1.355728457498541,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048183145232013995,
      "loss": 5.141,
      "mean_token_accuracy": 0.1911247953772545,
      "num_tokens": 30235945.0,
      "step": 17425
    },
    {
      "entropy": 5.568153810501099,
      "epoch": 1.3561174868702586,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048182053989522883,
      "loss": 5.1064,
      "mean_token_accuracy": 0.18803365528583527,
      "num_tokens": 30244912.0,
      "step": 17430
    },
    {
      "entropy": 5.543595457077027,
      "epoch": 1.3565065162419763,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048180962433214515,
      "loss": 5.049,
      "mean_token_accuracy": 0.19550030678510666,
      "num_tokens": 30253807.0,
      "step": 17435
    },
    {
      "entropy": 5.544771575927735,
      "epoch": 1.356895545613694,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048179870563105453,
      "loss": 5.1601,
      "mean_token_accuracy": 0.19282080680131913,
      "num_tokens": 30262380.0,
      "step": 17440
    },
    {
      "entropy": 5.623108959197998,
      "epoch": 1.3572845749854114,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004817877837921228,
      "loss": 5.082,
      "mean_token_accuracy": 0.1914823532104492,
      "num_tokens": 30271374.0,
      "step": 17445
    },
    {
      "entropy": 5.574206590652466,
      "epoch": 1.357673604357129,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004817768588155155,
      "loss": 5.1129,
      "mean_token_accuracy": 0.19587183743715286,
      "num_tokens": 30280037.0,
      "step": 17450
    },
    {
      "entropy": 5.483479738235474,
      "epoch": 1.3580626337288466,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048176593070139865,
      "loss": 4.908,
      "mean_token_accuracy": 0.20995635837316512,
      "num_tokens": 30288335.0,
      "step": 17455
    },
    {
      "entropy": 5.492858219146728,
      "epoch": 1.358451663100564,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048175499944993767,
      "loss": 5.0306,
      "mean_token_accuracy": 0.19311318993568422,
      "num_tokens": 30297109.0,
      "step": 17460
    },
    {
      "entropy": 5.5455790042877195,
      "epoch": 1.3588406924722816,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048174406506129884,
      "loss": 5.0362,
      "mean_token_accuracy": 0.19513872861862183,
      "num_tokens": 30305594.0,
      "step": 17465
    },
    {
      "entropy": 5.500927686691284,
      "epoch": 1.3592297218439993,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048173312753564787,
      "loss": 4.958,
      "mean_token_accuracy": 0.19964597076177598,
      "num_tokens": 30313106.0,
      "step": 17470
    },
    {
      "entropy": 5.533057117462159,
      "epoch": 1.3596187512157167,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004817221868731506,
      "loss": 5.1603,
      "mean_token_accuracy": 0.1874983176589012,
      "num_tokens": 30322194.0,
      "step": 17475
    },
    {
      "entropy": 5.491589117050171,
      "epoch": 1.3600077805874344,
      "grad_norm": 1.0,
      "learning_rate": 0.00048171124307397334,
      "loss": 4.9949,
      "mean_token_accuracy": 0.20160888135433197,
      "num_tokens": 30330247.0,
      "step": 17480
    },
    {
      "entropy": 5.56473937034607,
      "epoch": 1.360396809959152,
      "grad_norm": 1.125,
      "learning_rate": 0.0004817002961382819,
      "loss": 5.0886,
      "mean_token_accuracy": 0.1946029081940651,
      "num_tokens": 30338738.0,
      "step": 17485
    },
    {
      "entropy": 5.522298049926758,
      "epoch": 1.3607858393308696,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048168934606624255,
      "loss": 5.0752,
      "mean_token_accuracy": 0.18871772289276123,
      "num_tokens": 30347729.0,
      "step": 17490
    },
    {
      "entropy": 5.577666854858398,
      "epoch": 1.361174868702587,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048167839285802116,
      "loss": 5.0815,
      "mean_token_accuracy": 0.19029493480920792,
      "num_tokens": 30356408.0,
      "step": 17495
    },
    {
      "entropy": 5.543626499176026,
      "epoch": 1.3615638980743046,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004816674365137843,
      "loss": 5.0447,
      "mean_token_accuracy": 0.19172577261924745,
      "num_tokens": 30364910.0,
      "step": 17500
    },
    {
      "entropy": 5.51476263999939,
      "epoch": 1.3619529274460223,
      "grad_norm": 1.140625,
      "learning_rate": 0.000481656477033698,
      "loss": 5.0657,
      "mean_token_accuracy": 0.1929220139980316,
      "num_tokens": 30373882.0,
      "step": 17505
    },
    {
      "entropy": 5.601848077774048,
      "epoch": 1.3623419568177397,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004816455144179286,
      "loss": 5.1237,
      "mean_token_accuracy": 0.18255161494016647,
      "num_tokens": 30382884.0,
      "step": 17510
    },
    {
      "entropy": 5.5524969577789305,
      "epoch": 1.3627309861894572,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004816345486666424,
      "loss": 5.0176,
      "mean_token_accuracy": 0.1949968546628952,
      "num_tokens": 30391798.0,
      "step": 17515
    },
    {
      "entropy": 5.548796701431274,
      "epoch": 1.363120015561175,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004816235797800058,
      "loss": 5.0379,
      "mean_token_accuracy": 0.19791061878204347,
      "num_tokens": 30400194.0,
      "step": 17520
    },
    {
      "entropy": 5.522584009170532,
      "epoch": 1.3635090449328924,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004816126077581852,
      "loss": 4.9724,
      "mean_token_accuracy": 0.19731647968292237,
      "num_tokens": 30408653.0,
      "step": 17525
    },
    {
      "entropy": 5.536618375778199,
      "epoch": 1.36389807430461,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048160163260134724,
      "loss": 5.1015,
      "mean_token_accuracy": 0.19110390692949294,
      "num_tokens": 30417467.0,
      "step": 17530
    },
    {
      "entropy": 5.549066066741943,
      "epoch": 1.3642871036763276,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004815906543096583,
      "loss": 5.0827,
      "mean_token_accuracy": 0.19031040668487548,
      "num_tokens": 30425993.0,
      "step": 17535
    },
    {
      "entropy": 5.54912371635437,
      "epoch": 1.3646761330480452,
      "grad_norm": 1.046875,
      "learning_rate": 0.000481579672883285,
      "loss": 5.062,
      "mean_token_accuracy": 0.19586015492677689,
      "num_tokens": 30434809.0,
      "step": 17540
    },
    {
      "entropy": 5.558625364303589,
      "epoch": 1.3650651624197627,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048156868832239396,
      "loss": 5.084,
      "mean_token_accuracy": 0.19773411005735397,
      "num_tokens": 30443536.0,
      "step": 17545
    },
    {
      "entropy": 5.507848787307739,
      "epoch": 1.3654541917914802,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004815577006271519,
      "loss": 5.0751,
      "mean_token_accuracy": 0.19477519243955613,
      "num_tokens": 30452163.0,
      "step": 17550
    },
    {
      "entropy": 5.510320949554443,
      "epoch": 1.365843221163198,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048154670979772555,
      "loss": 5.0353,
      "mean_token_accuracy": 0.19959058463573456,
      "num_tokens": 30461234.0,
      "step": 17555
    },
    {
      "entropy": 5.662606191635132,
      "epoch": 1.3662322505349154,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004815357158342815,
      "loss": 5.203,
      "mean_token_accuracy": 0.17945901304483414,
      "num_tokens": 30470438.0,
      "step": 17560
    },
    {
      "entropy": 5.585896778106689,
      "epoch": 1.3666212799066328,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004815247187369868,
      "loss": 4.9935,
      "mean_token_accuracy": 0.19709022492170333,
      "num_tokens": 30479454.0,
      "step": 17565
    },
    {
      "entropy": 5.58470892906189,
      "epoch": 1.3670103092783505,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048151371850600814,
      "loss": 5.0972,
      "mean_token_accuracy": 0.19227968007326127,
      "num_tokens": 30487968.0,
      "step": 17570
    },
    {
      "entropy": 5.54062294960022,
      "epoch": 1.367399338650068,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048150271514151255,
      "loss": 5.0285,
      "mean_token_accuracy": 0.19773666709661483,
      "num_tokens": 30495712.0,
      "step": 17575
    },
    {
      "entropy": 5.56768045425415,
      "epoch": 1.3677883680217857,
      "grad_norm": 1.0,
      "learning_rate": 0.00048149170864366693,
      "loss": 5.1316,
      "mean_token_accuracy": 0.18999616354703902,
      "num_tokens": 30505008.0,
      "step": 17580
    },
    {
      "entropy": 5.563160037994384,
      "epoch": 1.3681773973935032,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048148069901263836,
      "loss": 5.0286,
      "mean_token_accuracy": 0.19441289603710174,
      "num_tokens": 30514496.0,
      "step": 17585
    },
    {
      "entropy": 5.5507340908050535,
      "epoch": 1.368566426765221,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048146968624859375,
      "loss": 5.0202,
      "mean_token_accuracy": 0.19551381468772888,
      "num_tokens": 30522942.0,
      "step": 17590
    },
    {
      "entropy": 5.598671102523804,
      "epoch": 1.3689554561369384,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004814586703517003,
      "loss": 5.1809,
      "mean_token_accuracy": 0.19080182313919067,
      "num_tokens": 30532124.0,
      "step": 17595
    },
    {
      "entropy": 5.59537935256958,
      "epoch": 1.3693444855086558,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004814476513221251,
      "loss": 5.282,
      "mean_token_accuracy": 0.17841803580522536,
      "num_tokens": 30541063.0,
      "step": 17600
    },
    {
      "entropy": 5.609765625,
      "epoch": 1.3697335148803735,
      "grad_norm": 1.125,
      "learning_rate": 0.00048143662916003545,
      "loss": 5.0051,
      "mean_token_accuracy": 0.2002350300550461,
      "num_tokens": 30549062.0,
      "step": 17605
    },
    {
      "entropy": 5.628254175186157,
      "epoch": 1.370122544252091,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004814256038655985,
      "loss": 5.1096,
      "mean_token_accuracy": 0.18862653523683548,
      "num_tokens": 30557211.0,
      "step": 17610
    },
    {
      "entropy": 5.570678949356079,
      "epoch": 1.3705115736238085,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004814145754389816,
      "loss": 5.1427,
      "mean_token_accuracy": 0.18625256419181824,
      "num_tokens": 30566603.0,
      "step": 17615
    },
    {
      "entropy": 5.505239725112915,
      "epoch": 1.3709006029955262,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048140354388035204,
      "loss": 5.0211,
      "mean_token_accuracy": 0.19429576992988587,
      "num_tokens": 30574903.0,
      "step": 17620
    },
    {
      "entropy": 5.591094017028809,
      "epoch": 1.3712896323672437,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004813925091898772,
      "loss": 5.1186,
      "mean_token_accuracy": 0.18633642941713333,
      "num_tokens": 30584925.0,
      "step": 17625
    },
    {
      "entropy": 5.637942218780518,
      "epoch": 1.3716786617389614,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004813814713677246,
      "loss": 5.1758,
      "mean_token_accuracy": 0.18917116671800613,
      "num_tokens": 30593376.0,
      "step": 17630
    },
    {
      "entropy": 5.576561069488525,
      "epoch": 1.3720676911106788,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004813704304140616,
      "loss": 5.0956,
      "mean_token_accuracy": 0.19653935581445695,
      "num_tokens": 30602306.0,
      "step": 17635
    },
    {
      "entropy": 5.528047800064087,
      "epoch": 1.3724567204823965,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004813593863290559,
      "loss": 5.0404,
      "mean_token_accuracy": 0.19463779032230377,
      "num_tokens": 30611340.0,
      "step": 17640
    },
    {
      "entropy": 5.584700870513916,
      "epoch": 1.372845749854114,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004813483391128748,
      "loss": 5.1336,
      "mean_token_accuracy": 0.19378792196512223,
      "num_tokens": 30619507.0,
      "step": 17645
    },
    {
      "entropy": 5.578923034667969,
      "epoch": 1.3732347792258315,
      "grad_norm": 1.0,
      "learning_rate": 0.0004813372887656862,
      "loss": 5.1371,
      "mean_token_accuracy": 0.19276241660118104,
      "num_tokens": 30629897.0,
      "step": 17650
    },
    {
      "entropy": 5.581551551818848,
      "epoch": 1.3736238085975492,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004813262352876576,
      "loss": 5.0771,
      "mean_token_accuracy": 0.19110633283853531,
      "num_tokens": 30638517.0,
      "step": 17655
    },
    {
      "entropy": 5.603126049041748,
      "epoch": 1.3740128379692667,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048131517867895684,
      "loss": 5.1106,
      "mean_token_accuracy": 0.19446144104003907,
      "num_tokens": 30646816.0,
      "step": 17660
    },
    {
      "entropy": 5.583583879470825,
      "epoch": 1.3744018673409841,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048130411893975155,
      "loss": 5.0792,
      "mean_token_accuracy": 0.19811022877693177,
      "num_tokens": 30655199.0,
      "step": 17665
    },
    {
      "entropy": 5.515999984741211,
      "epoch": 1.3747908967127018,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004812930560702097,
      "loss": 4.9601,
      "mean_token_accuracy": 0.20013821721076966,
      "num_tokens": 30663562.0,
      "step": 17670
    },
    {
      "entropy": 5.581359577178955,
      "epoch": 1.3751799260844193,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048128199007049907,
      "loss": 5.1247,
      "mean_token_accuracy": 0.189364193379879,
      "num_tokens": 30672611.0,
      "step": 17675
    },
    {
      "entropy": 5.556504487991333,
      "epoch": 1.375568955456137,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048127092094078756,
      "loss": 5.0464,
      "mean_token_accuracy": 0.19969519078731537,
      "num_tokens": 30680766.0,
      "step": 17680
    },
    {
      "entropy": 5.482817554473877,
      "epoch": 1.3759579848278545,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048125984868124317,
      "loss": 4.9236,
      "mean_token_accuracy": 0.20279396325349808,
      "num_tokens": 30689192.0,
      "step": 17685
    },
    {
      "entropy": 5.506543684005737,
      "epoch": 1.3763470141995722,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004812487732920338,
      "loss": 5.0348,
      "mean_token_accuracy": 0.18978818953037263,
      "num_tokens": 30697889.0,
      "step": 17690
    },
    {
      "entropy": 5.612115716934204,
      "epoch": 1.3767360435712896,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004812376947733277,
      "loss": 5.0863,
      "mean_token_accuracy": 0.19444066882133484,
      "num_tokens": 30706285.0,
      "step": 17695
    },
    {
      "entropy": 5.549118089675903,
      "epoch": 1.3771250729430071,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048122661312529263,
      "loss": 5.053,
      "mean_token_accuracy": 0.1928205043077469,
      "num_tokens": 30715159.0,
      "step": 17700
    },
    {
      "entropy": 5.495332956314087,
      "epoch": 1.3775141023147248,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004812155283480971,
      "loss": 4.9455,
      "mean_token_accuracy": 0.19942924678325652,
      "num_tokens": 30723133.0,
      "step": 17705
    },
    {
      "entropy": 5.616144466400146,
      "epoch": 1.3779031316864423,
      "grad_norm": 1.125,
      "learning_rate": 0.0004812044404419091,
      "loss": 5.1449,
      "mean_token_accuracy": 0.18886625319719313,
      "num_tokens": 30732272.0,
      "step": 17710
    },
    {
      "entropy": 5.563917350769043,
      "epoch": 1.3782921610581598,
      "grad_norm": 1.0625,
      "learning_rate": 0.000481193349406897,
      "loss": 5.0126,
      "mean_token_accuracy": 0.1946540132164955,
      "num_tokens": 30740439.0,
      "step": 17715
    },
    {
      "entropy": 5.5266951560974125,
      "epoch": 1.3786811904298775,
      "grad_norm": 1.0625,
      "learning_rate": 0.000481182255243229,
      "loss": 5.0885,
      "mean_token_accuracy": 0.1949324294924736,
      "num_tokens": 30749722.0,
      "step": 17720
    },
    {
      "entropy": 5.652087831497193,
      "epoch": 1.379070219801595,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048117115795107335,
      "loss": 5.2578,
      "mean_token_accuracy": 0.18782196342945098,
      "num_tokens": 30759261.0,
      "step": 17725
    },
    {
      "entropy": 5.639539957046509,
      "epoch": 1.3794592491733126,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004811600575305987,
      "loss": 5.0971,
      "mean_token_accuracy": 0.19457124322652816,
      "num_tokens": 30769133.0,
      "step": 17730
    },
    {
      "entropy": 5.56157488822937,
      "epoch": 1.3798482785450301,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004811489539819731,
      "loss": 5.0308,
      "mean_token_accuracy": 0.19893912822008133,
      "num_tokens": 30777879.0,
      "step": 17735
    },
    {
      "entropy": 5.5561573028564455,
      "epoch": 1.3802373079167478,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048113784730536544,
      "loss": 5.1481,
      "mean_token_accuracy": 0.19803484082221984,
      "num_tokens": 30786307.0,
      "step": 17740
    },
    {
      "entropy": 5.5917370319366455,
      "epoch": 1.3806263372884653,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048112673750094396,
      "loss": 5.1782,
      "mean_token_accuracy": 0.18679643869400026,
      "num_tokens": 30796153.0,
      "step": 17745
    },
    {
      "entropy": 5.5418017387390135,
      "epoch": 1.3810153666601828,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004811156245688773,
      "loss": 4.9347,
      "mean_token_accuracy": 0.1995364710688591,
      "num_tokens": 30805441.0,
      "step": 17750
    },
    {
      "entropy": 5.5273401737213135,
      "epoch": 1.3814043960319005,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004811045085093342,
      "loss": 5.1078,
      "mean_token_accuracy": 0.18242234140634536,
      "num_tokens": 30814898.0,
      "step": 17755
    },
    {
      "entropy": 5.644738817214966,
      "epoch": 1.381793425403618,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004810933893224831,
      "loss": 5.2709,
      "mean_token_accuracy": 0.18093016296625136,
      "num_tokens": 30823995.0,
      "step": 17760
    },
    {
      "entropy": 5.581730508804322,
      "epoch": 1.3821824547753354,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048108226700849287,
      "loss": 5.1041,
      "mean_token_accuracy": 0.19191443175077438,
      "num_tokens": 30832963.0,
      "step": 17765
    },
    {
      "entropy": 5.5758106231689455,
      "epoch": 1.382571484147053,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004810711415675323,
      "loss": 5.1074,
      "mean_token_accuracy": 0.18946473002433778,
      "num_tokens": 30841559.0,
      "step": 17770
    },
    {
      "entropy": 5.606370544433593,
      "epoch": 1.3829605135187706,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004810600129997702,
      "loss": 5.1495,
      "mean_token_accuracy": 0.19180584996938704,
      "num_tokens": 30850481.0,
      "step": 17775
    },
    {
      "entropy": 5.551694059371949,
      "epoch": 1.3833495428904883,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048104888130537534,
      "loss": 4.996,
      "mean_token_accuracy": 0.20071802288293839,
      "num_tokens": 30859632.0,
      "step": 17780
    },
    {
      "entropy": 5.592933893203735,
      "epoch": 1.3837385722622058,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048103774648451664,
      "loss": 5.0591,
      "mean_token_accuracy": 0.1899753227829933,
      "num_tokens": 30867909.0,
      "step": 17785
    },
    {
      "entropy": 5.576524496078491,
      "epoch": 1.3841276016339235,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048102660853736314,
      "loss": 5.0822,
      "mean_token_accuracy": 0.19792128950357438,
      "num_tokens": 30876339.0,
      "step": 17790
    },
    {
      "entropy": 5.543550777435303,
      "epoch": 1.384516631005641,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004810154674640837,
      "loss": 5.0636,
      "mean_token_accuracy": 0.19140105098485946,
      "num_tokens": 30884998.0,
      "step": 17795
    },
    {
      "entropy": 5.5799602508544925,
      "epoch": 1.3849056603773584,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048100432326484744,
      "loss": 5.0935,
      "mean_token_accuracy": 0.18713977932929993,
      "num_tokens": 30893554.0,
      "step": 17800
    },
    {
      "entropy": 5.529959487915039,
      "epoch": 1.385294689749076,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004809931759398235,
      "loss": 5.11,
      "mean_token_accuracy": 0.1893826976418495,
      "num_tokens": 30901954.0,
      "step": 17805
    },
    {
      "entropy": 5.556656312942505,
      "epoch": 1.3856837191207936,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048098202548918094,
      "loss": 5.0485,
      "mean_token_accuracy": 0.1950295940041542,
      "num_tokens": 30911145.0,
      "step": 17810
    },
    {
      "entropy": 5.519692087173462,
      "epoch": 1.386072748492511,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048097087191308914,
      "loss": 4.9505,
      "mean_token_accuracy": 0.1988508179783821,
      "num_tokens": 30918924.0,
      "step": 17815
    },
    {
      "entropy": 5.542035341262817,
      "epoch": 1.3864617778642288,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004809597152117171,
      "loss": 5.1208,
      "mean_token_accuracy": 0.19037587195634842,
      "num_tokens": 30927620.0,
      "step": 17820
    },
    {
      "entropy": 5.535134983062744,
      "epoch": 1.3868508072359464,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004809485553852342,
      "loss": 4.9999,
      "mean_token_accuracy": 0.19581740647554396,
      "num_tokens": 30936140.0,
      "step": 17825
    },
    {
      "entropy": 5.629930448532105,
      "epoch": 1.387239836607664,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004809373924338098,
      "loss": 5.2073,
      "mean_token_accuracy": 0.18628117591142654,
      "num_tokens": 30944450.0,
      "step": 17830
    },
    {
      "entropy": 5.529106616973877,
      "epoch": 1.3876288659793814,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048092622635761316,
      "loss": 5.0344,
      "mean_token_accuracy": 0.19644004702568055,
      "num_tokens": 30953536.0,
      "step": 17835
    },
    {
      "entropy": 5.57500810623169,
      "epoch": 1.388017895351099,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048091505715681395,
      "loss": 5.0475,
      "mean_token_accuracy": 0.19390478432178498,
      "num_tokens": 30962556.0,
      "step": 17840
    },
    {
      "entropy": 5.5235267162323,
      "epoch": 1.3884069247228166,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004809038848315814,
      "loss": 5.0009,
      "mean_token_accuracy": 0.20344025790691375,
      "num_tokens": 30970476.0,
      "step": 17845
    },
    {
      "entropy": 5.609632539749145,
      "epoch": 1.388795954094534,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004808927093820851,
      "loss": 5.037,
      "mean_token_accuracy": 0.18904352337121963,
      "num_tokens": 30978945.0,
      "step": 17850
    },
    {
      "entropy": 5.502901077270508,
      "epoch": 1.3891849834662517,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004808815308084947,
      "loss": 4.9967,
      "mean_token_accuracy": 0.1952747255563736,
      "num_tokens": 30988135.0,
      "step": 17855
    },
    {
      "entropy": 5.598600101470947,
      "epoch": 1.3895740128379692,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004808703491109798,
      "loss": 5.0872,
      "mean_token_accuracy": 0.20162756741046906,
      "num_tokens": 30996408.0,
      "step": 17860
    },
    {
      "entropy": 5.536910724639893,
      "epoch": 1.3899630422096867,
      "grad_norm": 1.125,
      "learning_rate": 0.0004808591642897099,
      "loss": 5.1247,
      "mean_token_accuracy": 0.18928430378437042,
      "num_tokens": 31004156.0,
      "step": 17865
    },
    {
      "entropy": 5.522326564788818,
      "epoch": 1.3903520715814044,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000480847976344855,
      "loss": 5.0477,
      "mean_token_accuracy": 0.19676471054553984,
      "num_tokens": 31013518.0,
      "step": 17870
    },
    {
      "entropy": 5.509045934677124,
      "epoch": 1.390741100953122,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048083678527658465,
      "loss": 4.937,
      "mean_token_accuracy": 0.20250283926725388,
      "num_tokens": 31022116.0,
      "step": 17875
    },
    {
      "entropy": 5.553902673721313,
      "epoch": 1.3911301303248396,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004808255910850687,
      "loss": 5.101,
      "mean_token_accuracy": 0.1916371166706085,
      "num_tokens": 31030320.0,
      "step": 17880
    },
    {
      "entropy": 5.536445140838623,
      "epoch": 1.391519159696557,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004808143937704769,
      "loss": 5.0736,
      "mean_token_accuracy": 0.20021251142024993,
      "num_tokens": 31038410.0,
      "step": 17885
    },
    {
      "entropy": 5.509433555603027,
      "epoch": 1.3919081890682747,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048080319333297937,
      "loss": 4.9912,
      "mean_token_accuracy": 0.19662870615720748,
      "num_tokens": 31047719.0,
      "step": 17890
    },
    {
      "entropy": 5.545101118087769,
      "epoch": 1.3922972184399922,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048079198977274597,
      "loss": 5.1237,
      "mean_token_accuracy": 0.19389263689517974,
      "num_tokens": 31057536.0,
      "step": 17895
    },
    {
      "entropy": 5.60426893234253,
      "epoch": 1.3926862478117097,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048078078308994666,
      "loss": 5.0781,
      "mean_token_accuracy": 0.19347328543663025,
      "num_tokens": 31066975.0,
      "step": 17900
    },
    {
      "entropy": 5.591017389297486,
      "epoch": 1.3930752771834274,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004807695732847515,
      "loss": 5.0578,
      "mean_token_accuracy": 0.20267266631126404,
      "num_tokens": 31075573.0,
      "step": 17905
    },
    {
      "entropy": 5.499698686599731,
      "epoch": 1.3934643065551449,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048075836035733053,
      "loss": 4.9857,
      "mean_token_accuracy": 0.20540899634361268,
      "num_tokens": 31084486.0,
      "step": 17910
    },
    {
      "entropy": 5.524811601638794,
      "epoch": 1.3938533359268626,
      "grad_norm": 1.09375,
      "learning_rate": 0.000480747144307854,
      "loss": 5.032,
      "mean_token_accuracy": 0.19896805584430693,
      "num_tokens": 31093045.0,
      "step": 17915
    },
    {
      "entropy": 5.5754108905792235,
      "epoch": 1.39424236529858,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048073592513649203,
      "loss": 5.0651,
      "mean_token_accuracy": 0.19914070814847945,
      "num_tokens": 31102024.0,
      "step": 17920
    },
    {
      "entropy": 5.556309938430786,
      "epoch": 1.3946313946702977,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004807247028434148,
      "loss": 5.1563,
      "mean_token_accuracy": 0.1848788172006607,
      "num_tokens": 31110656.0,
      "step": 17925
    },
    {
      "entropy": 5.562730550765991,
      "epoch": 1.3950204240420152,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004807134774287927,
      "loss": 5.0239,
      "mean_token_accuracy": 0.19971562772989274,
      "num_tokens": 31119199.0,
      "step": 17930
    },
    {
      "entropy": 5.5469917297363285,
      "epoch": 1.3954094534137327,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048070224889279603,
      "loss": 5.0235,
      "mean_token_accuracy": 0.2022391676902771,
      "num_tokens": 31127301.0,
      "step": 17935
    },
    {
      "entropy": 5.529315710067749,
      "epoch": 1.3957984827854504,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048069101723559505,
      "loss": 5.1091,
      "mean_token_accuracy": 0.19404977560043335,
      "num_tokens": 31136714.0,
      "step": 17940
    },
    {
      "entropy": 5.569016361236573,
      "epoch": 1.3961875121571679,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004806797824573603,
      "loss": 5.0975,
      "mean_token_accuracy": 0.18793601542711258,
      "num_tokens": 31145663.0,
      "step": 17945
    },
    {
      "entropy": 5.495148038864135,
      "epoch": 1.3965765415288853,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004806685445582624,
      "loss": 4.8486,
      "mean_token_accuracy": 0.21121444255113603,
      "num_tokens": 31154335.0,
      "step": 17950
    },
    {
      "entropy": 5.512045097351074,
      "epoch": 1.396965570900603,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048065730353847143,
      "loss": 5.0672,
      "mean_token_accuracy": 0.19605384021997452,
      "num_tokens": 31163683.0,
      "step": 17955
    },
    {
      "entropy": 5.540881824493408,
      "epoch": 1.3973546002723205,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048064605939815837,
      "loss": 5.1438,
      "mean_token_accuracy": 0.1950014352798462,
      "num_tokens": 31172361.0,
      "step": 17960
    },
    {
      "entropy": 5.568410921096802,
      "epoch": 1.3977436296440382,
      "grad_norm": 1.125,
      "learning_rate": 0.0004806348121374936,
      "loss": 4.975,
      "mean_token_accuracy": 0.20192594081163406,
      "num_tokens": 31180572.0,
      "step": 17965
    },
    {
      "entropy": 5.517013025283814,
      "epoch": 1.3981326590157557,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004806235617566479,
      "loss": 4.9983,
      "mean_token_accuracy": 0.19979233294725418,
      "num_tokens": 31188751.0,
      "step": 17970
    },
    {
      "entropy": 5.604952573776245,
      "epoch": 1.3985216883874734,
      "grad_norm": 1.140625,
      "learning_rate": 0.000480612308255792,
      "loss": 5.1794,
      "mean_token_accuracy": 0.18921997547149658,
      "num_tokens": 31197762.0,
      "step": 17975
    },
    {
      "entropy": 5.587277269363403,
      "epoch": 1.3989107177591908,
      "grad_norm": 1.125,
      "learning_rate": 0.00048060105163509647,
      "loss": 5.111,
      "mean_token_accuracy": 0.1892549380660057,
      "num_tokens": 31207203.0,
      "step": 17980
    },
    {
      "entropy": 5.602642583847046,
      "epoch": 1.3992997471309083,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004805897918947323,
      "loss": 5.0866,
      "mean_token_accuracy": 0.19446944892406465,
      "num_tokens": 31215711.0,
      "step": 17985
    },
    {
      "entropy": 5.535875749588013,
      "epoch": 1.399688776502626,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004805785290348702,
      "loss": 5.0461,
      "mean_token_accuracy": 0.19532687664031984,
      "num_tokens": 31224564.0,
      "step": 17990
    },
    {
      "entropy": 5.5441070079803465,
      "epoch": 1.4000778058743435,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048056726305568124,
      "loss": 5.0143,
      "mean_token_accuracy": 0.1948551580309868,
      "num_tokens": 31233252.0,
      "step": 17995
    },
    {
      "entropy": 5.589895343780517,
      "epoch": 1.400466835246061,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048055599395733617,
      "loss": 5.1052,
      "mean_token_accuracy": 0.19427623003721237,
      "num_tokens": 31242539.0,
      "step": 18000
    },
    {
      "epoch": 1.400466835246061,
      "eval_entropy": 5.370003930368139,
      "eval_loss": 5.140840530395508,
      "eval_mean_token_accuracy": 0.20023161241851023,
      "eval_num_tokens": 31242539.0,
      "eval_runtime": 21.6224,
      "eval_samples_per_second": 1921.986,
      "eval_steps_per_second": 240.26,
      "step": 18000
    },
    {
      "entropy": 5.587586212158203,
      "epoch": 1.4008558646177787,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048054472174000615,
      "loss": 5.0759,
      "mean_token_accuracy": 0.19416010677814483,
      "num_tokens": 31250770.0,
      "step": 18005
    },
    {
      "entropy": 5.498182535171509,
      "epoch": 1.4012448939894961,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048053344640386207,
      "loss": 5.0179,
      "mean_token_accuracy": 0.20035575181245804,
      "num_tokens": 31259507.0,
      "step": 18010
    },
    {
      "entropy": 5.543242645263672,
      "epoch": 1.4016339233612138,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048052216794907505,
      "loss": 5.1281,
      "mean_token_accuracy": 0.19109611213207245,
      "num_tokens": 31268106.0,
      "step": 18015
    },
    {
      "entropy": 5.501767587661743,
      "epoch": 1.4020229527329313,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048051088637581624,
      "loss": 5.018,
      "mean_token_accuracy": 0.198416368663311,
      "num_tokens": 31276495.0,
      "step": 18020
    },
    {
      "entropy": 5.5984395980834964,
      "epoch": 1.402411982104649,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000480499601684257,
      "loss": 5.1065,
      "mean_token_accuracy": 0.1878896251320839,
      "num_tokens": 31284962.0,
      "step": 18025
    },
    {
      "entropy": 5.603269004821778,
      "epoch": 1.4028010114763665,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004804883138745682,
      "loss": 5.1412,
      "mean_token_accuracy": 0.18735029995441438,
      "num_tokens": 31294497.0,
      "step": 18030
    },
    {
      "entropy": 5.556841421127319,
      "epoch": 1.403190040848084,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004804770229469214,
      "loss": 5.0634,
      "mean_token_accuracy": 0.1966368466615677,
      "num_tokens": 31302607.0,
      "step": 18035
    },
    {
      "entropy": 5.6279784679412845,
      "epoch": 1.4035790702198017,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004804657289014878,
      "loss": 5.1234,
      "mean_token_accuracy": 0.18979247212409972,
      "num_tokens": 31311989.0,
      "step": 18040
    },
    {
      "entropy": 5.631705045700073,
      "epoch": 1.4039680995915191,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048045443173843884,
      "loss": 5.0391,
      "mean_token_accuracy": 0.19928252547979355,
      "num_tokens": 31320681.0,
      "step": 18045
    },
    {
      "entropy": 5.564608764648438,
      "epoch": 1.4043571289632366,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048044313145794587,
      "loss": 5.0607,
      "mean_token_accuracy": 0.18982219845056533,
      "num_tokens": 31328985.0,
      "step": 18050
    },
    {
      "entropy": 5.539751243591309,
      "epoch": 1.4047461583349543,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048043182806018034,
      "loss": 5.0182,
      "mean_token_accuracy": 0.19533439576625825,
      "num_tokens": 31337486.0,
      "step": 18055
    },
    {
      "entropy": 5.5585414409637455,
      "epoch": 1.4051351877066718,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048042052154531384,
      "loss": 5.1282,
      "mean_token_accuracy": 0.19507600963115693,
      "num_tokens": 31345831.0,
      "step": 18060
    },
    {
      "entropy": 5.507851505279541,
      "epoch": 1.4055242170783895,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004804092119135179,
      "loss": 5.0225,
      "mean_token_accuracy": 0.20095583647489548,
      "num_tokens": 31354791.0,
      "step": 18065
    },
    {
      "entropy": 5.529274797439575,
      "epoch": 1.405913246450107,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004803978991649641,
      "loss": 4.9873,
      "mean_token_accuracy": 0.19792200922966002,
      "num_tokens": 31362652.0,
      "step": 18070
    },
    {
      "entropy": 5.513758945465088,
      "epoch": 1.4063022758218247,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048038658329982404,
      "loss": 5.0661,
      "mean_token_accuracy": 0.19375650882720946,
      "num_tokens": 31371439.0,
      "step": 18075
    },
    {
      "entropy": 5.633718347549438,
      "epoch": 1.4066913051935421,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004803752643182695,
      "loss": 5.0862,
      "mean_token_accuracy": 0.18813119679689408,
      "num_tokens": 31380039.0,
      "step": 18080
    },
    {
      "entropy": 5.599810075759888,
      "epoch": 1.4070803345652596,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004803639422204723,
      "loss": 5.1886,
      "mean_token_accuracy": 0.18489259779453276,
      "num_tokens": 31389318.0,
      "step": 18085
    },
    {
      "entropy": 5.582911825180053,
      "epoch": 1.4074693639369773,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004803526170066041,
      "loss": 5.0724,
      "mean_token_accuracy": 0.19448621422052384,
      "num_tokens": 31398442.0,
      "step": 18090
    },
    {
      "entropy": 5.563860034942627,
      "epoch": 1.4078583933086948,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048034128867683674,
      "loss": 5.0634,
      "mean_token_accuracy": 0.1970498576760292,
      "num_tokens": 31406566.0,
      "step": 18095
    },
    {
      "entropy": 5.55968337059021,
      "epoch": 1.4082474226804123,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004803299572313422,
      "loss": 5.1318,
      "mean_token_accuracy": 0.18702000975608826,
      "num_tokens": 31415207.0,
      "step": 18100
    },
    {
      "entropy": 5.666746711730957,
      "epoch": 1.40863645205213,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004803186226702924,
      "loss": 5.12,
      "mean_token_accuracy": 0.18178979009389878,
      "num_tokens": 31424175.0,
      "step": 18105
    },
    {
      "entropy": 5.610285091400146,
      "epoch": 1.4090254814238474,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004803072849938592,
      "loss": 5.1526,
      "mean_token_accuracy": 0.18573977500200273,
      "num_tokens": 31434022.0,
      "step": 18110
    },
    {
      "entropy": 5.588834476470947,
      "epoch": 1.4094145107955651,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004802959442022149,
      "loss": 5.0182,
      "mean_token_accuracy": 0.19994013607501984,
      "num_tokens": 31442838.0,
      "step": 18115
    },
    {
      "entropy": 5.590457201004028,
      "epoch": 1.4098035401672826,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048028460029553126,
      "loss": 5.0257,
      "mean_token_accuracy": 0.1917799726128578,
      "num_tokens": 31451482.0,
      "step": 18120
    },
    {
      "entropy": 5.563886928558349,
      "epoch": 1.4101925695390003,
      "grad_norm": 1.125,
      "learning_rate": 0.00048027325327398065,
      "loss": 5.0683,
      "mean_token_accuracy": 0.19909143447875977,
      "num_tokens": 31460234.0,
      "step": 18125
    },
    {
      "entropy": 5.5280561447143555,
      "epoch": 1.4105815989107178,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004802619031377351,
      "loss": 5.0387,
      "mean_token_accuracy": 0.20026593804359435,
      "num_tokens": 31470466.0,
      "step": 18130
    },
    {
      "entropy": 5.509233713150024,
      "epoch": 1.4109706282824352,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048025054988696684,
      "loss": 5.1366,
      "mean_token_accuracy": 0.19033488035202026,
      "num_tokens": 31480901.0,
      "step": 18135
    },
    {
      "entropy": 5.571567058563232,
      "epoch": 1.411359657654153,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048023919352184827,
      "loss": 5.0393,
      "mean_token_accuracy": 0.19016336947679519,
      "num_tokens": 31490329.0,
      "step": 18140
    },
    {
      "entropy": 5.549619579315186,
      "epoch": 1.4117486870258704,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048022783404255156,
      "loss": 5.0826,
      "mean_token_accuracy": 0.19608550369739533,
      "num_tokens": 31499519.0,
      "step": 18145
    },
    {
      "entropy": 5.597281265258789,
      "epoch": 1.412137716397588,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004802164714492491,
      "loss": 5.1131,
      "mean_token_accuracy": 0.19269932955503463,
      "num_tokens": 31507899.0,
      "step": 18150
    },
    {
      "entropy": 5.651979970932007,
      "epoch": 1.4125267457693056,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048020510574211335,
      "loss": 5.2543,
      "mean_token_accuracy": 0.18247729837894439,
      "num_tokens": 31517861.0,
      "step": 18155
    },
    {
      "entropy": 5.606938791275025,
      "epoch": 1.412915775141023,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048019373692131674,
      "loss": 5.1387,
      "mean_token_accuracy": 0.18567328304052352,
      "num_tokens": 31527273.0,
      "step": 18160
    },
    {
      "entropy": 5.607219982147217,
      "epoch": 1.4133048045127408,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048018236498703176,
      "loss": 5.0655,
      "mean_token_accuracy": 0.1965159773826599,
      "num_tokens": 31536221.0,
      "step": 18165
    },
    {
      "entropy": 5.568474578857422,
      "epoch": 1.4136938338844582,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048017098993943097,
      "loss": 5.021,
      "mean_token_accuracy": 0.19845947325229646,
      "num_tokens": 31544329.0,
      "step": 18170
    },
    {
      "entropy": 5.611068677902222,
      "epoch": 1.414082863256176,
      "grad_norm": 1.109375,
      "learning_rate": 0.000480159611778687,
      "loss": 5.0907,
      "mean_token_accuracy": 0.19424625635147094,
      "num_tokens": 31552323.0,
      "step": 18175
    },
    {
      "entropy": 5.560807466506958,
      "epoch": 1.4144718926278934,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048014823050497243,
      "loss": 5.0526,
      "mean_token_accuracy": 0.1948814034461975,
      "num_tokens": 31560432.0,
      "step": 18180
    },
    {
      "entropy": 5.472802782058716,
      "epoch": 1.4148609219996109,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048013684611846,
      "loss": 5.0118,
      "mean_token_accuracy": 0.20406582653522493,
      "num_tokens": 31569294.0,
      "step": 18185
    },
    {
      "entropy": 5.527368497848511,
      "epoch": 1.4152499513713286,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048012545861932245,
      "loss": 5.0358,
      "mean_token_accuracy": 0.1995076820254326,
      "num_tokens": 31577860.0,
      "step": 18190
    },
    {
      "entropy": 5.537588977813721,
      "epoch": 1.415638980743046,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004801140680077325,
      "loss": 4.9971,
      "mean_token_accuracy": 0.19913579970598222,
      "num_tokens": 31586034.0,
      "step": 18195
    },
    {
      "entropy": 5.578590726852417,
      "epoch": 1.4160280101147635,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004801026742838631,
      "loss": 5.0231,
      "mean_token_accuracy": 0.1959235206246376,
      "num_tokens": 31594911.0,
      "step": 18200
    },
    {
      "entropy": 5.592549848556518,
      "epoch": 1.4164170394864812,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004800912774478871,
      "loss": 5.1242,
      "mean_token_accuracy": 0.19008546620607375,
      "num_tokens": 31603477.0,
      "step": 18205
    },
    {
      "entropy": 5.630913209915161,
      "epoch": 1.416806068858199,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004800798774999773,
      "loss": 5.0898,
      "mean_token_accuracy": 0.19573789238929748,
      "num_tokens": 31612288.0,
      "step": 18210
    },
    {
      "entropy": 5.56613712310791,
      "epoch": 1.4171950982299164,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048006847444030686,
      "loss": 5.0112,
      "mean_token_accuracy": 0.20468000918626786,
      "num_tokens": 31621580.0,
      "step": 18215
    },
    {
      "entropy": 5.497746849060059,
      "epoch": 1.4175841276016339,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004800570682690487,
      "loss": 5.0306,
      "mean_token_accuracy": 0.1915210574865341,
      "num_tokens": 31629860.0,
      "step": 18220
    },
    {
      "entropy": 5.629140853881836,
      "epoch": 1.4179731569733516,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004800456589863759,
      "loss": 5.0922,
      "mean_token_accuracy": 0.18835911750793458,
      "num_tokens": 31639012.0,
      "step": 18225
    },
    {
      "entropy": 5.58698582649231,
      "epoch": 1.418362186345069,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004800342465924616,
      "loss": 5.1129,
      "mean_token_accuracy": 0.19246853291988372,
      "num_tokens": 31648675.0,
      "step": 18230
    },
    {
      "entropy": 5.58170371055603,
      "epoch": 1.4187512157167865,
      "grad_norm": 1.140625,
      "learning_rate": 0.000480022831087479,
      "loss": 5.0721,
      "mean_token_accuracy": 0.2007998287677765,
      "num_tokens": 31656570.0,
      "step": 18235
    },
    {
      "entropy": 5.511694478988647,
      "epoch": 1.4191402450885042,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004800114124716012,
      "loss": 5.0058,
      "mean_token_accuracy": 0.20106196999549866,
      "num_tokens": 31665636.0,
      "step": 18240
    },
    {
      "entropy": 5.613012313842773,
      "epoch": 1.4195292744602217,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004799999907450015,
      "loss": 5.13,
      "mean_token_accuracy": 0.19319438189268112,
      "num_tokens": 31673638.0,
      "step": 18245
    },
    {
      "entropy": 5.583807420730591,
      "epoch": 1.4199183038319392,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004799885659078533,
      "loss": 5.1011,
      "mean_token_accuracy": 0.19507333189249038,
      "num_tokens": 31681886.0,
      "step": 18250
    },
    {
      "entropy": 5.631104373931885,
      "epoch": 1.4203073332036569,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004799771379603299,
      "loss": 5.2108,
      "mean_token_accuracy": 0.18666794896125793,
      "num_tokens": 31690349.0,
      "step": 18255
    },
    {
      "entropy": 5.5384868621826175,
      "epoch": 1.4206963625753746,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004799657069026046,
      "loss": 5.0804,
      "mean_token_accuracy": 0.18777843564748764,
      "num_tokens": 31699183.0,
      "step": 18260
    },
    {
      "entropy": 5.546294021606445,
      "epoch": 1.421085391947092,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047995427273485097,
      "loss": 5.0518,
      "mean_token_accuracy": 0.2011060208082199,
      "num_tokens": 31707759.0,
      "step": 18265
    },
    {
      "entropy": 5.642793369293213,
      "epoch": 1.4214744213188095,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047994283545724247,
      "loss": 5.1843,
      "mean_token_accuracy": 0.18489649146795273,
      "num_tokens": 31716383.0,
      "step": 18270
    },
    {
      "entropy": 5.604320096969604,
      "epoch": 1.4218634506905272,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047993139506995273,
      "loss": 5.0848,
      "mean_token_accuracy": 0.1846178337931633,
      "num_tokens": 31724617.0,
      "step": 18275
    },
    {
      "entropy": 5.586402225494385,
      "epoch": 1.4222524800622447,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004799199515731551,
      "loss": 5.0323,
      "mean_token_accuracy": 0.19798973798751832,
      "num_tokens": 31732027.0,
      "step": 18280
    },
    {
      "entropy": 5.485814428329467,
      "epoch": 1.4226415094339622,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047990850496702346,
      "loss": 5.0009,
      "mean_token_accuracy": 0.20008625090122223,
      "num_tokens": 31740707.0,
      "step": 18285
    },
    {
      "entropy": 5.556511878967285,
      "epoch": 1.4230305388056799,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004798970552517314,
      "loss": 5.0263,
      "mean_token_accuracy": 0.19690660685300826,
      "num_tokens": 31748864.0,
      "step": 18290
    },
    {
      "entropy": 5.652910423278809,
      "epoch": 1.4234195681773973,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047988560242745264,
      "loss": 5.1992,
      "mean_token_accuracy": 0.1871165856719017,
      "num_tokens": 31758356.0,
      "step": 18295
    },
    {
      "entropy": 5.596076726913452,
      "epoch": 1.423808597549115,
      "grad_norm": 1.1875,
      "learning_rate": 0.000479874146494361,
      "loss": 5.0762,
      "mean_token_accuracy": 0.20283491164445877,
      "num_tokens": 31766911.0,
      "step": 18300
    },
    {
      "entropy": 5.575757598876953,
      "epoch": 1.4241976269208325,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004798626874526302,
      "loss": 5.0877,
      "mean_token_accuracy": 0.19178476780653,
      "num_tokens": 31775551.0,
      "step": 18305
    },
    {
      "entropy": 5.610908174514771,
      "epoch": 1.4245866562925502,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047985122530243426,
      "loss": 5.2317,
      "mean_token_accuracy": 0.18173079937696457,
      "num_tokens": 31784064.0,
      "step": 18310
    },
    {
      "entropy": 5.631493711471558,
      "epoch": 1.4249756856642677,
      "grad_norm": 1.09375,
      "learning_rate": 0.000479839760043947,
      "loss": 5.0987,
      "mean_token_accuracy": 0.18938282132148743,
      "num_tokens": 31792574.0,
      "step": 18315
    },
    {
      "entropy": 5.566015338897705,
      "epoch": 1.4253647150359852,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047982829167734245,
      "loss": 4.9375,
      "mean_token_accuracy": 0.20298290401697158,
      "num_tokens": 31800331.0,
      "step": 18320
    },
    {
      "entropy": 5.530305337905884,
      "epoch": 1.4257537444077029,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047981682020279456,
      "loss": 4.9884,
      "mean_token_accuracy": 0.19619510620832442,
      "num_tokens": 31808474.0,
      "step": 18325
    },
    {
      "entropy": 5.584475660324097,
      "epoch": 1.4261427737794203,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004798053456204775,
      "loss": 5.116,
      "mean_token_accuracy": 0.18598145991563797,
      "num_tokens": 31817026.0,
      "step": 18330
    },
    {
      "entropy": 5.510889339447021,
      "epoch": 1.4265318031511378,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004797938679305651,
      "loss": 4.9596,
      "mean_token_accuracy": 0.19875158369541168,
      "num_tokens": 31825096.0,
      "step": 18335
    },
    {
      "entropy": 5.542738342285157,
      "epoch": 1.4269208325228555,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047978238713323193,
      "loss": 5.1227,
      "mean_token_accuracy": 0.19513861387968062,
      "num_tokens": 31833650.0,
      "step": 18340
    },
    {
      "entropy": 5.566265726089478,
      "epoch": 1.427309861894573,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047977090322865183,
      "loss": 5.0512,
      "mean_token_accuracy": 0.1949874132871628,
      "num_tokens": 31841803.0,
      "step": 18345
    },
    {
      "entropy": 5.6203724384307865,
      "epoch": 1.4276988912662907,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004797594162169993,
      "loss": 5.1024,
      "mean_token_accuracy": 0.19450120031833648,
      "num_tokens": 31850121.0,
      "step": 18350
    },
    {
      "entropy": 5.554633712768554,
      "epoch": 1.4280879206380082,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004797479260984485,
      "loss": 5.06,
      "mean_token_accuracy": 0.19070458561182022,
      "num_tokens": 31858585.0,
      "step": 18355
    },
    {
      "entropy": 5.548305988311768,
      "epoch": 1.4284769500097259,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047973643287317386,
      "loss": 5.046,
      "mean_token_accuracy": 0.1958570122718811,
      "num_tokens": 31866199.0,
      "step": 18360
    },
    {
      "entropy": 5.532158422470093,
      "epoch": 1.4288659793814433,
      "grad_norm": 1.0,
      "learning_rate": 0.00047972493654134963,
      "loss": 5.0621,
      "mean_token_accuracy": 0.19767487347126006,
      "num_tokens": 31875423.0,
      "step": 18365
    },
    {
      "entropy": 5.513716554641723,
      "epoch": 1.4292550087531608,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047971343710315043,
      "loss": 4.9667,
      "mean_token_accuracy": 0.20076639503240584,
      "num_tokens": 31883110.0,
      "step": 18370
    },
    {
      "entropy": 5.514259958267212,
      "epoch": 1.4296440381248785,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004797019345587506,
      "loss": 5.0745,
      "mean_token_accuracy": 0.19770575016736985,
      "num_tokens": 31891588.0,
      "step": 18375
    },
    {
      "entropy": 5.548969173431397,
      "epoch": 1.430033067496596,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004796904289083248,
      "loss": 5.0205,
      "mean_token_accuracy": 0.19764809608459472,
      "num_tokens": 31900671.0,
      "step": 18380
    },
    {
      "entropy": 5.607488679885864,
      "epoch": 1.4304220968683135,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004796789201520474,
      "loss": 5.1029,
      "mean_token_accuracy": 0.1922275975346565,
      "num_tokens": 31909188.0,
      "step": 18385
    },
    {
      "entropy": 5.5605306148529055,
      "epoch": 1.4308111262400312,
      "grad_norm": 1.125,
      "learning_rate": 0.00047966740829009333,
      "loss": 5.0333,
      "mean_token_accuracy": 0.20165584236383438,
      "num_tokens": 31917607.0,
      "step": 18390
    },
    {
      "entropy": 5.568596649169922,
      "epoch": 1.4312001556117486,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000479655893322637,
      "loss": 5.0857,
      "mean_token_accuracy": 0.18877284228801727,
      "num_tokens": 31926510.0,
      "step": 18395
    },
    {
      "entropy": 5.580822372436524,
      "epoch": 1.4315891849834663,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004796443752498532,
      "loss": 5.1778,
      "mean_token_accuracy": 0.18380894958972932,
      "num_tokens": 31935550.0,
      "step": 18400
    },
    {
      "entropy": 5.6675046443939205,
      "epoch": 1.4319782143551838,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004796328540719169,
      "loss": 5.2419,
      "mean_token_accuracy": 0.1793557345867157,
      "num_tokens": 31944499.0,
      "step": 18405
    },
    {
      "entropy": 5.64169659614563,
      "epoch": 1.4323672437269015,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004796213297890026,
      "loss": 4.9718,
      "mean_token_accuracy": 0.2024772971868515,
      "num_tokens": 31952622.0,
      "step": 18410
    },
    {
      "entropy": 5.513220500946045,
      "epoch": 1.432756273098619,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004796098024012853,
      "loss": 5.0222,
      "mean_token_accuracy": 0.19552865028381347,
      "num_tokens": 31961606.0,
      "step": 18415
    },
    {
      "entropy": 5.527184867858887,
      "epoch": 1.4331453024703364,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047959827190894,
      "loss": 5.1237,
      "mean_token_accuracy": 0.19069355726242065,
      "num_tokens": 31970312.0,
      "step": 18420
    },
    {
      "entropy": 5.581167793273925,
      "epoch": 1.4335343318420541,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047958673831214157,
      "loss": 5.0874,
      "mean_token_accuracy": 0.19458332657814026,
      "num_tokens": 31978718.0,
      "step": 18425
    },
    {
      "entropy": 5.617711687088013,
      "epoch": 1.4339233612137716,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047957520161106496,
      "loss": 5.1133,
      "mean_token_accuracy": 0.18958386331796645,
      "num_tokens": 31987451.0,
      "step": 18430
    },
    {
      "entropy": 5.60126748085022,
      "epoch": 1.434312390585489,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004795636618058853,
      "loss": 4.9235,
      "mean_token_accuracy": 0.20605753362178802,
      "num_tokens": 31994977.0,
      "step": 18435
    },
    {
      "entropy": 5.500937986373901,
      "epoch": 1.4347014199572068,
      "grad_norm": 1.125,
      "learning_rate": 0.0004795521188967777,
      "loss": 5.0023,
      "mean_token_accuracy": 0.20137885957956314,
      "num_tokens": 32003804.0,
      "step": 18440
    },
    {
      "entropy": 5.605511903762817,
      "epoch": 1.4350904493289243,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004795405728839172,
      "loss": 5.1304,
      "mean_token_accuracy": 0.18806419670581817,
      "num_tokens": 32012090.0,
      "step": 18445
    },
    {
      "entropy": 5.56484956741333,
      "epoch": 1.435479478700642,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004795290237674792,
      "loss": 5.0906,
      "mean_token_accuracy": 0.18754435926675797,
      "num_tokens": 32019664.0,
      "step": 18450
    },
    {
      "entropy": 5.577170991897583,
      "epoch": 1.4358685080723594,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047951747154763866,
      "loss": 5.141,
      "mean_token_accuracy": 0.19115390479564667,
      "num_tokens": 32028457.0,
      "step": 18455
    },
    {
      "entropy": 5.541293144226074,
      "epoch": 1.4362575374440771,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047950591622457106,
      "loss": 4.9626,
      "mean_token_accuracy": 0.1959880918264389,
      "num_tokens": 32036730.0,
      "step": 18460
    },
    {
      "entropy": 5.625739002227784,
      "epoch": 1.4366465668157946,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047949435779845174,
      "loss": 5.1478,
      "mean_token_accuracy": 0.19264086037874223,
      "num_tokens": 32045657.0,
      "step": 18465
    },
    {
      "entropy": 5.5570303440094,
      "epoch": 1.437035596187512,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047948279626945596,
      "loss": 5.0636,
      "mean_token_accuracy": 0.1928187519311905,
      "num_tokens": 32053973.0,
      "step": 18470
    },
    {
      "entropy": 5.653687810897827,
      "epoch": 1.4374246255592298,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047947123163775924,
      "loss": 5.1649,
      "mean_token_accuracy": 0.18393130600452423,
      "num_tokens": 32062833.0,
      "step": 18475
    },
    {
      "entropy": 5.548023509979248,
      "epoch": 1.4378136549309473,
      "grad_norm": 1.078125,
      "learning_rate": 0.000479459663903537,
      "loss": 5.0212,
      "mean_token_accuracy": 0.20152370929718016,
      "num_tokens": 32071578.0,
      "step": 18480
    },
    {
      "entropy": 5.559324407577515,
      "epoch": 1.4382026843026647,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004794480930669649,
      "loss": 4.9648,
      "mean_token_accuracy": 0.197468800842762,
      "num_tokens": 32079847.0,
      "step": 18485
    },
    {
      "entropy": 5.538774633407593,
      "epoch": 1.4385917136743824,
      "grad_norm": 1.125,
      "learning_rate": 0.0004794365191282183,
      "loss": 5.0268,
      "mean_token_accuracy": 0.19269732981920243,
      "num_tokens": 32089751.0,
      "step": 18490
    },
    {
      "entropy": 5.601061820983887,
      "epoch": 1.4389807430461,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047942494208747297,
      "loss": 5.1858,
      "mean_token_accuracy": 0.18932649344205857,
      "num_tokens": 32097791.0,
      "step": 18495
    },
    {
      "entropy": 5.608306074142456,
      "epoch": 1.4393697724178176,
      "grad_norm": 1.125,
      "learning_rate": 0.0004794133619449045,
      "loss": 4.998,
      "mean_token_accuracy": 0.20100484937429428,
      "num_tokens": 32106417.0,
      "step": 18500
    },
    {
      "entropy": 5.552778720855713,
      "epoch": 1.439758801789535,
      "grad_norm": 1.125,
      "learning_rate": 0.0004794017787006886,
      "loss": 5.0942,
      "mean_token_accuracy": 0.1918921485543251,
      "num_tokens": 32115035.0,
      "step": 18505
    },
    {
      "entropy": 5.535371017456055,
      "epoch": 1.4401478311612528,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000479390192355001,
      "loss": 5.0431,
      "mean_token_accuracy": 0.19497817903757095,
      "num_tokens": 32123704.0,
      "step": 18510
    },
    {
      "entropy": 5.622191429138184,
      "epoch": 1.4405368605329703,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004793786029080176,
      "loss": 5.1229,
      "mean_token_accuracy": 0.1895605206489563,
      "num_tokens": 32132672.0,
      "step": 18515
    },
    {
      "entropy": 5.567349004745483,
      "epoch": 1.4409258899046877,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004793670103599143,
      "loss": 5.0326,
      "mean_token_accuracy": 0.19571120738983155,
      "num_tokens": 32142135.0,
      "step": 18520
    },
    {
      "entropy": 5.5992950916290285,
      "epoch": 1.4413149192764054,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047935541471086677,
      "loss": 5.1343,
      "mean_token_accuracy": 0.18213363885879516,
      "num_tokens": 32151261.0,
      "step": 18525
    },
    {
      "entropy": 5.520450592041016,
      "epoch": 1.441703948648123,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004793438159610511,
      "loss": 5.0258,
      "mean_token_accuracy": 0.1970395788550377,
      "num_tokens": 32159454.0,
      "step": 18530
    },
    {
      "entropy": 5.617293882369995,
      "epoch": 1.4420929780198404,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047933221411064334,
      "loss": 5.1208,
      "mean_token_accuracy": 0.18518660068511963,
      "num_tokens": 32168706.0,
      "step": 18535
    },
    {
      "entropy": 5.688238620758057,
      "epoch": 1.442482007391558,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004793206091598194,
      "loss": 5.1716,
      "mean_token_accuracy": 0.1869201824069023,
      "num_tokens": 32177892.0,
      "step": 18540
    },
    {
      "entropy": 5.58838586807251,
      "epoch": 1.4428710367632755,
      "grad_norm": 1.125,
      "learning_rate": 0.0004793090011087554,
      "loss": 5.0649,
      "mean_token_accuracy": 0.1936390817165375,
      "num_tokens": 32186081.0,
      "step": 18545
    },
    {
      "entropy": 5.572668790817261,
      "epoch": 1.4432600661349932,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047929738995762755,
      "loss": 5.0256,
      "mean_token_accuracy": 0.19540503919124602,
      "num_tokens": 32193778.0,
      "step": 18550
    },
    {
      "entropy": 5.497219371795654,
      "epoch": 1.4436490955067107,
      "grad_norm": 1.125,
      "learning_rate": 0.000479285775706612,
      "loss": 5.01,
      "mean_token_accuracy": 0.1967809021472931,
      "num_tokens": 32202773.0,
      "step": 18555
    },
    {
      "entropy": 5.571686935424805,
      "epoch": 1.4440381248784284,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047927415835588496,
      "loss": 5.0412,
      "mean_token_accuracy": 0.1946617603302002,
      "num_tokens": 32212105.0,
      "step": 18560
    },
    {
      "entropy": 5.510843181610108,
      "epoch": 1.444427154250146,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047926253790562265,
      "loss": 4.9309,
      "mean_token_accuracy": 0.20008269399404527,
      "num_tokens": 32220550.0,
      "step": 18565
    },
    {
      "entropy": 5.518900918960571,
      "epoch": 1.4448161836218634,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047925091435600147,
      "loss": 4.9124,
      "mean_token_accuracy": 0.20338450670242308,
      "num_tokens": 32228562.0,
      "step": 18570
    },
    {
      "entropy": 5.578421926498413,
      "epoch": 1.445205212993581,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047923928770719776,
      "loss": 5.0924,
      "mean_token_accuracy": 0.19500968605279922,
      "num_tokens": 32237172.0,
      "step": 18575
    },
    {
      "entropy": 5.512694454193115,
      "epoch": 1.4455942423652985,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047922765795938797,
      "loss": 5.0724,
      "mean_token_accuracy": 0.19186332821846008,
      "num_tokens": 32246211.0,
      "step": 18580
    },
    {
      "entropy": 5.470756340026855,
      "epoch": 1.445983271737016,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004792160251127485,
      "loss": 4.9998,
      "mean_token_accuracy": 0.19970700591802598,
      "num_tokens": 32255366.0,
      "step": 18585
    },
    {
      "entropy": 5.650744342803955,
      "epoch": 1.4463723011087337,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047920438916745586,
      "loss": 5.1791,
      "mean_token_accuracy": 0.18737162202596663,
      "num_tokens": 32264342.0,
      "step": 18590
    },
    {
      "entropy": 5.601012706756592,
      "epoch": 1.4467613304804512,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004791927501236866,
      "loss": 5.1048,
      "mean_token_accuracy": 0.18978851288557053,
      "num_tokens": 32273184.0,
      "step": 18595
    },
    {
      "entropy": 5.594601202011108,
      "epoch": 1.4471503598521689,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047918110798161754,
      "loss": 5.1312,
      "mean_token_accuracy": 0.1859067752957344,
      "num_tokens": 32282303.0,
      "step": 18600
    },
    {
      "entropy": 5.596266508102417,
      "epoch": 1.4475393892238864,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047916946274142503,
      "loss": 5.0258,
      "mean_token_accuracy": 0.19668152183294296,
      "num_tokens": 32291118.0,
      "step": 18605
    },
    {
      "entropy": 5.597677993774414,
      "epoch": 1.447928418595604,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047915781440328593,
      "loss": 5.0666,
      "mean_token_accuracy": 0.18761706054210664,
      "num_tokens": 32299746.0,
      "step": 18610
    },
    {
      "entropy": 5.546115446090698,
      "epoch": 1.4483174479673215,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004791461629673769,
      "loss": 5.0579,
      "mean_token_accuracy": 0.19625270515680313,
      "num_tokens": 32308515.0,
      "step": 18615
    },
    {
      "entropy": 5.601655912399292,
      "epoch": 1.448706477339039,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004791345084338748,
      "loss": 5.1762,
      "mean_token_accuracy": 0.18931110054254532,
      "num_tokens": 32317607.0,
      "step": 18620
    },
    {
      "entropy": 5.632953119277954,
      "epoch": 1.4490955067107567,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004791228508029565,
      "loss": 5.0785,
      "mean_token_accuracy": 0.19761386662721633,
      "num_tokens": 32325822.0,
      "step": 18625
    },
    {
      "entropy": 5.644744634628296,
      "epoch": 1.4494845360824742,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047911119007479873,
      "loss": 5.1697,
      "mean_token_accuracy": 0.18785120397806168,
      "num_tokens": 32336297.0,
      "step": 18630
    },
    {
      "entropy": 5.57390832901001,
      "epoch": 1.4498735654541917,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047909952624957866,
      "loss": 5.0693,
      "mean_token_accuracy": 0.1904062271118164,
      "num_tokens": 32344306.0,
      "step": 18635
    },
    {
      "entropy": 5.620115900039673,
      "epoch": 1.4502625948259094,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000479087859327473,
      "loss": 5.1724,
      "mean_token_accuracy": 0.18820504248142242,
      "num_tokens": 32353668.0,
      "step": 18640
    },
    {
      "entropy": 5.599977254867554,
      "epoch": 1.450651624197627,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000479076189308659,
      "loss": 5.0526,
      "mean_token_accuracy": 0.19765406399965285,
      "num_tokens": 32361563.0,
      "step": 18645
    },
    {
      "entropy": 5.623808908462524,
      "epoch": 1.4510406535693445,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047906451619331364,
      "loss": 5.1714,
      "mean_token_accuracy": 0.1899188682436943,
      "num_tokens": 32369612.0,
      "step": 18650
    },
    {
      "entropy": 5.527958965301513,
      "epoch": 1.451429682941062,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047905283998161404,
      "loss": 4.9581,
      "mean_token_accuracy": 0.20327578037977218,
      "num_tokens": 32379013.0,
      "step": 18655
    },
    {
      "entropy": 5.543435573577881,
      "epoch": 1.4518187123127797,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047904116067373743,
      "loss": 5.0692,
      "mean_token_accuracy": 0.19372039437294006,
      "num_tokens": 32387933.0,
      "step": 18660
    },
    {
      "entropy": 5.5895215511322025,
      "epoch": 1.4522077416844972,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004790294782698609,
      "loss": 5.1056,
      "mean_token_accuracy": 0.19261073768138887,
      "num_tokens": 32395669.0,
      "step": 18665
    },
    {
      "entropy": 5.518316030502319,
      "epoch": 1.4525967710562147,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004790177927701618,
      "loss": 4.9574,
      "mean_token_accuracy": 0.20497757345438003,
      "num_tokens": 32403676.0,
      "step": 18670
    },
    {
      "entropy": 5.580837202072144,
      "epoch": 1.4529858004279324,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004790061041748174,
      "loss": 5.0365,
      "mean_token_accuracy": 0.19945041239261627,
      "num_tokens": 32412089.0,
      "step": 18675
    },
    {
      "entropy": 5.578955793380738,
      "epoch": 1.4533748297996498,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004789944124840051,
      "loss": 5.0109,
      "mean_token_accuracy": 0.19762406200170518,
      "num_tokens": 32420357.0,
      "step": 18680
    },
    {
      "entropy": 5.6209557056427,
      "epoch": 1.4537638591713673,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004789827176979022,
      "loss": 5.1075,
      "mean_token_accuracy": 0.19459480792284012,
      "num_tokens": 32429118.0,
      "step": 18685
    },
    {
      "entropy": 5.61270956993103,
      "epoch": 1.454152888543085,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004789710198166864,
      "loss": 5.0226,
      "mean_token_accuracy": 0.19631969481706618,
      "num_tokens": 32437266.0,
      "step": 18690
    },
    {
      "entropy": 5.583590841293335,
      "epoch": 1.4545419179148027,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047895931884053497,
      "loss": 5.1226,
      "mean_token_accuracy": 0.19126349538564683,
      "num_tokens": 32446014.0,
      "step": 18695
    },
    {
      "entropy": 5.550280332565308,
      "epoch": 1.4549309472865202,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047894761476962547,
      "loss": 5.1083,
      "mean_token_accuracy": 0.18722242414951323,
      "num_tokens": 32454591.0,
      "step": 18700
    },
    {
      "entropy": 5.575741958618164,
      "epoch": 1.4553199766582376,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047893590760413545,
      "loss": 5.0597,
      "mean_token_accuracy": 0.1984192058444023,
      "num_tokens": 32462748.0,
      "step": 18705
    },
    {
      "entropy": 5.581970691680908,
      "epoch": 1.4557090060299553,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047892419734424276,
      "loss": 5.0358,
      "mean_token_accuracy": 0.19910039901733398,
      "num_tokens": 32471792.0,
      "step": 18710
    },
    {
      "entropy": 5.598289346694946,
      "epoch": 1.4560980354016728,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047891248399012495,
      "loss": 5.1146,
      "mean_token_accuracy": 0.19560667425394057,
      "num_tokens": 32480804.0,
      "step": 18715
    },
    {
      "entropy": 5.611759996414184,
      "epoch": 1.4564870647733903,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004789007675419597,
      "loss": 5.0657,
      "mean_token_accuracy": 0.19722296297550201,
      "num_tokens": 32489591.0,
      "step": 18720
    },
    {
      "entropy": 5.5477344512939455,
      "epoch": 1.456876094145108,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047888904799992486,
      "loss": 5.0679,
      "mean_token_accuracy": 0.19624695479869841,
      "num_tokens": 32498164.0,
      "step": 18725
    },
    {
      "entropy": 5.526124954223633,
      "epoch": 1.4572651235168255,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047887732536419826,
      "loss": 4.912,
      "mean_token_accuracy": 0.20134677588939667,
      "num_tokens": 32505909.0,
      "step": 18730
    },
    {
      "entropy": 5.573574018478394,
      "epoch": 1.4576541528885432,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004788655996349577,
      "loss": 5.0376,
      "mean_token_accuracy": 0.19465433061122894,
      "num_tokens": 32513990.0,
      "step": 18735
    },
    {
      "entropy": 5.53552656173706,
      "epoch": 1.4580431822602606,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047885387081238125,
      "loss": 5.0763,
      "mean_token_accuracy": 0.1893923059105873,
      "num_tokens": 32522932.0,
      "step": 18740
    },
    {
      "entropy": 5.5612226009368895,
      "epoch": 1.4584322116319783,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004788421388966467,
      "loss": 5.1491,
      "mean_token_accuracy": 0.18405068814754486,
      "num_tokens": 32530953.0,
      "step": 18745
    },
    {
      "entropy": 5.558132600784302,
      "epoch": 1.4588212410036958,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047883040388793207,
      "loss": 4.9984,
      "mean_token_accuracy": 0.1950299322605133,
      "num_tokens": 32539812.0,
      "step": 18750
    },
    {
      "entropy": 5.573992347717285,
      "epoch": 1.4592102703754133,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047881866578641563,
      "loss": 4.9797,
      "mean_token_accuracy": 0.19788700640201567,
      "num_tokens": 32548425.0,
      "step": 18755
    },
    {
      "entropy": 5.554071235656738,
      "epoch": 1.459599299747131,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004788069245922753,
      "loss": 5.0395,
      "mean_token_accuracy": 0.19586289525032044,
      "num_tokens": 32557619.0,
      "step": 18760
    },
    {
      "entropy": 5.59688811302185,
      "epoch": 1.4599883291188485,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004787951803056893,
      "loss": 5.0637,
      "mean_token_accuracy": 0.18809051364660262,
      "num_tokens": 32565602.0,
      "step": 18765
    },
    {
      "entropy": 5.596247243881225,
      "epoch": 1.460377358490566,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004787834329268358,
      "loss": 5.074,
      "mean_token_accuracy": 0.1922636702656746,
      "num_tokens": 32573691.0,
      "step": 18770
    },
    {
      "entropy": 5.600850200653076,
      "epoch": 1.4607663878622836,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000478771682455893,
      "loss": 5.1574,
      "mean_token_accuracy": 0.18532035648822784,
      "num_tokens": 32582642.0,
      "step": 18775
    },
    {
      "entropy": 5.58724536895752,
      "epoch": 1.461155417234001,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004787599288930393,
      "loss": 5.0654,
      "mean_token_accuracy": 0.19453577399253846,
      "num_tokens": 32591328.0,
      "step": 18780
    },
    {
      "entropy": 5.5703472137451175,
      "epoch": 1.4615444466057188,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000478748172238453,
      "loss": 5.0261,
      "mean_token_accuracy": 0.1928997665643692,
      "num_tokens": 32599286.0,
      "step": 18785
    },
    {
      "entropy": 5.603278923034668,
      "epoch": 1.4619334759774363,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004787364124923125,
      "loss": 5.1898,
      "mean_token_accuracy": 0.18411661833524703,
      "num_tokens": 32608156.0,
      "step": 18790
    },
    {
      "entropy": 5.572177934646606,
      "epoch": 1.462322505349154,
      "grad_norm": 1.125,
      "learning_rate": 0.00047872464965479625,
      "loss": 5.0074,
      "mean_token_accuracy": 0.19800837486982345,
      "num_tokens": 32616134.0,
      "step": 18795
    },
    {
      "entropy": 5.605358695983886,
      "epoch": 1.4627115347208715,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004787128837260826,
      "loss": 5.057,
      "mean_token_accuracy": 0.1914464935660362,
      "num_tokens": 32624711.0,
      "step": 18800
    },
    {
      "entropy": 5.5844427108764645,
      "epoch": 1.463100564092589,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004787011147063503,
      "loss": 5.0671,
      "mean_token_accuracy": 0.1945657417178154,
      "num_tokens": 32633049.0,
      "step": 18805
    },
    {
      "entropy": 5.5803216934204105,
      "epoch": 1.4634895934643066,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004786893425957777,
      "loss": 5.0202,
      "mean_token_accuracy": 0.19808992594480515,
      "num_tokens": 32641173.0,
      "step": 18810
    },
    {
      "entropy": 5.561643028259278,
      "epoch": 1.463878622836024,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004786775673945436,
      "loss": 5.006,
      "mean_token_accuracy": 0.20172724723815919,
      "num_tokens": 32650275.0,
      "step": 18815
    },
    {
      "entropy": 5.57724347114563,
      "epoch": 1.4642676522077416,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047866578910282656,
      "loss": 5.028,
      "mean_token_accuracy": 0.20235637873411177,
      "num_tokens": 32658675.0,
      "step": 18820
    },
    {
      "entropy": 5.582719945907593,
      "epoch": 1.4646566815794593,
      "grad_norm": 1.125,
      "learning_rate": 0.00047865400772080535,
      "loss": 5.0318,
      "mean_token_accuracy": 0.19783681333065034,
      "num_tokens": 32666643.0,
      "step": 18825
    },
    {
      "entropy": 5.550979137420654,
      "epoch": 1.4650457109511767,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047864222324865875,
      "loss": 4.948,
      "mean_token_accuracy": 0.1993061661720276,
      "num_tokens": 32674791.0,
      "step": 18830
    },
    {
      "entropy": 5.584186697006226,
      "epoch": 1.4654347403228944,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004786304356865655,
      "loss": 5.0423,
      "mean_token_accuracy": 0.20804574489593505,
      "num_tokens": 32682790.0,
      "step": 18835
    },
    {
      "entropy": 5.594402551651001,
      "epoch": 1.465823769694612,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047861864503470457,
      "loss": 5.1548,
      "mean_token_accuracy": 0.18853792250156404,
      "num_tokens": 32692701.0,
      "step": 18840
    },
    {
      "entropy": 5.583247470855713,
      "epoch": 1.4662127990663296,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004786068512932547,
      "loss": 5.0914,
      "mean_token_accuracy": 0.19674921333789824,
      "num_tokens": 32701297.0,
      "step": 18845
    },
    {
      "entropy": 5.58556752204895,
      "epoch": 1.466601828438047,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000478595054462395,
      "loss": 5.0077,
      "mean_token_accuracy": 0.19780241698026657,
      "num_tokens": 32710188.0,
      "step": 18850
    },
    {
      "entropy": 5.514643907546997,
      "epoch": 1.4669908578097646,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047858325454230437,
      "loss": 5.0548,
      "mean_token_accuracy": 0.1983303740620613,
      "num_tokens": 32719255.0,
      "step": 18855
    },
    {
      "entropy": 5.550378799438477,
      "epoch": 1.4673798871814823,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004785714515331619,
      "loss": 5.0911,
      "mean_token_accuracy": 0.202760012447834,
      "num_tokens": 32728133.0,
      "step": 18860
    },
    {
      "entropy": 5.609722375869751,
      "epoch": 1.4677689165531997,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047855964543514666,
      "loss": 5.1087,
      "mean_token_accuracy": 0.19485201239585875,
      "num_tokens": 32736333.0,
      "step": 18865
    },
    {
      "entropy": 5.507652664184571,
      "epoch": 1.4681579459249172,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047854783624843786,
      "loss": 5.0202,
      "mean_token_accuracy": 0.20020298063755035,
      "num_tokens": 32744954.0,
      "step": 18870
    },
    {
      "entropy": 5.536999750137329,
      "epoch": 1.468546975296635,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047853602397321453,
      "loss": 5.0978,
      "mean_token_accuracy": 0.19216215312480928,
      "num_tokens": 32753720.0,
      "step": 18875
    },
    {
      "entropy": 5.563452577590942,
      "epoch": 1.4689360046683524,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047852420860965605,
      "loss": 4.9855,
      "mean_token_accuracy": 0.19806430637836456,
      "num_tokens": 32762903.0,
      "step": 18880
    },
    {
      "entropy": 5.63886866569519,
      "epoch": 1.46932503404007,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047851239015794166,
      "loss": 5.1693,
      "mean_token_accuracy": 0.19156209975481034,
      "num_tokens": 32771367.0,
      "step": 18885
    },
    {
      "entropy": 5.592373847961426,
      "epoch": 1.4697140634117876,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047850056861825066,
      "loss": 5.0735,
      "mean_token_accuracy": 0.198457470536232,
      "num_tokens": 32779782.0,
      "step": 18890
    },
    {
      "entropy": 5.479941320419312,
      "epoch": 1.4701030927835053,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047848874399076245,
      "loss": 4.9637,
      "mean_token_accuracy": 0.19765836745500565,
      "num_tokens": 32788157.0,
      "step": 18895
    },
    {
      "entropy": 5.475327348709106,
      "epoch": 1.4704921221552227,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004784769162756563,
      "loss": 4.9495,
      "mean_token_accuracy": 0.2017843633890152,
      "num_tokens": 32796590.0,
      "step": 18900
    },
    {
      "entropy": 5.652024364471435,
      "epoch": 1.4708811515269402,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000478465085473112,
      "loss": 5.1714,
      "mean_token_accuracy": 0.18636193722486497,
      "num_tokens": 32806297.0,
      "step": 18905
    },
    {
      "entropy": 5.602399063110352,
      "epoch": 1.471270180898658,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004784532515833088,
      "loss": 5.0366,
      "mean_token_accuracy": 0.20525036603212357,
      "num_tokens": 32815370.0,
      "step": 18910
    },
    {
      "entropy": 5.517171478271484,
      "epoch": 1.4716592102703754,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047844141460642636,
      "loss": 5.071,
      "mean_token_accuracy": 0.18807777166366577,
      "num_tokens": 32824678.0,
      "step": 18915
    },
    {
      "entropy": 5.613802337646485,
      "epoch": 1.4720482396420929,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047842957454264425,
      "loss": 5.1526,
      "mean_token_accuracy": 0.18967153280973434,
      "num_tokens": 32834759.0,
      "step": 18920
    },
    {
      "entropy": 5.61039457321167,
      "epoch": 1.4724372690138106,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047841773139214213,
      "loss": 5.0834,
      "mean_token_accuracy": 0.19385363459587096,
      "num_tokens": 32843152.0,
      "step": 18925
    },
    {
      "entropy": 5.641839408874512,
      "epoch": 1.472826298385528,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004784058851550997,
      "loss": 5.277,
      "mean_token_accuracy": 0.17639375627040862,
      "num_tokens": 32852702.0,
      "step": 18930
    },
    {
      "entropy": 5.559104251861572,
      "epoch": 1.4732153277572457,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004783940358316967,
      "loss": 4.9541,
      "mean_token_accuracy": 0.2102203384041786,
      "num_tokens": 32860201.0,
      "step": 18935
    },
    {
      "entropy": 5.48592529296875,
      "epoch": 1.4736043571289632,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047838218342211296,
      "loss": 4.9297,
      "mean_token_accuracy": 0.20592311918735504,
      "num_tokens": 32868731.0,
      "step": 18940
    },
    {
      "entropy": 5.607624912261963,
      "epoch": 1.473993386500681,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047837032792652837,
      "loss": 5.1341,
      "mean_token_accuracy": 0.1935158595442772,
      "num_tokens": 32876960.0,
      "step": 18945
    },
    {
      "entropy": 5.6421088695526125,
      "epoch": 1.4743824158723984,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004783584693451226,
      "loss": 5.0962,
      "mean_token_accuracy": 0.1993286967277527,
      "num_tokens": 32886446.0,
      "step": 18950
    },
    {
      "entropy": 5.650771141052246,
      "epoch": 1.4747714452441159,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004783466076780759,
      "loss": 5.1644,
      "mean_token_accuracy": 0.18645775616168975,
      "num_tokens": 32895141.0,
      "step": 18955
    },
    {
      "entropy": 5.6041257858276365,
      "epoch": 1.4751604746158336,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004783347429255679,
      "loss": 5.0569,
      "mean_token_accuracy": 0.19232889115810395,
      "num_tokens": 32904191.0,
      "step": 18960
    },
    {
      "entropy": 5.61749415397644,
      "epoch": 1.475549503987551,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000478322875087779,
      "loss": 5.0682,
      "mean_token_accuracy": 0.19540086835622789,
      "num_tokens": 32912805.0,
      "step": 18965
    },
    {
      "entropy": 5.568792915344238,
      "epoch": 1.4759385333592685,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047831100416488906,
      "loss": 5.0285,
      "mean_token_accuracy": 0.1942672073841095,
      "num_tokens": 32921275.0,
      "step": 18970
    },
    {
      "entropy": 5.617499113082886,
      "epoch": 1.4763275627309862,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047829913015707816,
      "loss": 5.0848,
      "mean_token_accuracy": 0.19569036662578582,
      "num_tokens": 32929472.0,
      "step": 18975
    },
    {
      "entropy": 5.5695373058319095,
      "epoch": 1.4767165921027037,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047828725306452657,
      "loss": 5.0786,
      "mean_token_accuracy": 0.1941814735531807,
      "num_tokens": 32937697.0,
      "step": 18980
    },
    {
      "entropy": 5.619187831878662,
      "epoch": 1.4771056214744214,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047827537288741453,
      "loss": 5.0613,
      "mean_token_accuracy": 0.19765759408473968,
      "num_tokens": 32946516.0,
      "step": 18985
    },
    {
      "entropy": 5.555873394012451,
      "epoch": 1.4774946508461388,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004782634896259222,
      "loss": 4.9585,
      "mean_token_accuracy": 0.20265400111675264,
      "num_tokens": 32954841.0,
      "step": 18990
    },
    {
      "entropy": 5.560416650772095,
      "epoch": 1.4778836802178565,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047825160328023,
      "loss": 5.1076,
      "mean_token_accuracy": 0.19306646585464476,
      "num_tokens": 32963678.0,
      "step": 18995
    },
    {
      "entropy": 5.487839984893799,
      "epoch": 1.478272709589574,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004782397138505181,
      "loss": 4.9074,
      "mean_token_accuracy": 0.20588816702365875,
      "num_tokens": 32973163.0,
      "step": 19000
    },
    {
      "entropy": 5.657270622253418,
      "epoch": 1.4786617389612915,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047822782133696715,
      "loss": 5.1143,
      "mean_token_accuracy": 0.18870944231748582,
      "num_tokens": 32982473.0,
      "step": 19005
    },
    {
      "entropy": 5.604637861251831,
      "epoch": 1.4790507683330092,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004782159257397574,
      "loss": 4.966,
      "mean_token_accuracy": 0.21140444576740264,
      "num_tokens": 32990328.0,
      "step": 19010
    },
    {
      "entropy": 5.4945343971252445,
      "epoch": 1.4794397977047267,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047820402705906953,
      "loss": 5.043,
      "mean_token_accuracy": 0.19526928663253784,
      "num_tokens": 32998432.0,
      "step": 19015
    },
    {
      "entropy": 5.556626653671264,
      "epoch": 1.4798288270764441,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047819212529508394,
      "loss": 5.1176,
      "mean_token_accuracy": 0.20131382644176482,
      "num_tokens": 33007017.0,
      "step": 19020
    },
    {
      "entropy": 5.637542581558227,
      "epoch": 1.4802178564481618,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004781802204479812,
      "loss": 5.0357,
      "mean_token_accuracy": 0.20269588381052017,
      "num_tokens": 33015212.0,
      "step": 19025
    },
    {
      "entropy": 5.603157806396484,
      "epoch": 1.4806068858198795,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004781683125179421,
      "loss": 5.0723,
      "mean_token_accuracy": 0.193496997654438,
      "num_tokens": 33023555.0,
      "step": 19030
    },
    {
      "entropy": 5.537458419799805,
      "epoch": 1.480995915191597,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004781564015051472,
      "loss": 5.0621,
      "mean_token_accuracy": 0.19310159385204315,
      "num_tokens": 33033048.0,
      "step": 19035
    },
    {
      "entropy": 5.474830389022827,
      "epoch": 1.4813849445633145,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004781444874097772,
      "loss": 4.9446,
      "mean_token_accuracy": 0.2005469560623169,
      "num_tokens": 33041553.0,
      "step": 19040
    },
    {
      "entropy": 5.527775716781616,
      "epoch": 1.4817739739350322,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047813257023201307,
      "loss": 5.049,
      "mean_token_accuracy": 0.19772643148899077,
      "num_tokens": 33050864.0,
      "step": 19045
    },
    {
      "entropy": 5.504838657379151,
      "epoch": 1.4821630033067497,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004781206499720354,
      "loss": 4.9565,
      "mean_token_accuracy": 0.20115589946508408,
      "num_tokens": 33059700.0,
      "step": 19050
    },
    {
      "entropy": 5.470868110656738,
      "epoch": 1.4825520326784671,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047810872663002523,
      "loss": 4.9181,
      "mean_token_accuracy": 0.20432905554771424,
      "num_tokens": 33067784.0,
      "step": 19055
    },
    {
      "entropy": 5.571485185623169,
      "epoch": 1.4829410620501848,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047809680020616333,
      "loss": 5.0822,
      "mean_token_accuracy": 0.18967503309249878,
      "num_tokens": 33077174.0,
      "step": 19060
    },
    {
      "entropy": 5.603109979629517,
      "epoch": 1.4833300914219023,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047808487070063083,
      "loss": 5.0948,
      "mean_token_accuracy": 0.1976098135113716,
      "num_tokens": 33085492.0,
      "step": 19065
    },
    {
      "entropy": 5.573793220520019,
      "epoch": 1.4837191207936198,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004780729381136086,
      "loss": 5.0415,
      "mean_token_accuracy": 0.2006871670484543,
      "num_tokens": 33094388.0,
      "step": 19070
    },
    {
      "entropy": 5.479659223556519,
      "epoch": 1.4841081501653375,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004780610024452778,
      "loss": 4.887,
      "mean_token_accuracy": 0.20236905217170714,
      "num_tokens": 33101996.0,
      "step": 19075
    },
    {
      "entropy": 5.532952737808228,
      "epoch": 1.4844971795370552,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047804906369581954,
      "loss": 5.1195,
      "mean_token_accuracy": 0.19214823693037034,
      "num_tokens": 33111139.0,
      "step": 19080
    },
    {
      "entropy": 5.533510065078735,
      "epoch": 1.4848862089087727,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004780371218654148,
      "loss": 4.9485,
      "mean_token_accuracy": 0.2004660740494728,
      "num_tokens": 33119159.0,
      "step": 19085
    },
    {
      "entropy": 5.637355613708496,
      "epoch": 1.4852752382804901,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047802517695424496,
      "loss": 5.1099,
      "mean_token_accuracy": 0.19767505377531053,
      "num_tokens": 33128317.0,
      "step": 19090
    },
    {
      "entropy": 5.540646600723266,
      "epoch": 1.4856642676522078,
      "grad_norm": 1.125,
      "learning_rate": 0.0004780132289624913,
      "loss": 5.042,
      "mean_token_accuracy": 0.2068367838859558,
      "num_tokens": 33136936.0,
      "step": 19095
    },
    {
      "entropy": 5.6162255764007565,
      "epoch": 1.4860532970239253,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004780012778903349,
      "loss": 5.0963,
      "mean_token_accuracy": 0.19549003690481187,
      "num_tokens": 33145059.0,
      "step": 19100
    },
    {
      "entropy": 5.5247434139251705,
      "epoch": 1.4864423263956428,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047798932373795724,
      "loss": 4.9924,
      "mean_token_accuracy": 0.19375913441181183,
      "num_tokens": 33153683.0,
      "step": 19105
    },
    {
      "entropy": 5.636048412322998,
      "epoch": 1.4868313557673605,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047797736650553977,
      "loss": 5.1174,
      "mean_token_accuracy": 0.19583724439144135,
      "num_tokens": 33163179.0,
      "step": 19110
    },
    {
      "entropy": 5.54756383895874,
      "epoch": 1.487220385139078,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047796540619326366,
      "loss": 4.9324,
      "mean_token_accuracy": 0.20565094500780107,
      "num_tokens": 33172029.0,
      "step": 19115
    },
    {
      "entropy": 5.602601146697998,
      "epoch": 1.4876094145107956,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004779534428013107,
      "loss": 5.1249,
      "mean_token_accuracy": 0.1932734355330467,
      "num_tokens": 33181168.0,
      "step": 19120
    },
    {
      "entropy": 5.600402116775513,
      "epoch": 1.4879984438825131,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047794147632986223,
      "loss": 5.1237,
      "mean_token_accuracy": 0.18885552287101745,
      "num_tokens": 33191320.0,
      "step": 19125
    },
    {
      "entropy": 5.524675178527832,
      "epoch": 1.4883874732542308,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004779295067790999,
      "loss": 5.0436,
      "mean_token_accuracy": 0.19526771903038026,
      "num_tokens": 33199625.0,
      "step": 19130
    },
    {
      "entropy": 5.5658196926116945,
      "epoch": 1.4887765026259483,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004779175341492053,
      "loss": 5.0222,
      "mean_token_accuracy": 0.1994064062833786,
      "num_tokens": 33208145.0,
      "step": 19135
    },
    {
      "entropy": 5.599393749237061,
      "epoch": 1.4891655319976658,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004779055584403601,
      "loss": 5.0333,
      "mean_token_accuracy": 0.19652648866176606,
      "num_tokens": 33216441.0,
      "step": 19140
    },
    {
      "entropy": 5.60699725151062,
      "epoch": 1.4895545613693835,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000477893579652746,
      "loss": 5.0475,
      "mean_token_accuracy": 0.19165287166833878,
      "num_tokens": 33225612.0,
      "step": 19145
    },
    {
      "entropy": 5.532086372375488,
      "epoch": 1.489943590741101,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047788159778654475,
      "loss": 5.0237,
      "mean_token_accuracy": 0.19535877257585527,
      "num_tokens": 33234410.0,
      "step": 19150
    },
    {
      "entropy": 5.602610111236572,
      "epoch": 1.4903326201128184,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047786961284193813,
      "loss": 4.9725,
      "mean_token_accuracy": 0.2012861981987953,
      "num_tokens": 33242838.0,
      "step": 19155
    },
    {
      "entropy": 5.5760826587677,
      "epoch": 1.4907216494845361,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004778576248191081,
      "loss": 5.1145,
      "mean_token_accuracy": 0.19367948919534683,
      "num_tokens": 33251266.0,
      "step": 19160
    },
    {
      "entropy": 5.554506158828735,
      "epoch": 1.4911106788562536,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004778456337182365,
      "loss": 5.0314,
      "mean_token_accuracy": 0.20112899839878082,
      "num_tokens": 33259644.0,
      "step": 19165
    },
    {
      "entropy": 5.617341613769531,
      "epoch": 1.4914997082279713,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004778336395395052,
      "loss": 4.9987,
      "mean_token_accuracy": 0.19537637233734131,
      "num_tokens": 33267547.0,
      "step": 19170
    },
    {
      "entropy": 5.559826946258545,
      "epoch": 1.4918887375996888,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047782164228309636,
      "loss": 5.0667,
      "mean_token_accuracy": 0.19532513320446016,
      "num_tokens": 33276808.0,
      "step": 19175
    },
    {
      "entropy": 5.575062561035156,
      "epoch": 1.4922777669714065,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047780964194919193,
      "loss": 5.0914,
      "mean_token_accuracy": 0.19577698558568954,
      "num_tokens": 33285569.0,
      "step": 19180
    },
    {
      "entropy": 5.592564821243286,
      "epoch": 1.492666796343124,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047779763853797387,
      "loss": 5.0694,
      "mean_token_accuracy": 0.196518974006176,
      "num_tokens": 33295214.0,
      "step": 19185
    },
    {
      "entropy": 5.5705180168151855,
      "epoch": 1.4930558257148414,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004777856320496245,
      "loss": 4.998,
      "mean_token_accuracy": 0.20373905152082444,
      "num_tokens": 33304140.0,
      "step": 19190
    },
    {
      "entropy": 5.5535133361816404,
      "epoch": 1.493444855086559,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000477773622484326,
      "loss": 5.109,
      "mean_token_accuracy": 0.1947051018476486,
      "num_tokens": 33312879.0,
      "step": 19195
    },
    {
      "entropy": 5.549249696731567,
      "epoch": 1.4938338844582766,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004777616098422604,
      "loss": 5.0584,
      "mean_token_accuracy": 0.19798233956098557,
      "num_tokens": 33320769.0,
      "step": 19200
    },
    {
      "entropy": 5.499776363372803,
      "epoch": 1.494222913829994,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047774959412361014,
      "loss": 4.9151,
      "mean_token_accuracy": 0.20723643153905869,
      "num_tokens": 33328911.0,
      "step": 19205
    },
    {
      "entropy": 5.565856313705444,
      "epoch": 1.4946119432017118,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004777375753285575,
      "loss": 5.0079,
      "mean_token_accuracy": 0.2016614630818367,
      "num_tokens": 33337607.0,
      "step": 19210
    },
    {
      "entropy": 5.584451198577881,
      "epoch": 1.4950009725734292,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047772555345728486,
      "loss": 5.0525,
      "mean_token_accuracy": 0.19763896018266677,
      "num_tokens": 33346663.0,
      "step": 19215
    },
    {
      "entropy": 5.654525375366211,
      "epoch": 1.495390001945147,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004777135285099746,
      "loss": 5.1009,
      "mean_token_accuracy": 0.1854400858283043,
      "num_tokens": 33355501.0,
      "step": 19220
    },
    {
      "entropy": 5.538935947418213,
      "epoch": 1.4957790313168644,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004777015004868092,
      "loss": 4.9837,
      "mean_token_accuracy": 0.1891601040959358,
      "num_tokens": 33363497.0,
      "step": 19225
    },
    {
      "entropy": 5.521978378295898,
      "epoch": 1.496168060688582,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004776894693879712,
      "loss": 4.9354,
      "mean_token_accuracy": 0.21100567132234574,
      "num_tokens": 33371349.0,
      "step": 19230
    },
    {
      "entropy": 5.516081619262695,
      "epoch": 1.4965570900602996,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004776774352136431,
      "loss": 4.9046,
      "mean_token_accuracy": 0.20327480882406235,
      "num_tokens": 33379889.0,
      "step": 19235
    },
    {
      "entropy": 5.604160213470459,
      "epoch": 1.496946119432017,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047766539796400756,
      "loss": 5.0702,
      "mean_token_accuracy": 0.19519925266504287,
      "num_tokens": 33388452.0,
      "step": 19240
    },
    {
      "entropy": 5.522110843658448,
      "epoch": 1.4973351488037348,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004776533576392471,
      "loss": 4.998,
      "mean_token_accuracy": 0.1975472465157509,
      "num_tokens": 33397598.0,
      "step": 19245
    },
    {
      "entropy": 5.601174545288086,
      "epoch": 1.4977241781754522,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004776413142395445,
      "loss": 5.062,
      "mean_token_accuracy": 0.19426544904708862,
      "num_tokens": 33406509.0,
      "step": 19250
    },
    {
      "entropy": 5.545873022079467,
      "epoch": 1.4981132075471697,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047762926776508244,
      "loss": 5.0578,
      "mean_token_accuracy": 0.1902050644159317,
      "num_tokens": 33415129.0,
      "step": 19255
    },
    {
      "entropy": 5.473169660568237,
      "epoch": 1.4985022369188874,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047761721821604387,
      "loss": 4.8993,
      "mean_token_accuracy": 0.20065039694309234,
      "num_tokens": 33423588.0,
      "step": 19260
    },
    {
      "entropy": 5.603838300704956,
      "epoch": 1.4988912662906049,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004776051655926115,
      "loss": 5.1019,
      "mean_token_accuracy": 0.1859123796224594,
      "num_tokens": 33431598.0,
      "step": 19265
    },
    {
      "entropy": 5.676783895492553,
      "epoch": 1.4992802956623226,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047759310989496813,
      "loss": 5.1635,
      "mean_token_accuracy": 0.19246651232242584,
      "num_tokens": 33441139.0,
      "step": 19270
    },
    {
      "entropy": 5.609130191802978,
      "epoch": 1.49966932503404,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004775810511232969,
      "loss": 5.0112,
      "mean_token_accuracy": 0.19229902774095536,
      "num_tokens": 33449558.0,
      "step": 19275
    },
    {
      "entropy": 5.578229808807373,
      "epoch": 1.5000583544057577,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047756898927778056,
      "loss": 5.0867,
      "mean_token_accuracy": 0.19262377470731734,
      "num_tokens": 33457429.0,
      "step": 19280
    },
    {
      "entropy": 5.52520980834961,
      "epoch": 1.5004473837774752,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047755692435860227,
      "loss": 4.9563,
      "mean_token_accuracy": 0.2032209262251854,
      "num_tokens": 33465781.0,
      "step": 19285
    },
    {
      "entropy": 5.616564226150513,
      "epoch": 1.5008364131491927,
      "grad_norm": 1.078125,
      "learning_rate": 0.000477544856365945,
      "loss": 5.0227,
      "mean_token_accuracy": 0.1925642654299736,
      "num_tokens": 33473865.0,
      "step": 19290
    },
    {
      "entropy": 5.572822380065918,
      "epoch": 1.5012254425209104,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047753278529999205,
      "loss": 5.0722,
      "mean_token_accuracy": 0.19841428399085997,
      "num_tokens": 33482497.0,
      "step": 19295
    },
    {
      "entropy": 5.560092973709106,
      "epoch": 1.5016144718926279,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047752071116092637,
      "loss": 5.0794,
      "mean_token_accuracy": 0.1939171776175499,
      "num_tokens": 33490933.0,
      "step": 19300
    },
    {
      "entropy": 5.577777767181397,
      "epoch": 1.5020035012643453,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004775086339489312,
      "loss": 5.0705,
      "mean_token_accuracy": 0.1964462071657181,
      "num_tokens": 33500192.0,
      "step": 19305
    },
    {
      "entropy": 5.597309255599976,
      "epoch": 1.502392530636063,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004774965536641899,
      "loss": 5.0548,
      "mean_token_accuracy": 0.19409905821084977,
      "num_tokens": 33508543.0,
      "step": 19310
    },
    {
      "entropy": 5.545638799667358,
      "epoch": 1.5027815600077807,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047748447030688575,
      "loss": 5.0667,
      "mean_token_accuracy": 0.19157154262065887,
      "num_tokens": 33517310.0,
      "step": 19315
    },
    {
      "entropy": 5.589897537231446,
      "epoch": 1.503170589379498,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047747238387720194,
      "loss": 5.1175,
      "mean_token_accuracy": 0.18972918540239334,
      "num_tokens": 33526274.0,
      "step": 19320
    },
    {
      "entropy": 5.617796325683594,
      "epoch": 1.5035596187512157,
      "grad_norm": 1.125,
      "learning_rate": 0.000477460294375322,
      "loss": 5.123,
      "mean_token_accuracy": 0.19104223400354386,
      "num_tokens": 33534438.0,
      "step": 19325
    },
    {
      "entropy": 5.576346826553345,
      "epoch": 1.5039486481229334,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047744820180142935,
      "loss": 5.0241,
      "mean_token_accuracy": 0.1926095813512802,
      "num_tokens": 33542943.0,
      "step": 19330
    },
    {
      "entropy": 5.567524194717407,
      "epoch": 1.5043376774946509,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047743610615570746,
      "loss": 5.0392,
      "mean_token_accuracy": 0.1967159017920494,
      "num_tokens": 33550567.0,
      "step": 19335
    },
    {
      "entropy": 5.584063005447388,
      "epoch": 1.5047267068663683,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047742400743833993,
      "loss": 5.122,
      "mean_token_accuracy": 0.19178835302591324,
      "num_tokens": 33559851.0,
      "step": 19340
    },
    {
      "entropy": 5.602731800079345,
      "epoch": 1.505115736238086,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004774119056495102,
      "loss": 5.063,
      "mean_token_accuracy": 0.19006472826004028,
      "num_tokens": 33568645.0,
      "step": 19345
    },
    {
      "entropy": 5.554150485992432,
      "epoch": 1.5055047656098035,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047739980078940206,
      "loss": 5.0225,
      "mean_token_accuracy": 0.20502635091543198,
      "num_tokens": 33577331.0,
      "step": 19350
    },
    {
      "entropy": 5.486583042144775,
      "epoch": 1.505893794981521,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047738769285819894,
      "loss": 4.8751,
      "mean_token_accuracy": 0.20783884823322296,
      "num_tokens": 33586196.0,
      "step": 19355
    },
    {
      "entropy": 5.527578926086425,
      "epoch": 1.5062828243532387,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004773755818560849,
      "loss": 5.1015,
      "mean_token_accuracy": 0.1951591670513153,
      "num_tokens": 33595200.0,
      "step": 19360
    },
    {
      "entropy": 5.543863582611084,
      "epoch": 1.5066718537249564,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004773634677832434,
      "loss": 4.9857,
      "mean_token_accuracy": 0.20338316410779952,
      "num_tokens": 33604130.0,
      "step": 19365
    },
    {
      "entropy": 5.606501150131225,
      "epoch": 1.5070608830966739,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004773513506398584,
      "loss": 5.1717,
      "mean_token_accuracy": 0.18432357907295227,
      "num_tokens": 33613254.0,
      "step": 19370
    },
    {
      "entropy": 5.578696537017822,
      "epoch": 1.5074499124683913,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004773392304261137,
      "loss": 5.0343,
      "mean_token_accuracy": 0.19818809926509856,
      "num_tokens": 33621395.0,
      "step": 19375
    },
    {
      "entropy": 5.562988328933716,
      "epoch": 1.507838941840109,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004773271071421933,
      "loss": 4.9828,
      "mean_token_accuracy": 0.19403541535139085,
      "num_tokens": 33630293.0,
      "step": 19380
    },
    {
      "entropy": 5.614288282394409,
      "epoch": 1.5082279712118265,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047731498078828105,
      "loss": 5.0956,
      "mean_token_accuracy": 0.19194564670324327,
      "num_tokens": 33638224.0,
      "step": 19385
    },
    {
      "entropy": 5.647917461395264,
      "epoch": 1.508617000583544,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000477302851364561,
      "loss": 5.149,
      "mean_token_accuracy": 0.1849290817975998,
      "num_tokens": 33646974.0,
      "step": 19390
    },
    {
      "entropy": 5.602269124984741,
      "epoch": 1.5090060299552617,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047729071887121717,
      "loss": 5.112,
      "mean_token_accuracy": 0.1931400179862976,
      "num_tokens": 33656171.0,
      "step": 19395
    },
    {
      "entropy": 5.569032764434814,
      "epoch": 1.5093950593269791,
      "grad_norm": 1.125,
      "learning_rate": 0.0004772785833084337,
      "loss": 5.1024,
      "mean_token_accuracy": 0.1861686572432518,
      "num_tokens": 33664605.0,
      "step": 19400
    },
    {
      "entropy": 5.626187324523926,
      "epoch": 1.5097840886986966,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004772664446763946,
      "loss": 5.0828,
      "mean_token_accuracy": 0.19436684548854827,
      "num_tokens": 33673128.0,
      "step": 19405
    },
    {
      "entropy": 5.56131272315979,
      "epoch": 1.5101731180704143,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004772543029752842,
      "loss": 5.0484,
      "mean_token_accuracy": 0.19543803483247757,
      "num_tokens": 33681358.0,
      "step": 19410
    },
    {
      "entropy": 5.502350664138794,
      "epoch": 1.510562147442132,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047724215820528666,
      "loss": 4.914,
      "mean_token_accuracy": 0.208049413561821,
      "num_tokens": 33690282.0,
      "step": 19415
    },
    {
      "entropy": 5.52520899772644,
      "epoch": 1.5109511768138495,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004772300103665863,
      "loss": 4.9859,
      "mean_token_accuracy": 0.19778398275375367,
      "num_tokens": 33698294.0,
      "step": 19420
    },
    {
      "entropy": 5.520120096206665,
      "epoch": 1.511340206185567,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047721785945936733,
      "loss": 4.9624,
      "mean_token_accuracy": 0.20365458875894546,
      "num_tokens": 33706552.0,
      "step": 19425
    },
    {
      "entropy": 5.585129308700561,
      "epoch": 1.5117292355572847,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004772057054838143,
      "loss": 5.0206,
      "mean_token_accuracy": 0.1980569466948509,
      "num_tokens": 33714704.0,
      "step": 19430
    },
    {
      "entropy": 5.6245622634887695,
      "epoch": 1.5121182649290021,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047719354844011146,
      "loss": 5.0733,
      "mean_token_accuracy": 0.19992723166942597,
      "num_tokens": 33723823.0,
      "step": 19435
    },
    {
      "entropy": 5.671552944183349,
      "epoch": 1.5125072943007196,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004771813883284434,
      "loss": 5.1359,
      "mean_token_accuracy": 0.19441195726394653,
      "num_tokens": 33732616.0,
      "step": 19440
    },
    {
      "entropy": 5.527145767211914,
      "epoch": 1.5128963236724373,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047716922514899455,
      "loss": 4.9974,
      "mean_token_accuracy": 0.20298608690500258,
      "num_tokens": 33741542.0,
      "step": 19445
    },
    {
      "entropy": 5.561437606811523,
      "epoch": 1.5132853530441548,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047715705890194953,
      "loss": 5.0465,
      "mean_token_accuracy": 0.199357508122921,
      "num_tokens": 33750651.0,
      "step": 19450
    },
    {
      "entropy": 5.508872032165527,
      "epoch": 1.5136743824158723,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047714488958749285,
      "loss": 5.1055,
      "mean_token_accuracy": 0.19082214683294296,
      "num_tokens": 33760741.0,
      "step": 19455
    },
    {
      "entropy": 5.639386177062988,
      "epoch": 1.51406341178759,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047713271720580924,
      "loss": 5.092,
      "mean_token_accuracy": 0.19224275648593903,
      "num_tokens": 33769541.0,
      "step": 19460
    },
    {
      "entropy": 5.604441452026367,
      "epoch": 1.5144524411593077,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004771205417570834,
      "loss": 5.0973,
      "mean_token_accuracy": 0.19790313094854356,
      "num_tokens": 33778684.0,
      "step": 19465
    },
    {
      "entropy": 5.59993314743042,
      "epoch": 1.5148414705310251,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047710836324150004,
      "loss": 5.0877,
      "mean_token_accuracy": 0.19678373336791993,
      "num_tokens": 33787782.0,
      "step": 19470
    },
    {
      "entropy": 5.5572850704193115,
      "epoch": 1.5152304999027426,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000477096181659244,
      "loss": 5.0302,
      "mean_token_accuracy": 0.19933279156684874,
      "num_tokens": 33796194.0,
      "step": 19475
    },
    {
      "entropy": 5.602280521392823,
      "epoch": 1.5156195292744603,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004770839970105,
      "loss": 5.0406,
      "mean_token_accuracy": 0.19467580318450928,
      "num_tokens": 33804615.0,
      "step": 19480
    },
    {
      "entropy": 5.642336177825928,
      "epoch": 1.5160085586461778,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047707180929545295,
      "loss": 5.1195,
      "mean_token_accuracy": 0.19626032710075378,
      "num_tokens": 33812937.0,
      "step": 19485
    },
    {
      "entropy": 5.57250452041626,
      "epoch": 1.5163975880178953,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047705961851428786,
      "loss": 5.0074,
      "mean_token_accuracy": 0.193632872402668,
      "num_tokens": 33821454.0,
      "step": 19490
    },
    {
      "entropy": 5.600674343109131,
      "epoch": 1.516786617389613,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047704742466718973,
      "loss": 5.077,
      "mean_token_accuracy": 0.19773407578468322,
      "num_tokens": 33829736.0,
      "step": 19495
    },
    {
      "entropy": 5.620756959915161,
      "epoch": 1.5171756467613304,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047703522775434354,
      "loss": 5.0815,
      "mean_token_accuracy": 0.19734029173851014,
      "num_tokens": 33838403.0,
      "step": 19500
    },
    {
      "entropy": 5.591812181472778,
      "epoch": 1.517564676133048,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047702302777593425,
      "loss": 5.0551,
      "mean_token_accuracy": 0.20051679760217667,
      "num_tokens": 33847149.0,
      "step": 19505
    },
    {
      "entropy": 5.561823320388794,
      "epoch": 1.5179537055047656,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047701082473214715,
      "loss": 5.0398,
      "mean_token_accuracy": 0.19589455872774125,
      "num_tokens": 33855924.0,
      "step": 19510
    },
    {
      "entropy": 5.542061853408813,
      "epoch": 1.5183427348764833,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047699861862316725,
      "loss": 5.0356,
      "mean_token_accuracy": 0.19539591521024705,
      "num_tokens": 33864314.0,
      "step": 19515
    },
    {
      "entropy": 5.5685662746429445,
      "epoch": 1.5187317642482008,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004769864094491798,
      "loss": 4.9984,
      "mean_token_accuracy": 0.19744833558797836,
      "num_tokens": 33873468.0,
      "step": 19520
    },
    {
      "entropy": 5.57979531288147,
      "epoch": 1.5191207936199183,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004769741972103702,
      "loss": 5.0502,
      "mean_token_accuracy": 0.1949549213051796,
      "num_tokens": 33881627.0,
      "step": 19525
    },
    {
      "entropy": 5.5815106391906735,
      "epoch": 1.519509822991636,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047696198190692353,
      "loss": 5.052,
      "mean_token_accuracy": 0.19259062260389329,
      "num_tokens": 33889839.0,
      "step": 19530
    },
    {
      "entropy": 5.5627600193023685,
      "epoch": 1.5198988523633534,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004769497635390253,
      "loss": 4.9773,
      "mean_token_accuracy": 0.199481500685215,
      "num_tokens": 33898370.0,
      "step": 19535
    },
    {
      "entropy": 5.600575876235962,
      "epoch": 1.520287881735071,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004769375421068608,
      "loss": 5.0825,
      "mean_token_accuracy": 0.1904037356376648,
      "num_tokens": 33907419.0,
      "step": 19540
    },
    {
      "entropy": 5.602925157546997,
      "epoch": 1.5206769111067886,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047692531761061555,
      "loss": 5.0575,
      "mean_token_accuracy": 0.189690899848938,
      "num_tokens": 33915957.0,
      "step": 19545
    },
    {
      "entropy": 5.586113166809082,
      "epoch": 1.521065940478506,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000476913090050475,
      "loss": 5.0321,
      "mean_token_accuracy": 0.1981988102197647,
      "num_tokens": 33924012.0,
      "step": 19550
    },
    {
      "entropy": 5.548090553283691,
      "epoch": 1.5214549698502235,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047690085942662464,
      "loss": 5.08,
      "mean_token_accuracy": 0.19278478473424912,
      "num_tokens": 33932652.0,
      "step": 19555
    },
    {
      "entropy": 5.579479598999024,
      "epoch": 1.5218439992219412,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047688862573925015,
      "loss": 5.0066,
      "mean_token_accuracy": 0.1998341754078865,
      "num_tokens": 33941208.0,
      "step": 19560
    },
    {
      "entropy": 5.581429529190063,
      "epoch": 1.522233028593659,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047687638898853707,
      "loss": 5.0178,
      "mean_token_accuracy": 0.19614847302436828,
      "num_tokens": 33950435.0,
      "step": 19565
    },
    {
      "entropy": 5.564855766296387,
      "epoch": 1.5226220579653764,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004768641491746711,
      "loss": 4.9976,
      "mean_token_accuracy": 0.2005571559071541,
      "num_tokens": 33959659.0,
      "step": 19570
    },
    {
      "entropy": 5.572706460952759,
      "epoch": 1.523011087337094,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000476851906297838,
      "loss": 5.0264,
      "mean_token_accuracy": 0.19316282421350478,
      "num_tokens": 33969105.0,
      "step": 19575
    },
    {
      "entropy": 5.575748491287231,
      "epoch": 1.5234001167088116,
      "grad_norm": 1.125,
      "learning_rate": 0.00047683966035822346,
      "loss": 5.113,
      "mean_token_accuracy": 0.19172277599573134,
      "num_tokens": 33977314.0,
      "step": 19580
    },
    {
      "entropy": 5.53893404006958,
      "epoch": 1.523789146080529,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047682741135601336,
      "loss": 5.0535,
      "mean_token_accuracy": 0.19928955286741257,
      "num_tokens": 33985696.0,
      "step": 19585
    },
    {
      "entropy": 5.466497898101807,
      "epoch": 1.5241781754522465,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047681515929139356,
      "loss": 4.8326,
      "mean_token_accuracy": 0.2092184156179428,
      "num_tokens": 33994539.0,
      "step": 19590
    },
    {
      "entropy": 5.538030958175659,
      "epoch": 1.5245672048239642,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047680290416454995,
      "loss": 5.1073,
      "mean_token_accuracy": 0.18450858443975449,
      "num_tokens": 34003101.0,
      "step": 19595
    },
    {
      "entropy": 5.624976634979248,
      "epoch": 1.524956234195682,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004767906459756684,
      "loss": 5.0497,
      "mean_token_accuracy": 0.19662797898054124,
      "num_tokens": 34011344.0,
      "step": 19600
    },
    {
      "entropy": 5.6196732997894285,
      "epoch": 1.5253452635673992,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047677838472493504,
      "loss": 5.0544,
      "mean_token_accuracy": 0.19773033559322356,
      "num_tokens": 34019628.0,
      "step": 19605
    },
    {
      "entropy": 5.5861822128295895,
      "epoch": 1.5257342929391169,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004767661204125358,
      "loss": 5.0182,
      "mean_token_accuracy": 0.19757231771945954,
      "num_tokens": 34027709.0,
      "step": 19610
    },
    {
      "entropy": 5.5696399211883545,
      "epoch": 1.5261233223108346,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004767538530386569,
      "loss": 5.0877,
      "mean_token_accuracy": 0.1964617297053337,
      "num_tokens": 34036412.0,
      "step": 19615
    },
    {
      "entropy": 5.5558857917785645,
      "epoch": 1.526512351682552,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047674158260348437,
      "loss": 5.0202,
      "mean_token_accuracy": 0.20062774568796157,
      "num_tokens": 34044842.0,
      "step": 19620
    },
    {
      "entropy": 5.498097991943359,
      "epoch": 1.5269013810542695,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047672930910720436,
      "loss": 4.9709,
      "mean_token_accuracy": 0.1978704497218132,
      "num_tokens": 34053286.0,
      "step": 19625
    },
    {
      "entropy": 5.548878860473633,
      "epoch": 1.5272904104259872,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047671703255000326,
      "loss": 5.015,
      "mean_token_accuracy": 0.20002574622631072,
      "num_tokens": 34062059.0,
      "step": 19630
    },
    {
      "entropy": 5.6085916519165036,
      "epoch": 1.5276794397977047,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004767047529320673,
      "loss": 5.0791,
      "mean_token_accuracy": 0.1960703343153,
      "num_tokens": 34070800.0,
      "step": 19635
    },
    {
      "entropy": 5.597479581832886,
      "epoch": 1.5280684691694222,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047669247025358257,
      "loss": 5.1267,
      "mean_token_accuracy": 0.19158907383680343,
      "num_tokens": 34079935.0,
      "step": 19640
    },
    {
      "entropy": 5.5205841064453125,
      "epoch": 1.5284574985411399,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047668018451473584,
      "loss": 4.9959,
      "mean_token_accuracy": 0.20108865946531296,
      "num_tokens": 34088614.0,
      "step": 19645
    },
    {
      "entropy": 5.588642644882202,
      "epoch": 1.5288465279128576,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004766678957157132,
      "loss": 5.0201,
      "mean_token_accuracy": 0.20080013275146485,
      "num_tokens": 34096899.0,
      "step": 19650
    },
    {
      "entropy": 5.63423523902893,
      "epoch": 1.5292355572845748,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004766556038567013,
      "loss": 5.1219,
      "mean_token_accuracy": 0.19054164588451386,
      "num_tokens": 34105936.0,
      "step": 19655
    },
    {
      "entropy": 5.555719232559204,
      "epoch": 1.5296245866562925,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004766433089378865,
      "loss": 4.9821,
      "mean_token_accuracy": 0.19837913364171983,
      "num_tokens": 34113675.0,
      "step": 19660
    },
    {
      "entropy": 5.555165863037109,
      "epoch": 1.5300136160280102,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047663101095945544,
      "loss": 4.9013,
      "mean_token_accuracy": 0.20620633214712142,
      "num_tokens": 34121944.0,
      "step": 19665
    },
    {
      "entropy": 5.486474895477295,
      "epoch": 1.5304026453997277,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047661870992159476,
      "loss": 5.0275,
      "mean_token_accuracy": 0.2022565260529518,
      "num_tokens": 34130403.0,
      "step": 19670
    },
    {
      "entropy": 5.52857813835144,
      "epoch": 1.5307916747714452,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047660640582449106,
      "loss": 5.0989,
      "mean_token_accuracy": 0.20228696018457412,
      "num_tokens": 34139853.0,
      "step": 19675
    },
    {
      "entropy": 5.68403992652893,
      "epoch": 1.5311807041431629,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047659409866833104,
      "loss": 5.1485,
      "mean_token_accuracy": 0.20078216791152953,
      "num_tokens": 34147833.0,
      "step": 19680
    },
    {
      "entropy": 5.614011192321778,
      "epoch": 1.5315697335148803,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004765817884533014,
      "loss": 5.0334,
      "mean_token_accuracy": 0.2000156283378601,
      "num_tokens": 34156248.0,
      "step": 19685
    },
    {
      "entropy": 5.468405246734619,
      "epoch": 1.5319587628865978,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000476569475179589,
      "loss": 4.9264,
      "mean_token_accuracy": 0.2064286708831787,
      "num_tokens": 34164523.0,
      "step": 19690
    },
    {
      "entropy": 5.58081088066101,
      "epoch": 1.5323477922583155,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047655715884738074,
      "loss": 5.1846,
      "mean_token_accuracy": 0.19613184183835983,
      "num_tokens": 34174084.0,
      "step": 19695
    },
    {
      "entropy": 5.61304931640625,
      "epoch": 1.5327368216300332,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004765448394568632,
      "loss": 5.0678,
      "mean_token_accuracy": 0.19733116328716277,
      "num_tokens": 34182551.0,
      "step": 19700
    },
    {
      "entropy": 5.533194303512573,
      "epoch": 1.5331258510017505,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004765325170082237,
      "loss": 4.9298,
      "mean_token_accuracy": 0.20900073349475862,
      "num_tokens": 34190899.0,
      "step": 19705
    },
    {
      "entropy": 5.604460191726685,
      "epoch": 1.5335148803734682,
      "grad_norm": 1.0625,
      "learning_rate": 0.000476520191501649,
      "loss": 5.1433,
      "mean_token_accuracy": 0.19249052554368973,
      "num_tokens": 34200394.0,
      "step": 19710
    },
    {
      "entropy": 5.6358174800872805,
      "epoch": 1.5339039097451859,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047650786293732607,
      "loss": 5.1109,
      "mean_token_accuracy": 0.18892403692007065,
      "num_tokens": 34209773.0,
      "step": 19715
    },
    {
      "entropy": 5.611033391952515,
      "epoch": 1.5342929391169033,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004764955313154421,
      "loss": 5.0664,
      "mean_token_accuracy": 0.19133815169334412,
      "num_tokens": 34218487.0,
      "step": 19720
    },
    {
      "entropy": 5.616184902191162,
      "epoch": 1.5346819684886208,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047648319663618415,
      "loss": 5.108,
      "mean_token_accuracy": 0.19806885868310928,
      "num_tokens": 34227309.0,
      "step": 19725
    },
    {
      "entropy": 5.606688976287842,
      "epoch": 1.5350709978603385,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047647085889973936,
      "loss": 5.057,
      "mean_token_accuracy": 0.19754924476146699,
      "num_tokens": 34235877.0,
      "step": 19730
    },
    {
      "entropy": 5.611206674575806,
      "epoch": 1.535460027232056,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047645851810629503,
      "loss": 5.017,
      "mean_token_accuracy": 0.19970910400152206,
      "num_tokens": 34244073.0,
      "step": 19735
    },
    {
      "entropy": 5.578699398040771,
      "epoch": 1.5358490566037735,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047644617425603825,
      "loss": 5.0567,
      "mean_token_accuracy": 0.19914185106754304,
      "num_tokens": 34252255.0,
      "step": 19740
    },
    {
      "entropy": 5.571052312850952,
      "epoch": 1.5362380859754912,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004764338273491565,
      "loss": 5.04,
      "mean_token_accuracy": 0.2035927250981331,
      "num_tokens": 34261063.0,
      "step": 19745
    },
    {
      "entropy": 5.59115948677063,
      "epoch": 1.5366271153472089,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047642147738583695,
      "loss": 5.0887,
      "mean_token_accuracy": 0.1943797990679741,
      "num_tokens": 34269072.0,
      "step": 19750
    },
    {
      "entropy": 5.630272436141968,
      "epoch": 1.5370161447189261,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004764091243662671,
      "loss": 5.0653,
      "mean_token_accuracy": 0.18938371241092683,
      "num_tokens": 34277845.0,
      "step": 19755
    },
    {
      "entropy": 5.472293710708618,
      "epoch": 1.5374051740906438,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047639676829063437,
      "loss": 4.9064,
      "mean_token_accuracy": 0.2081210732460022,
      "num_tokens": 34285831.0,
      "step": 19760
    },
    {
      "entropy": 5.604944896697998,
      "epoch": 1.5377942034623615,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047638440915912623,
      "loss": 5.1149,
      "mean_token_accuracy": 0.1914842903614044,
      "num_tokens": 34294095.0,
      "step": 19765
    },
    {
      "entropy": 5.651477718353272,
      "epoch": 1.538183232834079,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004763720469719303,
      "loss": 5.0538,
      "mean_token_accuracy": 0.19778647571802138,
      "num_tokens": 34301570.0,
      "step": 19770
    },
    {
      "entropy": 5.6162529468536375,
      "epoch": 1.5385722622057965,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000476359681729234,
      "loss": 5.0787,
      "mean_token_accuracy": 0.193239164352417,
      "num_tokens": 34310119.0,
      "step": 19775
    },
    {
      "entropy": 5.574469375610351,
      "epoch": 1.5389612915775142,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000476347313431225,
      "loss": 5.0199,
      "mean_token_accuracy": 0.2011852040886879,
      "num_tokens": 34319622.0,
      "step": 19780
    },
    {
      "entropy": 5.542970180511475,
      "epoch": 1.5393503209492316,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047633494207809096,
      "loss": 5.0253,
      "mean_token_accuracy": 0.2008110210299492,
      "num_tokens": 34328376.0,
      "step": 19785
    },
    {
      "entropy": 5.528694200515747,
      "epoch": 1.539739350320949,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047632256767001977,
      "loss": 5.0646,
      "mean_token_accuracy": 0.19802765250205995,
      "num_tokens": 34337430.0,
      "step": 19790
    },
    {
      "entropy": 5.618781280517578,
      "epoch": 1.5401283796926668,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000476310190207199,
      "loss": 5.0676,
      "mean_token_accuracy": 0.19930807799100875,
      "num_tokens": 34345639.0,
      "step": 19795
    },
    {
      "entropy": 5.6438757419586185,
      "epoch": 1.5405174090643845,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047629780968981653,
      "loss": 5.0537,
      "mean_token_accuracy": 0.19548204243183137,
      "num_tokens": 34354083.0,
      "step": 19800
    },
    {
      "entropy": 5.535804414749146,
      "epoch": 1.540906438436102,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047628542611806007,
      "loss": 5.0411,
      "mean_token_accuracy": 0.19648046642541886,
      "num_tokens": 34363882.0,
      "step": 19805
    },
    {
      "entropy": 5.5137416362762455,
      "epoch": 1.5412954678078195,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047627303949211773,
      "loss": 4.9815,
      "mean_token_accuracy": 0.20216081887483597,
      "num_tokens": 34372487.0,
      "step": 19810
    },
    {
      "entropy": 5.539914274215699,
      "epoch": 1.5416844971795371,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004762606498121774,
      "loss": 5.0156,
      "mean_token_accuracy": 0.1989564374089241,
      "num_tokens": 34380729.0,
      "step": 19815
    },
    {
      "entropy": 5.570079326629639,
      "epoch": 1.5420735265512546,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000476248257078427,
      "loss": 5.0507,
      "mean_token_accuracy": 0.19525644332170486,
      "num_tokens": 34390121.0,
      "step": 19820
    },
    {
      "entropy": 5.566248750686645,
      "epoch": 1.542462555922972,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004762358612910547,
      "loss": 5.0282,
      "mean_token_accuracy": 0.1983843356370926,
      "num_tokens": 34398460.0,
      "step": 19825
    },
    {
      "entropy": 5.565110397338867,
      "epoch": 1.5428515852946898,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004762234624502484,
      "loss": 5.0246,
      "mean_token_accuracy": 0.19438279718160628,
      "num_tokens": 34407406.0,
      "step": 19830
    },
    {
      "entropy": 5.568144464492798,
      "epoch": 1.5432406146664073,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047621106055619644,
      "loss": 4.9752,
      "mean_token_accuracy": 0.19961445033550262,
      "num_tokens": 34416588.0,
      "step": 19835
    },
    {
      "entropy": 5.5808424949646,
      "epoch": 1.5436296440381247,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047619865560908687,
      "loss": 5.0597,
      "mean_token_accuracy": 0.1943053498864174,
      "num_tokens": 34425266.0,
      "step": 19840
    },
    {
      "entropy": 5.5848808765411375,
      "epoch": 1.5440186734098424,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004761862476091079,
      "loss": 5.0046,
      "mean_token_accuracy": 0.1964513748884201,
      "num_tokens": 34433790.0,
      "step": 19845
    },
    {
      "entropy": 5.639099073410034,
      "epoch": 1.5444077027815601,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047617383655644785,
      "loss": 5.1329,
      "mean_token_accuracy": 0.1871665522456169,
      "num_tokens": 34442733.0,
      "step": 19850
    },
    {
      "entropy": 5.642767000198364,
      "epoch": 1.5447967321532776,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004761614224512951,
      "loss": 5.0282,
      "mean_token_accuracy": 0.19980663061141968,
      "num_tokens": 34451756.0,
      "step": 19855
    },
    {
      "entropy": 5.572693824768066,
      "epoch": 1.545185761524995,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004761490052938379,
      "loss": 5.0433,
      "mean_token_accuracy": 0.19933335930109025,
      "num_tokens": 34460003.0,
      "step": 19860
    },
    {
      "entropy": 5.5568413734436035,
      "epoch": 1.5455747908967128,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004761365850842646,
      "loss": 5.0533,
      "mean_token_accuracy": 0.20289468467235566,
      "num_tokens": 34468485.0,
      "step": 19865
    },
    {
      "entropy": 5.594424772262573,
      "epoch": 1.5459638202684303,
      "grad_norm": 1.125,
      "learning_rate": 0.0004761241618227639,
      "loss": 4.9807,
      "mean_token_accuracy": 0.19829490929841995,
      "num_tokens": 34477221.0,
      "step": 19870
    },
    {
      "entropy": 5.588550138473511,
      "epoch": 1.5463528496401477,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047611173550952414,
      "loss": 5.1263,
      "mean_token_accuracy": 0.18963509649038315,
      "num_tokens": 34486339.0,
      "step": 19875
    },
    {
      "entropy": 5.579809761047363,
      "epoch": 1.5467418790118654,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047609930614473387,
      "loss": 4.9861,
      "mean_token_accuracy": 0.20371175557374954,
      "num_tokens": 34494595.0,
      "step": 19880
    },
    {
      "entropy": 5.5361815929412845,
      "epoch": 1.547130908383583,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047608687372858175,
      "loss": 4.9579,
      "mean_token_accuracy": 0.19888206869363784,
      "num_tokens": 34502702.0,
      "step": 19885
    },
    {
      "entropy": 5.5999082088470455,
      "epoch": 1.5475199377553004,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047607443826125633,
      "loss": 5.0995,
      "mean_token_accuracy": 0.19613149315118789,
      "num_tokens": 34510828.0,
      "step": 19890
    },
    {
      "entropy": 5.654301452636719,
      "epoch": 1.547908967127018,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047606199974294637,
      "loss": 5.1716,
      "mean_token_accuracy": 0.19079936146736146,
      "num_tokens": 34520288.0,
      "step": 19895
    },
    {
      "entropy": 5.633886766433716,
      "epoch": 1.5482979964987358,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004760495581738406,
      "loss": 5.0259,
      "mean_token_accuracy": 0.197757488489151,
      "num_tokens": 34528178.0,
      "step": 19900
    },
    {
      "entropy": 5.599398136138916,
      "epoch": 1.5486870258704533,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004760371135541278,
      "loss": 5.076,
      "mean_token_accuracy": 0.18469796031713487,
      "num_tokens": 34536915.0,
      "step": 19905
    },
    {
      "entropy": 5.605526494979858,
      "epoch": 1.5490760552421707,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004760246658839967,
      "loss": 5.0518,
      "mean_token_accuracy": 0.1931252047419548,
      "num_tokens": 34545009.0,
      "step": 19910
    },
    {
      "entropy": 5.644918823242188,
      "epoch": 1.5494650846138884,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004760122151636364,
      "loss": 5.1299,
      "mean_token_accuracy": 0.19367597997188568,
      "num_tokens": 34553936.0,
      "step": 19915
    },
    {
      "entropy": 5.615521955490112,
      "epoch": 1.549854113985606,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004759997613932356,
      "loss": 5.0097,
      "mean_token_accuracy": 0.20323496311903,
      "num_tokens": 34563068.0,
      "step": 19920
    },
    {
      "entropy": 5.632466459274292,
      "epoch": 1.5502431433573234,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047598730457298335,
      "loss": 5.1467,
      "mean_token_accuracy": 0.18695403039455413,
      "num_tokens": 34571787.0,
      "step": 19925
    },
    {
      "entropy": 5.644014120101929,
      "epoch": 1.550632172729041,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047597484470306866,
      "loss": 5.0498,
      "mean_token_accuracy": 0.19249477088451386,
      "num_tokens": 34580375.0,
      "step": 19930
    },
    {
      "entropy": 5.6548888206481935,
      "epoch": 1.5510212021007586,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004759623817836806,
      "loss": 5.0837,
      "mean_token_accuracy": 0.19720218926668168,
      "num_tokens": 34588762.0,
      "step": 19935
    },
    {
      "entropy": 5.539520359039306,
      "epoch": 1.551410231472476,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004759499158150082,
      "loss": 4.9807,
      "mean_token_accuracy": 0.200630085170269,
      "num_tokens": 34596986.0,
      "step": 19940
    },
    {
      "entropy": 5.650885200500488,
      "epoch": 1.5517992608441937,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047593744679724076,
      "loss": 5.0996,
      "mean_token_accuracy": 0.19646360874176025,
      "num_tokens": 34605445.0,
      "step": 19945
    },
    {
      "entropy": 5.584255361557007,
      "epoch": 1.5521882902159114,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047592497473056733,
      "loss": 5.091,
      "mean_token_accuracy": 0.19710955023765564,
      "num_tokens": 34614793.0,
      "step": 19950
    },
    {
      "entropy": 5.713321399688721,
      "epoch": 1.552577319587629,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047591249961517724,
      "loss": 5.1795,
      "mean_token_accuracy": 0.1833881214261055,
      "num_tokens": 34623790.0,
      "step": 19955
    },
    {
      "entropy": 5.591821718215942,
      "epoch": 1.5529663489593464,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004759000214512598,
      "loss": 5.034,
      "mean_token_accuracy": 0.19421738535165786,
      "num_tokens": 34632387.0,
      "step": 19960
    },
    {
      "entropy": 5.5276631832122805,
      "epoch": 1.553355378331064,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004758875402390042,
      "loss": 5.0224,
      "mean_token_accuracy": 0.2058748260140419,
      "num_tokens": 34640097.0,
      "step": 19965
    },
    {
      "entropy": 5.578539657592773,
      "epoch": 1.5537444077027815,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047587505597859996,
      "loss": 5.0432,
      "mean_token_accuracy": 0.1926708221435547,
      "num_tokens": 34649525.0,
      "step": 19970
    },
    {
      "entropy": 5.534431028366089,
      "epoch": 1.554133437074499,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047586256867023646,
      "loss": 4.9929,
      "mean_token_accuracy": 0.20284031182527543,
      "num_tokens": 34658467.0,
      "step": 19975
    },
    {
      "entropy": 5.645889806747436,
      "epoch": 1.5545224664462167,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004758500783141032,
      "loss": 5.118,
      "mean_token_accuracy": 0.1930898219347,
      "num_tokens": 34666926.0,
      "step": 19980
    },
    {
      "entropy": 5.639770746231079,
      "epoch": 1.5549114958179344,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004758375849103897,
      "loss": 5.0522,
      "mean_token_accuracy": 0.1991471454501152,
      "num_tokens": 34675023.0,
      "step": 19985
    },
    {
      "entropy": 5.5654443264007565,
      "epoch": 1.5553005251896517,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004758250884592855,
      "loss": 4.9982,
      "mean_token_accuracy": 0.20224488377571107,
      "num_tokens": 34683191.0,
      "step": 19990
    },
    {
      "entropy": 5.5716594696044925,
      "epoch": 1.5556895545613694,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047581258896098024,
      "loss": 5.0134,
      "mean_token_accuracy": 0.2008790448307991,
      "num_tokens": 34691899.0,
      "step": 19995
    },
    {
      "entropy": 5.58690242767334,
      "epoch": 1.556078583933087,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004758000864156636,
      "loss": 5.1023,
      "mean_token_accuracy": 0.1934252232313156,
      "num_tokens": 34700942.0,
      "step": 20000
    },
    {
      "entropy": 5.6374091625213625,
      "epoch": 1.5564676133048045,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047578758082352527,
      "loss": 5.1246,
      "mean_token_accuracy": 0.19306767880916595,
      "num_tokens": 34710780.0,
      "step": 20005
    },
    {
      "entropy": 5.616314029693603,
      "epoch": 1.556856642676522,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047577507218475487,
      "loss": 5.0531,
      "mean_token_accuracy": 0.1941116690635681,
      "num_tokens": 34719332.0,
      "step": 20010
    },
    {
      "entropy": 5.634790515899658,
      "epoch": 1.5572456720482397,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047576256049954236,
      "loss": 5.1008,
      "mean_token_accuracy": 0.1875286340713501,
      "num_tokens": 34727890.0,
      "step": 20015
    },
    {
      "entropy": 5.574285793304443,
      "epoch": 1.5576347014199572,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004757500457680776,
      "loss": 5.0796,
      "mean_token_accuracy": 0.192035573720932,
      "num_tokens": 34736942.0,
      "step": 20020
    },
    {
      "entropy": 5.5662578582763675,
      "epoch": 1.5580237307916747,
      "grad_norm": 1.125,
      "learning_rate": 0.0004757375279905504,
      "loss": 4.9341,
      "mean_token_accuracy": 0.2048136055469513,
      "num_tokens": 34745612.0,
      "step": 20025
    },
    {
      "entropy": 5.601886367797851,
      "epoch": 1.5584127601633924,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047572500716715075,
      "loss": 5.0379,
      "mean_token_accuracy": 0.19600876420736313,
      "num_tokens": 34754276.0,
      "step": 20030
    },
    {
      "entropy": 5.563464117050171,
      "epoch": 1.55880178953511,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047571248329806855,
      "loss": 4.9287,
      "mean_token_accuracy": 0.20608661472797393,
      "num_tokens": 34762042.0,
      "step": 20035
    },
    {
      "entropy": 5.527154970169067,
      "epoch": 1.5591908189068273,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047569995638349383,
      "loss": 4.9784,
      "mean_token_accuracy": 0.19811805039644242,
      "num_tokens": 34770172.0,
      "step": 20040
    },
    {
      "entropy": 5.5666392803192135,
      "epoch": 1.559579848278545,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004756874264236168,
      "loss": 4.9957,
      "mean_token_accuracy": 0.20177559107542037,
      "num_tokens": 34778647.0,
      "step": 20045
    },
    {
      "entropy": 5.580323600769043,
      "epoch": 1.5599688776502627,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047567489341862753,
      "loss": 5.0264,
      "mean_token_accuracy": 0.1945127174258232,
      "num_tokens": 34787409.0,
      "step": 20050
    },
    {
      "entropy": 5.625707292556763,
      "epoch": 1.5603579070219802,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047566235736871607,
      "loss": 5.1125,
      "mean_token_accuracy": 0.1873391956090927,
      "num_tokens": 34795634.0,
      "step": 20055
    },
    {
      "entropy": 5.577753353118896,
      "epoch": 1.5607469363936977,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047564981827407277,
      "loss": 4.9637,
      "mean_token_accuracy": 0.2052878975868225,
      "num_tokens": 34803803.0,
      "step": 20060
    },
    {
      "entropy": 5.575420475006103,
      "epoch": 1.5611359657654154,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047563727613488785,
      "loss": 4.9792,
      "mean_token_accuracy": 0.1994999274611473,
      "num_tokens": 34813202.0,
      "step": 20065
    },
    {
      "entropy": 5.591044950485229,
      "epoch": 1.5615249951371328,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047562473095135154,
      "loss": 5.0942,
      "mean_token_accuracy": 0.19303883314132692,
      "num_tokens": 34821717.0,
      "step": 20070
    },
    {
      "entropy": 5.582557725906372,
      "epoch": 1.5619140245088503,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004756121827236544,
      "loss": 4.9969,
      "mean_token_accuracy": 0.1997735992074013,
      "num_tokens": 34830366.0,
      "step": 20075
    },
    {
      "entropy": 5.618888568878174,
      "epoch": 1.562303053880568,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004755996314519866,
      "loss": 5.0667,
      "mean_token_accuracy": 0.19839146137237548,
      "num_tokens": 34838734.0,
      "step": 20080
    },
    {
      "entropy": 5.534379529953003,
      "epoch": 1.5626920832522857,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004755870771365387,
      "loss": 4.9786,
      "mean_token_accuracy": 0.20598955899477006,
      "num_tokens": 34846873.0,
      "step": 20085
    },
    {
      "entropy": 5.582231855392456,
      "epoch": 1.563081112624003,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047557451977750113,
      "loss": 5.0458,
      "mean_token_accuracy": 0.19527384787797927,
      "num_tokens": 34855542.0,
      "step": 20090
    },
    {
      "entropy": 5.558034229278564,
      "epoch": 1.5634701419957207,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004755619593750645,
      "loss": 4.977,
      "mean_token_accuracy": 0.2025115445256233,
      "num_tokens": 34864656.0,
      "step": 20095
    },
    {
      "entropy": 5.616631746292114,
      "epoch": 1.5638591713674383,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004755493959294194,
      "loss": 5.0727,
      "mean_token_accuracy": 0.1959698647260666,
      "num_tokens": 34873891.0,
      "step": 20100
    },
    {
      "entropy": 5.610802745819091,
      "epoch": 1.5642482007391558,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004755368294407564,
      "loss": 5.0748,
      "mean_token_accuracy": 0.1968161165714264,
      "num_tokens": 34883278.0,
      "step": 20105
    },
    {
      "entropy": 5.505993318557739,
      "epoch": 1.5646372301108733,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004755242599092662,
      "loss": 4.9913,
      "mean_token_accuracy": 0.1968425452709198,
      "num_tokens": 34892323.0,
      "step": 20110
    },
    {
      "entropy": 5.615243244171142,
      "epoch": 1.565026259482591,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047551168733513956,
      "loss": 5.0321,
      "mean_token_accuracy": 0.19448182880878448,
      "num_tokens": 34901774.0,
      "step": 20115
    },
    {
      "entropy": 5.667083597183227,
      "epoch": 1.5654152888543085,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047549911171856717,
      "loss": 5.0515,
      "mean_token_accuracy": 0.19572762697935103,
      "num_tokens": 34911082.0,
      "step": 20120
    },
    {
      "entropy": 5.606286573410034,
      "epoch": 1.565804318226026,
      "grad_norm": 1.125,
      "learning_rate": 0.0004754865330597398,
      "loss": 5.0496,
      "mean_token_accuracy": 0.19445881247520447,
      "num_tokens": 34919631.0,
      "step": 20125
    },
    {
      "entropy": 5.585497665405273,
      "epoch": 1.5661933475977436,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047547395135884854,
      "loss": 5.0603,
      "mean_token_accuracy": 0.19418118447065352,
      "num_tokens": 34928258.0,
      "step": 20130
    },
    {
      "entropy": 5.578464460372925,
      "epoch": 1.5665823769694613,
      "grad_norm": 1.125,
      "learning_rate": 0.0004754613666160841,
      "loss": 5.0423,
      "mean_token_accuracy": 0.19600022286176683,
      "num_tokens": 34936590.0,
      "step": 20135
    },
    {
      "entropy": 5.552773284912109,
      "epoch": 1.5669714063411786,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047544877883163753,
      "loss": 5.0646,
      "mean_token_accuracy": 0.19047437906265258,
      "num_tokens": 34945468.0,
      "step": 20140
    },
    {
      "entropy": 5.542993831634521,
      "epoch": 1.5673604357128963,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047543618800569975,
      "loss": 4.9773,
      "mean_token_accuracy": 0.20318094342947007,
      "num_tokens": 34954005.0,
      "step": 20145
    },
    {
      "entropy": 5.502168893814087,
      "epoch": 1.567749465084614,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047542359413846184,
      "loss": 4.9947,
      "mean_token_accuracy": 0.20074473470449447,
      "num_tokens": 34962852.0,
      "step": 20150
    },
    {
      "entropy": 5.579649353027344,
      "epoch": 1.5681384944563315,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004754109972301149,
      "loss": 5.0698,
      "mean_token_accuracy": 0.19368595480918885,
      "num_tokens": 34971718.0,
      "step": 20155
    },
    {
      "entropy": 5.597169256210327,
      "epoch": 1.568527523828049,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047539839728085007,
      "loss": 4.9871,
      "mean_token_accuracy": 0.20480787009000778,
      "num_tokens": 34980155.0,
      "step": 20160
    },
    {
      "entropy": 5.664648580551147,
      "epoch": 1.5689165531997666,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004753857942908585,
      "loss": 5.215,
      "mean_token_accuracy": 0.18520424515008926,
      "num_tokens": 34989971.0,
      "step": 20165
    },
    {
      "entropy": 5.55683741569519,
      "epoch": 1.5693055825714841,
      "grad_norm": 1.125,
      "learning_rate": 0.0004753731882603315,
      "loss": 5.0146,
      "mean_token_accuracy": 0.19009589403867722,
      "num_tokens": 34998760.0,
      "step": 20170
    },
    {
      "entropy": 5.587511873245239,
      "epoch": 1.5696946119432016,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047536057918946026,
      "loss": 5.0526,
      "mean_token_accuracy": 0.19795449078083038,
      "num_tokens": 35008020.0,
      "step": 20175
    },
    {
      "entropy": 5.527889728546143,
      "epoch": 1.5700836413149193,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004753479670784361,
      "loss": 4.9188,
      "mean_token_accuracy": 0.20395711660385132,
      "num_tokens": 35016357.0,
      "step": 20180
    },
    {
      "entropy": 5.542853116989136,
      "epoch": 1.570472670686637,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004753353519274505,
      "loss": 5.0311,
      "mean_token_accuracy": 0.20119148641824722,
      "num_tokens": 35025312.0,
      "step": 20185
    },
    {
      "entropy": 5.612212991714477,
      "epoch": 1.5708617000583545,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004753227337366948,
      "loss": 5.1108,
      "mean_token_accuracy": 0.196427121758461,
      "num_tokens": 35033781.0,
      "step": 20190
    },
    {
      "entropy": 5.645495223999023,
      "epoch": 1.571250729430072,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047531011250636046,
      "loss": 5.1264,
      "mean_token_accuracy": 0.1907256379723549,
      "num_tokens": 35042884.0,
      "step": 20195
    },
    {
      "entropy": 5.5680383205413815,
      "epoch": 1.5716397588017896,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047529748823663896,
      "loss": 4.9594,
      "mean_token_accuracy": 0.20116744190454483,
      "num_tokens": 35051789.0,
      "step": 20200
    },
    {
      "entropy": 5.6125627040863035,
      "epoch": 1.572028788173507,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004752848609277219,
      "loss": 5.0667,
      "mean_token_accuracy": 0.19924165904521943,
      "num_tokens": 35059958.0,
      "step": 20205
    },
    {
      "entropy": 5.5964439868927,
      "epoch": 1.5724178175452246,
      "grad_norm": 1.125,
      "learning_rate": 0.00047527223057980086,
      "loss": 5.0324,
      "mean_token_accuracy": 0.1986391693353653,
      "num_tokens": 35068384.0,
      "step": 20210
    },
    {
      "entropy": 5.663403749465942,
      "epoch": 1.5728068469169423,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047525959719306736,
      "loss": 5.1086,
      "mean_token_accuracy": 0.1908140629529953,
      "num_tokens": 35076660.0,
      "step": 20215
    },
    {
      "entropy": 5.609164476394653,
      "epoch": 1.5731958762886598,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004752469607677134,
      "loss": 5.0129,
      "mean_token_accuracy": 0.20330344438552855,
      "num_tokens": 35085527.0,
      "step": 20220
    },
    {
      "entropy": 5.562790870666504,
      "epoch": 1.5735849056603772,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004752343213039305,
      "loss": 4.9962,
      "mean_token_accuracy": 0.1984825074672699,
      "num_tokens": 35094066.0,
      "step": 20225
    },
    {
      "entropy": 5.648116588592529,
      "epoch": 1.573973935032095,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004752216788019104,
      "loss": 5.1546,
      "mean_token_accuracy": 0.1864498019218445,
      "num_tokens": 35103219.0,
      "step": 20230
    },
    {
      "entropy": 5.63135724067688,
      "epoch": 1.5743629644038126,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004752090332618451,
      "loss": 5.0678,
      "mean_token_accuracy": 0.1967295989394188,
      "num_tokens": 35111336.0,
      "step": 20235
    },
    {
      "entropy": 5.5947425842285154,
      "epoch": 1.57475199377553,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004751963846839263,
      "loss": 5.024,
      "mean_token_accuracy": 0.2028147116303444,
      "num_tokens": 35119758.0,
      "step": 20240
    },
    {
      "entropy": 5.499973392486572,
      "epoch": 1.5751410231472476,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047518373306834605,
      "loss": 4.9585,
      "mean_token_accuracy": 0.20007056295871734,
      "num_tokens": 35128535.0,
      "step": 20245
    },
    {
      "entropy": 5.578498983383179,
      "epoch": 1.5755300525189653,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047517107841529626,
      "loss": 5.1018,
      "mean_token_accuracy": 0.19355902075767517,
      "num_tokens": 35137443.0,
      "step": 20250
    },
    {
      "entropy": 5.533048915863037,
      "epoch": 1.5759190818906827,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000475158420724969,
      "loss": 4.9079,
      "mean_token_accuracy": 0.21057977974414827,
      "num_tokens": 35145801.0,
      "step": 20255
    },
    {
      "entropy": 5.640761232376098,
      "epoch": 1.5763081112624002,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047514575999755627,
      "loss": 5.1448,
      "mean_token_accuracy": 0.18846935778856277,
      "num_tokens": 35154661.0,
      "step": 20260
    },
    {
      "entropy": 5.6143097400665285,
      "epoch": 1.576697140634118,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047513309623325024,
      "loss": 5.1471,
      "mean_token_accuracy": 0.1856221467256546,
      "num_tokens": 35163762.0,
      "step": 20265
    },
    {
      "entropy": 5.562805080413819,
      "epoch": 1.5770861700058354,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004751204294322429,
      "loss": 5.0499,
      "mean_token_accuracy": 0.19497397691011428,
      "num_tokens": 35172059.0,
      "step": 20270
    },
    {
      "entropy": 5.583526515960694,
      "epoch": 1.5774751993775529,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047510775959472675,
      "loss": 5.099,
      "mean_token_accuracy": 0.18765577226877211,
      "num_tokens": 35179805.0,
      "step": 20275
    },
    {
      "entropy": 5.5387366771697994,
      "epoch": 1.5778642287492706,
      "grad_norm": 1.125,
      "learning_rate": 0.0004750950867208937,
      "loss": 5.0597,
      "mean_token_accuracy": 0.20366354137659073,
      "num_tokens": 35188136.0,
      "step": 20280
    },
    {
      "entropy": 5.688149738311767,
      "epoch": 1.5782532581209883,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004750824108109362,
      "loss": 5.1419,
      "mean_token_accuracy": 0.18975001871585845,
      "num_tokens": 35196577.0,
      "step": 20285
    },
    {
      "entropy": 5.580220174789429,
      "epoch": 1.5786422874927057,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004750697318650466,
      "loss": 5.0584,
      "mean_token_accuracy": 0.1920944258570671,
      "num_tokens": 35205223.0,
      "step": 20290
    },
    {
      "entropy": 5.660600662231445,
      "epoch": 1.5790313168644232,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004750570498834173,
      "loss": 5.1235,
      "mean_token_accuracy": 0.19122048318386078,
      "num_tokens": 35213776.0,
      "step": 20295
    },
    {
      "entropy": 5.577645397186279,
      "epoch": 1.579420346236141,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004750443648662407,
      "loss": 5.0419,
      "mean_token_accuracy": 0.20026295483112336,
      "num_tokens": 35222596.0,
      "step": 20300
    },
    {
      "entropy": 5.670751667022705,
      "epoch": 1.5798093756078584,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047503167681370924,
      "loss": 5.1811,
      "mean_token_accuracy": 0.18609293699264526,
      "num_tokens": 35232402.0,
      "step": 20305
    },
    {
      "entropy": 5.60352668762207,
      "epoch": 1.5801984049795759,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004750189857260154,
      "loss": 4.9721,
      "mean_token_accuracy": 0.2055818647146225,
      "num_tokens": 35240651.0,
      "step": 20310
    },
    {
      "entropy": 5.532120037078857,
      "epoch": 1.5805874343512936,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004750062916033519,
      "loss": 5.0122,
      "mean_token_accuracy": 0.19325342029333115,
      "num_tokens": 35249101.0,
      "step": 20315
    },
    {
      "entropy": 5.587149572372437,
      "epoch": 1.580976463723011,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004749935944459113,
      "loss": 5.1074,
      "mean_token_accuracy": 0.1891570582985878,
      "num_tokens": 35257909.0,
      "step": 20320
    },
    {
      "entropy": 5.66335096359253,
      "epoch": 1.5813654930947285,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004749808942538862,
      "loss": 5.1743,
      "mean_token_accuracy": 0.1812183141708374,
      "num_tokens": 35267128.0,
      "step": 20325
    },
    {
      "entropy": 5.615903520584107,
      "epoch": 1.5817545224664462,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004749681910274693,
      "loss": 5.0436,
      "mean_token_accuracy": 0.19352351129055023,
      "num_tokens": 35275751.0,
      "step": 20330
    },
    {
      "entropy": 5.571080923080444,
      "epoch": 1.582143551838164,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047495548476685334,
      "loss": 4.9522,
      "mean_token_accuracy": 0.20331335365772246,
      "num_tokens": 35285359.0,
      "step": 20335
    },
    {
      "entropy": 5.527443790435791,
      "epoch": 1.5825325812098814,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047494277547223125,
      "loss": 4.9585,
      "mean_token_accuracy": 0.19941352009773256,
      "num_tokens": 35293691.0,
      "step": 20340
    },
    {
      "entropy": 5.528579187393189,
      "epoch": 1.5829216105815989,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047493006314379573,
      "loss": 4.996,
      "mean_token_accuracy": 0.20230142623186112,
      "num_tokens": 35302751.0,
      "step": 20345
    },
    {
      "entropy": 5.665782880783081,
      "epoch": 1.5833106399533166,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047491734778173976,
      "loss": 5.0727,
      "mean_token_accuracy": 0.19230367988348007,
      "num_tokens": 35311849.0,
      "step": 20350
    },
    {
      "entropy": 5.560361528396607,
      "epoch": 1.583699669325034,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047490462938625617,
      "loss": 5.0079,
      "mean_token_accuracy": 0.19786733388900757,
      "num_tokens": 35320248.0,
      "step": 20355
    },
    {
      "entropy": 5.593536281585694,
      "epoch": 1.5840886986967515,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047489190795753806,
      "loss": 5.0056,
      "mean_token_accuracy": 0.2081570953130722,
      "num_tokens": 35328565.0,
      "step": 20360
    },
    {
      "entropy": 5.579427480697632,
      "epoch": 1.5844777280684692,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004748791834957784,
      "loss": 5.0217,
      "mean_token_accuracy": 0.2008778691291809,
      "num_tokens": 35337828.0,
      "step": 20365
    },
    {
      "entropy": 5.60620527267456,
      "epoch": 1.5848667574401867,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047486645600117037,
      "loss": 5.0908,
      "mean_token_accuracy": 0.18780517876148223,
      "num_tokens": 35347107.0,
      "step": 20370
    },
    {
      "entropy": 5.601256275177002,
      "epoch": 1.5852557868119042,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004748537254739068,
      "loss": 5.0226,
      "mean_token_accuracy": 0.19478196799755096,
      "num_tokens": 35355925.0,
      "step": 20375
    },
    {
      "entropy": 5.568712854385376,
      "epoch": 1.5856448161836219,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004748409919141812,
      "loss": 5.0053,
      "mean_token_accuracy": 0.20000686049461364,
      "num_tokens": 35364319.0,
      "step": 20380
    },
    {
      "entropy": 5.50074520111084,
      "epoch": 1.5860338455553395,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004748282553221865,
      "loss": 4.9402,
      "mean_token_accuracy": 0.204921455681324,
      "num_tokens": 35372358.0,
      "step": 20385
    },
    {
      "entropy": 5.589786005020142,
      "epoch": 1.586422874927057,
      "grad_norm": 1.125,
      "learning_rate": 0.0004748155156981162,
      "loss": 5.1736,
      "mean_token_accuracy": 0.18421153277158736,
      "num_tokens": 35381301.0,
      "step": 20390
    },
    {
      "entropy": 5.5876593589782715,
      "epoch": 1.5868119042987745,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047480277304216346,
      "loss": 5.0008,
      "mean_token_accuracy": 0.19653751701116562,
      "num_tokens": 35389778.0,
      "step": 20395
    },
    {
      "entropy": 5.577833747863769,
      "epoch": 1.5872009336704922,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047479002735452164,
      "loss": 4.9646,
      "mean_token_accuracy": 0.20017237812280655,
      "num_tokens": 35397926.0,
      "step": 20400
    },
    {
      "entropy": 5.562922239303589,
      "epoch": 1.5875899630422097,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047477727863538415,
      "loss": 5.0014,
      "mean_token_accuracy": 0.20306000113487244,
      "num_tokens": 35406462.0,
      "step": 20405
    },
    {
      "entropy": 5.6313729763031,
      "epoch": 1.5879789924139271,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047476452688494444,
      "loss": 5.0112,
      "mean_token_accuracy": 0.19846069812774658,
      "num_tokens": 35414989.0,
      "step": 20410
    },
    {
      "entropy": 5.604921007156372,
      "epoch": 1.5883680217856448,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000474751772103396,
      "loss": 4.9893,
      "mean_token_accuracy": 0.20431422889232637,
      "num_tokens": 35423738.0,
      "step": 20415
    },
    {
      "entropy": 5.609814882278442,
      "epoch": 1.5887570511573625,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004747390142909323,
      "loss": 5.0483,
      "mean_token_accuracy": 0.19522486627101898,
      "num_tokens": 35432377.0,
      "step": 20420
    },
    {
      "entropy": 5.543861389160156,
      "epoch": 1.5891460805290798,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047472625344774713,
      "loss": 5.0318,
      "mean_token_accuracy": 0.20262193977832793,
      "num_tokens": 35441188.0,
      "step": 20425
    },
    {
      "entropy": 5.585594749450683,
      "epoch": 1.5895351099007975,
      "grad_norm": 1.125,
      "learning_rate": 0.00047471348957403374,
      "loss": 5.0855,
      "mean_token_accuracy": 0.19431216269731522,
      "num_tokens": 35449715.0,
      "step": 20430
    },
    {
      "entropy": 5.593348360061645,
      "epoch": 1.5899241392725152,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004747007226699862,
      "loss": 5.043,
      "mean_token_accuracy": 0.1976607546210289,
      "num_tokens": 35458420.0,
      "step": 20435
    },
    {
      "entropy": 5.621132230758667,
      "epoch": 1.5903131686442327,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047468795273579803,
      "loss": 4.9659,
      "mean_token_accuracy": 0.20256307870149612,
      "num_tokens": 35467072.0,
      "step": 20440
    },
    {
      "entropy": 5.584652709960937,
      "epoch": 1.5907021980159501,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004746751797716629,
      "loss": 5.0288,
      "mean_token_accuracy": 0.19904178828001023,
      "num_tokens": 35475539.0,
      "step": 20445
    },
    {
      "entropy": 5.592234325408936,
      "epoch": 1.5910912273876678,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004746624037777748,
      "loss": 5.0401,
      "mean_token_accuracy": 0.20247142761945724,
      "num_tokens": 35484260.0,
      "step": 20450
    },
    {
      "entropy": 5.606941366195679,
      "epoch": 1.5914802567593853,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047464962475432754,
      "loss": 4.9761,
      "mean_token_accuracy": 0.20188238024711608,
      "num_tokens": 35492539.0,
      "step": 20455
    },
    {
      "entropy": 5.629344320297241,
      "epoch": 1.5918692861311028,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000474636842701515,
      "loss": 5.13,
      "mean_token_accuracy": 0.19017908424139024,
      "num_tokens": 35502008.0,
      "step": 20460
    },
    {
      "entropy": 5.549961853027344,
      "epoch": 1.5922583155028205,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004746240576195311,
      "loss": 4.9886,
      "mean_token_accuracy": 0.20168758928775787,
      "num_tokens": 35510677.0,
      "step": 20465
    },
    {
      "entropy": 5.559269523620605,
      "epoch": 1.5926473448745382,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047461126950856987,
      "loss": 5.0269,
      "mean_token_accuracy": 0.19858846068382263,
      "num_tokens": 35519368.0,
      "step": 20470
    },
    {
      "entropy": 5.6469896793365475,
      "epoch": 1.5930363742462554,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004745984783688253,
      "loss": 5.1715,
      "mean_token_accuracy": 0.18591942638158798,
      "num_tokens": 35527952.0,
      "step": 20475
    },
    {
      "entropy": 5.647821664810181,
      "epoch": 1.5934254036179731,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047458568420049153,
      "loss": 5.0863,
      "mean_token_accuracy": 0.1939651146531105,
      "num_tokens": 35537806.0,
      "step": 20480
    },
    {
      "entropy": 5.551797103881836,
      "epoch": 1.5938144329896908,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047457288700376274,
      "loss": 4.9349,
      "mean_token_accuracy": 0.19671977013349534,
      "num_tokens": 35546135.0,
      "step": 20485
    },
    {
      "entropy": 5.5138627052307125,
      "epoch": 1.5942034623614083,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047456008677883304,
      "loss": 5.004,
      "mean_token_accuracy": 0.19634348899126053,
      "num_tokens": 35555065.0,
      "step": 20490
    },
    {
      "entropy": 5.5616833686828615,
      "epoch": 1.5945924917331258,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004745472835258966,
      "loss": 5.1244,
      "mean_token_accuracy": 0.18762902915477753,
      "num_tokens": 35564222.0,
      "step": 20495
    },
    {
      "entropy": 5.64038200378418,
      "epoch": 1.5949815211048435,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047453447724514773,
      "loss": 5.0796,
      "mean_token_accuracy": 0.1916970580816269,
      "num_tokens": 35573021.0,
      "step": 20500
    },
    {
      "entropy": 5.648864316940307,
      "epoch": 1.595370550476561,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004745216679367808,
      "loss": 5.1123,
      "mean_token_accuracy": 0.19242366701364516,
      "num_tokens": 35581105.0,
      "step": 20505
    },
    {
      "entropy": 5.642642402648926,
      "epoch": 1.5957595798482784,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004745088556009901,
      "loss": 5.1566,
      "mean_token_accuracy": 0.18715530186891555,
      "num_tokens": 35590169.0,
      "step": 20510
    },
    {
      "entropy": 5.628381061553955,
      "epoch": 1.5961486092199961,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004744960402379701,
      "loss": 5.0416,
      "mean_token_accuracy": 0.2013960674405098,
      "num_tokens": 35598760.0,
      "step": 20515
    },
    {
      "entropy": 5.650123357772827,
      "epoch": 1.5965376385917138,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047448322184791525,
      "loss": 5.0853,
      "mean_token_accuracy": 0.19448831379413606,
      "num_tokens": 35607014.0,
      "step": 20520
    },
    {
      "entropy": 5.617592620849609,
      "epoch": 1.596926667963431,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047447040043101994,
      "loss": 5.0535,
      "mean_token_accuracy": 0.19487944692373277,
      "num_tokens": 35615319.0,
      "step": 20525
    },
    {
      "entropy": 5.548729515075683,
      "epoch": 1.5973156973351488,
      "grad_norm": 1.125,
      "learning_rate": 0.00047445757598747886,
      "loss": 5.0724,
      "mean_token_accuracy": 0.19767432659864426,
      "num_tokens": 35623836.0,
      "step": 20530
    },
    {
      "entropy": 5.536630439758301,
      "epoch": 1.5977047267068665,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004744447485174864,
      "loss": 4.9449,
      "mean_token_accuracy": 0.19615080803632737,
      "num_tokens": 35632601.0,
      "step": 20535
    },
    {
      "entropy": 5.671441888809204,
      "epoch": 1.598093756078584,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047443191802123746,
      "loss": 5.115,
      "mean_token_accuracy": 0.19679517596960067,
      "num_tokens": 35642035.0,
      "step": 20540
    },
    {
      "entropy": 5.600108194351196,
      "epoch": 1.5984827854503014,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047441908449892664,
      "loss": 5.0283,
      "mean_token_accuracy": 0.20052443891763688,
      "num_tokens": 35650454.0,
      "step": 20545
    },
    {
      "entropy": 5.605909395217895,
      "epoch": 1.5988718148220191,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047440624795074855,
      "loss": 5.0746,
      "mean_token_accuracy": 0.19575013518333434,
      "num_tokens": 35658945.0,
      "step": 20550
    },
    {
      "entropy": 5.629741287231445,
      "epoch": 1.5992608441937366,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047439340837689804,
      "loss": 5.062,
      "mean_token_accuracy": 0.19770074039697647,
      "num_tokens": 35667193.0,
      "step": 20555
    },
    {
      "entropy": 5.567346906661987,
      "epoch": 1.599649873565454,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047438056577756986,
      "loss": 4.9256,
      "mean_token_accuracy": 0.2032573327422142,
      "num_tokens": 35676100.0,
      "step": 20560
    },
    {
      "entropy": 5.578464078903198,
      "epoch": 1.6000389029371718,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047436772015295903,
      "loss": 5.0471,
      "mean_token_accuracy": 0.1980154901742935,
      "num_tokens": 35684352.0,
      "step": 20565
    },
    {
      "entropy": 5.61439962387085,
      "epoch": 1.6004279323088895,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047435487150326036,
      "loss": 5.0062,
      "mean_token_accuracy": 0.20416561514139175,
      "num_tokens": 35693044.0,
      "step": 20570
    },
    {
      "entropy": 5.544478988647461,
      "epoch": 1.6008169616806067,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004743420198286688,
      "loss": 4.9925,
      "mean_token_accuracy": 0.20231689810752868,
      "num_tokens": 35701172.0,
      "step": 20575
    },
    {
      "entropy": 5.598053741455078,
      "epoch": 1.6012059910523244,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047432916512937936,
      "loss": 5.0925,
      "mean_token_accuracy": 0.19265136867761612,
      "num_tokens": 35709905.0,
      "step": 20580
    },
    {
      "entropy": 5.560888433456421,
      "epoch": 1.6015950204240421,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004743163074055871,
      "loss": 5.0584,
      "mean_token_accuracy": 0.19510424584150315,
      "num_tokens": 35718404.0,
      "step": 20585
    },
    {
      "entropy": 5.619507455825806,
      "epoch": 1.6019840497957596,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004743034466574871,
      "loss": 5.0657,
      "mean_token_accuracy": 0.19765877723693848,
      "num_tokens": 35726475.0,
      "step": 20590
    },
    {
      "entropy": 5.619621801376343,
      "epoch": 1.602373079167477,
      "grad_norm": 1.125,
      "learning_rate": 0.0004742905828852746,
      "loss": 5.0857,
      "mean_token_accuracy": 0.19610037356615068,
      "num_tokens": 35734819.0,
      "step": 20595
    },
    {
      "entropy": 5.524581718444824,
      "epoch": 1.6027621085391948,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004742777160891447,
      "loss": 4.9701,
      "mean_token_accuracy": 0.20106784552335738,
      "num_tokens": 35743098.0,
      "step": 20600
    },
    {
      "entropy": 5.606900358200074,
      "epoch": 1.6031511379109122,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004742648462692926,
      "loss": 5.0845,
      "mean_token_accuracy": 0.19541829228401184,
      "num_tokens": 35752502.0,
      "step": 20605
    },
    {
      "entropy": 5.637093019485474,
      "epoch": 1.6035401672826297,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047425197342591363,
      "loss": 5.0276,
      "mean_token_accuracy": 0.20529707968235017,
      "num_tokens": 35761293.0,
      "step": 20610
    },
    {
      "entropy": 5.622268342971802,
      "epoch": 1.6039291966543474,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004742390975592031,
      "loss": 5.1535,
      "mean_token_accuracy": 0.18202961683273317,
      "num_tokens": 35771031.0,
      "step": 20615
    },
    {
      "entropy": 5.599624395370483,
      "epoch": 1.604318226026065,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047422621866935645,
      "loss": 5.0477,
      "mean_token_accuracy": 0.19335551857948302,
      "num_tokens": 35778894.0,
      "step": 20620
    },
    {
      "entropy": 5.675481796264648,
      "epoch": 1.6047072553977826,
      "grad_norm": 1.171875,
      "learning_rate": 0.000474213336756569,
      "loss": 5.1047,
      "mean_token_accuracy": 0.19172658920288085,
      "num_tokens": 35787159.0,
      "step": 20625
    },
    {
      "entropy": 5.561765670776367,
      "epoch": 1.6050962847695,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004742004518210362,
      "loss": 5.0366,
      "mean_token_accuracy": 0.20181388258934022,
      "num_tokens": 35795858.0,
      "step": 20630
    },
    {
      "entropy": 5.567565202713013,
      "epoch": 1.6054853141412178,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004741875638629536,
      "loss": 5.038,
      "mean_token_accuracy": 0.19847733825445174,
      "num_tokens": 35804754.0,
      "step": 20635
    },
    {
      "entropy": 5.638898944854736,
      "epoch": 1.6058743435129352,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004741746728825168,
      "loss": 5.1304,
      "mean_token_accuracy": 0.18828104734420775,
      "num_tokens": 35813622.0,
      "step": 20640
    },
    {
      "entropy": 5.588763046264648,
      "epoch": 1.6062633728846527,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004741617788799213,
      "loss": 5.0029,
      "mean_token_accuracy": 0.1989662989974022,
      "num_tokens": 35821345.0,
      "step": 20645
    },
    {
      "entropy": 5.562273788452148,
      "epoch": 1.6066524022563704,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047414888185536285,
      "loss": 5.0269,
      "mean_token_accuracy": 0.20523292124271392,
      "num_tokens": 35830058.0,
      "step": 20650
    },
    {
      "entropy": 5.608818483352661,
      "epoch": 1.6070414316280879,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004741359818090371,
      "loss": 5.0507,
      "mean_token_accuracy": 0.1980800971388817,
      "num_tokens": 35838951.0,
      "step": 20655
    },
    {
      "entropy": 5.597017860412597,
      "epoch": 1.6074304609998054,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047412307874113976,
      "loss": 5.0301,
      "mean_token_accuracy": 0.20165429562330245,
      "num_tokens": 35847245.0,
      "step": 20660
    },
    {
      "entropy": 5.634465503692627,
      "epoch": 1.607819490371523,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004741101726518667,
      "loss": 5.142,
      "mean_token_accuracy": 0.19017770737409592,
      "num_tokens": 35856037.0,
      "step": 20665
    },
    {
      "entropy": 5.562431764602661,
      "epoch": 1.6082085197432407,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004740972635414136,
      "loss": 4.979,
      "mean_token_accuracy": 0.20021205842494966,
      "num_tokens": 35864881.0,
      "step": 20670
    },
    {
      "entropy": 5.674786901473999,
      "epoch": 1.6085975491149582,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004740843514099765,
      "loss": 5.0849,
      "mean_token_accuracy": 0.19342207163572311,
      "num_tokens": 35872941.0,
      "step": 20675
    },
    {
      "entropy": 5.556571197509766,
      "epoch": 1.6089865784866757,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004740714362577512,
      "loss": 5.058,
      "mean_token_accuracy": 0.1919024795293808,
      "num_tokens": 35882068.0,
      "step": 20680
    },
    {
      "entropy": 5.573685979843139,
      "epoch": 1.6093756078583934,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047405851808493364,
      "loss": 4.9974,
      "mean_token_accuracy": 0.19965358674526215,
      "num_tokens": 35890306.0,
      "step": 20685
    },
    {
      "entropy": 5.611655950546265,
      "epoch": 1.6097646372301109,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047404559689172006,
      "loss": 5.0593,
      "mean_token_accuracy": 0.20373389273881912,
      "num_tokens": 35899311.0,
      "step": 20690
    },
    {
      "entropy": 5.604390096664429,
      "epoch": 1.6101536666018283,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047403267267830617,
      "loss": 5.1198,
      "mean_token_accuracy": 0.18639786690473556,
      "num_tokens": 35908599.0,
      "step": 20695
    },
    {
      "entropy": 5.600953006744385,
      "epoch": 1.610542695973546,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047401974544488844,
      "loss": 5.0677,
      "mean_token_accuracy": 0.19232304245233536,
      "num_tokens": 35917318.0,
      "step": 20700
    },
    {
      "entropy": 5.628634881973267,
      "epoch": 1.6109317253452635,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004740068151916628,
      "loss": 5.0371,
      "mean_token_accuracy": 0.1977323532104492,
      "num_tokens": 35926179.0,
      "step": 20705
    },
    {
      "entropy": 5.533683156967163,
      "epoch": 1.611320754716981,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004739938819188255,
      "loss": 4.8991,
      "mean_token_accuracy": 0.21258203089237213,
      "num_tokens": 35934155.0,
      "step": 20710
    },
    {
      "entropy": 5.5131481170654295,
      "epoch": 1.6117097840886987,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004739809456265727,
      "loss": 4.9884,
      "mean_token_accuracy": 0.20170480459928514,
      "num_tokens": 35942301.0,
      "step": 20715
    },
    {
      "entropy": 5.517496538162232,
      "epoch": 1.6120988134604164,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004739680063151008,
      "loss": 4.9844,
      "mean_token_accuracy": 0.20154403299093246,
      "num_tokens": 35951005.0,
      "step": 20720
    },
    {
      "entropy": 5.558914518356323,
      "epoch": 1.6124878428321339,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004739550639846061,
      "loss": 5.0876,
      "mean_token_accuracy": 0.19566754549741744,
      "num_tokens": 35959425.0,
      "step": 20725
    },
    {
      "entropy": 5.609003114700317,
      "epoch": 1.6128768722038513,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047394211863528496,
      "loss": 5.022,
      "mean_token_accuracy": 0.19801850318908693,
      "num_tokens": 35967803.0,
      "step": 20730
    },
    {
      "entropy": 5.621663904190063,
      "epoch": 1.613265901575569,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004739291702673338,
      "loss": 5.1413,
      "mean_token_accuracy": 0.19048357009887695,
      "num_tokens": 35977313.0,
      "step": 20735
    },
    {
      "entropy": 5.687045288085938,
      "epoch": 1.6136549309472865,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004739162188809492,
      "loss": 5.115,
      "mean_token_accuracy": 0.19212632328271867,
      "num_tokens": 35986404.0,
      "step": 20740
    },
    {
      "entropy": 5.62446346282959,
      "epoch": 1.614043960319004,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047390326447632747,
      "loss": 5.0935,
      "mean_token_accuracy": 0.19979857206344603,
      "num_tokens": 35995865.0,
      "step": 20745
    },
    {
      "entropy": 5.5454638481140135,
      "epoch": 1.6144329896907217,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004738903070536654,
      "loss": 4.9867,
      "mean_token_accuracy": 0.20248697847127914,
      "num_tokens": 36005098.0,
      "step": 20750
    },
    {
      "entropy": 5.550710248947143,
      "epoch": 1.6148220190624392,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004738773466131594,
      "loss": 5.0375,
      "mean_token_accuracy": 0.2022942289710045,
      "num_tokens": 36013642.0,
      "step": 20755
    },
    {
      "entropy": 5.6437986373901365,
      "epoch": 1.6152110484341566,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004738643831550063,
      "loss": 5.1326,
      "mean_token_accuracy": 0.1917092338204384,
      "num_tokens": 36022565.0,
      "step": 20760
    },
    {
      "entropy": 5.62987322807312,
      "epoch": 1.6156000778058743,
      "grad_norm": 1.125,
      "learning_rate": 0.0004738514166794026,
      "loss": 5.0909,
      "mean_token_accuracy": 0.1914977326989174,
      "num_tokens": 36031406.0,
      "step": 20765
    },
    {
      "entropy": 5.627246379852295,
      "epoch": 1.615989107177592,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047383844718654525,
      "loss": 5.0942,
      "mean_token_accuracy": 0.194160595536232,
      "num_tokens": 36040282.0,
      "step": 20770
    },
    {
      "entropy": 5.710406494140625,
      "epoch": 1.6163781365493095,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004738254746766309,
      "loss": 5.154,
      "mean_token_accuracy": 0.19395956099033357,
      "num_tokens": 36049647.0,
      "step": 20775
    },
    {
      "entropy": 5.583831310272217,
      "epoch": 1.616767165921027,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004738124991498565,
      "loss": 4.9771,
      "mean_token_accuracy": 0.2017980694770813,
      "num_tokens": 36057992.0,
      "step": 20780
    },
    {
      "entropy": 5.575826025009155,
      "epoch": 1.6171561952927447,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047379952060641866,
      "loss": 5.0323,
      "mean_token_accuracy": 0.20144667029380797,
      "num_tokens": 36066145.0,
      "step": 20785
    },
    {
      "entropy": 5.55538763999939,
      "epoch": 1.6175452246644622,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047378653904651475,
      "loss": 5.0317,
      "mean_token_accuracy": 0.19647867679595948,
      "num_tokens": 36074596.0,
      "step": 20790
    },
    {
      "entropy": 5.6219950199127195,
      "epoch": 1.6179342540361796,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047377355447034136,
      "loss": 4.9486,
      "mean_token_accuracy": 0.20715731680393218,
      "num_tokens": 36083499.0,
      "step": 20795
    },
    {
      "entropy": 5.6103228569030765,
      "epoch": 1.6183232834078973,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004737605668780958,
      "loss": 5.0632,
      "mean_token_accuracy": 0.19658583998680115,
      "num_tokens": 36092293.0,
      "step": 20800
    },
    {
      "entropy": 5.5501752376556395,
      "epoch": 1.618712312779615,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047374757626997494,
      "loss": 5.0459,
      "mean_token_accuracy": 0.19150787591934204,
      "num_tokens": 36100926.0,
      "step": 20805
    },
    {
      "entropy": 5.668728303909302,
      "epoch": 1.6191013421513323,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004737345826461759,
      "loss": 5.0915,
      "mean_token_accuracy": 0.19880053997039795,
      "num_tokens": 36109831.0,
      "step": 20810
    },
    {
      "entropy": 5.66787805557251,
      "epoch": 1.61949037152305,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004737215860068959,
      "loss": 5.1494,
      "mean_token_accuracy": 0.19307994842529297,
      "num_tokens": 36118759.0,
      "step": 20815
    },
    {
      "entropy": 5.635524892807007,
      "epoch": 1.6198794008947677,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047370858635233223,
      "loss": 5.1819,
      "mean_token_accuracy": 0.1869405224919319,
      "num_tokens": 36127854.0,
      "step": 20820
    },
    {
      "entropy": 5.596605396270752,
      "epoch": 1.6202684302664851,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047369558368268194,
      "loss": 5.0783,
      "mean_token_accuracy": 0.1983827143907547,
      "num_tokens": 36137686.0,
      "step": 20825
    },
    {
      "entropy": 5.642459154129028,
      "epoch": 1.6206574596382026,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004736825779981424,
      "loss": 5.0686,
      "mean_token_accuracy": 0.19523619562387468,
      "num_tokens": 36146352.0,
      "step": 20830
    },
    {
      "entropy": 5.582647228240967,
      "epoch": 1.6210464890099203,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004736695692989111,
      "loss": 4.9996,
      "mean_token_accuracy": 0.20211186856031418,
      "num_tokens": 36156106.0,
      "step": 20835
    },
    {
      "entropy": 5.5328929901123045,
      "epoch": 1.6214355183816378,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004736565575851852,
      "loss": 4.9149,
      "mean_token_accuracy": 0.2003731831908226,
      "num_tokens": 36164169.0,
      "step": 20840
    },
    {
      "entropy": 5.614877223968506,
      "epoch": 1.6218245477533553,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004736435428571622,
      "loss": 5.0682,
      "mean_token_accuracy": 0.1957612544298172,
      "num_tokens": 36173368.0,
      "step": 20845
    },
    {
      "entropy": 5.604243993759155,
      "epoch": 1.622213577125073,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004736305251150397,
      "loss": 5.0427,
      "mean_token_accuracy": 0.1916337102651596,
      "num_tokens": 36182405.0,
      "step": 20850
    },
    {
      "entropy": 5.6107336521148685,
      "epoch": 1.6226026064967907,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004736175043590151,
      "loss": 5.0975,
      "mean_token_accuracy": 0.1956746682524681,
      "num_tokens": 36191683.0,
      "step": 20855
    },
    {
      "entropy": 5.607086181640625,
      "epoch": 1.622991635868508,
      "grad_norm": 1.140625,
      "learning_rate": 0.000473604480589286,
      "loss": 4.9823,
      "mean_token_accuracy": 0.19905490428209305,
      "num_tokens": 36200173.0,
      "step": 20860
    },
    {
      "entropy": 5.517714309692383,
      "epoch": 1.6233806652402256,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047359145380605007,
      "loss": 5.014,
      "mean_token_accuracy": 0.19795836955308915,
      "num_tokens": 36209057.0,
      "step": 20865
    },
    {
      "entropy": 5.591701507568359,
      "epoch": 1.6237696946119433,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004735784240095049,
      "loss": 5.0193,
      "mean_token_accuracy": 0.19638884365558623,
      "num_tokens": 36217645.0,
      "step": 20870
    },
    {
      "entropy": 5.664293813705444,
      "epoch": 1.6241587239836608,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047356539119984813,
      "loss": 5.097,
      "mean_token_accuracy": 0.19613057374954224,
      "num_tokens": 36226282.0,
      "step": 20875
    },
    {
      "entropy": 5.6145209789276125,
      "epoch": 1.6245477533553783,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004735523553772777,
      "loss": 5.0754,
      "mean_token_accuracy": 0.19918007552623748,
      "num_tokens": 36235322.0,
      "step": 20880
    },
    {
      "entropy": 5.59035005569458,
      "epoch": 1.624936782727096,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004735393165419912,
      "loss": 5.0115,
      "mean_token_accuracy": 0.2092021495103836,
      "num_tokens": 36244212.0,
      "step": 20885
    },
    {
      "entropy": 5.546533679962158,
      "epoch": 1.6253258120988134,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047352627469418666,
      "loss": 5.0201,
      "mean_token_accuracy": 0.200291445851326,
      "num_tokens": 36252837.0,
      "step": 20890
    },
    {
      "entropy": 5.635124349594117,
      "epoch": 1.625714841470531,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004735132298340619,
      "loss": 5.0628,
      "mean_token_accuracy": 0.19072452783584595,
      "num_tokens": 36261492.0,
      "step": 20895
    },
    {
      "entropy": 5.636734914779663,
      "epoch": 1.6261038708422486,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004735001819618148,
      "loss": 5.0474,
      "mean_token_accuracy": 0.19585788995027542,
      "num_tokens": 36270195.0,
      "step": 20900
    },
    {
      "entropy": 5.603625917434693,
      "epoch": 1.6264929002139663,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004734871310776434,
      "loss": 4.9834,
      "mean_token_accuracy": 0.21196918785572053,
      "num_tokens": 36278546.0,
      "step": 20905
    },
    {
      "entropy": 5.616297054290771,
      "epoch": 1.6268819295856836,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004734740771817457,
      "loss": 5.1164,
      "mean_token_accuracy": 0.18655333071947097,
      "num_tokens": 36287433.0,
      "step": 20910
    },
    {
      "entropy": 5.527868223190308,
      "epoch": 1.6272709589574013,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004734610202743198,
      "loss": 4.979,
      "mean_token_accuracy": 0.1922747254371643,
      "num_tokens": 36296213.0,
      "step": 20915
    },
    {
      "entropy": 5.548872947692871,
      "epoch": 1.627659988329119,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004734479603555638,
      "loss": 4.9486,
      "mean_token_accuracy": 0.214207661151886,
      "num_tokens": 36303989.0,
      "step": 20920
    },
    {
      "entropy": 5.651712417602539,
      "epoch": 1.6280490177008364,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047343489742567593,
      "loss": 5.2291,
      "mean_token_accuracy": 0.18782870322465897,
      "num_tokens": 36312939.0,
      "step": 20925
    },
    {
      "entropy": 5.661740159988403,
      "epoch": 1.628438047072554,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047342183148485424,
      "loss": 5.0079,
      "mean_token_accuracy": 0.19800445437431335,
      "num_tokens": 36322130.0,
      "step": 20930
    },
    {
      "entropy": 5.592354583740234,
      "epoch": 1.6288270764442716,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047340876253329706,
      "loss": 4.9613,
      "mean_token_accuracy": 0.2005373001098633,
      "num_tokens": 36330376.0,
      "step": 20935
    },
    {
      "entropy": 5.525694847106934,
      "epoch": 1.629216105815989,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047339569057120275,
      "loss": 4.9834,
      "mean_token_accuracy": 0.20261211991310119,
      "num_tokens": 36339614.0,
      "step": 20940
    },
    {
      "entropy": 5.610240936279297,
      "epoch": 1.6296051351877066,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047338261559876966,
      "loss": 5.087,
      "mean_token_accuracy": 0.19515491873025895,
      "num_tokens": 36349561.0,
      "step": 20945
    },
    {
      "entropy": 5.685820198059082,
      "epoch": 1.6299941645594243,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047336953761619604,
      "loss": 5.1143,
      "mean_token_accuracy": 0.1891432821750641,
      "num_tokens": 36358277.0,
      "step": 20950
    },
    {
      "entropy": 5.6725646495819095,
      "epoch": 1.630383193931142,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047335645662368046,
      "loss": 5.0902,
      "mean_token_accuracy": 0.20042209923267365,
      "num_tokens": 36366736.0,
      "step": 20955
    },
    {
      "entropy": 5.601393032073974,
      "epoch": 1.6307722233028592,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004733433726214214,
      "loss": 5.0422,
      "mean_token_accuracy": 0.20678918361663817,
      "num_tokens": 36374802.0,
      "step": 20960
    },
    {
      "entropy": 5.607021856307983,
      "epoch": 1.631161252674577,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047333028560961733,
      "loss": 5.0728,
      "mean_token_accuracy": 0.1930759146809578,
      "num_tokens": 36384195.0,
      "step": 20965
    },
    {
      "entropy": 5.601344585418701,
      "epoch": 1.6315502820462946,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047331719558846687,
      "loss": 5.0766,
      "mean_token_accuracy": 0.1924204096198082,
      "num_tokens": 36393001.0,
      "step": 20970
    },
    {
      "entropy": 5.617854309082031,
      "epoch": 1.631939311418012,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047330410255816863,
      "loss": 5.0251,
      "mean_token_accuracy": 0.19239550083875656,
      "num_tokens": 36401674.0,
      "step": 20975
    },
    {
      "entropy": 5.552043342590332,
      "epoch": 1.6323283407897295,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047329100651892123,
      "loss": 5.0005,
      "mean_token_accuracy": 0.2087088033556938,
      "num_tokens": 36409817.0,
      "step": 20980
    },
    {
      "entropy": 5.584095191955567,
      "epoch": 1.6327173701614472,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004732779074709234,
      "loss": 4.9755,
      "mean_token_accuracy": 0.2001844972372055,
      "num_tokens": 36417541.0,
      "step": 20985
    },
    {
      "entropy": 5.5701454162597654,
      "epoch": 1.6331063995331647,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000473264805414374,
      "loss": 5.0786,
      "mean_token_accuracy": 0.1908455029129982,
      "num_tokens": 36426613.0,
      "step": 20990
    },
    {
      "entropy": 5.604685258865357,
      "epoch": 1.6334954289048822,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047325170034947167,
      "loss": 5.0843,
      "mean_token_accuracy": 0.19605784863233566,
      "num_tokens": 36436052.0,
      "step": 20995
    },
    {
      "entropy": 5.573258686065674,
      "epoch": 1.6338844582766,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047323859227641537,
      "loss": 5.0006,
      "mean_token_accuracy": 0.2038291722536087,
      "num_tokens": 36444386.0,
      "step": 21000
    },
    {
      "epoch": 1.6338844582766,
      "eval_entropy": 5.405573300057816,
      "eval_loss": 5.110541820526123,
      "eval_mean_token_accuracy": 0.20385344626957458,
      "eval_num_tokens": 36444386.0,
      "eval_runtime": 21.8671,
      "eval_samples_per_second": 1900.479,
      "eval_steps_per_second": 237.571,
      "step": 21000
    },
    {
      "entropy": 5.641724395751953,
      "epoch": 1.6342734876483176,
      "grad_norm": 1.125,
      "learning_rate": 0.00047322548119540397,
      "loss": 5.0429,
      "mean_token_accuracy": 0.19551848322153093,
      "num_tokens": 36452759.0,
      "step": 21005
    },
    {
      "entropy": 5.623278856277466,
      "epoch": 1.634662517020035,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047321236710663636,
      "loss": 4.992,
      "mean_token_accuracy": 0.19861437380313873,
      "num_tokens": 36460581.0,
      "step": 21010
    },
    {
      "entropy": 5.576382350921631,
      "epoch": 1.6350515463917525,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047319925001031165,
      "loss": 5.0331,
      "mean_token_accuracy": 0.2002054825425148,
      "num_tokens": 36468995.0,
      "step": 21015
    },
    {
      "entropy": 5.660507345199585,
      "epoch": 1.6354405757634702,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004731861299066287,
      "loss": 5.0945,
      "mean_token_accuracy": 0.19583260118961335,
      "num_tokens": 36477901.0,
      "step": 21020
    },
    {
      "entropy": 5.635953283309936,
      "epoch": 1.6358296051351877,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004731730067957867,
      "loss": 5.0988,
      "mean_token_accuracy": 0.1976763814687729,
      "num_tokens": 36486500.0,
      "step": 21025
    },
    {
      "entropy": 5.5837701797485355,
      "epoch": 1.6362186345069052,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047315988067798476,
      "loss": 5.0331,
      "mean_token_accuracy": 0.19408315420150757,
      "num_tokens": 36495247.0,
      "step": 21030
    },
    {
      "entropy": 5.6036810874938965,
      "epoch": 1.6366076638786229,
      "grad_norm": 1.171875,
      "learning_rate": 0.000473146751553422,
      "loss": 4.997,
      "mean_token_accuracy": 0.2110765278339386,
      "num_tokens": 36503633.0,
      "step": 21035
    },
    {
      "entropy": 5.626317262649536,
      "epoch": 1.6369966932503404,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004731336194222978,
      "loss": 5.0548,
      "mean_token_accuracy": 0.19802806079387664,
      "num_tokens": 36512138.0,
      "step": 21040
    },
    {
      "entropy": 5.631261491775513,
      "epoch": 1.6373857226220578,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047312048428481114,
      "loss": 5.04,
      "mean_token_accuracy": 0.20009175986051558,
      "num_tokens": 36521029.0,
      "step": 21045
    },
    {
      "entropy": 5.555566024780274,
      "epoch": 1.6377747519937755,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047310734614116153,
      "loss": 4.9764,
      "mean_token_accuracy": 0.2042452722787857,
      "num_tokens": 36530658.0,
      "step": 21050
    },
    {
      "entropy": 5.569853210449219,
      "epoch": 1.6381637813654932,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004730942049915483,
      "loss": 4.9689,
      "mean_token_accuracy": 0.19934993386268615,
      "num_tokens": 36539037.0,
      "step": 21055
    },
    {
      "entropy": 5.587121486663818,
      "epoch": 1.6385528107372107,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004730810608361707,
      "loss": 5.0531,
      "mean_token_accuracy": 0.20214777886867524,
      "num_tokens": 36547863.0,
      "step": 21060
    },
    {
      "entropy": 5.631935930252075,
      "epoch": 1.6389418401089282,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047306791367522833,
      "loss": 5.0729,
      "mean_token_accuracy": 0.1970382958650589,
      "num_tokens": 36555896.0,
      "step": 21065
    },
    {
      "entropy": 5.567766094207764,
      "epoch": 1.6393308694806459,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004730547635089206,
      "loss": 5.0068,
      "mean_token_accuracy": 0.20656665414571762,
      "num_tokens": 36564088.0,
      "step": 21070
    },
    {
      "entropy": 5.600091791152954,
      "epoch": 1.6397198988523634,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047304161033744714,
      "loss": 4.9747,
      "mean_token_accuracy": 0.20273981392383575,
      "num_tokens": 36572490.0,
      "step": 21075
    },
    {
      "entropy": 5.627602386474609,
      "epoch": 1.6401089282240808,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004730284541610075,
      "loss": 5.0496,
      "mean_token_accuracy": 0.19069170504808425,
      "num_tokens": 36580834.0,
      "step": 21080
    },
    {
      "entropy": 5.576250600814819,
      "epoch": 1.6404979575957985,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004730152949798012,
      "loss": 5.0207,
      "mean_token_accuracy": 0.20018489956855773,
      "num_tokens": 36589420.0,
      "step": 21085
    },
    {
      "entropy": 5.540608263015747,
      "epoch": 1.640886986967516,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047300213279402787,
      "loss": 5.0317,
      "mean_token_accuracy": 0.1982778176665306,
      "num_tokens": 36598419.0,
      "step": 21090
    },
    {
      "entropy": 5.580012083053589,
      "epoch": 1.6412760163392335,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047298896760388745,
      "loss": 5.0488,
      "mean_token_accuracy": 0.19693606346845627,
      "num_tokens": 36607367.0,
      "step": 21095
    },
    {
      "entropy": 5.641748380661011,
      "epoch": 1.6416650457109512,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004729757994095796,
      "loss": 5.1343,
      "mean_token_accuracy": 0.19228486120700836,
      "num_tokens": 36616197.0,
      "step": 21100
    },
    {
      "entropy": 5.663556671142578,
      "epoch": 1.6420540750826689,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047296262821130405,
      "loss": 5.1743,
      "mean_token_accuracy": 0.1908399745821953,
      "num_tokens": 36625452.0,
      "step": 21105
    },
    {
      "entropy": 5.608483505249024,
      "epoch": 1.6424431044543863,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047294945400926065,
      "loss": 5.1074,
      "mean_token_accuracy": 0.19364217817783355,
      "num_tokens": 36634356.0,
      "step": 21110
    },
    {
      "entropy": 5.6074357509613035,
      "epoch": 1.6428321338261038,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004729362768036494,
      "loss": 5.0578,
      "mean_token_accuracy": 0.19648833870887755,
      "num_tokens": 36642548.0,
      "step": 21115
    },
    {
      "entropy": 5.595019769668579,
      "epoch": 1.6432211631978215,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047292309659467024,
      "loss": 4.9667,
      "mean_token_accuracy": 0.21047018468379974,
      "num_tokens": 36650548.0,
      "step": 21120
    },
    {
      "entropy": 5.533414316177368,
      "epoch": 1.643610192569539,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004729099133825231,
      "loss": 4.9941,
      "mean_token_accuracy": 0.20174986720085145,
      "num_tokens": 36659322.0,
      "step": 21125
    },
    {
      "entropy": 5.557530164718628,
      "epoch": 1.6439992219412565,
      "grad_norm": 1.078125,
      "learning_rate": 0.000472896727167408,
      "loss": 4.9816,
      "mean_token_accuracy": 0.20263081341981887,
      "num_tokens": 36667699.0,
      "step": 21130
    },
    {
      "entropy": 5.66042652130127,
      "epoch": 1.6443882513129742,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047288353794952505,
      "loss": 5.1073,
      "mean_token_accuracy": 0.19515934735536575,
      "num_tokens": 36677842.0,
      "step": 21135
    },
    {
      "entropy": 5.548838520050049,
      "epoch": 1.6447772806846916,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004728703457290744,
      "loss": 4.9196,
      "mean_token_accuracy": 0.20510536879301072,
      "num_tokens": 36686488.0,
      "step": 21140
    },
    {
      "entropy": 5.554946994781494,
      "epoch": 1.6451663100564091,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004728571505062562,
      "loss": 5.0813,
      "mean_token_accuracy": 0.19801685959100723,
      "num_tokens": 36694955.0,
      "step": 21145
    },
    {
      "entropy": 5.524270248413086,
      "epoch": 1.6455553394281268,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004728439522812707,
      "loss": 4.9784,
      "mean_token_accuracy": 0.20682270526885987,
      "num_tokens": 36703481.0,
      "step": 21150
    },
    {
      "entropy": 5.653072452545166,
      "epoch": 1.6459443687998445,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047283075105431806,
      "loss": 5.1802,
      "mean_token_accuracy": 0.19131255000829697,
      "num_tokens": 36711812.0,
      "step": 21155
    },
    {
      "entropy": 5.629106426239014,
      "epoch": 1.646333398171562,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004728175468255987,
      "loss": 5.0343,
      "mean_token_accuracy": 0.20296893864870072,
      "num_tokens": 36720496.0,
      "step": 21160
    },
    {
      "entropy": 5.591406583786011,
      "epoch": 1.6467224275432795,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047280433959531287,
      "loss": 4.9187,
      "mean_token_accuracy": 0.2038732051849365,
      "num_tokens": 36728959.0,
      "step": 21165
    },
    {
      "entropy": 5.546081686019898,
      "epoch": 1.6471114569149972,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000472791129363661,
      "loss": 5.0187,
      "mean_token_accuracy": 0.19617637246847153,
      "num_tokens": 36737492.0,
      "step": 21170
    },
    {
      "entropy": 5.607508707046509,
      "epoch": 1.6475004862867146,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004727779161308436,
      "loss": 4.9811,
      "mean_token_accuracy": 0.2019236907362938,
      "num_tokens": 36745852.0,
      "step": 21175
    },
    {
      "entropy": 5.624356126785278,
      "epoch": 1.6478895156584321,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004727646998970612,
      "loss": 5.0531,
      "mean_token_accuracy": 0.1960201531648636,
      "num_tokens": 36753897.0,
      "step": 21180
    },
    {
      "entropy": 5.566350078582763,
      "epoch": 1.6482785450301498,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047275148066251417,
      "loss": 4.99,
      "mean_token_accuracy": 0.20034620612859727,
      "num_tokens": 36762737.0,
      "step": 21185
    },
    {
      "entropy": 5.649991178512574,
      "epoch": 1.6486675744018673,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004727382584274032,
      "loss": 5.0918,
      "mean_token_accuracy": 0.19357527643442154,
      "num_tokens": 36770742.0,
      "step": 21190
    },
    {
      "entropy": 5.686472654342651,
      "epoch": 1.6490566037735848,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047272503319192887,
      "loss": 5.1543,
      "mean_token_accuracy": 0.1870390221476555,
      "num_tokens": 36779420.0,
      "step": 21195
    },
    {
      "entropy": 5.579587936401367,
      "epoch": 1.6494456331453025,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004727118049562919,
      "loss": 4.9751,
      "mean_token_accuracy": 0.1996660366654396,
      "num_tokens": 36787888.0,
      "step": 21200
    },
    {
      "entropy": 5.530983638763428,
      "epoch": 1.6498346625170202,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000472698573720693,
      "loss": 4.9156,
      "mean_token_accuracy": 0.2062940552830696,
      "num_tokens": 36797131.0,
      "step": 21205
    },
    {
      "entropy": 5.5598304271698,
      "epoch": 1.6502236918887376,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004726853394853329,
      "loss": 5.0853,
      "mean_token_accuracy": 0.19564976245164872,
      "num_tokens": 36805429.0,
      "step": 21210
    },
    {
      "entropy": 5.596779155731201,
      "epoch": 1.650612721260455,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047267210225041234,
      "loss": 5.1382,
      "mean_token_accuracy": 0.18940980285406112,
      "num_tokens": 36814204.0,
      "step": 21215
    },
    {
      "entropy": 5.692402362823486,
      "epoch": 1.6510017506321728,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047265886201613234,
      "loss": 5.0897,
      "mean_token_accuracy": 0.19122100621461868,
      "num_tokens": 36823229.0,
      "step": 21220
    },
    {
      "entropy": 5.674637174606323,
      "epoch": 1.6513907800038903,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047264561878269365,
      "loss": 5.1638,
      "mean_token_accuracy": 0.19215769469738006,
      "num_tokens": 36832252.0,
      "step": 21225
    },
    {
      "entropy": 5.5992189884185795,
      "epoch": 1.6517798093756078,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004726323725502973,
      "loss": 5.0285,
      "mean_token_accuracy": 0.19869712591171265,
      "num_tokens": 36841456.0,
      "step": 21230
    },
    {
      "entropy": 5.625344753265381,
      "epoch": 1.6521688387473255,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004726191233191443,
      "loss": 4.9997,
      "mean_token_accuracy": 0.20772081464529038,
      "num_tokens": 36849717.0,
      "step": 21235
    },
    {
      "entropy": 5.640730142593384,
      "epoch": 1.6525578681190431,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004726058710894357,
      "loss": 5.0778,
      "mean_token_accuracy": 0.1952358067035675,
      "num_tokens": 36858222.0,
      "step": 21240
    },
    {
      "entropy": 5.567594194412232,
      "epoch": 1.6529468974907604,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004725926158613724,
      "loss": 5.0808,
      "mean_token_accuracy": 0.1902633234858513,
      "num_tokens": 36866899.0,
      "step": 21245
    },
    {
      "entropy": 5.626479959487915,
      "epoch": 1.653335926862478,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004725793576351557,
      "loss": 5.0414,
      "mean_token_accuracy": 0.20733904093503952,
      "num_tokens": 36876124.0,
      "step": 21250
    },
    {
      "entropy": 5.648554086685181,
      "epoch": 1.6537249562341958,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004725660964109867,
      "loss": 5.0986,
      "mean_token_accuracy": 0.19696181863546372,
      "num_tokens": 36884689.0,
      "step": 21255
    },
    {
      "entropy": 5.632748079299927,
      "epoch": 1.6541139856059133,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047255283218906673,
      "loss": 5.0523,
      "mean_token_accuracy": 0.1899647369980812,
      "num_tokens": 36893073.0,
      "step": 21260
    },
    {
      "entropy": 5.637504005432129,
      "epoch": 1.6545030149776307,
      "grad_norm": 1.046875,
      "learning_rate": 0.000472539564969597,
      "loss": 5.068,
      "mean_token_accuracy": 0.1918485000729561,
      "num_tokens": 36902428.0,
      "step": 21265
    },
    {
      "entropy": 5.65430588722229,
      "epoch": 1.6548920443493484,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047252629475277863,
      "loss": 5.0889,
      "mean_token_accuracy": 0.20354003757238387,
      "num_tokens": 36910297.0,
      "step": 21270
    },
    {
      "entropy": 5.619797039031982,
      "epoch": 1.655281073721066,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004725130215388132,
      "loss": 5.0304,
      "mean_token_accuracy": 0.20086280554533004,
      "num_tokens": 36919115.0,
      "step": 21275
    },
    {
      "entropy": 5.627178287506103,
      "epoch": 1.6556701030927834,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004724997453279021,
      "loss": 5.064,
      "mean_token_accuracy": 0.1941582292318344,
      "num_tokens": 36928419.0,
      "step": 21280
    },
    {
      "entropy": 5.6247961044311525,
      "epoch": 1.656059132464501,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004724864661202467,
      "loss": 5.013,
      "mean_token_accuracy": 0.19971290677785875,
      "num_tokens": 36936511.0,
      "step": 21285
    },
    {
      "entropy": 5.649073696136474,
      "epoch": 1.6564481618362188,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047247318391604846,
      "loss": 5.0763,
      "mean_token_accuracy": 0.19819322675466539,
      "num_tokens": 36945316.0,
      "step": 21290
    },
    {
      "entropy": 5.647952461242676,
      "epoch": 1.656837191207936,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047245989871550897,
      "loss": 5.1028,
      "mean_token_accuracy": 0.18941883444786073,
      "num_tokens": 36954140.0,
      "step": 21295
    },
    {
      "entropy": 5.598030042648316,
      "epoch": 1.6572262205796537,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004724466105188299,
      "loss": 5.0254,
      "mean_token_accuracy": 0.1976059243083,
      "num_tokens": 36962852.0,
      "step": 21300
    },
    {
      "entropy": 5.665014028549194,
      "epoch": 1.6576152499513714,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047243331932621263,
      "loss": 5.0646,
      "mean_token_accuracy": 0.19564741104841232,
      "num_tokens": 36971617.0,
      "step": 21305
    },
    {
      "entropy": 5.611910104751587,
      "epoch": 1.658004279323089,
      "grad_norm": 1.125,
      "learning_rate": 0.0004724200251378591,
      "loss": 5.0874,
      "mean_token_accuracy": 0.19729828089475632,
      "num_tokens": 36981078.0,
      "step": 21310
    },
    {
      "entropy": 5.61009521484375,
      "epoch": 1.6583933086948064,
      "grad_norm": 1.125,
      "learning_rate": 0.0004724067279539709,
      "loss": 4.9952,
      "mean_token_accuracy": 0.19736217260360717,
      "num_tokens": 36989860.0,
      "step": 21315
    },
    {
      "entropy": 5.611889934539795,
      "epoch": 1.658782338066524,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047239342777474975,
      "loss": 5.0201,
      "mean_token_accuracy": 0.20354832708835602,
      "num_tokens": 36998037.0,
      "step": 21320
    },
    {
      "entropy": 5.667009019851685,
      "epoch": 1.6591713674382416,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047238012460039765,
      "loss": 5.0641,
      "mean_token_accuracy": 0.19685031473636627,
      "num_tokens": 37006375.0,
      "step": 21325
    },
    {
      "entropy": 5.615655946731567,
      "epoch": 1.659560396809959,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004723668184311162,
      "loss": 5.125,
      "mean_token_accuracy": 0.18976180553436278,
      "num_tokens": 37014650.0,
      "step": 21330
    },
    {
      "entropy": 5.586802339553833,
      "epoch": 1.6599494261816767,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004723535092671074,
      "loss": 5.0766,
      "mean_token_accuracy": 0.18870789110660552,
      "num_tokens": 37023147.0,
      "step": 21335
    },
    {
      "entropy": 5.67778754234314,
      "epoch": 1.6603384555533944,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047234019710857335,
      "loss": 5.0726,
      "mean_token_accuracy": 0.19449642449617385,
      "num_tokens": 37031461.0,
      "step": 21340
    },
    {
      "entropy": 5.683797788619995,
      "epoch": 1.6607274849251117,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004723268819557158,
      "loss": 5.1136,
      "mean_token_accuracy": 0.19374279528856278,
      "num_tokens": 37039924.0,
      "step": 21345
    },
    {
      "entropy": 5.583129644393921,
      "epoch": 1.6611165142968294,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047231356380873687,
      "loss": 5.0344,
      "mean_token_accuracy": 0.19770515710115433,
      "num_tokens": 37048631.0,
      "step": 21350
    },
    {
      "entropy": 5.580436706542969,
      "epoch": 1.661505543668547,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004723002426678388,
      "loss": 5.0193,
      "mean_token_accuracy": 0.20376285761594773,
      "num_tokens": 37057018.0,
      "step": 21355
    },
    {
      "entropy": 5.636303234100342,
      "epoch": 1.6618945730402646,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004722869185332235,
      "loss": 5.0503,
      "mean_token_accuracy": 0.19928723573684692,
      "num_tokens": 37065762.0,
      "step": 21360
    },
    {
      "entropy": 5.587451934814453,
      "epoch": 1.662283602411982,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047227359140509324,
      "loss": 5.0441,
      "mean_token_accuracy": 0.19854637682437898,
      "num_tokens": 37073836.0,
      "step": 21365
    },
    {
      "entropy": 5.601717948913574,
      "epoch": 1.6626726317836997,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004722602612836501,
      "loss": 5.102,
      "mean_token_accuracy": 0.19312269240617752,
      "num_tokens": 37082788.0,
      "step": 21370
    },
    {
      "entropy": 5.671463680267334,
      "epoch": 1.6630616611554172,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047224692816909665,
      "loss": 5.1401,
      "mean_token_accuracy": 0.1904866576194763,
      "num_tokens": 37091228.0,
      "step": 21375
    },
    {
      "entropy": 5.651086807250977,
      "epoch": 1.6634506905271347,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004722335920616349,
      "loss": 5.0307,
      "mean_token_accuracy": 0.1992301568388939,
      "num_tokens": 37099819.0,
      "step": 21380
    },
    {
      "entropy": 5.602416324615478,
      "epoch": 1.6638397198988524,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047222025296146745,
      "loss": 5.0868,
      "mean_token_accuracy": 0.18942280411720275,
      "num_tokens": 37108129.0,
      "step": 21385
    },
    {
      "entropy": 5.562743854522705,
      "epoch": 1.66422874927057,
      "grad_norm": 1.125,
      "learning_rate": 0.00047220691086879643,
      "loss": 4.9712,
      "mean_token_accuracy": 0.2057488217949867,
      "num_tokens": 37116093.0,
      "step": 21390
    },
    {
      "entropy": 5.560021591186524,
      "epoch": 1.6646177786422873,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004721935657838245,
      "loss": 5.0347,
      "mean_token_accuracy": 0.19663818031549454,
      "num_tokens": 37124098.0,
      "step": 21395
    },
    {
      "entropy": 5.664295625686646,
      "epoch": 1.665006808014005,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000472180217706754,
      "loss": 5.1037,
      "mean_token_accuracy": 0.18548205345869065,
      "num_tokens": 37132608.0,
      "step": 21400
    },
    {
      "entropy": 5.645784187316894,
      "epoch": 1.6653958373857227,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004721668666377876,
      "loss": 5.0983,
      "mean_token_accuracy": 0.19707471877336502,
      "num_tokens": 37141365.0,
      "step": 21405
    },
    {
      "entropy": 5.5506734371185305,
      "epoch": 1.6657848667574402,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047215351257712783,
      "loss": 4.9823,
      "mean_token_accuracy": 0.20428500026464463,
      "num_tokens": 37150146.0,
      "step": 21410
    },
    {
      "entropy": 5.6480506420135494,
      "epoch": 1.6661738961291577,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047214015552497735,
      "loss": 5.1603,
      "mean_token_accuracy": 0.18554372042417527,
      "num_tokens": 37158887.0,
      "step": 21415
    },
    {
      "entropy": 5.660572099685669,
      "epoch": 1.6665629255008754,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047212679548153867,
      "loss": 5.0091,
      "mean_token_accuracy": 0.19732916504144668,
      "num_tokens": 37167902.0,
      "step": 21420
    },
    {
      "entropy": 5.678050994873047,
      "epoch": 1.6669519548725928,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047211343244701475,
      "loss": 5.1379,
      "mean_token_accuracy": 0.19153306633234024,
      "num_tokens": 37175853.0,
      "step": 21425
    },
    {
      "entropy": 5.594776773452759,
      "epoch": 1.6673409842443103,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004721000664216081,
      "loss": 5.0273,
      "mean_token_accuracy": 0.20073224157094954,
      "num_tokens": 37184036.0,
      "step": 21430
    },
    {
      "entropy": 5.630748558044433,
      "epoch": 1.667730013616028,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004720866974055219,
      "loss": 5.0867,
      "mean_token_accuracy": 0.19522957354784012,
      "num_tokens": 37193277.0,
      "step": 21435
    },
    {
      "entropy": 5.63545069694519,
      "epoch": 1.6681190429877457,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004720733253989585,
      "loss": 5.0414,
      "mean_token_accuracy": 0.20022166669368743,
      "num_tokens": 37201925.0,
      "step": 21440
    },
    {
      "entropy": 5.656448698043823,
      "epoch": 1.6685080723594632,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047205995040212114,
      "loss": 5.1284,
      "mean_token_accuracy": 0.1892026036977768,
      "num_tokens": 37210552.0,
      "step": 21445
    },
    {
      "entropy": 5.6526281356811525,
      "epoch": 1.6688971017311807,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004720465724152127,
      "loss": 5.0649,
      "mean_token_accuracy": 0.19977136105298995,
      "num_tokens": 37219001.0,
      "step": 21450
    },
    {
      "entropy": 5.601140308380127,
      "epoch": 1.6692861311028984,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047203319143843613,
      "loss": 5.0008,
      "mean_token_accuracy": 0.20496486723423005,
      "num_tokens": 37227635.0,
      "step": 21455
    },
    {
      "entropy": 5.545358276367187,
      "epoch": 1.6696751604746158,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004720198074719946,
      "loss": 5.0452,
      "mean_token_accuracy": 0.19702990502119064,
      "num_tokens": 37236872.0,
      "step": 21460
    },
    {
      "entropy": 5.6623540878295895,
      "epoch": 1.6700641898463333,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047200642051609094,
      "loss": 5.0999,
      "mean_token_accuracy": 0.19148070067167283,
      "num_tokens": 37245026.0,
      "step": 21465
    },
    {
      "entropy": 5.643272542953492,
      "epoch": 1.670453219218051,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004719930305709285,
      "loss": 5.0613,
      "mean_token_accuracy": 0.1968997076153755,
      "num_tokens": 37253295.0,
      "step": 21470
    },
    {
      "entropy": 5.61349573135376,
      "epoch": 1.6708422485897685,
      "grad_norm": 1.125,
      "learning_rate": 0.00047197963763671033,
      "loss": 5.0141,
      "mean_token_accuracy": 0.2089267000555992,
      "num_tokens": 37262785.0,
      "step": 21475
    },
    {
      "entropy": 5.593543910980225,
      "epoch": 1.671231277961486,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004719662417136397,
      "loss": 5.1072,
      "mean_token_accuracy": 0.19766231030225753,
      "num_tokens": 37271937.0,
      "step": 21480
    },
    {
      "entropy": 5.613740873336792,
      "epoch": 1.6716203073332037,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047195284280191987,
      "loss": 5.002,
      "mean_token_accuracy": 0.2016749456524849,
      "num_tokens": 37280861.0,
      "step": 21485
    },
    {
      "entropy": 5.662140941619873,
      "epoch": 1.6720093367049214,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004719394409017541,
      "loss": 5.0401,
      "mean_token_accuracy": 0.20485589802265167,
      "num_tokens": 37289235.0,
      "step": 21490
    },
    {
      "entropy": 5.598332357406616,
      "epoch": 1.6723983660766388,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047192603601334585,
      "loss": 5.043,
      "mean_token_accuracy": 0.19603785127401352,
      "num_tokens": 37297778.0,
      "step": 21495
    },
    {
      "entropy": 5.5919841766357425,
      "epoch": 1.6727873954483563,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047191262813689835,
      "loss": 5.0686,
      "mean_token_accuracy": 0.19123001247644425,
      "num_tokens": 37306960.0,
      "step": 21500
    },
    {
      "entropy": 5.648209857940674,
      "epoch": 1.673176424820074,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004718992172726152,
      "loss": 5.0787,
      "mean_token_accuracy": 0.19498798847198487,
      "num_tokens": 37316050.0,
      "step": 21505
    },
    {
      "entropy": 5.5621373653411865,
      "epoch": 1.6735654541917915,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047188580342069983,
      "loss": 5.0475,
      "mean_token_accuracy": 0.2012204870581627,
      "num_tokens": 37325058.0,
      "step": 21510
    },
    {
      "entropy": 5.662901163101196,
      "epoch": 1.673954483563509,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004718723865813557,
      "loss": 5.1306,
      "mean_token_accuracy": 0.19502254575490952,
      "num_tokens": 37333512.0,
      "step": 21515
    },
    {
      "entropy": 5.620169067382813,
      "epoch": 1.6743435129352267,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004718589667547865,
      "loss": 5.021,
      "mean_token_accuracy": 0.19990231096744537,
      "num_tokens": 37341695.0,
      "step": 21520
    },
    {
      "entropy": 5.693024444580078,
      "epoch": 1.6747325423069441,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004718455439411958,
      "loss": 5.1885,
      "mean_token_accuracy": 0.18235407918691635,
      "num_tokens": 37350917.0,
      "step": 21525
    },
    {
      "entropy": 5.681368017196656,
      "epoch": 1.6751215716786616,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004718321181407873,
      "loss": 5.1613,
      "mean_token_accuracy": 0.18665122389793395,
      "num_tokens": 37360206.0,
      "step": 21530
    },
    {
      "entropy": 5.704204511642456,
      "epoch": 1.6755106010503793,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004718186893537647,
      "loss": 5.13,
      "mean_token_accuracy": 0.19519478380680083,
      "num_tokens": 37369378.0,
      "step": 21535
    },
    {
      "entropy": 5.634978246688843,
      "epoch": 1.675899630422097,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004718052575803318,
      "loss": 5.0169,
      "mean_token_accuracy": 0.2059697464108467,
      "num_tokens": 37377744.0,
      "step": 21540
    },
    {
      "entropy": 5.648777008056641,
      "epoch": 1.6762886597938145,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004717918228206923,
      "loss": 5.0243,
      "mean_token_accuracy": 0.19610974788665772,
      "num_tokens": 37387097.0,
      "step": 21545
    },
    {
      "entropy": 5.658017873764038,
      "epoch": 1.676677689165532,
      "grad_norm": 1.125,
      "learning_rate": 0.0004717783850750501,
      "loss": 5.1301,
      "mean_token_accuracy": 0.19374706000089645,
      "num_tokens": 37395551.0,
      "step": 21550
    },
    {
      "entropy": 5.595207357406617,
      "epoch": 1.6770667185372496,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004717649443436091,
      "loss": 4.9058,
      "mean_token_accuracy": 0.20657152086496353,
      "num_tokens": 37403876.0,
      "step": 21555
    },
    {
      "entropy": 5.589859294891357,
      "epoch": 1.6774557479089671,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047175150062657336,
      "loss": 5.0393,
      "mean_token_accuracy": 0.19815760254859924,
      "num_tokens": 37412658.0,
      "step": 21560
    },
    {
      "entropy": 5.586010837554932,
      "epoch": 1.6778447772806846,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047173805392414664,
      "loss": 5.0212,
      "mean_token_accuracy": 0.19506628066301346,
      "num_tokens": 37420790.0,
      "step": 21565
    },
    {
      "entropy": 5.6234354972839355,
      "epoch": 1.6782338066524023,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004717246042365332,
      "loss": 5.0759,
      "mean_token_accuracy": 0.19952067136764526,
      "num_tokens": 37428866.0,
      "step": 21570
    },
    {
      "entropy": 5.598287534713745,
      "epoch": 1.6786228360241198,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047171115156393695,
      "loss": 5.0119,
      "mean_token_accuracy": 0.20278420448303222,
      "num_tokens": 37437636.0,
      "step": 21575
    },
    {
      "entropy": 5.625813770294189,
      "epoch": 1.6790118653958372,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047169769590656207,
      "loss": 5.0618,
      "mean_token_accuracy": 0.19751233756542205,
      "num_tokens": 37446455.0,
      "step": 21580
    },
    {
      "entropy": 5.611213350296021,
      "epoch": 1.679400894767555,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047168423726461273,
      "loss": 5.1464,
      "mean_token_accuracy": 0.19106802195310593,
      "num_tokens": 37454997.0,
      "step": 21585
    },
    {
      "entropy": 5.586915349960327,
      "epoch": 1.6797899241392726,
      "grad_norm": 1.125,
      "learning_rate": 0.0004716707756382931,
      "loss": 4.9644,
      "mean_token_accuracy": 0.20687023997306825,
      "num_tokens": 37463701.0,
      "step": 21590
    },
    {
      "entropy": 5.646985816955566,
      "epoch": 1.6801789535109901,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047165731102780757,
      "loss": 5.1183,
      "mean_token_accuracy": 0.19007940888404845,
      "num_tokens": 37472460.0,
      "step": 21595
    },
    {
      "entropy": 5.622792673110962,
      "epoch": 1.6805679828827076,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047164384343336037,
      "loss": 5.0163,
      "mean_token_accuracy": 0.20255867391824722,
      "num_tokens": 37481114.0,
      "step": 21600
    },
    {
      "entropy": 5.620723676681519,
      "epoch": 1.6809570122544253,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047163037285515583,
      "loss": 5.0974,
      "mean_token_accuracy": 0.18801950365304948,
      "num_tokens": 37489593.0,
      "step": 21605
    },
    {
      "entropy": 5.675616264343262,
      "epoch": 1.6813460416261428,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004716168992933982,
      "loss": 5.1408,
      "mean_token_accuracy": 0.19550462812185287,
      "num_tokens": 37498044.0,
      "step": 21610
    },
    {
      "entropy": 5.703295516967773,
      "epoch": 1.6817350709978602,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047160342274829234,
      "loss": 5.0446,
      "mean_token_accuracy": 0.19380800127983094,
      "num_tokens": 37507339.0,
      "step": 21615
    },
    {
      "entropy": 5.554900169372559,
      "epoch": 1.682124100369578,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047158994322004223,
      "loss": 4.9398,
      "mean_token_accuracy": 0.2054002657532692,
      "num_tokens": 37515919.0,
      "step": 21620
    },
    {
      "entropy": 5.5543115615844725,
      "epoch": 1.6825131297412956,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004715764607088527,
      "loss": 4.952,
      "mean_token_accuracy": 0.21057135611772537,
      "num_tokens": 37524990.0,
      "step": 21625
    },
    {
      "entropy": 5.567415237426758,
      "epoch": 1.6829021591130129,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004715629752149283,
      "loss": 5.0755,
      "mean_token_accuracy": 0.1948564350605011,
      "num_tokens": 37534125.0,
      "step": 21630
    },
    {
      "entropy": 5.583069229125977,
      "epoch": 1.6832911884847306,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047154948673847356,
      "loss": 4.9904,
      "mean_token_accuracy": 0.19955350756645202,
      "num_tokens": 37541715.0,
      "step": 21635
    },
    {
      "entropy": 5.679392147064209,
      "epoch": 1.6836802178564483,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004715359952796933,
      "loss": 5.132,
      "mean_token_accuracy": 0.19678231626749038,
      "num_tokens": 37550097.0,
      "step": 21640
    },
    {
      "entropy": 5.532448196411133,
      "epoch": 1.6840692472281658,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047152250083879203,
      "loss": 4.9981,
      "mean_token_accuracy": 0.20245244354009628,
      "num_tokens": 37558950.0,
      "step": 21645
    },
    {
      "entropy": 5.618267154693603,
      "epoch": 1.6844582765998832,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047150900341597464,
      "loss": 5.0198,
      "mean_token_accuracy": 0.20316656529903412,
      "num_tokens": 37567737.0,
      "step": 21650
    },
    {
      "entropy": 5.636998796463013,
      "epoch": 1.684847305971601,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004714955030114459,
      "loss": 5.066,
      "mean_token_accuracy": 0.19484193325042726,
      "num_tokens": 37576649.0,
      "step": 21655
    },
    {
      "entropy": 5.613708543777466,
      "epoch": 1.6852363353433184,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047148199962541065,
      "loss": 5.0289,
      "mean_token_accuracy": 0.20365013778209687,
      "num_tokens": 37584455.0,
      "step": 21660
    },
    {
      "entropy": 5.625494956970215,
      "epoch": 1.6856253647150359,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004714684932580738,
      "loss": 5.1289,
      "mean_token_accuracy": 0.1881230130791664,
      "num_tokens": 37593653.0,
      "step": 21665
    },
    {
      "entropy": 5.632204008102417,
      "epoch": 1.6860143940867536,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004714549839096403,
      "loss": 5.0836,
      "mean_token_accuracy": 0.19123605042695999,
      "num_tokens": 37603212.0,
      "step": 21670
    },
    {
      "entropy": 5.646163606643677,
      "epoch": 1.6864034234584713,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047144147158031507,
      "loss": 5.0132,
      "mean_token_accuracy": 0.19662821143865586,
      "num_tokens": 37611871.0,
      "step": 21675
    },
    {
      "entropy": 5.681102180480957,
      "epoch": 1.6867924528301885,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004714279562703032,
      "loss": 5.1991,
      "mean_token_accuracy": 0.18252745866775513,
      "num_tokens": 37621410.0,
      "step": 21680
    },
    {
      "entropy": 5.605821990966797,
      "epoch": 1.6871814822019062,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004714144379798098,
      "loss": 5.0094,
      "mean_token_accuracy": 0.1992748886346817,
      "num_tokens": 37629525.0,
      "step": 21685
    },
    {
      "entropy": 5.668889093399048,
      "epoch": 1.687570511573624,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047140091670903986,
      "loss": 5.0683,
      "mean_token_accuracy": 0.19742657244205475,
      "num_tokens": 37638183.0,
      "step": 21690
    },
    {
      "entropy": 5.631321573257447,
      "epoch": 1.6879595409453414,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047138739245819857,
      "loss": 5.0515,
      "mean_token_accuracy": 0.19697533696889877,
      "num_tokens": 37646884.0,
      "step": 21695
    },
    {
      "entropy": 5.69326376914978,
      "epoch": 1.6883485703170589,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047137386522749124,
      "loss": 5.0952,
      "mean_token_accuracy": 0.19137585908174515,
      "num_tokens": 37655482.0,
      "step": 21700
    },
    {
      "entropy": 5.5896728992462155,
      "epoch": 1.6887375996887766,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004713603350171231,
      "loss": 4.9125,
      "mean_token_accuracy": 0.20493291467428207,
      "num_tokens": 37664056.0,
      "step": 21705
    },
    {
      "entropy": 5.63419919013977,
      "epoch": 1.689126629060494,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047134680182729926,
      "loss": 5.2219,
      "mean_token_accuracy": 0.1882893204689026,
      "num_tokens": 37673198.0,
      "step": 21710
    },
    {
      "entropy": 5.583607006072998,
      "epoch": 1.6895156584322115,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004713332656582254,
      "loss": 5.0473,
      "mean_token_accuracy": 0.195000259578228,
      "num_tokens": 37682660.0,
      "step": 21715
    },
    {
      "entropy": 5.583115577697754,
      "epoch": 1.6899046878039292,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004713197265101066,
      "loss": 5.0122,
      "mean_token_accuracy": 0.20022477954626083,
      "num_tokens": 37691232.0,
      "step": 21720
    },
    {
      "entropy": 5.613849210739136,
      "epoch": 1.690293717175647,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047130618438314856,
      "loss": 4.9723,
      "mean_token_accuracy": 0.20278759002685548,
      "num_tokens": 37699734.0,
      "step": 21725
    },
    {
      "entropy": 5.620531797409058,
      "epoch": 1.6906827465473642,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004712926392775565,
      "loss": 5.0704,
      "mean_token_accuracy": 0.2015253409743309,
      "num_tokens": 37708529.0,
      "step": 21730
    },
    {
      "entropy": 5.583433151245117,
      "epoch": 1.6910717759190819,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004712790911935362,
      "loss": 4.9423,
      "mean_token_accuracy": 0.20783183872699737,
      "num_tokens": 37717197.0,
      "step": 21735
    },
    {
      "entropy": 5.580004405975342,
      "epoch": 1.6914608052907996,
      "grad_norm": 1.125,
      "learning_rate": 0.000471265540131293,
      "loss": 5.0724,
      "mean_token_accuracy": 0.194974385201931,
      "num_tokens": 37726276.0,
      "step": 21740
    },
    {
      "entropy": 5.6626749515533445,
      "epoch": 1.691849834662517,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047125198609103267,
      "loss": 5.118,
      "mean_token_accuracy": 0.19248749166727067,
      "num_tokens": 37735016.0,
      "step": 21745
    },
    {
      "entropy": 5.669690656661987,
      "epoch": 1.6922388640342345,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004712384290729607,
      "loss": 5.0845,
      "mean_token_accuracy": 0.2006140187382698,
      "num_tokens": 37744332.0,
      "step": 21750
    },
    {
      "entropy": 5.696141719818115,
      "epoch": 1.6926278934059522,
      "grad_norm": 1.15625,
      "learning_rate": 0.000471224869077283,
      "loss": 5.1338,
      "mean_token_accuracy": 0.18522909432649612,
      "num_tokens": 37752805.0,
      "step": 21755
    },
    {
      "entropy": 5.67636923789978,
      "epoch": 1.6930169227776697,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047121130610420527,
      "loss": 5.0618,
      "mean_token_accuracy": 0.19436305463314058,
      "num_tokens": 37760979.0,
      "step": 21760
    },
    {
      "entropy": 5.555144500732422,
      "epoch": 1.6934059521493872,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047119774015393323,
      "loss": 5.0009,
      "mean_token_accuracy": 0.19952594637870788,
      "num_tokens": 37769096.0,
      "step": 21765
    },
    {
      "entropy": 5.603054094314575,
      "epoch": 1.6937949815211049,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047118417122667285,
      "loss": 5.0132,
      "mean_token_accuracy": 0.19807154238224028,
      "num_tokens": 37776998.0,
      "step": 21770
    },
    {
      "entropy": 5.654562091827392,
      "epoch": 1.6941840108928226,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004711705993226298,
      "loss": 5.1268,
      "mean_token_accuracy": 0.19315582066774367,
      "num_tokens": 37785547.0,
      "step": 21775
    },
    {
      "entropy": 5.604728555679321,
      "epoch": 1.6945730402645398,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004711570244420102,
      "loss": 5.088,
      "mean_token_accuracy": 0.19292886853218078,
      "num_tokens": 37793659.0,
      "step": 21780
    },
    {
      "entropy": 5.620196199417114,
      "epoch": 1.6949620696362575,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004711434465850199,
      "loss": 4.9949,
      "mean_token_accuracy": 0.200394643843174,
      "num_tokens": 37802282.0,
      "step": 21785
    },
    {
      "entropy": 5.659757423400879,
      "epoch": 1.6953510990079752,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004711298657518651,
      "loss": 5.0643,
      "mean_token_accuracy": 0.1961624428629875,
      "num_tokens": 37810692.0,
      "step": 21790
    },
    {
      "entropy": 5.566381406784058,
      "epoch": 1.6957401283796927,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004711162819427518,
      "loss": 5.0086,
      "mean_token_accuracy": 0.2052112728357315,
      "num_tokens": 37819264.0,
      "step": 21795
    },
    {
      "entropy": 5.607199668884277,
      "epoch": 1.6961291577514102,
      "grad_norm": 1.15625,
      "learning_rate": 0.000471102695157886,
      "loss": 5.0465,
      "mean_token_accuracy": 0.20193013846874236,
      "num_tokens": 37827650.0,
      "step": 21800
    },
    {
      "entropy": 5.542062997817993,
      "epoch": 1.6965181871231279,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004710891053974739,
      "loss": 4.9991,
      "mean_token_accuracy": 0.20087489038705825,
      "num_tokens": 37835683.0,
      "step": 21805
    },
    {
      "entropy": 5.620373392105103,
      "epoch": 1.6969072164948453,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004710755126617217,
      "loss": 5.0033,
      "mean_token_accuracy": 0.19744803458452226,
      "num_tokens": 37844334.0,
      "step": 21810
    },
    {
      "entropy": 5.586679172515869,
      "epoch": 1.6972962458665628,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004710619169508358,
      "loss": 4.9842,
      "mean_token_accuracy": 0.1974950209259987,
      "num_tokens": 37853609.0,
      "step": 21815
    },
    {
      "entropy": 5.619667339324951,
      "epoch": 1.6976852752382805,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004710483182650223,
      "loss": 5.0405,
      "mean_token_accuracy": 0.19896610975265502,
      "num_tokens": 37862817.0,
      "step": 21820
    },
    {
      "entropy": 5.5404700756073,
      "epoch": 1.6980743046099982,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047103471660448767,
      "loss": 4.9883,
      "mean_token_accuracy": 0.20614997446537017,
      "num_tokens": 37871630.0,
      "step": 21825
    },
    {
      "entropy": 5.509577751159668,
      "epoch": 1.6984633339817157,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047102111196943813,
      "loss": 4.9808,
      "mean_token_accuracy": 0.20710501521825792,
      "num_tokens": 37881098.0,
      "step": 21830
    },
    {
      "entropy": 5.613763427734375,
      "epoch": 1.6988523633534331,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047100750436008035,
      "loss": 5.0026,
      "mean_token_accuracy": 0.20495725274086,
      "num_tokens": 37889760.0,
      "step": 21835
    },
    {
      "entropy": 5.680889892578125,
      "epoch": 1.6992413927251508,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047099389377662054,
      "loss": 5.1114,
      "mean_token_accuracy": 0.19521128088235856,
      "num_tokens": 37898417.0,
      "step": 21840
    },
    {
      "entropy": 5.576249027252198,
      "epoch": 1.6996304220968683,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004709802802192654,
      "loss": 5.0677,
      "mean_token_accuracy": 0.19635892063379287,
      "num_tokens": 37907117.0,
      "step": 21845
    },
    {
      "entropy": 5.559094667434692,
      "epoch": 1.7000194514685858,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047096666368822153,
      "loss": 4.9965,
      "mean_token_accuracy": 0.19987713992595674,
      "num_tokens": 37915781.0,
      "step": 21850
    },
    {
      "entropy": 5.626416110992432,
      "epoch": 1.7004084808403035,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047095304418369536,
      "loss": 5.0789,
      "mean_token_accuracy": 0.1932198151946068,
      "num_tokens": 37924770.0,
      "step": 21855
    },
    {
      "entropy": 5.628818845748901,
      "epoch": 1.700797510212021,
      "grad_norm": 1.125,
      "learning_rate": 0.0004709394217058936,
      "loss": 5.0954,
      "mean_token_accuracy": 0.19124641567468642,
      "num_tokens": 37933512.0,
      "step": 21860
    },
    {
      "entropy": 5.558410787582398,
      "epoch": 1.7011865395837384,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004709257962550231,
      "loss": 4.9185,
      "mean_token_accuracy": 0.2027181029319763,
      "num_tokens": 37942439.0,
      "step": 21865
    },
    {
      "entropy": 5.637639331817627,
      "epoch": 1.7015755689554561,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047091216783129035,
      "loss": 5.1046,
      "mean_token_accuracy": 0.1973971724510193,
      "num_tokens": 37951388.0,
      "step": 21870
    },
    {
      "entropy": 5.652983617782593,
      "epoch": 1.7019645983271738,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004708985364349024,
      "loss": 5.0958,
      "mean_token_accuracy": 0.19401235729455948,
      "num_tokens": 37960372.0,
      "step": 21875
    },
    {
      "entropy": 5.687587404251099,
      "epoch": 1.7023536276988913,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047088490206606586,
      "loss": 5.1953,
      "mean_token_accuracy": 0.18200006932020188,
      "num_tokens": 37969472.0,
      "step": 21880
    },
    {
      "entropy": 5.575918674468994,
      "epoch": 1.7027426570706088,
      "grad_norm": 1.25,
      "learning_rate": 0.00047087126472498785,
      "loss": 4.9135,
      "mean_token_accuracy": 0.20755328834056855,
      "num_tokens": 37977422.0,
      "step": 21885
    },
    {
      "entropy": 5.542408132553101,
      "epoch": 1.7031316864423265,
      "grad_norm": 1.203125,
      "learning_rate": 0.000470857624411875,
      "loss": 4.8893,
      "mean_token_accuracy": 0.20672637820243836,
      "num_tokens": 37985861.0,
      "step": 21890
    },
    {
      "entropy": 5.615883255004883,
      "epoch": 1.703520715814044,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047084398112693456,
      "loss": 5.0159,
      "mean_token_accuracy": 0.202946974337101,
      "num_tokens": 37994535.0,
      "step": 21895
    },
    {
      "entropy": 5.5972161293029785,
      "epoch": 1.7039097451857614,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004708303348703734,
      "loss": 5.0655,
      "mean_token_accuracy": 0.19671754240989686,
      "num_tokens": 38003351.0,
      "step": 21900
    },
    {
      "entropy": 5.6489684104919435,
      "epoch": 1.7042987745574791,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004708166856423986,
      "loss": 5.0582,
      "mean_token_accuracy": 0.19544261693954468,
      "num_tokens": 38012210.0,
      "step": 21905
    },
    {
      "entropy": 5.652815628051758,
      "epoch": 1.7046878039291966,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047080303344321727,
      "loss": 5.0482,
      "mean_token_accuracy": 0.20523087829351425,
      "num_tokens": 38020548.0,
      "step": 21910
    },
    {
      "entropy": 5.6231691360473635,
      "epoch": 1.705076833300914,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047078937827303653,
      "loss": 5.0241,
      "mean_token_accuracy": 0.19584515392780305,
      "num_tokens": 38029451.0,
      "step": 21915
    },
    {
      "entropy": 5.601893568038941,
      "epoch": 1.7054658626726318,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004707757201320636,
      "loss": 5.0473,
      "mean_token_accuracy": 0.19684240370988845,
      "num_tokens": 38037821.0,
      "step": 21920
    },
    {
      "entropy": 5.590384435653687,
      "epoch": 1.7058548920443495,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047076205902050577,
      "loss": 5.1079,
      "mean_token_accuracy": 0.1883827954530716,
      "num_tokens": 38046188.0,
      "step": 21925
    },
    {
      "entropy": 5.633674240112304,
      "epoch": 1.706243921416067,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047074839493857024,
      "loss": 5.1132,
      "mean_token_accuracy": 0.19555674344301224,
      "num_tokens": 38054821.0,
      "step": 21930
    },
    {
      "entropy": 5.619605541229248,
      "epoch": 1.7066329507877844,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004707347278864644,
      "loss": 5.0242,
      "mean_token_accuracy": 0.20205829590559005,
      "num_tokens": 38063057.0,
      "step": 21935
    },
    {
      "entropy": 5.673763370513916,
      "epoch": 1.7070219801595021,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047072105786439563,
      "loss": 5.1902,
      "mean_token_accuracy": 0.1913291707634926,
      "num_tokens": 38071694.0,
      "step": 21940
    },
    {
      "entropy": 5.6164937019348145,
      "epoch": 1.7074110095312196,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047070738487257137,
      "loss": 5.0014,
      "mean_token_accuracy": 0.19819054007530212,
      "num_tokens": 38080965.0,
      "step": 21945
    },
    {
      "entropy": 5.635694885253907,
      "epoch": 1.707800038902937,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047069370891119895,
      "loss": 5.034,
      "mean_token_accuracy": 0.19620207846164703,
      "num_tokens": 38089786.0,
      "step": 21950
    },
    {
      "entropy": 5.614178657531738,
      "epoch": 1.7081890682746548,
      "grad_norm": 1.125,
      "learning_rate": 0.000470680029980486,
      "loss": 5.0305,
      "mean_token_accuracy": 0.19973579049110413,
      "num_tokens": 38097994.0,
      "step": 21955
    },
    {
      "entropy": 5.6129923343658445,
      "epoch": 1.7085780976463723,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047066634808064006,
      "loss": 5.1008,
      "mean_token_accuracy": 0.1921934649348259,
      "num_tokens": 38106910.0,
      "step": 21960
    },
    {
      "entropy": 5.588732481002808,
      "epoch": 1.7089671270180897,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047065266321186873,
      "loss": 5.0103,
      "mean_token_accuracy": 0.2006385311484337,
      "num_tokens": 38115162.0,
      "step": 21965
    },
    {
      "entropy": 5.658104467391968,
      "epoch": 1.7093561563898074,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004706389753743797,
      "loss": 5.0712,
      "mean_token_accuracy": 0.20191887021064758,
      "num_tokens": 38123776.0,
      "step": 21970
    },
    {
      "entropy": 5.654714059829712,
      "epoch": 1.7097451857615251,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004706252845683805,
      "loss": 5.164,
      "mean_token_accuracy": 0.18981072306632996,
      "num_tokens": 38132591.0,
      "step": 21975
    },
    {
      "entropy": 5.596533203125,
      "epoch": 1.7101342151332426,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047061159079407903,
      "loss": 5.0073,
      "mean_token_accuracy": 0.20352110415697097,
      "num_tokens": 38141092.0,
      "step": 21980
    },
    {
      "entropy": 5.598656892776489,
      "epoch": 1.71052324450496,
      "grad_norm": 1.171875,
      "learning_rate": 0.000470597894051683,
      "loss": 4.9891,
      "mean_token_accuracy": 0.2021651491522789,
      "num_tokens": 38149315.0,
      "step": 21985
    },
    {
      "entropy": 5.560493278503418,
      "epoch": 1.7109122738766778,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004705841943414003,
      "loss": 4.9577,
      "mean_token_accuracy": 0.1955380395054817,
      "num_tokens": 38158208.0,
      "step": 21990
    },
    {
      "entropy": 5.735902452468872,
      "epoch": 1.7113013032483952,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047057049166343876,
      "loss": 5.1085,
      "mean_token_accuracy": 0.1928066521883011,
      "num_tokens": 38166339.0,
      "step": 21995
    },
    {
      "entropy": 5.6498880863189695,
      "epoch": 1.7116903326201127,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047055678601800623,
      "loss": 5.0647,
      "mean_token_accuracy": 0.19326005280017852,
      "num_tokens": 38174813.0,
      "step": 22000
    },
    {
      "entropy": 5.56307282447815,
      "epoch": 1.7120793619918304,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047054307740531076,
      "loss": 4.9517,
      "mean_token_accuracy": 0.20143619626760484,
      "num_tokens": 38183440.0,
      "step": 22005
    },
    {
      "entropy": 5.543004035949707,
      "epoch": 1.712468391363548,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047052936582556035,
      "loss": 4.9982,
      "mean_token_accuracy": 0.19977577328681945,
      "num_tokens": 38192510.0,
      "step": 22010
    },
    {
      "entropy": 5.603354358673096,
      "epoch": 1.7128574207352654,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047051565127896307,
      "loss": 5.0019,
      "mean_token_accuracy": 0.20090843141078948,
      "num_tokens": 38200854.0,
      "step": 22015
    },
    {
      "entropy": 5.630614805221557,
      "epoch": 1.713246450106983,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047050193376572686,
      "loss": 5.0503,
      "mean_token_accuracy": 0.19628504812717437,
      "num_tokens": 38209632.0,
      "step": 22020
    },
    {
      "entropy": 5.55157585144043,
      "epoch": 1.7136354794787008,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047048821328606,
      "loss": 4.981,
      "mean_token_accuracy": 0.20299442261457443,
      "num_tokens": 38218657.0,
      "step": 22025
    },
    {
      "entropy": 5.664433288574219,
      "epoch": 1.7140245088504182,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004704744898401708,
      "loss": 5.0542,
      "mean_token_accuracy": 0.19763325154781342,
      "num_tokens": 38227507.0,
      "step": 22030
    },
    {
      "entropy": 5.643199300765991,
      "epoch": 1.7144135382221357,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004704607634282672,
      "loss": 5.014,
      "mean_token_accuracy": 0.19684745520353317,
      "num_tokens": 38236822.0,
      "step": 22035
    },
    {
      "entropy": 5.502330636978149,
      "epoch": 1.7148025675938534,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047044703405055765,
      "loss": 4.9449,
      "mean_token_accuracy": 0.20614244192838668,
      "num_tokens": 38245144.0,
      "step": 22040
    },
    {
      "entropy": 5.570398902893066,
      "epoch": 1.7151915969655709,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047043330170725046,
      "loss": 5.0183,
      "mean_token_accuracy": 0.19679562300443648,
      "num_tokens": 38254363.0,
      "step": 22045
    },
    {
      "entropy": 5.579965877532959,
      "epoch": 1.7155806263372884,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047041956639855406,
      "loss": 5.014,
      "mean_token_accuracy": 0.19745832681655884,
      "num_tokens": 38263117.0,
      "step": 22050
    },
    {
      "entropy": 5.55317440032959,
      "epoch": 1.715969655709006,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004704058281246766,
      "loss": 5.0093,
      "mean_token_accuracy": 0.196611750125885,
      "num_tokens": 38271217.0,
      "step": 22055
    },
    {
      "entropy": 5.670649766921997,
      "epoch": 1.7163586850807238,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004703920868858268,
      "loss": 5.1362,
      "mean_token_accuracy": 0.1912349730730057,
      "num_tokens": 38279789.0,
      "step": 22060
    },
    {
      "entropy": 5.614987230300903,
      "epoch": 1.716747714452441,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047037834268221315,
      "loss": 4.9647,
      "mean_token_accuracy": 0.2047826960682869,
      "num_tokens": 38288845.0,
      "step": 22065
    },
    {
      "entropy": 5.626345539093018,
      "epoch": 1.7171367438241587,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004703645955140441,
      "loss": 5.0186,
      "mean_token_accuracy": 0.1981974720954895,
      "num_tokens": 38297237.0,
      "step": 22070
    },
    {
      "entropy": 5.677994108200073,
      "epoch": 1.7175257731958764,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047035084538152815,
      "loss": 5.1366,
      "mean_token_accuracy": 0.18971950113773345,
      "num_tokens": 38306582.0,
      "step": 22075
    },
    {
      "entropy": 5.621893739700317,
      "epoch": 1.7179148025675939,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004703370922848742,
      "loss": 4.958,
      "mean_token_accuracy": 0.21022734940052032,
      "num_tokens": 38315406.0,
      "step": 22080
    },
    {
      "entropy": 5.561059379577637,
      "epoch": 1.7183038319393114,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047032333622429074,
      "loss": 4.9466,
      "mean_token_accuracy": 0.20267179012298583,
      "num_tokens": 38324271.0,
      "step": 22085
    },
    {
      "entropy": 5.612152099609375,
      "epoch": 1.718692861311029,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047030957719998656,
      "loss": 5.0584,
      "mean_token_accuracy": 0.1959070384502411,
      "num_tokens": 38332831.0,
      "step": 22090
    },
    {
      "entropy": 5.596252298355102,
      "epoch": 1.7190818906827465,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004702958152121704,
      "loss": 4.9724,
      "mean_token_accuracy": 0.20565009415149688,
      "num_tokens": 38341263.0,
      "step": 22095
    },
    {
      "entropy": 5.61143307685852,
      "epoch": 1.719470920054464,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004702820502610511,
      "loss": 5.1574,
      "mean_token_accuracy": 0.18411071449518204,
      "num_tokens": 38349890.0,
      "step": 22100
    },
    {
      "entropy": 5.562587451934815,
      "epoch": 1.7198599494261817,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004702682823468375,
      "loss": 4.94,
      "mean_token_accuracy": 0.20408442467451096,
      "num_tokens": 38358071.0,
      "step": 22105
    },
    {
      "entropy": 5.648678302764893,
      "epoch": 1.7202489787978994,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047025451146973844,
      "loss": 5.1091,
      "mean_token_accuracy": 0.18887461721897125,
      "num_tokens": 38366859.0,
      "step": 22110
    },
    {
      "entropy": 5.742717123031616,
      "epoch": 1.7206380081696167,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047024073762996305,
      "loss": 5.2601,
      "mean_token_accuracy": 0.1797910898923874,
      "num_tokens": 38375897.0,
      "step": 22115
    },
    {
      "entropy": 5.61982307434082,
      "epoch": 1.7210270375413343,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004702269608277202,
      "loss": 5.026,
      "mean_token_accuracy": 0.20267996639013292,
      "num_tokens": 38384700.0,
      "step": 22120
    },
    {
      "entropy": 5.615942144393921,
      "epoch": 1.721416066913052,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004702131810632189,
      "loss": 5.0957,
      "mean_token_accuracy": 0.19702455997467042,
      "num_tokens": 38393212.0,
      "step": 22125
    },
    {
      "entropy": 5.571683359146118,
      "epoch": 1.7218050962847695,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047019939833666837,
      "loss": 5.0059,
      "mean_token_accuracy": 0.20378904044628143,
      "num_tokens": 38401572.0,
      "step": 22130
    },
    {
      "entropy": 5.5830878734588625,
      "epoch": 1.722194125656487,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004701856126482776,
      "loss": 5.0231,
      "mean_token_accuracy": 0.20075735747814177,
      "num_tokens": 38410669.0,
      "step": 22135
    },
    {
      "entropy": 5.62599401473999,
      "epoch": 1.7225831550282047,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004701718239982559,
      "loss": 5.0386,
      "mean_token_accuracy": 0.19647755473852158,
      "num_tokens": 38419038.0,
      "step": 22140
    },
    {
      "entropy": 5.576609897613525,
      "epoch": 1.7229721843999222,
      "grad_norm": 1.125,
      "learning_rate": 0.00047015803238681234,
      "loss": 4.9709,
      "mean_token_accuracy": 0.20945372730493544,
      "num_tokens": 38427314.0,
      "step": 22145
    },
    {
      "entropy": 5.5433111667633055,
      "epoch": 1.7233612137716396,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004701442378141563,
      "loss": 4.9738,
      "mean_token_accuracy": 0.20084090977907182,
      "num_tokens": 38435373.0,
      "step": 22150
    },
    {
      "entropy": 5.55924916267395,
      "epoch": 1.7237502431433573,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000470130440280497,
      "loss": 4.9962,
      "mean_token_accuracy": 0.20518484115600585,
      "num_tokens": 38444448.0,
      "step": 22155
    },
    {
      "entropy": 5.626925134658814,
      "epoch": 1.724139272515075,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004701166397860439,
      "loss": 4.956,
      "mean_token_accuracy": 0.19871326237916948,
      "num_tokens": 38453576.0,
      "step": 22160
    },
    {
      "entropy": 5.635059022903443,
      "epoch": 1.7245283018867923,
      "grad_norm": 1.125,
      "learning_rate": 0.0004701028363310064,
      "loss": 4.9954,
      "mean_token_accuracy": 0.204918210208416,
      "num_tokens": 38462848.0,
      "step": 22165
    },
    {
      "entropy": 5.545461130142212,
      "epoch": 1.72491733125851,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047008902991559385,
      "loss": 4.9798,
      "mean_token_accuracy": 0.20151522308588027,
      "num_tokens": 38470953.0,
      "step": 22170
    },
    {
      "entropy": 5.64134111404419,
      "epoch": 1.7253063606302277,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004700752205400158,
      "loss": 5.0995,
      "mean_token_accuracy": 0.18773148357868194,
      "num_tokens": 38479835.0,
      "step": 22175
    },
    {
      "entropy": 5.6186822891235355,
      "epoch": 1.7256953900019452,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047006140820448174,
      "loss": 5.0461,
      "mean_token_accuracy": 0.19586482495069504,
      "num_tokens": 38488342.0,
      "step": 22180
    },
    {
      "entropy": 5.629008102416992,
      "epoch": 1.7260844193736626,
      "grad_norm": 1.125,
      "learning_rate": 0.00047004759290920136,
      "loss": 5.0464,
      "mean_token_accuracy": 0.19680440723896026,
      "num_tokens": 38496716.0,
      "step": 22185
    },
    {
      "entropy": 5.61487774848938,
      "epoch": 1.7264734487453803,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004700337746543841,
      "loss": 5.0626,
      "mean_token_accuracy": 0.19789986908435822,
      "num_tokens": 38506615.0,
      "step": 22190
    },
    {
      "entropy": 5.5246048927307125,
      "epoch": 1.7268624781170978,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004700199534402398,
      "loss": 4.882,
      "mean_token_accuracy": 0.20460295230150222,
      "num_tokens": 38515936.0,
      "step": 22195
    },
    {
      "entropy": 5.581892204284668,
      "epoch": 1.7272515074888153,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004700061292669781,
      "loss": 5.0132,
      "mean_token_accuracy": 0.19651895016431808,
      "num_tokens": 38524829.0,
      "step": 22200
    },
    {
      "entropy": 5.662121534347534,
      "epoch": 1.727640536860533,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004699923021348088,
      "loss": 5.1618,
      "mean_token_accuracy": 0.1871777206659317,
      "num_tokens": 38533686.0,
      "step": 22205
    },
    {
      "entropy": 5.671685075759887,
      "epoch": 1.7280295662322507,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046997847204394166,
      "loss": 5.0354,
      "mean_token_accuracy": 0.1999412328004837,
      "num_tokens": 38542449.0,
      "step": 22210
    },
    {
      "entropy": 5.61876745223999,
      "epoch": 1.728418595603968,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046996463899458653,
      "loss": 5.0026,
      "mean_token_accuracy": 0.19994359463453293,
      "num_tokens": 38550888.0,
      "step": 22215
    },
    {
      "entropy": 5.5458001613616945,
      "epoch": 1.7288076249756856,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004699508029869533,
      "loss": 5.0408,
      "mean_token_accuracy": 0.20018828213214873,
      "num_tokens": 38559366.0,
      "step": 22220
    },
    {
      "entropy": 5.643072414398193,
      "epoch": 1.7291966543474033,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046993696402125205,
      "loss": 5.0744,
      "mean_token_accuracy": 0.1992821589112282,
      "num_tokens": 38567080.0,
      "step": 22225
    },
    {
      "entropy": 5.60291051864624,
      "epoch": 1.7295856837191208,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004699231220976925,
      "loss": 5.0648,
      "mean_token_accuracy": 0.20158005952835084,
      "num_tokens": 38575520.0,
      "step": 22230
    },
    {
      "entropy": 5.659439945220948,
      "epoch": 1.7299747130908383,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004699092772164849,
      "loss": 5.1272,
      "mean_token_accuracy": 0.18940088301897048,
      "num_tokens": 38583974.0,
      "step": 22235
    },
    {
      "entropy": 5.559523344039917,
      "epoch": 1.730363742462556,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004698954293778392,
      "loss": 5.0374,
      "mean_token_accuracy": 0.19782306104898453,
      "num_tokens": 38592677.0,
      "step": 22240
    },
    {
      "entropy": 5.585797691345215,
      "epoch": 1.7307527718342735,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046988157858196555,
      "loss": 4.8879,
      "mean_token_accuracy": 0.2065796971321106,
      "num_tokens": 38601342.0,
      "step": 22245
    },
    {
      "entropy": 5.675569343566894,
      "epoch": 1.731141801205991,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046986772482907426,
      "loss": 5.1455,
      "mean_token_accuracy": 0.18814289271831514,
      "num_tokens": 38610895.0,
      "step": 22250
    },
    {
      "entropy": 5.633636617660523,
      "epoch": 1.7315308305777086,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004698538681193754,
      "loss": 5.1021,
      "mean_token_accuracy": 0.19708590507507323,
      "num_tokens": 38619059.0,
      "step": 22255
    },
    {
      "entropy": 5.60432653427124,
      "epoch": 1.7319198599494263,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046984000845307907,
      "loss": 5.0577,
      "mean_token_accuracy": 0.19824582636356353,
      "num_tokens": 38628084.0,
      "step": 22260
    },
    {
      "entropy": 5.613573312759399,
      "epoch": 1.7323088893211438,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046982614583039584,
      "loss": 5.045,
      "mean_token_accuracy": 0.19509783238172532,
      "num_tokens": 38636205.0,
      "step": 22265
    },
    {
      "entropy": 5.585463523864746,
      "epoch": 1.7326979186928613,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004698122802515359,
      "loss": 5.0041,
      "mean_token_accuracy": 0.19653920084238052,
      "num_tokens": 38644583.0,
      "step": 22270
    },
    {
      "entropy": 5.584925603866577,
      "epoch": 1.733086948064579,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004697984117167097,
      "loss": 5.001,
      "mean_token_accuracy": 0.1998422622680664,
      "num_tokens": 38653517.0,
      "step": 22275
    },
    {
      "entropy": 5.599721956253052,
      "epoch": 1.7334759774362964,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046978454022612767,
      "loss": 5.0574,
      "mean_token_accuracy": 0.1968567743897438,
      "num_tokens": 38661926.0,
      "step": 22280
    },
    {
      "entropy": 5.709655094146728,
      "epoch": 1.733865006808014,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046977066578000025,
      "loss": 5.1027,
      "mean_token_accuracy": 0.20009998977184296,
      "num_tokens": 38671567.0,
      "step": 22285
    },
    {
      "entropy": 5.624595594406128,
      "epoch": 1.7342540361797316,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046975678837853806,
      "loss": 5.023,
      "mean_token_accuracy": 0.20285670310258866,
      "num_tokens": 38680439.0,
      "step": 22290
    },
    {
      "entropy": 5.599853134155273,
      "epoch": 1.734643065551449,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046974290802195156,
      "loss": 5.0765,
      "mean_token_accuracy": 0.19931709319353103,
      "num_tokens": 38689168.0,
      "step": 22295
    },
    {
      "entropy": 5.637978506088257,
      "epoch": 1.7350320949231666,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004697290247104513,
      "loss": 5.0113,
      "mean_token_accuracy": 0.19943537563085556,
      "num_tokens": 38697592.0,
      "step": 22300
    },
    {
      "entropy": 5.612720012664795,
      "epoch": 1.7354211242948843,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046971513844424814,
      "loss": 5.0607,
      "mean_token_accuracy": 0.19440296292304993,
      "num_tokens": 38705561.0,
      "step": 22305
    },
    {
      "entropy": 5.579755210876465,
      "epoch": 1.735810153666602,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046970124922355265,
      "loss": 5.0226,
      "mean_token_accuracy": 0.20079128444194794,
      "num_tokens": 38714694.0,
      "step": 22310
    },
    {
      "entropy": 5.622611045837402,
      "epoch": 1.7361991830383194,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004696873570485756,
      "loss": 5.0005,
      "mean_token_accuracy": 0.20074739754199983,
      "num_tokens": 38722797.0,
      "step": 22315
    },
    {
      "entropy": 5.639300155639648,
      "epoch": 1.736588212410037,
      "grad_norm": 1.125,
      "learning_rate": 0.0004696734619195278,
      "loss": 5.0789,
      "mean_token_accuracy": 0.1965402603149414,
      "num_tokens": 38732076.0,
      "step": 22320
    },
    {
      "entropy": 5.687313032150269,
      "epoch": 1.7369772417817546,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046965956383662,
      "loss": 5.0815,
      "mean_token_accuracy": 0.2024357348680496,
      "num_tokens": 38740992.0,
      "step": 22325
    },
    {
      "entropy": 5.645163869857788,
      "epoch": 1.737366271153472,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004696456628000632,
      "loss": 5.0479,
      "mean_token_accuracy": 0.197814379632473,
      "num_tokens": 38749266.0,
      "step": 22330
    },
    {
      "entropy": 5.611548900604248,
      "epoch": 1.7377553005251896,
      "grad_norm": 1.125,
      "learning_rate": 0.00046963175881006825,
      "loss": 5.1296,
      "mean_token_accuracy": 0.19693208187818528,
      "num_tokens": 38757983.0,
      "step": 22335
    },
    {
      "entropy": 5.602069282531739,
      "epoch": 1.7381443298969073,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004696178518668462,
      "loss": 4.9691,
      "mean_token_accuracy": 0.21298617422580718,
      "num_tokens": 38766258.0,
      "step": 22340
    },
    {
      "entropy": 5.623243141174316,
      "epoch": 1.7385333592686247,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046960394197060804,
      "loss": 5.0821,
      "mean_token_accuracy": 0.19330163300037384,
      "num_tokens": 38775105.0,
      "step": 22345
    },
    {
      "entropy": 5.6474607467651365,
      "epoch": 1.7389223886403422,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046959002912156473,
      "loss": 5.0943,
      "mean_token_accuracy": 0.1924824595451355,
      "num_tokens": 38783415.0,
      "step": 22350
    },
    {
      "entropy": 5.619459581375122,
      "epoch": 1.73931141801206,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004695761133199275,
      "loss": 5.0972,
      "mean_token_accuracy": 0.2004402458667755,
      "num_tokens": 38791703.0,
      "step": 22355
    },
    {
      "entropy": 5.628544330596924,
      "epoch": 1.7397004473837776,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004695621945659074,
      "loss": 5.0859,
      "mean_token_accuracy": 0.19474948197603226,
      "num_tokens": 38800823.0,
      "step": 22360
    },
    {
      "entropy": 5.619390535354614,
      "epoch": 1.740089476755495,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004695482728597157,
      "loss": 5.0653,
      "mean_token_accuracy": 0.19545395374298097,
      "num_tokens": 38810607.0,
      "step": 22365
    },
    {
      "entropy": 5.624842357635498,
      "epoch": 1.7404785061272126,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046953434820156363,
      "loss": 5.0307,
      "mean_token_accuracy": 0.1922731801867485,
      "num_tokens": 38819288.0,
      "step": 22370
    },
    {
      "entropy": 5.596852350234985,
      "epoch": 1.7408675354989303,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004695204205916624,
      "loss": 4.9746,
      "mean_token_accuracy": 0.2028667837381363,
      "num_tokens": 38827791.0,
      "step": 22375
    },
    {
      "entropy": 5.635534000396729,
      "epoch": 1.7412565648706477,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004695064900302235,
      "loss": 5.0248,
      "mean_token_accuracy": 0.2029082491993904,
      "num_tokens": 38836338.0,
      "step": 22380
    },
    {
      "entropy": 5.6671514987945555,
      "epoch": 1.7416455942423652,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004694925565174581,
      "loss": 4.987,
      "mean_token_accuracy": 0.1979934349656105,
      "num_tokens": 38844768.0,
      "step": 22385
    },
    {
      "entropy": 5.7007737159729,
      "epoch": 1.742034623614083,
      "grad_norm": 1.125,
      "learning_rate": 0.00046947862005357777,
      "loss": 5.1307,
      "mean_token_accuracy": 0.19242537766695023,
      "num_tokens": 38854022.0,
      "step": 22390
    },
    {
      "entropy": 5.585982704162598,
      "epoch": 1.7424236529858004,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004694646806387939,
      "loss": 4.9947,
      "mean_token_accuracy": 0.20192920416593552,
      "num_tokens": 38862035.0,
      "step": 22395
    },
    {
      "entropy": 5.607659101486206,
      "epoch": 1.7428126823575179,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004694507382733181,
      "loss": 5.0382,
      "mean_token_accuracy": 0.20076656192541123,
      "num_tokens": 38870762.0,
      "step": 22400
    },
    {
      "entropy": 5.5907834529876705,
      "epoch": 1.7432017117292355,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004694367929573618,
      "loss": 4.9888,
      "mean_token_accuracy": 0.2040885418653488,
      "num_tokens": 38879791.0,
      "step": 22405
    },
    {
      "entropy": 5.721454763412476,
      "epoch": 1.7435907411009532,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004694228446911367,
      "loss": 5.126,
      "mean_token_accuracy": 0.19045703411102294,
      "num_tokens": 38889926.0,
      "step": 22410
    },
    {
      "entropy": 5.670068216323853,
      "epoch": 1.7439797704726707,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004694088934748542,
      "loss": 5.0532,
      "mean_token_accuracy": 0.19268347024917604,
      "num_tokens": 38898428.0,
      "step": 22415
    },
    {
      "entropy": 5.580760955810547,
      "epoch": 1.7443687998443882,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004693949393087263,
      "loss": 5.0736,
      "mean_token_accuracy": 0.19253176748752593,
      "num_tokens": 38906800.0,
      "step": 22420
    },
    {
      "entropy": 5.595942640304566,
      "epoch": 1.744757829216106,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004693809821929647,
      "loss": 5.0442,
      "mean_token_accuracy": 0.19183948636054993,
      "num_tokens": 38915947.0,
      "step": 22425
    },
    {
      "entropy": 5.700219058990479,
      "epoch": 1.7451468585878234,
      "grad_norm": 1.078125,
      "learning_rate": 0.000469367022127781,
      "loss": 5.0902,
      "mean_token_accuracy": 0.19053598195314408,
      "num_tokens": 38924895.0,
      "step": 22430
    },
    {
      "entropy": 5.670387125015258,
      "epoch": 1.7455358879595408,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046935305911338703,
      "loss": 5.0708,
      "mean_token_accuracy": 0.19062747061252594,
      "num_tokens": 38933393.0,
      "step": 22435
    },
    {
      "entropy": 5.657281589508057,
      "epoch": 1.7459249173312585,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004693390931499948,
      "loss": 5.0595,
      "mean_token_accuracy": 0.20176222771406174,
      "num_tokens": 38942219.0,
      "step": 22440
    },
    {
      "entropy": 5.620717525482178,
      "epoch": 1.7463139467029762,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004693251242378162,
      "loss": 5.0062,
      "mean_token_accuracy": 0.1986298531293869,
      "num_tokens": 38950938.0,
      "step": 22445
    },
    {
      "entropy": 5.615544366836548,
      "epoch": 1.7467029760746935,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046931115237706304,
      "loss": 5.0354,
      "mean_token_accuracy": 0.19712853729724883,
      "num_tokens": 38959805.0,
      "step": 22450
    },
    {
      "entropy": 5.608055686950683,
      "epoch": 1.7470920054464112,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004692971775679475,
      "loss": 4.9661,
      "mean_token_accuracy": 0.2008902534842491,
      "num_tokens": 38968308.0,
      "step": 22455
    },
    {
      "entropy": 5.731195735931396,
      "epoch": 1.7474810348181289,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046928319981068154,
      "loss": 5.1088,
      "mean_token_accuracy": 0.1972878247499466,
      "num_tokens": 38976409.0,
      "step": 22460
    },
    {
      "entropy": 5.566067314147949,
      "epoch": 1.7478700641898464,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004692692191054773,
      "loss": 5.0382,
      "mean_token_accuracy": 0.1974053904414177,
      "num_tokens": 38985193.0,
      "step": 22465
    },
    {
      "entropy": 5.591451168060303,
      "epoch": 1.7482590935615638,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004692552354525468,
      "loss": 5.0266,
      "mean_token_accuracy": 0.20146097987890244,
      "num_tokens": 38994141.0,
      "step": 22470
    },
    {
      "entropy": 5.681043243408203,
      "epoch": 1.7486481229332815,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004692412488521023,
      "loss": 5.1729,
      "mean_token_accuracy": 0.1890342116355896,
      "num_tokens": 39002953.0,
      "step": 22475
    },
    {
      "entropy": 5.664262676239014,
      "epoch": 1.749037152304999,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000469227259304356,
      "loss": 5.0723,
      "mean_token_accuracy": 0.20161379128694534,
      "num_tokens": 39010672.0,
      "step": 22480
    },
    {
      "entropy": 5.622957324981689,
      "epoch": 1.7494261816767165,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046921326680952023,
      "loss": 5.1056,
      "mean_token_accuracy": 0.19200315475463867,
      "num_tokens": 39019677.0,
      "step": 22485
    },
    {
      "entropy": 5.651808166503907,
      "epoch": 1.7498152110484342,
      "grad_norm": 1.125,
      "learning_rate": 0.0004691992713678072,
      "loss": 5.0584,
      "mean_token_accuracy": 0.19479953348636628,
      "num_tokens": 39028165.0,
      "step": 22490
    },
    {
      "entropy": 5.588686513900757,
      "epoch": 1.7502042404201519,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004691852729794293,
      "loss": 4.9812,
      "mean_token_accuracy": 0.2045218303799629,
      "num_tokens": 39037537.0,
      "step": 22495
    },
    {
      "entropy": 5.597321319580078,
      "epoch": 1.7505932697918691,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004691712716445991,
      "loss": 5.0065,
      "mean_token_accuracy": 0.19972130805253982,
      "num_tokens": 39045861.0,
      "step": 22500
    },
    {
      "entropy": 5.665545845031739,
      "epoch": 1.7509822991635868,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004691572673635288,
      "loss": 5.0349,
      "mean_token_accuracy": 0.20176538676023484,
      "num_tokens": 39054526.0,
      "step": 22505
    },
    {
      "entropy": 5.626248407363891,
      "epoch": 1.7513713285353045,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000469143260136431,
      "loss": 5.0505,
      "mean_token_accuracy": 0.19129231721162795,
      "num_tokens": 39063296.0,
      "step": 22510
    },
    {
      "entropy": 5.607632255554199,
      "epoch": 1.751760357907022,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004691292499635183,
      "loss": 4.9743,
      "mean_token_accuracy": 0.19862579107284545,
      "num_tokens": 39071570.0,
      "step": 22515
    },
    {
      "entropy": 5.583535242080688,
      "epoch": 1.7521493872787395,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004691152368450032,
      "loss": 5.0705,
      "mean_token_accuracy": 0.20157907754182816,
      "num_tokens": 39080798.0,
      "step": 22520
    },
    {
      "entropy": 5.615162658691406,
      "epoch": 1.7525384166504572,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046910122078109835,
      "loss": 5.0365,
      "mean_token_accuracy": 0.19796667844057084,
      "num_tokens": 39089572.0,
      "step": 22525
    },
    {
      "entropy": 5.699023580551147,
      "epoch": 1.7529274460221747,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004690872017720163,
      "loss": 5.1322,
      "mean_token_accuracy": 0.19475738406181337,
      "num_tokens": 39098391.0,
      "step": 22530
    },
    {
      "entropy": 5.65471773147583,
      "epoch": 1.7533164753938921,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046907317981797006,
      "loss": 5.095,
      "mean_token_accuracy": 0.1988322228193283,
      "num_tokens": 39106892.0,
      "step": 22535
    },
    {
      "entropy": 5.6489990711212155,
      "epoch": 1.7537055047656098,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046905915491917213,
      "loss": 5.036,
      "mean_token_accuracy": 0.19630955904722214,
      "num_tokens": 39115635.0,
      "step": 22540
    },
    {
      "entropy": 5.634373521804809,
      "epoch": 1.7540945341373275,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004690451270758354,
      "loss": 5.0243,
      "mean_token_accuracy": 0.19801483899354935,
      "num_tokens": 39123665.0,
      "step": 22545
    },
    {
      "entropy": 5.673346042633057,
      "epoch": 1.7544835635090448,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046903109628817276,
      "loss": 5.0849,
      "mean_token_accuracy": 0.1909360945224762,
      "num_tokens": 39132727.0,
      "step": 22550
    },
    {
      "entropy": 5.651640892028809,
      "epoch": 1.7548725928807625,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046901706255639703,
      "loss": 5.063,
      "mean_token_accuracy": 0.1909584328532219,
      "num_tokens": 39140814.0,
      "step": 22555
    },
    {
      "entropy": 5.706284093856811,
      "epoch": 1.7552616222524802,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004690030258807212,
      "loss": 5.139,
      "mean_token_accuracy": 0.1881895422935486,
      "num_tokens": 39149877.0,
      "step": 22560
    },
    {
      "entropy": 5.614695262908936,
      "epoch": 1.7556506516241976,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046898898626135827,
      "loss": 5.0711,
      "mean_token_accuracy": 0.19244237691164018,
      "num_tokens": 39158698.0,
      "step": 22565
    },
    {
      "entropy": 5.572536993026733,
      "epoch": 1.7560396809959151,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004689749436985211,
      "loss": 4.983,
      "mean_token_accuracy": 0.20743250995874404,
      "num_tokens": 39167546.0,
      "step": 22570
    },
    {
      "entropy": 5.623037099838257,
      "epoch": 1.7564287103676328,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046896089819242304,
      "loss": 5.028,
      "mean_token_accuracy": 0.19573469310998917,
      "num_tokens": 39176547.0,
      "step": 22575
    },
    {
      "entropy": 5.702540493011474,
      "epoch": 1.7568177397393503,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004689468497432771,
      "loss": 5.0322,
      "mean_token_accuracy": 0.20036076456308366,
      "num_tokens": 39185273.0,
      "step": 22580
    },
    {
      "entropy": 5.668255710601807,
      "epoch": 1.7572067691110678,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004689327983512963,
      "loss": 5.0119,
      "mean_token_accuracy": 0.2043660968542099,
      "num_tokens": 39194204.0,
      "step": 22585
    },
    {
      "entropy": 5.615302515029907,
      "epoch": 1.7575957984827855,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046891874401669404,
      "loss": 4.9714,
      "mean_token_accuracy": 0.20304716974496842,
      "num_tokens": 39202951.0,
      "step": 22590
    },
    {
      "entropy": 5.524821186065674,
      "epoch": 1.7579848278545032,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004689046867396834,
      "loss": 5.0245,
      "mean_token_accuracy": 0.19657075256109238,
      "num_tokens": 39211508.0,
      "step": 22595
    },
    {
      "entropy": 5.67389702796936,
      "epoch": 1.7583738572262204,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004688906265204779,
      "loss": 5.0915,
      "mean_token_accuracy": 0.19024554044008254,
      "num_tokens": 39219804.0,
      "step": 22600
    },
    {
      "entropy": 5.71880259513855,
      "epoch": 1.7587628865979381,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004688765633592907,
      "loss": 5.115,
      "mean_token_accuracy": 0.19325451999902726,
      "num_tokens": 39228204.0,
      "step": 22605
    },
    {
      "entropy": 5.623468542098999,
      "epoch": 1.7591519159696558,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004688624972563352,
      "loss": 5.0399,
      "mean_token_accuracy": 0.1918053150177002,
      "num_tokens": 39237003.0,
      "step": 22610
    },
    {
      "entropy": 5.562715721130371,
      "epoch": 1.7595409453413733,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004688484282118249,
      "loss": 4.9983,
      "mean_token_accuracy": 0.19092916548252106,
      "num_tokens": 39245636.0,
      "step": 22615
    },
    {
      "entropy": 5.542520189285279,
      "epoch": 1.7599299747130908,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004688343562259732,
      "loss": 4.8806,
      "mean_token_accuracy": 0.20764474868774413,
      "num_tokens": 39254288.0,
      "step": 22620
    },
    {
      "entropy": 5.6226293563842775,
      "epoch": 1.7603190040848085,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004688202812989937,
      "loss": 5.0804,
      "mean_token_accuracy": 0.1971711367368698,
      "num_tokens": 39263217.0,
      "step": 22625
    },
    {
      "entropy": 5.56850118637085,
      "epoch": 1.760708033456526,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004688062034311,
      "loss": 5.0483,
      "mean_token_accuracy": 0.19963135868310927,
      "num_tokens": 39271240.0,
      "step": 22630
    },
    {
      "entropy": 5.5523176193237305,
      "epoch": 1.7610970628282434,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046879212262250546,
      "loss": 4.9714,
      "mean_token_accuracy": 0.20715888738632202,
      "num_tokens": 39279665.0,
      "step": 22635
    },
    {
      "entropy": 5.636352252960205,
      "epoch": 1.761486092199961,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046877803887342406,
      "loss": 5.1554,
      "mean_token_accuracy": 0.18949809074401855,
      "num_tokens": 39287975.0,
      "step": 22640
    },
    {
      "entropy": 5.658179044723511,
      "epoch": 1.7618751215716788,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004687639521840693,
      "loss": 5.0921,
      "mean_token_accuracy": 0.19391151815652846,
      "num_tokens": 39296541.0,
      "step": 22645
    },
    {
      "entropy": 5.643190860748291,
      "epoch": 1.7622641509433963,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046874986255465495,
      "loss": 4.9898,
      "mean_token_accuracy": 0.20430760979652404,
      "num_tokens": 39304710.0,
      "step": 22650
    },
    {
      "entropy": 5.60588231086731,
      "epoch": 1.7626531803151138,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046873576998539485,
      "loss": 4.9353,
      "mean_token_accuracy": 0.1998494029045105,
      "num_tokens": 39313236.0,
      "step": 22655
    },
    {
      "entropy": 5.550936317443847,
      "epoch": 1.7630422096868315,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004687216744765028,
      "loss": 4.8758,
      "mean_token_accuracy": 0.20712209939956666,
      "num_tokens": 39322656.0,
      "step": 22660
    },
    {
      "entropy": 5.587309169769287,
      "epoch": 1.763431239058549,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004687075760281927,
      "loss": 4.9638,
      "mean_token_accuracy": 0.20811061561107635,
      "num_tokens": 39331074.0,
      "step": 22665
    },
    {
      "entropy": 5.626010704040527,
      "epoch": 1.7638202684302664,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004686934746406784,
      "loss": 5.0313,
      "mean_token_accuracy": 0.19273855835199355,
      "num_tokens": 39339450.0,
      "step": 22670
    },
    {
      "entropy": 5.636268854141235,
      "epoch": 1.764209297801984,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046867937031417397,
      "loss": 5.1306,
      "mean_token_accuracy": 0.1921254202723503,
      "num_tokens": 39347270.0,
      "step": 22675
    },
    {
      "entropy": 5.58633828163147,
      "epoch": 1.7645983271737016,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004686652630488933,
      "loss": 5.0434,
      "mean_token_accuracy": 0.19059428423643113,
      "num_tokens": 39355853.0,
      "step": 22680
    },
    {
      "entropy": 5.559527015686035,
      "epoch": 1.764987356545419,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046865115284505063,
      "loss": 4.9501,
      "mean_token_accuracy": 0.21121643334627152,
      "num_tokens": 39364160.0,
      "step": 22685
    },
    {
      "entropy": 5.61551866531372,
      "epoch": 1.7653763859171367,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046863703970285986,
      "loss": 5.0413,
      "mean_token_accuracy": 0.20251805931329728,
      "num_tokens": 39372504.0,
      "step": 22690
    },
    {
      "entropy": 5.593826150894165,
      "epoch": 1.7657654152888544,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004686229236225352,
      "loss": 5.0234,
      "mean_token_accuracy": 0.20164479613304137,
      "num_tokens": 39380883.0,
      "step": 22695
    },
    {
      "entropy": 5.691967487335205,
      "epoch": 1.766154444660572,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004686088046042909,
      "loss": 5.096,
      "mean_token_accuracy": 0.1906644433736801,
      "num_tokens": 39389943.0,
      "step": 22700
    },
    {
      "entropy": 5.568107986450196,
      "epoch": 1.7665434740322894,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046859468264834114,
      "loss": 5.0396,
      "mean_token_accuracy": 0.20075453966856002,
      "num_tokens": 39398919.0,
      "step": 22705
    },
    {
      "entropy": 5.631412315368652,
      "epoch": 1.766932503404007,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046858055775490017,
      "loss": 5.1135,
      "mean_token_accuracy": 0.19490496963262557,
      "num_tokens": 39407547.0,
      "step": 22710
    },
    {
      "entropy": 5.595941543579102,
      "epoch": 1.7673215327757246,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004685664299241825,
      "loss": 4.9473,
      "mean_token_accuracy": 0.20129703283309935,
      "num_tokens": 39416264.0,
      "step": 22715
    },
    {
      "entropy": 5.604745817184448,
      "epoch": 1.767710562147442,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004685522991564022,
      "loss": 5.0167,
      "mean_token_accuracy": 0.19666850417852402,
      "num_tokens": 39424104.0,
      "step": 22720
    },
    {
      "entropy": 5.646996402740479,
      "epoch": 1.7680995915191597,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004685381654517738,
      "loss": 5.0607,
      "mean_token_accuracy": 0.19386076927185059,
      "num_tokens": 39432443.0,
      "step": 22725
    },
    {
      "entropy": 5.563793039321899,
      "epoch": 1.7684886208908772,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004685240288105119,
      "loss": 5.025,
      "mean_token_accuracy": 0.19606747925281526,
      "num_tokens": 39441620.0,
      "step": 22730
    },
    {
      "entropy": 5.585921907424927,
      "epoch": 1.7688776502625947,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046850988923283085,
      "loss": 5.0048,
      "mean_token_accuracy": 0.20375807583332062,
      "num_tokens": 39449854.0,
      "step": 22735
    },
    {
      "entropy": 5.607060241699219,
      "epoch": 1.7692666796343124,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046849574671894523,
      "loss": 5.0793,
      "mean_token_accuracy": 0.19520506411790847,
      "num_tokens": 39458496.0,
      "step": 22740
    },
    {
      "entropy": 5.6429595947265625,
      "epoch": 1.76965570900603,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046848160126906956,
      "loss": 5.0974,
      "mean_token_accuracy": 0.19544232934713363,
      "num_tokens": 39466545.0,
      "step": 22745
    },
    {
      "entropy": 5.631972789764404,
      "epoch": 1.7700447383777476,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004684674528834186,
      "loss": 5.0811,
      "mean_token_accuracy": 0.192271426320076,
      "num_tokens": 39475771.0,
      "step": 22750
    },
    {
      "entropy": 5.6377147197723385,
      "epoch": 1.770433767749465,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046845330156220703,
      "loss": 5.0882,
      "mean_token_accuracy": 0.19829896539449693,
      "num_tokens": 39484244.0,
      "step": 22755
    },
    {
      "entropy": 5.67882890701294,
      "epoch": 1.7708227971211827,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004684391473056495,
      "loss": 5.1049,
      "mean_token_accuracy": 0.19099098592996597,
      "num_tokens": 39493138.0,
      "step": 22760
    },
    {
      "entropy": 5.679558277130127,
      "epoch": 1.7712118264929002,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046842499011396076,
      "loss": 5.0594,
      "mean_token_accuracy": 0.19432098269462586,
      "num_tokens": 39502065.0,
      "step": 22765
    },
    {
      "entropy": 5.616088581085205,
      "epoch": 1.7716008558646177,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046841082998735575,
      "loss": 5.0516,
      "mean_token_accuracy": 0.2049824982881546,
      "num_tokens": 39510095.0,
      "step": 22770
    },
    {
      "entropy": 5.5582022190094,
      "epoch": 1.7719898852363354,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046839666692604916,
      "loss": 5.0905,
      "mean_token_accuracy": 0.19606523364782333,
      "num_tokens": 39518104.0,
      "step": 22775
    },
    {
      "entropy": 5.659603881835937,
      "epoch": 1.7723789146080529,
      "grad_norm": 1.453125,
      "learning_rate": 0.000468382500930256,
      "loss": 5.0534,
      "mean_token_accuracy": 0.1983247220516205,
      "num_tokens": 39526768.0,
      "step": 22780
    },
    {
      "entropy": 5.627512502670288,
      "epoch": 1.7727679439797703,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046836833200019116,
      "loss": 5.0173,
      "mean_token_accuracy": 0.19492925107479095,
      "num_tokens": 39535847.0,
      "step": 22785
    },
    {
      "entropy": 5.6733832359313965,
      "epoch": 1.773156973351488,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004683541601360697,
      "loss": 5.087,
      "mean_token_accuracy": 0.2005626752972603,
      "num_tokens": 39544364.0,
      "step": 22790
    },
    {
      "entropy": 5.63012318611145,
      "epoch": 1.7735460027232057,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046833998533810653,
      "loss": 5.105,
      "mean_token_accuracy": 0.19127426594495772,
      "num_tokens": 39552733.0,
      "step": 22795
    },
    {
      "entropy": 5.638961124420166,
      "epoch": 1.7739350320949232,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004683258076065169,
      "loss": 5.0359,
      "mean_token_accuracy": 0.19908232539892196,
      "num_tokens": 39561580.0,
      "step": 22800
    },
    {
      "entropy": 5.575855731964111,
      "epoch": 1.7743240614666407,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046831162694151586,
      "loss": 5.0166,
      "mean_token_accuracy": 0.1958705946803093,
      "num_tokens": 39570283.0,
      "step": 22805
    },
    {
      "entropy": 5.633059358596801,
      "epoch": 1.7747130908383584,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046829744334331855,
      "loss": 4.986,
      "mean_token_accuracy": 0.20528567135334014,
      "num_tokens": 39578835.0,
      "step": 22810
    },
    {
      "entropy": 5.619071960449219,
      "epoch": 1.7751021202100759,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046828325681214013,
      "loss": 5.0621,
      "mean_token_accuracy": 0.2053024113178253,
      "num_tokens": 39587053.0,
      "step": 22815
    },
    {
      "entropy": 5.655352926254272,
      "epoch": 1.7754911495817933,
      "grad_norm": 1.1875,
      "learning_rate": 0.000468269067348196,
      "loss": 5.1407,
      "mean_token_accuracy": 0.18540143817663193,
      "num_tokens": 39595799.0,
      "step": 22820
    },
    {
      "entropy": 5.652642107009887,
      "epoch": 1.775880178953511,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004682548749517014,
      "loss": 5.0113,
      "mean_token_accuracy": 0.19996405839920045,
      "num_tokens": 39604248.0,
      "step": 22825
    },
    {
      "entropy": 5.557036304473877,
      "epoch": 1.7762692083252287,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046824067962287163,
      "loss": 5.0153,
      "mean_token_accuracy": 0.19849045425653458,
      "num_tokens": 39612690.0,
      "step": 22830
    },
    {
      "entropy": 5.632285070419312,
      "epoch": 1.776658237696946,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004682264813619221,
      "loss": 5.0469,
      "mean_token_accuracy": 0.19531194120645523,
      "num_tokens": 39621784.0,
      "step": 22835
    },
    {
      "entropy": 5.6317747116088865,
      "epoch": 1.7770472670686637,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046821228016906836,
      "loss": 4.9848,
      "mean_token_accuracy": 0.2053050622344017,
      "num_tokens": 39630455.0,
      "step": 22840
    },
    {
      "entropy": 5.590449666976928,
      "epoch": 1.7774362964403814,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004681980760445257,
      "loss": 4.9935,
      "mean_token_accuracy": 0.20124854743480683,
      "num_tokens": 39639017.0,
      "step": 22845
    },
    {
      "entropy": 5.589898252487183,
      "epoch": 1.7778253258120988,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004681838689885098,
      "loss": 4.9485,
      "mean_token_accuracy": 0.2057025760412216,
      "num_tokens": 39648083.0,
      "step": 22850
    },
    {
      "entropy": 5.582891273498535,
      "epoch": 1.7782143551838163,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004681696590012362,
      "loss": 5.0382,
      "mean_token_accuracy": 0.2028049796819687,
      "num_tokens": 39656887.0,
      "step": 22855
    },
    {
      "entropy": 5.571018362045288,
      "epoch": 1.778603384555534,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046815544608292043,
      "loss": 5.064,
      "mean_token_accuracy": 0.1959187299013138,
      "num_tokens": 39665658.0,
      "step": 22860
    },
    {
      "entropy": 5.641104888916016,
      "epoch": 1.7789924139272515,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004681412302337782,
      "loss": 5.0367,
      "mean_token_accuracy": 0.20009022057056428,
      "num_tokens": 39673423.0,
      "step": 22865
    },
    {
      "entropy": 5.654715347290039,
      "epoch": 1.779381443298969,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004681270114540252,
      "loss": 5.0502,
      "mean_token_accuracy": 0.19800397008657455,
      "num_tokens": 39682382.0,
      "step": 22870
    },
    {
      "entropy": 5.633648920059204,
      "epoch": 1.7797704726706867,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004681127897438772,
      "loss": 5.088,
      "mean_token_accuracy": 0.19494579881429672,
      "num_tokens": 39691089.0,
      "step": 22875
    },
    {
      "entropy": 5.6912885189056395,
      "epoch": 1.7801595020424044,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046809856510355007,
      "loss": 5.1391,
      "mean_token_accuracy": 0.19422789067029952,
      "num_tokens": 39700888.0,
      "step": 22880
    },
    {
      "entropy": 5.648356246948242,
      "epoch": 1.7805485314141216,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004680843375332595,
      "loss": 5.0726,
      "mean_token_accuracy": 0.20570575296878815,
      "num_tokens": 39708515.0,
      "step": 22885
    },
    {
      "entropy": 5.544374704360962,
      "epoch": 1.7809375607858393,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046807010703322143,
      "loss": 4.9426,
      "mean_token_accuracy": 0.1983916640281677,
      "num_tokens": 39716790.0,
      "step": 22890
    },
    {
      "entropy": 5.66735987663269,
      "epoch": 1.781326590157557,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004680558736036518,
      "loss": 5.0131,
      "mean_token_accuracy": 0.19904389083385468,
      "num_tokens": 39726292.0,
      "step": 22895
    },
    {
      "entropy": 5.584895515441895,
      "epoch": 1.7817156195292745,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004680416372447666,
      "loss": 5.0174,
      "mean_token_accuracy": 0.20312283784151078,
      "num_tokens": 39735222.0,
      "step": 22900
    },
    {
      "entropy": 5.619968891143799,
      "epoch": 1.782104648900992,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046802739795678177,
      "loss": 5.0716,
      "mean_token_accuracy": 0.19360256642103196,
      "num_tokens": 39744619.0,
      "step": 22905
    },
    {
      "entropy": 5.633113956451416,
      "epoch": 1.7824936782727097,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046801315573991346,
      "loss": 5.0706,
      "mean_token_accuracy": 0.19817846417427062,
      "num_tokens": 39752394.0,
      "step": 22910
    },
    {
      "entropy": 5.54956316947937,
      "epoch": 1.7828827076444271,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046799891059437764,
      "loss": 4.9658,
      "mean_token_accuracy": 0.20080901980400084,
      "num_tokens": 39760529.0,
      "step": 22915
    },
    {
      "entropy": 5.651667070388794,
      "epoch": 1.7832717370161446,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046798466252039056,
      "loss": 5.1195,
      "mean_token_accuracy": 0.19348373413085937,
      "num_tokens": 39768231.0,
      "step": 22920
    },
    {
      "entropy": 5.7124701023101805,
      "epoch": 1.7836607663878623,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046797041151816845,
      "loss": 5.1758,
      "mean_token_accuracy": 0.19045044481754303,
      "num_tokens": 39776773.0,
      "step": 22925
    },
    {
      "entropy": 5.686195135116577,
      "epoch": 1.78404979575958,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046795615758792747,
      "loss": 5.0429,
      "mean_token_accuracy": 0.18977955877780914,
      "num_tokens": 39785248.0,
      "step": 22930
    },
    {
      "entropy": 5.60693507194519,
      "epoch": 1.7844388251312973,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004679419007298839,
      "loss": 5.0398,
      "mean_token_accuracy": 0.19851401895284654,
      "num_tokens": 39794794.0,
      "step": 22935
    },
    {
      "entropy": 5.621718311309815,
      "epoch": 1.784827854503015,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004679276409442541,
      "loss": 5.0277,
      "mean_token_accuracy": 0.19571128487586975,
      "num_tokens": 39803662.0,
      "step": 22940
    },
    {
      "entropy": 5.660523080825806,
      "epoch": 1.7852168838747327,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004679133782312544,
      "loss": 5.0279,
      "mean_token_accuracy": 0.1910773292183876,
      "num_tokens": 39812189.0,
      "step": 22945
    },
    {
      "entropy": 5.642277097702026,
      "epoch": 1.7856059132464501,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004678991125911013,
      "loss": 5.0757,
      "mean_token_accuracy": 0.19846590012311935,
      "num_tokens": 39820831.0,
      "step": 22950
    },
    {
      "entropy": 5.534529829025269,
      "epoch": 1.7859949426181676,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004678848440240111,
      "loss": 4.9067,
      "mean_token_accuracy": 0.2052549973130226,
      "num_tokens": 39829380.0,
      "step": 22955
    },
    {
      "entropy": 5.598503494262696,
      "epoch": 1.7863839719898853,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004678705725302005,
      "loss": 5.0465,
      "mean_token_accuracy": 0.2032509282231331,
      "num_tokens": 39838243.0,
      "step": 22960
    },
    {
      "entropy": 5.640819025039673,
      "epoch": 1.7867730013616028,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004678562981098859,
      "loss": 5.0134,
      "mean_token_accuracy": 0.20074148327112198,
      "num_tokens": 39846885.0,
      "step": 22965
    },
    {
      "entropy": 5.618864917755127,
      "epoch": 1.7871620307333203,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046784202076328394,
      "loss": 4.9742,
      "mean_token_accuracy": 0.20472084432840348,
      "num_tokens": 39855337.0,
      "step": 22970
    },
    {
      "entropy": 5.57243800163269,
      "epoch": 1.787551060105038,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046782774049061124,
      "loss": 5.0477,
      "mean_token_accuracy": 0.2021118849515915,
      "num_tokens": 39864606.0,
      "step": 22975
    },
    {
      "entropy": 5.703754615783692,
      "epoch": 1.7879400894767556,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004678134572920845,
      "loss": 5.0539,
      "mean_token_accuracy": 0.19163644313812256,
      "num_tokens": 39872621.0,
      "step": 22980
    },
    {
      "entropy": 5.618570041656494,
      "epoch": 1.788329118848473,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004677991711679204,
      "loss": 5.0199,
      "mean_token_accuracy": 0.20278891175985336,
      "num_tokens": 39880525.0,
      "step": 22985
    },
    {
      "entropy": 5.581645917892456,
      "epoch": 1.7887181482201906,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046778488211833585,
      "loss": 4.9541,
      "mean_token_accuracy": 0.20556234568357468,
      "num_tokens": 39889541.0,
      "step": 22990
    },
    {
      "entropy": 5.670374536514283,
      "epoch": 1.7891071775919083,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004677705901435475,
      "loss": 5.0219,
      "mean_token_accuracy": 0.20049411058425903,
      "num_tokens": 39897682.0,
      "step": 22995
    },
    {
      "entropy": 5.667968463897705,
      "epoch": 1.7894962069636258,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004677562952437723,
      "loss": 5.0305,
      "mean_token_accuracy": 0.19325620383024217,
      "num_tokens": 39906656.0,
      "step": 23000
    },
    {
      "entropy": 5.64656286239624,
      "epoch": 1.7898852363353432,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046774199741922705,
      "loss": 5.1082,
      "mean_token_accuracy": 0.19324425011873245,
      "num_tokens": 39915202.0,
      "step": 23005
    },
    {
      "entropy": 5.593799114227295,
      "epoch": 1.790274265707061,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046772769667012884,
      "loss": 4.9583,
      "mean_token_accuracy": 0.19882046580314636,
      "num_tokens": 39923510.0,
      "step": 23010
    },
    {
      "entropy": 5.539695692062378,
      "epoch": 1.7906632950787784,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004677133929966946,
      "loss": 5.0021,
      "mean_token_accuracy": 0.20754729956388474,
      "num_tokens": 39932468.0,
      "step": 23015
    },
    {
      "entropy": 5.664415550231934,
      "epoch": 1.791052324450496,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046769908639914134,
      "loss": 5.108,
      "mean_token_accuracy": 0.19654156416654586,
      "num_tokens": 39942758.0,
      "step": 23020
    },
    {
      "entropy": 5.7177956104278564,
      "epoch": 1.7914413538222136,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004676847768776861,
      "loss": 5.0902,
      "mean_token_accuracy": 0.19549465626478196,
      "num_tokens": 39950829.0,
      "step": 23025
    },
    {
      "entropy": 5.614761257171631,
      "epoch": 1.7918303831939313,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004676704644325462,
      "loss": 5.0645,
      "mean_token_accuracy": 0.19321791529655458,
      "num_tokens": 39960204.0,
      "step": 23030
    },
    {
      "entropy": 5.617123413085937,
      "epoch": 1.7922194125656488,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004676561490639385,
      "loss": 5.0288,
      "mean_token_accuracy": 0.1986328184604645,
      "num_tokens": 39968513.0,
      "step": 23035
    },
    {
      "entropy": 5.614787149429321,
      "epoch": 1.7926084419373662,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004676418307720805,
      "loss": 5.0642,
      "mean_token_accuracy": 0.19716532081365584,
      "num_tokens": 39977181.0,
      "step": 23040
    },
    {
      "entropy": 5.544361543655396,
      "epoch": 1.792997471309084,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004676275095571892,
      "loss": 4.8729,
      "mean_token_accuracy": 0.21068945676088333,
      "num_tokens": 39986760.0,
      "step": 23045
    },
    {
      "entropy": 5.635046195983887,
      "epoch": 1.7933865006808014,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046761318541948215,
      "loss": 5.0925,
      "mean_token_accuracy": 0.18915906101465224,
      "num_tokens": 39995203.0,
      "step": 23050
    },
    {
      "entropy": 5.599363803863525,
      "epoch": 1.7937755300525189,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004675988583591766,
      "loss": 4.9717,
      "mean_token_accuracy": 0.21010951548814774,
      "num_tokens": 40003526.0,
      "step": 23055
    },
    {
      "entropy": 5.632573032379151,
      "epoch": 1.7941645594242366,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046758452837648997,
      "loss": 5.029,
      "mean_token_accuracy": 0.1966042250394821,
      "num_tokens": 40012036.0,
      "step": 23060
    },
    {
      "entropy": 5.605696439743042,
      "epoch": 1.794553588795954,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004675701954716395,
      "loss": 5.0451,
      "mean_token_accuracy": 0.19894930273294448,
      "num_tokens": 40020237.0,
      "step": 23065
    },
    {
      "entropy": 5.558801460266113,
      "epoch": 1.7949426181676715,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046755585964484286,
      "loss": 4.9443,
      "mean_token_accuracy": 0.20942775011062623,
      "num_tokens": 40028749.0,
      "step": 23070
    },
    {
      "entropy": 5.606195640563965,
      "epoch": 1.7953316475393892,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004675415208963177,
      "loss": 4.9327,
      "mean_token_accuracy": 0.2019081547856331,
      "num_tokens": 40036799.0,
      "step": 23075
    },
    {
      "entropy": 5.58902759552002,
      "epoch": 1.795720676911107,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046752717922628125,
      "loss": 5.0737,
      "mean_token_accuracy": 0.1900378629565239,
      "num_tokens": 40045416.0,
      "step": 23080
    },
    {
      "entropy": 5.680563163757324,
      "epoch": 1.7961097062828244,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004675128346349514,
      "loss": 5.0804,
      "mean_token_accuracy": 0.201175494492054,
      "num_tokens": 40054128.0,
      "step": 23085
    },
    {
      "entropy": 5.720880508422852,
      "epoch": 1.7964987356545419,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046749848712254554,
      "loss": 5.1142,
      "mean_token_accuracy": 0.19448594003915787,
      "num_tokens": 40062765.0,
      "step": 23090
    },
    {
      "entropy": 5.586523056030273,
      "epoch": 1.7968877650262596,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046748413668928164,
      "loss": 5.044,
      "mean_token_accuracy": 0.19767448157072068,
      "num_tokens": 40071190.0,
      "step": 23095
    },
    {
      "entropy": 5.605634641647339,
      "epoch": 1.797276794397977,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004674697833353774,
      "loss": 5.0537,
      "mean_token_accuracy": 0.19914634972810746,
      "num_tokens": 40080552.0,
      "step": 23100
    },
    {
      "entropy": 5.636071634292603,
      "epoch": 1.7976658237696945,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046745542706105045,
      "loss": 5.0662,
      "mean_token_accuracy": 0.20074681043624878,
      "num_tokens": 40089407.0,
      "step": 23105
    },
    {
      "entropy": 5.7634786605834964,
      "epoch": 1.7980548531414122,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046744106786651875,
      "loss": 5.2552,
      "mean_token_accuracy": 0.18388169407844543,
      "num_tokens": 40097946.0,
      "step": 23110
    },
    {
      "entropy": 5.61965856552124,
      "epoch": 1.7984438825131297,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004674267057520001,
      "loss": 4.9933,
      "mean_token_accuracy": 0.2065201297402382,
      "num_tokens": 40105901.0,
      "step": 23115
    },
    {
      "entropy": 5.548855543136597,
      "epoch": 1.7988329118848472,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004674123407177125,
      "loss": 4.9142,
      "mean_token_accuracy": 0.19971174448728563,
      "num_tokens": 40114481.0,
      "step": 23120
    },
    {
      "entropy": 5.6078376293182375,
      "epoch": 1.7992219412565649,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046739797276387394,
      "loss": 4.9669,
      "mean_token_accuracy": 0.20634620487689972,
      "num_tokens": 40123086.0,
      "step": 23125
    },
    {
      "entropy": 5.647294378280639,
      "epoch": 1.7996109706282826,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046738360189070235,
      "loss": 4.9869,
      "mean_token_accuracy": 0.20457558929920197,
      "num_tokens": 40131497.0,
      "step": 23130
    },
    {
      "entropy": 5.592572975158691,
      "epoch": 1.8,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004673692280984157,
      "loss": 5.0127,
      "mean_token_accuracy": 0.19852383881807328,
      "num_tokens": 40139632.0,
      "step": 23135
    },
    {
      "entropy": 5.587947988510132,
      "epoch": 1.8003890293717175,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004673548513872324,
      "loss": 4.9999,
      "mean_token_accuracy": 0.2045734092593193,
      "num_tokens": 40148312.0,
      "step": 23140
    },
    {
      "entropy": 5.633117580413819,
      "epoch": 1.8007780587434352,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046734047175737026,
      "loss": 5.067,
      "mean_token_accuracy": 0.1953944444656372,
      "num_tokens": 40156750.0,
      "step": 23145
    },
    {
      "entropy": 5.600761127471924,
      "epoch": 1.8011670881151527,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046732608920904754,
      "loss": 5.0375,
      "mean_token_accuracy": 0.19682238399982452,
      "num_tokens": 40165334.0,
      "step": 23150
    },
    {
      "entropy": 5.597075033187866,
      "epoch": 1.8015561174868702,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004673117037424827,
      "loss": 5.0212,
      "mean_token_accuracy": 0.19980016350746155,
      "num_tokens": 40174217.0,
      "step": 23155
    },
    {
      "entropy": 5.538127088546753,
      "epoch": 1.8019451468585879,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046729731535789375,
      "loss": 4.9883,
      "mean_token_accuracy": 0.20436269640922547,
      "num_tokens": 40182889.0,
      "step": 23160
    },
    {
      "entropy": 5.625492763519287,
      "epoch": 1.8023341762303053,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046728292405549913,
      "loss": 5.0434,
      "mean_token_accuracy": 0.20408593565225602,
      "num_tokens": 40191428.0,
      "step": 23165
    },
    {
      "entropy": 5.6284246921539305,
      "epoch": 1.8027232056020228,
      "grad_norm": 1.125,
      "learning_rate": 0.0004672685298355172,
      "loss": 4.9996,
      "mean_token_accuracy": 0.2096492290496826,
      "num_tokens": 40200204.0,
      "step": 23170
    },
    {
      "entropy": 5.5758730411529545,
      "epoch": 1.8031122349737405,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004672541326981664,
      "loss": 4.9084,
      "mean_token_accuracy": 0.21313364952802658,
      "num_tokens": 40208326.0,
      "step": 23175
    },
    {
      "entropy": 5.5254319190979,
      "epoch": 1.8035012643454582,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000467239732643665,
      "loss": 5.0378,
      "mean_token_accuracy": 0.20054250806570054,
      "num_tokens": 40217300.0,
      "step": 23180
    },
    {
      "entropy": 5.628619384765625,
      "epoch": 1.8038902937171757,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046722532967223183,
      "loss": 5.116,
      "mean_token_accuracy": 0.19628379195928575,
      "num_tokens": 40226695.0,
      "step": 23185
    },
    {
      "entropy": 5.647273397445678,
      "epoch": 1.8042793230888932,
      "grad_norm": 1.125,
      "learning_rate": 0.0004672109237840851,
      "loss": 5.0098,
      "mean_token_accuracy": 0.19701189398765565,
      "num_tokens": 40235286.0,
      "step": 23190
    },
    {
      "entropy": 5.5947012424469,
      "epoch": 1.8046683524606109,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046719651497944355,
      "loss": 4.9774,
      "mean_token_accuracy": 0.20165529400110244,
      "num_tokens": 40244065.0,
      "step": 23195
    },
    {
      "entropy": 5.694451522827149,
      "epoch": 1.8050573818323283,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004671821032585259,
      "loss": 5.179,
      "mean_token_accuracy": 0.18821609169244766,
      "num_tokens": 40253058.0,
      "step": 23200
    },
    {
      "entropy": 5.673509788513184,
      "epoch": 1.8054464112040458,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004671676886215506,
      "loss": 5.0568,
      "mean_token_accuracy": 0.19948756843805313,
      "num_tokens": 40261797.0,
      "step": 23205
    },
    {
      "entropy": 5.592724704742432,
      "epoch": 1.8058354405757635,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004671532710687365,
      "loss": 5.0209,
      "mean_token_accuracy": 0.20477258116006852,
      "num_tokens": 40270003.0,
      "step": 23210
    },
    {
      "entropy": 5.595245742797852,
      "epoch": 1.806224469947481,
      "grad_norm": 1.125,
      "learning_rate": 0.0004671388506003024,
      "loss": 5.0355,
      "mean_token_accuracy": 0.20503741055727004,
      "num_tokens": 40278241.0,
      "step": 23215
    },
    {
      "entropy": 5.633118152618408,
      "epoch": 1.8066134993191985,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004671244272164671,
      "loss": 5.0921,
      "mean_token_accuracy": 0.19270867854356766,
      "num_tokens": 40286316.0,
      "step": 23220
    },
    {
      "entropy": 5.63343300819397,
      "epoch": 1.8070025286909162,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046711000091744935,
      "loss": 5.0556,
      "mean_token_accuracy": 0.19763821512460708,
      "num_tokens": 40295378.0,
      "step": 23225
    },
    {
      "entropy": 5.675390386581421,
      "epoch": 1.8073915580626339,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004670955717034681,
      "loss": 5.062,
      "mean_token_accuracy": 0.194849593937397,
      "num_tokens": 40305379.0,
      "step": 23230
    },
    {
      "entropy": 5.6832112789154055,
      "epoch": 1.8077805874343513,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046708113957474233,
      "loss": 5.0944,
      "mean_token_accuracy": 0.19446978271007537,
      "num_tokens": 40314174.0,
      "step": 23235
    },
    {
      "entropy": 5.669077634811401,
      "epoch": 1.8081696168060688,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000467066704531491,
      "loss": 5.0785,
      "mean_token_accuracy": 0.1998366430401802,
      "num_tokens": 40323708.0,
      "step": 23240
    },
    {
      "entropy": 5.603174829483033,
      "epoch": 1.8085586461777865,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004670522665739332,
      "loss": 4.973,
      "mean_token_accuracy": 0.2021640881896019,
      "num_tokens": 40333664.0,
      "step": 23245
    },
    {
      "entropy": 5.631776285171509,
      "epoch": 1.808947675549504,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004670378257022878,
      "loss": 5.0319,
      "mean_token_accuracy": 0.20200446993112564,
      "num_tokens": 40342037.0,
      "step": 23250
    },
    {
      "entropy": 5.610338973999023,
      "epoch": 1.8093367049212215,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046702338191677414,
      "loss": 4.967,
      "mean_token_accuracy": 0.2091684564948082,
      "num_tokens": 40350472.0,
      "step": 23255
    },
    {
      "entropy": 5.615092611312866,
      "epoch": 1.8097257342929391,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004670089352176113,
      "loss": 5.0465,
      "mean_token_accuracy": 0.19918030500411987,
      "num_tokens": 40358829.0,
      "step": 23260
    },
    {
      "entropy": 5.605515098571777,
      "epoch": 1.8101147636646568,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046699448560501847,
      "loss": 5.0424,
      "mean_token_accuracy": 0.20094968378543854,
      "num_tokens": 40367414.0,
      "step": 23265
    },
    {
      "entropy": 5.6542649269104,
      "epoch": 1.810503793036374,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004669800330792149,
      "loss": 5.11,
      "mean_token_accuracy": 0.18898468911647798,
      "num_tokens": 40375969.0,
      "step": 23270
    },
    {
      "entropy": 5.6899800300598145,
      "epoch": 1.8108928224080918,
      "grad_norm": 1.125,
      "learning_rate": 0.00046696557764041994,
      "loss": 5.0352,
      "mean_token_accuracy": 0.1950753539800644,
      "num_tokens": 40384200.0,
      "step": 23275
    },
    {
      "entropy": 5.683923292160034,
      "epoch": 1.8112818517798095,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004669511192888528,
      "loss": 5.1396,
      "mean_token_accuracy": 0.19239381700754166,
      "num_tokens": 40392891.0,
      "step": 23280
    },
    {
      "entropy": 5.680789756774902,
      "epoch": 1.811670881151527,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046693665802473294,
      "loss": 5.0943,
      "mean_token_accuracy": 0.1943405330181122,
      "num_tokens": 40402009.0,
      "step": 23285
    },
    {
      "entropy": 5.6732964515686035,
      "epoch": 1.8120599105232444,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046692219384827986,
      "loss": 5.0804,
      "mean_token_accuracy": 0.18799724131822587,
      "num_tokens": 40410756.0,
      "step": 23290
    },
    {
      "entropy": 5.6894214153289795,
      "epoch": 1.8124489398949621,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004669077267597129,
      "loss": 5.1112,
      "mean_token_accuracy": 0.19085838794708251,
      "num_tokens": 40420087.0,
      "step": 23295
    },
    {
      "entropy": 5.58405613899231,
      "epoch": 1.8128379692666796,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046689325675925174,
      "loss": 4.9792,
      "mean_token_accuracy": 0.20499624013900758,
      "num_tokens": 40428741.0,
      "step": 23300
    },
    {
      "entropy": 5.7090095520019535,
      "epoch": 1.813226998638397,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046687878384711574,
      "loss": 5.1738,
      "mean_token_accuracy": 0.19049516171216965,
      "num_tokens": 40438128.0,
      "step": 23305
    },
    {
      "entropy": 5.660978651046753,
      "epoch": 1.8136160280101148,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046686430802352476,
      "loss": 5.1363,
      "mean_token_accuracy": 0.1953614756464958,
      "num_tokens": 40446990.0,
      "step": 23310
    },
    {
      "entropy": 5.62388424873352,
      "epoch": 1.8140050573818325,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004668498292886982,
      "loss": 5.0125,
      "mean_token_accuracy": 0.20001835376024246,
      "num_tokens": 40454627.0,
      "step": 23315
    },
    {
      "entropy": 5.577475070953369,
      "epoch": 1.8143940867535497,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046683534764285577,
      "loss": 5.0581,
      "mean_token_accuracy": 0.20503107756376265,
      "num_tokens": 40463321.0,
      "step": 23320
    },
    {
      "entropy": 5.58239598274231,
      "epoch": 1.8147831161252674,
      "grad_norm": 1.125,
      "learning_rate": 0.00046682086308621725,
      "loss": 4.912,
      "mean_token_accuracy": 0.20495291352272033,
      "num_tokens": 40472156.0,
      "step": 23325
    },
    {
      "entropy": 5.615819883346558,
      "epoch": 1.8151721454969851,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004668063756190026,
      "loss": 5.0472,
      "mean_token_accuracy": 0.19533439427614213,
      "num_tokens": 40480938.0,
      "step": 23330
    },
    {
      "entropy": 5.566732549667359,
      "epoch": 1.8155611748687026,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046679188524143136,
      "loss": 4.9985,
      "mean_token_accuracy": 0.20331875532865523,
      "num_tokens": 40489740.0,
      "step": 23335
    },
    {
      "entropy": 5.645817661285401,
      "epoch": 1.81595020424042,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004667773919537235,
      "loss": 4.9804,
      "mean_token_accuracy": 0.20023264437913896,
      "num_tokens": 40498420.0,
      "step": 23340
    },
    {
      "entropy": 5.589276695251465,
      "epoch": 1.8163392336121378,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046676289575609914,
      "loss": 4.8997,
      "mean_token_accuracy": 0.20963711887598038,
      "num_tokens": 40506634.0,
      "step": 23345
    },
    {
      "entropy": 5.660476589202881,
      "epoch": 1.8167282629838553,
      "grad_norm": 1.125,
      "learning_rate": 0.00046674839664877796,
      "loss": 5.1098,
      "mean_token_accuracy": 0.20263600796461106,
      "num_tokens": 40515310.0,
      "step": 23350
    },
    {
      "entropy": 5.666203308105469,
      "epoch": 1.8171172923555727,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004667338946319801,
      "loss": 4.9802,
      "mean_token_accuracy": 0.20835888534784316,
      "num_tokens": 40523739.0,
      "step": 23355
    },
    {
      "entropy": 5.6660542488098145,
      "epoch": 1.8175063217272904,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004667193897059255,
      "loss": 5.1391,
      "mean_token_accuracy": 0.19278421401977539,
      "num_tokens": 40533070.0,
      "step": 23360
    },
    {
      "entropy": 5.664768123626709,
      "epoch": 1.8178953510990081,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046670488187083436,
      "loss": 4.9967,
      "mean_token_accuracy": 0.20563114434480667,
      "num_tokens": 40541975.0,
      "step": 23365
    },
    {
      "entropy": 5.696281433105469,
      "epoch": 1.8182843804707254,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004666903711269267,
      "loss": 5.1271,
      "mean_token_accuracy": 0.19356989115476608,
      "num_tokens": 40550896.0,
      "step": 23370
    },
    {
      "entropy": 5.618568181991577,
      "epoch": 1.818673409842443,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004666758574744228,
      "loss": 5.0939,
      "mean_token_accuracy": 0.20155951976776124,
      "num_tokens": 40559719.0,
      "step": 23375
    },
    {
      "entropy": 5.668618392944336,
      "epoch": 1.8190624392141608,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004666613409135429,
      "loss": 5.0917,
      "mean_token_accuracy": 0.19836247712373734,
      "num_tokens": 40568448.0,
      "step": 23380
    },
    {
      "entropy": 5.735010194778442,
      "epoch": 1.8194514685858783,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004666468214445072,
      "loss": 5.1217,
      "mean_token_accuracy": 0.19426717311143876,
      "num_tokens": 40576443.0,
      "step": 23385
    },
    {
      "entropy": 5.609504747390747,
      "epoch": 1.8198404979575957,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046663229906753595,
      "loss": 4.8785,
      "mean_token_accuracy": 0.2150632083415985,
      "num_tokens": 40584440.0,
      "step": 23390
    },
    {
      "entropy": 5.605626583099365,
      "epoch": 1.8202295273293134,
      "grad_norm": 1.125,
      "learning_rate": 0.00046661777378284965,
      "loss": 5.1571,
      "mean_token_accuracy": 0.19661157876253127,
      "num_tokens": 40593696.0,
      "step": 23395
    },
    {
      "entropy": 5.590264368057251,
      "epoch": 1.820618556701031,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004666032455906685,
      "loss": 4.9759,
      "mean_token_accuracy": 0.2084008902311325,
      "num_tokens": 40601368.0,
      "step": 23400
    },
    {
      "entropy": 5.716005134582519,
      "epoch": 1.8210075860727484,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046658871449121325,
      "loss": 5.0597,
      "mean_token_accuracy": 0.20483262240886688,
      "num_tokens": 40609143.0,
      "step": 23405
    },
    {
      "entropy": 5.681556367874146,
      "epoch": 1.821396615444466,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004665741804847041,
      "loss": 5.0446,
      "mean_token_accuracy": 0.19661843478679658,
      "num_tokens": 40617602.0,
      "step": 23410
    },
    {
      "entropy": 5.547113418579102,
      "epoch": 1.8217856448161838,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046655964357136164,
      "loss": 4.9386,
      "mean_token_accuracy": 0.20660165101289749,
      "num_tokens": 40626412.0,
      "step": 23415
    },
    {
      "entropy": 5.617480087280273,
      "epoch": 1.822174674187901,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046654510375140657,
      "loss": 5.0111,
      "mean_token_accuracy": 0.20171340703964233,
      "num_tokens": 40635095.0,
      "step": 23420
    },
    {
      "entropy": 5.641251277923584,
      "epoch": 1.8225637035596187,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004665305610250594,
      "loss": 5.0127,
      "mean_token_accuracy": 0.199694462120533,
      "num_tokens": 40643421.0,
      "step": 23425
    },
    {
      "entropy": 5.6266560554504395,
      "epoch": 1.8229527329313364,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004665160153925408,
      "loss": 5.0507,
      "mean_token_accuracy": 0.2015625447034836,
      "num_tokens": 40652180.0,
      "step": 23430
    },
    {
      "entropy": 5.707278060913086,
      "epoch": 1.823341762303054,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046650146685407154,
      "loss": 5.1214,
      "mean_token_accuracy": 0.19078605026006698,
      "num_tokens": 40660398.0,
      "step": 23435
    },
    {
      "entropy": 5.638275766372681,
      "epoch": 1.8237307916747714,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046648691540987226,
      "loss": 5.1415,
      "mean_token_accuracy": 0.18793484270572663,
      "num_tokens": 40668827.0,
      "step": 23440
    },
    {
      "entropy": 5.653947019577027,
      "epoch": 1.824119821046489,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046647236106016387,
      "loss": 5.038,
      "mean_token_accuracy": 0.204281485080719,
      "num_tokens": 40676920.0,
      "step": 23445
    },
    {
      "entropy": 5.627874565124512,
      "epoch": 1.8245088504182065,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004664578038051672,
      "loss": 4.985,
      "mean_token_accuracy": 0.20252078622579575,
      "num_tokens": 40685115.0,
      "step": 23450
    },
    {
      "entropy": 5.63462724685669,
      "epoch": 1.824897879789924,
      "grad_norm": 1.171875,
      "learning_rate": 0.000466443243645103,
      "loss": 5.022,
      "mean_token_accuracy": 0.20258653312921523,
      "num_tokens": 40693870.0,
      "step": 23455
    },
    {
      "entropy": 5.640165996551514,
      "epoch": 1.8252869091616417,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004664286805801923,
      "loss": 4.9996,
      "mean_token_accuracy": 0.19670678675174713,
      "num_tokens": 40703031.0,
      "step": 23460
    },
    {
      "entropy": 5.56710638999939,
      "epoch": 1.8256759385333594,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004664141146106562,
      "loss": 4.9733,
      "mean_token_accuracy": 0.20350362211465836,
      "num_tokens": 40711690.0,
      "step": 23465
    },
    {
      "entropy": 5.637618827819824,
      "epoch": 1.8260649679050769,
      "grad_norm": 1.125,
      "learning_rate": 0.00046639954573671547,
      "loss": 5.0782,
      "mean_token_accuracy": 0.19567493051290513,
      "num_tokens": 40720086.0,
      "step": 23470
    },
    {
      "entropy": 5.713523292541504,
      "epoch": 1.8264539972767944,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046638497395859127,
      "loss": 5.0936,
      "mean_token_accuracy": 0.19538353979587555,
      "num_tokens": 40728117.0,
      "step": 23475
    },
    {
      "entropy": 5.672002935409546,
      "epoch": 1.826843026648512,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004663703992765047,
      "loss": 5.0891,
      "mean_token_accuracy": 0.20182826220989228,
      "num_tokens": 40736761.0,
      "step": 23480
    },
    {
      "entropy": 5.623796176910401,
      "epoch": 1.8272320560202295,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046635582169067693,
      "loss": 5.0372,
      "mean_token_accuracy": 0.19532569199800492,
      "num_tokens": 40745291.0,
      "step": 23485
    },
    {
      "entropy": 5.710185718536377,
      "epoch": 1.827621085391947,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046634124120132915,
      "loss": 5.0643,
      "mean_token_accuracy": 0.200746650993824,
      "num_tokens": 40754308.0,
      "step": 23490
    },
    {
      "entropy": 5.585196256637573,
      "epoch": 1.8280101147636647,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004663266578086826,
      "loss": 4.9514,
      "mean_token_accuracy": 0.20119321942329407,
      "num_tokens": 40763046.0,
      "step": 23495
    },
    {
      "entropy": 5.560595607757568,
      "epoch": 1.8283991441353822,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004663120715129585,
      "loss": 5.0093,
      "mean_token_accuracy": 0.19790664613246917,
      "num_tokens": 40770912.0,
      "step": 23500
    },
    {
      "entropy": 5.662583208084106,
      "epoch": 1.8287881735070997,
      "grad_norm": 1.125,
      "learning_rate": 0.0004662974823143783,
      "loss": 5.0973,
      "mean_token_accuracy": 0.18717847019433975,
      "num_tokens": 40779748.0,
      "step": 23505
    },
    {
      "entropy": 5.70459246635437,
      "epoch": 1.8291772028788174,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004662828902131632,
      "loss": 5.076,
      "mean_token_accuracy": 0.19280010014772414,
      "num_tokens": 40787732.0,
      "step": 23510
    },
    {
      "entropy": 5.620465898513794,
      "epoch": 1.829566232250535,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004662682952095348,
      "loss": 4.9837,
      "mean_token_accuracy": 0.20278068333864213,
      "num_tokens": 40796521.0,
      "step": 23515
    },
    {
      "entropy": 5.6513392448425295,
      "epoch": 1.8299552616222525,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046625369730371436,
      "loss": 5.1351,
      "mean_token_accuracy": 0.19270308315753937,
      "num_tokens": 40805479.0,
      "step": 23520
    },
    {
      "entropy": 5.617520999908447,
      "epoch": 1.83034429099397,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004662390964959235,
      "loss": 4.9483,
      "mean_token_accuracy": 0.20246266275644303,
      "num_tokens": 40813647.0,
      "step": 23525
    },
    {
      "entropy": 5.56342453956604,
      "epoch": 1.8307333203656877,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046622449278638375,
      "loss": 4.9291,
      "mean_token_accuracy": 0.20142456144094467,
      "num_tokens": 40821753.0,
      "step": 23530
    },
    {
      "entropy": 5.5992988586425785,
      "epoch": 1.8311223497374052,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004662098861753167,
      "loss": 5.0795,
      "mean_token_accuracy": 0.20021137744188308,
      "num_tokens": 40830391.0,
      "step": 23535
    },
    {
      "entropy": 5.631122827529907,
      "epoch": 1.8315113791091227,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046619527666294394,
      "loss": 5.0281,
      "mean_token_accuracy": 0.1989116281270981,
      "num_tokens": 40839372.0,
      "step": 23540
    },
    {
      "entropy": 5.671263551712036,
      "epoch": 1.8319004084808403,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004661806642494872,
      "loss": 5.118,
      "mean_token_accuracy": 0.1946151077747345,
      "num_tokens": 40847945.0,
      "step": 23545
    },
    {
      "entropy": 5.723411893844604,
      "epoch": 1.8322894378525578,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004661660489351681,
      "loss": 5.1568,
      "mean_token_accuracy": 0.19021772891283034,
      "num_tokens": 40857600.0,
      "step": 23550
    },
    {
      "entropy": 5.615023183822632,
      "epoch": 1.8326784672242753,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004661514307202086,
      "loss": 4.9757,
      "mean_token_accuracy": 0.20282722115516663,
      "num_tokens": 40866672.0,
      "step": 23555
    },
    {
      "entropy": 5.641484069824219,
      "epoch": 1.833067496595993,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046613680960483036,
      "loss": 5.0327,
      "mean_token_accuracy": 0.1948842450976372,
      "num_tokens": 40875240.0,
      "step": 23560
    },
    {
      "entropy": 5.600885677337646,
      "epoch": 1.8334565259677107,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004661221855892552,
      "loss": 4.9899,
      "mean_token_accuracy": 0.20269301384687424,
      "num_tokens": 40883731.0,
      "step": 23565
    },
    {
      "entropy": 5.558529043197632,
      "epoch": 1.8338455553394282,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046610755867370506,
      "loss": 4.95,
      "mean_token_accuracy": 0.2098532199859619,
      "num_tokens": 40892572.0,
      "step": 23570
    },
    {
      "entropy": 5.589279842376709,
      "epoch": 1.8342345847111456,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000466092928858402,
      "loss": 5.0444,
      "mean_token_accuracy": 0.2021312430500984,
      "num_tokens": 40901603.0,
      "step": 23575
    },
    {
      "entropy": 5.61164116859436,
      "epoch": 1.8346236140828633,
      "grad_norm": 1.125,
      "learning_rate": 0.00046607829614356787,
      "loss": 5.0085,
      "mean_token_accuracy": 0.20301983803510665,
      "num_tokens": 40910502.0,
      "step": 23580
    },
    {
      "entropy": 5.657184314727783,
      "epoch": 1.8350126434545808,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004660636605294247,
      "loss": 4.9748,
      "mean_token_accuracy": 0.20652611553668976,
      "num_tokens": 40919858.0,
      "step": 23585
    },
    {
      "entropy": 5.686249208450318,
      "epoch": 1.8354016728262983,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004660490220161946,
      "loss": 5.021,
      "mean_token_accuracy": 0.20412850379943848,
      "num_tokens": 40928795.0,
      "step": 23590
    },
    {
      "entropy": 5.60613603591919,
      "epoch": 1.835790702198016,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046603438060409966,
      "loss": 5.011,
      "mean_token_accuracy": 0.20072602927684785,
      "num_tokens": 40937240.0,
      "step": 23595
    },
    {
      "entropy": 5.62018404006958,
      "epoch": 1.8361797315697335,
      "grad_norm": 1.125,
      "learning_rate": 0.000466019736293362,
      "loss": 5.0688,
      "mean_token_accuracy": 0.2002474144101143,
      "num_tokens": 40945974.0,
      "step": 23600
    },
    {
      "entropy": 5.599502086639404,
      "epoch": 1.836568760941451,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046600508908420396,
      "loss": 4.9625,
      "mean_token_accuracy": 0.20582503229379653,
      "num_tokens": 40953553.0,
      "step": 23605
    },
    {
      "entropy": 5.6148271560668945,
      "epoch": 1.8369577903131686,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046599043897684766,
      "loss": 5.0057,
      "mean_token_accuracy": 0.2088996633887291,
      "num_tokens": 40961287.0,
      "step": 23610
    },
    {
      "entropy": 5.583562898635864,
      "epoch": 1.8373468196848863,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004659757859715155,
      "loss": 4.9577,
      "mean_token_accuracy": 0.2002980262041092,
      "num_tokens": 40970266.0,
      "step": 23615
    },
    {
      "entropy": 5.652315282821656,
      "epoch": 1.8377358490566038,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046596113006842975,
      "loss": 4.9915,
      "mean_token_accuracy": 0.20945734828710555,
      "num_tokens": 40978303.0,
      "step": 23620
    },
    {
      "entropy": 5.719200801849365,
      "epoch": 1.8381248784283213,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004659464712678128,
      "loss": 5.051,
      "mean_token_accuracy": 0.19932790249586105,
      "num_tokens": 40986360.0,
      "step": 23625
    },
    {
      "entropy": 5.633736228942871,
      "epoch": 1.838513907800039,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004659318095698871,
      "loss": 5.0936,
      "mean_token_accuracy": 0.1960252806544304,
      "num_tokens": 40995323.0,
      "step": 23630
    },
    {
      "entropy": 5.622292375564575,
      "epoch": 1.8389029371717565,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000465917144974875,
      "loss": 4.9919,
      "mean_token_accuracy": 0.20539594143629075,
      "num_tokens": 41004032.0,
      "step": 23635
    },
    {
      "entropy": 5.636664628982544,
      "epoch": 1.839291966543474,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004659024774829992,
      "loss": 5.0199,
      "mean_token_accuracy": 0.20131931602954864,
      "num_tokens": 41011992.0,
      "step": 23640
    },
    {
      "entropy": 5.653627109527588,
      "epoch": 1.8396809959151916,
      "grad_norm": 1.21875,
      "learning_rate": 0.000465887807094482,
      "loss": 5.0669,
      "mean_token_accuracy": 0.1950000822544098,
      "num_tokens": 41020252.0,
      "step": 23645
    },
    {
      "entropy": 5.631339263916016,
      "epoch": 1.8400700252869093,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046587313380954635,
      "loss": 4.9717,
      "mean_token_accuracy": 0.1985732689499855,
      "num_tokens": 41028858.0,
      "step": 23650
    },
    {
      "entropy": 5.660921096801758,
      "epoch": 1.8404590546586266,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046585845762841453,
      "loss": 4.9573,
      "mean_token_accuracy": 0.20440082550048827,
      "num_tokens": 41037539.0,
      "step": 23655
    },
    {
      "entropy": 5.618165397644043,
      "epoch": 1.8408480840303443,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004658437785513095,
      "loss": 5.0671,
      "mean_token_accuracy": 0.1988501101732254,
      "num_tokens": 41045809.0,
      "step": 23660
    },
    {
      "entropy": 5.549845743179321,
      "epoch": 1.841237113402062,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004658290965784539,
      "loss": 4.9727,
      "mean_token_accuracy": 0.20460626035928725,
      "num_tokens": 41054675.0,
      "step": 23665
    },
    {
      "entropy": 5.679784107208252,
      "epoch": 1.8416261427737795,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004658144117100704,
      "loss": 5.138,
      "mean_token_accuracy": 0.1915805906057358,
      "num_tokens": 41063493.0,
      "step": 23670
    },
    {
      "entropy": 5.6620213985443115,
      "epoch": 1.842015172145497,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046579972394638204,
      "loss": 5.0638,
      "mean_token_accuracy": 0.19647962599992752,
      "num_tokens": 41072985.0,
      "step": 23675
    },
    {
      "entropy": 5.647083711624146,
      "epoch": 1.8424042015172146,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046578503328761146,
      "loss": 5.0378,
      "mean_token_accuracy": 0.204992313683033,
      "num_tokens": 41080926.0,
      "step": 23680
    },
    {
      "entropy": 5.557557487487793,
      "epoch": 1.842793230888932,
      "grad_norm": 1.125,
      "learning_rate": 0.00046577033973398173,
      "loss": 4.9875,
      "mean_token_accuracy": 0.20369023233652114,
      "num_tokens": 41090251.0,
      "step": 23685
    },
    {
      "entropy": 5.560831880569458,
      "epoch": 1.8431822602606496,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004657556432857157,
      "loss": 4.9567,
      "mean_token_accuracy": 0.20601247102022172,
      "num_tokens": 41098681.0,
      "step": 23690
    },
    {
      "entropy": 5.6283434391021725,
      "epoch": 1.8435712896323673,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004657409439430364,
      "loss": 5.0696,
      "mean_token_accuracy": 0.2005976065993309,
      "num_tokens": 41107399.0,
      "step": 23695
    },
    {
      "entropy": 5.700315189361572,
      "epoch": 1.843960319004085,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004657262417061669,
      "loss": 5.0372,
      "mean_token_accuracy": 0.19821143448352813,
      "num_tokens": 41116476.0,
      "step": 23700
    },
    {
      "entropy": 5.600835657119751,
      "epoch": 1.8443493483758022,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004657115365753302,
      "loss": 4.9593,
      "mean_token_accuracy": 0.20475750863552095,
      "num_tokens": 41125191.0,
      "step": 23705
    },
    {
      "entropy": 5.572400856018066,
      "epoch": 1.84473837774752,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046569682855074953,
      "loss": 4.9778,
      "mean_token_accuracy": 0.20813532769680024,
      "num_tokens": 41133811.0,
      "step": 23710
    },
    {
      "entropy": 5.653465032577515,
      "epoch": 1.8451274071192376,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046568211763264796,
      "loss": 5.108,
      "mean_token_accuracy": 0.19387288093566896,
      "num_tokens": 41143093.0,
      "step": 23715
    },
    {
      "entropy": 5.725752353668213,
      "epoch": 1.845516436490955,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004656674038212488,
      "loss": 5.1208,
      "mean_token_accuracy": 0.1990572080016136,
      "num_tokens": 41152106.0,
      "step": 23720
    },
    {
      "entropy": 5.62844033241272,
      "epoch": 1.8459054658626726,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046565268711677525,
      "loss": 4.9997,
      "mean_token_accuracy": 0.19969366639852523,
      "num_tokens": 41161100.0,
      "step": 23725
    },
    {
      "entropy": 5.626623821258545,
      "epoch": 1.8462944952343903,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046563796751945065,
      "loss": 5.0461,
      "mean_token_accuracy": 0.19808444231748581,
      "num_tokens": 41170339.0,
      "step": 23730
    },
    {
      "entropy": 5.680894804000855,
      "epoch": 1.8466835246061077,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046562324502949834,
      "loss": 5.038,
      "mean_token_accuracy": 0.1946217119693756,
      "num_tokens": 41178821.0,
      "step": 23735
    },
    {
      "entropy": 5.667541885375977,
      "epoch": 1.8470725539778252,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004656085196471416,
      "loss": 5.0385,
      "mean_token_accuracy": 0.200397127866745,
      "num_tokens": 41187578.0,
      "step": 23740
    },
    {
      "entropy": 5.511444997787476,
      "epoch": 1.847461583349543,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046559379137260404,
      "loss": 4.9153,
      "mean_token_accuracy": 0.20791668593883514,
      "num_tokens": 41196799.0,
      "step": 23745
    },
    {
      "entropy": 5.624306964874267,
      "epoch": 1.8478506127212606,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000465579060206109,
      "loss": 5.0985,
      "mean_token_accuracy": 0.19296039640903473,
      "num_tokens": 41206300.0,
      "step": 23750
    },
    {
      "entropy": 5.712112379074097,
      "epoch": 1.8482396420929779,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046556432614788015,
      "loss": 5.0934,
      "mean_token_accuracy": 0.1984875023365021,
      "num_tokens": 41214377.0,
      "step": 23755
    },
    {
      "entropy": 5.614333629608154,
      "epoch": 1.8486286714646956,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004655495891981409,
      "loss": 5.0023,
      "mean_token_accuracy": 0.19942827373743058,
      "num_tokens": 41223303.0,
      "step": 23760
    },
    {
      "entropy": 5.655520629882813,
      "epoch": 1.8490177008364133,
      "grad_norm": 1.078125,
      "learning_rate": 0.000465534849357115,
      "loss": 5.1183,
      "mean_token_accuracy": 0.19205860495567323,
      "num_tokens": 41232323.0,
      "step": 23765
    },
    {
      "entropy": 5.666510105133057,
      "epoch": 1.8494067302081307,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046552010662502595,
      "loss": 5.0753,
      "mean_token_accuracy": 0.19606455862522126,
      "num_tokens": 41241605.0,
      "step": 23770
    },
    {
      "entropy": 5.698424673080444,
      "epoch": 1.8497957595798482,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004655053610020976,
      "loss": 5.1501,
      "mean_token_accuracy": 0.1843496799468994,
      "num_tokens": 41250604.0,
      "step": 23775
    },
    {
      "entropy": 5.625069189071655,
      "epoch": 1.850184788951566,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004654906124885535,
      "loss": 4.9521,
      "mean_token_accuracy": 0.20801903009414674,
      "num_tokens": 41258640.0,
      "step": 23780
    },
    {
      "entropy": 5.653652477264404,
      "epoch": 1.8505738183232834,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004654758610846176,
      "loss": 5.0938,
      "mean_token_accuracy": 0.2022511512041092,
      "num_tokens": 41268104.0,
      "step": 23785
    },
    {
      "entropy": 5.6686872959136965,
      "epoch": 1.8509628476950009,
      "grad_norm": 1.25,
      "learning_rate": 0.0004654611067905137,
      "loss": 5.1168,
      "mean_token_accuracy": 0.19575035870075225,
      "num_tokens": 41276276.0,
      "step": 23790
    },
    {
      "entropy": 5.640301084518432,
      "epoch": 1.8513518770667186,
      "grad_norm": 1.125,
      "learning_rate": 0.00046544634960646563,
      "loss": 4.9588,
      "mean_token_accuracy": 0.20497772395610808,
      "num_tokens": 41284464.0,
      "step": 23795
    },
    {
      "entropy": 5.627992153167725,
      "epoch": 1.8517409064384363,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004654315895326974,
      "loss": 4.9579,
      "mean_token_accuracy": 0.20381938070058822,
      "num_tokens": 41293108.0,
      "step": 23800
    },
    {
      "entropy": 5.62423152923584,
      "epoch": 1.8521299358101535,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004654168265694328,
      "loss": 5.0092,
      "mean_token_accuracy": 0.198760886490345,
      "num_tokens": 41301830.0,
      "step": 23805
    },
    {
      "entropy": 5.671004009246826,
      "epoch": 1.8525189651818712,
      "grad_norm": 1.1875,
      "learning_rate": 0.000465402060716896,
      "loss": 5.0533,
      "mean_token_accuracy": 0.19943101108074188,
      "num_tokens": 41310487.0,
      "step": 23810
    },
    {
      "entropy": 5.645656681060791,
      "epoch": 1.852907994553589,
      "grad_norm": 1.125,
      "learning_rate": 0.000465387291975311,
      "loss": 5.1213,
      "mean_token_accuracy": 0.1944141760468483,
      "num_tokens": 41319730.0,
      "step": 23815
    },
    {
      "entropy": 5.659071969985962,
      "epoch": 1.8532970239253064,
      "grad_norm": 1.25,
      "learning_rate": 0.00046537252034490185,
      "loss": 5.0275,
      "mean_token_accuracy": 0.20051229894161224,
      "num_tokens": 41328914.0,
      "step": 23820
    },
    {
      "entropy": 5.641670274734497,
      "epoch": 1.8536860532970239,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046535774582589275,
      "loss": 5.0674,
      "mean_token_accuracy": 0.19726224839687348,
      "num_tokens": 41337060.0,
      "step": 23825
    },
    {
      "entropy": 5.644946193695068,
      "epoch": 1.8540750826687415,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046534296841850776,
      "loss": 4.9857,
      "mean_token_accuracy": 0.20408716052770615,
      "num_tokens": 41344966.0,
      "step": 23830
    },
    {
      "entropy": 5.558892774581909,
      "epoch": 1.854464112040459,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046532818812297124,
      "loss": 4.9036,
      "mean_token_accuracy": 0.2084477350115776,
      "num_tokens": 41353678.0,
      "step": 23835
    },
    {
      "entropy": 5.704274845123291,
      "epoch": 1.8548531414121765,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004653134049395074,
      "loss": 5.165,
      "mean_token_accuracy": 0.19792997539043428,
      "num_tokens": 41362782.0,
      "step": 23840
    },
    {
      "entropy": 5.606107568740844,
      "epoch": 1.8552421707838942,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004652986188683406,
      "loss": 4.9902,
      "mean_token_accuracy": 0.20383159518241883,
      "num_tokens": 41370668.0,
      "step": 23845
    },
    {
      "entropy": 5.56073842048645,
      "epoch": 1.855631200155612,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046528382990969504,
      "loss": 4.9819,
      "mean_token_accuracy": 0.20325508564710618,
      "num_tokens": 41378945.0,
      "step": 23850
    },
    {
      "entropy": 5.677216100692749,
      "epoch": 1.8560202295273294,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004652690380637953,
      "loss": 5.0618,
      "mean_token_accuracy": 0.19687515497207642,
      "num_tokens": 41387579.0,
      "step": 23855
    },
    {
      "entropy": 5.652396202087402,
      "epoch": 1.8564092588990468,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004652542433308657,
      "loss": 5.0498,
      "mean_token_accuracy": 0.2057149201631546,
      "num_tokens": 41396472.0,
      "step": 23860
    },
    {
      "entropy": 5.664175510406494,
      "epoch": 1.8567982882707645,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004652394457111309,
      "loss": 5.1032,
      "mean_token_accuracy": 0.19284926801919938,
      "num_tokens": 41404599.0,
      "step": 23865
    },
    {
      "entropy": 5.6889043807983395,
      "epoch": 1.857187317642482,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046522464520481517,
      "loss": 5.129,
      "mean_token_accuracy": 0.1895430013537407,
      "num_tokens": 41412799.0,
      "step": 23870
    },
    {
      "entropy": 5.611789655685425,
      "epoch": 1.8575763470141995,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046520984181214333,
      "loss": 4.9305,
      "mean_token_accuracy": 0.20889392197132112,
      "num_tokens": 41421193.0,
      "step": 23875
    },
    {
      "entropy": 5.606680154800415,
      "epoch": 1.8579653763859172,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004651950355333398,
      "loss": 5.0074,
      "mean_token_accuracy": 0.19574365764856339,
      "num_tokens": 41430201.0,
      "step": 23880
    },
    {
      "entropy": 5.513489151000977,
      "epoch": 1.8583544057576347,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004651802263686294,
      "loss": 4.9666,
      "mean_token_accuracy": 0.2042250543832779,
      "num_tokens": 41439122.0,
      "step": 23885
    },
    {
      "entropy": 5.601866436004639,
      "epoch": 1.8587434351293521,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004651654143182367,
      "loss": 5.008,
      "mean_token_accuracy": 0.19989681392908096,
      "num_tokens": 41448553.0,
      "step": 23890
    },
    {
      "entropy": 5.61417818069458,
      "epoch": 1.8591324645010698,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004651505993823866,
      "loss": 4.9919,
      "mean_token_accuracy": 0.20259442776441575,
      "num_tokens": 41457098.0,
      "step": 23895
    },
    {
      "entropy": 5.63874192237854,
      "epoch": 1.8595214938727875,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046513578156130383,
      "loss": 5.0547,
      "mean_token_accuracy": 0.19884778261184693,
      "num_tokens": 41465714.0,
      "step": 23900
    },
    {
      "entropy": 5.687614917755127,
      "epoch": 1.859910523244505,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004651209608552131,
      "loss": 5.0655,
      "mean_token_accuracy": 0.20034240931272507,
      "num_tokens": 41474418.0,
      "step": 23905
    },
    {
      "entropy": 5.580568408966064,
      "epoch": 1.8602995526162225,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046510613726433945,
      "loss": 4.9515,
      "mean_token_accuracy": 0.20689060240983964,
      "num_tokens": 41482430.0,
      "step": 23910
    },
    {
      "entropy": 5.68475308418274,
      "epoch": 1.8606885819879402,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004650913107889078,
      "loss": 5.1105,
      "mean_token_accuracy": 0.1981865346431732,
      "num_tokens": 41490759.0,
      "step": 23915
    },
    {
      "entropy": 5.555719614028931,
      "epoch": 1.8610776113596577,
      "grad_norm": 1.09375,
      "learning_rate": 0.000465076481429143,
      "loss": 4.9791,
      "mean_token_accuracy": 0.20109595358371735,
      "num_tokens": 41499660.0,
      "step": 23920
    },
    {
      "entropy": 5.598991966247558,
      "epoch": 1.8614666407313751,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004650616491852701,
      "loss": 4.9884,
      "mean_token_accuracy": 0.20436572134494782,
      "num_tokens": 41507515.0,
      "step": 23925
    },
    {
      "entropy": 5.634622383117676,
      "epoch": 1.8618556701030928,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046504681405751426,
      "loss": 5.0646,
      "mean_token_accuracy": 0.19026773571968078,
      "num_tokens": 41515859.0,
      "step": 23930
    },
    {
      "entropy": 5.626671171188354,
      "epoch": 1.8622446994748103,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046503197604610047,
      "loss": 5.084,
      "mean_token_accuracy": 0.20466479063034057,
      "num_tokens": 41524776.0,
      "step": 23935
    },
    {
      "entropy": 5.689475965499878,
      "epoch": 1.8626337288465278,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046501713515125393,
      "loss": 5.0087,
      "mean_token_accuracy": 0.20334266871213913,
      "num_tokens": 41532854.0,
      "step": 23940
    },
    {
      "entropy": 5.669583988189697,
      "epoch": 1.8630227582182455,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004650022913731998,
      "loss": 4.9973,
      "mean_token_accuracy": 0.20839875787496567,
      "num_tokens": 41542038.0,
      "step": 23945
    },
    {
      "entropy": 5.605361080169677,
      "epoch": 1.8634117875899632,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046498744471216335,
      "loss": 5.0107,
      "mean_token_accuracy": 0.19885413944721222,
      "num_tokens": 41550428.0,
      "step": 23950
    },
    {
      "entropy": 5.586767625808716,
      "epoch": 1.8638008169616807,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046497259516836974,
      "loss": 4.9821,
      "mean_token_accuracy": 0.1993659868836403,
      "num_tokens": 41559289.0,
      "step": 23955
    },
    {
      "entropy": 5.643695497512818,
      "epoch": 1.8641898463333981,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046495774274204446,
      "loss": 4.9927,
      "mean_token_accuracy": 0.21045035719871522,
      "num_tokens": 41567901.0,
      "step": 23960
    },
    {
      "entropy": 5.724760055541992,
      "epoch": 1.8645788757051158,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004649428874334127,
      "loss": 5.1054,
      "mean_token_accuracy": 0.19264122247695922,
      "num_tokens": 41577589.0,
      "step": 23965
    },
    {
      "entropy": 5.664262962341309,
      "epoch": 1.8649679050768333,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004649280292427,
      "loss": 5.1103,
      "mean_token_accuracy": 0.19481078684329986,
      "num_tokens": 41586319.0,
      "step": 23970
    },
    {
      "entropy": 5.680643463134766,
      "epoch": 1.8653569344485508,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004649131681701318,
      "loss": 5.0688,
      "mean_token_accuracy": 0.19689726829528809,
      "num_tokens": 41594316.0,
      "step": 23975
    },
    {
      "entropy": 5.635032749176025,
      "epoch": 1.8657459638202685,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046489830421593347,
      "loss": 5.0182,
      "mean_token_accuracy": 0.20163055658340454,
      "num_tokens": 41603221.0,
      "step": 23980
    },
    {
      "entropy": 5.588497972488403,
      "epoch": 1.866134993191986,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004648834373803307,
      "loss": 5.0244,
      "mean_token_accuracy": 0.19905508905649186,
      "num_tokens": 41612082.0,
      "step": 23985
    },
    {
      "entropy": 5.661404466629028,
      "epoch": 1.8665240225637034,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046486856766354893,
      "loss": 4.9748,
      "mean_token_accuracy": 0.20424366593360901,
      "num_tokens": 41619762.0,
      "step": 23990
    },
    {
      "entropy": 5.58872356414795,
      "epoch": 1.8669130519354211,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046485369506581387,
      "loss": 4.9785,
      "mean_token_accuracy": 0.20352459847927093,
      "num_tokens": 41627968.0,
      "step": 23995
    },
    {
      "entropy": 5.646211910247803,
      "epoch": 1.8673020813071388,
      "grad_norm": 1.125,
      "learning_rate": 0.00046483881958735124,
      "loss": 5.0746,
      "mean_token_accuracy": 0.19413445442914962,
      "num_tokens": 41636841.0,
      "step": 24000
    },
    {
      "epoch": 1.8673020813071388,
      "eval_entropy": 5.451980794680819,
      "eval_loss": 5.086483001708984,
      "eval_mean_token_accuracy": 0.2072724711433182,
      "eval_num_tokens": 41636841.0,
      "eval_runtime": 21.6418,
      "eval_samples_per_second": 1920.268,
      "eval_steps_per_second": 240.045,
      "step": 24000
    },
    {
      "entropy": 5.715573835372925,
      "epoch": 1.8676911106788563,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046482394122838663,
      "loss": 5.0195,
      "mean_token_accuracy": 0.2034873127937317,
      "num_tokens": 41645352.0,
      "step": 24005
    },
    {
      "entropy": 5.62004017829895,
      "epoch": 1.8680801400505738,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046480905998914587,
      "loss": 4.9362,
      "mean_token_accuracy": 0.207021501660347,
      "num_tokens": 41654296.0,
      "step": 24010
    },
    {
      "entropy": 5.557055282592773,
      "epoch": 1.8684691694222915,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004647941758698547,
      "loss": 4.9949,
      "mean_token_accuracy": 0.19812235981225967,
      "num_tokens": 41662008.0,
      "step": 24015
    },
    {
      "entropy": 5.639769124984741,
      "epoch": 1.868858198794009,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000464779288870739,
      "loss": 5.1423,
      "mean_token_accuracy": 0.19511993527412413,
      "num_tokens": 41671765.0,
      "step": 24020
    },
    {
      "entropy": 5.711396169662476,
      "epoch": 1.8692472281657264,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046476439899202464,
      "loss": 5.1274,
      "mean_token_accuracy": 0.1911906123161316,
      "num_tokens": 41681035.0,
      "step": 24025
    },
    {
      "entropy": 5.657670354843139,
      "epoch": 1.8696362575374441,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046474950623393756,
      "loss": 4.9883,
      "mean_token_accuracy": 0.20573211908340455,
      "num_tokens": 41689974.0,
      "step": 24030
    },
    {
      "entropy": 5.604171657562256,
      "epoch": 1.8700252869091616,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004647346105967038,
      "loss": 5.0467,
      "mean_token_accuracy": 0.19555254876613617,
      "num_tokens": 41698560.0,
      "step": 24035
    },
    {
      "entropy": 5.633584499359131,
      "epoch": 1.870414316280879,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004647197120805493,
      "loss": 5.0147,
      "mean_token_accuracy": 0.2044930338859558,
      "num_tokens": 41706741.0,
      "step": 24040
    },
    {
      "entropy": 5.624234676361084,
      "epoch": 1.8708033456525968,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046470481068570013,
      "loss": 4.9924,
      "mean_token_accuracy": 0.197985877096653,
      "num_tokens": 41715685.0,
      "step": 24045
    },
    {
      "entropy": 5.5964837074279785,
      "epoch": 1.8711923750243145,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004646899064123824,
      "loss": 5.071,
      "mean_token_accuracy": 0.19954553246498108,
      "num_tokens": 41724123.0,
      "step": 24050
    },
    {
      "entropy": 5.664126586914063,
      "epoch": 1.871581404396032,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004646749992608223,
      "loss": 5.0293,
      "mean_token_accuracy": 0.20310772359371185,
      "num_tokens": 41732604.0,
      "step": 24055
    },
    {
      "entropy": 5.567569637298584,
      "epoch": 1.8719704337677494,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004646600892312459,
      "loss": 4.9966,
      "mean_token_accuracy": 0.20097895860671997,
      "num_tokens": 41740444.0,
      "step": 24060
    },
    {
      "entropy": 5.603746557235718,
      "epoch": 1.872359463139467,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004646451763238796,
      "loss": 4.9912,
      "mean_token_accuracy": 0.19906289577484132,
      "num_tokens": 41748800.0,
      "step": 24065
    },
    {
      "entropy": 5.604099082946777,
      "epoch": 1.8727484925111846,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004646302605389496,
      "loss": 5.0307,
      "mean_token_accuracy": 0.1970682844519615,
      "num_tokens": 41757402.0,
      "step": 24070
    },
    {
      "entropy": 5.656908798217773,
      "epoch": 1.873137521882902,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004646153418766822,
      "loss": 5.0052,
      "mean_token_accuracy": 0.2088899090886116,
      "num_tokens": 41766604.0,
      "step": 24075
    },
    {
      "entropy": 5.703961753845215,
      "epoch": 1.8735265512546198,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046460042033730377,
      "loss": 5.1785,
      "mean_token_accuracy": 0.18879443705081939,
      "num_tokens": 41775450.0,
      "step": 24080
    },
    {
      "entropy": 5.664543533325196,
      "epoch": 1.8739155806263375,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004645854959210408,
      "loss": 5.1566,
      "mean_token_accuracy": 0.1986447349190712,
      "num_tokens": 41784355.0,
      "step": 24085
    },
    {
      "entropy": 5.609068393707275,
      "epoch": 1.8743046099980547,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046457056862811956,
      "loss": 4.9852,
      "mean_token_accuracy": 0.1970805287361145,
      "num_tokens": 41793441.0,
      "step": 24090
    },
    {
      "entropy": 5.638676881790161,
      "epoch": 1.8746936393697724,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004645556384587668,
      "loss": 4.9605,
      "mean_token_accuracy": 0.2019784852862358,
      "num_tokens": 41802407.0,
      "step": 24095
    },
    {
      "entropy": 5.662277507781982,
      "epoch": 1.87508266874149,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004645407054132089,
      "loss": 5.0325,
      "mean_token_accuracy": 0.1958734482526779,
      "num_tokens": 41810362.0,
      "step": 24100
    },
    {
      "entropy": 5.625180006027222,
      "epoch": 1.8754716981132076,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004645257694916725,
      "loss": 5.1369,
      "mean_token_accuracy": 0.19154489487409593,
      "num_tokens": 41819332.0,
      "step": 24105
    },
    {
      "entropy": 5.64078803062439,
      "epoch": 1.875860727484925,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004645108306943842,
      "loss": 4.9931,
      "mean_token_accuracy": 0.20537394881248475,
      "num_tokens": 41828048.0,
      "step": 24110
    },
    {
      "entropy": 5.600185203552246,
      "epoch": 1.8762497568566427,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004644958890215707,
      "loss": 4.9725,
      "mean_token_accuracy": 0.20183290541172028,
      "num_tokens": 41836583.0,
      "step": 24115
    },
    {
      "entropy": 5.604400587081909,
      "epoch": 1.8766387862283602,
      "grad_norm": 1.25,
      "learning_rate": 0.0004644809444734586,
      "loss": 5.0245,
      "mean_token_accuracy": 0.19665726274251938,
      "num_tokens": 41845454.0,
      "step": 24120
    },
    {
      "entropy": 5.622054147720337,
      "epoch": 1.8770278156000777,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046446599705027487,
      "loss": 5.0096,
      "mean_token_accuracy": 0.19856580942869187,
      "num_tokens": 41854481.0,
      "step": 24125
    },
    {
      "entropy": 5.66301474571228,
      "epoch": 1.8774168449717954,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046445104675224607,
      "loss": 5.0618,
      "mean_token_accuracy": 0.19987966418266295,
      "num_tokens": 41863015.0,
      "step": 24130
    },
    {
      "entropy": 5.666498851776123,
      "epoch": 1.877805874343513,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004644360935795993,
      "loss": 5.044,
      "mean_token_accuracy": 0.1952373594045639,
      "num_tokens": 41871703.0,
      "step": 24135
    },
    {
      "entropy": 5.652878332138061,
      "epoch": 1.8781949037152303,
      "grad_norm": 1.125,
      "learning_rate": 0.00046442113753256126,
      "loss": 4.9716,
      "mean_token_accuracy": 0.1986891120672226,
      "num_tokens": 41880136.0,
      "step": 24140
    },
    {
      "entropy": 5.684490394592285,
      "epoch": 1.878583933086948,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046440617861135905,
      "loss": 5.1034,
      "mean_token_accuracy": 0.19204966723918915,
      "num_tokens": 41889065.0,
      "step": 24145
    },
    {
      "entropy": 5.589878225326538,
      "epoch": 1.8789729624586657,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004643912168162194,
      "loss": 4.9584,
      "mean_token_accuracy": 0.2080230951309204,
      "num_tokens": 41898160.0,
      "step": 24150
    },
    {
      "entropy": 5.657149934768677,
      "epoch": 1.8793619918303832,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004643762521473696,
      "loss": 5.0464,
      "mean_token_accuracy": 0.1975500091910362,
      "num_tokens": 41907037.0,
      "step": 24155
    },
    {
      "entropy": 5.686665344238281,
      "epoch": 1.8797510212021007,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004643612846050366,
      "loss": 5.1001,
      "mean_token_accuracy": 0.19433003664016724,
      "num_tokens": 41915176.0,
      "step": 24160
    },
    {
      "entropy": 5.7039406299591064,
      "epoch": 1.8801400505738184,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046434631418944744,
      "loss": 5.076,
      "mean_token_accuracy": 0.1939019188284874,
      "num_tokens": 41923586.0,
      "step": 24165
    },
    {
      "entropy": 5.652913188934326,
      "epoch": 1.8805290799455359,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004643313409008294,
      "loss": 5.1581,
      "mean_token_accuracy": 0.1885446935892105,
      "num_tokens": 41932803.0,
      "step": 24170
    },
    {
      "entropy": 5.651459836959839,
      "epoch": 1.8809181093172533,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046431636473940956,
      "loss": 5.06,
      "mean_token_accuracy": 0.197680002450943,
      "num_tokens": 41941737.0,
      "step": 24175
    },
    {
      "entropy": 5.659139347076416,
      "epoch": 1.881307138688971,
      "grad_norm": 1.25,
      "learning_rate": 0.0004643013857054152,
      "loss": 4.9879,
      "mean_token_accuracy": 0.19936260879039763,
      "num_tokens": 41950099.0,
      "step": 24180
    },
    {
      "entropy": 5.6279871463775635,
      "epoch": 1.8816961680606887,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046428640379907367,
      "loss": 5.0476,
      "mean_token_accuracy": 0.19482465088367462,
      "num_tokens": 41959091.0,
      "step": 24185
    },
    {
      "entropy": 5.651910829544067,
      "epoch": 1.882085197432406,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046427141902061225,
      "loss": 4.9544,
      "mean_token_accuracy": 0.20624804645776748,
      "num_tokens": 41968462.0,
      "step": 24190
    },
    {
      "entropy": 5.633303689956665,
      "epoch": 1.8824742268041237,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004642564313702582,
      "loss": 5.0242,
      "mean_token_accuracy": 0.19739639908075332,
      "num_tokens": 41976514.0,
      "step": 24195
    },
    {
      "entropy": 5.5897644519805905,
      "epoch": 1.8828632561758414,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046424144084823916,
      "loss": 4.9327,
      "mean_token_accuracy": 0.2036871388554573,
      "num_tokens": 41985583.0,
      "step": 24200
    },
    {
      "entropy": 5.566992712020874,
      "epoch": 1.8832522855475589,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004642264474547824,
      "loss": 4.9615,
      "mean_token_accuracy": 0.19973803907632828,
      "num_tokens": 41993684.0,
      "step": 24205
    },
    {
      "entropy": 5.568357229232788,
      "epoch": 1.8836413149192763,
      "grad_norm": 1.125,
      "learning_rate": 0.00046421145119011556,
      "loss": 4.9269,
      "mean_token_accuracy": 0.20974594801664354,
      "num_tokens": 42002239.0,
      "step": 24210
    },
    {
      "entropy": 5.54041485786438,
      "epoch": 1.884030344290994,
      "grad_norm": 1.15625,
      "learning_rate": 0.000464196452054466,
      "loss": 4.9022,
      "mean_token_accuracy": 0.21332722902297974,
      "num_tokens": 42010886.0,
      "step": 24215
    },
    {
      "entropy": 5.642314338684082,
      "epoch": 1.8844193736627115,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004641814500480615,
      "loss": 5.0832,
      "mean_token_accuracy": 0.19305527210235596,
      "num_tokens": 42019703.0,
      "step": 24220
    },
    {
      "entropy": 5.685856103897095,
      "epoch": 1.884808403034429,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004641664451711296,
      "loss": 5.1275,
      "mean_token_accuracy": 0.19278821647167205,
      "num_tokens": 42028467.0,
      "step": 24225
    },
    {
      "entropy": 5.652223443984985,
      "epoch": 1.8851974324061467,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046415143742389793,
      "loss": 5.0144,
      "mean_token_accuracy": 0.19974922388792038,
      "num_tokens": 42037220.0,
      "step": 24230
    },
    {
      "entropy": 5.686418676376343,
      "epoch": 1.8855864617778644,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004641364268065943,
      "loss": 5.0725,
      "mean_token_accuracy": 0.1949046492576599,
      "num_tokens": 42046034.0,
      "step": 24235
    },
    {
      "entropy": 5.612969255447387,
      "epoch": 1.8859754911495816,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046412141331944644,
      "loss": 5.0165,
      "mean_token_accuracy": 0.20125429034233094,
      "num_tokens": 42054659.0,
      "step": 24240
    },
    {
      "entropy": 5.669411516189575,
      "epoch": 1.8863645205212993,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046410639696268206,
      "loss": 5.0774,
      "mean_token_accuracy": 0.19198661744594575,
      "num_tokens": 42063130.0,
      "step": 24245
    },
    {
      "entropy": 5.669471645355225,
      "epoch": 1.886753549893017,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004640913777365292,
      "loss": 5.085,
      "mean_token_accuracy": 0.193919537961483,
      "num_tokens": 42071162.0,
      "step": 24250
    },
    {
      "entropy": 5.633266735076904,
      "epoch": 1.8871425792647345,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046407635564121565,
      "loss": 5.0806,
      "mean_token_accuracy": 0.20278292149305344,
      "num_tokens": 42079948.0,
      "step": 24255
    },
    {
      "entropy": 5.644159841537475,
      "epoch": 1.887531608636452,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046406133067696936,
      "loss": 5.0276,
      "mean_token_accuracy": 0.19592273980379105,
      "num_tokens": 42089010.0,
      "step": 24260
    },
    {
      "entropy": 5.699525642395019,
      "epoch": 1.8879206380081697,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004640463028440182,
      "loss": 5.0347,
      "mean_token_accuracy": 0.19488841742277146,
      "num_tokens": 42097096.0,
      "step": 24265
    },
    {
      "entropy": 5.566258478164673,
      "epoch": 1.8883096673798871,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004640312721425904,
      "loss": 4.9415,
      "mean_token_accuracy": 0.20562813580036163,
      "num_tokens": 42105870.0,
      "step": 24270
    },
    {
      "entropy": 5.570469236373901,
      "epoch": 1.8886986967516046,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004640162385729139,
      "loss": 4.9583,
      "mean_token_accuracy": 0.20743005722761154,
      "num_tokens": 42114246.0,
      "step": 24275
    },
    {
      "entropy": 5.585826826095581,
      "epoch": 1.8890877261233223,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004640012021352168,
      "loss": 4.9712,
      "mean_token_accuracy": 0.19971460700035096,
      "num_tokens": 42122163.0,
      "step": 24280
    },
    {
      "entropy": 5.639885330200196,
      "epoch": 1.88947675549504,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004639861628297273,
      "loss": 5.005,
      "mean_token_accuracy": 0.20067329555749894,
      "num_tokens": 42130255.0,
      "step": 24285
    },
    {
      "entropy": 5.684164381027221,
      "epoch": 1.8898657848667575,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046397112065667354,
      "loss": 5.0577,
      "mean_token_accuracy": 0.196845705807209,
      "num_tokens": 42138615.0,
      "step": 24290
    },
    {
      "entropy": 5.632627964019775,
      "epoch": 1.890254814238475,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046395607561628387,
      "loss": 4.9512,
      "mean_token_accuracy": 0.20552507489919664,
      "num_tokens": 42146502.0,
      "step": 24295
    },
    {
      "entropy": 5.59929347038269,
      "epoch": 1.8906438436101927,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046394102770878643,
      "loss": 5.0402,
      "mean_token_accuracy": 0.19964609593153,
      "num_tokens": 42155548.0,
      "step": 24300
    },
    {
      "entropy": 5.525604486465454,
      "epoch": 1.8910328729819101,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046392597693440974,
      "loss": 4.8488,
      "mean_token_accuracy": 0.2119812026619911,
      "num_tokens": 42163813.0,
      "step": 24305
    },
    {
      "entropy": 5.689184045791626,
      "epoch": 1.8914219023536276,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004639109232933819,
      "loss": 5.0695,
      "mean_token_accuracy": 0.19636989533901214,
      "num_tokens": 42172702.0,
      "step": 24310
    },
    {
      "entropy": 5.700816106796265,
      "epoch": 1.8918109317253453,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004638958667859316,
      "loss": 5.1022,
      "mean_token_accuracy": 0.19660286754369735,
      "num_tokens": 42180776.0,
      "step": 24315
    },
    {
      "entropy": 5.686199712753296,
      "epoch": 1.8921999610970628,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004638808074122872,
      "loss": 5.0535,
      "mean_token_accuracy": 0.19616233706474304,
      "num_tokens": 42189695.0,
      "step": 24320
    },
    {
      "entropy": 5.626562452316284,
      "epoch": 1.8925889904687803,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004638657451726771,
      "loss": 5.0171,
      "mean_token_accuracy": 0.19584813863039016,
      "num_tokens": 42197884.0,
      "step": 24325
    },
    {
      "entropy": 5.675503778457641,
      "epoch": 1.892978019840498,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046385068006732995,
      "loss": 5.1275,
      "mean_token_accuracy": 0.19184932857751846,
      "num_tokens": 42206800.0,
      "step": 24330
    },
    {
      "entropy": 5.634459829330444,
      "epoch": 1.8933670492122157,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046383561209647437,
      "loss": 4.9908,
      "mean_token_accuracy": 0.2106849119067192,
      "num_tokens": 42216091.0,
      "step": 24335
    },
    {
      "entropy": 5.616139984130859,
      "epoch": 1.8937560785839331,
      "grad_norm": 1.125,
      "learning_rate": 0.00046382054126033884,
      "loss": 5.0336,
      "mean_token_accuracy": 0.19986066967248917,
      "num_tokens": 42224675.0,
      "step": 24340
    },
    {
      "entropy": 5.659064626693725,
      "epoch": 1.8941451079556506,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004638054675591523,
      "loss": 5.0606,
      "mean_token_accuracy": 0.19514855295419692,
      "num_tokens": 42232964.0,
      "step": 24345
    },
    {
      "entropy": 5.6641490936279295,
      "epoch": 1.8945341373273683,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004637903909931432,
      "loss": 5.0573,
      "mean_token_accuracy": 0.19220776706933976,
      "num_tokens": 42241932.0,
      "step": 24350
    },
    {
      "entropy": 5.600286483764648,
      "epoch": 1.8949231666990858,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004637753115625404,
      "loss": 4.8701,
      "mean_token_accuracy": 0.2033713459968567,
      "num_tokens": 42250146.0,
      "step": 24355
    },
    {
      "entropy": 5.667975759506225,
      "epoch": 1.8953121960708033,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004637602292675726,
      "loss": 5.1215,
      "mean_token_accuracy": 0.19113409519195557,
      "num_tokens": 42259183.0,
      "step": 24360
    },
    {
      "entropy": 5.669849300384522,
      "epoch": 1.895701225442521,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004637451441084689,
      "loss": 5.1325,
      "mean_token_accuracy": 0.1883016988635063,
      "num_tokens": 42268436.0,
      "step": 24365
    },
    {
      "entropy": 5.704381847381592,
      "epoch": 1.8960902548142384,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004637300560854581,
      "loss": 5.0892,
      "mean_token_accuracy": 0.19984538108110428,
      "num_tokens": 42277146.0,
      "step": 24370
    },
    {
      "entropy": 5.739028835296631,
      "epoch": 1.896479284185956,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000463714965198769,
      "loss": 5.0371,
      "mean_token_accuracy": 0.1973297968506813,
      "num_tokens": 42286571.0,
      "step": 24375
    },
    {
      "entropy": 5.627263879776001,
      "epoch": 1.8968683135576736,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004636998714486307,
      "loss": 4.9678,
      "mean_token_accuracy": 0.20563387572765351,
      "num_tokens": 42295415.0,
      "step": 24380
    },
    {
      "entropy": 5.613515329360962,
      "epoch": 1.8972573429293913,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046368477483527224,
      "loss": 5.0408,
      "mean_token_accuracy": 0.19516167640686036,
      "num_tokens": 42305105.0,
      "step": 24385
    },
    {
      "entropy": 5.634998083114624,
      "epoch": 1.8976463723011088,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004636696753589226,
      "loss": 4.9434,
      "mean_token_accuracy": 0.20307771265506744,
      "num_tokens": 42313034.0,
      "step": 24390
    },
    {
      "entropy": 5.628355979919434,
      "epoch": 1.8980354016728263,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004636545730198109,
      "loss": 5.0121,
      "mean_token_accuracy": 0.20169672667980193,
      "num_tokens": 42321249.0,
      "step": 24395
    },
    {
      "entropy": 5.600574254989624,
      "epoch": 1.898424431044544,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046363946781816643,
      "loss": 5.0277,
      "mean_token_accuracy": 0.20129622668027877,
      "num_tokens": 42329519.0,
      "step": 24400
    },
    {
      "entropy": 5.681989908218384,
      "epoch": 1.8988134604162614,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046362435975421816,
      "loss": 5.0933,
      "mean_token_accuracy": 0.19677205085754396,
      "num_tokens": 42338318.0,
      "step": 24405
    },
    {
      "entropy": 5.64445858001709,
      "epoch": 1.899202489787979,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046360924882819554,
      "loss": 5.0232,
      "mean_token_accuracy": 0.20437194854021073,
      "num_tokens": 42346038.0,
      "step": 24410
    },
    {
      "entropy": 5.611570739746094,
      "epoch": 1.8995915191596966,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004635941350403277,
      "loss": 4.9864,
      "mean_token_accuracy": 0.20351084917783738,
      "num_tokens": 42356280.0,
      "step": 24415
    },
    {
      "entropy": 5.577451086044311,
      "epoch": 1.899980548531414,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004635790183908442,
      "loss": 4.9458,
      "mean_token_accuracy": 0.21141845136880874,
      "num_tokens": 42364716.0,
      "step": 24420
    },
    {
      "entropy": 5.639925479888916,
      "epoch": 1.9003695779031315,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046356389887997415,
      "loss": 5.0846,
      "mean_token_accuracy": 0.1967342808842659,
      "num_tokens": 42373137.0,
      "step": 24425
    },
    {
      "entropy": 5.6386326313018795,
      "epoch": 1.9007586072748492,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046354877650794707,
      "loss": 4.9879,
      "mean_token_accuracy": 0.1931294709444046,
      "num_tokens": 42382054.0,
      "step": 24430
    },
    {
      "entropy": 5.600449562072754,
      "epoch": 1.901147636646567,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046353365127499235,
      "loss": 4.9889,
      "mean_token_accuracy": 0.1989274278283119,
      "num_tokens": 42391228.0,
      "step": 24435
    },
    {
      "entropy": 5.66278395652771,
      "epoch": 1.9015366660182844,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004635185231813396,
      "loss": 4.967,
      "mean_token_accuracy": 0.20482051521539688,
      "num_tokens": 42400178.0,
      "step": 24440
    },
    {
      "entropy": 5.633172559738159,
      "epoch": 1.901925695390002,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004635033922272183,
      "loss": 5.0128,
      "mean_token_accuracy": 0.20630544573068618,
      "num_tokens": 42408515.0,
      "step": 24445
    },
    {
      "entropy": 5.6902485370635985,
      "epoch": 1.9023147247617196,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046348825841285813,
      "loss": 5.0722,
      "mean_token_accuracy": 0.19617781043052673,
      "num_tokens": 42417291.0,
      "step": 24450
    },
    {
      "entropy": 5.745232057571411,
      "epoch": 1.902703754133437,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004634731217384886,
      "loss": 5.2083,
      "mean_token_accuracy": 0.1841219663619995,
      "num_tokens": 42426371.0,
      "step": 24455
    },
    {
      "entropy": 5.72010440826416,
      "epoch": 1.9030927835051545,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004634579822043394,
      "loss": 5.1599,
      "mean_token_accuracy": 0.19807369709014894,
      "num_tokens": 42434767.0,
      "step": 24460
    },
    {
      "entropy": 5.639829063415528,
      "epoch": 1.9034818128768722,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046344283981064035,
      "loss": 5.0483,
      "mean_token_accuracy": 0.19719869941473006,
      "num_tokens": 42443618.0,
      "step": 24465
    },
    {
      "entropy": 5.630080509185791,
      "epoch": 1.90387084224859,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046342769455762114,
      "loss": 5.012,
      "mean_token_accuracy": 0.19952094852924346,
      "num_tokens": 42453134.0,
      "step": 24470
    },
    {
      "entropy": 5.675983524322509,
      "epoch": 1.9042598716203072,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004634125464455116,
      "loss": 5.0389,
      "mean_token_accuracy": 0.19640731513500215,
      "num_tokens": 42462037.0,
      "step": 24475
    },
    {
      "entropy": 5.616878652572632,
      "epoch": 1.9046489009920249,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046339739547454146,
      "loss": 4.9435,
      "mean_token_accuracy": 0.20552055388689042,
      "num_tokens": 42471318.0,
      "step": 24480
    },
    {
      "entropy": 5.631363439559936,
      "epoch": 1.9050379303637426,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046338224164494074,
      "loss": 5.0615,
      "mean_token_accuracy": 0.1985990047454834,
      "num_tokens": 42478851.0,
      "step": 24485
    },
    {
      "entropy": 5.572282218933106,
      "epoch": 1.90542695973546,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046336708495693933,
      "loss": 4.9861,
      "mean_token_accuracy": 0.19799603372812272,
      "num_tokens": 42488406.0,
      "step": 24490
    },
    {
      "entropy": 5.656839895248413,
      "epoch": 1.9058159891071775,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046335192541076725,
      "loss": 4.9043,
      "mean_token_accuracy": 0.2085791438817978,
      "num_tokens": 42496001.0,
      "step": 24495
    },
    {
      "entropy": 5.640024042129516,
      "epoch": 1.9062050184788952,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004633367630066545,
      "loss": 5.0815,
      "mean_token_accuracy": 0.19553401619195937,
      "num_tokens": 42504808.0,
      "step": 24500
    },
    {
      "entropy": 5.60738844871521,
      "epoch": 1.9065940478506127,
      "grad_norm": 1.25,
      "learning_rate": 0.00046332159774483116,
      "loss": 5.0024,
      "mean_token_accuracy": 0.20108923614025115,
      "num_tokens": 42512885.0,
      "step": 24505
    },
    {
      "entropy": 5.6044553279876705,
      "epoch": 1.9069830772223302,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004633064296255273,
      "loss": 4.9864,
      "mean_token_accuracy": 0.19766912013292312,
      "num_tokens": 42521481.0,
      "step": 24510
    },
    {
      "entropy": 5.687718868255615,
      "epoch": 1.9073721065940479,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046329125864897307,
      "loss": 5.0408,
      "mean_token_accuracy": 0.194117534160614,
      "num_tokens": 42529784.0,
      "step": 24515
    },
    {
      "entropy": 5.602256965637207,
      "epoch": 1.9077611359657656,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004632760848153987,
      "loss": 4.921,
      "mean_token_accuracy": 0.20549528747797013,
      "num_tokens": 42538802.0,
      "step": 24520
    },
    {
      "entropy": 5.586233615875244,
      "epoch": 1.9081501653374828,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004632609081250345,
      "loss": 4.9523,
      "mean_token_accuracy": 0.1994928613305092,
      "num_tokens": 42547699.0,
      "step": 24525
    },
    {
      "entropy": 5.611535835266113,
      "epoch": 1.9085391947092005,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046324572857811054,
      "loss": 5.0227,
      "mean_token_accuracy": 0.19892506301403046,
      "num_tokens": 42556583.0,
      "step": 24530
    },
    {
      "entropy": 5.556941223144531,
      "epoch": 1.9089282240809182,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004632305461748573,
      "loss": 4.9346,
      "mean_token_accuracy": 0.2049015700817108,
      "num_tokens": 42565025.0,
      "step": 24535
    },
    {
      "entropy": 5.607936239242553,
      "epoch": 1.9093172534526357,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046321536091550517,
      "loss": 5.0347,
      "mean_token_accuracy": 0.19777579605579376,
      "num_tokens": 42573798.0,
      "step": 24540
    },
    {
      "entropy": 5.660494136810303,
      "epoch": 1.9097062828243532,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004632001728002845,
      "loss": 4.9949,
      "mean_token_accuracy": 0.19820922315120698,
      "num_tokens": 42582002.0,
      "step": 24545
    },
    {
      "entropy": 5.725435066223144,
      "epoch": 1.9100953121960709,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004631849818294257,
      "loss": 5.011,
      "mean_token_accuracy": 0.19780345559120177,
      "num_tokens": 42591387.0,
      "step": 24550
    },
    {
      "entropy": 5.700288105010986,
      "epoch": 1.9104843415677883,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046316978800315934,
      "loss": 5.0861,
      "mean_token_accuracy": 0.1946798548102379,
      "num_tokens": 42599769.0,
      "step": 24555
    },
    {
      "entropy": 5.541125392913818,
      "epoch": 1.9108733709395058,
      "grad_norm": 1.125,
      "learning_rate": 0.000463154591321716,
      "loss": 4.9371,
      "mean_token_accuracy": 0.20612243860960006,
      "num_tokens": 42608279.0,
      "step": 24560
    },
    {
      "entropy": 5.592624616622925,
      "epoch": 1.9112624003112235,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046313939178532624,
      "loss": 5.0037,
      "mean_token_accuracy": 0.19894545674324035,
      "num_tokens": 42616652.0,
      "step": 24565
    },
    {
      "entropy": 5.581622076034546,
      "epoch": 1.9116514296829412,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004631241893942206,
      "loss": 4.9755,
      "mean_token_accuracy": 0.20120550692081451,
      "num_tokens": 42625629.0,
      "step": 24570
    },
    {
      "entropy": 5.649332475662232,
      "epoch": 1.9120404590546585,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046310898414862983,
      "loss": 5.0615,
      "mean_token_accuracy": 0.20079586356878282,
      "num_tokens": 42634159.0,
      "step": 24575
    },
    {
      "entropy": 5.720401382446289,
      "epoch": 1.9124294884263762,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004630937760487847,
      "loss": 5.1323,
      "mean_token_accuracy": 0.1872016668319702,
      "num_tokens": 42642880.0,
      "step": 24580
    },
    {
      "entropy": 5.664138412475586,
      "epoch": 1.9128185177980939,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004630785650949158,
      "loss": 5.0576,
      "mean_token_accuracy": 0.1971852406859398,
      "num_tokens": 42651481.0,
      "step": 24585
    },
    {
      "entropy": 5.661550283432007,
      "epoch": 1.9132075471698113,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004630633512872541,
      "loss": 5.0284,
      "mean_token_accuracy": 0.20767067819833757,
      "num_tokens": 42660131.0,
      "step": 24590
    },
    {
      "entropy": 5.595924997329712,
      "epoch": 1.9135965765415288,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046304813462603035,
      "loss": 5.0517,
      "mean_token_accuracy": 0.20472236275672911,
      "num_tokens": 42669094.0,
      "step": 24595
    },
    {
      "entropy": 5.633868217468262,
      "epoch": 1.9139856059132465,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004630329151114754,
      "loss": 4.9862,
      "mean_token_accuracy": 0.2096072256565094,
      "num_tokens": 42677792.0,
      "step": 24600
    },
    {
      "entropy": 5.660708332061768,
      "epoch": 1.914374635284964,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046301769274382034,
      "loss": 5.0741,
      "mean_token_accuracy": 0.1958266466856003,
      "num_tokens": 42686397.0,
      "step": 24605
    },
    {
      "entropy": 5.643627071380616,
      "epoch": 1.9147636646566815,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004630024675232961,
      "loss": 4.9897,
      "mean_token_accuracy": 0.2036696821451187,
      "num_tokens": 42695832.0,
      "step": 24610
    },
    {
      "entropy": 5.642992162704468,
      "epoch": 1.9151526940283992,
      "grad_norm": 1.125,
      "learning_rate": 0.00046298723945013354,
      "loss": 5.03,
      "mean_token_accuracy": 0.19840102791786193,
      "num_tokens": 42704378.0,
      "step": 24615
    },
    {
      "entropy": 5.589532995223999,
      "epoch": 1.9155417234001169,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004629720085245639,
      "loss": 4.976,
      "mean_token_accuracy": 0.2029490739107132,
      "num_tokens": 42713926.0,
      "step": 24620
    },
    {
      "entropy": 5.674699354171753,
      "epoch": 1.9159307527718341,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004629567747468182,
      "loss": 4.9622,
      "mean_token_accuracy": 0.20398446768522263,
      "num_tokens": 42722779.0,
      "step": 24625
    },
    {
      "entropy": 5.644221305847168,
      "epoch": 1.9163197821435518,
      "grad_norm": 1.125,
      "learning_rate": 0.0004629415381171276,
      "loss": 4.9435,
      "mean_token_accuracy": 0.20241658985614777,
      "num_tokens": 42731460.0,
      "step": 24630
    },
    {
      "entropy": 5.654909420013428,
      "epoch": 1.9167088115152695,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046292629863572325,
      "loss": 5.0229,
      "mean_token_accuracy": 0.20356272757053376,
      "num_tokens": 42739733.0,
      "step": 24635
    },
    {
      "entropy": 5.698343563079834,
      "epoch": 1.917097840886987,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004629110563028365,
      "loss": 5.044,
      "mean_token_accuracy": 0.1995990738272667,
      "num_tokens": 42748330.0,
      "step": 24640
    },
    {
      "entropy": 5.662912225723266,
      "epoch": 1.9174868702587045,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004628958111186985,
      "loss": 5.0506,
      "mean_token_accuracy": 0.19397861361503602,
      "num_tokens": 42757423.0,
      "step": 24645
    },
    {
      "entropy": 5.640391254425049,
      "epoch": 1.9178758996304222,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004628805630835407,
      "loss": 5.0529,
      "mean_token_accuracy": 0.19197241067886353,
      "num_tokens": 42766213.0,
      "step": 24650
    },
    {
      "entropy": 5.708678483963013,
      "epoch": 1.9182649290021396,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004628653121975944,
      "loss": 5.0656,
      "mean_token_accuracy": 0.2049526184797287,
      "num_tokens": 42774037.0,
      "step": 24655
    },
    {
      "entropy": 5.654391622543335,
      "epoch": 1.918653958373857,
      "grad_norm": 1.1875,
      "learning_rate": 0.000462850058461091,
      "loss": 5.1267,
      "mean_token_accuracy": 0.19167830646038056,
      "num_tokens": 42782806.0,
      "step": 24660
    },
    {
      "entropy": 5.586967277526855,
      "epoch": 1.9190429877455748,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004628348018742619,
      "loss": 4.9478,
      "mean_token_accuracy": 0.19925517737865447,
      "num_tokens": 42791790.0,
      "step": 24665
    },
    {
      "entropy": 5.6678112030029295,
      "epoch": 1.9194320171172925,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004628195424373387,
      "loss": 5.1206,
      "mean_token_accuracy": 0.19638921916484833,
      "num_tokens": 42799966.0,
      "step": 24670
    },
    {
      "entropy": 5.6609070777893065,
      "epoch": 1.91982104648901,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046280428015055286,
      "loss": 5.0112,
      "mean_token_accuracy": 0.20267692506313323,
      "num_tokens": 42808022.0,
      "step": 24675
    },
    {
      "entropy": 5.644557523727417,
      "epoch": 1.9202100758607274,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046278901501413606,
      "loss": 4.9364,
      "mean_token_accuracy": 0.21018209606409072,
      "num_tokens": 42816191.0,
      "step": 24680
    },
    {
      "entropy": 5.6555256843566895,
      "epoch": 1.9205991052324451,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046277374702831983,
      "loss": 5.0374,
      "mean_token_accuracy": 0.19775919169187545,
      "num_tokens": 42824840.0,
      "step": 24685
    },
    {
      "entropy": 5.632498550415039,
      "epoch": 1.9209881346041626,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046275847619333576,
      "loss": 5.0202,
      "mean_token_accuracy": 0.19889149367809295,
      "num_tokens": 42833788.0,
      "step": 24690
    },
    {
      "entropy": 5.688897132873535,
      "epoch": 1.92137716397588,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046274320250941576,
      "loss": 5.0525,
      "mean_token_accuracy": 0.2003769814968109,
      "num_tokens": 42842666.0,
      "step": 24695
    },
    {
      "entropy": 5.715423488616944,
      "epoch": 1.9217661933475978,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046272792597679146,
      "loss": 5.0961,
      "mean_token_accuracy": 0.19091131091117858,
      "num_tokens": 42850725.0,
      "step": 24700
    },
    {
      "entropy": 5.6071258068084715,
      "epoch": 1.9221552227193153,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046271264659569474,
      "loss": 5.0471,
      "mean_token_accuracy": 0.19348591715097427,
      "num_tokens": 42859732.0,
      "step": 24705
    },
    {
      "entropy": 5.645140075683594,
      "epoch": 1.9225442520910327,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004626973643663573,
      "loss": 5.0309,
      "mean_token_accuracy": 0.19824828952550888,
      "num_tokens": 42868858.0,
      "step": 24710
    },
    {
      "entropy": 5.66028847694397,
      "epoch": 1.9229332814627504,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004626820792890112,
      "loss": 5.0527,
      "mean_token_accuracy": 0.19698149263858794,
      "num_tokens": 42877483.0,
      "step": 24715
    },
    {
      "entropy": 5.6464990139007565,
      "epoch": 1.9233223108344681,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004626667913638882,
      "loss": 4.91,
      "mean_token_accuracy": 0.21170711219310762,
      "num_tokens": 42885156.0,
      "step": 24720
    },
    {
      "entropy": 5.674964094161988,
      "epoch": 1.9237113402061856,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004626515005912203,
      "loss": 5.0972,
      "mean_token_accuracy": 0.19178426414728164,
      "num_tokens": 42893599.0,
      "step": 24725
    },
    {
      "entropy": 5.625675678253174,
      "epoch": 1.924100369577903,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004626362069712397,
      "loss": 4.9755,
      "mean_token_accuracy": 0.20820729434490204,
      "num_tokens": 42902861.0,
      "step": 24730
    },
    {
      "entropy": 5.6438439846038815,
      "epoch": 1.9244893989496208,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004626209105041782,
      "loss": 5.0193,
      "mean_token_accuracy": 0.19692981243133545,
      "num_tokens": 42911246.0,
      "step": 24735
    },
    {
      "entropy": 5.638027000427246,
      "epoch": 1.9248784283213383,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000462605611190268,
      "loss": 5.1106,
      "mean_token_accuracy": 0.19920943230390548,
      "num_tokens": 42919526.0,
      "step": 24740
    },
    {
      "entropy": 5.654398345947266,
      "epoch": 1.9252674576930557,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046259030902974133,
      "loss": 4.9559,
      "mean_token_accuracy": 0.20410550385713577,
      "num_tokens": 42927473.0,
      "step": 24745
    },
    {
      "entropy": 5.576043605804443,
      "epoch": 1.9256564870647734,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004625750040228302,
      "loss": 4.9929,
      "mean_token_accuracy": 0.20215248465538024,
      "num_tokens": 42936108.0,
      "step": 24750
    },
    {
      "entropy": 5.556648206710816,
      "epoch": 1.926045516436491,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046255969616976704,
      "loss": 4.9161,
      "mean_token_accuracy": 0.20890791416168214,
      "num_tokens": 42944147.0,
      "step": 24755
    },
    {
      "entropy": 5.627501583099365,
      "epoch": 1.9264345458082084,
      "grad_norm": 1.046875,
      "learning_rate": 0.000462544385470784,
      "loss": 5.1133,
      "mean_token_accuracy": 0.19114366471767424,
      "num_tokens": 42952674.0,
      "step": 24760
    },
    {
      "entropy": 5.610938644409179,
      "epoch": 1.926823575179926,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004625290719261134,
      "loss": 4.9999,
      "mean_token_accuracy": 0.1999828040599823,
      "num_tokens": 42960907.0,
      "step": 24765
    },
    {
      "entropy": 5.616477346420288,
      "epoch": 1.9272126045516438,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004625137555359876,
      "loss": 4.975,
      "mean_token_accuracy": 0.2017730012536049,
      "num_tokens": 42968875.0,
      "step": 24770
    },
    {
      "entropy": 5.629356241226196,
      "epoch": 1.9276016339233613,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046249843630063905,
      "loss": 5.0092,
      "mean_token_accuracy": 0.20193208754062653,
      "num_tokens": 42977062.0,
      "step": 24775
    },
    {
      "entropy": 5.623213243484497,
      "epoch": 1.9279906632950787,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004624831142203001,
      "loss": 4.9735,
      "mean_token_accuracy": 0.2052836149930954,
      "num_tokens": 42985346.0,
      "step": 24780
    },
    {
      "entropy": 5.576091241836548,
      "epoch": 1.9283796926667964,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046246778929520346,
      "loss": 5.0485,
      "mean_token_accuracy": 0.18774661272764206,
      "num_tokens": 42993529.0,
      "step": 24785
    },
    {
      "entropy": 5.663331317901611,
      "epoch": 1.928768722038514,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004624524615255814,
      "loss": 5.0378,
      "mean_token_accuracy": 0.20091119408607483,
      "num_tokens": 43002254.0,
      "step": 24790
    },
    {
      "entropy": 5.6411339282989506,
      "epoch": 1.9291577514102314,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046243713091166655,
      "loss": 5.023,
      "mean_token_accuracy": 0.20111434459686278,
      "num_tokens": 43011130.0,
      "step": 24795
    },
    {
      "entropy": 5.613749027252197,
      "epoch": 1.929546780781949,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004624217974536916,
      "loss": 4.9748,
      "mean_token_accuracy": 0.21016593277454376,
      "num_tokens": 43019185.0,
      "step": 24800
    },
    {
      "entropy": 5.686710977554322,
      "epoch": 1.9299358101536666,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046240646115188925,
      "loss": 5.1847,
      "mean_token_accuracy": 0.1894476518034935,
      "num_tokens": 43028817.0,
      "step": 24805
    },
    {
      "entropy": 5.6527642726898195,
      "epoch": 1.930324839525384,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004623911220064921,
      "loss": 4.9975,
      "mean_token_accuracy": 0.20213748812675475,
      "num_tokens": 43038205.0,
      "step": 24810
    },
    {
      "entropy": 5.597604131698608,
      "epoch": 1.9307138688971017,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046237578001773297,
      "loss": 5.0189,
      "mean_token_accuracy": 0.20230717062950135,
      "num_tokens": 43046844.0,
      "step": 24815
    },
    {
      "entropy": 5.629145908355713,
      "epoch": 1.9311028982688194,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046236043518584454,
      "loss": 5.0959,
      "mean_token_accuracy": 0.19502410888671876,
      "num_tokens": 43055928.0,
      "step": 24820
    },
    {
      "entropy": 5.723018836975098,
      "epoch": 1.931491927640537,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004623450875110597,
      "loss": 5.0393,
      "mean_token_accuracy": 0.2012139841914177,
      "num_tokens": 43065371.0,
      "step": 24825
    },
    {
      "entropy": 5.6945921897888185,
      "epoch": 1.9318809570122544,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046232973699361147,
      "loss": 4.9818,
      "mean_token_accuracy": 0.19539664685726166,
      "num_tokens": 43073486.0,
      "step": 24830
    },
    {
      "entropy": 5.6217896938323975,
      "epoch": 1.932269986383972,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004623143836337326,
      "loss": 5.0389,
      "mean_token_accuracy": 0.19931581020355224,
      "num_tokens": 43083028.0,
      "step": 24835
    },
    {
      "entropy": 5.602948474884033,
      "epoch": 1.9326590157556895,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046229902743165607,
      "loss": 5.0082,
      "mean_token_accuracy": 0.20272698998451233,
      "num_tokens": 43091134.0,
      "step": 24840
    },
    {
      "entropy": 5.675520706176758,
      "epoch": 1.933048045127407,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004622836683876149,
      "loss": 5.076,
      "mean_token_accuracy": 0.19590371251106262,
      "num_tokens": 43099849.0,
      "step": 24845
    },
    {
      "entropy": 5.6430878162384035,
      "epoch": 1.9334370744991247,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004622683065018422,
      "loss": 5.0061,
      "mean_token_accuracy": 0.19762473553419113,
      "num_tokens": 43108311.0,
      "step": 24850
    },
    {
      "entropy": 5.588192129135132,
      "epoch": 1.9338261038708422,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046225294177457105,
      "loss": 4.9008,
      "mean_token_accuracy": 0.20076172947883605,
      "num_tokens": 43116179.0,
      "step": 24855
    },
    {
      "entropy": 5.56018476486206,
      "epoch": 1.9342151332425597,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046223757420603445,
      "loss": 4.9435,
      "mean_token_accuracy": 0.20267283022403718,
      "num_tokens": 43125728.0,
      "step": 24860
    },
    {
      "entropy": 5.556919860839844,
      "epoch": 1.9346041626142774,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004622222037964658,
      "loss": 4.9152,
      "mean_token_accuracy": 0.20083318650722504,
      "num_tokens": 43134512.0,
      "step": 24865
    },
    {
      "entropy": 5.610207748413086,
      "epoch": 1.934993191985995,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046220683054609815,
      "loss": 4.9617,
      "mean_token_accuracy": 0.20174091309309006,
      "num_tokens": 43144831.0,
      "step": 24870
    },
    {
      "entropy": 5.653454303741455,
      "epoch": 1.9353822213577125,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046219145445516483,
      "loss": 5.0518,
      "mean_token_accuracy": 0.20016671270132064,
      "num_tokens": 43153614.0,
      "step": 24875
    },
    {
      "entropy": 5.630557298660278,
      "epoch": 1.93577125072943,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046217607552389905,
      "loss": 5.0545,
      "mean_token_accuracy": 0.195446740090847,
      "num_tokens": 43162004.0,
      "step": 24880
    },
    {
      "entropy": 5.593694686889648,
      "epoch": 1.9361602801011477,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046216069375253435,
      "loss": 4.9488,
      "mean_token_accuracy": 0.2111530989408493,
      "num_tokens": 43170255.0,
      "step": 24885
    },
    {
      "entropy": 5.563721990585327,
      "epoch": 1.9365493094728652,
      "grad_norm": 1.125,
      "learning_rate": 0.000462145309141304,
      "loss": 5.0477,
      "mean_token_accuracy": 0.1995652809739113,
      "num_tokens": 43179036.0,
      "step": 24890
    },
    {
      "entropy": 5.644901752471924,
      "epoch": 1.9369383388445827,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004621299216904414,
      "loss": 5.0643,
      "mean_token_accuracy": 0.2006633073091507,
      "num_tokens": 43187207.0,
      "step": 24895
    },
    {
      "entropy": 5.617110967636108,
      "epoch": 1.9373273682163004,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046211453140018006,
      "loss": 5.0527,
      "mean_token_accuracy": 0.19078547805547713,
      "num_tokens": 43195600.0,
      "step": 24900
    },
    {
      "entropy": 5.645773696899414,
      "epoch": 1.937716397588018,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004620991382707537,
      "loss": 5.0572,
      "mean_token_accuracy": 0.20281705260276794,
      "num_tokens": 43204069.0,
      "step": 24905
    },
    {
      "entropy": 5.653849363327026,
      "epoch": 1.9381054269597353,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046208374230239556,
      "loss": 5.1164,
      "mean_token_accuracy": 0.18887075483798982,
      "num_tokens": 43212998.0,
      "step": 24910
    },
    {
      "entropy": 5.704898977279663,
      "epoch": 1.938494456331453,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004620683434953394,
      "loss": 5.0588,
      "mean_token_accuracy": 0.1996948465704918,
      "num_tokens": 43222411.0,
      "step": 24915
    },
    {
      "entropy": 5.636697912216187,
      "epoch": 1.9388834857031707,
      "grad_norm": 1.25,
      "learning_rate": 0.00046205294184981896,
      "loss": 5.0211,
      "mean_token_accuracy": 0.19601618349552155,
      "num_tokens": 43231265.0,
      "step": 24920
    },
    {
      "entropy": 5.616773843765259,
      "epoch": 1.9392725150748882,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046203753736606787,
      "loss": 5.0124,
      "mean_token_accuracy": 0.19903927892446518,
      "num_tokens": 43240609.0,
      "step": 24925
    },
    {
      "entropy": 5.674380683898926,
      "epoch": 1.9396615444466057,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004620221300443197,
      "loss": 5.0286,
      "mean_token_accuracy": 0.20214620381593704,
      "num_tokens": 43249381.0,
      "step": 24930
    },
    {
      "entropy": 5.6355180740356445,
      "epoch": 1.9400505738183234,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004620067198848086,
      "loss": 5.0037,
      "mean_token_accuracy": 0.20035636872053147,
      "num_tokens": 43257653.0,
      "step": 24935
    },
    {
      "entropy": 5.630123519897461,
      "epoch": 1.9404396031900408,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046199130688776805,
      "loss": 5.112,
      "mean_token_accuracy": 0.19244824200868607,
      "num_tokens": 43267111.0,
      "step": 24940
    },
    {
      "entropy": 5.6152183532714846,
      "epoch": 1.9408286325617583,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046197589105343204,
      "loss": 4.9678,
      "mean_token_accuracy": 0.20231392085552216,
      "num_tokens": 43276207.0,
      "step": 24945
    },
    {
      "entropy": 5.695230388641358,
      "epoch": 1.941217661933476,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004619604723820345,
      "loss": 5.0213,
      "mean_token_accuracy": 0.20445334315299987,
      "num_tokens": 43285031.0,
      "step": 24950
    },
    {
      "entropy": 5.696539258956909,
      "epoch": 1.9416066913051937,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004619450508738094,
      "loss": 5.0674,
      "mean_token_accuracy": 0.19647687673568726,
      "num_tokens": 43292802.0,
      "step": 24955
    },
    {
      "entropy": 5.666595602035523,
      "epoch": 1.941995720676911,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046192962652899056,
      "loss": 4.997,
      "mean_token_accuracy": 0.2038459822535515,
      "num_tokens": 43302349.0,
      "step": 24960
    },
    {
      "entropy": 5.58628396987915,
      "epoch": 1.9423847500486286,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046191419934781236,
      "loss": 4.9831,
      "mean_token_accuracy": 0.20070502460002898,
      "num_tokens": 43311455.0,
      "step": 24965
    },
    {
      "entropy": 5.625572776794433,
      "epoch": 1.9427737794203463,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004618987693305086,
      "loss": 5.0005,
      "mean_token_accuracy": 0.19921856969594956,
      "num_tokens": 43319924.0,
      "step": 24970
    },
    {
      "entropy": 5.667898178100586,
      "epoch": 1.9431628087920638,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004618833364773135,
      "loss": 5.1286,
      "mean_token_accuracy": 0.19586402475833892,
      "num_tokens": 43328772.0,
      "step": 24975
    },
    {
      "entropy": 5.590522766113281,
      "epoch": 1.9435518381637813,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046186790078846127,
      "loss": 5.0028,
      "mean_token_accuracy": 0.20137498676776885,
      "num_tokens": 43338862.0,
      "step": 24980
    },
    {
      "entropy": 5.677141857147217,
      "epoch": 1.943940867535499,
      "grad_norm": 1.125,
      "learning_rate": 0.00046185246226418603,
      "loss": 5.0365,
      "mean_token_accuracy": 0.1979084849357605,
      "num_tokens": 43347292.0,
      "step": 24985
    },
    {
      "entropy": 5.607594108581543,
      "epoch": 1.9443298969072165,
      "grad_norm": 1.125,
      "learning_rate": 0.00046183702090472206,
      "loss": 4.9678,
      "mean_token_accuracy": 0.20476548820734025,
      "num_tokens": 43356676.0,
      "step": 24990
    },
    {
      "entropy": 5.67404580116272,
      "epoch": 1.944718926278934,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004618215767103037,
      "loss": 5.1069,
      "mean_token_accuracy": 0.19462481141090393,
      "num_tokens": 43364746.0,
      "step": 24995
    },
    {
      "entropy": 5.6351189613342285,
      "epoch": 1.9451079556506516,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004618061296811653,
      "loss": 4.9551,
      "mean_token_accuracy": 0.19834737926721574,
      "num_tokens": 43373541.0,
      "step": 25000
    },
    {
      "entropy": 5.631911516189575,
      "epoch": 1.9454969850223693,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046179067981754124,
      "loss": 4.9595,
      "mean_token_accuracy": 0.2020665407180786,
      "num_tokens": 43382524.0,
      "step": 25005
    },
    {
      "entropy": 5.582877445220947,
      "epoch": 1.9458860143940866,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046177522711966584,
      "loss": 4.9916,
      "mean_token_accuracy": 0.20211981981992722,
      "num_tokens": 43391875.0,
      "step": 25010
    },
    {
      "entropy": 5.667497348785401,
      "epoch": 1.9462750437658043,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046175977158777377,
      "loss": 5.0826,
      "mean_token_accuracy": 0.19222914129495622,
      "num_tokens": 43399942.0,
      "step": 25015
    },
    {
      "entropy": 5.653568744659424,
      "epoch": 1.946664073137522,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004617443132220993,
      "loss": 5.0622,
      "mean_token_accuracy": 0.19539483338594438,
      "num_tokens": 43409371.0,
      "step": 25020
    },
    {
      "entropy": 5.631257486343384,
      "epoch": 1.9470531025092395,
      "grad_norm": 1.125,
      "learning_rate": 0.00046172885202287717,
      "loss": 5.0441,
      "mean_token_accuracy": 0.2009055495262146,
      "num_tokens": 43418181.0,
      "step": 25025
    },
    {
      "entropy": 5.590766143798828,
      "epoch": 1.947442131880957,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046171338799034194,
      "loss": 4.9935,
      "mean_token_accuracy": 0.20404358208179474,
      "num_tokens": 43426004.0,
      "step": 25030
    },
    {
      "entropy": 5.609193229675293,
      "epoch": 1.9478311612526746,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004616979211247282,
      "loss": 4.9212,
      "mean_token_accuracy": 0.20013620853424072,
      "num_tokens": 43433732.0,
      "step": 25035
    },
    {
      "entropy": 5.653647089004517,
      "epoch": 1.9482201906243921,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046168245142627065,
      "loss": 5.0545,
      "mean_token_accuracy": 0.2003475844860077,
      "num_tokens": 43442374.0,
      "step": 25040
    },
    {
      "entropy": 5.670967864990234,
      "epoch": 1.9486092199961096,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004616669788952041,
      "loss": 5.0783,
      "mean_token_accuracy": 0.19947095811367035,
      "num_tokens": 43451335.0,
      "step": 25045
    },
    {
      "entropy": 5.694469785690307,
      "epoch": 1.9489982493678273,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046165150353176315,
      "loss": 5.0963,
      "mean_token_accuracy": 0.19339118003845215,
      "num_tokens": 43459520.0,
      "step": 25050
    },
    {
      "entropy": 5.62667293548584,
      "epoch": 1.949387278739545,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004616360253361828,
      "loss": 4.9964,
      "mean_token_accuracy": 0.20136125832796098,
      "num_tokens": 43467695.0,
      "step": 25055
    },
    {
      "entropy": 5.59513840675354,
      "epoch": 1.9497763081112622,
      "grad_norm": 1.125,
      "learning_rate": 0.00046162054430869777,
      "loss": 5.0603,
      "mean_token_accuracy": 0.20271592438220978,
      "num_tokens": 43476161.0,
      "step": 25060
    },
    {
      "entropy": 5.606297874450684,
      "epoch": 1.95016533748298,
      "grad_norm": 1.109375,
      "learning_rate": 0.000461605060449543,
      "loss": 4.9809,
      "mean_token_accuracy": 0.2020869955420494,
      "num_tokens": 43485106.0,
      "step": 25065
    },
    {
      "entropy": 5.664118909835816,
      "epoch": 1.9505543668546976,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046158957375895353,
      "loss": 5.0795,
      "mean_token_accuracy": 0.1873122498393059,
      "num_tokens": 43494232.0,
      "step": 25070
    },
    {
      "entropy": 5.665037298202515,
      "epoch": 1.950943396226415,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004615740842371643,
      "loss": 4.9705,
      "mean_token_accuracy": 0.20532150864601134,
      "num_tokens": 43502556.0,
      "step": 25075
    },
    {
      "entropy": 5.587154531478882,
      "epoch": 1.9513324255981326,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046155859188441023,
      "loss": 4.928,
      "mean_token_accuracy": 0.20323332101106645,
      "num_tokens": 43511594.0,
      "step": 25080
    },
    {
      "entropy": 5.597598934173584,
      "epoch": 1.9517214549698503,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004615430967009265,
      "loss": 5.0659,
      "mean_token_accuracy": 0.1988156855106354,
      "num_tokens": 43520532.0,
      "step": 25085
    },
    {
      "entropy": 5.615779638290405,
      "epoch": 1.9521104843415678,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046152759868694815,
      "loss": 5.0931,
      "mean_token_accuracy": 0.19540077894926072,
      "num_tokens": 43529286.0,
      "step": 25090
    },
    {
      "entropy": 5.662170362472534,
      "epoch": 1.9524995137132852,
      "grad_norm": 1.125,
      "learning_rate": 0.0004615120978427104,
      "loss": 5.0243,
      "mean_token_accuracy": 0.19368369728326798,
      "num_tokens": 43537983.0,
      "step": 25095
    },
    {
      "entropy": 5.6424524784088135,
      "epoch": 1.952888543085003,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004614965941684485,
      "loss": 4.989,
      "mean_token_accuracy": 0.20542954206466674,
      "num_tokens": 43546550.0,
      "step": 25100
    },
    {
      "entropy": 5.645759630203247,
      "epoch": 1.9532775724567206,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004614810876643975,
      "loss": 5.062,
      "mean_token_accuracy": 0.20392390787601472,
      "num_tokens": 43555006.0,
      "step": 25105
    },
    {
      "entropy": 5.649363946914673,
      "epoch": 1.953666601828438,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046146557833079286,
      "loss": 5.1336,
      "mean_token_accuracy": 0.19307100474834443,
      "num_tokens": 43563925.0,
      "step": 25110
    },
    {
      "entropy": 5.67637619972229,
      "epoch": 1.9540556312001556,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004614500661678698,
      "loss": 5.0291,
      "mean_token_accuracy": 0.19548413902521133,
      "num_tokens": 43572269.0,
      "step": 25115
    },
    {
      "entropy": 5.6539754390716555,
      "epoch": 1.9544446605718733,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004614345511758639,
      "loss": 5.0566,
      "mean_token_accuracy": 0.19636407047510146,
      "num_tokens": 43579908.0,
      "step": 25120
    },
    {
      "entropy": 5.6206940650939945,
      "epoch": 1.9548336899435907,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046141903335501034,
      "loss": 4.9932,
      "mean_token_accuracy": 0.199165241420269,
      "num_tokens": 43589328.0,
      "step": 25125
    },
    {
      "entropy": 5.628673934936524,
      "epoch": 1.9552227193153082,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004614035127055447,
      "loss": 5.0475,
      "mean_token_accuracy": 0.20080305337905885,
      "num_tokens": 43597840.0,
      "step": 25130
    },
    {
      "entropy": 5.615436363220215,
      "epoch": 1.955611748687026,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004613879892277024,
      "loss": 4.9758,
      "mean_token_accuracy": 0.20515798181295394,
      "num_tokens": 43606496.0,
      "step": 25135
    },
    {
      "entropy": 5.670128250122071,
      "epoch": 1.9560007780587434,
      "grad_norm": 1.109375,
      "learning_rate": 0.000461372462921719,
      "loss": 5.0153,
      "mean_token_accuracy": 0.19746301025152208,
      "num_tokens": 43614411.0,
      "step": 25140
    },
    {
      "entropy": 5.555499696731568,
      "epoch": 1.9563898074304609,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004613569337878302,
      "loss": 4.936,
      "mean_token_accuracy": 0.20453592985868455,
      "num_tokens": 43622842.0,
      "step": 25145
    },
    {
      "entropy": 5.612040281295776,
      "epoch": 1.9567788368021786,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004613414018262715,
      "loss": 5.059,
      "mean_token_accuracy": 0.19443465322256087,
      "num_tokens": 43631535.0,
      "step": 25150
    },
    {
      "entropy": 5.666794013977051,
      "epoch": 1.9571678661738963,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004613258670372786,
      "loss": 5.048,
      "mean_token_accuracy": 0.20596069991588592,
      "num_tokens": 43640197.0,
      "step": 25155
    },
    {
      "entropy": 5.588799858093262,
      "epoch": 1.9575568955456137,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004613103294210873,
      "loss": 4.9815,
      "mean_token_accuracy": 0.2022147685289383,
      "num_tokens": 43648685.0,
      "step": 25160
    },
    {
      "entropy": 5.58948335647583,
      "epoch": 1.9579459249173312,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046129478897793323,
      "loss": 5.0117,
      "mean_token_accuracy": 0.19989572018384932,
      "num_tokens": 43657526.0,
      "step": 25165
    },
    {
      "entropy": 5.672930097579956,
      "epoch": 1.958334954289049,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046127924570805236,
      "loss": 5.0333,
      "mean_token_accuracy": 0.20916518419981003,
      "num_tokens": 43666119.0,
      "step": 25170
    },
    {
      "entropy": 5.699412822723389,
      "epoch": 1.9587239836607664,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004612636996116803,
      "loss": 5.0076,
      "mean_token_accuracy": 0.1953686699271202,
      "num_tokens": 43674634.0,
      "step": 25175
    },
    {
      "entropy": 5.532100629806519,
      "epoch": 1.9591130130324839,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004612481506890532,
      "loss": 4.9444,
      "mean_token_accuracy": 0.2038511425256729,
      "num_tokens": 43683329.0,
      "step": 25180
    },
    {
      "entropy": 5.615132427215576,
      "epoch": 1.9595020424042016,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046123259894040677,
      "loss": 4.9678,
      "mean_token_accuracy": 0.19947278946638108,
      "num_tokens": 43692598.0,
      "step": 25185
    },
    {
      "entropy": 5.5849957942962645,
      "epoch": 1.959891071775919,
      "grad_norm": 1.25,
      "learning_rate": 0.000461217044365977,
      "loss": 4.9473,
      "mean_token_accuracy": 0.20274491161108016,
      "num_tokens": 43702260.0,
      "step": 25190
    },
    {
      "entropy": 5.6672674179077145,
      "epoch": 1.9602801011476365,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004612014869660002,
      "loss": 5.0725,
      "mean_token_accuracy": 0.1998094543814659,
      "num_tokens": 43710003.0,
      "step": 25195
    },
    {
      "entropy": 5.644912958145142,
      "epoch": 1.9606691305193542,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046118592674071197,
      "loss": 4.9895,
      "mean_token_accuracy": 0.20102166682481765,
      "num_tokens": 43718687.0,
      "step": 25200
    },
    {
      "entropy": 5.583854913711548,
      "epoch": 1.961058159891072,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004611703636903488,
      "loss": 4.9141,
      "mean_token_accuracy": 0.21215902715921403,
      "num_tokens": 43727051.0,
      "step": 25205
    },
    {
      "entropy": 5.626040697097778,
      "epoch": 1.9614471892627894,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004611547978151466,
      "loss": 4.9618,
      "mean_token_accuracy": 0.2068934381008148,
      "num_tokens": 43735572.0,
      "step": 25210
    },
    {
      "entropy": 5.636909627914429,
      "epoch": 1.9618362186345069,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046113922911534167,
      "loss": 5.0112,
      "mean_token_accuracy": 0.2012847900390625,
      "num_tokens": 43744286.0,
      "step": 25215
    },
    {
      "entropy": 5.632687950134278,
      "epoch": 1.9622252480062246,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004611236575911702,
      "loss": 5.1011,
      "mean_token_accuracy": 0.20091122388839722,
      "num_tokens": 43754042.0,
      "step": 25220
    },
    {
      "entropy": 5.607291412353516,
      "epoch": 1.962614277377942,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046110808324286844,
      "loss": 4.899,
      "mean_token_accuracy": 0.20896745920181276,
      "num_tokens": 43762031.0,
      "step": 25225
    },
    {
      "entropy": 5.651480722427368,
      "epoch": 1.9630033067496595,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046109250607067275,
      "loss": 5.0402,
      "mean_token_accuracy": 0.20543597042560577,
      "num_tokens": 43770546.0,
      "step": 25230
    },
    {
      "entropy": 5.607420587539673,
      "epoch": 1.9633923361213772,
      "grad_norm": 1.125,
      "learning_rate": 0.0004610769260748196,
      "loss": 5.0299,
      "mean_token_accuracy": 0.19978648275136948,
      "num_tokens": 43779918.0,
      "step": 25235
    },
    {
      "entropy": 5.636054182052613,
      "epoch": 1.9637813654930947,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004610613432555451,
      "loss": 5.0201,
      "mean_token_accuracy": 0.19870051443576814,
      "num_tokens": 43788425.0,
      "step": 25240
    },
    {
      "entropy": 5.639534282684326,
      "epoch": 1.9641703948648122,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046104575761308597,
      "loss": 5.0668,
      "mean_token_accuracy": 0.1951361730694771,
      "num_tokens": 43797080.0,
      "step": 25245
    },
    {
      "entropy": 5.61166296005249,
      "epoch": 1.9645594242365298,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004610301691476786,
      "loss": 5.0138,
      "mean_token_accuracy": 0.2090237781405449,
      "num_tokens": 43806026.0,
      "step": 25250
    },
    {
      "entropy": 5.7103277206420895,
      "epoch": 1.9649484536082475,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004610145778595594,
      "loss": 5.0566,
      "mean_token_accuracy": 0.20122942477464675,
      "num_tokens": 43814333.0,
      "step": 25255
    },
    {
      "entropy": 5.611333274841309,
      "epoch": 1.965337482979965,
      "grad_norm": 1.125,
      "learning_rate": 0.0004609989837489651,
      "loss": 4.9736,
      "mean_token_accuracy": 0.2049986883997917,
      "num_tokens": 43823235.0,
      "step": 25260
    },
    {
      "entropy": 5.644299459457398,
      "epoch": 1.9657265123516825,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046098338681613234,
      "loss": 5.0287,
      "mean_token_accuracy": 0.20356353521347045,
      "num_tokens": 43832257.0,
      "step": 25265
    },
    {
      "entropy": 5.637997102737427,
      "epoch": 1.9661155417234002,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004609677870612976,
      "loss": 4.9621,
      "mean_token_accuracy": 0.21006757915019988,
      "num_tokens": 43840359.0,
      "step": 25270
    },
    {
      "entropy": 5.56435079574585,
      "epoch": 1.9665045710951177,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004609521844846978,
      "loss": 5.0825,
      "mean_token_accuracy": 0.19741379469633102,
      "num_tokens": 43849374.0,
      "step": 25275
    },
    {
      "entropy": 5.635733318328858,
      "epoch": 1.9668936004668351,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004609365790865695,
      "loss": 5.0626,
      "mean_token_accuracy": 0.19691859036684037,
      "num_tokens": 43857861.0,
      "step": 25280
    },
    {
      "entropy": 5.685753965377808,
      "epoch": 1.9672826298385528,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046092097086714956,
      "loss": 5.0541,
      "mean_token_accuracy": 0.20009537488222123,
      "num_tokens": 43865964.0,
      "step": 25285
    },
    {
      "entropy": 5.593512296676636,
      "epoch": 1.9676716592102705,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046090535982667486,
      "loss": 4.9678,
      "mean_token_accuracy": 0.20924752652645112,
      "num_tokens": 43874434.0,
      "step": 25290
    },
    {
      "entropy": 5.5867602825164795,
      "epoch": 1.9680606885819878,
      "grad_norm": 1.125,
      "learning_rate": 0.00046088974596538216,
      "loss": 5.0303,
      "mean_token_accuracy": 0.20201183259487152,
      "num_tokens": 43882691.0,
      "step": 25295
    },
    {
      "entropy": 5.685683488845825,
      "epoch": 1.9684497179537055,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004608741292835085,
      "loss": 5.0618,
      "mean_token_accuracy": 0.19685016125440596,
      "num_tokens": 43890589.0,
      "step": 25300
    },
    {
      "entropy": 5.685256385803223,
      "epoch": 1.9688387473254232,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004608585097812907,
      "loss": 5.0465,
      "mean_token_accuracy": 0.20567833930253981,
      "num_tokens": 43898908.0,
      "step": 25305
    },
    {
      "entropy": 5.673302268981933,
      "epoch": 1.9692277766971407,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004608428874589659,
      "loss": 5.0667,
      "mean_token_accuracy": 0.19665531069040298,
      "num_tokens": 43907640.0,
      "step": 25310
    },
    {
      "entropy": 5.623785877227784,
      "epoch": 1.9696168060688581,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004608272623167711,
      "loss": 4.9333,
      "mean_token_accuracy": 0.21087561696767806,
      "num_tokens": 43915757.0,
      "step": 25315
    },
    {
      "entropy": 5.691026067733764,
      "epoch": 1.9700058354405758,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046081163435494335,
      "loss": 5.0994,
      "mean_token_accuracy": 0.19721119999885559,
      "num_tokens": 43924910.0,
      "step": 25320
    },
    {
      "entropy": 5.658428192138672,
      "epoch": 1.9703948648122933,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004607960035737198,
      "loss": 5.0362,
      "mean_token_accuracy": 0.2079536259174347,
      "num_tokens": 43933122.0,
      "step": 25325
    },
    {
      "entropy": 5.564061784744263,
      "epoch": 1.9707838941840108,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004607803699733376,
      "loss": 5.007,
      "mean_token_accuracy": 0.20335709154605866,
      "num_tokens": 43941996.0,
      "step": 25330
    },
    {
      "entropy": 5.602984428405762,
      "epoch": 1.9711729235557285,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000460764733554034,
      "loss": 4.9436,
      "mean_token_accuracy": 0.2117355838418007,
      "num_tokens": 43950637.0,
      "step": 25335
    },
    {
      "entropy": 5.662865734100341,
      "epoch": 1.9715619529274462,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046074909431604623,
      "loss": 5.0291,
      "mean_token_accuracy": 0.19787351042032242,
      "num_tokens": 43959252.0,
      "step": 25340
    },
    {
      "entropy": 5.607434463500977,
      "epoch": 1.9719509822991634,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004607334522596116,
      "loss": 5.0106,
      "mean_token_accuracy": 0.1948935255408287,
      "num_tokens": 43968391.0,
      "step": 25345
    },
    {
      "entropy": 5.62926664352417,
      "epoch": 1.9723400116708811,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004607178073849675,
      "loss": 4.9446,
      "mean_token_accuracy": 0.20762541890144348,
      "num_tokens": 43977066.0,
      "step": 25350
    },
    {
      "entropy": 5.677835464477539,
      "epoch": 1.9727290410425988,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004607021596923512,
      "loss": 5.0083,
      "mean_token_accuracy": 0.209498630464077,
      "num_tokens": 43985072.0,
      "step": 25355
    },
    {
      "entropy": 5.75250391960144,
      "epoch": 1.9731180704143163,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004606865091820003,
      "loss": 5.2008,
      "mean_token_accuracy": 0.19260186851024627,
      "num_tokens": 43993321.0,
      "step": 25360
    },
    {
      "entropy": 5.567666101455688,
      "epoch": 1.9735070997860338,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004606708558541521,
      "loss": 4.8846,
      "mean_token_accuracy": 0.2144096687436104,
      "num_tokens": 44002476.0,
      "step": 25365
    },
    {
      "entropy": 5.661514377593994,
      "epoch": 1.9738961291577515,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004606551997090442,
      "loss": 5.0892,
      "mean_token_accuracy": 0.19390449672937393,
      "num_tokens": 44011790.0,
      "step": 25370
    },
    {
      "entropy": 5.611005783081055,
      "epoch": 1.974285158529469,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004606395407469141,
      "loss": 4.9872,
      "mean_token_accuracy": 0.2004820927977562,
      "num_tokens": 44021086.0,
      "step": 25375
    },
    {
      "entropy": 5.599969863891602,
      "epoch": 1.9746741879011864,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004606238789679995,
      "loss": 4.9547,
      "mean_token_accuracy": 0.20880282819271087,
      "num_tokens": 44029789.0,
      "step": 25380
    },
    {
      "entropy": 5.5799132823944095,
      "epoch": 1.9750632172729041,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004606082143725381,
      "loss": 4.994,
      "mean_token_accuracy": 0.19691853374242782,
      "num_tokens": 44039508.0,
      "step": 25385
    },
    {
      "entropy": 5.603135490417481,
      "epoch": 1.9754522466446218,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004605925469607673,
      "loss": 4.98,
      "mean_token_accuracy": 0.21062097549438477,
      "num_tokens": 44048464.0,
      "step": 25390
    },
    {
      "entropy": 5.75105390548706,
      "epoch": 1.975841276016339,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046057687673292506,
      "loss": 5.1033,
      "mean_token_accuracy": 0.19382342845201492,
      "num_tokens": 44057613.0,
      "step": 25395
    },
    {
      "entropy": 5.609554624557495,
      "epoch": 1.9762303053880568,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046056120368924907,
      "loss": 4.9005,
      "mean_token_accuracy": 0.21266730278730392,
      "num_tokens": 44066334.0,
      "step": 25400
    },
    {
      "entropy": 5.625293111801147,
      "epoch": 1.9766193347597745,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004605455278299772,
      "loss": 5.1422,
      "mean_token_accuracy": 0.19469816982746124,
      "num_tokens": 44076355.0,
      "step": 25405
    },
    {
      "entropy": 5.668260478973389,
      "epoch": 1.977008364131492,
      "grad_norm": 1.125,
      "learning_rate": 0.0004605298491553472,
      "loss": 5.0728,
      "mean_token_accuracy": 0.19277391880750655,
      "num_tokens": 44084445.0,
      "step": 25410
    },
    {
      "entropy": 5.654285526275634,
      "epoch": 1.9773973935032094,
      "grad_norm": 1.125,
      "learning_rate": 0.0004605141676655971,
      "loss": 5.0071,
      "mean_token_accuracy": 0.2106776475906372,
      "num_tokens": 44093711.0,
      "step": 25415
    },
    {
      "entropy": 5.628231477737427,
      "epoch": 1.9777864228749271,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046049848336096485,
      "loss": 5.0084,
      "mean_token_accuracy": 0.19742907881736754,
      "num_tokens": 44102647.0,
      "step": 25420
    },
    {
      "entropy": 5.629535055160522,
      "epoch": 1.9781754522466446,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004604827962416883,
      "loss": 4.9206,
      "mean_token_accuracy": 0.2070244938135147,
      "num_tokens": 44110661.0,
      "step": 25425
    },
    {
      "entropy": 5.696330404281616,
      "epoch": 1.978564481618362,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004604671063080055,
      "loss": 5.1623,
      "mean_token_accuracy": 0.18638489097356797,
      "num_tokens": 44119441.0,
      "step": 25430
    },
    {
      "entropy": 5.628379201889038,
      "epoch": 1.9789535109900798,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004604514135601546,
      "loss": 4.9654,
      "mean_token_accuracy": 0.20146813839673997,
      "num_tokens": 44127774.0,
      "step": 25435
    },
    {
      "entropy": 5.695941209793091,
      "epoch": 1.9793425403617975,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004604357179983736,
      "loss": 5.1919,
      "mean_token_accuracy": 0.1838580921292305,
      "num_tokens": 44137075.0,
      "step": 25440
    },
    {
      "entropy": 5.57067904472351,
      "epoch": 1.9797315697335147,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004604200196229009,
      "loss": 4.9268,
      "mean_token_accuracy": 0.20352620780467987,
      "num_tokens": 44145647.0,
      "step": 25445
    },
    {
      "entropy": 5.586453247070312,
      "epoch": 1.9801205991052324,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004604043184339744,
      "loss": 4.9102,
      "mean_token_accuracy": 0.20952169150114058,
      "num_tokens": 44154267.0,
      "step": 25450
    },
    {
      "entropy": 5.630558824539184,
      "epoch": 1.9805096284769501,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004603886144318325,
      "loss": 5.0485,
      "mean_token_accuracy": 0.2020800605416298,
      "num_tokens": 44162532.0,
      "step": 25455
    },
    {
      "entropy": 5.66807451248169,
      "epoch": 1.9808986578486676,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046037290761671346,
      "loss": 5.0225,
      "mean_token_accuracy": 0.20350697189569472,
      "num_tokens": 44171218.0,
      "step": 25460
    },
    {
      "entropy": 5.603969717025757,
      "epoch": 1.981287687220385,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004603571979888556,
      "loss": 5.0961,
      "mean_token_accuracy": 0.1953939124941826,
      "num_tokens": 44180189.0,
      "step": 25465
    },
    {
      "entropy": 5.625427484512329,
      "epoch": 1.9816767165921028,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004603414855484973,
      "loss": 4.9509,
      "mean_token_accuracy": 0.20667276978492738,
      "num_tokens": 44189048.0,
      "step": 25470
    },
    {
      "entropy": 5.658818864822388,
      "epoch": 1.9820657459638202,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000460325770295877,
      "loss": 5.0289,
      "mean_token_accuracy": 0.2022498816251755,
      "num_tokens": 44197995.0,
      "step": 25475
    },
    {
      "entropy": 5.685598373413086,
      "epoch": 1.9824547753355377,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046031005223123297,
      "loss": 5.1315,
      "mean_token_accuracy": 0.18635321259498597,
      "num_tokens": 44207042.0,
      "step": 25480
    },
    {
      "entropy": 5.639273452758789,
      "epoch": 1.9828438047072554,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000460294331354804,
      "loss": 4.9394,
      "mean_token_accuracy": 0.21081544011831282,
      "num_tokens": 44214678.0,
      "step": 25485
    },
    {
      "entropy": 5.61578779220581,
      "epoch": 1.983232834078973,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004602786076668283,
      "loss": 4.9711,
      "mean_token_accuracy": 0.20548944622278215,
      "num_tokens": 44223752.0,
      "step": 25490
    },
    {
      "entropy": 5.667809772491455,
      "epoch": 1.9836218634506906,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046026288116754477,
      "loss": 5.042,
      "mean_token_accuracy": 0.19623966217041017,
      "num_tokens": 44232109.0,
      "step": 25495
    },
    {
      "entropy": 5.730648756027222,
      "epoch": 1.984010892822408,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004602471518571919,
      "loss": 5.1612,
      "mean_token_accuracy": 0.1914820745587349,
      "num_tokens": 44239957.0,
      "step": 25500
    },
    {
      "entropy": 5.703612613677978,
      "epoch": 1.9843999221941258,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004602314197360083,
      "loss": 5.0146,
      "mean_token_accuracy": 0.20351981818675996,
      "num_tokens": 44248582.0,
      "step": 25505
    },
    {
      "entropy": 5.694765520095825,
      "epoch": 1.9847889515658432,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004602156848042328,
      "loss": 5.0218,
      "mean_token_accuracy": 0.2002855896949768,
      "num_tokens": 44256711.0,
      "step": 25510
    },
    {
      "entropy": 5.626265573501587,
      "epoch": 1.9851779809375607,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000460199947062104,
      "loss": 4.9955,
      "mean_token_accuracy": 0.20539263933897017,
      "num_tokens": 44265140.0,
      "step": 25515
    },
    {
      "entropy": 5.701619052886963,
      "epoch": 1.9855670103092784,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046018420650986074,
      "loss": 5.002,
      "mean_token_accuracy": 0.1958703950047493,
      "num_tokens": 44273560.0,
      "step": 25520
    },
    {
      "entropy": 5.579987335205078,
      "epoch": 1.9859560396809959,
      "grad_norm": 1.125,
      "learning_rate": 0.000460168463147742,
      "loss": 4.906,
      "mean_token_accuracy": 0.2073463648557663,
      "num_tokens": 44282155.0,
      "step": 25525
    },
    {
      "entropy": 5.604988765716553,
      "epoch": 1.9863450690527134,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004601527169759865,
      "loss": 4.9545,
      "mean_token_accuracy": 0.2013138011097908,
      "num_tokens": 44291237.0,
      "step": 25530
    },
    {
      "entropy": 5.606105709075928,
      "epoch": 1.986734098424431,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004601369679948333,
      "loss": 4.9753,
      "mean_token_accuracy": 0.20221568644046783,
      "num_tokens": 44299886.0,
      "step": 25535
    },
    {
      "entropy": 5.665149831771851,
      "epoch": 1.9871231277961487,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046012121620452127,
      "loss": 5.0879,
      "mean_token_accuracy": 0.20545906573534012,
      "num_tokens": 44308247.0,
      "step": 25540
    },
    {
      "entropy": 5.636614370346069,
      "epoch": 1.9875121571678662,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004601054616052893,
      "loss": 4.9827,
      "mean_token_accuracy": 0.1998549148440361,
      "num_tokens": 44316301.0,
      "step": 25545
    },
    {
      "entropy": 5.740137767791748,
      "epoch": 1.9879011865395837,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046008970419737674,
      "loss": 5.1094,
      "mean_token_accuracy": 0.18824613839387894,
      "num_tokens": 44324758.0,
      "step": 25550
    },
    {
      "entropy": 5.7089118480682375,
      "epoch": 1.9882902159113014,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004600739439810225,
      "loss": 5.1049,
      "mean_token_accuracy": 0.1910092368721962,
      "num_tokens": 44333139.0,
      "step": 25555
    },
    {
      "entropy": 5.694560098648071,
      "epoch": 1.9886792452830189,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046005818095646564,
      "loss": 5.0707,
      "mean_token_accuracy": 0.20021645873785018,
      "num_tokens": 44341992.0,
      "step": 25560
    },
    {
      "entropy": 5.6200854778289795,
      "epoch": 1.9890682746547363,
      "grad_norm": 1.375,
      "learning_rate": 0.00046004241512394544,
      "loss": 5.0335,
      "mean_token_accuracy": 0.19744022488594054,
      "num_tokens": 44350947.0,
      "step": 25565
    },
    {
      "entropy": 5.627526712417603,
      "epoch": 1.989457304026454,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004600266464837012,
      "loss": 4.9689,
      "mean_token_accuracy": 0.20670282691717148,
      "num_tokens": 44358934.0,
      "step": 25570
    },
    {
      "entropy": 5.615651178359985,
      "epoch": 1.9898463333981715,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004600108750359721,
      "loss": 4.9557,
      "mean_token_accuracy": 0.20935382097959518,
      "num_tokens": 44367117.0,
      "step": 25575
    },
    {
      "entropy": 5.644807958602906,
      "epoch": 1.990235362769889,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045999510078099736,
      "loss": 5.02,
      "mean_token_accuracy": 0.19504447877407075,
      "num_tokens": 44375591.0,
      "step": 25580
    },
    {
      "entropy": 5.584261035919189,
      "epoch": 1.9906243921416067,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045997932371901645,
      "loss": 4.9134,
      "mean_token_accuracy": 0.20993310660123826,
      "num_tokens": 44384602.0,
      "step": 25585
    },
    {
      "entropy": 5.550891733169555,
      "epoch": 1.9910134215133244,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004599635438502687,
      "loss": 4.8969,
      "mean_token_accuracy": 0.2152763843536377,
      "num_tokens": 44393451.0,
      "step": 25590
    },
    {
      "entropy": 5.587131404876709,
      "epoch": 1.9914024508850419,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045994776117499363,
      "loss": 4.9715,
      "mean_token_accuracy": 0.2011899918317795,
      "num_tokens": 44401638.0,
      "step": 25595
    },
    {
      "entropy": 5.638331413269043,
      "epoch": 1.9917914802567593,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045993197569343063,
      "loss": 4.9883,
      "mean_token_accuracy": 0.20341338217258453,
      "num_tokens": 44410337.0,
      "step": 25600
    },
    {
      "entropy": 5.671945428848266,
      "epoch": 1.992180509628477,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045991618740581926,
      "loss": 5.0437,
      "mean_token_accuracy": 0.19910795688629152,
      "num_tokens": 44418178.0,
      "step": 25605
    },
    {
      "entropy": 5.652189636230469,
      "epoch": 1.9925695390001945,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004599003963123991,
      "loss": 5.0637,
      "mean_token_accuracy": 0.1915079414844513,
      "num_tokens": 44427056.0,
      "step": 25610
    },
    {
      "entropy": 5.71173243522644,
      "epoch": 1.992958568371912,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045988460241340966,
      "loss": 5.0081,
      "mean_token_accuracy": 0.20151230543851853,
      "num_tokens": 44434687.0,
      "step": 25615
    },
    {
      "entropy": 5.665513753890991,
      "epoch": 1.9933475977436297,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045986880570909075,
      "loss": 5.0089,
      "mean_token_accuracy": 0.20644039064645767,
      "num_tokens": 44442763.0,
      "step": 25620
    },
    {
      "entropy": 5.517037439346313,
      "epoch": 1.9937366271153472,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045985300619968186,
      "loss": 4.9006,
      "mean_token_accuracy": 0.20334762334823608,
      "num_tokens": 44451850.0,
      "step": 25625
    },
    {
      "entropy": 5.558308410644531,
      "epoch": 1.9941256564870646,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045983720388542286,
      "loss": 4.953,
      "mean_token_accuracy": 0.21013179272413254,
      "num_tokens": 44459915.0,
      "step": 25630
    },
    {
      "entropy": 5.576267433166504,
      "epoch": 1.9945146858587823,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004598213987665535,
      "loss": 5.012,
      "mean_token_accuracy": 0.20058779120445253,
      "num_tokens": 44469252.0,
      "step": 25635
    },
    {
      "entropy": 5.737783432006836,
      "epoch": 1.9949037152305,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045980559084331354,
      "loss": 5.0858,
      "mean_token_accuracy": 0.19604782462120057,
      "num_tokens": 44477748.0,
      "step": 25640
    },
    {
      "entropy": 5.636568450927735,
      "epoch": 1.9952927446022175,
      "grad_norm": 1.046875,
      "learning_rate": 0.000459789780115943,
      "loss": 5.0191,
      "mean_token_accuracy": 0.19964341968297958,
      "num_tokens": 44485796.0,
      "step": 25645
    },
    {
      "entropy": 5.557905387878418,
      "epoch": 1.995681773973935,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045977396658468156,
      "loss": 5.0209,
      "mean_token_accuracy": 0.1956699460744858,
      "num_tokens": 44494604.0,
      "step": 25650
    },
    {
      "entropy": 5.573694944381714,
      "epoch": 1.9960708033456527,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004597581502497693,
      "loss": 4.9982,
      "mean_token_accuracy": 0.1985325485467911,
      "num_tokens": 44503728.0,
      "step": 25655
    },
    {
      "entropy": 5.661129331588745,
      "epoch": 1.9964598327173702,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004597423311114462,
      "loss": 4.9865,
      "mean_token_accuracy": 0.2076873004436493,
      "num_tokens": 44512201.0,
      "step": 25660
    },
    {
      "entropy": 5.703567981719971,
      "epoch": 1.9968488620890876,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004597265091699522,
      "loss": 5.0161,
      "mean_token_accuracy": 0.1997128590941429,
      "num_tokens": 44520122.0,
      "step": 25665
    },
    {
      "entropy": 5.562481498718261,
      "epoch": 1.9972378914608053,
      "grad_norm": 1.125,
      "learning_rate": 0.0004597106844255276,
      "loss": 4.9984,
      "mean_token_accuracy": 0.20076329708099366,
      "num_tokens": 44528928.0,
      "step": 25670
    },
    {
      "entropy": 5.652709913253784,
      "epoch": 1.9976269208325228,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045969485687841227,
      "loss": 5.1262,
      "mean_token_accuracy": 0.19472890347242355,
      "num_tokens": 44537330.0,
      "step": 25675
    },
    {
      "entropy": 5.697073698043823,
      "epoch": 1.9980159502042403,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045967902652884645,
      "loss": 5.0037,
      "mean_token_accuracy": 0.20233794897794724,
      "num_tokens": 44545748.0,
      "step": 25680
    },
    {
      "entropy": 5.6518495082855225,
      "epoch": 1.998404979575958,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004596631933770704,
      "loss": 5.1157,
      "mean_token_accuracy": 0.19582041800022126,
      "num_tokens": 44554562.0,
      "step": 25685
    },
    {
      "entropy": 5.616053247451783,
      "epoch": 1.9987940089476757,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045964735742332427,
      "loss": 4.9583,
      "mean_token_accuracy": 0.19985282570123672,
      "num_tokens": 44563544.0,
      "step": 25690
    },
    {
      "entropy": 5.676188564300537,
      "epoch": 1.9991830383193931,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004596315186678484,
      "loss": 5.026,
      "mean_token_accuracy": 0.1987316220998764,
      "num_tokens": 44571179.0,
      "step": 25695
    },
    {
      "entropy": 5.686815309524536,
      "epoch": 1.9995720676911106,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045961567711088307,
      "loss": 5.0591,
      "mean_token_accuracy": 0.19998359382152558,
      "num_tokens": 44580237.0,
      "step": 25700
    },
    {
      "entropy": 5.6182520389556885,
      "epoch": 1.9999610970628283,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045959983275266873,
      "loss": 5.008,
      "mean_token_accuracy": 0.20599913001060485,
      "num_tokens": 44588362.0,
      "step": 25705
    },
    {
      "entropy": 5.602212693956163,
      "epoch": 2.000311223497374,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045958398559344573,
      "loss": 4.9271,
      "mean_token_accuracy": 0.20771176285213894,
      "num_tokens": 44596037.0,
      "step": 25710
    },
    {
      "entropy": 5.6889026165008545,
      "epoch": 2.0007002528690916,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045956813563345454,
      "loss": 4.9812,
      "mean_token_accuracy": 0.21024697571992873,
      "num_tokens": 44604913.0,
      "step": 25715
    },
    {
      "entropy": 5.685275220870972,
      "epoch": 2.0010892822408093,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004595522828729357,
      "loss": 5.0306,
      "mean_token_accuracy": 0.20595474243164064,
      "num_tokens": 44613523.0,
      "step": 25720
    },
    {
      "entropy": 5.6259174823760985,
      "epoch": 2.0014783116125265,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004595364273121296,
      "loss": 4.8335,
      "mean_token_accuracy": 0.21599423438310622,
      "num_tokens": 44622259.0,
      "step": 25725
    },
    {
      "entropy": 5.613101434707642,
      "epoch": 2.0018673409842442,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004595205689512771,
      "loss": 4.9081,
      "mean_token_accuracy": 0.21179116070270537,
      "num_tokens": 44630793.0,
      "step": 25730
    },
    {
      "entropy": 5.638196229934692,
      "epoch": 2.002256370355962,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045950470779061855,
      "loss": 4.9421,
      "mean_token_accuracy": 0.2082168996334076,
      "num_tokens": 44638986.0,
      "step": 25735
    },
    {
      "entropy": 5.66730375289917,
      "epoch": 2.0026453997276796,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045948884383039475,
      "loss": 5.0017,
      "mean_token_accuracy": 0.19579940438270568,
      "num_tokens": 44647491.0,
      "step": 25740
    },
    {
      "entropy": 5.578490543365478,
      "epoch": 2.003034429099397,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045947297707084637,
      "loss": 4.9234,
      "mean_token_accuracy": 0.2060213416814804,
      "num_tokens": 44655431.0,
      "step": 25745
    },
    {
      "entropy": 5.593864631652832,
      "epoch": 2.0034234584711146,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004594571075122142,
      "loss": 4.9348,
      "mean_token_accuracy": 0.2071908235549927,
      "num_tokens": 44664398.0,
      "step": 25750
    },
    {
      "entropy": 5.647603988647461,
      "epoch": 2.0038124878428323,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045944123515473905,
      "loss": 4.9149,
      "mean_token_accuracy": 0.2097979575395584,
      "num_tokens": 44673372.0,
      "step": 25755
    },
    {
      "entropy": 5.5615349292755125,
      "epoch": 2.0042015172145495,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045942535999866175,
      "loss": 4.8497,
      "mean_token_accuracy": 0.2112073630094528,
      "num_tokens": 44681736.0,
      "step": 25760
    },
    {
      "entropy": 5.6290534973144535,
      "epoch": 2.0045905465862672,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004594094820442231,
      "loss": 4.9962,
      "mean_token_accuracy": 0.20321001559495927,
      "num_tokens": 44691858.0,
      "step": 25765
    },
    {
      "entropy": 5.709609317779541,
      "epoch": 2.004979575957985,
      "grad_norm": 1.40625,
      "learning_rate": 0.000459393601291664,
      "loss": 5.0094,
      "mean_token_accuracy": 0.201983280479908,
      "num_tokens": 44701002.0,
      "step": 25770
    },
    {
      "entropy": 5.63389778137207,
      "epoch": 2.005368605329702,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045937771774122563,
      "loss": 4.9084,
      "mean_token_accuracy": 0.21504546850919723,
      "num_tokens": 44709158.0,
      "step": 25775
    },
    {
      "entropy": 5.592244958877563,
      "epoch": 2.00575763470142,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004593618313931488,
      "loss": 4.9441,
      "mean_token_accuracy": 0.2064715340733528,
      "num_tokens": 44717495.0,
      "step": 25780
    },
    {
      "entropy": 5.632651853561401,
      "epoch": 2.0061466640731376,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045934594224767463,
      "loss": 4.9254,
      "mean_token_accuracy": 0.20571477860212325,
      "num_tokens": 44726040.0,
      "step": 25785
    },
    {
      "entropy": 5.63297028541565,
      "epoch": 2.0065356934448553,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045933005030504424,
      "loss": 4.9612,
      "mean_token_accuracy": 0.201497782766819,
      "num_tokens": 44735098.0,
      "step": 25790
    },
    {
      "entropy": 5.6099405765533445,
      "epoch": 2.0069247228165725,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045931415556549863,
      "loss": 4.9361,
      "mean_token_accuracy": 0.20973822772502898,
      "num_tokens": 44743764.0,
      "step": 25795
    },
    {
      "entropy": 5.610128164291382,
      "epoch": 2.0073137521882902,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004592982580292792,
      "loss": 4.8523,
      "mean_token_accuracy": 0.20498300343751907,
      "num_tokens": 44751554.0,
      "step": 25800
    },
    {
      "entropy": 5.708305978775025,
      "epoch": 2.007702781560008,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045928235769662697,
      "loss": 5.0334,
      "mean_token_accuracy": 0.1978297308087349,
      "num_tokens": 44760158.0,
      "step": 25805
    },
    {
      "entropy": 5.633608007431031,
      "epoch": 2.008091810931725,
      "grad_norm": 1.125,
      "learning_rate": 0.00045926645456778327,
      "loss": 4.9248,
      "mean_token_accuracy": 0.21309804022312165,
      "num_tokens": 44769209.0,
      "step": 25810
    },
    {
      "entropy": 5.621290588378907,
      "epoch": 2.008480840303443,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045925054864298946,
      "loss": 4.9751,
      "mean_token_accuracy": 0.20010165125131607,
      "num_tokens": 44777991.0,
      "step": 25815
    },
    {
      "entropy": 5.670830917358399,
      "epoch": 2.0088698696751606,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045923463992248684,
      "loss": 4.9933,
      "mean_token_accuracy": 0.20064879357814788,
      "num_tokens": 44786492.0,
      "step": 25820
    },
    {
      "entropy": 5.701191043853759,
      "epoch": 2.009258899046878,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045921872840651675,
      "loss": 5.0398,
      "mean_token_accuracy": 0.20214534401893616,
      "num_tokens": 44795021.0,
      "step": 25825
    },
    {
      "entropy": 5.621680021286011,
      "epoch": 2.0096479284185955,
      "grad_norm": 1.25,
      "learning_rate": 0.00045920281409532064,
      "loss": 4.9329,
      "mean_token_accuracy": 0.20348845273256302,
      "num_tokens": 44803267.0,
      "step": 25830
    },
    {
      "entropy": 5.659497404098511,
      "epoch": 2.010036957790313,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004591868969891401,
      "loss": 5.0038,
      "mean_token_accuracy": 0.20323139876127244,
      "num_tokens": 44811954.0,
      "step": 25835
    },
    {
      "entropy": 5.6048705101013185,
      "epoch": 2.010425987162031,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004591709770882165,
      "loss": 4.8583,
      "mean_token_accuracy": 0.20969469100236893,
      "num_tokens": 44820748.0,
      "step": 25840
    },
    {
      "entropy": 5.67017970085144,
      "epoch": 2.010815016533748,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004591550543927915,
      "loss": 5.045,
      "mean_token_accuracy": 0.20047615319490433,
      "num_tokens": 44829400.0,
      "step": 25845
    },
    {
      "entropy": 5.683247137069702,
      "epoch": 2.011204045905466,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004591391289031067,
      "loss": 4.978,
      "mean_token_accuracy": 0.20444614738225936,
      "num_tokens": 44838663.0,
      "step": 25850
    },
    {
      "entropy": 5.639876747131348,
      "epoch": 2.0115930752771836,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004591232006194036,
      "loss": 4.8802,
      "mean_token_accuracy": 0.21119816303253175,
      "num_tokens": 44847365.0,
      "step": 25855
    },
    {
      "entropy": 5.646116447448731,
      "epoch": 2.011982104648901,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045910726954192404,
      "loss": 4.9486,
      "mean_token_accuracy": 0.20773670822381973,
      "num_tokens": 44856093.0,
      "step": 25860
    },
    {
      "entropy": 5.6047163009643555,
      "epoch": 2.0123711340206185,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004590913356709097,
      "loss": 4.9379,
      "mean_token_accuracy": 0.2084503710269928,
      "num_tokens": 44865116.0,
      "step": 25865
    },
    {
      "entropy": 5.624687719345093,
      "epoch": 2.012760163392336,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045907539900660237,
      "loss": 4.995,
      "mean_token_accuracy": 0.1994251176714897,
      "num_tokens": 44874037.0,
      "step": 25870
    },
    {
      "entropy": 5.608638143539428,
      "epoch": 2.013149192764054,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004590594595492438,
      "loss": 4.8343,
      "mean_token_accuracy": 0.212248994410038,
      "num_tokens": 44882603.0,
      "step": 25875
    },
    {
      "entropy": 5.64333701133728,
      "epoch": 2.013538222135771,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045904351729907593,
      "loss": 4.8897,
      "mean_token_accuracy": 0.2131791591644287,
      "num_tokens": 44890645.0,
      "step": 25880
    },
    {
      "entropy": 5.697340202331543,
      "epoch": 2.013927251507489,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045902757225634066,
      "loss": 5.0066,
      "mean_token_accuracy": 0.2054777041077614,
      "num_tokens": 44898972.0,
      "step": 25885
    },
    {
      "entropy": 5.590086650848389,
      "epoch": 2.0143162808792066,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004590116244212799,
      "loss": 4.979,
      "mean_token_accuracy": 0.20843791961669922,
      "num_tokens": 44907427.0,
      "step": 25890
    },
    {
      "entropy": 5.634195280075073,
      "epoch": 2.014705310250924,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004589956737941355,
      "loss": 5.0134,
      "mean_token_accuracy": 0.2046400472521782,
      "num_tokens": 44916465.0,
      "step": 25895
    },
    {
      "entropy": 5.632593774795533,
      "epoch": 2.0150943396226415,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004589797203751497,
      "loss": 4.9379,
      "mean_token_accuracy": 0.213020621240139,
      "num_tokens": 44925102.0,
      "step": 25900
    },
    {
      "entropy": 5.63078236579895,
      "epoch": 2.015483368994359,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004589637641645645,
      "loss": 4.9128,
      "mean_token_accuracy": 0.2100828319787979,
      "num_tokens": 44933061.0,
      "step": 25905
    },
    {
      "entropy": 5.576472997665405,
      "epoch": 2.0158723983660765,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045894780516262193,
      "loss": 4.9534,
      "mean_token_accuracy": 0.20516390949487687,
      "num_tokens": 44942342.0,
      "step": 25910
    },
    {
      "entropy": 5.653770780563354,
      "epoch": 2.016261427737794,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004589318433695642,
      "loss": 4.94,
      "mean_token_accuracy": 0.2114293560385704,
      "num_tokens": 44951163.0,
      "step": 25915
    },
    {
      "entropy": 5.6057243824005125,
      "epoch": 2.016650457109512,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004589158787856336,
      "loss": 4.9295,
      "mean_token_accuracy": 0.20823193788528443,
      "num_tokens": 44960208.0,
      "step": 25920
    },
    {
      "entropy": 5.586337566375732,
      "epoch": 2.0170394864812295,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004588999114110721,
      "loss": 4.8233,
      "mean_token_accuracy": 0.21183962523937225,
      "num_tokens": 44969051.0,
      "step": 25925
    },
    {
      "entropy": 5.57568039894104,
      "epoch": 2.017428515852947,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004588839412461223,
      "loss": 4.8569,
      "mean_token_accuracy": 0.22637201398611068,
      "num_tokens": 44977980.0,
      "step": 25930
    },
    {
      "entropy": 5.608126831054688,
      "epoch": 2.0178175452246645,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004588679682910264,
      "loss": 4.9449,
      "mean_token_accuracy": 0.20136756151914598,
      "num_tokens": 44986520.0,
      "step": 25935
    },
    {
      "entropy": 5.625402927398682,
      "epoch": 2.018206574596382,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004588519925460266,
      "loss": 4.9393,
      "mean_token_accuracy": 0.20873477309942245,
      "num_tokens": 44995011.0,
      "step": 25940
    },
    {
      "entropy": 5.52529993057251,
      "epoch": 2.0185956039680995,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004588360140113655,
      "loss": 4.9227,
      "mean_token_accuracy": 0.21274462938308716,
      "num_tokens": 45004611.0,
      "step": 25945
    },
    {
      "entropy": 5.660669231414795,
      "epoch": 2.018984633339817,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004588200326872855,
      "loss": 4.967,
      "mean_token_accuracy": 0.19807026982307435,
      "num_tokens": 45013832.0,
      "step": 25950
    },
    {
      "entropy": 5.646408319473267,
      "epoch": 2.019373662711535,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004588040485740291,
      "loss": 4.9139,
      "mean_token_accuracy": 0.20660153329372405,
      "num_tokens": 45022751.0,
      "step": 25955
    },
    {
      "entropy": 5.687806463241577,
      "epoch": 2.019762692083252,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004587880616718387,
      "loss": 4.9797,
      "mean_token_accuracy": 0.20526391565799712,
      "num_tokens": 45030826.0,
      "step": 25960
    },
    {
      "entropy": 5.652901220321655,
      "epoch": 2.02015172145497,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004587720719809572,
      "loss": 4.9643,
      "mean_token_accuracy": 0.20641106218099595,
      "num_tokens": 45039625.0,
      "step": 25965
    },
    {
      "entropy": 5.605326223373413,
      "epoch": 2.0205407508266875,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004587560795016269,
      "loss": 4.9402,
      "mean_token_accuracy": 0.21145935654640197,
      "num_tokens": 45048229.0,
      "step": 25970
    },
    {
      "entropy": 5.666127872467041,
      "epoch": 2.020929780198405,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004587400842340905,
      "loss": 5.0239,
      "mean_token_accuracy": 0.19960868060588838,
      "num_tokens": 45057414.0,
      "step": 25975
    },
    {
      "entropy": 5.577289390563965,
      "epoch": 2.0213188095701224,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045872408617859083,
      "loss": 4.8649,
      "mean_token_accuracy": 0.2089320093393326,
      "num_tokens": 45065785.0,
      "step": 25980
    },
    {
      "entropy": 5.573744916915894,
      "epoch": 2.02170783894184,
      "grad_norm": 1.125,
      "learning_rate": 0.00045870808533537066,
      "loss": 4.8846,
      "mean_token_accuracy": 0.2090733453631401,
      "num_tokens": 45074896.0,
      "step": 25985
    },
    {
      "entropy": 5.543414306640625,
      "epoch": 2.022096868313558,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045869208170467256,
      "loss": 4.9254,
      "mean_token_accuracy": 0.20179485380649567,
      "num_tokens": 45083554.0,
      "step": 25990
    },
    {
      "entropy": 5.673861408233643,
      "epoch": 2.022485897685275,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004586760752867396,
      "loss": 4.9301,
      "mean_token_accuracy": 0.20531890988349916,
      "num_tokens": 45092204.0,
      "step": 25995
    },
    {
      "entropy": 5.665485000610351,
      "epoch": 2.022874927056993,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045866006608181456,
      "loss": 4.9589,
      "mean_token_accuracy": 0.19454343616962433,
      "num_tokens": 45101665.0,
      "step": 26000
    },
    {
      "entropy": 5.628693008422852,
      "epoch": 2.0232639564287105,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004586440540901403,
      "loss": 5.0074,
      "mean_token_accuracy": 0.20306645035743714,
      "num_tokens": 45111093.0,
      "step": 26005
    },
    {
      "entropy": 5.648394775390625,
      "epoch": 2.0236529858004277,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045862803931195976,
      "loss": 4.9419,
      "mean_token_accuracy": 0.2079176664352417,
      "num_tokens": 45118826.0,
      "step": 26010
    },
    {
      "entropy": 5.632839202880859,
      "epoch": 2.0240420151721454,
      "grad_norm": 1.125,
      "learning_rate": 0.0004586120217475161,
      "loss": 4.9641,
      "mean_token_accuracy": 0.21168821454048156,
      "num_tokens": 45128265.0,
      "step": 26015
    },
    {
      "entropy": 5.646838712692261,
      "epoch": 2.024431044543863,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004585960013970522,
      "loss": 5.014,
      "mean_token_accuracy": 0.20075671821832658,
      "num_tokens": 45137922.0,
      "step": 26020
    },
    {
      "entropy": 5.549643802642822,
      "epoch": 2.024820073915581,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004585799782608112,
      "loss": 4.8419,
      "mean_token_accuracy": 0.22199385464191437,
      "num_tokens": 45146242.0,
      "step": 26025
    },
    {
      "entropy": 5.670653009414673,
      "epoch": 2.025209103287298,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045856395233903624,
      "loss": 5.0058,
      "mean_token_accuracy": 0.1997980758547783,
      "num_tokens": 45154897.0,
      "step": 26030
    },
    {
      "entropy": 5.641484117507934,
      "epoch": 2.025598132659016,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004585479236319705,
      "loss": 4.9621,
      "mean_token_accuracy": 0.20345127880573272,
      "num_tokens": 45164786.0,
      "step": 26035
    },
    {
      "entropy": 5.6514880657196045,
      "epoch": 2.0259871620307335,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045853189213985714,
      "loss": 4.9998,
      "mean_token_accuracy": 0.1980445384979248,
      "num_tokens": 45173652.0,
      "step": 26040
    },
    {
      "entropy": 5.628904199600219,
      "epoch": 2.0263761914024507,
      "grad_norm": 1.125,
      "learning_rate": 0.00045851585786293943,
      "loss": 4.8814,
      "mean_token_accuracy": 0.20898549854755402,
      "num_tokens": 45182399.0,
      "step": 26045
    },
    {
      "entropy": 5.636854791641236,
      "epoch": 2.0267652207741684,
      "grad_norm": 1.125,
      "learning_rate": 0.00045849982080146067,
      "loss": 4.9524,
      "mean_token_accuracy": 0.19723586589097977,
      "num_tokens": 45191587.0,
      "step": 26050
    },
    {
      "entropy": 5.600432395935059,
      "epoch": 2.027154250145886,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004584837809556642,
      "loss": 4.9196,
      "mean_token_accuracy": 0.2080520585179329,
      "num_tokens": 45200032.0,
      "step": 26055
    },
    {
      "entropy": 5.606080675125122,
      "epoch": 2.0275432795176034,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045846773832579346,
      "loss": 4.9447,
      "mean_token_accuracy": 0.2100594937801361,
      "num_tokens": 45209205.0,
      "step": 26060
    },
    {
      "entropy": 5.7015380859375,
      "epoch": 2.027932308889321,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045845169291209176,
      "loss": 4.9449,
      "mean_token_accuracy": 0.20696453452110292,
      "num_tokens": 45217603.0,
      "step": 26065
    },
    {
      "entropy": 5.618218040466308,
      "epoch": 2.0283213382610388,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045843564471480263,
      "loss": 4.8949,
      "mean_token_accuracy": 0.20571276992559434,
      "num_tokens": 45225971.0,
      "step": 26070
    },
    {
      "entropy": 5.673703193664551,
      "epoch": 2.0287103676327565,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004584195937341695,
      "loss": 4.962,
      "mean_token_accuracy": 0.19924225509166718,
      "num_tokens": 45234494.0,
      "step": 26075
    },
    {
      "entropy": 5.58463225364685,
      "epoch": 2.0290993970044737,
      "grad_norm": 1.25,
      "learning_rate": 0.00045840353997043606,
      "loss": 4.7976,
      "mean_token_accuracy": 0.21467165052890777,
      "num_tokens": 45242685.0,
      "step": 26080
    },
    {
      "entropy": 5.582918262481689,
      "epoch": 2.0294884263761914,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004583874834238458,
      "loss": 4.8744,
      "mean_token_accuracy": 0.20863621532917023,
      "num_tokens": 45252282.0,
      "step": 26085
    },
    {
      "entropy": 5.640808200836181,
      "epoch": 2.029877455747909,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004583714240946423,
      "loss": 5.0418,
      "mean_token_accuracy": 0.19608078747987748,
      "num_tokens": 45261390.0,
      "step": 26090
    },
    {
      "entropy": 5.6888503074646,
      "epoch": 2.0302664851196264,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045835536198306936,
      "loss": 5.0561,
      "mean_token_accuracy": 0.1985889494419098,
      "num_tokens": 45270244.0,
      "step": 26095
    },
    {
      "entropy": 5.686486148834229,
      "epoch": 2.030655514491344,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045833929708937067,
      "loss": 4.9733,
      "mean_token_accuracy": 0.1979078993201256,
      "num_tokens": 45279404.0,
      "step": 26100
    },
    {
      "entropy": 5.624703502655029,
      "epoch": 2.0310445438630618,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004583232294137899,
      "loss": 4.9253,
      "mean_token_accuracy": 0.2153921514749527,
      "num_tokens": 45288064.0,
      "step": 26105
    },
    {
      "entropy": 5.65843186378479,
      "epoch": 2.031433573234779,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004583071589565709,
      "loss": 4.9787,
      "mean_token_accuracy": 0.202467879652977,
      "num_tokens": 45296528.0,
      "step": 26110
    },
    {
      "entropy": 5.578810405731201,
      "epoch": 2.0318226026064967,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004582910857179576,
      "loss": 4.8864,
      "mean_token_accuracy": 0.20989707112312317,
      "num_tokens": 45305524.0,
      "step": 26115
    },
    {
      "entropy": 5.674742603302002,
      "epoch": 2.0322116319782144,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004582750096981938,
      "loss": 5.0771,
      "mean_token_accuracy": 0.19384129345417023,
      "num_tokens": 45314386.0,
      "step": 26120
    },
    {
      "entropy": 5.6356751918792725,
      "epoch": 2.032600661349932,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004582589308975234,
      "loss": 4.9681,
      "mean_token_accuracy": 0.2020436316728592,
      "num_tokens": 45322901.0,
      "step": 26125
    },
    {
      "entropy": 5.605420875549316,
      "epoch": 2.0329896907216494,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045824284931619044,
      "loss": 4.8976,
      "mean_token_accuracy": 0.20827918946743013,
      "num_tokens": 45331509.0,
      "step": 26130
    },
    {
      "entropy": 5.650594568252563,
      "epoch": 2.033378720093367,
      "grad_norm": 1.125,
      "learning_rate": 0.00045822676495443893,
      "loss": 4.9428,
      "mean_token_accuracy": 0.2078060194849968,
      "num_tokens": 45340356.0,
      "step": 26135
    },
    {
      "entropy": 5.678124952316284,
      "epoch": 2.0337677494650848,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045821067781251284,
      "loss": 4.9768,
      "mean_token_accuracy": 0.2082965061068535,
      "num_tokens": 45349890.0,
      "step": 26140
    },
    {
      "entropy": 5.613535308837891,
      "epoch": 2.034156778836802,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045819458789065633,
      "loss": 4.9569,
      "mean_token_accuracy": 0.2071606770157814,
      "num_tokens": 45359249.0,
      "step": 26145
    },
    {
      "entropy": 5.626892995834351,
      "epoch": 2.0345458082085197,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004581784951891135,
      "loss": 4.9226,
      "mean_token_accuracy": 0.21266525387763976,
      "num_tokens": 45367832.0,
      "step": 26150
    },
    {
      "entropy": 5.638589334487915,
      "epoch": 2.0349348375802374,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004581623997081286,
      "loss": 4.9276,
      "mean_token_accuracy": 0.20842116624116896,
      "num_tokens": 45376316.0,
      "step": 26155
    },
    {
      "entropy": 5.6590207576751705,
      "epoch": 2.0353238669519547,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004581463014479459,
      "loss": 5.0001,
      "mean_token_accuracy": 0.1962370052933693,
      "num_tokens": 45385801.0,
      "step": 26160
    },
    {
      "entropy": 5.634700107574463,
      "epoch": 2.0357128963236724,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004581302004088095,
      "loss": 4.9503,
      "mean_token_accuracy": 0.20776505321264266,
      "num_tokens": 45394669.0,
      "step": 26165
    },
    {
      "entropy": 5.616348361968994,
      "epoch": 2.03610192569539,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004581140965909638,
      "loss": 4.9084,
      "mean_token_accuracy": 0.21000946015119554,
      "num_tokens": 45402896.0,
      "step": 26170
    },
    {
      "entropy": 5.601734161376953,
      "epoch": 2.0364909550671078,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004580979899946531,
      "loss": 4.9539,
      "mean_token_accuracy": 0.20440015941858292,
      "num_tokens": 45411047.0,
      "step": 26175
    },
    {
      "entropy": 5.664836311340332,
      "epoch": 2.036879984438825,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004580818806201219,
      "loss": 5.0632,
      "mean_token_accuracy": 0.20106602013111113,
      "num_tokens": 45420661.0,
      "step": 26180
    },
    {
      "entropy": 5.6778710842132565,
      "epoch": 2.0372690138105427,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045806576846761453,
      "loss": 4.9158,
      "mean_token_accuracy": 0.20963267236948013,
      "num_tokens": 45429199.0,
      "step": 26185
    },
    {
      "entropy": 5.676671648025513,
      "epoch": 2.0376580431822604,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045804965353737556,
      "loss": 4.9154,
      "mean_token_accuracy": 0.21138497292995453,
      "num_tokens": 45437504.0,
      "step": 26190
    },
    {
      "entropy": 5.563845539093018,
      "epoch": 2.0380470725539777,
      "grad_norm": 1.125,
      "learning_rate": 0.0004580335358296494,
      "loss": 4.8653,
      "mean_token_accuracy": 0.21282437443733215,
      "num_tokens": 45445696.0,
      "step": 26195
    },
    {
      "entropy": 5.593537473678589,
      "epoch": 2.0384361019256954,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045801741534468065,
      "loss": 4.9689,
      "mean_token_accuracy": 0.2060506284236908,
      "num_tokens": 45455058.0,
      "step": 26200
    },
    {
      "entropy": 5.773765802383423,
      "epoch": 2.038825131297413,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004580012920827139,
      "loss": 5.0528,
      "mean_token_accuracy": 0.20553303509950638,
      "num_tokens": 45464193.0,
      "step": 26205
    },
    {
      "entropy": 5.580555486679077,
      "epoch": 2.0392141606691303,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004579851660439939,
      "loss": 4.8209,
      "mean_token_accuracy": 0.21999395340681077,
      "num_tokens": 45471922.0,
      "step": 26210
    },
    {
      "entropy": 5.6359411716461185,
      "epoch": 2.039603190040848,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045796903722876523,
      "loss": 5.0041,
      "mean_token_accuracy": 0.19785224199295043,
      "num_tokens": 45480181.0,
      "step": 26215
    },
    {
      "entropy": 5.711096525192261,
      "epoch": 2.0399922194125657,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004579529056372726,
      "loss": 5.0314,
      "mean_token_accuracy": 0.2001571998000145,
      "num_tokens": 45488970.0,
      "step": 26220
    },
    {
      "entropy": 5.751931428909302,
      "epoch": 2.0403812487842834,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004579367712697609,
      "loss": 4.9344,
      "mean_token_accuracy": 0.20439580529928209,
      "num_tokens": 45497066.0,
      "step": 26225
    },
    {
      "entropy": 5.626629877090454,
      "epoch": 2.0407702781560006,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045792063412647475,
      "loss": 4.9197,
      "mean_token_accuracy": 0.20418355911970137,
      "num_tokens": 45506399.0,
      "step": 26230
    },
    {
      "entropy": 5.582385873794555,
      "epoch": 2.0411593075277183,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045790449420765925,
      "loss": 4.9341,
      "mean_token_accuracy": 0.199679172039032,
      "num_tokens": 45516625.0,
      "step": 26235
    },
    {
      "entropy": 5.6861964702606205,
      "epoch": 2.041548336899436,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045788835151355914,
      "loss": 5.0404,
      "mean_token_accuracy": 0.1987278938293457,
      "num_tokens": 45525539.0,
      "step": 26240
    },
    {
      "entropy": 5.691116237640381,
      "epoch": 2.0419373662711533,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045787220604441933,
      "loss": 4.9561,
      "mean_token_accuracy": 0.20885669887065889,
      "num_tokens": 45535388.0,
      "step": 26245
    },
    {
      "entropy": 5.685454988479615,
      "epoch": 2.042326395642871,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045785605780048497,
      "loss": 5.0089,
      "mean_token_accuracy": 0.2019638791680336,
      "num_tokens": 45543729.0,
      "step": 26250
    },
    {
      "entropy": 5.677403688430786,
      "epoch": 2.0427154250145887,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045783990678200086,
      "loss": 4.9996,
      "mean_token_accuracy": 0.20386107116937638,
      "num_tokens": 45551887.0,
      "step": 26255
    },
    {
      "entropy": 5.732798862457275,
      "epoch": 2.043104454386306,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045782375298921227,
      "loss": 5.0628,
      "mean_token_accuracy": 0.2009087935090065,
      "num_tokens": 45560267.0,
      "step": 26260
    },
    {
      "entropy": 5.679188585281372,
      "epoch": 2.0434934837580236,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004578075964223642,
      "loss": 4.9688,
      "mean_token_accuracy": 0.20053595304489136,
      "num_tokens": 45569459.0,
      "step": 26265
    },
    {
      "entropy": 5.734493350982666,
      "epoch": 2.0438825131297413,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004577914370817018,
      "loss": 4.9525,
      "mean_token_accuracy": 0.20151663869619368,
      "num_tokens": 45577713.0,
      "step": 26270
    },
    {
      "entropy": 5.595556640625,
      "epoch": 2.044271542501459,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045777527496747034,
      "loss": 4.8888,
      "mean_token_accuracy": 0.2062281087040901,
      "num_tokens": 45586521.0,
      "step": 26275
    },
    {
      "entropy": 5.640928077697754,
      "epoch": 2.0446605718731763,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045775911007991493,
      "loss": 4.9307,
      "mean_token_accuracy": 0.2075049564242363,
      "num_tokens": 45595210.0,
      "step": 26280
    },
    {
      "entropy": 5.643621921539307,
      "epoch": 2.045049601244894,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045774294241928093,
      "loss": 4.9864,
      "mean_token_accuracy": 0.20238704234361649,
      "num_tokens": 45603472.0,
      "step": 26285
    },
    {
      "entropy": 5.63580813407898,
      "epoch": 2.0454386306166117,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004577267719858137,
      "loss": 5.0239,
      "mean_token_accuracy": 0.19817969799041749,
      "num_tokens": 45612149.0,
      "step": 26290
    },
    {
      "entropy": 5.748331117630005,
      "epoch": 2.045827659988329,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045771059877975853,
      "loss": 5.0346,
      "mean_token_accuracy": 0.2027564078569412,
      "num_tokens": 45619903.0,
      "step": 26295
    },
    {
      "entropy": 5.681639957427978,
      "epoch": 2.0462166893600466,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004576944228013608,
      "loss": 4.9474,
      "mean_token_accuracy": 0.2076179191470146,
      "num_tokens": 45628012.0,
      "step": 26300
    },
    {
      "entropy": 5.6106048107147215,
      "epoch": 2.0466057187317643,
      "grad_norm": 1.21875,
      "learning_rate": 0.000457678244050866,
      "loss": 4.9792,
      "mean_token_accuracy": 0.2022384375333786,
      "num_tokens": 45636587.0,
      "step": 26305
    },
    {
      "entropy": 5.630101919174194,
      "epoch": 2.046994748103482,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004576620625285196,
      "loss": 4.9276,
      "mean_token_accuracy": 0.20198894888162613,
      "num_tokens": 45645189.0,
      "step": 26310
    },
    {
      "entropy": 5.694034004211426,
      "epoch": 2.0473837774751993,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045764587823456717,
      "loss": 5.0113,
      "mean_token_accuracy": 0.201967491209507,
      "num_tokens": 45654059.0,
      "step": 26315
    },
    {
      "entropy": 5.634521007537842,
      "epoch": 2.047772806846917,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045762969116925424,
      "loss": 4.87,
      "mean_token_accuracy": 0.21535642743110656,
      "num_tokens": 45662125.0,
      "step": 26320
    },
    {
      "entropy": 5.645720148086548,
      "epoch": 2.0481618362186347,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045761350133282643,
      "loss": 4.9427,
      "mean_token_accuracy": 0.2078005328774452,
      "num_tokens": 45670068.0,
      "step": 26325
    },
    {
      "entropy": 5.604268169403076,
      "epoch": 2.048550865590352,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045759730872552937,
      "loss": 4.8236,
      "mean_token_accuracy": 0.21853054016828538,
      "num_tokens": 45678717.0,
      "step": 26330
    },
    {
      "entropy": 5.626949167251587,
      "epoch": 2.0489398949620696,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004575811133476088,
      "loss": 4.9405,
      "mean_token_accuracy": 0.20620205849409104,
      "num_tokens": 45686882.0,
      "step": 26335
    },
    {
      "entropy": 5.5663713932037355,
      "epoch": 2.0493289243337873,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045756491519931047,
      "loss": 4.8584,
      "mean_token_accuracy": 0.21039746403694154,
      "num_tokens": 45695460.0,
      "step": 26340
    },
    {
      "entropy": 5.733092737197876,
      "epoch": 2.0497179537055046,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004575487142808801,
      "loss": 5.0207,
      "mean_token_accuracy": 0.19638052880764006,
      "num_tokens": 45703287.0,
      "step": 26345
    },
    {
      "entropy": 5.734856033325196,
      "epoch": 2.0501069830772223,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004575325105925636,
      "loss": 5.0494,
      "mean_token_accuracy": 0.19611111730337144,
      "num_tokens": 45712830.0,
      "step": 26350
    },
    {
      "entropy": 5.635013914108276,
      "epoch": 2.05049601244894,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045751630413460665,
      "loss": 5.0323,
      "mean_token_accuracy": 0.19877097010612488,
      "num_tokens": 45720701.0,
      "step": 26355
    },
    {
      "entropy": 5.633447265625,
      "epoch": 2.0508850418206577,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004575000949072554,
      "loss": 4.9079,
      "mean_token_accuracy": 0.21525898873806,
      "num_tokens": 45728663.0,
      "step": 26360
    },
    {
      "entropy": 5.6398577213287355,
      "epoch": 2.051274071192375,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004574838829107557,
      "loss": 4.9511,
      "mean_token_accuracy": 0.2039007917046547,
      "num_tokens": 45737661.0,
      "step": 26365
    },
    {
      "entropy": 5.57434401512146,
      "epoch": 2.0516631005640926,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004574676681453535,
      "loss": 4.899,
      "mean_token_accuracy": 0.21216239929199218,
      "num_tokens": 45746343.0,
      "step": 26370
    },
    {
      "entropy": 5.592216825485229,
      "epoch": 2.0520521299358103,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045745145061129485,
      "loss": 4.95,
      "mean_token_accuracy": 0.20669371485710145,
      "num_tokens": 45755025.0,
      "step": 26375
    },
    {
      "entropy": 5.646818780899048,
      "epoch": 2.0524411593075276,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045743523030882584,
      "loss": 5.0233,
      "mean_token_accuracy": 0.20067696571350097,
      "num_tokens": 45763107.0,
      "step": 26380
    },
    {
      "entropy": 5.642067575454712,
      "epoch": 2.0528301886792453,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004574190072381926,
      "loss": 4.8891,
      "mean_token_accuracy": 0.21034082174301147,
      "num_tokens": 45771754.0,
      "step": 26385
    },
    {
      "entropy": 5.675598239898681,
      "epoch": 2.053219218050963,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004574027813996413,
      "loss": 5.0595,
      "mean_token_accuracy": 0.19678059965372086,
      "num_tokens": 45779282.0,
      "step": 26390
    },
    {
      "entropy": 5.675804615020752,
      "epoch": 2.05360824742268,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004573865527934181,
      "loss": 5.0194,
      "mean_token_accuracy": 0.19829467833042144,
      "num_tokens": 45788002.0,
      "step": 26395
    },
    {
      "entropy": 5.72514328956604,
      "epoch": 2.053997276794398,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004573703214197692,
      "loss": 5.0249,
      "mean_token_accuracy": 0.1953348085284233,
      "num_tokens": 45796563.0,
      "step": 26400
    },
    {
      "entropy": 5.636849641799927,
      "epoch": 2.0543863061661156,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045735408727894095,
      "loss": 4.9229,
      "mean_token_accuracy": 0.20516045093536378,
      "num_tokens": 45806311.0,
      "step": 26405
    },
    {
      "entropy": 5.6359052658081055,
      "epoch": 2.0547753355378333,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045733785037117977,
      "loss": 4.9105,
      "mean_token_accuracy": 0.21418850123882294,
      "num_tokens": 45814007.0,
      "step": 26410
    },
    {
      "entropy": 5.63319730758667,
      "epoch": 2.0551643649095506,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004573216106967319,
      "loss": 4.9617,
      "mean_token_accuracy": 0.20315697193145751,
      "num_tokens": 45822492.0,
      "step": 26415
    },
    {
      "entropy": 5.599131965637207,
      "epoch": 2.0555533942812683,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045730536825584365,
      "loss": 4.8926,
      "mean_token_accuracy": 0.20820895284414292,
      "num_tokens": 45830853.0,
      "step": 26420
    },
    {
      "entropy": 5.715424871444702,
      "epoch": 2.055942423652986,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045728912304876176,
      "loss": 4.9859,
      "mean_token_accuracy": 0.19848760068416596,
      "num_tokens": 45839948.0,
      "step": 26425
    },
    {
      "entropy": 5.758290386199951,
      "epoch": 2.056331453024703,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004572728750757326,
      "loss": 5.1012,
      "mean_token_accuracy": 0.1859108477830887,
      "num_tokens": 45848264.0,
      "step": 26430
    },
    {
      "entropy": 5.562944650650024,
      "epoch": 2.056720482396421,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004572566243370025,
      "loss": 4.8477,
      "mean_token_accuracy": 0.21397773176431656,
      "num_tokens": 45856523.0,
      "step": 26435
    },
    {
      "entropy": 5.611719608306885,
      "epoch": 2.0571095117681386,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004572403708328183,
      "loss": 4.9683,
      "mean_token_accuracy": 0.20735953599214554,
      "num_tokens": 45864736.0,
      "step": 26440
    },
    {
      "entropy": 5.642543458938599,
      "epoch": 2.057498541139856,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004572241145634266,
      "loss": 4.973,
      "mean_token_accuracy": 0.20111494660377502,
      "num_tokens": 45873979.0,
      "step": 26445
    },
    {
      "entropy": 5.604004907608032,
      "epoch": 2.0578875705115736,
      "grad_norm": 1.15625,
      "learning_rate": 0.000457207855529074,
      "loss": 4.9414,
      "mean_token_accuracy": 0.20514785647392272,
      "num_tokens": 45882545.0,
      "step": 26450
    },
    {
      "entropy": 5.63154935836792,
      "epoch": 2.0582765998832913,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045719159373000713,
      "loss": 5.0199,
      "mean_token_accuracy": 0.19547130316495895,
      "num_tokens": 45890830.0,
      "step": 26455
    },
    {
      "entropy": 5.728578424453735,
      "epoch": 2.058665629255009,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004571753291664729,
      "loss": 4.9734,
      "mean_token_accuracy": 0.2012726277112961,
      "num_tokens": 45899457.0,
      "step": 26460
    },
    {
      "entropy": 5.657454252243042,
      "epoch": 2.059054658626726,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004571590618387179,
      "loss": 4.9712,
      "mean_token_accuracy": 0.20497499108314515,
      "num_tokens": 45907998.0,
      "step": 26465
    },
    {
      "entropy": 5.597486400604248,
      "epoch": 2.059443687998444,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004571427917469892,
      "loss": 4.9178,
      "mean_token_accuracy": 0.2081871137022972,
      "num_tokens": 45916102.0,
      "step": 26470
    },
    {
      "entropy": 5.617994165420532,
      "epoch": 2.0598327173701616,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045712651889153345,
      "loss": 4.9394,
      "mean_token_accuracy": 0.20682152807712556,
      "num_tokens": 45924347.0,
      "step": 26475
    },
    {
      "entropy": 5.704796648025512,
      "epoch": 2.060221746741879,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045711024327259764,
      "loss": 4.9612,
      "mean_token_accuracy": 0.20622393041849135,
      "num_tokens": 45932702.0,
      "step": 26480
    },
    {
      "entropy": 5.556574583053589,
      "epoch": 2.0606107761135966,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045709396489042884,
      "loss": 4.9184,
      "mean_token_accuracy": 0.20257857590913772,
      "num_tokens": 45941235.0,
      "step": 26485
    },
    {
      "entropy": 5.637005233764649,
      "epoch": 2.0609998054853143,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045707768374527385,
      "loss": 4.9084,
      "mean_token_accuracy": 0.21213483661413193,
      "num_tokens": 45949582.0,
      "step": 26490
    },
    {
      "entropy": 5.638558721542358,
      "epoch": 2.0613888348570315,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004570613998373799,
      "loss": 4.9875,
      "mean_token_accuracy": 0.2043515920639038,
      "num_tokens": 45957920.0,
      "step": 26495
    },
    {
      "entropy": 5.6318505764007565,
      "epoch": 2.061777864228749,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045704511316699395,
      "loss": 5.0363,
      "mean_token_accuracy": 0.20439658015966417,
      "num_tokens": 45966302.0,
      "step": 26500
    },
    {
      "entropy": 5.673410415649414,
      "epoch": 2.062166893600467,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045702882373436317,
      "loss": 5.0381,
      "mean_token_accuracy": 0.20185166597366333,
      "num_tokens": 45975481.0,
      "step": 26505
    },
    {
      "entropy": 5.789321947097778,
      "epoch": 2.0625559229721846,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004570125315397347,
      "loss": 5.0357,
      "mean_token_accuracy": 0.20039262473583222,
      "num_tokens": 45984466.0,
      "step": 26510
    },
    {
      "entropy": 5.60607328414917,
      "epoch": 2.062944952343902,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004569962365833558,
      "loss": 4.8755,
      "mean_token_accuracy": 0.20871899425983428,
      "num_tokens": 45992862.0,
      "step": 26515
    },
    {
      "entropy": 5.623969650268554,
      "epoch": 2.0633339817156195,
      "grad_norm": 1.25,
      "learning_rate": 0.00045697993886547374,
      "loss": 4.9517,
      "mean_token_accuracy": 0.2035495236515999,
      "num_tokens": 46001211.0,
      "step": 26520
    },
    {
      "entropy": 5.686485624313354,
      "epoch": 2.0637230110873372,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045696363838633566,
      "loss": 4.9896,
      "mean_token_accuracy": 0.2002142995595932,
      "num_tokens": 46009574.0,
      "step": 26525
    },
    {
      "entropy": 5.636230611801148,
      "epoch": 2.0641120404590545,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045694733514618904,
      "loss": 4.9116,
      "mean_token_accuracy": 0.2045785516500473,
      "num_tokens": 46017686.0,
      "step": 26530
    },
    {
      "entropy": 5.6358551502227785,
      "epoch": 2.064501069830772,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004569310291452812,
      "loss": 4.9338,
      "mean_token_accuracy": 0.2003798305988312,
      "num_tokens": 46026842.0,
      "step": 26535
    },
    {
      "entropy": 5.629838132858277,
      "epoch": 2.06489009920249,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004569147203838595,
      "loss": 4.9378,
      "mean_token_accuracy": 0.20564806908369065,
      "num_tokens": 46036308.0,
      "step": 26540
    },
    {
      "entropy": 5.6653971672058105,
      "epoch": 2.065279128574207,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045689840886217157,
      "loss": 4.947,
      "mean_token_accuracy": 0.20572903603315354,
      "num_tokens": 46045424.0,
      "step": 26545
    },
    {
      "entropy": 5.6023633003234865,
      "epoch": 2.065668157945925,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045688209458046475,
      "loss": 4.9313,
      "mean_token_accuracy": 0.2061443418264389,
      "num_tokens": 46054386.0,
      "step": 26550
    },
    {
      "entropy": 5.579394149780273,
      "epoch": 2.0660571873176425,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045686577753898663,
      "loss": 4.8595,
      "mean_token_accuracy": 0.213695752620697,
      "num_tokens": 46063617.0,
      "step": 26555
    },
    {
      "entropy": 5.668267822265625,
      "epoch": 2.0664462166893602,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045684945773798483,
      "loss": 5.0625,
      "mean_token_accuracy": 0.19994451105594635,
      "num_tokens": 46072915.0,
      "step": 26560
    },
    {
      "entropy": 5.688279485702514,
      "epoch": 2.0668352460610775,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000456833135177707,
      "loss": 5.0845,
      "mean_token_accuracy": 0.1986884132027626,
      "num_tokens": 46081890.0,
      "step": 26565
    },
    {
      "entropy": 5.746045160293579,
      "epoch": 2.067224275432795,
      "grad_norm": 1.25,
      "learning_rate": 0.0004568168098584007,
      "loss": 5.0457,
      "mean_token_accuracy": 0.1940393403172493,
      "num_tokens": 46090445.0,
      "step": 26570
    },
    {
      "entropy": 5.734163236618042,
      "epoch": 2.067613304804513,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004568004817803137,
      "loss": 5.0085,
      "mean_token_accuracy": 0.1993106409907341,
      "num_tokens": 46099866.0,
      "step": 26575
    },
    {
      "entropy": 5.617708158493042,
      "epoch": 2.06800233417623,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004567841509436937,
      "loss": 4.9103,
      "mean_token_accuracy": 0.21352583467960357,
      "num_tokens": 46108036.0,
      "step": 26580
    },
    {
      "entropy": 5.646441888809204,
      "epoch": 2.068391363547948,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004567678173487887,
      "loss": 4.884,
      "mean_token_accuracy": 0.2175465703010559,
      "num_tokens": 46116087.0,
      "step": 26585
    },
    {
      "entropy": 5.606901454925537,
      "epoch": 2.0687803929196655,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004567514809958462,
      "loss": 4.9626,
      "mean_token_accuracy": 0.21432460099458694,
      "num_tokens": 46124934.0,
      "step": 26590
    },
    {
      "entropy": 5.6626299858093265,
      "epoch": 2.069169422291383,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004567351418851144,
      "loss": 4.9744,
      "mean_token_accuracy": 0.20717561841011048,
      "num_tokens": 46133691.0,
      "step": 26595
    },
    {
      "entropy": 5.671142578125,
      "epoch": 2.0695584516631005,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045671880001684113,
      "loss": 5.0326,
      "mean_token_accuracy": 0.19388167560100555,
      "num_tokens": 46141933.0,
      "step": 26600
    },
    {
      "entropy": 5.601749563217163,
      "epoch": 2.069947481034818,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045670245539127413,
      "loss": 4.8919,
      "mean_token_accuracy": 0.20567837059497834,
      "num_tokens": 46150656.0,
      "step": 26605
    },
    {
      "entropy": 5.567006826400757,
      "epoch": 2.070336510406536,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045668610800866173,
      "loss": 4.9301,
      "mean_token_accuracy": 0.20382735133171082,
      "num_tokens": 46160353.0,
      "step": 26610
    },
    {
      "entropy": 5.634338092803955,
      "epoch": 2.070725539778253,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045666975786925177,
      "loss": 4.9756,
      "mean_token_accuracy": 0.20893378853797911,
      "num_tokens": 46168810.0,
      "step": 26615
    },
    {
      "entropy": 5.655444765090943,
      "epoch": 2.071114569149971,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004566534049732923,
      "loss": 4.9436,
      "mean_token_accuracy": 0.2044213518500328,
      "num_tokens": 46177593.0,
      "step": 26620
    },
    {
      "entropy": 5.665983867645264,
      "epoch": 2.0715035985216885,
      "grad_norm": 1.25,
      "learning_rate": 0.00045663704932103166,
      "loss": 4.9508,
      "mean_token_accuracy": 0.2078593611717224,
      "num_tokens": 46186027.0,
      "step": 26625
    },
    {
      "entropy": 5.658973503112793,
      "epoch": 2.071892627893406,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045662069091271785,
      "loss": 5.0546,
      "mean_token_accuracy": 0.19743530303239823,
      "num_tokens": 46194705.0,
      "step": 26630
    },
    {
      "entropy": 5.630601596832276,
      "epoch": 2.0722816572651235,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045660432974859924,
      "loss": 4.8891,
      "mean_token_accuracy": 0.21460187286138535,
      "num_tokens": 46203498.0,
      "step": 26635
    },
    {
      "entropy": 5.664354467391968,
      "epoch": 2.072670686636841,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045658796582892383,
      "loss": 4.9532,
      "mean_token_accuracy": 0.2090080574154854,
      "num_tokens": 46211918.0,
      "step": 26640
    },
    {
      "entropy": 5.693994569778442,
      "epoch": 2.073059716008559,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045657159915394013,
      "loss": 5.0472,
      "mean_token_accuracy": 0.2011939987540245,
      "num_tokens": 46220689.0,
      "step": 26645
    },
    {
      "entropy": 5.6612037181854244,
      "epoch": 2.073448745380276,
      "grad_norm": 1.125,
      "learning_rate": 0.0004565552297238964,
      "loss": 5.023,
      "mean_token_accuracy": 0.19944491535425185,
      "num_tokens": 46229645.0,
      "step": 26650
    },
    {
      "entropy": 5.717441606521606,
      "epoch": 2.073837774751994,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000456538857539041,
      "loss": 4.9726,
      "mean_token_accuracy": 0.1966388627886772,
      "num_tokens": 46237691.0,
      "step": 26655
    },
    {
      "entropy": 5.655142164230346,
      "epoch": 2.0742268041237115,
      "grad_norm": 1.25,
      "learning_rate": 0.00045652248259962254,
      "loss": 4.9545,
      "mean_token_accuracy": 0.20381524860858918,
      "num_tokens": 46245756.0,
      "step": 26660
    },
    {
      "entropy": 5.708768653869629,
      "epoch": 2.0746158334954288,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004565061049058892,
      "loss": 5.041,
      "mean_token_accuracy": 0.19739940315485,
      "num_tokens": 46255233.0,
      "step": 26665
    },
    {
      "entropy": 5.683301401138306,
      "epoch": 2.0750048628671465,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00045648972445808963,
      "loss": 4.9887,
      "mean_token_accuracy": 0.19785669445991516,
      "num_tokens": 46263569.0,
      "step": 26670
    },
    {
      "entropy": 5.559755468368531,
      "epoch": 2.075393892238864,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045647334125647235,
      "loss": 4.8676,
      "mean_token_accuracy": 0.21105239987373353,
      "num_tokens": 46273010.0,
      "step": 26675
    },
    {
      "entropy": 5.686338615417481,
      "epoch": 2.0757829216105814,
      "grad_norm": 1.109375,
      "learning_rate": 0.000456456955301286,
      "loss": 4.9605,
      "mean_token_accuracy": 0.20988255590200425,
      "num_tokens": 46281566.0,
      "step": 26680
    },
    {
      "entropy": 5.560998582839966,
      "epoch": 2.076171950982299,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004564405665927791,
      "loss": 4.8495,
      "mean_token_accuracy": 0.21517568230628967,
      "num_tokens": 46290267.0,
      "step": 26685
    },
    {
      "entropy": 5.598146867752075,
      "epoch": 2.076560980354017,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045642417513120043,
      "loss": 4.8765,
      "mean_token_accuracy": 0.20851069390773774,
      "num_tokens": 46298694.0,
      "step": 26690
    },
    {
      "entropy": 5.647942066192627,
      "epoch": 2.076950009725734,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045640778091679876,
      "loss": 4.8363,
      "mean_token_accuracy": 0.2128787413239479,
      "num_tokens": 46307358.0,
      "step": 26695
    },
    {
      "entropy": 5.620609617233276,
      "epoch": 2.0773390390974518,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004563913839498226,
      "loss": 4.8886,
      "mean_token_accuracy": 0.2124302253127098,
      "num_tokens": 46316351.0,
      "step": 26700
    },
    {
      "entropy": 5.617027521133423,
      "epoch": 2.0777280684691695,
      "grad_norm": 1.171875,
      "learning_rate": 0.000456374984230521,
      "loss": 4.945,
      "mean_token_accuracy": 0.20778462439775466,
      "num_tokens": 46325591.0,
      "step": 26705
    },
    {
      "entropy": 5.6449486255645756,
      "epoch": 2.078117097840887,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004563585817591427,
      "loss": 4.9352,
      "mean_token_accuracy": 0.2057796224951744,
      "num_tokens": 46334595.0,
      "step": 26710
    },
    {
      "entropy": 5.733292913436889,
      "epoch": 2.0785061272126044,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004563421765359365,
      "loss": 5.0212,
      "mean_token_accuracy": 0.19988455772399902,
      "num_tokens": 46343587.0,
      "step": 26715
    },
    {
      "entropy": 5.755514097213745,
      "epoch": 2.078895156584322,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045632576856115156,
      "loss": 4.9953,
      "mean_token_accuracy": 0.20242587476968765,
      "num_tokens": 46351900.0,
      "step": 26720
    },
    {
      "entropy": 5.686200380325317,
      "epoch": 2.07928418595604,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045630935783503657,
      "loss": 5.0058,
      "mean_token_accuracy": 0.20042734891176223,
      "num_tokens": 46359854.0,
      "step": 26725
    },
    {
      "entropy": 5.6967932224273685,
      "epoch": 2.079673215327757,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004562929443578407,
      "loss": 5.0846,
      "mean_token_accuracy": 0.19680868983268737,
      "num_tokens": 46368930.0,
      "step": 26730
    },
    {
      "entropy": 5.645068550109864,
      "epoch": 2.0800622446994748,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004562765281298129,
      "loss": 4.8754,
      "mean_token_accuracy": 0.20406321585178375,
      "num_tokens": 46377522.0,
      "step": 26735
    },
    {
      "entropy": 5.64232611656189,
      "epoch": 2.0804512740711925,
      "grad_norm": 1.25,
      "learning_rate": 0.0004562601091512024,
      "loss": 4.9759,
      "mean_token_accuracy": 0.199060919880867,
      "num_tokens": 46386368.0,
      "step": 26740
    },
    {
      "entropy": 5.644097137451172,
      "epoch": 2.08084030344291,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004562436874222582,
      "loss": 4.9338,
      "mean_token_accuracy": 0.20560528486967086,
      "num_tokens": 46395363.0,
      "step": 26745
    },
    {
      "entropy": 5.692240142822266,
      "epoch": 2.0812293328146274,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045622726294322955,
      "loss": 4.9897,
      "mean_token_accuracy": 0.20273148119449616,
      "num_tokens": 46403998.0,
      "step": 26750
    },
    {
      "entropy": 5.559898090362549,
      "epoch": 2.081618362186345,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045621083571436563,
      "loss": 4.796,
      "mean_token_accuracy": 0.21877406984567643,
      "num_tokens": 46412138.0,
      "step": 26755
    },
    {
      "entropy": 5.623029994964599,
      "epoch": 2.082007391558063,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004561944057359157,
      "loss": 4.9595,
      "mean_token_accuracy": 0.20859382748603822,
      "num_tokens": 46420761.0,
      "step": 26760
    },
    {
      "entropy": 5.580400180816651,
      "epoch": 2.08239642092978,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004561779730081291,
      "loss": 4.9106,
      "mean_token_accuracy": 0.21232731491327286,
      "num_tokens": 46428938.0,
      "step": 26765
    },
    {
      "entropy": 5.67702841758728,
      "epoch": 2.0827854503014978,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004561615375312551,
      "loss": 4.9172,
      "mean_token_accuracy": 0.21263327300548554,
      "num_tokens": 46436552.0,
      "step": 26770
    },
    {
      "entropy": 5.587342834472656,
      "epoch": 2.0831744796732155,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045614509930554304,
      "loss": 4.8881,
      "mean_token_accuracy": 0.20578060299158096,
      "num_tokens": 46445223.0,
      "step": 26775
    },
    {
      "entropy": 5.619335556030274,
      "epoch": 2.0835635090449327,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045612865833124253,
      "loss": 4.9935,
      "mean_token_accuracy": 0.20630675107240676,
      "num_tokens": 46453232.0,
      "step": 26780
    },
    {
      "entropy": 5.514754867553711,
      "epoch": 2.0839525384166504,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004561122146086029,
      "loss": 4.8111,
      "mean_token_accuracy": 0.21386877000331878,
      "num_tokens": 46462562.0,
      "step": 26785
    },
    {
      "entropy": 5.638877820968628,
      "epoch": 2.084341567788368,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004560957681378737,
      "loss": 5.0083,
      "mean_token_accuracy": 0.1961371809244156,
      "num_tokens": 46470599.0,
      "step": 26790
    },
    {
      "entropy": 5.694039964675904,
      "epoch": 2.084730597160086,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045607931891930445,
      "loss": 4.9618,
      "mean_token_accuracy": 0.20422324985265733,
      "num_tokens": 46478707.0,
      "step": 26795
    },
    {
      "entropy": 5.613685846328735,
      "epoch": 2.085119626531803,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004560628669531447,
      "loss": 4.9417,
      "mean_token_accuracy": 0.2121308669447899,
      "num_tokens": 46487044.0,
      "step": 26800
    },
    {
      "entropy": 5.6448140144348145,
      "epoch": 2.0855086559035207,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045604641223964416,
      "loss": 4.9564,
      "mean_token_accuracy": 0.19777119904756546,
      "num_tokens": 46495278.0,
      "step": 26805
    },
    {
      "entropy": 5.62329421043396,
      "epoch": 2.0858976852752384,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045602995477905247,
      "loss": 4.9523,
      "mean_token_accuracy": 0.21106358915567397,
      "num_tokens": 46503545.0,
      "step": 26810
    },
    {
      "entropy": 5.626003313064575,
      "epoch": 2.0862867146469557,
      "grad_norm": 1.125,
      "learning_rate": 0.0004560134945716194,
      "loss": 5.0131,
      "mean_token_accuracy": 0.2157517269253731,
      "num_tokens": 46513058.0,
      "step": 26815
    },
    {
      "entropy": 5.663360166549682,
      "epoch": 2.0866757440186734,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00045599703161759464,
      "loss": 4.9227,
      "mean_token_accuracy": 0.20297565758228303,
      "num_tokens": 46520651.0,
      "step": 26820
    },
    {
      "entropy": 5.632119274139404,
      "epoch": 2.087064773390391,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045598056591722805,
      "loss": 4.895,
      "mean_token_accuracy": 0.20758887082338334,
      "num_tokens": 46528899.0,
      "step": 26825
    },
    {
      "entropy": 5.6920829772949215,
      "epoch": 2.0874538027621083,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045596409747076936,
      "loss": 5.0822,
      "mean_token_accuracy": 0.19562147110700606,
      "num_tokens": 46537921.0,
      "step": 26830
    },
    {
      "entropy": 5.656803226470947,
      "epoch": 2.087842832133826,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004559476262784686,
      "loss": 4.9258,
      "mean_token_accuracy": 0.20856411755084991,
      "num_tokens": 46547432.0,
      "step": 26835
    },
    {
      "entropy": 5.647641944885254,
      "epoch": 2.0882318615055437,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004559311523405755,
      "loss": 4.9748,
      "mean_token_accuracy": 0.2019300252199173,
      "num_tokens": 46555406.0,
      "step": 26840
    },
    {
      "entropy": 5.661641836166382,
      "epoch": 2.0886208908772614,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004559146756573402,
      "loss": 4.9816,
      "mean_token_accuracy": 0.2056744784116745,
      "num_tokens": 46564428.0,
      "step": 26845
    },
    {
      "entropy": 5.6842903137207035,
      "epoch": 2.0890099202489787,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045589819622901274,
      "loss": 4.8964,
      "mean_token_accuracy": 0.2122207224369049,
      "num_tokens": 46573222.0,
      "step": 26850
    },
    {
      "entropy": 5.631218957901001,
      "epoch": 2.0893989496206964,
      "grad_norm": 1.203125,
      "learning_rate": 0.000455881714055843,
      "loss": 4.912,
      "mean_token_accuracy": 0.20891210287809373,
      "num_tokens": 46581951.0,
      "step": 26855
    },
    {
      "entropy": 5.68305606842041,
      "epoch": 2.089787978992414,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045586522913808114,
      "loss": 5.0412,
      "mean_token_accuracy": 0.20667289197444916,
      "num_tokens": 46590658.0,
      "step": 26860
    },
    {
      "entropy": 5.646737957000733,
      "epoch": 2.0901770083641313,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004558487414759773,
      "loss": 4.8771,
      "mean_token_accuracy": 0.22141017019748688,
      "num_tokens": 46599068.0,
      "step": 26865
    },
    {
      "entropy": 5.6099982261657715,
      "epoch": 2.090566037735849,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045583225106978175,
      "loss": 4.9697,
      "mean_token_accuracy": 0.20863139778375625,
      "num_tokens": 46608024.0,
      "step": 26870
    },
    {
      "entropy": 5.6124852180480955,
      "epoch": 2.0909550671075667,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004558157579197446,
      "loss": 4.9029,
      "mean_token_accuracy": 0.20791249424219133,
      "num_tokens": 46616558.0,
      "step": 26875
    },
    {
      "entropy": 5.728190660476685,
      "epoch": 2.091344096479284,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045579926202611603,
      "loss": 5.105,
      "mean_token_accuracy": 0.1906801924109459,
      "num_tokens": 46624943.0,
      "step": 26880
    },
    {
      "entropy": 5.63717622756958,
      "epoch": 2.0917331258510017,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004557827633891465,
      "loss": 4.9601,
      "mean_token_accuracy": 0.1995410978794098,
      "num_tokens": 46633753.0,
      "step": 26885
    },
    {
      "entropy": 5.702869462966919,
      "epoch": 2.0921221552227194,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004557662620090863,
      "loss": 4.9813,
      "mean_token_accuracy": 0.20546361058950424,
      "num_tokens": 46642210.0,
      "step": 26890
    },
    {
      "entropy": 5.660136938095093,
      "epoch": 2.092511184594437,
      "grad_norm": 1.25,
      "learning_rate": 0.00045574975788618564,
      "loss": 4.9173,
      "mean_token_accuracy": 0.2108284205198288,
      "num_tokens": 46650327.0,
      "step": 26895
    },
    {
      "entropy": 5.561326599121093,
      "epoch": 2.0929002139661543,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004557332510206953,
      "loss": 4.8701,
      "mean_token_accuracy": 0.20414306223392487,
      "num_tokens": 46659453.0,
      "step": 26900
    },
    {
      "entropy": 5.701846742630005,
      "epoch": 2.093289243337872,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004557167414128654,
      "loss": 4.9412,
      "mean_token_accuracy": 0.20632577389478685,
      "num_tokens": 46667378.0,
      "step": 26905
    },
    {
      "entropy": 5.714741182327271,
      "epoch": 2.0936782727095897,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004557002290629467,
      "loss": 5.109,
      "mean_token_accuracy": 0.1892647221684456,
      "num_tokens": 46675949.0,
      "step": 26910
    },
    {
      "entropy": 5.744656133651733,
      "epoch": 2.094067302081307,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045568371397118954,
      "loss": 5.1317,
      "mean_token_accuracy": 0.19753847122192383,
      "num_tokens": 46684808.0,
      "step": 26915
    },
    {
      "entropy": 5.69275598526001,
      "epoch": 2.0944563314530247,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004556671961378446,
      "loss": 4.9487,
      "mean_token_accuracy": 0.20452860593795777,
      "num_tokens": 46693209.0,
      "step": 26920
    },
    {
      "entropy": 5.59790301322937,
      "epoch": 2.0948453608247424,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045565067556316264,
      "loss": 4.9181,
      "mean_token_accuracy": 0.217241533100605,
      "num_tokens": 46702405.0,
      "step": 26925
    },
    {
      "entropy": 5.552121591567993,
      "epoch": 2.0952343901964596,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004556341522473941,
      "loss": 4.9297,
      "mean_token_accuracy": 0.20776440799236298,
      "num_tokens": 46711568.0,
      "step": 26930
    },
    {
      "entropy": 5.632434272766114,
      "epoch": 2.0956234195681773,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004556176261907899,
      "loss": 4.9633,
      "mean_token_accuracy": 0.2118593841791153,
      "num_tokens": 46719886.0,
      "step": 26935
    },
    {
      "entropy": 5.616064357757568,
      "epoch": 2.096012448939895,
      "grad_norm": 1.125,
      "learning_rate": 0.0004556010973936006,
      "loss": 4.8079,
      "mean_token_accuracy": 0.21635161340236664,
      "num_tokens": 46728293.0,
      "step": 26940
    },
    {
      "entropy": 5.673112630844116,
      "epoch": 2.0964014783116127,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004555845658560772,
      "loss": 4.9735,
      "mean_token_accuracy": 0.19880374968051912,
      "num_tokens": 46736730.0,
      "step": 26945
    },
    {
      "entropy": 5.6528278350830075,
      "epoch": 2.09679050768333,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045556803157847036,
      "loss": 5.014,
      "mean_token_accuracy": 0.1957776889204979,
      "num_tokens": 46745563.0,
      "step": 26950
    },
    {
      "entropy": 5.676190185546875,
      "epoch": 2.0971795370550477,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045555149456103113,
      "loss": 4.96,
      "mean_token_accuracy": 0.20189842134714125,
      "num_tokens": 46754285.0,
      "step": 26955
    },
    {
      "entropy": 5.652914571762085,
      "epoch": 2.0975685664267654,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004555349548040102,
      "loss": 4.9459,
      "mean_token_accuracy": 0.21316161006689072,
      "num_tokens": 46763114.0,
      "step": 26960
    },
    {
      "entropy": 5.676901197433471,
      "epoch": 2.0979575957984826,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004555184123076589,
      "loss": 4.9783,
      "mean_token_accuracy": 0.20448049455881118,
      "num_tokens": 46771150.0,
      "step": 26965
    },
    {
      "entropy": 5.596397399902344,
      "epoch": 2.0983466251702003,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045550186707222785,
      "loss": 4.8611,
      "mean_token_accuracy": 0.21743495315313338,
      "num_tokens": 46779955.0,
      "step": 26970
    },
    {
      "entropy": 5.640749979019165,
      "epoch": 2.098735654541918,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004554853190979683,
      "loss": 5.0001,
      "mean_token_accuracy": 0.19638883769512178,
      "num_tokens": 46788769.0,
      "step": 26975
    },
    {
      "entropy": 5.69490270614624,
      "epoch": 2.0991246839136353,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00045546876838513134,
      "loss": 4.9971,
      "mean_token_accuracy": 0.20097410976886748,
      "num_tokens": 46797649.0,
      "step": 26980
    },
    {
      "entropy": 5.629510688781738,
      "epoch": 2.099513713285353,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045545221493396805,
      "loss": 4.9127,
      "mean_token_accuracy": 0.20845791846513748,
      "num_tokens": 46806140.0,
      "step": 26985
    },
    {
      "entropy": 5.620686340332031,
      "epoch": 2.0999027426570707,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045543565874472963,
      "loss": 4.8695,
      "mean_token_accuracy": 0.21568753272295,
      "num_tokens": 46815476.0,
      "step": 26990
    },
    {
      "entropy": 5.618377637863159,
      "epoch": 2.1002917720287884,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045541909981766725,
      "loss": 5.0042,
      "mean_token_accuracy": 0.2029227763414383,
      "num_tokens": 46824171.0,
      "step": 26995
    },
    {
      "entropy": 5.60492787361145,
      "epoch": 2.1006808014005056,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004554025381530322,
      "loss": 4.8502,
      "mean_token_accuracy": 0.21458748877048492,
      "num_tokens": 46832474.0,
      "step": 27000
    },
    {
      "epoch": 2.1006808014005056,
      "eval_entropy": 5.399869043902782,
      "eval_loss": 5.073859214782715,
      "eval_mean_token_accuracy": 0.2090760173994951,
      "eval_num_tokens": 46832474.0,
      "eval_runtime": 21.7148,
      "eval_samples_per_second": 1913.806,
      "eval_steps_per_second": 239.237,
      "step": 27000
    },
    {
      "entropy": 5.613234233856201,
      "epoch": 2.1010698307722233,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004553859737510758,
      "loss": 4.829,
      "mean_token_accuracy": 0.21281650066375732,
      "num_tokens": 46840626.0,
      "step": 27005
    },
    {
      "entropy": 5.601217555999756,
      "epoch": 2.101458860143941,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004553694066120493,
      "loss": 4.8236,
      "mean_token_accuracy": 0.21641796082258224,
      "num_tokens": 46849626.0,
      "step": 27010
    },
    {
      "entropy": 5.683456754684448,
      "epoch": 2.1018478895156583,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045535283673620426,
      "loss": 4.9959,
      "mean_token_accuracy": 0.20372324883937837,
      "num_tokens": 46858520.0,
      "step": 27015
    },
    {
      "entropy": 5.642015790939331,
      "epoch": 2.102236918887376,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045533626412379194,
      "loss": 4.9631,
      "mean_token_accuracy": 0.20724524557590485,
      "num_tokens": 46866532.0,
      "step": 27020
    },
    {
      "entropy": 5.668253660202026,
      "epoch": 2.1026259482590937,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004553196887750638,
      "loss": 5.0141,
      "mean_token_accuracy": 0.20169663280248643,
      "num_tokens": 46875815.0,
      "step": 27025
    },
    {
      "entropy": 5.648970365524292,
      "epoch": 2.103014977630811,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004553031106902714,
      "loss": 4.9253,
      "mean_token_accuracy": 0.20133455246686935,
      "num_tokens": 46884708.0,
      "step": 27030
    },
    {
      "entropy": 5.649463701248169,
      "epoch": 2.1034040070025286,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045528652986966627,
      "loss": 4.8558,
      "mean_token_accuracy": 0.21517056375741958,
      "num_tokens": 46894089.0,
      "step": 27035
    },
    {
      "entropy": 5.731872415542602,
      "epoch": 2.1037930363742463,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045526994631350006,
      "loss": 5.041,
      "mean_token_accuracy": 0.20477527230978013,
      "num_tokens": 46903318.0,
      "step": 27040
    },
    {
      "entropy": 5.6384350776672365,
      "epoch": 2.104182065745964,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004552533600220243,
      "loss": 4.8713,
      "mean_token_accuracy": 0.21197218000888823,
      "num_tokens": 46912138.0,
      "step": 27045
    },
    {
      "entropy": 5.649442338943482,
      "epoch": 2.1045710951176813,
      "grad_norm": 1.390625,
      "learning_rate": 0.00045523677099549073,
      "loss": 4.9379,
      "mean_token_accuracy": 0.20431977212429048,
      "num_tokens": 46919544.0,
      "step": 27050
    },
    {
      "entropy": 5.639273262023925,
      "epoch": 2.104960124489399,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000455220179234151,
      "loss": 5.0359,
      "mean_token_accuracy": 0.20379054695367813,
      "num_tokens": 46928760.0,
      "step": 27055
    },
    {
      "entropy": 5.639343214035034,
      "epoch": 2.1053491538611167,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004552035847382569,
      "loss": 4.9636,
      "mean_token_accuracy": 0.2090733289718628,
      "num_tokens": 46936860.0,
      "step": 27060
    },
    {
      "entropy": 5.654239559173584,
      "epoch": 2.105738183232834,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045518698750806024,
      "loss": 4.9652,
      "mean_token_accuracy": 0.202738219499588,
      "num_tokens": 46945157.0,
      "step": 27065
    },
    {
      "entropy": 5.706164455413818,
      "epoch": 2.1061272126045516,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045517038754381287,
      "loss": 4.9604,
      "mean_token_accuracy": 0.2036540389060974,
      "num_tokens": 46954006.0,
      "step": 27070
    },
    {
      "entropy": 5.631262397766113,
      "epoch": 2.1065162419762693,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004551537848457666,
      "loss": 4.9162,
      "mean_token_accuracy": 0.2140851229429245,
      "num_tokens": 46962437.0,
      "step": 27075
    },
    {
      "entropy": 5.6672468185424805,
      "epoch": 2.106905271347987,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004551371794141734,
      "loss": 5.0271,
      "mean_token_accuracy": 0.20912189930677413,
      "num_tokens": 46972379.0,
      "step": 27080
    },
    {
      "entropy": 5.7228635311126705,
      "epoch": 2.1072943007197042,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004551205712492852,
      "loss": 4.9942,
      "mean_token_accuracy": 0.20503181219100952,
      "num_tokens": 46981410.0,
      "step": 27085
    },
    {
      "entropy": 5.638900947570801,
      "epoch": 2.107683330091422,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004551039603513541,
      "loss": 4.8627,
      "mean_token_accuracy": 0.2083207458257675,
      "num_tokens": 46989662.0,
      "step": 27090
    },
    {
      "entropy": 5.681561231613159,
      "epoch": 2.1080723594631396,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045508734672063204,
      "loss": 4.9969,
      "mean_token_accuracy": 0.20572306215763092,
      "num_tokens": 46998726.0,
      "step": 27095
    },
    {
      "entropy": 5.686280727386475,
      "epoch": 2.108461388834857,
      "grad_norm": 1.25,
      "learning_rate": 0.0004550707303573711,
      "loss": 5.0219,
      "mean_token_accuracy": 0.2005669504404068,
      "num_tokens": 47006980.0,
      "step": 27100
    },
    {
      "entropy": 5.648754119873047,
      "epoch": 2.1088504182065746,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045505411126182347,
      "loss": 4.9986,
      "mean_token_accuracy": 0.19914681613445281,
      "num_tokens": 47015693.0,
      "step": 27105
    },
    {
      "entropy": 5.659471607208252,
      "epoch": 2.1092394475782923,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004550374894342412,
      "loss": 4.9348,
      "mean_token_accuracy": 0.21607737988233566,
      "num_tokens": 47024256.0,
      "step": 27110
    },
    {
      "entropy": 5.7053979396820065,
      "epoch": 2.1096284769500095,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004550208648748767,
      "loss": 5.0252,
      "mean_token_accuracy": 0.19807718992233275,
      "num_tokens": 47032962.0,
      "step": 27115
    },
    {
      "entropy": 5.661650085449219,
      "epoch": 2.1100175063217272,
      "grad_norm": 1.1875,
      "learning_rate": 0.000455004237583982,
      "loss": 4.9811,
      "mean_token_accuracy": 0.20571606755256652,
      "num_tokens": 47041507.0,
      "step": 27120
    },
    {
      "entropy": 5.602028512954712,
      "epoch": 2.110406535693445,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004549876075618096,
      "loss": 4.9075,
      "mean_token_accuracy": 0.21057561337947844,
      "num_tokens": 47050088.0,
      "step": 27125
    },
    {
      "entropy": 5.715896797180176,
      "epoch": 2.110795565065162,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004549709748086117,
      "loss": 5.0468,
      "mean_token_accuracy": 0.2057209715247154,
      "num_tokens": 47058900.0,
      "step": 27130
    },
    {
      "entropy": 5.676351261138916,
      "epoch": 2.11118459443688,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045495433932464063,
      "loss": 4.9465,
      "mean_token_accuracy": 0.20517079085111617,
      "num_tokens": 47067689.0,
      "step": 27135
    },
    {
      "entropy": 5.639160203933716,
      "epoch": 2.1115736238085976,
      "grad_norm": 1.09375,
      "learning_rate": 0.000454937701110149,
      "loss": 4.9062,
      "mean_token_accuracy": 0.20840339809656144,
      "num_tokens": 47076309.0,
      "step": 27140
    },
    {
      "entropy": 5.649688577651977,
      "epoch": 2.1119626531803153,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045492106016538917,
      "loss": 4.9087,
      "mean_token_accuracy": 0.21411127150058745,
      "num_tokens": 47084462.0,
      "step": 27145
    },
    {
      "entropy": 5.637322282791137,
      "epoch": 2.1123516825520325,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045490441649061354,
      "loss": 5.0083,
      "mean_token_accuracy": 0.2038608968257904,
      "num_tokens": 47093586.0,
      "step": 27150
    },
    {
      "entropy": 5.656395959854126,
      "epoch": 2.1127407119237502,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004548877700860747,
      "loss": 4.9133,
      "mean_token_accuracy": 0.20982620418071746,
      "num_tokens": 47101296.0,
      "step": 27155
    },
    {
      "entropy": 5.690453815460205,
      "epoch": 2.113129741295468,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045487112095202544,
      "loss": 5.0018,
      "mean_token_accuracy": 0.20231125205755235,
      "num_tokens": 47110033.0,
      "step": 27160
    },
    {
      "entropy": 5.625979471206665,
      "epoch": 2.113518770667185,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004548544690887181,
      "loss": 5.0349,
      "mean_token_accuracy": 0.1993177905678749,
      "num_tokens": 47119292.0,
      "step": 27165
    },
    {
      "entropy": 5.683827829360962,
      "epoch": 2.113907800038903,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004548378144964054,
      "loss": 4.9959,
      "mean_token_accuracy": 0.2053084120154381,
      "num_tokens": 47128081.0,
      "step": 27170
    },
    {
      "entropy": 5.751273536682129,
      "epoch": 2.1142968294106206,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045482115717534024,
      "loss": 5.019,
      "mean_token_accuracy": 0.20371666848659514,
      "num_tokens": 47136639.0,
      "step": 27175
    },
    {
      "entropy": 5.685723972320557,
      "epoch": 2.1146858587823383,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004548044971257752,
      "loss": 5.0451,
      "mean_token_accuracy": 0.19916887134313582,
      "num_tokens": 47145830.0,
      "step": 27180
    },
    {
      "entropy": 5.630901575088501,
      "epoch": 2.1150748881540555,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045478783434796307,
      "loss": 4.994,
      "mean_token_accuracy": 0.20066393315792083,
      "num_tokens": 47154695.0,
      "step": 27185
    },
    {
      "entropy": 5.670967435836792,
      "epoch": 2.1154639175257732,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045477116884215675,
      "loss": 5.0037,
      "mean_token_accuracy": 0.2023532897233963,
      "num_tokens": 47163818.0,
      "step": 27190
    },
    {
      "entropy": 5.6202317237854,
      "epoch": 2.115852946897491,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000454754500608609,
      "loss": 4.8881,
      "mean_token_accuracy": 0.21655764132738115,
      "num_tokens": 47172426.0,
      "step": 27195
    },
    {
      "entropy": 5.70129976272583,
      "epoch": 2.116241976269208,
      "grad_norm": 1.125,
      "learning_rate": 0.0004547378296475729,
      "loss": 5.0823,
      "mean_token_accuracy": 0.19865068197250366,
      "num_tokens": 47182034.0,
      "step": 27200
    },
    {
      "entropy": 5.673218727111816,
      "epoch": 2.116631005640926,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045472115595930124,
      "loss": 4.9872,
      "mean_token_accuracy": 0.20043184459209443,
      "num_tokens": 47190968.0,
      "step": 27205
    },
    {
      "entropy": 5.602432823181152,
      "epoch": 2.1170200350126436,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045470447954404716,
      "loss": 4.8728,
      "mean_token_accuracy": 0.20573307424783707,
      "num_tokens": 47199917.0,
      "step": 27210
    },
    {
      "entropy": 5.663146257400513,
      "epoch": 2.117409064384361,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045468780040206364,
      "loss": 4.9728,
      "mean_token_accuracy": 0.20519523471593856,
      "num_tokens": 47208467.0,
      "step": 27215
    },
    {
      "entropy": 5.595748662948608,
      "epoch": 2.1177980937560785,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004546711185336037,
      "loss": 4.9661,
      "mean_token_accuracy": 0.2030106782913208,
      "num_tokens": 47216784.0,
      "step": 27220
    },
    {
      "entropy": 5.705761575698853,
      "epoch": 2.1181871231277962,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045465443393892047,
      "loss": 5.0002,
      "mean_token_accuracy": 0.20709338784217834,
      "num_tokens": 47226215.0,
      "step": 27225
    },
    {
      "entropy": 5.6469254970550535,
      "epoch": 2.118576152499514,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004546377466182673,
      "loss": 4.9401,
      "mean_token_accuracy": 0.21024397164583206,
      "num_tokens": 47234952.0,
      "step": 27230
    },
    {
      "entropy": 5.699376249313355,
      "epoch": 2.118965181871231,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004546210565718972,
      "loss": 5.0135,
      "mean_token_accuracy": 0.1958385244011879,
      "num_tokens": 47243897.0,
      "step": 27235
    },
    {
      "entropy": 5.651567602157593,
      "epoch": 2.119354211242949,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045460436380006337,
      "loss": 4.9455,
      "mean_token_accuracy": 0.20446202009916306,
      "num_tokens": 47252171.0,
      "step": 27240
    },
    {
      "entropy": 5.6806893825531,
      "epoch": 2.1197432406146666,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004545876683030192,
      "loss": 4.9284,
      "mean_token_accuracy": 0.21119514107704163,
      "num_tokens": 47260494.0,
      "step": 27245
    },
    {
      "entropy": 5.753199863433838,
      "epoch": 2.120132269986384,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004545709700810181,
      "loss": 5.1151,
      "mean_token_accuracy": 0.2012203425168991,
      "num_tokens": 47269485.0,
      "step": 27250
    },
    {
      "entropy": 5.542975282669067,
      "epoch": 2.1205212993581015,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004545542691343133,
      "loss": 4.7969,
      "mean_token_accuracy": 0.21646440774202347,
      "num_tokens": 47278275.0,
      "step": 27255
    },
    {
      "entropy": 5.6425775527954105,
      "epoch": 2.120910328729819,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004545375654631582,
      "loss": 4.9152,
      "mean_token_accuracy": 0.20735618472099304,
      "num_tokens": 47287267.0,
      "step": 27260
    },
    {
      "entropy": 5.671722316741944,
      "epoch": 2.1212993581015365,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045452085906780645,
      "loss": 5.0447,
      "mean_token_accuracy": 0.20067078471183777,
      "num_tokens": 47295785.0,
      "step": 27265
    },
    {
      "entropy": 5.697864198684693,
      "epoch": 2.121688387473254,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045450414994851123,
      "loss": 4.9814,
      "mean_token_accuracy": 0.20833688527345656,
      "num_tokens": 47303951.0,
      "step": 27270
    },
    {
      "entropy": 5.654378509521484,
      "epoch": 2.122077416844972,
      "grad_norm": 1.25,
      "learning_rate": 0.00045448743810552636,
      "loss": 4.9478,
      "mean_token_accuracy": 0.20670368671417236,
      "num_tokens": 47312226.0,
      "step": 27275
    },
    {
      "entropy": 5.635615015029908,
      "epoch": 2.1224664462166896,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045447072353910537,
      "loss": 5.0407,
      "mean_token_accuracy": 0.20630938708782195,
      "num_tokens": 47320657.0,
      "step": 27280
    },
    {
      "entropy": 5.613146257400513,
      "epoch": 2.122855475588407,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004544540062495016,
      "loss": 4.9496,
      "mean_token_accuracy": 0.21435826569795607,
      "num_tokens": 47329063.0,
      "step": 27285
    },
    {
      "entropy": 5.646435499191284,
      "epoch": 2.1232445049601245,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045443728623696905,
      "loss": 4.8993,
      "mean_token_accuracy": 0.20285990685224534,
      "num_tokens": 47338304.0,
      "step": 27290
    },
    {
      "entropy": 5.642085838317871,
      "epoch": 2.123633534331842,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045442056350176123,
      "loss": 5.0319,
      "mean_token_accuracy": 0.20336118340492249,
      "num_tokens": 47347248.0,
      "step": 27295
    },
    {
      "entropy": 5.624567031860352,
      "epoch": 2.1240225637035595,
      "grad_norm": 1.203125,
      "learning_rate": 0.000454403838044132,
      "loss": 5.0218,
      "mean_token_accuracy": 0.20095341205596923,
      "num_tokens": 47356586.0,
      "step": 27300
    },
    {
      "entropy": 5.689911031723023,
      "epoch": 2.124411593075277,
      "grad_norm": 1.1875,
      "learning_rate": 0.000454387109864335,
      "loss": 4.9093,
      "mean_token_accuracy": 0.20587568283081054,
      "num_tokens": 47365195.0,
      "step": 27305
    },
    {
      "entropy": 5.676011943817139,
      "epoch": 2.124800622446995,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045437037896262425,
      "loss": 5.1017,
      "mean_token_accuracy": 0.19270495027303697,
      "num_tokens": 47374154.0,
      "step": 27310
    },
    {
      "entropy": 5.692531871795654,
      "epoch": 2.125189651818712,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004543536453392534,
      "loss": 4.9056,
      "mean_token_accuracy": 0.21137844175100326,
      "num_tokens": 47383253.0,
      "step": 27315
    },
    {
      "entropy": 5.692821645736695,
      "epoch": 2.12557868119043,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045433690899447643,
      "loss": 4.9816,
      "mean_token_accuracy": 0.20413635671138763,
      "num_tokens": 47391387.0,
      "step": 27320
    },
    {
      "entropy": 5.683674049377442,
      "epoch": 2.1259677105621475,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045432016992854734,
      "loss": 5.0053,
      "mean_token_accuracy": 0.2035363271832466,
      "num_tokens": 47400294.0,
      "step": 27325
    },
    {
      "entropy": 5.674733114242554,
      "epoch": 2.126356739933865,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004543034281417201,
      "loss": 4.933,
      "mean_token_accuracy": 0.20517879873514175,
      "num_tokens": 47409024.0,
      "step": 27330
    },
    {
      "entropy": 5.673489046096802,
      "epoch": 2.1267457693055825,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004542866836342488,
      "loss": 4.9253,
      "mean_token_accuracy": 0.2108193814754486,
      "num_tokens": 47416974.0,
      "step": 27335
    },
    {
      "entropy": 5.658822679519654,
      "epoch": 2.1271347986773,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004542699364063873,
      "loss": 5.0385,
      "mean_token_accuracy": 0.20563113540410996,
      "num_tokens": 47425934.0,
      "step": 27340
    },
    {
      "entropy": 5.614219808578492,
      "epoch": 2.127523828049018,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004542531864583899,
      "loss": 5.0189,
      "mean_token_accuracy": 0.20435035824775696,
      "num_tokens": 47434764.0,
      "step": 27345
    },
    {
      "entropy": 5.741429138183594,
      "epoch": 2.127912857420735,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004542364337905108,
      "loss": 5.0423,
      "mean_token_accuracy": 0.19391341656446456,
      "num_tokens": 47443435.0,
      "step": 27350
    },
    {
      "entropy": 5.732747316360474,
      "epoch": 2.128301886792453,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045421967840300403,
      "loss": 5.091,
      "mean_token_accuracy": 0.1965980738401413,
      "num_tokens": 47452109.0,
      "step": 27355
    },
    {
      "entropy": 5.694426965713501,
      "epoch": 2.1286909161641705,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004542029202961239,
      "loss": 4.9209,
      "mean_token_accuracy": 0.208268141746521,
      "num_tokens": 47460319.0,
      "step": 27360
    },
    {
      "entropy": 5.721395111083984,
      "epoch": 2.1290799455358878,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045418615947012464,
      "loss": 5.0772,
      "mean_token_accuracy": 0.19648429453372956,
      "num_tokens": 47468806.0,
      "step": 27365
    },
    {
      "entropy": 5.643730688095093,
      "epoch": 2.1294689749076054,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004541693959252607,
      "loss": 5.0037,
      "mean_token_accuracy": 0.1998947575688362,
      "num_tokens": 47477121.0,
      "step": 27370
    },
    {
      "entropy": 5.534208536148071,
      "epoch": 2.129858004279323,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045415262966178635,
      "loss": 4.8308,
      "mean_token_accuracy": 0.21480167657136917,
      "num_tokens": 47485379.0,
      "step": 27375
    },
    {
      "entropy": 5.661724805831909,
      "epoch": 2.130247033651041,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004541358606799559,
      "loss": 5.0023,
      "mean_token_accuracy": 0.19952017217874526,
      "num_tokens": 47493402.0,
      "step": 27380
    },
    {
      "entropy": 5.716940402984619,
      "epoch": 2.130636063022758,
      "grad_norm": 1.4375,
      "learning_rate": 0.000454119088980024,
      "loss": 4.9785,
      "mean_token_accuracy": 0.2086973249912262,
      "num_tokens": 47503148.0,
      "step": 27385
    },
    {
      "entropy": 5.663738107681274,
      "epoch": 2.131025092394476,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004541023145622449,
      "loss": 4.9113,
      "mean_token_accuracy": 0.2086601138114929,
      "num_tokens": 47511681.0,
      "step": 27390
    },
    {
      "entropy": 5.608212900161743,
      "epoch": 2.1314141217661935,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045408553742687336,
      "loss": 4.9066,
      "mean_token_accuracy": 0.2117876023054123,
      "num_tokens": 47520240.0,
      "step": 27395
    },
    {
      "entropy": 5.6712775230407715,
      "epoch": 2.1318031511379107,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004540687575741638,
      "loss": 4.9584,
      "mean_token_accuracy": 0.20362551510334015,
      "num_tokens": 47528553.0,
      "step": 27400
    },
    {
      "entropy": 5.680309629440307,
      "epoch": 2.1321921805096284,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045405197500437077,
      "loss": 5.013,
      "mean_token_accuracy": 0.2037097156047821,
      "num_tokens": 47537735.0,
      "step": 27405
    },
    {
      "entropy": 5.633607578277588,
      "epoch": 2.132581209881346,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045403518971774915,
      "loss": 4.8842,
      "mean_token_accuracy": 0.20978721529245375,
      "num_tokens": 47545251.0,
      "step": 27410
    },
    {
      "entropy": 5.657035064697266,
      "epoch": 2.132970239253064,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045401840171455343,
      "loss": 4.9738,
      "mean_token_accuracy": 0.20088404715061187,
      "num_tokens": 47553962.0,
      "step": 27415
    },
    {
      "entropy": 5.632261323928833,
      "epoch": 2.133359268624781,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004540016109950384,
      "loss": 5.0203,
      "mean_token_accuracy": 0.2031080335378647,
      "num_tokens": 47562401.0,
      "step": 27420
    },
    {
      "entropy": 5.685746431350708,
      "epoch": 2.133748297996499,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004539848175594589,
      "loss": 4.9554,
      "mean_token_accuracy": 0.20848299711942672,
      "num_tokens": 47570097.0,
      "step": 27425
    },
    {
      "entropy": 5.68476710319519,
      "epoch": 2.1341373273682165,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004539680214080695,
      "loss": 5.0117,
      "mean_token_accuracy": 0.20040542334318162,
      "num_tokens": 47578705.0,
      "step": 27430
    },
    {
      "entropy": 5.6244128227233885,
      "epoch": 2.1345263567399337,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045395122254112536,
      "loss": 5.0028,
      "mean_token_accuracy": 0.19810067117214203,
      "num_tokens": 47588075.0,
      "step": 27435
    },
    {
      "entropy": 5.591316604614258,
      "epoch": 2.1349153861116514,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004539344209588812,
      "loss": 4.9234,
      "mean_token_accuracy": 0.20926258265972136,
      "num_tokens": 47596906.0,
      "step": 27440
    },
    {
      "entropy": 5.65273871421814,
      "epoch": 2.135304415483369,
      "grad_norm": 1.125,
      "learning_rate": 0.00045391761666159204,
      "loss": 4.9544,
      "mean_token_accuracy": 0.21180009543895723,
      "num_tokens": 47605427.0,
      "step": 27445
    },
    {
      "entropy": 5.739815330505371,
      "epoch": 2.1356934448550864,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045390080964951294,
      "loss": 5.1011,
      "mean_token_accuracy": 0.1974730223417282,
      "num_tokens": 47614249.0,
      "step": 27450
    },
    {
      "entropy": 5.619745302200317,
      "epoch": 2.136082474226804,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045388399992289864,
      "loss": 4.9088,
      "mean_token_accuracy": 0.20457417517900467,
      "num_tokens": 47623200.0,
      "step": 27455
    },
    {
      "entropy": 5.690948390960694,
      "epoch": 2.136471503598522,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045386718748200436,
      "loss": 5.0205,
      "mean_token_accuracy": 0.20281428694725037,
      "num_tokens": 47632392.0,
      "step": 27460
    },
    {
      "entropy": 5.756765413284302,
      "epoch": 2.136860532970239,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045385037232708534,
      "loss": 5.0867,
      "mean_token_accuracy": 0.20347670763731002,
      "num_tokens": 47641648.0,
      "step": 27465
    },
    {
      "entropy": 5.720460271835327,
      "epoch": 2.1372495623419567,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004538335544583964,
      "loss": 4.9609,
      "mean_token_accuracy": 0.21140029430389404,
      "num_tokens": 47651113.0,
      "step": 27470
    },
    {
      "entropy": 5.737417125701905,
      "epoch": 2.1376385917136744,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045381673387619306,
      "loss": 5.1064,
      "mean_token_accuracy": 0.19741086214780806,
      "num_tokens": 47659959.0,
      "step": 27475
    },
    {
      "entropy": 5.681628608703614,
      "epoch": 2.138027621085392,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045379991058073035,
      "loss": 5.0479,
      "mean_token_accuracy": 0.1997288718819618,
      "num_tokens": 47668738.0,
      "step": 27480
    },
    {
      "entropy": 5.606743335723877,
      "epoch": 2.1384166504571094,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004537830845722636,
      "loss": 4.902,
      "mean_token_accuracy": 0.2129257634282112,
      "num_tokens": 47677268.0,
      "step": 27485
    },
    {
      "entropy": 5.639599561691284,
      "epoch": 2.138805679828827,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004537662558510481,
      "loss": 4.8838,
      "mean_token_accuracy": 0.2137099862098694,
      "num_tokens": 47685339.0,
      "step": 27490
    },
    {
      "entropy": 5.664579582214356,
      "epoch": 2.1391947092005448,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045374942441733925,
      "loss": 4.9766,
      "mean_token_accuracy": 0.20466979295015336,
      "num_tokens": 47694729.0,
      "step": 27495
    },
    {
      "entropy": 5.658829975128174,
      "epoch": 2.139583738572262,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004537325902713923,
      "loss": 4.9465,
      "mean_token_accuracy": 0.20435376912355424,
      "num_tokens": 47703555.0,
      "step": 27500
    },
    {
      "entropy": 5.697295141220093,
      "epoch": 2.1399727679439797,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004537157534134629,
      "loss": 5.014,
      "mean_token_accuracy": 0.19566967338323593,
      "num_tokens": 47713115.0,
      "step": 27505
    },
    {
      "entropy": 5.666584300994873,
      "epoch": 2.1403617973156974,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045369891384380623,
      "loss": 4.9975,
      "mean_token_accuracy": 0.2091094046831131,
      "num_tokens": 47722263.0,
      "step": 27510
    },
    {
      "entropy": 5.654507780075074,
      "epoch": 2.140750826687415,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004536820715626781,
      "loss": 5.0034,
      "mean_token_accuracy": 0.20624600499868392,
      "num_tokens": 47730037.0,
      "step": 27515
    },
    {
      "entropy": 5.597562170028686,
      "epoch": 2.1411398560591324,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045366522657033385,
      "loss": 4.8962,
      "mean_token_accuracy": 0.2219584509730339,
      "num_tokens": 47738968.0,
      "step": 27520
    },
    {
      "entropy": 5.650251150131226,
      "epoch": 2.14152888543085,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045364837886702924,
      "loss": 4.9629,
      "mean_token_accuracy": 0.20619574189186096,
      "num_tokens": 47747515.0,
      "step": 27525
    },
    {
      "entropy": 5.64413423538208,
      "epoch": 2.1419179148025678,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004536315284530198,
      "loss": 4.9316,
      "mean_token_accuracy": 0.20930712074041366,
      "num_tokens": 47756440.0,
      "step": 27530
    },
    {
      "entropy": 5.652896738052368,
      "epoch": 2.142306944174285,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004536146753285612,
      "loss": 5.0063,
      "mean_token_accuracy": 0.19525465965270997,
      "num_tokens": 47765823.0,
      "step": 27535
    },
    {
      "entropy": 5.678119421005249,
      "epoch": 2.1426959735460027,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004535978194939093,
      "loss": 4.9597,
      "mean_token_accuracy": 0.20953722447156906,
      "num_tokens": 47775423.0,
      "step": 27540
    },
    {
      "entropy": 5.673740720748901,
      "epoch": 2.1430850029177204,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045358096094931966,
      "loss": 5.0231,
      "mean_token_accuracy": 0.1950572595000267,
      "num_tokens": 47784276.0,
      "step": 27545
    },
    {
      "entropy": 5.656785297393799,
      "epoch": 2.1434740322894377,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004535640996950482,
      "loss": 4.9166,
      "mean_token_accuracy": 0.21487697213888168,
      "num_tokens": 47792608.0,
      "step": 27550
    },
    {
      "entropy": 5.633726930618286,
      "epoch": 2.1438630616611554,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004535472357313507,
      "loss": 4.9946,
      "mean_token_accuracy": 0.20173333138227462,
      "num_tokens": 47802356.0,
      "step": 27555
    },
    {
      "entropy": 5.667555141448974,
      "epoch": 2.144252091032873,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045353036905848316,
      "loss": 4.9863,
      "mean_token_accuracy": 0.2040463000535965,
      "num_tokens": 47811307.0,
      "step": 27560
    },
    {
      "entropy": 5.770419645309448,
      "epoch": 2.1446411204045903,
      "grad_norm": 1.125,
      "learning_rate": 0.0004535134996767014,
      "loss": 5.0916,
      "mean_token_accuracy": 0.1956991210579872,
      "num_tokens": 47820572.0,
      "step": 27565
    },
    {
      "entropy": 5.715929126739502,
      "epoch": 2.145030149776308,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004534966275862614,
      "loss": 5.0483,
      "mean_token_accuracy": 0.19835762083530425,
      "num_tokens": 47829053.0,
      "step": 27570
    },
    {
      "entropy": 5.688897514343262,
      "epoch": 2.1454191791480257,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004534797527874192,
      "loss": 4.9617,
      "mean_token_accuracy": 0.20267821103334427,
      "num_tokens": 47838244.0,
      "step": 27575
    },
    {
      "entropy": 5.656722640991211,
      "epoch": 2.1458082085197434,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045346287528043083,
      "loss": 4.9067,
      "mean_token_accuracy": 0.21080998629331588,
      "num_tokens": 47846908.0,
      "step": 27580
    },
    {
      "entropy": 5.699781274795532,
      "epoch": 2.1461972378914607,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045344599506555244,
      "loss": 5.0566,
      "mean_token_accuracy": 0.2020363613963127,
      "num_tokens": 47855824.0,
      "step": 27585
    },
    {
      "entropy": 5.633361291885376,
      "epoch": 2.1465862672631784,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045342911214304,
      "loss": 4.9004,
      "mean_token_accuracy": 0.20326099842786788,
      "num_tokens": 47864641.0,
      "step": 27590
    },
    {
      "entropy": 5.697423648834229,
      "epoch": 2.146975296634896,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004534122265131498,
      "loss": 5.0722,
      "mean_token_accuracy": 0.19777345806360244,
      "num_tokens": 47873721.0,
      "step": 27595
    },
    {
      "entropy": 5.637767028808594,
      "epoch": 2.1473643260066133,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045339533817613807,
      "loss": 4.9002,
      "mean_token_accuracy": 0.21434686332941055,
      "num_tokens": 47882485.0,
      "step": 27600
    },
    {
      "entropy": 5.603484010696411,
      "epoch": 2.147753355378331,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045337844713226106,
      "loss": 4.8576,
      "mean_token_accuracy": 0.20903533101081848,
      "num_tokens": 47891574.0,
      "step": 27605
    },
    {
      "entropy": 5.597446870803833,
      "epoch": 2.1481423847500487,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045336155338177497,
      "loss": 4.8378,
      "mean_token_accuracy": 0.21507161110639572,
      "num_tokens": 47900123.0,
      "step": 27610
    },
    {
      "entropy": 5.648542594909668,
      "epoch": 2.1485314141217664,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004533446569249362,
      "loss": 4.9903,
      "mean_token_accuracy": 0.2088017076253891,
      "num_tokens": 47908342.0,
      "step": 27615
    },
    {
      "entropy": 5.640203237533569,
      "epoch": 2.1489204434934837,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045332775776200116,
      "loss": 4.9339,
      "mean_token_accuracy": 0.2093158781528473,
      "num_tokens": 47917365.0,
      "step": 27620
    },
    {
      "entropy": 5.622347402572632,
      "epoch": 2.1493094728652014,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045331085589322626,
      "loss": 4.9106,
      "mean_token_accuracy": 0.2084290489554405,
      "num_tokens": 47925839.0,
      "step": 27625
    },
    {
      "entropy": 5.609056377410889,
      "epoch": 2.149698502236919,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004532939513188679,
      "loss": 4.8599,
      "mean_token_accuracy": 0.21751902848482133,
      "num_tokens": 47934408.0,
      "step": 27630
    },
    {
      "entropy": 5.647518587112427,
      "epoch": 2.1500875316086363,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004532770440391826,
      "loss": 4.995,
      "mean_token_accuracy": 0.1958807498216629,
      "num_tokens": 47943486.0,
      "step": 27635
    },
    {
      "entropy": 5.672332239151001,
      "epoch": 2.150476560980354,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004532601340544269,
      "loss": 5.0219,
      "mean_token_accuracy": 0.20350443869829177,
      "num_tokens": 47952809.0,
      "step": 27640
    },
    {
      "entropy": 5.702562284469605,
      "epoch": 2.1508655903520717,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004532432213648574,
      "loss": 4.9209,
      "mean_token_accuracy": 0.21032371520996093,
      "num_tokens": 47961104.0,
      "step": 27645
    },
    {
      "entropy": 5.662651062011719,
      "epoch": 2.151254619723789,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045322630597073067,
      "loss": 4.9603,
      "mean_token_accuracy": 0.2003912776708603,
      "num_tokens": 47969442.0,
      "step": 27650
    },
    {
      "entropy": 5.548160886764526,
      "epoch": 2.1516436490955066,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045320938787230355,
      "loss": 4.8638,
      "mean_token_accuracy": 0.21644951701164244,
      "num_tokens": 47978703.0,
      "step": 27655
    },
    {
      "entropy": 5.682194328308105,
      "epoch": 2.1520326784672243,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045319246706983257,
      "loss": 5.0028,
      "mean_token_accuracy": 0.20092980414628983,
      "num_tokens": 47986886.0,
      "step": 27660
    },
    {
      "entropy": 5.674047803878784,
      "epoch": 2.152421707838942,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045317554356357446,
      "loss": 4.9017,
      "mean_token_accuracy": 0.20921812653541566,
      "num_tokens": 47995278.0,
      "step": 27665
    },
    {
      "entropy": 5.624450922012329,
      "epoch": 2.1528107372106593,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045315861735378615,
      "loss": 4.9536,
      "mean_token_accuracy": 0.20757301300764083,
      "num_tokens": 48004351.0,
      "step": 27670
    },
    {
      "entropy": 5.668675136566162,
      "epoch": 2.153199766582377,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045314168844072435,
      "loss": 4.9468,
      "mean_token_accuracy": 0.20282215178012847,
      "num_tokens": 48013346.0,
      "step": 27675
    },
    {
      "entropy": 5.6592741966247555,
      "epoch": 2.1535887959540947,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045312475682464604,
      "loss": 5.0367,
      "mean_token_accuracy": 0.20089065432548522,
      "num_tokens": 48023371.0,
      "step": 27680
    },
    {
      "entropy": 5.62676591873169,
      "epoch": 2.153977825325812,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045310782250580794,
      "loss": 4.9191,
      "mean_token_accuracy": 0.20598409026861192,
      "num_tokens": 48032095.0,
      "step": 27685
    },
    {
      "entropy": 5.647483682632446,
      "epoch": 2.1543668546975296,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004530908854844672,
      "loss": 4.889,
      "mean_token_accuracy": 0.2031468540430069,
      "num_tokens": 48040818.0,
      "step": 27690
    },
    {
      "entropy": 5.638616371154785,
      "epoch": 2.1547558840692473,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045307394576088076,
      "loss": 4.8917,
      "mean_token_accuracy": 0.21400319486856462,
      "num_tokens": 48049737.0,
      "step": 27695
    },
    {
      "entropy": 5.590885925292969,
      "epoch": 2.1551449134409646,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004530570033353056,
      "loss": 4.8465,
      "mean_token_accuracy": 0.21539207696914672,
      "num_tokens": 48057659.0,
      "step": 27700
    },
    {
      "entropy": 5.599354791641235,
      "epoch": 2.1555339428126823,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045304005820799874,
      "loss": 4.8229,
      "mean_token_accuracy": 0.2143756628036499,
      "num_tokens": 48066578.0,
      "step": 27705
    },
    {
      "entropy": 5.622882461547851,
      "epoch": 2.1559229721844,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004530231103792175,
      "loss": 5.013,
      "mean_token_accuracy": 0.202509406208992,
      "num_tokens": 48076051.0,
      "step": 27710
    },
    {
      "entropy": 5.69720344543457,
      "epoch": 2.1563120015561177,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004530061598492188,
      "loss": 5.0198,
      "mean_token_accuracy": 0.19917848706245422,
      "num_tokens": 48084722.0,
      "step": 27715
    },
    {
      "entropy": 5.640050315856934,
      "epoch": 2.156701030927835,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045298920661826004,
      "loss": 4.838,
      "mean_token_accuracy": 0.2134719729423523,
      "num_tokens": 48092890.0,
      "step": 27720
    },
    {
      "entropy": 5.652290153503418,
      "epoch": 2.1570900602995526,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004529722506865984,
      "loss": 4.9841,
      "mean_token_accuracy": 0.20191383957862855,
      "num_tokens": 48101146.0,
      "step": 27725
    },
    {
      "entropy": 5.67432050704956,
      "epoch": 2.1574790896712703,
      "grad_norm": 1.203125,
      "learning_rate": 0.000452955292054491,
      "loss": 4.9679,
      "mean_token_accuracy": 0.20345318019390107,
      "num_tokens": 48109974.0,
      "step": 27730
    },
    {
      "entropy": 5.723814868927002,
      "epoch": 2.1578681190429876,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045293833072219535,
      "loss": 5.0296,
      "mean_token_accuracy": 0.20078433454036712,
      "num_tokens": 48118325.0,
      "step": 27735
    },
    {
      "entropy": 5.724700307846069,
      "epoch": 2.1582571484147053,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045292136668996876,
      "loss": 5.0528,
      "mean_token_accuracy": 0.1994461327791214,
      "num_tokens": 48128255.0,
      "step": 27740
    },
    {
      "entropy": 5.67337007522583,
      "epoch": 2.158646177786423,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004529043999580686,
      "loss": 4.8912,
      "mean_token_accuracy": 0.21185566931962968,
      "num_tokens": 48136763.0,
      "step": 27745
    },
    {
      "entropy": 5.646401071548462,
      "epoch": 2.1590352071581402,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045288743052675234,
      "loss": 4.8824,
      "mean_token_accuracy": 0.2150391787290573,
      "num_tokens": 48146371.0,
      "step": 27750
    },
    {
      "entropy": 5.69774980545044,
      "epoch": 2.159424236529858,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045287045839627744,
      "loss": 4.9716,
      "mean_token_accuracy": 0.21376355439424516,
      "num_tokens": 48155180.0,
      "step": 27755
    },
    {
      "entropy": 5.610779333114624,
      "epoch": 2.1598132659015756,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045285348356690155,
      "loss": 4.9195,
      "mean_token_accuracy": 0.2071367844939232,
      "num_tokens": 48163728.0,
      "step": 27760
    },
    {
      "entropy": 5.607184267044067,
      "epoch": 2.1602022952732933,
      "grad_norm": 1.203125,
      "learning_rate": 0.000452836506038882,
      "loss": 4.9473,
      "mean_token_accuracy": 0.20093848258256913,
      "num_tokens": 48171844.0,
      "step": 27765
    },
    {
      "entropy": 5.691542768478394,
      "epoch": 2.1605913246450106,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045281952581247654,
      "loss": 5.028,
      "mean_token_accuracy": 0.20891666561365127,
      "num_tokens": 48180617.0,
      "step": 27770
    },
    {
      "entropy": 5.629547119140625,
      "epoch": 2.1609803540167283,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045280254288794286,
      "loss": 4.8956,
      "mean_token_accuracy": 0.20768706053495406,
      "num_tokens": 48188875.0,
      "step": 27775
    },
    {
      "entropy": 5.578186559677124,
      "epoch": 2.161369383388446,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045278555726553855,
      "loss": 4.8526,
      "mean_token_accuracy": 0.21173908859491347,
      "num_tokens": 48197280.0,
      "step": 27780
    },
    {
      "entropy": 5.682181453704834,
      "epoch": 2.1617584127601632,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045276856894552143,
      "loss": 5.0564,
      "mean_token_accuracy": 0.20132864713668824,
      "num_tokens": 48206907.0,
      "step": 27785
    },
    {
      "entropy": 5.624484395980835,
      "epoch": 2.162147442131881,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004527515779281492,
      "loss": 4.876,
      "mean_token_accuracy": 0.2076454132795334,
      "num_tokens": 48215334.0,
      "step": 27790
    },
    {
      "entropy": 5.6608904838562015,
      "epoch": 2.1625364715035986,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045273458421367976,
      "loss": 4.9821,
      "mean_token_accuracy": 0.20526863038539886,
      "num_tokens": 48224667.0,
      "step": 27795
    },
    {
      "entropy": 5.659607028961181,
      "epoch": 2.162925500875316,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004527175878023708,
      "loss": 5.0527,
      "mean_token_accuracy": 0.1994410902261734,
      "num_tokens": 48233237.0,
      "step": 27800
    },
    {
      "entropy": 5.677418422698975,
      "epoch": 2.1633145302470336,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004527005886944803,
      "loss": 4.9448,
      "mean_token_accuracy": 0.19903944581747054,
      "num_tokens": 48241048.0,
      "step": 27805
    },
    {
      "entropy": 5.735195255279541,
      "epoch": 2.1637035596187513,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045268358689026626,
      "loss": 5.0593,
      "mean_token_accuracy": 0.20043904781341554,
      "num_tokens": 48250013.0,
      "step": 27810
    },
    {
      "entropy": 5.717362594604492,
      "epoch": 2.164092588990469,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004526665823899866,
      "loss": 5.0239,
      "mean_token_accuracy": 0.20728191882371902,
      "num_tokens": 48258566.0,
      "step": 27815
    },
    {
      "entropy": 5.661785459518432,
      "epoch": 2.164481618362186,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045264957519389936,
      "loss": 4.9022,
      "mean_token_accuracy": 0.21449643075466157,
      "num_tokens": 48266843.0,
      "step": 27820
    },
    {
      "entropy": 5.5901538848876955,
      "epoch": 2.164870647733904,
      "grad_norm": 1.125,
      "learning_rate": 0.00045263256530226253,
      "loss": 4.8928,
      "mean_token_accuracy": 0.20730493813753129,
      "num_tokens": 48276138.0,
      "step": 27825
    },
    {
      "entropy": 5.6109185218811035,
      "epoch": 2.1652596771056216,
      "grad_norm": 1.25,
      "learning_rate": 0.0004526155527153343,
      "loss": 4.8255,
      "mean_token_accuracy": 0.21440553069114685,
      "num_tokens": 48284452.0,
      "step": 27830
    },
    {
      "entropy": 5.702252292633057,
      "epoch": 2.165648706477339,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004525985374333727,
      "loss": 4.9821,
      "mean_token_accuracy": 0.20239596217870712,
      "num_tokens": 48292871.0,
      "step": 27835
    },
    {
      "entropy": 5.6288128852844235,
      "epoch": 2.1660377358490566,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000452581519456636,
      "loss": 4.9706,
      "mean_token_accuracy": 0.2007668897509575,
      "num_tokens": 48300642.0,
      "step": 27840
    },
    {
      "entropy": 5.66637773513794,
      "epoch": 2.1664267652207743,
      "grad_norm": 1.125,
      "learning_rate": 0.00045256449878538243,
      "loss": 5.0477,
      "mean_token_accuracy": 0.19616979956626893,
      "num_tokens": 48309229.0,
      "step": 27845
    },
    {
      "entropy": 5.715106773376465,
      "epoch": 2.166815794592492,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045254747541987017,
      "loss": 4.9844,
      "mean_token_accuracy": 0.1951386421918869,
      "num_tokens": 48317474.0,
      "step": 27850
    },
    {
      "entropy": 5.718110704421997,
      "epoch": 2.167204823964209,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004525304493603576,
      "loss": 5.0368,
      "mean_token_accuracy": 0.2044723555445671,
      "num_tokens": 48325810.0,
      "step": 27855
    },
    {
      "entropy": 5.679185438156128,
      "epoch": 2.167593853335927,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045251342060710295,
      "loss": 4.9592,
      "mean_token_accuracy": 0.20560362040996552,
      "num_tokens": 48334348.0,
      "step": 27860
    },
    {
      "entropy": 5.602088403701782,
      "epoch": 2.1679828827076446,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004524963891603647,
      "loss": 4.8117,
      "mean_token_accuracy": 0.21733194589614868,
      "num_tokens": 48342565.0,
      "step": 27865
    },
    {
      "entropy": 5.631564044952393,
      "epoch": 2.168371912079362,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045247935502040136,
      "loss": 5.0906,
      "mean_token_accuracy": 0.19503789991140366,
      "num_tokens": 48351398.0,
      "step": 27870
    },
    {
      "entropy": 5.621806240081787,
      "epoch": 2.1687609414510796,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004524623181874712,
      "loss": 4.9621,
      "mean_token_accuracy": 0.19961805045604705,
      "num_tokens": 48360275.0,
      "step": 27875
    },
    {
      "entropy": 5.66812424659729,
      "epoch": 2.1691499708227973,
      "grad_norm": 1.25,
      "learning_rate": 0.0004524452786618329,
      "loss": 4.9188,
      "mean_token_accuracy": 0.20675500631332397,
      "num_tokens": 48369516.0,
      "step": 27880
    },
    {
      "entropy": 5.62058391571045,
      "epoch": 2.1695390001945145,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045242823644374484,
      "loss": 4.9306,
      "mean_token_accuracy": 0.20588523894548416,
      "num_tokens": 48378412.0,
      "step": 27885
    },
    {
      "entropy": 5.689672517776489,
      "epoch": 2.169928029566232,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004524111915334658,
      "loss": 5.0048,
      "mean_token_accuracy": 0.20229062139987947,
      "num_tokens": 48386566.0,
      "step": 27890
    },
    {
      "entropy": 5.738546991348267,
      "epoch": 2.17031705893795,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045239414393125424,
      "loss": 4.9784,
      "mean_token_accuracy": 0.19933017939329148,
      "num_tokens": 48394727.0,
      "step": 27895
    },
    {
      "entropy": 5.730101823806763,
      "epoch": 2.170706088309667,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004523770936373689,
      "loss": 5.0482,
      "mean_token_accuracy": 0.19867919534444808,
      "num_tokens": 48404440.0,
      "step": 27900
    },
    {
      "entropy": 5.6688902378082275,
      "epoch": 2.171095117681385,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004523600406520685,
      "loss": 4.9025,
      "mean_token_accuracy": 0.21221558302640914,
      "num_tokens": 48413441.0,
      "step": 27905
    },
    {
      "entropy": 5.701059484481812,
      "epoch": 2.1714841470531026,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004523429849756118,
      "loss": 5.06,
      "mean_token_accuracy": 0.19839730560779573,
      "num_tokens": 48422393.0,
      "step": 27910
    },
    {
      "entropy": 5.665910959243774,
      "epoch": 2.1718731764248203,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004523259266082576,
      "loss": 4.9783,
      "mean_token_accuracy": 0.21153788417577743,
      "num_tokens": 48431151.0,
      "step": 27915
    },
    {
      "entropy": 5.657994794845581,
      "epoch": 2.1722622057965375,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045230886555026455,
      "loss": 4.9231,
      "mean_token_accuracy": 0.21089061796665193,
      "num_tokens": 48439453.0,
      "step": 27920
    },
    {
      "entropy": 5.7094508647918705,
      "epoch": 2.172651235168255,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045229180180189176,
      "loss": 4.9968,
      "mean_token_accuracy": 0.20898679941892623,
      "num_tokens": 48448055.0,
      "step": 27925
    },
    {
      "entropy": 5.675336599349976,
      "epoch": 2.173040264539973,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045227473536339817,
      "loss": 4.8757,
      "mean_token_accuracy": 0.21619545072317123,
      "num_tokens": 48456015.0,
      "step": 27930
    },
    {
      "entropy": 5.637132692337036,
      "epoch": 2.17342929391169,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004522576662350425,
      "loss": 4.9725,
      "mean_token_accuracy": 0.2090619459748268,
      "num_tokens": 48464389.0,
      "step": 27935
    },
    {
      "entropy": 5.632385873794556,
      "epoch": 2.173818323283408,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000452240594417084,
      "loss": 4.9286,
      "mean_token_accuracy": 0.20619429349899293,
      "num_tokens": 48472403.0,
      "step": 27940
    },
    {
      "entropy": 5.582054090499878,
      "epoch": 2.1742073526551255,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004522235199097815,
      "loss": 4.9624,
      "mean_token_accuracy": 0.19890948832035066,
      "num_tokens": 48480879.0,
      "step": 27945
    },
    {
      "entropy": 5.617822551727295,
      "epoch": 2.1745963820268432,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004522064427133942,
      "loss": 4.9115,
      "mean_token_accuracy": 0.21241922676563263,
      "num_tokens": 48489722.0,
      "step": 27950
    },
    {
      "entropy": 5.632689809799194,
      "epoch": 2.1749854113985605,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045218936282818116,
      "loss": 4.9539,
      "mean_token_accuracy": 0.20353959649801254,
      "num_tokens": 48499023.0,
      "step": 27955
    },
    {
      "entropy": 5.727786731719971,
      "epoch": 2.175374440770278,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004521722802544016,
      "loss": 4.996,
      "mean_token_accuracy": 0.2084108904004097,
      "num_tokens": 48508352.0,
      "step": 27960
    },
    {
      "entropy": 5.641827774047852,
      "epoch": 2.175763470141996,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045215519499231465,
      "loss": 4.8584,
      "mean_token_accuracy": 0.21824876219034195,
      "num_tokens": 48517481.0,
      "step": 27965
    },
    {
      "entropy": 5.633270978927612,
      "epoch": 2.176152499513713,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045213810704217963,
      "loss": 4.9383,
      "mean_token_accuracy": 0.20597258806228638,
      "num_tokens": 48525777.0,
      "step": 27970
    },
    {
      "entropy": 5.560191440582275,
      "epoch": 2.176541528885431,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004521210164042557,
      "loss": 4.8605,
      "mean_token_accuracy": 0.20894976556301117,
      "num_tokens": 48534297.0,
      "step": 27975
    },
    {
      "entropy": 5.676234436035156,
      "epoch": 2.1769305582571485,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004521039230788024,
      "loss": 4.9762,
      "mean_token_accuracy": 0.20960656106472014,
      "num_tokens": 48543425.0,
      "step": 27980
    },
    {
      "entropy": 5.631540966033936,
      "epoch": 2.177319587628866,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004520868270660789,
      "loss": 4.9121,
      "mean_token_accuracy": 0.20616969764232634,
      "num_tokens": 48553114.0,
      "step": 27985
    },
    {
      "entropy": 5.78299913406372,
      "epoch": 2.1777086170005835,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004520697283663446,
      "loss": 5.0008,
      "mean_token_accuracy": 0.20264660120010375,
      "num_tokens": 48561409.0,
      "step": 27990
    },
    {
      "entropy": 5.681301403045654,
      "epoch": 2.178097646372301,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045205262697985897,
      "loss": 5.0024,
      "mean_token_accuracy": 0.19788658171892165,
      "num_tokens": 48569782.0,
      "step": 27995
    },
    {
      "entropy": 5.688127851486206,
      "epoch": 2.1784866757440184,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004520355229068816,
      "loss": 5.047,
      "mean_token_accuracy": 0.19718634188175202,
      "num_tokens": 48579283.0,
      "step": 28000
    },
    {
      "entropy": 5.717211532592773,
      "epoch": 2.178875705115736,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045201841614767203,
      "loss": 4.9844,
      "mean_token_accuracy": 0.204121033847332,
      "num_tokens": 48587649.0,
      "step": 28005
    },
    {
      "entropy": 5.741157054901123,
      "epoch": 2.179264734487454,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045200130670248974,
      "loss": 4.9767,
      "mean_token_accuracy": 0.20132287144660949,
      "num_tokens": 48596227.0,
      "step": 28010
    },
    {
      "entropy": 5.651079750061035,
      "epoch": 2.1796537638591715,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045198419457159424,
      "loss": 4.9191,
      "mean_token_accuracy": 0.19988436996936798,
      "num_tokens": 48605364.0,
      "step": 28015
    },
    {
      "entropy": 5.62879114151001,
      "epoch": 2.180042793230889,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004519670797552453,
      "loss": 5.0276,
      "mean_token_accuracy": 0.20301719158887863,
      "num_tokens": 48613794.0,
      "step": 28020
    },
    {
      "entropy": 5.697993564605713,
      "epoch": 2.1804318226026065,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045194996225370266,
      "loss": 4.9796,
      "mean_token_accuracy": 0.19910020530223846,
      "num_tokens": 48622381.0,
      "step": 28025
    },
    {
      "entropy": 5.770380163192749,
      "epoch": 2.180820851974324,
      "grad_norm": 1.25,
      "learning_rate": 0.00045193284206722595,
      "loss": 5.1211,
      "mean_token_accuracy": 0.1939905181527138,
      "num_tokens": 48631484.0,
      "step": 28030
    },
    {
      "entropy": 5.669340896606445,
      "epoch": 2.1812098813460414,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004519157191960749,
      "loss": 4.9968,
      "mean_token_accuracy": 0.20431322306394578,
      "num_tokens": 48641082.0,
      "step": 28035
    },
    {
      "entropy": 5.648237133026123,
      "epoch": 2.181598910717759,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004518985936405096,
      "loss": 4.9531,
      "mean_token_accuracy": 0.2037985071539879,
      "num_tokens": 48649974.0,
      "step": 28040
    },
    {
      "entropy": 5.7161674976348875,
      "epoch": 2.181987940089477,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004518814654007896,
      "loss": 4.9911,
      "mean_token_accuracy": 0.20701447129249573,
      "num_tokens": 48658633.0,
      "step": 28045
    },
    {
      "entropy": 5.640104913711548,
      "epoch": 2.1823769694611945,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045186433447717486,
      "loss": 4.9729,
      "mean_token_accuracy": 0.2011757642030716,
      "num_tokens": 48667313.0,
      "step": 28050
    },
    {
      "entropy": 5.697940158843994,
      "epoch": 2.182765998832912,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045184720086992536,
      "loss": 5.0361,
      "mean_token_accuracy": 0.20083847343921662,
      "num_tokens": 48676664.0,
      "step": 28055
    },
    {
      "entropy": 5.652869367599488,
      "epoch": 2.1831550282046295,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045183006457930107,
      "loss": 4.9083,
      "mean_token_accuracy": 0.2108423411846161,
      "num_tokens": 48684770.0,
      "step": 28060
    },
    {
      "entropy": 5.613597679138183,
      "epoch": 2.183544057576347,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004518129256055621,
      "loss": 4.8955,
      "mean_token_accuracy": 0.21060262471437455,
      "num_tokens": 48692965.0,
      "step": 28065
    },
    {
      "entropy": 5.704329252243042,
      "epoch": 2.1839330869480644,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004517957839489683,
      "loss": 5.0225,
      "mean_token_accuracy": 0.19920146018266677,
      "num_tokens": 48702273.0,
      "step": 28070
    },
    {
      "entropy": 5.6070469379425045,
      "epoch": 2.184322116319782,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045177863960977986,
      "loss": 4.9584,
      "mean_token_accuracy": 0.19512238055467607,
      "num_tokens": 48710884.0,
      "step": 28075
    },
    {
      "entropy": 5.7953338623046875,
      "epoch": 2.1847111456915,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000451761492588257,
      "loss": 5.1002,
      "mean_token_accuracy": 0.1897400513291359,
      "num_tokens": 48719673.0,
      "step": 28080
    },
    {
      "entropy": 5.663338708877563,
      "epoch": 2.185100175063217,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045174434288465986,
      "loss": 4.962,
      "mean_token_accuracy": 0.20649539530277253,
      "num_tokens": 48728084.0,
      "step": 28085
    },
    {
      "entropy": 5.715069150924682,
      "epoch": 2.1854892044349348,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004517271904992486,
      "loss": 5.0468,
      "mean_token_accuracy": 0.1969381630420685,
      "num_tokens": 48736621.0,
      "step": 28090
    },
    {
      "entropy": 5.7222432613372805,
      "epoch": 2.1858782338066525,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004517100354322835,
      "loss": 4.9812,
      "mean_token_accuracy": 0.20841848850250244,
      "num_tokens": 48744990.0,
      "step": 28095
    },
    {
      "entropy": 5.663848972320556,
      "epoch": 2.18626726317837,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045169287768402497,
      "loss": 5.0022,
      "mean_token_accuracy": 0.19363734275102615,
      "num_tokens": 48753640.0,
      "step": 28100
    },
    {
      "entropy": 5.567945766448974,
      "epoch": 2.1866562925500874,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004516757172547331,
      "loss": 4.8959,
      "mean_token_accuracy": 0.21001117676496506,
      "num_tokens": 48762836.0,
      "step": 28105
    },
    {
      "entropy": 5.688671970367432,
      "epoch": 2.187045321921805,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004516585541446685,
      "loss": 4.9862,
      "mean_token_accuracy": 0.20826555490493776,
      "num_tokens": 48771158.0,
      "step": 28110
    },
    {
      "entropy": 5.685334920883179,
      "epoch": 2.187434351293523,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045164138835409153,
      "loss": 5.0438,
      "mean_token_accuracy": 0.19650269150733948,
      "num_tokens": 48779631.0,
      "step": 28115
    },
    {
      "entropy": 5.693322944641113,
      "epoch": 2.18782338066524,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045162421988326266,
      "loss": 5.0196,
      "mean_token_accuracy": 0.19955471456050872,
      "num_tokens": 48788584.0,
      "step": 28120
    },
    {
      "entropy": 5.672060871124268,
      "epoch": 2.1882124100369578,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004516070487324424,
      "loss": 4.9213,
      "mean_token_accuracy": 0.20327730476856232,
      "num_tokens": 48797403.0,
      "step": 28125
    },
    {
      "entropy": 5.65683650970459,
      "epoch": 2.1886014394086755,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045158987490189124,
      "loss": 4.9708,
      "mean_token_accuracy": 0.21074854731559753,
      "num_tokens": 48805448.0,
      "step": 28130
    },
    {
      "entropy": 5.698904085159302,
      "epoch": 2.1889904687803927,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004515726983918697,
      "loss": 5.0326,
      "mean_token_accuracy": 0.20094090700149536,
      "num_tokens": 48814983.0,
      "step": 28135
    },
    {
      "entropy": 5.705971050262451,
      "epoch": 2.1893794981521104,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045155551920263864,
      "loss": 4.9858,
      "mean_token_accuracy": 0.20424688458442689,
      "num_tokens": 48823955.0,
      "step": 28140
    },
    {
      "entropy": 5.682925605773926,
      "epoch": 2.189768527523828,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045153833733445857,
      "loss": 5.018,
      "mean_token_accuracy": 0.2027760848402977,
      "num_tokens": 48832078.0,
      "step": 28145
    },
    {
      "entropy": 5.708161020278931,
      "epoch": 2.190157556895546,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004515211527875903,
      "loss": 4.9405,
      "mean_token_accuracy": 0.20755576640367507,
      "num_tokens": 48840351.0,
      "step": 28150
    },
    {
      "entropy": 5.642427778244018,
      "epoch": 2.190546586267263,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004515039655622943,
      "loss": 4.9211,
      "mean_token_accuracy": 0.21980521827936172,
      "num_tokens": 48848333.0,
      "step": 28155
    },
    {
      "entropy": 5.645604228973388,
      "epoch": 2.1909356156389808,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045148677565883167,
      "loss": 4.97,
      "mean_token_accuracy": 0.1988986998796463,
      "num_tokens": 48856806.0,
      "step": 28160
    },
    {
      "entropy": 5.721673107147216,
      "epoch": 2.1913246450106985,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004514695830774631,
      "loss": 5.0232,
      "mean_token_accuracy": 0.2059039905667305,
      "num_tokens": 48865848.0,
      "step": 28165
    },
    {
      "entropy": 5.728046560287476,
      "epoch": 2.1917136743824157,
      "grad_norm": 1.25,
      "learning_rate": 0.00045145238781844953,
      "loss": 5.0395,
      "mean_token_accuracy": 0.1950358495116234,
      "num_tokens": 48875899.0,
      "step": 28170
    },
    {
      "entropy": 5.669016027450562,
      "epoch": 2.1921027037541334,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045143518988205184,
      "loss": 4.9186,
      "mean_token_accuracy": 0.2089315429329872,
      "num_tokens": 48884026.0,
      "step": 28175
    },
    {
      "entropy": 5.669518327713012,
      "epoch": 2.192491733125851,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045141798926853094,
      "loss": 4.9638,
      "mean_token_accuracy": 0.20037296265363694,
      "num_tokens": 48892640.0,
      "step": 28180
    },
    {
      "entropy": 5.743628025054932,
      "epoch": 2.1928807624975684,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004514007859781478,
      "loss": 5.0314,
      "mean_token_accuracy": 0.2032069444656372,
      "num_tokens": 48901071.0,
      "step": 28185
    },
    {
      "entropy": 5.737603664398193,
      "epoch": 2.193269791869286,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004513835800111635,
      "loss": 4.896,
      "mean_token_accuracy": 0.21216563135385513,
      "num_tokens": 48909034.0,
      "step": 28190
    },
    {
      "entropy": 5.7160662651062015,
      "epoch": 2.1936588212410038,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004513663713678392,
      "loss": 5.075,
      "mean_token_accuracy": 0.19648731052875518,
      "num_tokens": 48918399.0,
      "step": 28195
    },
    {
      "entropy": 5.687309265136719,
      "epoch": 2.1940478506127215,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045134916004843596,
      "loss": 4.9684,
      "mean_token_accuracy": 0.2019705593585968,
      "num_tokens": 48926767.0,
      "step": 28200
    },
    {
      "entropy": 5.662343549728393,
      "epoch": 2.1944368799844387,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004513319460532148,
      "loss": 4.93,
      "mean_token_accuracy": 0.20594674199819565,
      "num_tokens": 48935666.0,
      "step": 28205
    },
    {
      "entropy": 5.671147775650025,
      "epoch": 2.1948259093561564,
      "grad_norm": 1.125,
      "learning_rate": 0.00045131472938243707,
      "loss": 4.983,
      "mean_token_accuracy": 0.20340932607650758,
      "num_tokens": 48944461.0,
      "step": 28210
    },
    {
      "entropy": 5.639473819732666,
      "epoch": 2.195214938727874,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004512975100363639,
      "loss": 4.9123,
      "mean_token_accuracy": 0.20613808184862137,
      "num_tokens": 48952368.0,
      "step": 28215
    },
    {
      "entropy": 5.534367036819458,
      "epoch": 2.1956039680995914,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045128028801525675,
      "loss": 4.7918,
      "mean_token_accuracy": 0.21258027255535125,
      "num_tokens": 48961866.0,
      "step": 28220
    },
    {
      "entropy": 5.647168779373169,
      "epoch": 2.195992997471309,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045126306331937674,
      "loss": 4.9547,
      "mean_token_accuracy": 0.20967807173728942,
      "num_tokens": 48970401.0,
      "step": 28225
    },
    {
      "entropy": 5.705751562118531,
      "epoch": 2.1963820268430267,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004512458359489853,
      "loss": 5.0071,
      "mean_token_accuracy": 0.19936636686325074,
      "num_tokens": 48979165.0,
      "step": 28230
    },
    {
      "entropy": 5.645679092407226,
      "epoch": 2.196771056214744,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004512286059043437,
      "loss": 4.8996,
      "mean_token_accuracy": 0.21581383496522905,
      "num_tokens": 48987700.0,
      "step": 28235
    },
    {
      "entropy": 5.639755916595459,
      "epoch": 2.1971600855864617,
      "grad_norm": 1.125,
      "learning_rate": 0.00045121137318571364,
      "loss": 4.971,
      "mean_token_accuracy": 0.20731021761894225,
      "num_tokens": 48996901.0,
      "step": 28240
    },
    {
      "entropy": 5.697044563293457,
      "epoch": 2.1975491149581794,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045119413779335645,
      "loss": 5.0754,
      "mean_token_accuracy": 0.19524544030427932,
      "num_tokens": 49006049.0,
      "step": 28245
    },
    {
      "entropy": 5.744020223617554,
      "epoch": 2.197938144329897,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004511768997275336,
      "loss": 4.9958,
      "mean_token_accuracy": 0.20554505288600922,
      "num_tokens": 49015092.0,
      "step": 28250
    },
    {
      "entropy": 5.645731973648071,
      "epoch": 2.1983271737016143,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004511596589885068,
      "loss": 4.9703,
      "mean_token_accuracy": 0.20340969115495683,
      "num_tokens": 49023388.0,
      "step": 28255
    },
    {
      "entropy": 5.61659870147705,
      "epoch": 2.198716203073332,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004511424155765375,
      "loss": 4.9092,
      "mean_token_accuracy": 0.21286104172468184,
      "num_tokens": 49031931.0,
      "step": 28260
    },
    {
      "entropy": 5.709329080581665,
      "epoch": 2.1991052324450497,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045112516949188735,
      "loss": 5.0597,
      "mean_token_accuracy": 0.20465295463800431,
      "num_tokens": 49040584.0,
      "step": 28265
    },
    {
      "entropy": 5.717591428756714,
      "epoch": 2.199494261816767,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004511079207348182,
      "loss": 4.9785,
      "mean_token_accuracy": 0.20859259665012359,
      "num_tokens": 49049107.0,
      "step": 28270
    },
    {
      "entropy": 5.660569572448731,
      "epoch": 2.1998832911884847,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004510906693055916,
      "loss": 4.9375,
      "mean_token_accuracy": 0.20482848733663558,
      "num_tokens": 49058038.0,
      "step": 28275
    },
    {
      "entropy": 5.633388519287109,
      "epoch": 2.2002723205602024,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004510734152044694,
      "loss": 4.9308,
      "mean_token_accuracy": 0.20798454880714418,
      "num_tokens": 49066875.0,
      "step": 28280
    },
    {
      "entropy": 5.651957702636719,
      "epoch": 2.20066134993192,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045105615843171336,
      "loss": 5.004,
      "mean_token_accuracy": 0.19870477616786958,
      "num_tokens": 49076374.0,
      "step": 28285
    },
    {
      "entropy": 5.683029699325561,
      "epoch": 2.2010503793036373,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045103889898758534,
      "loss": 4.9582,
      "mean_token_accuracy": 0.19961364716291427,
      "num_tokens": 49085729.0,
      "step": 28290
    },
    {
      "entropy": 5.64361801147461,
      "epoch": 2.201439408675355,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004510216368723472,
      "loss": 4.8903,
      "mean_token_accuracy": 0.21082836091518403,
      "num_tokens": 49093959.0,
      "step": 28295
    },
    {
      "entropy": 5.649766826629639,
      "epoch": 2.2018284380470727,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045100437208626086,
      "loss": 4.9733,
      "mean_token_accuracy": 0.20577706545591354,
      "num_tokens": 49102412.0,
      "step": 28300
    },
    {
      "entropy": 5.584586763381958,
      "epoch": 2.20221746741879,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045098710462958844,
      "loss": 4.9457,
      "mean_token_accuracy": 0.20917852222919464,
      "num_tokens": 49110655.0,
      "step": 28305
    },
    {
      "entropy": 5.661291885375976,
      "epoch": 2.2026064967905077,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045096983450259166,
      "loss": 4.9785,
      "mean_token_accuracy": 0.20109608471393586,
      "num_tokens": 49119117.0,
      "step": 28310
    },
    {
      "entropy": 5.629784059524536,
      "epoch": 2.2029955261622254,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045095256170553285,
      "loss": 4.9058,
      "mean_token_accuracy": 0.2053326651453972,
      "num_tokens": 49127513.0,
      "step": 28315
    },
    {
      "entropy": 5.670602703094483,
      "epoch": 2.2033845555339426,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045093528623867384,
      "loss": 4.9236,
      "mean_token_accuracy": 0.20540666431188584,
      "num_tokens": 49135851.0,
      "step": 28320
    },
    {
      "entropy": 5.676714134216309,
      "epoch": 2.2037735849056603,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045091800810227706,
      "loss": 5.0332,
      "mean_token_accuracy": 0.20141338557004929,
      "num_tokens": 49144775.0,
      "step": 28325
    },
    {
      "entropy": 5.709808254241944,
      "epoch": 2.204162614277378,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004509007272966044,
      "loss": 4.9705,
      "mean_token_accuracy": 0.20860728323459626,
      "num_tokens": 49152861.0,
      "step": 28330
    },
    {
      "entropy": 5.682247591018677,
      "epoch": 2.2045516436490953,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004508834438219182,
      "loss": 5.0457,
      "mean_token_accuracy": 0.19494961947202682,
      "num_tokens": 49161875.0,
      "step": 28335
    },
    {
      "entropy": 5.740041732788086,
      "epoch": 2.204940673020813,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004508661576784807,
      "loss": 5.0687,
      "mean_token_accuracy": 0.202705118060112,
      "num_tokens": 49170277.0,
      "step": 28340
    },
    {
      "entropy": 5.6499974727630615,
      "epoch": 2.2053297023925307,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004508488688665541,
      "loss": 4.863,
      "mean_token_accuracy": 0.21453144699335097,
      "num_tokens": 49179078.0,
      "step": 28345
    },
    {
      "entropy": 5.660826683044434,
      "epoch": 2.2057187317642484,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004508315773864009,
      "loss": 4.9445,
      "mean_token_accuracy": 0.19795700758695603,
      "num_tokens": 49187649.0,
      "step": 28350
    },
    {
      "entropy": 5.635974740982055,
      "epoch": 2.2061077611359656,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045081428323828337,
      "loss": 4.8942,
      "mean_token_accuracy": 0.22162368446588515,
      "num_tokens": 49195369.0,
      "step": 28355
    },
    {
      "entropy": 5.700757741928101,
      "epoch": 2.2064967905076833,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004507969864224639,
      "loss": 5.0279,
      "mean_token_accuracy": 0.2009781926870346,
      "num_tokens": 49203623.0,
      "step": 28360
    },
    {
      "entropy": 5.680846357345581,
      "epoch": 2.206885819879401,
      "grad_norm": 1.34375,
      "learning_rate": 0.000450779686939205,
      "loss": 4.9848,
      "mean_token_accuracy": 0.2047211930155754,
      "num_tokens": 49212251.0,
      "step": 28365
    },
    {
      "entropy": 5.668113470077515,
      "epoch": 2.2072748492511183,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045076238478876916,
      "loss": 4.9439,
      "mean_token_accuracy": 0.20206397622823716,
      "num_tokens": 49220580.0,
      "step": 28370
    },
    {
      "entropy": 5.694327068328858,
      "epoch": 2.207663878622836,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004507450799714188,
      "loss": 4.9701,
      "mean_token_accuracy": 0.20728468149900436,
      "num_tokens": 49229011.0,
      "step": 28375
    },
    {
      "entropy": 5.664829206466675,
      "epoch": 2.2080529079945537,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045072777248741664,
      "loss": 4.9526,
      "mean_token_accuracy": 0.20984037816524506,
      "num_tokens": 49237082.0,
      "step": 28380
    },
    {
      "entropy": 5.7288237571716305,
      "epoch": 2.2084419373662714,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004507104623370252,
      "loss": 5.0701,
      "mean_token_accuracy": 0.1952718749642372,
      "num_tokens": 49245768.0,
      "step": 28385
    },
    {
      "entropy": 5.671775484085083,
      "epoch": 2.2088309667379886,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004506931495205072,
      "loss": 4.9857,
      "mean_token_accuracy": 0.20744284093379975,
      "num_tokens": 49254932.0,
      "step": 28390
    },
    {
      "entropy": 5.635376214981079,
      "epoch": 2.2092199961097063,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004506758340381253,
      "loss": 4.9724,
      "mean_token_accuracy": 0.19883600920438765,
      "num_tokens": 49263798.0,
      "step": 28395
    },
    {
      "entropy": 5.636352014541626,
      "epoch": 2.209609025481424,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004506585158901422,
      "loss": 4.8963,
      "mean_token_accuracy": 0.20576135218143463,
      "num_tokens": 49272045.0,
      "step": 28400
    },
    {
      "entropy": 5.727859973907471,
      "epoch": 2.2099980548531413,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004506411950768207,
      "loss": 5.0821,
      "mean_token_accuracy": 0.19559378176927567,
      "num_tokens": 49281562.0,
      "step": 28405
    },
    {
      "entropy": 5.630318021774292,
      "epoch": 2.210387084224859,
      "grad_norm": 1.375,
      "learning_rate": 0.00045062387159842363,
      "loss": 4.8581,
      "mean_token_accuracy": 0.21722469925880433,
      "num_tokens": 49290113.0,
      "step": 28410
    },
    {
      "entropy": 5.664670848846436,
      "epoch": 2.2107761135965767,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045060654545521386,
      "loss": 4.969,
      "mean_token_accuracy": 0.2101212978363037,
      "num_tokens": 49298210.0,
      "step": 28415
    },
    {
      "entropy": 5.700596284866333,
      "epoch": 2.211165142968294,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045058921664745436,
      "loss": 4.981,
      "mean_token_accuracy": 0.2012764036655426,
      "num_tokens": 49306533.0,
      "step": 28420
    },
    {
      "entropy": 5.669741296768189,
      "epoch": 2.2115541723400116,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045057188517540783,
      "loss": 5.0396,
      "mean_token_accuracy": 0.20619214475154876,
      "num_tokens": 49315617.0,
      "step": 28425
    },
    {
      "entropy": 5.730917882919312,
      "epoch": 2.2119432017117293,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004505545510393374,
      "loss": 4.95,
      "mean_token_accuracy": 0.20605897903442383,
      "num_tokens": 49324351.0,
      "step": 28430
    },
    {
      "entropy": 5.657011842727661,
      "epoch": 2.2123322310834466,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045053721423950617,
      "loss": 4.9349,
      "mean_token_accuracy": 0.21191661059856415,
      "num_tokens": 49333036.0,
      "step": 28435
    },
    {
      "entropy": 5.631449556350708,
      "epoch": 2.2127212604551643,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004505198747761771,
      "loss": 4.923,
      "mean_token_accuracy": 0.20679266154766082,
      "num_tokens": 49341715.0,
      "step": 28440
    },
    {
      "entropy": 5.690121746063232,
      "epoch": 2.213110289826882,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004505025326496132,
      "loss": 5.0022,
      "mean_token_accuracy": 0.2016412600874901,
      "num_tokens": 49350505.0,
      "step": 28445
    },
    {
      "entropy": 5.693645048141479,
      "epoch": 2.2134993191985997,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045048518786007784,
      "loss": 5.0023,
      "mean_token_accuracy": 0.2014643132686615,
      "num_tokens": 49358902.0,
      "step": 28450
    },
    {
      "entropy": 5.720460653305054,
      "epoch": 2.213888348570317,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045046784040783395,
      "loss": 5.0354,
      "mean_token_accuracy": 0.19743326008319856,
      "num_tokens": 49366932.0,
      "step": 28455
    },
    {
      "entropy": 5.763073253631592,
      "epoch": 2.2142773779420346,
      "grad_norm": 1.15625,
      "learning_rate": 0.000450450490293145,
      "loss": 5.1161,
      "mean_token_accuracy": 0.19040509760379792,
      "num_tokens": 49375907.0,
      "step": 28460
    },
    {
      "entropy": 5.749464416503907,
      "epoch": 2.2146664073137523,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000450433137516274,
      "loss": 4.9919,
      "mean_token_accuracy": 0.20192063748836517,
      "num_tokens": 49383871.0,
      "step": 28465
    },
    {
      "entropy": 5.676358461380005,
      "epoch": 2.2150554366854696,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004504157820774844,
      "loss": 4.9919,
      "mean_token_accuracy": 0.20251664221286775,
      "num_tokens": 49392452.0,
      "step": 28470
    },
    {
      "entropy": 5.611610555648804,
      "epoch": 2.2154444660571873,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004503984239770396,
      "loss": 4.8416,
      "mean_token_accuracy": 0.21371807903051376,
      "num_tokens": 49401698.0,
      "step": 28475
    },
    {
      "entropy": 5.708220529556274,
      "epoch": 2.215833495428905,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004503810632152028,
      "loss": 5.0021,
      "mean_token_accuracy": 0.20487595945596696,
      "num_tokens": 49409505.0,
      "step": 28480
    },
    {
      "entropy": 5.694054126739502,
      "epoch": 2.2162225248006227,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004503636997922375,
      "loss": 4.9893,
      "mean_token_accuracy": 0.20739823430776597,
      "num_tokens": 49418489.0,
      "step": 28485
    },
    {
      "entropy": 5.711539459228516,
      "epoch": 2.21661155417234,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004503463337084072,
      "loss": 4.9449,
      "mean_token_accuracy": 0.21111829578876495,
      "num_tokens": 49426989.0,
      "step": 28490
    },
    {
      "entropy": 5.696656656265259,
      "epoch": 2.2170005835440576,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004503289649639754,
      "loss": 5.0033,
      "mean_token_accuracy": 0.20572878420352936,
      "num_tokens": 49436199.0,
      "step": 28495
    },
    {
      "entropy": 5.646987342834473,
      "epoch": 2.2173896129157753,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004503115935592056,
      "loss": 4.9088,
      "mean_token_accuracy": 0.2111718624830246,
      "num_tokens": 49444092.0,
      "step": 28500
    },
    {
      "entropy": 5.643121671676636,
      "epoch": 2.2177786422874926,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004502942194943614,
      "loss": 4.9959,
      "mean_token_accuracy": 0.20761531889438628,
      "num_tokens": 49452461.0,
      "step": 28505
    },
    {
      "entropy": 5.719602727890015,
      "epoch": 2.2181676716592102,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045027684276970645,
      "loss": 4.946,
      "mean_token_accuracy": 0.20923102647066116,
      "num_tokens": 49461972.0,
      "step": 28510
    },
    {
      "entropy": 5.748099946975708,
      "epoch": 2.218556701030928,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045025946338550434,
      "loss": 4.9538,
      "mean_token_accuracy": 0.2064840763807297,
      "num_tokens": 49470656.0,
      "step": 28515
    },
    {
      "entropy": 5.660231590270996,
      "epoch": 2.218945730402645,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004502420813420189,
      "loss": 4.9939,
      "mean_token_accuracy": 0.202444688975811,
      "num_tokens": 49479018.0,
      "step": 28520
    },
    {
      "entropy": 5.691646242141724,
      "epoch": 2.219334759774363,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004502246966395137,
      "loss": 4.9702,
      "mean_token_accuracy": 0.20609240233898163,
      "num_tokens": 49487521.0,
      "step": 28525
    },
    {
      "entropy": 5.710647773742676,
      "epoch": 2.2197237891460806,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045020730927825274,
      "loss": 5.0485,
      "mean_token_accuracy": 0.19773528277873992,
      "num_tokens": 49495382.0,
      "step": 28530
    },
    {
      "entropy": 5.660996437072754,
      "epoch": 2.2201128185177983,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004501899192584997,
      "loss": 4.9359,
      "mean_token_accuracy": 0.21080552488565446,
      "num_tokens": 49503819.0,
      "step": 28535
    },
    {
      "entropy": 5.66800856590271,
      "epoch": 2.2205018478895155,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004501725265805185,
      "loss": 4.971,
      "mean_token_accuracy": 0.2079767495393753,
      "num_tokens": 49512908.0,
      "step": 28540
    },
    {
      "entropy": 5.684831953048706,
      "epoch": 2.2208908772612332,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045015513124457303,
      "loss": 5.0086,
      "mean_token_accuracy": 0.20947207808494567,
      "num_tokens": 49522462.0,
      "step": 28545
    },
    {
      "entropy": 5.721297073364258,
      "epoch": 2.221279906632951,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004501377332509272,
      "loss": 5.0053,
      "mean_token_accuracy": 0.200913704931736,
      "num_tokens": 49531089.0,
      "step": 28550
    },
    {
      "entropy": 5.720208787918091,
      "epoch": 2.221668936004668,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045012033259984494,
      "loss": 4.9216,
      "mean_token_accuracy": 0.20799678564071655,
      "num_tokens": 49539074.0,
      "step": 28555
    },
    {
      "entropy": 5.6479880809783936,
      "epoch": 2.222057965376386,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004501029292915905,
      "loss": 4.9912,
      "mean_token_accuracy": 0.20569613873958587,
      "num_tokens": 49547879.0,
      "step": 28560
    },
    {
      "entropy": 5.7733283042907715,
      "epoch": 2.2224469947481036,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045008552332642774,
      "loss": 5.0539,
      "mean_token_accuracy": 0.20528892427682877,
      "num_tokens": 49556220.0,
      "step": 28565
    },
    {
      "entropy": 5.731104373931885,
      "epoch": 2.222836024119821,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045006811470462083,
      "loss": 4.9753,
      "mean_token_accuracy": 0.20574127584695817,
      "num_tokens": 49564710.0,
      "step": 28570
    },
    {
      "entropy": 5.655422019958496,
      "epoch": 2.2232250534915385,
      "grad_norm": 1.28125,
      "learning_rate": 0.000450050703426434,
      "loss": 4.8703,
      "mean_token_accuracy": 0.21152126044034958,
      "num_tokens": 49572829.0,
      "step": 28575
    },
    {
      "entropy": 5.672166872024536,
      "epoch": 2.2236140828632562,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004500332894921313,
      "loss": 5.0415,
      "mean_token_accuracy": 0.20736970454454423,
      "num_tokens": 49580355.0,
      "step": 28580
    },
    {
      "entropy": 5.67030143737793,
      "epoch": 2.224003112234974,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004500158729019771,
      "loss": 4.9701,
      "mean_token_accuracy": 0.21005136519670486,
      "num_tokens": 49589086.0,
      "step": 28585
    },
    {
      "entropy": 5.661626958847046,
      "epoch": 2.224392141606691,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004499984536562355,
      "loss": 4.9077,
      "mean_token_accuracy": 0.2092096135020256,
      "num_tokens": 49597434.0,
      "step": 28590
    },
    {
      "entropy": 5.7259471893310545,
      "epoch": 2.224781170978409,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000449981031755171,
      "loss": 4.9838,
      "mean_token_accuracy": 0.20647546350955964,
      "num_tokens": 49604981.0,
      "step": 28595
    },
    {
      "entropy": 5.719478797912598,
      "epoch": 2.2251702003501266,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044996360719904777,
      "loss": 4.9882,
      "mean_token_accuracy": 0.210093754529953,
      "num_tokens": 49614119.0,
      "step": 28600
    },
    {
      "entropy": 5.729244375228882,
      "epoch": 2.225559229721844,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004499461799881303,
      "loss": 5.0198,
      "mean_token_accuracy": 0.1975761204957962,
      "num_tokens": 49622705.0,
      "step": 28605
    },
    {
      "entropy": 5.707197284698486,
      "epoch": 2.2259482590935615,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004499287501226831,
      "loss": 5.019,
      "mean_token_accuracy": 0.19443641602993011,
      "num_tokens": 49632086.0,
      "step": 28610
    },
    {
      "entropy": 5.690701961517334,
      "epoch": 2.2263372884652792,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044991131760297045,
      "loss": 4.9953,
      "mean_token_accuracy": 0.20548304319381713,
      "num_tokens": 49640733.0,
      "step": 28615
    },
    {
      "entropy": 5.6291852474212645,
      "epoch": 2.226726317836997,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044989388242925693,
      "loss": 4.9644,
      "mean_token_accuracy": 0.2103191941976547,
      "num_tokens": 49649694.0,
      "step": 28620
    },
    {
      "entropy": 5.723702049255371,
      "epoch": 2.227115347208714,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004498764446018073,
      "loss": 5.0514,
      "mean_token_accuracy": 0.20421615540981292,
      "num_tokens": 49657536.0,
      "step": 28625
    },
    {
      "entropy": 5.684167003631591,
      "epoch": 2.227504376580432,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004498590041208857,
      "loss": 4.9247,
      "mean_token_accuracy": 0.20953033417463302,
      "num_tokens": 49666332.0,
      "step": 28630
    },
    {
      "entropy": 5.6207653999328615,
      "epoch": 2.2278934059521496,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004498415609867572,
      "loss": 4.8958,
      "mean_token_accuracy": 0.21054576337337494,
      "num_tokens": 49675364.0,
      "step": 28635
    },
    {
      "entropy": 5.725854301452637,
      "epoch": 2.228282435323867,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044982411519968625,
      "loss": 5.0849,
      "mean_token_accuracy": 0.1959310218691826,
      "num_tokens": 49684178.0,
      "step": 28640
    },
    {
      "entropy": 5.724223566055298,
      "epoch": 2.2286714646955845,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004498066667599377,
      "loss": 5.0252,
      "mean_token_accuracy": 0.20158506631851197,
      "num_tokens": 49692747.0,
      "step": 28645
    },
    {
      "entropy": 5.6909246921539305,
      "epoch": 2.2290604940673022,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044978921566777604,
      "loss": 4.9901,
      "mean_token_accuracy": 0.20546096861362456,
      "num_tokens": 49701534.0,
      "step": 28650
    },
    {
      "entropy": 5.721595048904419,
      "epoch": 2.2294495234390195,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004497717619234664,
      "loss": 5.0003,
      "mean_token_accuracy": 0.20464992225170137,
      "num_tokens": 49710021.0,
      "step": 28655
    },
    {
      "entropy": 5.656595945358276,
      "epoch": 2.229838552810737,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044975430552727337,
      "loss": 4.9146,
      "mean_token_accuracy": 0.2082376942038536,
      "num_tokens": 49718786.0,
      "step": 28660
    },
    {
      "entropy": 5.698299264907837,
      "epoch": 2.230227582182455,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004497368464794619,
      "loss": 4.9766,
      "mean_token_accuracy": 0.2066560745239258,
      "num_tokens": 49726840.0,
      "step": 28665
    },
    {
      "entropy": 5.7308910369873045,
      "epoch": 2.230616611554172,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044971938478029693,
      "loss": 5.0569,
      "mean_token_accuracy": 0.19669657498598098,
      "num_tokens": 49736130.0,
      "step": 28670
    },
    {
      "entropy": 5.683980321884155,
      "epoch": 2.23100564092589,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044970192043004343,
      "loss": 4.9568,
      "mean_token_accuracy": 0.19948695003986358,
      "num_tokens": 49744810.0,
      "step": 28675
    },
    {
      "entropy": 5.652162170410156,
      "epoch": 2.2313946702976075,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004496844534289663,
      "loss": 4.9739,
      "mean_token_accuracy": 0.20694368183612824,
      "num_tokens": 49754053.0,
      "step": 28680
    },
    {
      "entropy": 5.704453468322754,
      "epoch": 2.231783699669325,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044966698377733064,
      "loss": 4.9767,
      "mean_token_accuracy": 0.20244079679250718,
      "num_tokens": 49763735.0,
      "step": 28685
    },
    {
      "entropy": 5.729519891738891,
      "epoch": 2.2321727290410425,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004496495114754015,
      "loss": 5.0444,
      "mean_token_accuracy": 0.1996692880988121,
      "num_tokens": 49772867.0,
      "step": 28690
    },
    {
      "entropy": 5.628458261489868,
      "epoch": 2.23256175841276,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044963203652344404,
      "loss": 4.8646,
      "mean_token_accuracy": 0.21059111505746841,
      "num_tokens": 49781460.0,
      "step": 28695
    },
    {
      "entropy": 5.649736738204956,
      "epoch": 2.232950787784478,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004496145589217234,
      "loss": 4.9012,
      "mean_token_accuracy": 0.20927857160568236,
      "num_tokens": 49790249.0,
      "step": 28700
    },
    {
      "entropy": 5.698975276947022,
      "epoch": 2.233339817156195,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044959707867050467,
      "loss": 4.9445,
      "mean_token_accuracy": 0.20887715965509415,
      "num_tokens": 49797959.0,
      "step": 28705
    },
    {
      "entropy": 5.638365268707275,
      "epoch": 2.233728846527913,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004495795957700532,
      "loss": 4.8885,
      "mean_token_accuracy": 0.21395120918750762,
      "num_tokens": 49806669.0,
      "step": 28710
    },
    {
      "entropy": 5.684682607650757,
      "epoch": 2.2341178758996305,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044956211022063424,
      "loss": 5.0901,
      "mean_token_accuracy": 0.195084348320961,
      "num_tokens": 49816821.0,
      "step": 28715
    },
    {
      "entropy": 5.744039297103882,
      "epoch": 2.234506905271348,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044954462202251316,
      "loss": 4.999,
      "mean_token_accuracy": 0.204379865527153,
      "num_tokens": 49825411.0,
      "step": 28720
    },
    {
      "entropy": 5.7246973514556885,
      "epoch": 2.2348959346430655,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044952713117595516,
      "loss": 5.0207,
      "mean_token_accuracy": 0.1994592308998108,
      "num_tokens": 49833496.0,
      "step": 28725
    },
    {
      "entropy": 5.636021089553833,
      "epoch": 2.235284964014783,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044950963768122576,
      "loss": 4.9539,
      "mean_token_accuracy": 0.1975434511899948,
      "num_tokens": 49842935.0,
      "step": 28730
    },
    {
      "entropy": 5.655119466781616,
      "epoch": 2.235673993386501,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004494921415385905,
      "loss": 5.045,
      "mean_token_accuracy": 0.1971466913819313,
      "num_tokens": 49851197.0,
      "step": 28735
    },
    {
      "entropy": 5.676263284683228,
      "epoch": 2.236063022758218,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044947464274831464,
      "loss": 4.9121,
      "mean_token_accuracy": 0.21161855459213258,
      "num_tokens": 49859778.0,
      "step": 28740
    },
    {
      "entropy": 5.693059587478638,
      "epoch": 2.236452052129936,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004494571413106637,
      "loss": 4.9804,
      "mean_token_accuracy": 0.20127903372049333,
      "num_tokens": 49868548.0,
      "step": 28745
    },
    {
      "entropy": 5.700651359558106,
      "epoch": 2.2368410815016535,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044943963722590344,
      "loss": 4.9562,
      "mean_token_accuracy": 0.20496193915605546,
      "num_tokens": 49877070.0,
      "step": 28750
    },
    {
      "entropy": 5.720123815536499,
      "epoch": 2.2372301108733708,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044942213049429917,
      "loss": 5.0072,
      "mean_token_accuracy": 0.20534780472517014,
      "num_tokens": 49885225.0,
      "step": 28755
    },
    {
      "entropy": 5.6429948806762695,
      "epoch": 2.2376191402450885,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044940462111611685,
      "loss": 5.0909,
      "mean_token_accuracy": 0.19609288424253463,
      "num_tokens": 49894305.0,
      "step": 28760
    },
    {
      "entropy": 5.644094085693359,
      "epoch": 2.238008169616806,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044938710909162195,
      "loss": 4.9231,
      "mean_token_accuracy": 0.21245565563440322,
      "num_tokens": 49902587.0,
      "step": 28765
    },
    {
      "entropy": 5.723309326171875,
      "epoch": 2.2383971989885234,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004493695944210802,
      "loss": 4.866,
      "mean_token_accuracy": 0.2109153687953949,
      "num_tokens": 49911037.0,
      "step": 28770
    },
    {
      "entropy": 5.665737438201904,
      "epoch": 2.238786228360241,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004493520771047574,
      "loss": 5.0203,
      "mean_token_accuracy": 0.20227380841970444,
      "num_tokens": 49920724.0,
      "step": 28775
    },
    {
      "entropy": 5.631714105606079,
      "epoch": 2.239175257731959,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044933455714291944,
      "loss": 4.9865,
      "mean_token_accuracy": 0.212224280834198,
      "num_tokens": 49929173.0,
      "step": 28780
    },
    {
      "entropy": 5.699655151367187,
      "epoch": 2.2395642871036765,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044931703453583195,
      "loss": 4.9587,
      "mean_token_accuracy": 0.20062381625175477,
      "num_tokens": 49938213.0,
      "step": 28785
    },
    {
      "entropy": 5.673656177520752,
      "epoch": 2.2399533164753938,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044929950928376095,
      "loss": 4.8821,
      "mean_token_accuracy": 0.21025510132312775,
      "num_tokens": 49946493.0,
      "step": 28790
    },
    {
      "entropy": 5.609847784042358,
      "epoch": 2.2403423458471114,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004492819813869723,
      "loss": 4.9498,
      "mean_token_accuracy": 0.2060559019446373,
      "num_tokens": 49954730.0,
      "step": 28795
    },
    {
      "entropy": 5.597666597366333,
      "epoch": 2.240731375218829,
      "grad_norm": 1.125,
      "learning_rate": 0.00044926445084573206,
      "loss": 4.9017,
      "mean_token_accuracy": 0.20812904834747314,
      "num_tokens": 49962651.0,
      "step": 28800
    },
    {
      "entropy": 5.700533294677735,
      "epoch": 2.2411204045905464,
      "grad_norm": 1.109375,
      "learning_rate": 0.000449246917660306,
      "loss": 4.9742,
      "mean_token_accuracy": 0.19800878167152405,
      "num_tokens": 49972560.0,
      "step": 28805
    },
    {
      "entropy": 5.691119813919068,
      "epoch": 2.241509433962264,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004492293818309604,
      "loss": 4.9991,
      "mean_token_accuracy": 0.20763276517391205,
      "num_tokens": 49982292.0,
      "step": 28810
    },
    {
      "entropy": 5.704091262817383,
      "epoch": 2.241898463333982,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004492118433579613,
      "loss": 5.0782,
      "mean_token_accuracy": 0.19732230454683303,
      "num_tokens": 49990395.0,
      "step": 28815
    },
    {
      "entropy": 5.690478420257568,
      "epoch": 2.2422874927056995,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044919430224157463,
      "loss": 5.0134,
      "mean_token_accuracy": 0.20094195157289504,
      "num_tokens": 49998944.0,
      "step": 28820
    },
    {
      "entropy": 5.686327600479126,
      "epoch": 2.2426765220774167,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044917675848206684,
      "loss": 4.9185,
      "mean_token_accuracy": 0.2024995818734169,
      "num_tokens": 50007583.0,
      "step": 28825
    },
    {
      "entropy": 5.602932119369507,
      "epoch": 2.2430655514491344,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044915921207970387,
      "loss": 4.8388,
      "mean_token_accuracy": 0.2183154493570328,
      "num_tokens": 50015854.0,
      "step": 28830
    },
    {
      "entropy": 5.599769210815429,
      "epoch": 2.243454580820852,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044914166303475206,
      "loss": 4.8536,
      "mean_token_accuracy": 0.21773716062307358,
      "num_tokens": 50024757.0,
      "step": 28835
    },
    {
      "entropy": 5.690275192260742,
      "epoch": 2.2438436101925694,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044912411134747764,
      "loss": 4.9291,
      "mean_token_accuracy": 0.20915476977825165,
      "num_tokens": 50034592.0,
      "step": 28840
    },
    {
      "entropy": 5.667948007583618,
      "epoch": 2.244232639564287,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044910655701814705,
      "loss": 4.958,
      "mean_token_accuracy": 0.20083529353141785,
      "num_tokens": 50043034.0,
      "step": 28845
    },
    {
      "entropy": 5.6177990436553955,
      "epoch": 2.244621668936005,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004490890000470266,
      "loss": 4.9584,
      "mean_token_accuracy": 0.20431554913520814,
      "num_tokens": 50051683.0,
      "step": 28850
    },
    {
      "entropy": 5.653458833694458,
      "epoch": 2.245010698307722,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044907144043438265,
      "loss": 4.9636,
      "mean_token_accuracy": 0.20053597390651703,
      "num_tokens": 50060257.0,
      "step": 28855
    },
    {
      "entropy": 5.594372129440307,
      "epoch": 2.2453997276794397,
      "grad_norm": 1.25,
      "learning_rate": 0.0004490538781804816,
      "loss": 4.8912,
      "mean_token_accuracy": 0.20885107517242432,
      "num_tokens": 50068188.0,
      "step": 28860
    },
    {
      "entropy": 5.7045458316802975,
      "epoch": 2.2457887570511574,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044903631328559,
      "loss": 4.9528,
      "mean_token_accuracy": 0.2024354502558708,
      "num_tokens": 50076188.0,
      "step": 28865
    },
    {
      "entropy": 5.699604129791259,
      "epoch": 2.246177786422875,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044901874574997444,
      "loss": 4.96,
      "mean_token_accuracy": 0.21424244344234467,
      "num_tokens": 50085139.0,
      "step": 28870
    },
    {
      "entropy": 5.7349790096282955,
      "epoch": 2.2465668157945924,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044900117557390137,
      "loss": 5.046,
      "mean_token_accuracy": 0.20355013608932496,
      "num_tokens": 50093050.0,
      "step": 28875
    },
    {
      "entropy": 5.654831457138061,
      "epoch": 2.24695584516631,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044898360275763733,
      "loss": 5.0007,
      "mean_token_accuracy": 0.206048446893692,
      "num_tokens": 50101854.0,
      "step": 28880
    },
    {
      "entropy": 5.729950475692749,
      "epoch": 2.247344874538028,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044896602730144905,
      "loss": 5.0229,
      "mean_token_accuracy": 0.2008727714419365,
      "num_tokens": 50110430.0,
      "step": 28885
    },
    {
      "entropy": 5.727756595611572,
      "epoch": 2.247733903909745,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044894844920560323,
      "loss": 4.9492,
      "mean_token_accuracy": 0.2020856410264969,
      "num_tokens": 50118633.0,
      "step": 28890
    },
    {
      "entropy": 5.731924962997437,
      "epoch": 2.2481229332814627,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004489308684703666,
      "loss": 5.0326,
      "mean_token_accuracy": 0.20265383273363113,
      "num_tokens": 50129120.0,
      "step": 28895
    },
    {
      "entropy": 5.747328567504883,
      "epoch": 2.2485119626531804,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004489132850960059,
      "loss": 5.0271,
      "mean_token_accuracy": 0.1986215054988861,
      "num_tokens": 50138404.0,
      "step": 28900
    },
    {
      "entropy": 5.68381700515747,
      "epoch": 2.2489009920248977,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004488956990827878,
      "loss": 4.977,
      "mean_token_accuracy": 0.1966927245259285,
      "num_tokens": 50146908.0,
      "step": 28905
    },
    {
      "entropy": 5.617712545394897,
      "epoch": 2.2492900213966154,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004488781104309793,
      "loss": 4.8874,
      "mean_token_accuracy": 0.20636935234069825,
      "num_tokens": 50155922.0,
      "step": 28910
    },
    {
      "entropy": 5.6566321849823,
      "epoch": 2.249679050768333,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044886051914084726,
      "loss": 5.0118,
      "mean_token_accuracy": 0.20579150021076204,
      "num_tokens": 50164023.0,
      "step": 28915
    },
    {
      "entropy": 5.641082334518432,
      "epoch": 2.2500680801400508,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044884292521265846,
      "loss": 4.9398,
      "mean_token_accuracy": 0.2023431271314621,
      "num_tokens": 50172147.0,
      "step": 28920
    },
    {
      "entropy": 5.715638971328735,
      "epoch": 2.250457109511768,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004488253286466801,
      "loss": 5.0395,
      "mean_token_accuracy": 0.19719575196504593,
      "num_tokens": 50180484.0,
      "step": 28925
    },
    {
      "entropy": 5.666181230545044,
      "epoch": 2.2508461388834857,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044880772944317905,
      "loss": 4.9015,
      "mean_token_accuracy": 0.2028050199151039,
      "num_tokens": 50189176.0,
      "step": 28930
    },
    {
      "entropy": 5.662782192230225,
      "epoch": 2.2512351682552034,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044879012760242224,
      "loss": 4.9383,
      "mean_token_accuracy": 0.2144239366054535,
      "num_tokens": 50197674.0,
      "step": 28935
    },
    {
      "entropy": 5.682578611373901,
      "epoch": 2.2516241976269207,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044877252312467694,
      "loss": 5.0065,
      "mean_token_accuracy": 0.20204246640205384,
      "num_tokens": 50206681.0,
      "step": 28940
    },
    {
      "entropy": 5.7812501907348635,
      "epoch": 2.2520132269986384,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004487549160102101,
      "loss": 5.0563,
      "mean_token_accuracy": 0.19705769866704942,
      "num_tokens": 50215457.0,
      "step": 28945
    },
    {
      "entropy": 5.69146032333374,
      "epoch": 2.252402256370356,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044873730625928914,
      "loss": 4.9275,
      "mean_token_accuracy": 0.21140078604221343,
      "num_tokens": 50224757.0,
      "step": 28950
    },
    {
      "entropy": 5.722042751312256,
      "epoch": 2.2527912857420738,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044871969387218094,
      "loss": 4.9953,
      "mean_token_accuracy": 0.20348539501428603,
      "num_tokens": 50233570.0,
      "step": 28955
    },
    {
      "entropy": 5.729600667953491,
      "epoch": 2.253180315113791,
      "grad_norm": 1.28125,
      "learning_rate": 0.000448702078849153,
      "loss": 4.9573,
      "mean_token_accuracy": 0.2048760697245598,
      "num_tokens": 50241767.0,
      "step": 28960
    },
    {
      "entropy": 5.609009647369385,
      "epoch": 2.2535693444855087,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044868446119047234,
      "loss": 4.9155,
      "mean_token_accuracy": 0.21366450935602188,
      "num_tokens": 50250121.0,
      "step": 28965
    },
    {
      "entropy": 5.756006050109863,
      "epoch": 2.253958373857226,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004486668408964065,
      "loss": 5.0642,
      "mean_token_accuracy": 0.19718434810638427,
      "num_tokens": 50258717.0,
      "step": 28970
    },
    {
      "entropy": 5.7003124237060545,
      "epoch": 2.2543474032289437,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004486492179672228,
      "loss": 5.0035,
      "mean_token_accuracy": 0.20884720534086226,
      "num_tokens": 50267528.0,
      "step": 28975
    },
    {
      "entropy": 5.723517274856567,
      "epoch": 2.2547364326006614,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044863159240318863,
      "loss": 4.9806,
      "mean_token_accuracy": 0.20377870351076127,
      "num_tokens": 50276197.0,
      "step": 28980
    },
    {
      "entropy": 5.6395947456359865,
      "epoch": 2.255125461972379,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004486139642045714,
      "loss": 4.8875,
      "mean_token_accuracy": 0.21368484646081926,
      "num_tokens": 50284222.0,
      "step": 28985
    },
    {
      "entropy": 5.6487274169921875,
      "epoch": 2.2555144913440963,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004485963333716385,
      "loss": 4.9178,
      "mean_token_accuracy": 0.20188630372285843,
      "num_tokens": 50293061.0,
      "step": 28990
    },
    {
      "entropy": 5.655241775512695,
      "epoch": 2.255903520715814,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004485786999046576,
      "loss": 4.9534,
      "mean_token_accuracy": 0.20759957879781724,
      "num_tokens": 50301601.0,
      "step": 28995
    },
    {
      "entropy": 5.643373441696167,
      "epoch": 2.2562925500875317,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044856106380389624,
      "loss": 4.9908,
      "mean_token_accuracy": 0.20638565123081207,
      "num_tokens": 50311595.0,
      "step": 29000
    },
    {
      "entropy": 5.63364086151123,
      "epoch": 2.256681579459249,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000448543425069622,
      "loss": 4.9366,
      "mean_token_accuracy": 0.21395501345396042,
      "num_tokens": 50320205.0,
      "step": 29005
    },
    {
      "entropy": 5.712969827651977,
      "epoch": 2.2570706088309667,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004485257837021025,
      "loss": 4.9445,
      "mean_token_accuracy": 0.20548329502344131,
      "num_tokens": 50328414.0,
      "step": 29010
    },
    {
      "entropy": 5.646304655075073,
      "epoch": 2.2574596382026844,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044850813970160534,
      "loss": 4.9627,
      "mean_token_accuracy": 0.20800715535879136,
      "num_tokens": 50337206.0,
      "step": 29015
    },
    {
      "entropy": 5.671200323104858,
      "epoch": 2.257848667574402,
      "grad_norm": 1.125,
      "learning_rate": 0.0004484904930683984,
      "loss": 5.0062,
      "mean_token_accuracy": 0.20039883106946946,
      "num_tokens": 50346320.0,
      "step": 29020
    },
    {
      "entropy": 5.6965906620025635,
      "epoch": 2.2582376969461193,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004484728438027494,
      "loss": 4.9908,
      "mean_token_accuracy": 0.20056817978620528,
      "num_tokens": 50354199.0,
      "step": 29025
    },
    {
      "entropy": 5.699464988708496,
      "epoch": 2.258626726317837,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004484551919049261,
      "loss": 4.9893,
      "mean_token_accuracy": 0.2064412161707878,
      "num_tokens": 50363085.0,
      "step": 29030
    },
    {
      "entropy": 5.638885450363159,
      "epoch": 2.2590157556895547,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044843753737519625,
      "loss": 4.9385,
      "mean_token_accuracy": 0.20455342382192612,
      "num_tokens": 50371211.0,
      "step": 29035
    },
    {
      "entropy": 5.696005725860596,
      "epoch": 2.259404785061272,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004484198802138279,
      "loss": 4.9405,
      "mean_token_accuracy": 0.21405108124017716,
      "num_tokens": 50379426.0,
      "step": 29040
    },
    {
      "entropy": 5.688122320175171,
      "epoch": 2.2597938144329897,
      "grad_norm": 1.25,
      "learning_rate": 0.0004484022204210889,
      "loss": 5.0253,
      "mean_token_accuracy": 0.20341238081455232,
      "num_tokens": 50388073.0,
      "step": 29045
    },
    {
      "entropy": 5.639048624038696,
      "epoch": 2.2601828438047074,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044838455799724717,
      "loss": 4.9075,
      "mean_token_accuracy": 0.21038170903921127,
      "num_tokens": 50395943.0,
      "step": 29050
    },
    {
      "entropy": 5.6716564178466795,
      "epoch": 2.260571873176425,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044836689294257075,
      "loss": 4.9534,
      "mean_token_accuracy": 0.20146092772483826,
      "num_tokens": 50405218.0,
      "step": 29055
    },
    {
      "entropy": 5.6467078685760494,
      "epoch": 2.2609609025481423,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004483492252573276,
      "loss": 4.9367,
      "mean_token_accuracy": 0.20354214757680894,
      "num_tokens": 50414169.0,
      "step": 29060
    },
    {
      "entropy": 5.6424572467803955,
      "epoch": 2.26134993191986,
      "grad_norm": 1.328125,
      "learning_rate": 0.000448331554941786,
      "loss": 5.0229,
      "mean_token_accuracy": 0.20111193060874938,
      "num_tokens": 50422541.0,
      "step": 29065
    },
    {
      "entropy": 5.651554584503174,
      "epoch": 2.2617389612915777,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044831388199621385,
      "loss": 4.9059,
      "mean_token_accuracy": 0.21080812215805053,
      "num_tokens": 50431081.0,
      "step": 29070
    },
    {
      "entropy": 5.7153356075286865,
      "epoch": 2.262127990663295,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044829620642087946,
      "loss": 5.0006,
      "mean_token_accuracy": 0.20751287043094635,
      "num_tokens": 50439279.0,
      "step": 29075
    },
    {
      "entropy": 5.6551210403442385,
      "epoch": 2.2625170200350126,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004482785282160509,
      "loss": 4.9827,
      "mean_token_accuracy": 0.20528028160333633,
      "num_tokens": 50447628.0,
      "step": 29080
    },
    {
      "entropy": 5.703428888320923,
      "epoch": 2.2629060494067303,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044826084738199657,
      "loss": 4.9983,
      "mean_token_accuracy": 0.20377519130706787,
      "num_tokens": 50457109.0,
      "step": 29085
    },
    {
      "entropy": 5.6925492763519285,
      "epoch": 2.2632950787784476,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044824316391898464,
      "loss": 4.9127,
      "mean_token_accuracy": 0.20971190333366393,
      "num_tokens": 50466222.0,
      "step": 29090
    },
    {
      "entropy": 5.671778631210327,
      "epoch": 2.2636841081501653,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044822547782728334,
      "loss": 4.9389,
      "mean_token_accuracy": 0.20679860562086105,
      "num_tokens": 50475382.0,
      "step": 29095
    },
    {
      "entropy": 5.720492506027222,
      "epoch": 2.264073137521883,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004482077891071612,
      "loss": 5.0008,
      "mean_token_accuracy": 0.20674118548631668,
      "num_tokens": 50483501.0,
      "step": 29100
    },
    {
      "entropy": 5.694304704666138,
      "epoch": 2.2644621668936002,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044819009775888655,
      "loss": 4.9403,
      "mean_token_accuracy": 0.20562614500522614,
      "num_tokens": 50492005.0,
      "step": 29105
    },
    {
      "entropy": 5.640299654006958,
      "epoch": 2.264851196265318,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044817240378272784,
      "loss": 4.8553,
      "mean_token_accuracy": 0.21824472546577453,
      "num_tokens": 50500578.0,
      "step": 29110
    },
    {
      "entropy": 5.6453718662261965,
      "epoch": 2.2652402256370356,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044815470717895345,
      "loss": 4.9157,
      "mean_token_accuracy": 0.20263569205999374,
      "num_tokens": 50509285.0,
      "step": 29115
    },
    {
      "entropy": 5.627928781509399,
      "epoch": 2.2656292550087533,
      "grad_norm": 1.25,
      "learning_rate": 0.0004481370079478321,
      "loss": 4.9345,
      "mean_token_accuracy": 0.2070313423871994,
      "num_tokens": 50517519.0,
      "step": 29120
    },
    {
      "entropy": 5.676782608032227,
      "epoch": 2.2660182843804706,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004481193060896322,
      "loss": 4.9484,
      "mean_token_accuracy": 0.20466541051864623,
      "num_tokens": 50525843.0,
      "step": 29125
    },
    {
      "entropy": 5.698052978515625,
      "epoch": 2.2664073137521883,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004481016016046223,
      "loss": 4.9996,
      "mean_token_accuracy": 0.20431241095066072,
      "num_tokens": 50535638.0,
      "step": 29130
    },
    {
      "entropy": 5.723114633560181,
      "epoch": 2.266796343123906,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004480838944930712,
      "loss": 5.0225,
      "mean_token_accuracy": 0.20058311223983766,
      "num_tokens": 50544327.0,
      "step": 29135
    },
    {
      "entropy": 5.705422687530517,
      "epoch": 2.2671853724956232,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004480661847552474,
      "loss": 4.9623,
      "mean_token_accuracy": 0.20297568142414094,
      "num_tokens": 50552276.0,
      "step": 29140
    },
    {
      "entropy": 5.620291042327881,
      "epoch": 2.267574401867341,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004480484723914198,
      "loss": 4.8959,
      "mean_token_accuracy": 0.20897440910339354,
      "num_tokens": 50561247.0,
      "step": 29145
    },
    {
      "entropy": 5.726911973953247,
      "epoch": 2.2679634312390586,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004480307574018571,
      "loss": 4.9566,
      "mean_token_accuracy": 0.2100980430841446,
      "num_tokens": 50569401.0,
      "step": 29150
    },
    {
      "entropy": 5.701439332962036,
      "epoch": 2.2683524606107763,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000448013039786828,
      "loss": 4.9136,
      "mean_token_accuracy": 0.21291811466217042,
      "num_tokens": 50577346.0,
      "step": 29155
    },
    {
      "entropy": 5.612297105789184,
      "epoch": 2.2687414899824936,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044799531954660133,
      "loss": 4.9503,
      "mean_token_accuracy": 0.20680011361837386,
      "num_tokens": 50586370.0,
      "step": 29160
    },
    {
      "entropy": 5.641057300567627,
      "epoch": 2.2691305193542113,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004479775966814461,
      "loss": 4.8748,
      "mean_token_accuracy": 0.20872782319784164,
      "num_tokens": 50595326.0,
      "step": 29165
    },
    {
      "entropy": 5.747226715087891,
      "epoch": 2.269519548725929,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044795987119163115,
      "loss": 4.9711,
      "mean_token_accuracy": 0.20932436287403106,
      "num_tokens": 50603513.0,
      "step": 29170
    },
    {
      "entropy": 5.685132312774658,
      "epoch": 2.2699085780976462,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004479421430774255,
      "loss": 4.9356,
      "mean_token_accuracy": 0.19994676411151885,
      "num_tokens": 50612416.0,
      "step": 29175
    },
    {
      "entropy": 5.6654462814331055,
      "epoch": 2.270297607469364,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004479244123390981,
      "loss": 4.9576,
      "mean_token_accuracy": 0.2146904155611992,
      "num_tokens": 50620480.0,
      "step": 29180
    },
    {
      "entropy": 5.555679988861084,
      "epoch": 2.2706866368410816,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044790667897691796,
      "loss": 4.8598,
      "mean_token_accuracy": 0.217963807284832,
      "num_tokens": 50629044.0,
      "step": 29185
    },
    {
      "entropy": 5.725050687789917,
      "epoch": 2.271075666212799,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044788894299115415,
      "loss": 5.0697,
      "mean_token_accuracy": 0.199474561214447,
      "num_tokens": 50638610.0,
      "step": 29190
    },
    {
      "entropy": 5.63061671257019,
      "epoch": 2.2714646955845166,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004478712043820758,
      "loss": 4.949,
      "mean_token_accuracy": 0.20216356664896012,
      "num_tokens": 50647337.0,
      "step": 29195
    },
    {
      "entropy": 5.684944725036621,
      "epoch": 2.2718537249562343,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004478534631499521,
      "loss": 4.9817,
      "mean_token_accuracy": 0.20851464867591857,
      "num_tokens": 50654641.0,
      "step": 29200
    },
    {
      "entropy": 5.651462030410767,
      "epoch": 2.2722427543279515,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044783571929505223,
      "loss": 5.0065,
      "mean_token_accuracy": 0.19233741015195846,
      "num_tokens": 50663726.0,
      "step": 29205
    },
    {
      "entropy": 5.745954132080078,
      "epoch": 2.2726317836996692,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004478179728176454,
      "loss": 5.0332,
      "mean_token_accuracy": 0.2048352301120758,
      "num_tokens": 50672376.0,
      "step": 29210
    },
    {
      "entropy": 5.7330944538116455,
      "epoch": 2.273020813071387,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004478002237180009,
      "loss": 5.0283,
      "mean_token_accuracy": 0.1986946538090706,
      "num_tokens": 50680990.0,
      "step": 29215
    },
    {
      "entropy": 5.723233890533447,
      "epoch": 2.2734098424431046,
      "grad_norm": 1.109375,
      "learning_rate": 0.000447782471996388,
      "loss": 5.0686,
      "mean_token_accuracy": 0.19473170042037963,
      "num_tokens": 50690180.0,
      "step": 29220
    },
    {
      "entropy": 5.706169939041137,
      "epoch": 2.273798871814822,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004477647176530762,
      "loss": 4.9643,
      "mean_token_accuracy": 0.2066655933856964,
      "num_tokens": 50699094.0,
      "step": 29225
    },
    {
      "entropy": 5.712173509597778,
      "epoch": 2.2741879011865396,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004477469606883347,
      "loss": 4.9772,
      "mean_token_accuracy": 0.2010608658194542,
      "num_tokens": 50708312.0,
      "step": 29230
    },
    {
      "entropy": 5.610710573196411,
      "epoch": 2.2745769305582573,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004477292011024331,
      "loss": 4.9517,
      "mean_token_accuracy": 0.20605751872062683,
      "num_tokens": 50717092.0,
      "step": 29235
    },
    {
      "entropy": 5.6310817241668705,
      "epoch": 2.2749659599299745,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044771143889564075,
      "loss": 4.9273,
      "mean_token_accuracy": 0.2103450909256935,
      "num_tokens": 50725641.0,
      "step": 29240
    },
    {
      "entropy": 5.647262620925903,
      "epoch": 2.275354989301692,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044769367406822727,
      "loss": 4.9486,
      "mean_token_accuracy": 0.20846968293190002,
      "num_tokens": 50734082.0,
      "step": 29245
    },
    {
      "entropy": 5.715780258178711,
      "epoch": 2.27574401867341,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004476759066204621,
      "loss": 5.0504,
      "mean_token_accuracy": 0.19583817720413207,
      "num_tokens": 50743242.0,
      "step": 29250
    },
    {
      "entropy": 5.688461542129517,
      "epoch": 2.2761330480451276,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004476581365526149,
      "loss": 4.9099,
      "mean_token_accuracy": 0.20614150315523147,
      "num_tokens": 50751693.0,
      "step": 29255
    },
    {
      "entropy": 5.68257703781128,
      "epoch": 2.276522077416845,
      "grad_norm": 1.25,
      "learning_rate": 0.0004476403638649553,
      "loss": 4.9853,
      "mean_token_accuracy": 0.19477285593748092,
      "num_tokens": 50760031.0,
      "step": 29260
    },
    {
      "entropy": 5.627707004547119,
      "epoch": 2.2769111067885626,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044762258855775294,
      "loss": 4.9877,
      "mean_token_accuracy": 0.20667606741189956,
      "num_tokens": 50768236.0,
      "step": 29265
    },
    {
      "entropy": 5.68505482673645,
      "epoch": 2.2773001361602803,
      "grad_norm": 1.125,
      "learning_rate": 0.00044760481063127754,
      "loss": 4.9603,
      "mean_token_accuracy": 0.20635338127613068,
      "num_tokens": 50777055.0,
      "step": 29270
    },
    {
      "entropy": 5.719148778915406,
      "epoch": 2.2776891655319975,
      "grad_norm": 1.203125,
      "learning_rate": 0.000447587030085799,
      "loss": 4.9977,
      "mean_token_accuracy": 0.2014760732650757,
      "num_tokens": 50785574.0,
      "step": 29275
    },
    {
      "entropy": 5.637813234329224,
      "epoch": 2.278078194903715,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044756924692158684,
      "loss": 4.9838,
      "mean_token_accuracy": 0.20921201556921004,
      "num_tokens": 50794889.0,
      "step": 29280
    },
    {
      "entropy": 5.626026678085327,
      "epoch": 2.278467224275433,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004475514611389111,
      "loss": 5.0012,
      "mean_token_accuracy": 0.2037532076239586,
      "num_tokens": 50804823.0,
      "step": 29285
    },
    {
      "entropy": 5.716028833389283,
      "epoch": 2.2788562536471506,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004475336727380415,
      "loss": 4.8782,
      "mean_token_accuracy": 0.21451846063137053,
      "num_tokens": 50813233.0,
      "step": 29290
    },
    {
      "entropy": 5.671346759796142,
      "epoch": 2.279245283018868,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004475158817192481,
      "loss": 4.94,
      "mean_token_accuracy": 0.2055838882923126,
      "num_tokens": 50821925.0,
      "step": 29295
    },
    {
      "entropy": 5.669095611572265,
      "epoch": 2.2796343123905856,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044749808808280077,
      "loss": 4.993,
      "mean_token_accuracy": 0.20585575997829436,
      "num_tokens": 50831337.0,
      "step": 29300
    },
    {
      "entropy": 5.637554073333741,
      "epoch": 2.280023341762303,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044748029182896956,
      "loss": 4.9329,
      "mean_token_accuracy": 0.2113611415028572,
      "num_tokens": 50840493.0,
      "step": 29305
    },
    {
      "entropy": 5.720972919464112,
      "epoch": 2.2804123711340205,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004474624929580243,
      "loss": 5.0538,
      "mean_token_accuracy": 0.20116288363933563,
      "num_tokens": 50848843.0,
      "step": 29310
    },
    {
      "entropy": 5.731148195266724,
      "epoch": 2.280801400505738,
      "grad_norm": 1.125,
      "learning_rate": 0.0004474446914702354,
      "loss": 4.9696,
      "mean_token_accuracy": 0.20632649660110475,
      "num_tokens": 50857882.0,
      "step": 29315
    },
    {
      "entropy": 5.731100606918335,
      "epoch": 2.281190429877456,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004474268873658727,
      "loss": 5.0921,
      "mean_token_accuracy": 0.2023192197084427,
      "num_tokens": 50866058.0,
      "step": 29320
    },
    {
      "entropy": 5.636650276184082,
      "epoch": 2.281579459249173,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004474090806452064,
      "loss": 4.9928,
      "mean_token_accuracy": 0.202754969894886,
      "num_tokens": 50874175.0,
      "step": 29325
    },
    {
      "entropy": 5.641986131668091,
      "epoch": 2.281968488620891,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044739127130850675,
      "loss": 4.9497,
      "mean_token_accuracy": 0.20431893616914748,
      "num_tokens": 50883753.0,
      "step": 29330
    },
    {
      "entropy": 5.687885093688965,
      "epoch": 2.2823575179926086,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044737345935604397,
      "loss": 5.0189,
      "mean_token_accuracy": 0.2017875999212265,
      "num_tokens": 50891962.0,
      "step": 29335
    },
    {
      "entropy": 5.7277669429779055,
      "epoch": 2.282746547364326,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004473556447880883,
      "loss": 4.919,
      "mean_token_accuracy": 0.20891950279474258,
      "num_tokens": 50900405.0,
      "step": 29340
    },
    {
      "entropy": 5.727472400665283,
      "epoch": 2.2831355767360435,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044733782760490997,
      "loss": 5.0052,
      "mean_token_accuracy": 0.20217760950326918,
      "num_tokens": 50908850.0,
      "step": 29345
    },
    {
      "entropy": 5.669453907012939,
      "epoch": 2.283524606107761,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004473200078067795,
      "loss": 5.0645,
      "mean_token_accuracy": 0.19744303822517395,
      "num_tokens": 50918545.0,
      "step": 29350
    },
    {
      "entropy": 5.633996915817261,
      "epoch": 2.283913635479479,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044730218539396717,
      "loss": 4.9608,
      "mean_token_accuracy": 0.20658664405345917,
      "num_tokens": 50926664.0,
      "step": 29355
    },
    {
      "entropy": 5.701661252975464,
      "epoch": 2.284302664851196,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004472843603667434,
      "loss": 5.0443,
      "mean_token_accuracy": 0.1977180317044258,
      "num_tokens": 50935846.0,
      "step": 29360
    },
    {
      "entropy": 5.641158676147461,
      "epoch": 2.284691694222914,
      "grad_norm": 1.125,
      "learning_rate": 0.00044726653272537866,
      "loss": 4.9724,
      "mean_token_accuracy": 0.20433624237775802,
      "num_tokens": 50944289.0,
      "step": 29365
    },
    {
      "entropy": 5.664590644836426,
      "epoch": 2.2850807235946315,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004472487024701435,
      "loss": 4.9752,
      "mean_token_accuracy": 0.20093775987625123,
      "num_tokens": 50952949.0,
      "step": 29370
    },
    {
      "entropy": 5.76599235534668,
      "epoch": 2.285469752966349,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004472308696013085,
      "loss": 5.0744,
      "mean_token_accuracy": 0.19779651165008544,
      "num_tokens": 50961957.0,
      "step": 29375
    },
    {
      "entropy": 5.732619905471802,
      "epoch": 2.2858587823380665,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004472130341191441,
      "loss": 4.9738,
      "mean_token_accuracy": 0.20971042960882186,
      "num_tokens": 50970463.0,
      "step": 29380
    },
    {
      "entropy": 5.726377296447754,
      "epoch": 2.286247811709784,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004471951960239211,
      "loss": 4.9983,
      "mean_token_accuracy": 0.2012412205338478,
      "num_tokens": 50979350.0,
      "step": 29385
    },
    {
      "entropy": 5.709884834289551,
      "epoch": 2.286636841081502,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044717735531591,
      "loss": 5.0284,
      "mean_token_accuracy": 0.20080165266990663,
      "num_tokens": 50988048.0,
      "step": 29390
    },
    {
      "entropy": 5.679438829421997,
      "epoch": 2.287025870453219,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044715951199538156,
      "loss": 5.003,
      "mean_token_accuracy": 0.2052282363176346,
      "num_tokens": 50996714.0,
      "step": 29395
    },
    {
      "entropy": 5.7158972263336185,
      "epoch": 2.287414899824937,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044714166606260657,
      "loss": 4.9665,
      "mean_token_accuracy": 0.20451468080282212,
      "num_tokens": 51005557.0,
      "step": 29400
    },
    {
      "entropy": 5.746420669555664,
      "epoch": 2.2878039291966545,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004471238175178559,
      "loss": 4.9237,
      "mean_token_accuracy": 0.20573557317256927,
      "num_tokens": 51014024.0,
      "step": 29405
    },
    {
      "entropy": 5.6476668357849125,
      "epoch": 2.288192958568372,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004471059663614002,
      "loss": 4.8881,
      "mean_token_accuracy": 0.2060413882136345,
      "num_tokens": 51022211.0,
      "step": 29410
    },
    {
      "entropy": 5.673877763748169,
      "epoch": 2.2885819879400895,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004470881125935103,
      "loss": 5.0546,
      "mean_token_accuracy": 0.20810577273368835,
      "num_tokens": 51030459.0,
      "step": 29415
    },
    {
      "entropy": 5.736504983901978,
      "epoch": 2.288971017311807,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044707025621445735,
      "loss": 5.0427,
      "mean_token_accuracy": 0.20137473642826081,
      "num_tokens": 51038439.0,
      "step": 29420
    },
    {
      "entropy": 5.7015434265136715,
      "epoch": 2.2893600466835244,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044705239722451194,
      "loss": 4.9149,
      "mean_token_accuracy": 0.2088994488120079,
      "num_tokens": 51047698.0,
      "step": 29425
    },
    {
      "entropy": 5.6779866218566895,
      "epoch": 2.289749076055242,
      "grad_norm": 1.3125,
      "learning_rate": 0.00044703453562394544,
      "loss": 4.9436,
      "mean_token_accuracy": 0.2096534326672554,
      "num_tokens": 51056211.0,
      "step": 29430
    },
    {
      "entropy": 5.68499722480774,
      "epoch": 2.29013810542696,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004470166714130286,
      "loss": 4.9607,
      "mean_token_accuracy": 0.2061391741037369,
      "num_tokens": 51064321.0,
      "step": 29435
    },
    {
      "entropy": 5.673834753036499,
      "epoch": 2.290527134798677,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044699880459203264,
      "loss": 4.959,
      "mean_token_accuracy": 0.20158083736896515,
      "num_tokens": 51072537.0,
      "step": 29440
    },
    {
      "entropy": 5.678911304473877,
      "epoch": 2.290916164170395,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044698093516122844,
      "loss": 4.9535,
      "mean_token_accuracy": 0.201720729470253,
      "num_tokens": 51080873.0,
      "step": 29445
    },
    {
      "entropy": 5.716307210922241,
      "epoch": 2.2913051935421125,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004469630631208874,
      "loss": 5.026,
      "mean_token_accuracy": 0.20832799077033998,
      "num_tokens": 51089462.0,
      "step": 29450
    },
    {
      "entropy": 5.672102212905884,
      "epoch": 2.29169422291383,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004469451884712805,
      "loss": 4.9746,
      "mean_token_accuracy": 0.20351845622062684,
      "num_tokens": 51098918.0,
      "step": 29455
    },
    {
      "entropy": 5.717929029464722,
      "epoch": 2.2920832522855474,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004469273112126791,
      "loss": 5.0508,
      "mean_token_accuracy": 0.2033880665898323,
      "num_tokens": 51107483.0,
      "step": 29460
    },
    {
      "entropy": 5.748307228088379,
      "epoch": 2.292472281657265,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044690943134535445,
      "loss": 5.0284,
      "mean_token_accuracy": 0.20059556663036346,
      "num_tokens": 51116590.0,
      "step": 29465
    },
    {
      "entropy": 5.625221300125122,
      "epoch": 2.292861311028983,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004468915488695777,
      "loss": 4.8574,
      "mean_token_accuracy": 0.20738319158554078,
      "num_tokens": 51125296.0,
      "step": 29470
    },
    {
      "entropy": 5.6528740406036375,
      "epoch": 2.2932503404007,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044687366378562036,
      "loss": 4.9542,
      "mean_token_accuracy": 0.20194720327854157,
      "num_tokens": 51133369.0,
      "step": 29475
    },
    {
      "entropy": 5.760650157928467,
      "epoch": 2.293639369772418,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044685577609375373,
      "loss": 5.1069,
      "mean_token_accuracy": 0.20396720319986344,
      "num_tokens": 51142789.0,
      "step": 29480
    },
    {
      "entropy": 5.713109254837036,
      "epoch": 2.2940283991441355,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004468378857942492,
      "loss": 4.9576,
      "mean_token_accuracy": 0.1962758868932724,
      "num_tokens": 51151706.0,
      "step": 29485
    },
    {
      "entropy": 5.7486796379089355,
      "epoch": 2.294417428515853,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044681999288737826,
      "loss": 5.004,
      "mean_token_accuracy": 0.20968174785375596,
      "num_tokens": 51160924.0,
      "step": 29490
    },
    {
      "entropy": 5.720389938354492,
      "epoch": 2.2948064578875704,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00044680209737341244,
      "loss": 5.015,
      "mean_token_accuracy": 0.19941507875919343,
      "num_tokens": 51168871.0,
      "step": 29495
    },
    {
      "entropy": 5.671631240844727,
      "epoch": 2.295195487259288,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004467841992526233,
      "loss": 4.9267,
      "mean_token_accuracy": 0.2097158372402191,
      "num_tokens": 51177621.0,
      "step": 29500
    },
    {
      "entropy": 5.70870156288147,
      "epoch": 2.295584516631006,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044676629852528226,
      "loss": 4.9108,
      "mean_token_accuracy": 0.2091941237449646,
      "num_tokens": 51186004.0,
      "step": 29505
    },
    {
      "entropy": 5.674426794052124,
      "epoch": 2.295973546002723,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044674839519166104,
      "loss": 4.9099,
      "mean_token_accuracy": 0.20640100091695784,
      "num_tokens": 51194609.0,
      "step": 29510
    },
    {
      "entropy": 5.688918876647949,
      "epoch": 2.2963625753744408,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044673048925203136,
      "loss": 5.0259,
      "mean_token_accuracy": 0.20936882942914964,
      "num_tokens": 51203928.0,
      "step": 29515
    },
    {
      "entropy": 5.671715831756591,
      "epoch": 2.2967516047461585,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044671258070666476,
      "loss": 4.9656,
      "mean_token_accuracy": 0.21011485606431962,
      "num_tokens": 51212762.0,
      "step": 29520
    },
    {
      "entropy": 5.761320972442627,
      "epoch": 2.2971406341178757,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004466946695558331,
      "loss": 5.0783,
      "mean_token_accuracy": 0.1972859039902687,
      "num_tokens": 51221064.0,
      "step": 29525
    },
    {
      "entropy": 5.6770295143127445,
      "epoch": 2.2975296634895934,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044667675579980817,
      "loss": 4.9538,
      "mean_token_accuracy": 0.2028470128774643,
      "num_tokens": 51229633.0,
      "step": 29530
    },
    {
      "entropy": 5.718072080612183,
      "epoch": 2.297918692861311,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004466588394388616,
      "loss": 5.0024,
      "mean_token_accuracy": 0.2017356976866722,
      "num_tokens": 51239051.0,
      "step": 29535
    },
    {
      "entropy": 5.7224444389343265,
      "epoch": 2.2983077222330284,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004466409204732654,
      "loss": 4.9526,
      "mean_token_accuracy": 0.2047087073326111,
      "num_tokens": 51247703.0,
      "step": 29540
    },
    {
      "entropy": 5.648996067047119,
      "epoch": 2.298696751604746,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044662299890329144,
      "loss": 4.8933,
      "mean_token_accuracy": 0.21530297249555588,
      "num_tokens": 51256394.0,
      "step": 29545
    },
    {
      "entropy": 5.714378070831299,
      "epoch": 2.2990857809764638,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004466050747292116,
      "loss": 4.97,
      "mean_token_accuracy": 0.2055521473288536,
      "num_tokens": 51264181.0,
      "step": 29550
    },
    {
      "entropy": 5.645342111587524,
      "epoch": 2.2994748103481815,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000446587147951298,
      "loss": 4.8713,
      "mean_token_accuracy": 0.21444391310214997,
      "num_tokens": 51272758.0,
      "step": 29555
    },
    {
      "entropy": 5.714525985717773,
      "epoch": 2.2998638397198987,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004465692185698224,
      "loss": 5.0069,
      "mean_token_accuracy": 0.20195600539445877,
      "num_tokens": 51281681.0,
      "step": 29560
    },
    {
      "entropy": 5.7003857612609865,
      "epoch": 2.3002528690916164,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044655128658505717,
      "loss": 5.0092,
      "mean_token_accuracy": 0.2031571462750435,
      "num_tokens": 51290060.0,
      "step": 29565
    },
    {
      "entropy": 5.717109680175781,
      "epoch": 2.300641898463334,
      "grad_norm": 1.25,
      "learning_rate": 0.0004465333519972741,
      "loss": 4.9104,
      "mean_token_accuracy": 0.2083080753684044,
      "num_tokens": 51298262.0,
      "step": 29570
    },
    {
      "entropy": 5.608543395996094,
      "epoch": 2.3010309278350514,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004465154148067454,
      "loss": 4.852,
      "mean_token_accuracy": 0.21393874734640123,
      "num_tokens": 51306436.0,
      "step": 29575
    },
    {
      "entropy": 5.627600002288818,
      "epoch": 2.301419957206769,
      "grad_norm": 1.3125,
      "learning_rate": 0.00044649747501374336,
      "loss": 4.9494,
      "mean_token_accuracy": 0.20661789923906326,
      "num_tokens": 51315132.0,
      "step": 29580
    },
    {
      "entropy": 5.6241217136383055,
      "epoch": 2.3018089865784868,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004464795326185401,
      "loss": 4.8938,
      "mean_token_accuracy": 0.21037013232707977,
      "num_tokens": 51324189.0,
      "step": 29585
    },
    {
      "entropy": 5.636724185943604,
      "epoch": 2.3021980159502045,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004464615876214078,
      "loss": 4.9481,
      "mean_token_accuracy": 0.20687221586704255,
      "num_tokens": 51332680.0,
      "step": 29590
    },
    {
      "entropy": 5.698414611816406,
      "epoch": 2.3025870453219217,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004464436400226188,
      "loss": 5.0058,
      "mean_token_accuracy": 0.20218137204647063,
      "num_tokens": 51342036.0,
      "step": 29595
    },
    {
      "entropy": 5.691271018981934,
      "epoch": 2.3029760746936394,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004464256898224455,
      "loss": 4.9344,
      "mean_token_accuracy": 0.20007749795913696,
      "num_tokens": 51350987.0,
      "step": 29600
    },
    {
      "entropy": 5.691755867004394,
      "epoch": 2.303365104065357,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004464077370211602,
      "loss": 4.9391,
      "mean_token_accuracy": 0.2053200200200081,
      "num_tokens": 51359787.0,
      "step": 29605
    },
    {
      "entropy": 5.691194295883179,
      "epoch": 2.3037541334370744,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044638978161903533,
      "loss": 5.0425,
      "mean_token_accuracy": 0.19902438670396805,
      "num_tokens": 51368329.0,
      "step": 29610
    },
    {
      "entropy": 5.699222660064697,
      "epoch": 2.304143162808792,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004463718236163433,
      "loss": 5.0128,
      "mean_token_accuracy": 0.19826048761606216,
      "num_tokens": 51377104.0,
      "step": 29615
    },
    {
      "entropy": 5.72039213180542,
      "epoch": 2.3045321921805098,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044635386301335666,
      "loss": 5.0416,
      "mean_token_accuracy": 0.19737450182437896,
      "num_tokens": 51385210.0,
      "step": 29620
    },
    {
      "entropy": 5.700424718856811,
      "epoch": 2.304921221552227,
      "grad_norm": 1.25,
      "learning_rate": 0.0004463358998103478,
      "loss": 4.9782,
      "mean_token_accuracy": 0.20035565197467803,
      "num_tokens": 51393492.0,
      "step": 29625
    },
    {
      "entropy": 5.619831943511963,
      "epoch": 2.3053102509239447,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004463179340075894,
      "loss": 4.8388,
      "mean_token_accuracy": 0.216587632894516,
      "num_tokens": 51401953.0,
      "step": 29630
    },
    {
      "entropy": 5.663472080230713,
      "epoch": 2.3056992802956624,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004462999656053541,
      "loss": 5.0218,
      "mean_token_accuracy": 0.1992644652724266,
      "num_tokens": 51410874.0,
      "step": 29635
    },
    {
      "entropy": 5.670195198059082,
      "epoch": 2.3060883096673797,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044628199460391445,
      "loss": 4.9042,
      "mean_token_accuracy": 0.21523504704236984,
      "num_tokens": 51419189.0,
      "step": 29640
    },
    {
      "entropy": 5.7437238693237305,
      "epoch": 2.3064773390390974,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044626402100354307,
      "loss": 4.9757,
      "mean_token_accuracy": 0.20536222755908967,
      "num_tokens": 51427970.0,
      "step": 29645
    },
    {
      "entropy": 5.76452865600586,
      "epoch": 2.306866368410815,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004462460448045129,
      "loss": 4.9826,
      "mean_token_accuracy": 0.2053862690925598,
      "num_tokens": 51436069.0,
      "step": 29650
    },
    {
      "entropy": 5.660533332824707,
      "epoch": 2.3072553977825327,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044622806600709645,
      "loss": 4.9987,
      "mean_token_accuracy": 0.20407008081674577,
      "num_tokens": 51444424.0,
      "step": 29655
    },
    {
      "entropy": 5.673260498046875,
      "epoch": 2.30764442715425,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004462100846115667,
      "loss": 4.9608,
      "mean_token_accuracy": 0.20363647788763045,
      "num_tokens": 51453095.0,
      "step": 29660
    },
    {
      "entropy": 5.692152309417724,
      "epoch": 2.3080334565259677,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004461921006181964,
      "loss": 4.9108,
      "mean_token_accuracy": 0.2128586158156395,
      "num_tokens": 51461525.0,
      "step": 29665
    },
    {
      "entropy": 5.692513561248779,
      "epoch": 2.3084224858976854,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004461741140272584,
      "loss": 4.9334,
      "mean_token_accuracy": 0.20867202430963516,
      "num_tokens": 51469918.0,
      "step": 29670
    },
    {
      "entropy": 5.705195951461792,
      "epoch": 2.3088115152694026,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004461561248390257,
      "loss": 5.0229,
      "mean_token_accuracy": 0.20667574256658555,
      "num_tokens": 51478747.0,
      "step": 29675
    },
    {
      "entropy": 5.681801891326904,
      "epoch": 2.3092005446411203,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004461381330537713,
      "loss": 4.9772,
      "mean_token_accuracy": 0.20057612359523774,
      "num_tokens": 51487730.0,
      "step": 29680
    },
    {
      "entropy": 5.720447635650634,
      "epoch": 2.309589574012838,
      "grad_norm": 1.21875,
      "learning_rate": 0.000446120138671768,
      "loss": 4.9766,
      "mean_token_accuracy": 0.20250645279884338,
      "num_tokens": 51495672.0,
      "step": 29685
    },
    {
      "entropy": 5.7067454814910885,
      "epoch": 2.3099786033845557,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00044610214169328913,
      "loss": 4.9527,
      "mean_token_accuracy": 0.20430502742528917,
      "num_tokens": 51503377.0,
      "step": 29690
    },
    {
      "entropy": 5.624398612976075,
      "epoch": 2.310367632756273,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004460841421186075,
      "loss": 4.8975,
      "mean_token_accuracy": 0.21312921941280366,
      "num_tokens": 51511932.0,
      "step": 29695
    },
    {
      "entropy": 5.737968826293946,
      "epoch": 2.3107566621279907,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004460661399479963,
      "loss": 4.9774,
      "mean_token_accuracy": 0.20106977522373198,
      "num_tokens": 51520565.0,
      "step": 29700
    },
    {
      "entropy": 5.668567132949829,
      "epoch": 2.3111456914997084,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044604813518172876,
      "loss": 4.8625,
      "mean_token_accuracy": 0.21146301925182343,
      "num_tokens": 51529778.0,
      "step": 29705
    },
    {
      "entropy": 5.701326322555542,
      "epoch": 2.3115347208714256,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044603012782007796,
      "loss": 4.9914,
      "mean_token_accuracy": 0.20631556808948517,
      "num_tokens": 51539394.0,
      "step": 29710
    },
    {
      "entropy": 5.6826231479644775,
      "epoch": 2.3119237502431433,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004460121178633172,
      "loss": 4.9431,
      "mean_token_accuracy": 0.20364482402801515,
      "num_tokens": 51548314.0,
      "step": 29715
    },
    {
      "entropy": 5.619504070281982,
      "epoch": 2.312312779614861,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044599410531171986,
      "loss": 4.8627,
      "mean_token_accuracy": 0.21865388453006745,
      "num_tokens": 51556828.0,
      "step": 29720
    },
    {
      "entropy": 5.697992038726807,
      "epoch": 2.3127018089865787,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000445976090165559,
      "loss": 4.9887,
      "mean_token_accuracy": 0.20174259096384048,
      "num_tokens": 51565031.0,
      "step": 29725
    },
    {
      "entropy": 5.742313575744629,
      "epoch": 2.313090838358296,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004459580724251082,
      "loss": 5.0667,
      "mean_token_accuracy": 0.19750434756278992,
      "num_tokens": 51574218.0,
      "step": 29730
    },
    {
      "entropy": 5.639999485015869,
      "epoch": 2.3134798677300137,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044594005209064064,
      "loss": 4.9431,
      "mean_token_accuracy": 0.2074110507965088,
      "num_tokens": 51583229.0,
      "step": 29735
    },
    {
      "entropy": 5.645054578781128,
      "epoch": 2.313868897101731,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044592202916243003,
      "loss": 4.9339,
      "mean_token_accuracy": 0.20483045876026154,
      "num_tokens": 51591969.0,
      "step": 29740
    },
    {
      "entropy": 5.702998256683349,
      "epoch": 2.3142579264734486,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004459040036407495,
      "loss": 4.9471,
      "mean_token_accuracy": 0.2063742458820343,
      "num_tokens": 51600840.0,
      "step": 29745
    },
    {
      "entropy": 5.631151008605957,
      "epoch": 2.3146469558451663,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004458859755258729,
      "loss": 4.8006,
      "mean_token_accuracy": 0.22266144305467606,
      "num_tokens": 51608750.0,
      "step": 29750
    },
    {
      "entropy": 5.697806072235108,
      "epoch": 2.315035985216884,
      "grad_norm": 1.3125,
      "learning_rate": 0.00044586794481807357,
      "loss": 4.9888,
      "mean_token_accuracy": 0.21005109399557115,
      "num_tokens": 51617079.0,
      "step": 29755
    },
    {
      "entropy": 5.621458292007446,
      "epoch": 2.3154250145886013,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044584991151762506,
      "loss": 4.9753,
      "mean_token_accuracy": 0.20438900887966155,
      "num_tokens": 51625310.0,
      "step": 29760
    },
    {
      "entropy": 5.669579935073853,
      "epoch": 2.315814043960319,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004458318756248011,
      "loss": 4.9499,
      "mean_token_accuracy": 0.20532214790582656,
      "num_tokens": 51634071.0,
      "step": 29765
    },
    {
      "entropy": 5.677065515518189,
      "epoch": 2.3162030733320367,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044581383713987547,
      "loss": 4.9369,
      "mean_token_accuracy": 0.2069862574338913,
      "num_tokens": 51642761.0,
      "step": 29770
    },
    {
      "entropy": 5.682136392593383,
      "epoch": 2.316592102703754,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004457957960631216,
      "loss": 4.9607,
      "mean_token_accuracy": 0.21094863563776017,
      "num_tokens": 51652160.0,
      "step": 29775
    },
    {
      "entropy": 5.617931509017945,
      "epoch": 2.3169811320754716,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004457777523948134,
      "loss": 4.9056,
      "mean_token_accuracy": 0.20430764853954314,
      "num_tokens": 51661345.0,
      "step": 29780
    },
    {
      "entropy": 5.679696035385132,
      "epoch": 2.3173701614471893,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004457597061352247,
      "loss": 4.9687,
      "mean_token_accuracy": 0.20836330950260162,
      "num_tokens": 51669855.0,
      "step": 29785
    },
    {
      "entropy": 5.7489996433258055,
      "epoch": 2.317759190818907,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004457416572846291,
      "loss": 5.0589,
      "mean_token_accuracy": 0.20627965033054352,
      "num_tokens": 51679272.0,
      "step": 29790
    },
    {
      "entropy": 5.691824722290039,
      "epoch": 2.3181482201906243,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004457236058433008,
      "loss": 4.9478,
      "mean_token_accuracy": 0.20235888361930848,
      "num_tokens": 51688383.0,
      "step": 29795
    },
    {
      "entropy": 5.727585554122925,
      "epoch": 2.318537249562342,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044570555181151333,
      "loss": 4.9836,
      "mean_token_accuracy": 0.20171405822038652,
      "num_tokens": 51696996.0,
      "step": 29800
    },
    {
      "entropy": 5.683440160751343,
      "epoch": 2.3189262789340597,
      "grad_norm": 1.265625,
      "learning_rate": 0.000445687495189541,
      "loss": 5.0307,
      "mean_token_accuracy": 0.20353493839502335,
      "num_tokens": 51705830.0,
      "step": 29805
    },
    {
      "entropy": 5.635271644592285,
      "epoch": 2.319315308305777,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044566943597765745,
      "loss": 4.9419,
      "mean_token_accuracy": 0.2075018212199211,
      "num_tokens": 51715357.0,
      "step": 29810
    },
    {
      "entropy": 5.68923544883728,
      "epoch": 2.3197043376774946,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044565137417613694,
      "loss": 5.0012,
      "mean_token_accuracy": 0.2094678610563278,
      "num_tokens": 51725342.0,
      "step": 29815
    },
    {
      "entropy": 5.768256330490113,
      "epoch": 2.3200933670492123,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004456333097852534,
      "loss": 4.9726,
      "mean_token_accuracy": 0.20478148460388185,
      "num_tokens": 51733788.0,
      "step": 29820
    },
    {
      "entropy": 5.706807422637939,
      "epoch": 2.32048239642093,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044561524280528104,
      "loss": 5.0064,
      "mean_token_accuracy": 0.20504017770290375,
      "num_tokens": 51742713.0,
      "step": 29825
    },
    {
      "entropy": 5.734003400802612,
      "epoch": 2.3208714257926473,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044559717323649383,
      "loss": 5.0837,
      "mean_token_accuracy": 0.20199520885944366,
      "num_tokens": 51751482.0,
      "step": 29830
    },
    {
      "entropy": 5.608273220062256,
      "epoch": 2.321260455164365,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004455791010791662,
      "loss": 4.8346,
      "mean_token_accuracy": 0.21601275354623795,
      "num_tokens": 51760217.0,
      "step": 29835
    },
    {
      "entropy": 5.676750993728637,
      "epoch": 2.3216494845360827,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004455610263335721,
      "loss": 4.9922,
      "mean_token_accuracy": 0.20319125354290007,
      "num_tokens": 51767945.0,
      "step": 29840
    },
    {
      "entropy": 5.6727605819702145,
      "epoch": 2.3220385139078,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004455429489999859,
      "loss": 5.008,
      "mean_token_accuracy": 0.19976552277803422,
      "num_tokens": 51777318.0,
      "step": 29845
    },
    {
      "entropy": 5.689283657073974,
      "epoch": 2.3224275432795176,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044552486907868194,
      "loss": 4.9636,
      "mean_token_accuracy": 0.20612811744213105,
      "num_tokens": 51786571.0,
      "step": 29850
    },
    {
      "entropy": 5.740775299072266,
      "epoch": 2.3228165726512353,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044550678656993454,
      "loss": 5.0487,
      "mean_token_accuracy": 0.19938250333070756,
      "num_tokens": 51794872.0,
      "step": 29855
    },
    {
      "entropy": 5.745616340637207,
      "epoch": 2.3232056020229526,
      "grad_norm": 1.28125,
      "learning_rate": 0.000445488701474018,
      "loss": 5.002,
      "mean_token_accuracy": 0.19768717586994172,
      "num_tokens": 51803111.0,
      "step": 29860
    },
    {
      "entropy": 5.666545581817627,
      "epoch": 2.3235946313946703,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004454706137912067,
      "loss": 4.8879,
      "mean_token_accuracy": 0.21341898292303085,
      "num_tokens": 51811574.0,
      "step": 29865
    },
    {
      "entropy": 5.662443733215332,
      "epoch": 2.323983660766388,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004454525235217753,
      "loss": 4.8932,
      "mean_token_accuracy": 0.208584326505661,
      "num_tokens": 51820120.0,
      "step": 29870
    },
    {
      "entropy": 5.73737587928772,
      "epoch": 2.324372690138105,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044543443066599807,
      "loss": 5.0344,
      "mean_token_accuracy": 0.20604564994573593,
      "num_tokens": 51828293.0,
      "step": 29875
    },
    {
      "entropy": 5.727134466171265,
      "epoch": 2.324761719509823,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004454163352241498,
      "loss": 4.9421,
      "mean_token_accuracy": 0.20448226928710939,
      "num_tokens": 51836595.0,
      "step": 29880
    },
    {
      "entropy": 5.726908111572266,
      "epoch": 2.3251507488815406,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044539823719650463,
      "loss": 4.9938,
      "mean_token_accuracy": 0.20290320813655854,
      "num_tokens": 51845415.0,
      "step": 29885
    },
    {
      "entropy": 5.6966259479522705,
      "epoch": 2.3255397782532583,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004453801365833376,
      "loss": 4.9599,
      "mean_token_accuracy": 0.20741076171398162,
      "num_tokens": 51854091.0,
      "step": 29890
    },
    {
      "entropy": 5.665568447113037,
      "epoch": 2.3259288076249756,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044536203338492317,
      "loss": 4.9393,
      "mean_token_accuracy": 0.20820313543081284,
      "num_tokens": 51862767.0,
      "step": 29895
    },
    {
      "entropy": 5.658937692642212,
      "epoch": 2.3263178369966933,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044534392760153596,
      "loss": 4.8647,
      "mean_token_accuracy": 0.2145675465464592,
      "num_tokens": 51871593.0,
      "step": 29900
    },
    {
      "entropy": 5.7147924423217775,
      "epoch": 2.326706866368411,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004453258192334508,
      "loss": 4.9723,
      "mean_token_accuracy": 0.21021701544523239,
      "num_tokens": 51880775.0,
      "step": 29905
    },
    {
      "entropy": 5.666798543930054,
      "epoch": 2.327095895740128,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004453077082809425,
      "loss": 4.9603,
      "mean_token_accuracy": 0.20458897203207016,
      "num_tokens": 51889663.0,
      "step": 29910
    },
    {
      "entropy": 5.746243524551391,
      "epoch": 2.327484925111846,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044528959474428575,
      "loss": 4.9823,
      "mean_token_accuracy": 0.19908542633056642,
      "num_tokens": 51898565.0,
      "step": 29915
    },
    {
      "entropy": 5.689568996429443,
      "epoch": 2.3278739544835636,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004452714786237555,
      "loss": 4.9652,
      "mean_token_accuracy": 0.20675784796476365,
      "num_tokens": 51906828.0,
      "step": 29920
    },
    {
      "entropy": 5.656858491897583,
      "epoch": 2.3282629838552813,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004452533599196265,
      "loss": 4.9104,
      "mean_token_accuracy": 0.2050057142972946,
      "num_tokens": 51914305.0,
      "step": 29925
    },
    {
      "entropy": 5.684875535964966,
      "epoch": 2.3286520132269986,
      "grad_norm": 1.125,
      "learning_rate": 0.00044523523863217374,
      "loss": 5.0207,
      "mean_token_accuracy": 0.20663006454706193,
      "num_tokens": 51923544.0,
      "step": 29930
    },
    {
      "entropy": 5.71252851486206,
      "epoch": 2.3290410425987162,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004452171147616723,
      "loss": 4.9475,
      "mean_token_accuracy": 0.2061650425195694,
      "num_tokens": 51932198.0,
      "step": 29935
    },
    {
      "entropy": 5.662913656234741,
      "epoch": 2.329430071970434,
      "grad_norm": 1.234375,
      "learning_rate": 0.000445198988308397,
      "loss": 4.9133,
      "mean_token_accuracy": 0.21388867050409316,
      "num_tokens": 51940291.0,
      "step": 29940
    },
    {
      "entropy": 5.6071278095245365,
      "epoch": 2.329819101342151,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044518085927262293,
      "loss": 4.8552,
      "mean_token_accuracy": 0.213763627409935,
      "num_tokens": 51948899.0,
      "step": 29945
    },
    {
      "entropy": 5.6583771228790285,
      "epoch": 2.330208130713869,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004451627276546252,
      "loss": 4.9318,
      "mean_token_accuracy": 0.20397166758775712,
      "num_tokens": 51957726.0,
      "step": 29950
    },
    {
      "entropy": 5.71642279624939,
      "epoch": 2.3305971600855866,
      "grad_norm": 1.21875,
      "learning_rate": 0.000445144593454679,
      "loss": 4.9567,
      "mean_token_accuracy": 0.20640612989664078,
      "num_tokens": 51965922.0,
      "step": 29955
    },
    {
      "entropy": 5.662117195129395,
      "epoch": 2.330986189457304,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004451264566730593,
      "loss": 4.8964,
      "mean_token_accuracy": 0.20549058616161348,
      "num_tokens": 51974096.0,
      "step": 29960
    },
    {
      "entropy": 5.666009950637817,
      "epoch": 2.3313752188290215,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044510831731004146,
      "loss": 4.9673,
      "mean_token_accuracy": 0.20187125951051713,
      "num_tokens": 51982954.0,
      "step": 29965
    },
    {
      "entropy": 5.664199161529541,
      "epoch": 2.3317642482007392,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004450901753659007,
      "loss": 4.9404,
      "mean_token_accuracy": 0.21144961714744567,
      "num_tokens": 51991738.0,
      "step": 29970
    },
    {
      "entropy": 5.69935245513916,
      "epoch": 2.3321532775724565,
      "grad_norm": 1.25,
      "learning_rate": 0.00044507203084091215,
      "loss": 4.9383,
      "mean_token_accuracy": 0.20855033099651338,
      "num_tokens": 52000227.0,
      "step": 29975
    },
    {
      "entropy": 5.752443838119507,
      "epoch": 2.332542306944174,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004450538837353513,
      "loss": 5.0675,
      "mean_token_accuracy": 0.19365671873092652,
      "num_tokens": 52009336.0,
      "step": 29980
    },
    {
      "entropy": 5.692458009719848,
      "epoch": 2.332931336315892,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044503573404949343,
      "loss": 4.9838,
      "mean_token_accuracy": 0.207385878264904,
      "num_tokens": 52017850.0,
      "step": 29985
    },
    {
      "entropy": 5.70288200378418,
      "epoch": 2.3333203656876096,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004450175817836139,
      "loss": 4.9891,
      "mean_token_accuracy": 0.20155593752861023,
      "num_tokens": 52027672.0,
      "step": 29990
    },
    {
      "entropy": 5.709821653366089,
      "epoch": 2.333709395059327,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044499942693798823,
      "loss": 4.9684,
      "mean_token_accuracy": 0.20217744410037994,
      "num_tokens": 52036134.0,
      "step": 29995
    },
    {
      "entropy": 5.713846635818482,
      "epoch": 2.3340984244310445,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004449812695128918,
      "loss": 4.9638,
      "mean_token_accuracy": 0.20485790818929672,
      "num_tokens": 52044719.0,
      "step": 30000
    },
    {
      "epoch": 2.3340984244310445,
      "eval_entropy": 5.479735658625437,
      "eval_loss": 5.060657501220703,
      "eval_mean_token_accuracy": 0.21144516642444747,
      "eval_num_tokens": 52044719.0,
      "eval_runtime": 21.708,
      "eval_samples_per_second": 1914.411,
      "eval_steps_per_second": 239.313,
      "step": 30000
    },
    {
      "entropy": 5.701676750183106,
      "epoch": 2.3344874538027622,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044496310950860015,
      "loss": 5.0274,
      "mean_token_accuracy": 0.20307065546512604,
      "num_tokens": 52053238.0,
      "step": 30005
    },
    {
      "entropy": 5.715251064300537,
      "epoch": 2.3348764831744795,
      "grad_norm": 1.125,
      "learning_rate": 0.00044494494692538886,
      "loss": 5.0216,
      "mean_token_accuracy": 0.2022714614868164,
      "num_tokens": 52062258.0,
      "step": 30010
    },
    {
      "entropy": 5.739684629440307,
      "epoch": 2.335265512546197,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044492678176353347,
      "loss": 5.0085,
      "mean_token_accuracy": 0.20341558754444122,
      "num_tokens": 52070527.0,
      "step": 30015
    },
    {
      "entropy": 5.700058794021606,
      "epoch": 2.335654541917915,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044490861402330967,
      "loss": 5.0376,
      "mean_token_accuracy": 0.20443532913923262,
      "num_tokens": 52079827.0,
      "step": 30020
    },
    {
      "entropy": 5.6983497619628904,
      "epoch": 2.3360435712896326,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004448904437049931,
      "loss": 4.9621,
      "mean_token_accuracy": 0.20738120675086974,
      "num_tokens": 52088963.0,
      "step": 30025
    },
    {
      "entropy": 5.695399332046509,
      "epoch": 2.33643260066135,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004448722708088594,
      "loss": 4.9511,
      "mean_token_accuracy": 0.20227610617876052,
      "num_tokens": 52096885.0,
      "step": 30030
    },
    {
      "entropy": 5.749976015090942,
      "epoch": 2.3368216300330675,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004448540953351844,
      "loss": 4.9645,
      "mean_token_accuracy": 0.20222020149230957,
      "num_tokens": 52105530.0,
      "step": 30035
    },
    {
      "entropy": 5.663449478149414,
      "epoch": 2.3372106594047852,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004448359172842439,
      "loss": 4.8999,
      "mean_token_accuracy": 0.21206051409244536,
      "num_tokens": 52114233.0,
      "step": 30040
    },
    {
      "entropy": 5.685447931289673,
      "epoch": 2.3375996887765025,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044481773665631355,
      "loss": 4.9885,
      "mean_token_accuracy": 0.20781034231185913,
      "num_tokens": 52123561.0,
      "step": 30045
    },
    {
      "entropy": 5.6323071956634525,
      "epoch": 2.33798871814822,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004447995534516695,
      "loss": 4.9061,
      "mean_token_accuracy": 0.21150155663490294,
      "num_tokens": 52132329.0,
      "step": 30050
    },
    {
      "entropy": 5.6993842124938965,
      "epoch": 2.338377747519938,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044478136767058733,
      "loss": 4.9634,
      "mean_token_accuracy": 0.20552536398172377,
      "num_tokens": 52141019.0,
      "step": 30055
    },
    {
      "entropy": 5.644358348846436,
      "epoch": 2.338766776891655,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004447631793133432,
      "loss": 5.0204,
      "mean_token_accuracy": 0.20436219573020936,
      "num_tokens": 52149865.0,
      "step": 30060
    },
    {
      "entropy": 5.701110792160034,
      "epoch": 2.339155806263373,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004447449883802131,
      "loss": 5.0062,
      "mean_token_accuracy": 0.20145619958639144,
      "num_tokens": 52158991.0,
      "step": 30065
    },
    {
      "entropy": 5.647425317764283,
      "epoch": 2.3395448356350905,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044472679487147295,
      "loss": 4.9135,
      "mean_token_accuracy": 0.2094297453761101,
      "num_tokens": 52167141.0,
      "step": 30070
    },
    {
      "entropy": 5.6924670219421385,
      "epoch": 2.339933865006808,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044470859878739877,
      "loss": 4.9345,
      "mean_token_accuracy": 0.2071782350540161,
      "num_tokens": 52175483.0,
      "step": 30075
    },
    {
      "entropy": 5.646374320983886,
      "epoch": 2.3403228943785255,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044469040012826676,
      "loss": 4.9109,
      "mean_token_accuracy": 0.2096250668168068,
      "num_tokens": 52184290.0,
      "step": 30080
    },
    {
      "entropy": 5.638732671737671,
      "epoch": 2.340711923750243,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044467219889435304,
      "loss": 4.8852,
      "mean_token_accuracy": 0.2098342463374138,
      "num_tokens": 52193747.0,
      "step": 30085
    },
    {
      "entropy": 5.6837386131286625,
      "epoch": 2.341100953121961,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004446539950859337,
      "loss": 4.9999,
      "mean_token_accuracy": 0.20458538085222244,
      "num_tokens": 52202744.0,
      "step": 30090
    },
    {
      "entropy": 5.651950359344482,
      "epoch": 2.341489982493678,
      "grad_norm": 1.25,
      "learning_rate": 0.00044463578870328506,
      "loss": 4.9348,
      "mean_token_accuracy": 0.21027553975582122,
      "num_tokens": 52211257.0,
      "step": 30095
    },
    {
      "entropy": 5.728403091430664,
      "epoch": 2.341879011865396,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004446175797466834,
      "loss": 4.9678,
      "mean_token_accuracy": 0.20650721043348313,
      "num_tokens": 52219797.0,
      "step": 30100
    },
    {
      "entropy": 5.7734668254852295,
      "epoch": 2.3422680412371135,
      "grad_norm": 1.125,
      "learning_rate": 0.00044459936821640485,
      "loss": 5.0774,
      "mean_token_accuracy": 0.19797854870557785,
      "num_tokens": 52228157.0,
      "step": 30105
    },
    {
      "entropy": 5.705925703048706,
      "epoch": 2.3426570706088308,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004445811541127258,
      "loss": 5.0185,
      "mean_token_accuracy": 0.21527370661497117,
      "num_tokens": 52237136.0,
      "step": 30110
    },
    {
      "entropy": 5.702061605453491,
      "epoch": 2.3430460999805485,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044456293743592274,
      "loss": 4.99,
      "mean_token_accuracy": 0.206925068795681,
      "num_tokens": 52245966.0,
      "step": 30115
    },
    {
      "entropy": 5.713063144683838,
      "epoch": 2.343435129352266,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044454471818627196,
      "loss": 4.9806,
      "mean_token_accuracy": 0.1964236244559288,
      "num_tokens": 52254782.0,
      "step": 30120
    },
    {
      "entropy": 5.7399979591369625,
      "epoch": 2.343824158723984,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044452649636404995,
      "loss": 5.0446,
      "mean_token_accuracy": 0.19881800264120103,
      "num_tokens": 52264132.0,
      "step": 30125
    },
    {
      "entropy": 5.656832599639893,
      "epoch": 2.344213188095701,
      "grad_norm": 1.125,
      "learning_rate": 0.0004445082719695332,
      "loss": 4.8706,
      "mean_token_accuracy": 0.21263246089220048,
      "num_tokens": 52272763.0,
      "step": 30130
    },
    {
      "entropy": 5.7192402362823485,
      "epoch": 2.344602217467419,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044449004500299815,
      "loss": 5.0244,
      "mean_token_accuracy": 0.2017086401581764,
      "num_tokens": 52281505.0,
      "step": 30135
    },
    {
      "entropy": 5.677187252044678,
      "epoch": 2.3449912468391365,
      "grad_norm": 1.125,
      "learning_rate": 0.00044447181546472146,
      "loss": 4.8646,
      "mean_token_accuracy": 0.20859771519899367,
      "num_tokens": 52289559.0,
      "step": 30140
    },
    {
      "entropy": 5.635884714126587,
      "epoch": 2.3453802762108538,
      "grad_norm": 1.125,
      "learning_rate": 0.00044445358335497975,
      "loss": 4.9482,
      "mean_token_accuracy": 0.21297262609004974,
      "num_tokens": 52298318.0,
      "step": 30145
    },
    {
      "entropy": 5.641231489181519,
      "epoch": 2.3457693055825715,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004444353486740496,
      "loss": 4.95,
      "mean_token_accuracy": 0.20970804542303084,
      "num_tokens": 52307908.0,
      "step": 30150
    },
    {
      "entropy": 5.739203214645386,
      "epoch": 2.346158334954289,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004444171114222078,
      "loss": 4.9753,
      "mean_token_accuracy": 0.2070809409022331,
      "num_tokens": 52315879.0,
      "step": 30155
    },
    {
      "entropy": 5.684037113189698,
      "epoch": 2.346547364326007,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044439887159973096,
      "loss": 4.9737,
      "mean_token_accuracy": 0.19806987643241883,
      "num_tokens": 52324747.0,
      "step": 30160
    },
    {
      "entropy": 5.692579889297486,
      "epoch": 2.346936393697724,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004443806292068958,
      "loss": 4.9842,
      "mean_token_accuracy": 0.20247239172458648,
      "num_tokens": 52333655.0,
      "step": 30165
    },
    {
      "entropy": 5.651612043380737,
      "epoch": 2.347325423069442,
      "grad_norm": 1.25,
      "learning_rate": 0.0004443623842439792,
      "loss": 4.9893,
      "mean_token_accuracy": 0.20781946331262588,
      "num_tokens": 52342111.0,
      "step": 30170
    },
    {
      "entropy": 5.686558628082276,
      "epoch": 2.347714452441159,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000444344136711258,
      "loss": 4.9614,
      "mean_token_accuracy": 0.20871466249227524,
      "num_tokens": 52351390.0,
      "step": 30175
    },
    {
      "entropy": 5.720715665817261,
      "epoch": 2.3481034818128768,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044432588660900905,
      "loss": 4.9768,
      "mean_token_accuracy": 0.20486126989126205,
      "num_tokens": 52361602.0,
      "step": 30180
    },
    {
      "entropy": 5.708740711212158,
      "epoch": 2.3484925111845945,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004443076339375093,
      "loss": 4.9813,
      "mean_token_accuracy": 0.20958708226680756,
      "num_tokens": 52370304.0,
      "step": 30185
    },
    {
      "entropy": 5.662276792526245,
      "epoch": 2.348881540556312,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044428937869703575,
      "loss": 4.8344,
      "mean_token_accuracy": 0.21809003502130508,
      "num_tokens": 52378437.0,
      "step": 30190
    },
    {
      "entropy": 5.693362855911255,
      "epoch": 2.3492705699280294,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044427112088786525,
      "loss": 4.9598,
      "mean_token_accuracy": 0.20908398926258087,
      "num_tokens": 52388413.0,
      "step": 30195
    },
    {
      "entropy": 5.6839806079864506,
      "epoch": 2.349659599299747,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004442528605102749,
      "loss": 4.9402,
      "mean_token_accuracy": 0.2107618659734726,
      "num_tokens": 52396296.0,
      "step": 30200
    },
    {
      "entropy": 5.667974901199341,
      "epoch": 2.350048628671465,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004442345975645418,
      "loss": 4.9067,
      "mean_token_accuracy": 0.20804032385349275,
      "num_tokens": 52403967.0,
      "step": 30205
    },
    {
      "entropy": 5.632362937927246,
      "epoch": 2.350437658043182,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004442163320509431,
      "loss": 4.9666,
      "mean_token_accuracy": 0.21138082593679428,
      "num_tokens": 52413107.0,
      "step": 30210
    },
    {
      "entropy": 5.736687135696411,
      "epoch": 2.3508266874148998,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004441980639697558,
      "loss": 4.9484,
      "mean_token_accuracy": 0.20573663860559463,
      "num_tokens": 52421029.0,
      "step": 30215
    },
    {
      "entropy": 5.739324378967285,
      "epoch": 2.3512157167866174,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004441797933212573,
      "loss": 4.9836,
      "mean_token_accuracy": 0.1985979214310646,
      "num_tokens": 52430538.0,
      "step": 30220
    },
    {
      "entropy": 5.749335145950317,
      "epoch": 2.351604746158335,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004441615201057247,
      "loss": 5.0181,
      "mean_token_accuracy": 0.1978404238820076,
      "num_tokens": 52440013.0,
      "step": 30225
    },
    {
      "entropy": 5.726546001434326,
      "epoch": 2.3519937755300524,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004441432443234352,
      "loss": 5.1182,
      "mean_token_accuracy": 0.1975216507911682,
      "num_tokens": 52449034.0,
      "step": 30230
    },
    {
      "entropy": 5.716649436950684,
      "epoch": 2.35238280490177,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044412496597466626,
      "loss": 5.0311,
      "mean_token_accuracy": 0.19922008663415908,
      "num_tokens": 52457649.0,
      "step": 30235
    },
    {
      "entropy": 5.678254270553589,
      "epoch": 2.352771834273488,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004441066850596951,
      "loss": 4.8897,
      "mean_token_accuracy": 0.2105243906378746,
      "num_tokens": 52465861.0,
      "step": 30240
    },
    {
      "entropy": 5.761814975738526,
      "epoch": 2.353160863645205,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004440884015787992,
      "loss": 5.0262,
      "mean_token_accuracy": 0.20643251091241838,
      "num_tokens": 52474630.0,
      "step": 30245
    },
    {
      "entropy": 5.739989328384399,
      "epoch": 2.3535498930169227,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004440701155322558,
      "loss": 5.0644,
      "mean_token_accuracy": 0.20094598084688187,
      "num_tokens": 52482759.0,
      "step": 30250
    },
    {
      "entropy": 5.718815326690674,
      "epoch": 2.3539389223886404,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004440518269203427,
      "loss": 5.0733,
      "mean_token_accuracy": 0.19313051104545592,
      "num_tokens": 52491839.0,
      "step": 30255
    },
    {
      "entropy": 5.693753242492676,
      "epoch": 2.354327951760358,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044403353574333715,
      "loss": 4.905,
      "mean_token_accuracy": 0.20529360175132752,
      "num_tokens": 52500688.0,
      "step": 30260
    },
    {
      "entropy": 5.741302156448365,
      "epoch": 2.3547169811320754,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044401524200151667,
      "loss": 5.0121,
      "mean_token_accuracy": 0.2053874433040619,
      "num_tokens": 52509031.0,
      "step": 30265
    },
    {
      "entropy": 5.696054697036743,
      "epoch": 2.355106010503793,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044399694569515895,
      "loss": 4.9466,
      "mean_token_accuracy": 0.20526422709226608,
      "num_tokens": 52518337.0,
      "step": 30270
    },
    {
      "entropy": 5.69251594543457,
      "epoch": 2.355495039875511,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044397864682454156,
      "loss": 4.967,
      "mean_token_accuracy": 0.20859042257070542,
      "num_tokens": 52527257.0,
      "step": 30275
    },
    {
      "entropy": 5.737442922592163,
      "epoch": 2.355884069247228,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004439603453899421,
      "loss": 5.0288,
      "mean_token_accuracy": 0.20598723292350768,
      "num_tokens": 52536999.0,
      "step": 30280
    },
    {
      "entropy": 5.740789270401001,
      "epoch": 2.3562730986189457,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004439420413916384,
      "loss": 5.0399,
      "mean_token_accuracy": 0.19617908895015718,
      "num_tokens": 52545975.0,
      "step": 30285
    },
    {
      "entropy": 5.7288837909698485,
      "epoch": 2.3566621279906634,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044392373482990817,
      "loss": 5.0109,
      "mean_token_accuracy": 0.19849975109100343,
      "num_tokens": 52554667.0,
      "step": 30290
    },
    {
      "entropy": 5.706232929229737,
      "epoch": 2.3570511573623807,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004439054257050291,
      "loss": 4.9664,
      "mean_token_accuracy": 0.209281088411808,
      "num_tokens": 52563469.0,
      "step": 30295
    },
    {
      "entropy": 5.654235029220581,
      "epoch": 2.3574401867340984,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004438871140172789,
      "loss": 4.9194,
      "mean_token_accuracy": 0.21019026935100554,
      "num_tokens": 52572038.0,
      "step": 30300
    },
    {
      "entropy": 5.7148150444030765,
      "epoch": 2.357829216105816,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004438687997669357,
      "loss": 4.9918,
      "mean_token_accuracy": 0.19587428122758865,
      "num_tokens": 52579642.0,
      "step": 30305
    },
    {
      "entropy": 5.679110145568847,
      "epoch": 2.3582182454775333,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044385048295427724,
      "loss": 4.9073,
      "mean_token_accuracy": 0.210511015355587,
      "num_tokens": 52587425.0,
      "step": 30310
    },
    {
      "entropy": 5.648802471160889,
      "epoch": 2.358607274849251,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004438321635795814,
      "loss": 4.9288,
      "mean_token_accuracy": 0.21242012828588486,
      "num_tokens": 52596560.0,
      "step": 30315
    },
    {
      "entropy": 5.746258544921875,
      "epoch": 2.3589963042209687,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004438138416431262,
      "loss": 5.0625,
      "mean_token_accuracy": 0.2010079115629196,
      "num_tokens": 52604977.0,
      "step": 30320
    },
    {
      "entropy": 5.728331279754639,
      "epoch": 2.3593853335926864,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044379551714518966,
      "loss": 4.9961,
      "mean_token_accuracy": 0.20059742778539658,
      "num_tokens": 52614069.0,
      "step": 30325
    },
    {
      "entropy": 5.713247442245484,
      "epoch": 2.3597743629644037,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044377719008604984,
      "loss": 4.9914,
      "mean_token_accuracy": 0.20913219898939134,
      "num_tokens": 52623065.0,
      "step": 30330
    },
    {
      "entropy": 5.724383735656739,
      "epoch": 2.3601633923361214,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004437588604659848,
      "loss": 5.0185,
      "mean_token_accuracy": 0.20470532327890395,
      "num_tokens": 52632422.0,
      "step": 30335
    },
    {
      "entropy": 5.723710584640503,
      "epoch": 2.360552421707839,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004437405282852726,
      "loss": 5.0824,
      "mean_token_accuracy": 0.1973904028534889,
      "num_tokens": 52641358.0,
      "step": 30340
    },
    {
      "entropy": 5.722246980667114,
      "epoch": 2.3609414510795563,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044372219354419154,
      "loss": 4.9434,
      "mean_token_accuracy": 0.20737271755933762,
      "num_tokens": 52650011.0,
      "step": 30345
    },
    {
      "entropy": 5.7670676708221436,
      "epoch": 2.361330480451274,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004437038562430197,
      "loss": 5.027,
      "mean_token_accuracy": 0.1991714730858803,
      "num_tokens": 52658111.0,
      "step": 30350
    },
    {
      "entropy": 5.700721693038941,
      "epoch": 2.3617195098229917,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044368551638203536,
      "loss": 4.8905,
      "mean_token_accuracy": 0.2181832015514374,
      "num_tokens": 52666181.0,
      "step": 30355
    },
    {
      "entropy": 5.646384000778198,
      "epoch": 2.3621085391947094,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004436671739615168,
      "loss": 5.0072,
      "mean_token_accuracy": 0.20470435172319412,
      "num_tokens": 52675218.0,
      "step": 30360
    },
    {
      "entropy": 5.718186092376709,
      "epoch": 2.3624975685664267,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044364882898174235,
      "loss": 5.0431,
      "mean_token_accuracy": 0.20208911448717118,
      "num_tokens": 52684254.0,
      "step": 30365
    },
    {
      "entropy": 5.7428874492645265,
      "epoch": 2.3628865979381444,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004436304814429903,
      "loss": 4.9971,
      "mean_token_accuracy": 0.20249738544225693,
      "num_tokens": 52693106.0,
      "step": 30370
    },
    {
      "entropy": 5.756881427764893,
      "epoch": 2.363275627309862,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004436121313455391,
      "loss": 5.0018,
      "mean_token_accuracy": 0.20280634611845016,
      "num_tokens": 52702260.0,
      "step": 30375
    },
    {
      "entropy": 5.732424974441528,
      "epoch": 2.3636646566815793,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004435937786896673,
      "loss": 4.9803,
      "mean_token_accuracy": 0.21154899150133133,
      "num_tokens": 52710546.0,
      "step": 30380
    },
    {
      "entropy": 5.631473684310913,
      "epoch": 2.364053686053297,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044357542347565323,
      "loss": 4.9841,
      "mean_token_accuracy": 0.20053708106279372,
      "num_tokens": 52719197.0,
      "step": 30385
    },
    {
      "entropy": 5.815240859985352,
      "epoch": 2.3644427154250147,
      "grad_norm": 1.25,
      "learning_rate": 0.0004435570657037753,
      "loss": 5.1066,
      "mean_token_accuracy": 0.19827493280172348,
      "num_tokens": 52728242.0,
      "step": 30390
    },
    {
      "entropy": 5.699974489212036,
      "epoch": 2.364831744796732,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004435387053743123,
      "loss": 4.8901,
      "mean_token_accuracy": 0.20970286726951598,
      "num_tokens": 52736943.0,
      "step": 30395
    },
    {
      "entropy": 5.632663536071777,
      "epoch": 2.3652207741684497,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044352034248754265,
      "loss": 4.9427,
      "mean_token_accuracy": 0.2111756071448326,
      "num_tokens": 52744893.0,
      "step": 30400
    },
    {
      "entropy": 5.57286810874939,
      "epoch": 2.3656098035401674,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000443501977043745,
      "loss": 4.8506,
      "mean_token_accuracy": 0.21526092290878296,
      "num_tokens": 52753492.0,
      "step": 30405
    },
    {
      "entropy": 5.711662673950196,
      "epoch": 2.3659988329118846,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004434836090431981,
      "loss": 4.9708,
      "mean_token_accuracy": 0.20484145730733871,
      "num_tokens": 52762598.0,
      "step": 30410
    },
    {
      "entropy": 5.696292686462402,
      "epoch": 2.3663878622836023,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004434652384861806,
      "loss": 5.0026,
      "mean_token_accuracy": 0.20700121372938157,
      "num_tokens": 52770942.0,
      "step": 30415
    },
    {
      "entropy": 5.7142205238342285,
      "epoch": 2.36677689165532,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004434468653729712,
      "loss": 4.9492,
      "mean_token_accuracy": 0.20521363317966462,
      "num_tokens": 52779237.0,
      "step": 30420
    },
    {
      "entropy": 5.748275899887085,
      "epoch": 2.3671659210270377,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044342848970384876,
      "loss": 5.0223,
      "mean_token_accuracy": 0.19939265847206117,
      "num_tokens": 52787721.0,
      "step": 30425
    },
    {
      "entropy": 5.769319868087768,
      "epoch": 2.367554950398755,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044341011147909206,
      "loss": 4.9732,
      "mean_token_accuracy": 0.2015705943107605,
      "num_tokens": 52795765.0,
      "step": 30430
    },
    {
      "entropy": 5.6822107315063475,
      "epoch": 2.3679439797704727,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044339173069897996,
      "loss": 4.8966,
      "mean_token_accuracy": 0.21289556473493576,
      "num_tokens": 52804104.0,
      "step": 30435
    },
    {
      "entropy": 5.6194147109985355,
      "epoch": 2.3683330091421904,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044337334736379143,
      "loss": 4.9014,
      "mean_token_accuracy": 0.20616833716630936,
      "num_tokens": 52812817.0,
      "step": 30440
    },
    {
      "entropy": 5.645487642288208,
      "epoch": 2.3687220385139076,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004433549614738053,
      "loss": 4.9317,
      "mean_token_accuracy": 0.20428660064935683,
      "num_tokens": 52821924.0,
      "step": 30445
    },
    {
      "entropy": 5.702040910720825,
      "epoch": 2.3691110678856253,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044333657302930056,
      "loss": 4.9158,
      "mean_token_accuracy": 0.21273674964904785,
      "num_tokens": 52830943.0,
      "step": 30450
    },
    {
      "entropy": 5.724423122406006,
      "epoch": 2.369500097257343,
      "grad_norm": 1.125,
      "learning_rate": 0.0004433181820305564,
      "loss": 5.0501,
      "mean_token_accuracy": 0.20478523671627044,
      "num_tokens": 52839697.0,
      "step": 30455
    },
    {
      "entropy": 5.703110790252685,
      "epoch": 2.3698891266290607,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044329978847785156,
      "loss": 4.9908,
      "mean_token_accuracy": 0.20023351460695266,
      "num_tokens": 52848713.0,
      "step": 30460
    },
    {
      "entropy": 5.767563533782959,
      "epoch": 2.370278156000778,
      "grad_norm": 1.25,
      "learning_rate": 0.0004432813923714654,
      "loss": 5.0715,
      "mean_token_accuracy": 0.19697457998991014,
      "num_tokens": 52856852.0,
      "step": 30465
    },
    {
      "entropy": 5.643336963653565,
      "epoch": 2.3706671853724957,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044326299371167695,
      "loss": 4.9003,
      "mean_token_accuracy": 0.20981917083263396,
      "num_tokens": 52865235.0,
      "step": 30470
    },
    {
      "entropy": 5.689011144638061,
      "epoch": 2.3710562147442134,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044324459249876536,
      "loss": 4.91,
      "mean_token_accuracy": 0.20584203749895097,
      "num_tokens": 52873454.0,
      "step": 30475
    },
    {
      "entropy": 5.669601202011108,
      "epoch": 2.3714452441159306,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004432261887330099,
      "loss": 4.9497,
      "mean_token_accuracy": 0.21177210807800292,
      "num_tokens": 52882551.0,
      "step": 30480
    },
    {
      "entropy": 5.74150128364563,
      "epoch": 2.3718342734876483,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004432077824146898,
      "loss": 5.012,
      "mean_token_accuracy": 0.19828751385211946,
      "num_tokens": 52891145.0,
      "step": 30485
    },
    {
      "entropy": 5.700848913192749,
      "epoch": 2.372223302859366,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004431893735440843,
      "loss": 4.9671,
      "mean_token_accuracy": 0.20126706957817078,
      "num_tokens": 52899486.0,
      "step": 30490
    },
    {
      "entropy": 5.666141605377197,
      "epoch": 2.3726123322310833,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004431709621214727,
      "loss": 4.9471,
      "mean_token_accuracy": 0.20314858555793763,
      "num_tokens": 52907896.0,
      "step": 30495
    },
    {
      "entropy": 5.683284664154053,
      "epoch": 2.373001361602801,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044315254814713455,
      "loss": 4.9707,
      "mean_token_accuracy": 0.20181227922439576,
      "num_tokens": 52916489.0,
      "step": 30500
    },
    {
      "entropy": 5.686014890670776,
      "epoch": 2.3733903909745186,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044313413162134894,
      "loss": 4.9905,
      "mean_token_accuracy": 0.20366300493478776,
      "num_tokens": 52924559.0,
      "step": 30505
    },
    {
      "entropy": 5.700543832778931,
      "epoch": 2.373779420346236,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004431157125443958,
      "loss": 5.0158,
      "mean_token_accuracy": 0.20747122317552566,
      "num_tokens": 52934064.0,
      "step": 30510
    },
    {
      "entropy": 5.704973316192627,
      "epoch": 2.3741684497179536,
      "grad_norm": 1.21875,
      "learning_rate": 0.000443097290916554,
      "loss": 4.976,
      "mean_token_accuracy": 0.20694274008274077,
      "num_tokens": 52943283.0,
      "step": 30515
    },
    {
      "entropy": 5.628741025924683,
      "epoch": 2.3745574790896713,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004430788667381035,
      "loss": 4.8799,
      "mean_token_accuracy": 0.21135154962539673,
      "num_tokens": 52951694.0,
      "step": 30520
    },
    {
      "entropy": 5.686148166656494,
      "epoch": 2.374946508461389,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044306044000932374,
      "loss": 4.9895,
      "mean_token_accuracy": 0.20255753695964812,
      "num_tokens": 52959850.0,
      "step": 30525
    },
    {
      "entropy": 5.761051607131958,
      "epoch": 2.3753355378331062,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004430420107304943,
      "loss": 5.1363,
      "mean_token_accuracy": 0.19352102130651475,
      "num_tokens": 52969378.0,
      "step": 30530
    },
    {
      "entropy": 5.728963994979859,
      "epoch": 2.375724567204824,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00044302357890189475,
      "loss": 4.9784,
      "mean_token_accuracy": 0.21571914255619049,
      "num_tokens": 52977178.0,
      "step": 30535
    },
    {
      "entropy": 5.770847463607788,
      "epoch": 2.3761135965765416,
      "grad_norm": 1.5,
      "learning_rate": 0.0004430051445238049,
      "loss": 5.0036,
      "mean_token_accuracy": 0.20524838864803313,
      "num_tokens": 52986035.0,
      "step": 30540
    },
    {
      "entropy": 5.659987306594848,
      "epoch": 2.376502625948259,
      "grad_norm": 1.25,
      "learning_rate": 0.00044298670759650426,
      "loss": 5.0304,
      "mean_token_accuracy": 0.20628440231084824,
      "num_tokens": 52995399.0,
      "step": 30545
    },
    {
      "entropy": 5.700040197372436,
      "epoch": 2.3768916553199766,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004429682681202728,
      "loss": 4.9666,
      "mean_token_accuracy": 0.2097300574183464,
      "num_tokens": 53004500.0,
      "step": 30550
    },
    {
      "entropy": 5.77179970741272,
      "epoch": 2.3772806846916943,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044294982609539027,
      "loss": 5.0765,
      "mean_token_accuracy": 0.20099400132894515,
      "num_tokens": 53014161.0,
      "step": 30555
    },
    {
      "entropy": 5.660857105255127,
      "epoch": 2.377669714063412,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004429313815221363,
      "loss": 4.8553,
      "mean_token_accuracy": 0.21293341517448425,
      "num_tokens": 53022560.0,
      "step": 30560
    },
    {
      "entropy": 5.682220458984375,
      "epoch": 2.3780587434351292,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044291293440079107,
      "loss": 5.0081,
      "mean_token_accuracy": 0.20121063739061357,
      "num_tokens": 53031194.0,
      "step": 30565
    },
    {
      "entropy": 5.706054306030273,
      "epoch": 2.378447772806847,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004428944847316342,
      "loss": 5.018,
      "mean_token_accuracy": 0.20237772762775422,
      "num_tokens": 53039512.0,
      "step": 30570
    },
    {
      "entropy": 5.730585670471191,
      "epoch": 2.3788368021785646,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004428760325149458,
      "loss": 4.9929,
      "mean_token_accuracy": 0.20839864611625672,
      "num_tokens": 53048579.0,
      "step": 30575
    },
    {
      "entropy": 5.765998888015747,
      "epoch": 2.379225831550282,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044285757775100584,
      "loss": 5.1194,
      "mean_token_accuracy": 0.19846058487892151,
      "num_tokens": 53057770.0,
      "step": 30580
    },
    {
      "entropy": 5.681022644042969,
      "epoch": 2.3796148609219996,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044283912044009436,
      "loss": 4.9895,
      "mean_token_accuracy": 0.20465605407953263,
      "num_tokens": 53067127.0,
      "step": 30585
    },
    {
      "entropy": 5.73515682220459,
      "epoch": 2.3800038902937173,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044282066058249123,
      "loss": 4.9602,
      "mean_token_accuracy": 0.2102888137102127,
      "num_tokens": 53076351.0,
      "step": 30590
    },
    {
      "entropy": 5.725416755676269,
      "epoch": 2.380392919665435,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004428021981784768,
      "loss": 5.0101,
      "mean_token_accuracy": 0.20604050308465957,
      "num_tokens": 53085472.0,
      "step": 30595
    },
    {
      "entropy": 5.697421455383301,
      "epoch": 2.3807819490371522,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044278373322833106,
      "loss": 4.952,
      "mean_token_accuracy": 0.20982952862977983,
      "num_tokens": 53094181.0,
      "step": 30600
    },
    {
      "entropy": 5.668222522735595,
      "epoch": 2.38117097840887,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044276526573233416,
      "loss": 5.0291,
      "mean_token_accuracy": 0.19204013794660568,
      "num_tokens": 53103421.0,
      "step": 30605
    },
    {
      "entropy": 5.729308080673218,
      "epoch": 2.381560007780587,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004427467956907664,
      "loss": 4.9504,
      "mean_token_accuracy": 0.20751901268959044,
      "num_tokens": 53112058.0,
      "step": 30610
    },
    {
      "entropy": 5.772487497329712,
      "epoch": 2.381949037152305,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00044272832310390814,
      "loss": 5.0287,
      "mean_token_accuracy": 0.20354922115802765,
      "num_tokens": 53119755.0,
      "step": 30615
    },
    {
      "entropy": 5.6213428497314455,
      "epoch": 2.3823380665240226,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004427098479720394,
      "loss": 4.9162,
      "mean_token_accuracy": 0.21654124408960343,
      "num_tokens": 53128529.0,
      "step": 30620
    },
    {
      "entropy": 5.661910820007324,
      "epoch": 2.3827270958957403,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044269137029544073,
      "loss": 4.932,
      "mean_token_accuracy": 0.20995598733425141,
      "num_tokens": 53137594.0,
      "step": 30625
    },
    {
      "entropy": 5.752660083770752,
      "epoch": 2.3831161252674575,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004426728900743924,
      "loss": 4.9965,
      "mean_token_accuracy": 0.20239504873752595,
      "num_tokens": 53146521.0,
      "step": 30630
    },
    {
      "entropy": 5.720496892929077,
      "epoch": 2.3835051546391752,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044265440730917484,
      "loss": 4.9535,
      "mean_token_accuracy": 0.20930489897727966,
      "num_tokens": 53154832.0,
      "step": 30635
    },
    {
      "entropy": 5.693125152587891,
      "epoch": 2.383894184010893,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044263592200006845,
      "loss": 5.0444,
      "mean_token_accuracy": 0.19915098547935486,
      "num_tokens": 53163889.0,
      "step": 30640
    },
    {
      "entropy": 5.6846351146698,
      "epoch": 2.38428321338261,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044261743414735383,
      "loss": 4.991,
      "mean_token_accuracy": 0.20873599350452424,
      "num_tokens": 53172688.0,
      "step": 30645
    },
    {
      "entropy": 5.753062152862549,
      "epoch": 2.384672242754328,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004425989437513114,
      "loss": 4.9606,
      "mean_token_accuracy": 0.20272178202867508,
      "num_tokens": 53180742.0,
      "step": 30650
    },
    {
      "entropy": 5.638766622543335,
      "epoch": 2.3850612721260456,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004425804508122218,
      "loss": 4.942,
      "mean_token_accuracy": 0.2118618071079254,
      "num_tokens": 53188716.0,
      "step": 30655
    },
    {
      "entropy": 5.750921773910522,
      "epoch": 2.3854503014977633,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044256195533036566,
      "loss": 5.0843,
      "mean_token_accuracy": 0.19754649549722672,
      "num_tokens": 53197420.0,
      "step": 30660
    },
    {
      "entropy": 5.701545572280883,
      "epoch": 2.3858393308694805,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044254345730602345,
      "loss": 4.9989,
      "mean_token_accuracy": 0.20563812851905822,
      "num_tokens": 53205927.0,
      "step": 30665
    },
    {
      "entropy": 5.714105749130249,
      "epoch": 2.386228360241198,
      "grad_norm": 1.25,
      "learning_rate": 0.0004425249567394759,
      "loss": 4.9642,
      "mean_token_accuracy": 0.2086802288889885,
      "num_tokens": 53214793.0,
      "step": 30670
    },
    {
      "entropy": 5.691190099716186,
      "epoch": 2.386617389612916,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044250645363100386,
      "loss": 4.9539,
      "mean_token_accuracy": 0.20721821039915084,
      "num_tokens": 53223429.0,
      "step": 30675
    },
    {
      "entropy": 5.70162410736084,
      "epoch": 2.387006418984633,
      "grad_norm": 1.25,
      "learning_rate": 0.00044248794798088805,
      "loss": 4.8592,
      "mean_token_accuracy": 0.21638955920934677,
      "num_tokens": 53231697.0,
      "step": 30680
    },
    {
      "entropy": 5.73566780090332,
      "epoch": 2.387395448356351,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004424694397894091,
      "loss": 4.9859,
      "mean_token_accuracy": 0.20582906156778336,
      "num_tokens": 53239318.0,
      "step": 30685
    },
    {
      "entropy": 5.685762357711792,
      "epoch": 2.3877844777280686,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044245092905684794,
      "loss": 5.0778,
      "mean_token_accuracy": 0.1982678458094597,
      "num_tokens": 53248079.0,
      "step": 30690
    },
    {
      "entropy": 5.717021560668945,
      "epoch": 2.3881735070997863,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044243241578348553,
      "loss": 5.0244,
      "mean_token_accuracy": 0.20175449699163436,
      "num_tokens": 53256856.0,
      "step": 30695
    },
    {
      "entropy": 5.768334436416626,
      "epoch": 2.3885625364715035,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004424138999696027,
      "loss": 5.0515,
      "mean_token_accuracy": 0.20117962807416917,
      "num_tokens": 53265810.0,
      "step": 30700
    },
    {
      "entropy": 5.783043050765992,
      "epoch": 2.388951565843221,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004423953816154804,
      "loss": 5.0444,
      "mean_token_accuracy": 0.20439288020133972,
      "num_tokens": 53274485.0,
      "step": 30705
    },
    {
      "entropy": 5.722244024276733,
      "epoch": 2.389340595214939,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00044237686072139967,
      "loss": 4.9164,
      "mean_token_accuracy": 0.2052573785185814,
      "num_tokens": 53282322.0,
      "step": 30710
    },
    {
      "entropy": 5.753564834594727,
      "epoch": 2.389729624586656,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004423583372876414,
      "loss": 5.0298,
      "mean_token_accuracy": 0.20363051146268846,
      "num_tokens": 53291868.0,
      "step": 30715
    },
    {
      "entropy": 5.680089950561523,
      "epoch": 2.390118653958374,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044233981131448677,
      "loss": 4.9122,
      "mean_token_accuracy": 0.21827230006456375,
      "num_tokens": 53300654.0,
      "step": 30720
    },
    {
      "entropy": 5.664479923248291,
      "epoch": 2.3905076833300916,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044232128280221683,
      "loss": 4.8176,
      "mean_token_accuracy": 0.2174416184425354,
      "num_tokens": 53309178.0,
      "step": 30725
    },
    {
      "entropy": 5.704671621322632,
      "epoch": 2.390896712701809,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004423027517511128,
      "loss": 5.0131,
      "mean_token_accuracy": 0.20116270780563356,
      "num_tokens": 53317536.0,
      "step": 30730
    },
    {
      "entropy": 5.679006052017212,
      "epoch": 2.3912857420735265,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044228421816145573,
      "loss": 4.9769,
      "mean_token_accuracy": 0.20146695673465728,
      "num_tokens": 53325942.0,
      "step": 30735
    },
    {
      "entropy": 5.6894450187683105,
      "epoch": 2.391674771445244,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044226568203352694,
      "loss": 4.9631,
      "mean_token_accuracy": 0.20810642540454866,
      "num_tokens": 53335124.0,
      "step": 30740
    },
    {
      "entropy": 5.700615787506104,
      "epoch": 2.3920638008169615,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044224714336760766,
      "loss": 5.0522,
      "mean_token_accuracy": 0.20412603616714478,
      "num_tokens": 53344222.0,
      "step": 30745
    },
    {
      "entropy": 5.711447477340698,
      "epoch": 2.392452830188679,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004422286021639792,
      "loss": 4.963,
      "mean_token_accuracy": 0.20254070311784744,
      "num_tokens": 53352639.0,
      "step": 30750
    },
    {
      "entropy": 5.701373434066772,
      "epoch": 2.392841859560397,
      "grad_norm": 1.25,
      "learning_rate": 0.0004422100584229228,
      "loss": 4.9783,
      "mean_token_accuracy": 0.20497527420520784,
      "num_tokens": 53361357.0,
      "step": 30755
    },
    {
      "entropy": 5.741417264938354,
      "epoch": 2.3932308889321146,
      "grad_norm": 1.34375,
      "learning_rate": 0.00044219151214472,
      "loss": 5.006,
      "mean_token_accuracy": 0.21281603425741197,
      "num_tokens": 53369680.0,
      "step": 30760
    },
    {
      "entropy": 5.716237258911133,
      "epoch": 2.393619918303832,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004421729633296521,
      "loss": 4.9828,
      "mean_token_accuracy": 0.20617086738348006,
      "num_tokens": 53378619.0,
      "step": 30765
    },
    {
      "entropy": 5.717056131362915,
      "epoch": 2.3940089476755495,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044215441197800054,
      "loss": 4.9567,
      "mean_token_accuracy": 0.20952197164297104,
      "num_tokens": 53387630.0,
      "step": 30770
    },
    {
      "entropy": 5.701000595092774,
      "epoch": 2.394397977047267,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044213585809004685,
      "loss": 4.9769,
      "mean_token_accuracy": 0.1995824471116066,
      "num_tokens": 53396696.0,
      "step": 30775
    },
    {
      "entropy": 5.712011909484863,
      "epoch": 2.3947870064189845,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004421173016660725,
      "loss": 4.9794,
      "mean_token_accuracy": 0.20360222905874253,
      "num_tokens": 53405097.0,
      "step": 30780
    },
    {
      "entropy": 5.6822761535644535,
      "epoch": 2.395176035790702,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004420987427063592,
      "loss": 4.9502,
      "mean_token_accuracy": 0.20191132724285127,
      "num_tokens": 53413670.0,
      "step": 30785
    },
    {
      "entropy": 5.687037324905395,
      "epoch": 2.39556506516242,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044208018121118833,
      "loss": 5.0526,
      "mean_token_accuracy": 0.19744781255722046,
      "num_tokens": 53422370.0,
      "step": 30790
    },
    {
      "entropy": 5.760198593139648,
      "epoch": 2.3959540945341375,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044206161718084164,
      "loss": 5.0711,
      "mean_token_accuracy": 0.1970564067363739,
      "num_tokens": 53431607.0,
      "step": 30795
    },
    {
      "entropy": 5.718116712570191,
      "epoch": 2.396343123905855,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004420430506156009,
      "loss": 4.8744,
      "mean_token_accuracy": 0.21628762483596803,
      "num_tokens": 53440042.0,
      "step": 30800
    },
    {
      "entropy": 5.627287817001343,
      "epoch": 2.3967321532775725,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044202448151574764,
      "loss": 4.9664,
      "mean_token_accuracy": 0.21037106812000275,
      "num_tokens": 53448780.0,
      "step": 30805
    },
    {
      "entropy": 5.711191177368164,
      "epoch": 2.39712118264929,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004420059098815638,
      "loss": 5.0708,
      "mean_token_accuracy": 0.19735982716083528,
      "num_tokens": 53457598.0,
      "step": 30810
    },
    {
      "entropy": 5.778003263473511,
      "epoch": 2.3975102120210074,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004419873357133311,
      "loss": 5.0633,
      "mean_token_accuracy": 0.20849969387054443,
      "num_tokens": 53466387.0,
      "step": 30815
    },
    {
      "entropy": 5.751691150665283,
      "epoch": 2.397899241392725,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004419687590113313,
      "loss": 5.0109,
      "mean_token_accuracy": 0.2041358932852745,
      "num_tokens": 53475032.0,
      "step": 30820
    },
    {
      "entropy": 5.742014694213867,
      "epoch": 2.398288270764443,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044195017977584637,
      "loss": 5.0257,
      "mean_token_accuracy": 0.21066032350063324,
      "num_tokens": 53483142.0,
      "step": 30825
    },
    {
      "entropy": 5.724196052551269,
      "epoch": 2.39867730013616,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044193159800715823,
      "loss": 5.0848,
      "mean_token_accuracy": 0.20259960889816284,
      "num_tokens": 53492820.0,
      "step": 30830
    },
    {
      "entropy": 5.7266068935394285,
      "epoch": 2.399066329507878,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044191301370554874,
      "loss": 5.006,
      "mean_token_accuracy": 0.1978727951645851,
      "num_tokens": 53502018.0,
      "step": 30835
    },
    {
      "entropy": 5.719934558868408,
      "epoch": 2.3994553588795955,
      "grad_norm": 1.25,
      "learning_rate": 0.00044189442687129994,
      "loss": 4.9399,
      "mean_token_accuracy": 0.20603836625814437,
      "num_tokens": 53510287.0,
      "step": 30840
    },
    {
      "entropy": 5.687133121490478,
      "epoch": 2.3998443882513127,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004418758375046939,
      "loss": 4.969,
      "mean_token_accuracy": 0.20012108385562896,
      "num_tokens": 53519042.0,
      "step": 30845
    },
    {
      "entropy": 5.757483720779419,
      "epoch": 2.4002334176230304,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044185724560601267,
      "loss": 4.983,
      "mean_token_accuracy": 0.2065669134259224,
      "num_tokens": 53527678.0,
      "step": 30850
    },
    {
      "entropy": 5.780759620666504,
      "epoch": 2.400622446994748,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004418386511755382,
      "loss": 5.0317,
      "mean_token_accuracy": 0.20099728107452391,
      "num_tokens": 53536371.0,
      "step": 30855
    },
    {
      "entropy": 5.6850371837615965,
      "epoch": 2.401011476366466,
      "grad_norm": 1.125,
      "learning_rate": 0.0004418200542135528,
      "loss": 5.0448,
      "mean_token_accuracy": 0.20321519076824188,
      "num_tokens": 53545534.0,
      "step": 30860
    },
    {
      "entropy": 5.685451984405518,
      "epoch": 2.401400505738183,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004418014547203386,
      "loss": 4.9446,
      "mean_token_accuracy": 0.2088079497218132,
      "num_tokens": 53554904.0,
      "step": 30865
    },
    {
      "entropy": 5.77572603225708,
      "epoch": 2.401789535109901,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004417828526961778,
      "loss": 4.9722,
      "mean_token_accuracy": 0.20978059321641923,
      "num_tokens": 53563359.0,
      "step": 30870
    },
    {
      "entropy": 5.703126001358032,
      "epoch": 2.4021785644816185,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00044176424814135266,
      "loss": 5.0708,
      "mean_token_accuracy": 0.20332254618406295,
      "num_tokens": 53573428.0,
      "step": 30875
    },
    {
      "entropy": 5.731616449356079,
      "epoch": 2.4025675938533357,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004417456410561455,
      "loss": 5.0697,
      "mean_token_accuracy": 0.19640397429466247,
      "num_tokens": 53581719.0,
      "step": 30880
    },
    {
      "entropy": 5.79337248802185,
      "epoch": 2.4029566232250534,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004417270314408387,
      "loss": 5.1021,
      "mean_token_accuracy": 0.19791825413703917,
      "num_tokens": 53589715.0,
      "step": 30885
    },
    {
      "entropy": 5.715180540084839,
      "epoch": 2.403345652596771,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044170841929571454,
      "loss": 4.9869,
      "mean_token_accuracy": 0.20632381588220597,
      "num_tokens": 53598273.0,
      "step": 30890
    },
    {
      "entropy": 5.689082241058349,
      "epoch": 2.403734681968489,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044168980462105543,
      "loss": 5.0069,
      "mean_token_accuracy": 0.20256789326667785,
      "num_tokens": 53607824.0,
      "step": 30895
    },
    {
      "entropy": 5.716005563735962,
      "epoch": 2.404123711340206,
      "grad_norm": 1.125,
      "learning_rate": 0.0004416711874171439,
      "loss": 4.9322,
      "mean_token_accuracy": 0.21018763929605483,
      "num_tokens": 53616734.0,
      "step": 30900
    },
    {
      "entropy": 5.691632413864136,
      "epoch": 2.404512740711924,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004416525676842624,
      "loss": 4.9593,
      "mean_token_accuracy": 0.2085067391395569,
      "num_tokens": 53625877.0,
      "step": 30905
    },
    {
      "entropy": 5.72473373413086,
      "epoch": 2.4049017700836415,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004416339454226933,
      "loss": 5.0388,
      "mean_token_accuracy": 0.20700467973947526,
      "num_tokens": 53634872.0,
      "step": 30910
    },
    {
      "entropy": 5.670379304885865,
      "epoch": 2.4052907994553587,
      "grad_norm": 1.25,
      "learning_rate": 0.0004416153206327194,
      "loss": 4.9682,
      "mean_token_accuracy": 0.20880395472049712,
      "num_tokens": 53643916.0,
      "step": 30915
    },
    {
      "entropy": 5.7193577766418455,
      "epoch": 2.4056798288270764,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044159669331462326,
      "loss": 5.0003,
      "mean_token_accuracy": 0.20049893856048584,
      "num_tokens": 53652496.0,
      "step": 30920
    },
    {
      "entropy": 5.677342557907105,
      "epoch": 2.406068858198794,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044157806346868737,
      "loss": 4.9666,
      "mean_token_accuracy": 0.20404890477657317,
      "num_tokens": 53661191.0,
      "step": 30925
    },
    {
      "entropy": 5.70556378364563,
      "epoch": 2.406457887570512,
      "grad_norm": 1.25,
      "learning_rate": 0.00044155943109519454,
      "loss": 4.9822,
      "mean_token_accuracy": 0.20777849704027176,
      "num_tokens": 53669520.0,
      "step": 30930
    },
    {
      "entropy": 5.718815279006958,
      "epoch": 2.406846916942229,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004415407961944274,
      "loss": 5.0113,
      "mean_token_accuracy": 0.1981296107172966,
      "num_tokens": 53678398.0,
      "step": 30935
    },
    {
      "entropy": 5.7146430015563965,
      "epoch": 2.4072359463139468,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044152215876666883,
      "loss": 5.021,
      "mean_token_accuracy": 0.20283095687627792,
      "num_tokens": 53687870.0,
      "step": 30940
    },
    {
      "entropy": 5.7735680103302,
      "epoch": 2.407624975685664,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004415035188122016,
      "loss": 4.9795,
      "mean_token_accuracy": 0.20813732743263244,
      "num_tokens": 53697335.0,
      "step": 30945
    },
    {
      "entropy": 5.645122051239014,
      "epoch": 2.4080140050573817,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044148487633130837,
      "loss": 4.8915,
      "mean_token_accuracy": 0.2106337070465088,
      "num_tokens": 53705457.0,
      "step": 30950
    },
    {
      "entropy": 5.65149040222168,
      "epoch": 2.4084030344290994,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044146623132427213,
      "loss": 4.8061,
      "mean_token_accuracy": 0.21359836161136628,
      "num_tokens": 53714092.0,
      "step": 30955
    },
    {
      "entropy": 5.7227867603302,
      "epoch": 2.408792063800817,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044144758379137584,
      "loss": 5.0019,
      "mean_token_accuracy": 0.20538598746061326,
      "num_tokens": 53722972.0,
      "step": 30960
    },
    {
      "entropy": 5.738888883590699,
      "epoch": 2.4091810931725344,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044142893373290236,
      "loss": 4.9267,
      "mean_token_accuracy": 0.20870775282382964,
      "num_tokens": 53730842.0,
      "step": 30965
    },
    {
      "entropy": 5.73601770401001,
      "epoch": 2.409570122544252,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004414102811491348,
      "loss": 5.0205,
      "mean_token_accuracy": 0.20271632671356202,
      "num_tokens": 53740241.0,
      "step": 30970
    },
    {
      "entropy": 5.6587793827056885,
      "epoch": 2.4099591519159698,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044139162604035597,
      "loss": 4.9367,
      "mean_token_accuracy": 0.2048289105296135,
      "num_tokens": 53749653.0,
      "step": 30975
    },
    {
      "entropy": 5.7066967487335205,
      "epoch": 2.410348181287687,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044137296840684917,
      "loss": 4.9414,
      "mean_token_accuracy": 0.21393396854400634,
      "num_tokens": 53758161.0,
      "step": 30980
    },
    {
      "entropy": 5.6769123554229735,
      "epoch": 2.4107372106594047,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004413543082488973,
      "loss": 4.8722,
      "mean_token_accuracy": 0.22859333008527755,
      "num_tokens": 53766345.0,
      "step": 30985
    },
    {
      "entropy": 5.66603684425354,
      "epoch": 2.4111262400311224,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004413356455667836,
      "loss": 4.9099,
      "mean_token_accuracy": 0.21208098530769348,
      "num_tokens": 53774725.0,
      "step": 30990
    },
    {
      "entropy": 5.657475090026855,
      "epoch": 2.41151526940284,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004413169803607913,
      "loss": 4.941,
      "mean_token_accuracy": 0.20936129689216615,
      "num_tokens": 53783386.0,
      "step": 30995
    },
    {
      "entropy": 5.717900371551513,
      "epoch": 2.4119042987745574,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044129831263120344,
      "loss": 4.9491,
      "mean_token_accuracy": 0.2021641805768013,
      "num_tokens": 53791167.0,
      "step": 31000
    },
    {
      "entropy": 5.732458019256592,
      "epoch": 2.412293328146275,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004412796423783034,
      "loss": 4.9848,
      "mean_token_accuracy": 0.197972309589386,
      "num_tokens": 53800227.0,
      "step": 31005
    },
    {
      "entropy": 5.6941237449646,
      "epoch": 2.4126823575179928,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004412609696023745,
      "loss": 4.9738,
      "mean_token_accuracy": 0.2088466465473175,
      "num_tokens": 53808591.0,
      "step": 31010
    },
    {
      "entropy": 5.786401033401489,
      "epoch": 2.41307138688971,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004412422943037,
      "loss": 5.0577,
      "mean_token_accuracy": 0.203326053917408,
      "num_tokens": 53817738.0,
      "step": 31015
    },
    {
      "entropy": 5.720012855529785,
      "epoch": 2.4134604162614277,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044122361648256327,
      "loss": 5.0019,
      "mean_token_accuracy": 0.2086169183254242,
      "num_tokens": 53826919.0,
      "step": 31020
    },
    {
      "entropy": 5.684502601623535,
      "epoch": 2.4138494456331454,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044120493613924766,
      "loss": 4.9372,
      "mean_token_accuracy": 0.20957144796848298,
      "num_tokens": 53835415.0,
      "step": 31025
    },
    {
      "entropy": 5.72207498550415,
      "epoch": 2.414238475004863,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044118625327403677,
      "loss": 5.0242,
      "mean_token_accuracy": 0.20831262469291686,
      "num_tokens": 53844992.0,
      "step": 31030
    },
    {
      "entropy": 5.726292324066162,
      "epoch": 2.4146275043765804,
      "grad_norm": 1.265625,
      "learning_rate": 0.000441167567887214,
      "loss": 4.9882,
      "mean_token_accuracy": 0.20181672573089598,
      "num_tokens": 53853864.0,
      "step": 31035
    },
    {
      "entropy": 5.743131256103515,
      "epoch": 2.415016533748298,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004411488799790629,
      "loss": 4.9637,
      "mean_token_accuracy": 0.2081820011138916,
      "num_tokens": 53863021.0,
      "step": 31040
    },
    {
      "entropy": 5.721627807617187,
      "epoch": 2.4154055631200158,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044113018954986693,
      "loss": 5.0264,
      "mean_token_accuracy": 0.19995018988847732,
      "num_tokens": 53873259.0,
      "step": 31045
    },
    {
      "entropy": 5.6557399272918705,
      "epoch": 2.415794592491733,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004411114965999098,
      "loss": 4.9444,
      "mean_token_accuracy": 0.2107873111963272,
      "num_tokens": 53881861.0,
      "step": 31050
    },
    {
      "entropy": 5.723413944244385,
      "epoch": 2.4161836218634507,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000441092801129475,
      "loss": 5.0109,
      "mean_token_accuracy": 0.20508746504783631,
      "num_tokens": 53890852.0,
      "step": 31055
    },
    {
      "entropy": 5.682277202606201,
      "epoch": 2.4165726512351684,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044107410313884643,
      "loss": 4.8976,
      "mean_token_accuracy": 0.21048241406679152,
      "num_tokens": 53899452.0,
      "step": 31060
    },
    {
      "entropy": 5.732623815536499,
      "epoch": 2.4169616806068857,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044105540262830763,
      "loss": 5.0769,
      "mean_token_accuracy": 0.19938639402389527,
      "num_tokens": 53908458.0,
      "step": 31065
    },
    {
      "entropy": 5.719205331802368,
      "epoch": 2.4173507099786034,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004410366995981424,
      "loss": 5.0005,
      "mean_token_accuracy": 0.20371757745742797,
      "num_tokens": 53917998.0,
      "step": 31070
    },
    {
      "entropy": 5.775143957138061,
      "epoch": 2.417739739350321,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044101799404863457,
      "loss": 5.0979,
      "mean_token_accuracy": 0.19430523961782456,
      "num_tokens": 53926320.0,
      "step": 31075
    },
    {
      "entropy": 5.723098707199097,
      "epoch": 2.4181287687220383,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004409992859800679,
      "loss": 4.9929,
      "mean_token_accuracy": 0.19911582469940187,
      "num_tokens": 53934723.0,
      "step": 31080
    },
    {
      "entropy": 5.782225894927978,
      "epoch": 2.418517798093756,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004409805753927263,
      "loss": 5.0764,
      "mean_token_accuracy": 0.19909935891628266,
      "num_tokens": 53943125.0,
      "step": 31085
    },
    {
      "entropy": 5.7472423076629635,
      "epoch": 2.4189068274654737,
      "grad_norm": 1.25,
      "learning_rate": 0.0004409618622868936,
      "loss": 4.9975,
      "mean_token_accuracy": 0.1994920939207077,
      "num_tokens": 53951781.0,
      "step": 31090
    },
    {
      "entropy": 5.683186769485474,
      "epoch": 2.4192958568371914,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00044094314666285385,
      "loss": 4.982,
      "mean_token_accuracy": 0.20459204614162446,
      "num_tokens": 53960556.0,
      "step": 31095
    },
    {
      "entropy": 5.660785436630249,
      "epoch": 2.4196848862089086,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044092442852089095,
      "loss": 4.8982,
      "mean_token_accuracy": 0.20862288475036622,
      "num_tokens": 53969075.0,
      "step": 31100
    },
    {
      "entropy": 5.79287371635437,
      "epoch": 2.4200739155806263,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004409057078612889,
      "loss": 5.1386,
      "mean_token_accuracy": 0.19790823012590408,
      "num_tokens": 53978203.0,
      "step": 31105
    },
    {
      "entropy": 5.744794940948486,
      "epoch": 2.420462944952344,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044088698468433194,
      "loss": 4.9763,
      "mean_token_accuracy": 0.20337577015161515,
      "num_tokens": 53986935.0,
      "step": 31110
    },
    {
      "entropy": 5.725831842422485,
      "epoch": 2.4208519743240613,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044086825899030383,
      "loss": 4.9865,
      "mean_token_accuracy": 0.19874823838472366,
      "num_tokens": 53996167.0,
      "step": 31115
    },
    {
      "entropy": 5.687120246887207,
      "epoch": 2.421241003695779,
      "grad_norm": 1.171875,
      "learning_rate": 0.000440849530779489,
      "loss": 4.9874,
      "mean_token_accuracy": 0.20798343122005464,
      "num_tokens": 54004722.0,
      "step": 31120
    },
    {
      "entropy": 5.695720958709717,
      "epoch": 2.4216300330674967,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004408308000521715,
      "loss": 4.9424,
      "mean_token_accuracy": 0.21512996256351472,
      "num_tokens": 54012660.0,
      "step": 31125
    },
    {
      "entropy": 5.760531663894653,
      "epoch": 2.4220190624392144,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044081206680863556,
      "loss": 5.0152,
      "mean_token_accuracy": 0.205538772046566,
      "num_tokens": 54021517.0,
      "step": 31130
    },
    {
      "entropy": 5.760920619964599,
      "epoch": 2.4224080918109316,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044079333104916554,
      "loss": 4.9943,
      "mean_token_accuracy": 0.20841407924890518,
      "num_tokens": 54030081.0,
      "step": 31135
    },
    {
      "entropy": 5.695950841903686,
      "epoch": 2.4227971211826493,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004407745927740454,
      "loss": 4.9233,
      "mean_token_accuracy": 0.20991428792476655,
      "num_tokens": 54038214.0,
      "step": 31140
    },
    {
      "entropy": 5.638093090057373,
      "epoch": 2.423186150554367,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004407558519835598,
      "loss": 4.885,
      "mean_token_accuracy": 0.20571240931749343,
      "num_tokens": 54046813.0,
      "step": 31145
    },
    {
      "entropy": 5.673790359497071,
      "epoch": 2.4235751799260843,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044073710867799295,
      "loss": 4.9836,
      "mean_token_accuracy": 0.2052168995141983,
      "num_tokens": 54056095.0,
      "step": 31150
    },
    {
      "entropy": 5.681187295913697,
      "epoch": 2.423964209297802,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004407183628576293,
      "loss": 4.9002,
      "mean_token_accuracy": 0.2139764040708542,
      "num_tokens": 54065173.0,
      "step": 31155
    },
    {
      "entropy": 5.767171335220337,
      "epoch": 2.4243532386695197,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004406996145227532,
      "loss": 5.1184,
      "mean_token_accuracy": 0.19384173154830933,
      "num_tokens": 54073858.0,
      "step": 31160
    },
    {
      "entropy": 5.753905534744263,
      "epoch": 2.424742268041237,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004406808636736492,
      "loss": 5.0125,
      "mean_token_accuracy": 0.20371875166893005,
      "num_tokens": 54082129.0,
      "step": 31165
    },
    {
      "entropy": 5.718317317962646,
      "epoch": 2.4251312974129546,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044066211031060183,
      "loss": 4.9377,
      "mean_token_accuracy": 0.20650246292352675,
      "num_tokens": 54090360.0,
      "step": 31170
    },
    {
      "entropy": 5.763486671447754,
      "epoch": 2.4255203267846723,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004406433544338956,
      "loss": 4.9861,
      "mean_token_accuracy": 0.20372754335403442,
      "num_tokens": 54098309.0,
      "step": 31175
    },
    {
      "entropy": 5.769637775421143,
      "epoch": 2.4259093561563896,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004406245960438151,
      "loss": 4.9751,
      "mean_token_accuracy": 0.21131702661514282,
      "num_tokens": 54106505.0,
      "step": 31180
    },
    {
      "entropy": 5.781920433044434,
      "epoch": 2.4262983855281073,
      "grad_norm": 1.234375,
      "learning_rate": 0.000440605835140645,
      "loss": 5.1472,
      "mean_token_accuracy": 0.19213493168354034,
      "num_tokens": 54115791.0,
      "step": 31185
    },
    {
      "entropy": 5.710179567337036,
      "epoch": 2.426687414899825,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004405870717246699,
      "loss": 4.9266,
      "mean_token_accuracy": 0.20687406808137893,
      "num_tokens": 54124187.0,
      "step": 31190
    },
    {
      "entropy": 5.75041708946228,
      "epoch": 2.4270764442715427,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044056830579617464,
      "loss": 5.0208,
      "mean_token_accuracy": 0.20127360820770263,
      "num_tokens": 54133566.0,
      "step": 31195
    },
    {
      "entropy": 5.708799982070923,
      "epoch": 2.42746547364326,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044054953735544376,
      "loss": 4.9336,
      "mean_token_accuracy": 0.20634414106607438,
      "num_tokens": 54141698.0,
      "step": 31200
    },
    {
      "entropy": 5.733378839492798,
      "epoch": 2.4278545030149776,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004405307664027622,
      "loss": 5.0653,
      "mean_token_accuracy": 0.19988239407539368,
      "num_tokens": 54150094.0,
      "step": 31205
    },
    {
      "entropy": 5.693490934371948,
      "epoch": 2.4282435323866953,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004405119929384147,
      "loss": 4.9377,
      "mean_token_accuracy": 0.21428713351488113,
      "num_tokens": 54158770.0,
      "step": 31210
    },
    {
      "entropy": 5.691935777664185,
      "epoch": 2.4286325617584126,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004404932169626862,
      "loss": 4.9244,
      "mean_token_accuracy": 0.20729187428951262,
      "num_tokens": 54166912.0,
      "step": 31215
    },
    {
      "entropy": 5.70848913192749,
      "epoch": 2.4290215911301303,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004404744384758615,
      "loss": 4.962,
      "mean_token_accuracy": 0.20755135118961335,
      "num_tokens": 54176027.0,
      "step": 31220
    },
    {
      "entropy": 5.717586803436279,
      "epoch": 2.429410620501848,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004404556574782256,
      "loss": 4.9483,
      "mean_token_accuracy": 0.21022219508886336,
      "num_tokens": 54184116.0,
      "step": 31225
    },
    {
      "entropy": 5.677651834487915,
      "epoch": 2.4297996498735657,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044043687397006336,
      "loss": 4.9137,
      "mean_token_accuracy": 0.21353434175252914,
      "num_tokens": 54193043.0,
      "step": 31230
    },
    {
      "entropy": 5.7101246356964115,
      "epoch": 2.430188679245283,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044041808795166,
      "loss": 5.0147,
      "mean_token_accuracy": 0.20186586529016495,
      "num_tokens": 54201953.0,
      "step": 31235
    },
    {
      "entropy": 5.725139379501343,
      "epoch": 2.4305777086170006,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004403992994233004,
      "loss": 4.9953,
      "mean_token_accuracy": 0.2108522593975067,
      "num_tokens": 54210316.0,
      "step": 31240
    },
    {
      "entropy": 5.73111138343811,
      "epoch": 2.4309667379887183,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044038050838526974,
      "loss": 4.9314,
      "mean_token_accuracy": 0.21369454115629197,
      "num_tokens": 54218795.0,
      "step": 31245
    },
    {
      "entropy": 5.715928840637207,
      "epoch": 2.4313557673604356,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000440361714837853,
      "loss": 4.9577,
      "mean_token_accuracy": 0.2048373267054558,
      "num_tokens": 54227455.0,
      "step": 31250
    },
    {
      "entropy": 5.673466730117798,
      "epoch": 2.4317447967321533,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004403429187813355,
      "loss": 4.973,
      "mean_token_accuracy": 0.20863606333732604,
      "num_tokens": 54236124.0,
      "step": 31255
    },
    {
      "entropy": 5.653055810928345,
      "epoch": 2.432133826103871,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004403241202160024,
      "loss": 4.9221,
      "mean_token_accuracy": 0.21059010475873946,
      "num_tokens": 54244638.0,
      "step": 31260
    },
    {
      "entropy": 5.673806810379029,
      "epoch": 2.432522855475588,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004403053191421388,
      "loss": 4.9673,
      "mean_token_accuracy": 0.20429740846157074,
      "num_tokens": 54253678.0,
      "step": 31265
    },
    {
      "entropy": 5.8019084453582765,
      "epoch": 2.432911884847306,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004402865155600302,
      "loss": 5.1107,
      "mean_token_accuracy": 0.19972557127475737,
      "num_tokens": 54262414.0,
      "step": 31270
    },
    {
      "entropy": 5.632422304153442,
      "epoch": 2.4333009142190236,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044026770946996183,
      "loss": 4.886,
      "mean_token_accuracy": 0.21389907002449035,
      "num_tokens": 54270597.0,
      "step": 31275
    },
    {
      "entropy": 5.659053230285645,
      "epoch": 2.433689943590741,
      "grad_norm": 1.328125,
      "learning_rate": 0.00044024890087221896,
      "loss": 4.9691,
      "mean_token_accuracy": 0.2059006929397583,
      "num_tokens": 54279432.0,
      "step": 31280
    },
    {
      "entropy": 5.691193294525147,
      "epoch": 2.4340789729624586,
      "grad_norm": 1.25,
      "learning_rate": 0.00044023008976708705,
      "loss": 4.907,
      "mean_token_accuracy": 0.21236684620380403,
      "num_tokens": 54287578.0,
      "step": 31285
    },
    {
      "entropy": 5.744124460220337,
      "epoch": 2.4344680023341763,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004402112761548515,
      "loss": 5.0637,
      "mean_token_accuracy": 0.20120678842067719,
      "num_tokens": 54296119.0,
      "step": 31290
    },
    {
      "entropy": 5.636694812774659,
      "epoch": 2.434857031705894,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044019246003579776,
      "loss": 4.8933,
      "mean_token_accuracy": 0.2099173218011856,
      "num_tokens": 54304363.0,
      "step": 31295
    },
    {
      "entropy": 5.790515995025634,
      "epoch": 2.435246061077611,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004401736414102115,
      "loss": 5.0037,
      "mean_token_accuracy": 0.2092122569680214,
      "num_tokens": 54313672.0,
      "step": 31300
    },
    {
      "entropy": 5.734694004058838,
      "epoch": 2.435635090449329,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000440154820278378,
      "loss": 4.9741,
      "mean_token_accuracy": 0.21284423470497132,
      "num_tokens": 54322163.0,
      "step": 31305
    },
    {
      "entropy": 5.6897961616516115,
      "epoch": 2.4360241198210466,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004401359966405831,
      "loss": 4.9413,
      "mean_token_accuracy": 0.21117124259471892,
      "num_tokens": 54331176.0,
      "step": 31310
    },
    {
      "entropy": 5.673261499404907,
      "epoch": 2.436413149192764,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044011717049711216,
      "loss": 4.9907,
      "mean_token_accuracy": 0.20502465963363647,
      "num_tokens": 54340404.0,
      "step": 31315
    },
    {
      "entropy": 5.721183443069458,
      "epoch": 2.4368021785644816,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044009834184825103,
      "loss": 4.9344,
      "mean_token_accuracy": 0.21526179909706117,
      "num_tokens": 54349704.0,
      "step": 31320
    },
    {
      "entropy": 5.741689395904541,
      "epoch": 2.4371912079361993,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004400795106942853,
      "loss": 5.0157,
      "mean_token_accuracy": 0.20412552058696748,
      "num_tokens": 54358869.0,
      "step": 31325
    },
    {
      "entropy": 5.663774824142456,
      "epoch": 2.437580237307917,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044006067703550083,
      "loss": 4.912,
      "mean_token_accuracy": 0.21400506645441056,
      "num_tokens": 54367918.0,
      "step": 31330
    },
    {
      "entropy": 5.704087495803833,
      "epoch": 2.437969266679634,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004400418408721833,
      "loss": 4.9857,
      "mean_token_accuracy": 0.20165156275033952,
      "num_tokens": 54375798.0,
      "step": 31335
    },
    {
      "entropy": 5.732313537597657,
      "epoch": 2.438358296051352,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044002300220461843,
      "loss": 4.9393,
      "mean_token_accuracy": 0.20444128066301345,
      "num_tokens": 54384230.0,
      "step": 31340
    },
    {
      "entropy": 5.6941978931427,
      "epoch": 2.4387473254230696,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004400041610330922,
      "loss": 4.9744,
      "mean_token_accuracy": 0.20451862663030623,
      "num_tokens": 54392444.0,
      "step": 31345
    },
    {
      "entropy": 5.716174697875976,
      "epoch": 2.439136354794787,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004399853173578905,
      "loss": 4.9926,
      "mean_token_accuracy": 0.20363901257514955,
      "num_tokens": 54400788.0,
      "step": 31350
    },
    {
      "entropy": 5.68687219619751,
      "epoch": 2.4395253841665046,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043996647117929916,
      "loss": 5.0021,
      "mean_token_accuracy": 0.20676013827323914,
      "num_tokens": 54409749.0,
      "step": 31355
    },
    {
      "entropy": 5.810472059249878,
      "epoch": 2.4399144135382222,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004399476224976043,
      "loss": 5.1341,
      "mean_token_accuracy": 0.2020565912127495,
      "num_tokens": 54419677.0,
      "step": 31360
    },
    {
      "entropy": 5.825078344345092,
      "epoch": 2.44030344290994,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043992877131309165,
      "loss": 5.1121,
      "mean_token_accuracy": 0.19977762699127197,
      "num_tokens": 54428821.0,
      "step": 31365
    },
    {
      "entropy": 5.761816024780273,
      "epoch": 2.440692472281657,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004399099176260475,
      "loss": 5.0344,
      "mean_token_accuracy": 0.2010772481560707,
      "num_tokens": 54437549.0,
      "step": 31370
    },
    {
      "entropy": 5.671188735961914,
      "epoch": 2.441081501653375,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004398910614367579,
      "loss": 4.8825,
      "mean_token_accuracy": 0.21021064072847367,
      "num_tokens": 54446247.0,
      "step": 31375
    },
    {
      "entropy": 5.723113965988159,
      "epoch": 2.441470531025092,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043987220274550876,
      "loss": 5.0175,
      "mean_token_accuracy": 0.19985198527574538,
      "num_tokens": 54455191.0,
      "step": 31380
    },
    {
      "entropy": 5.7209803581237795,
      "epoch": 2.44185956039681,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043985334155258643,
      "loss": 5.0132,
      "mean_token_accuracy": 0.20116425603628157,
      "num_tokens": 54463094.0,
      "step": 31385
    },
    {
      "entropy": 5.723725461959839,
      "epoch": 2.4422485897685275,
      "grad_norm": 1.25,
      "learning_rate": 0.000439834477858277,
      "loss": 4.9459,
      "mean_token_accuracy": 0.20391057878732682,
      "num_tokens": 54471314.0,
      "step": 31390
    },
    {
      "entropy": 5.741238307952881,
      "epoch": 2.4426376191402452,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043981561166286676,
      "loss": 5.0643,
      "mean_token_accuracy": 0.19864479899406434,
      "num_tokens": 54480283.0,
      "step": 31395
    },
    {
      "entropy": 5.681617307662964,
      "epoch": 2.4430266485119625,
      "grad_norm": 1.21875,
      "learning_rate": 0.000439796742966642,
      "loss": 4.9642,
      "mean_token_accuracy": 0.20790818184614182,
      "num_tokens": 54488588.0,
      "step": 31400
    },
    {
      "entropy": 5.717163705825806,
      "epoch": 2.44341567788368,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043977787176988884,
      "loss": 4.9245,
      "mean_token_accuracy": 0.2095158502459526,
      "num_tokens": 54497532.0,
      "step": 31405
    },
    {
      "entropy": 5.797204208374024,
      "epoch": 2.443804707255398,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004397589980728938,
      "loss": 5.0401,
      "mean_token_accuracy": 0.2008937582373619,
      "num_tokens": 54506152.0,
      "step": 31410
    },
    {
      "entropy": 5.705220651626587,
      "epoch": 2.444193736627115,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043974012187594324,
      "loss": 4.999,
      "mean_token_accuracy": 0.20275918245315552,
      "num_tokens": 54515674.0,
      "step": 31415
    },
    {
      "entropy": 5.64618558883667,
      "epoch": 2.444582765998833,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004397212431793235,
      "loss": 4.8206,
      "mean_token_accuracy": 0.21482677906751632,
      "num_tokens": 54523399.0,
      "step": 31420
    },
    {
      "entropy": 5.627922439575196,
      "epoch": 2.4449717953705505,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043970236198332107,
      "loss": 4.9406,
      "mean_token_accuracy": 0.206563863158226,
      "num_tokens": 54532128.0,
      "step": 31425
    },
    {
      "entropy": 5.7204667091369625,
      "epoch": 2.4453608247422682,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043968347828822246,
      "loss": 4.9261,
      "mean_token_accuracy": 0.20579209327697753,
      "num_tokens": 54540474.0,
      "step": 31430
    },
    {
      "entropy": 5.702330207824707,
      "epoch": 2.4457498541139855,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004396645920943142,
      "loss": 4.9347,
      "mean_token_accuracy": 0.20686771124601364,
      "num_tokens": 54549534.0,
      "step": 31435
    },
    {
      "entropy": 5.720772075653076,
      "epoch": 2.446138883485703,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004396457034018828,
      "loss": 4.9628,
      "mean_token_accuracy": 0.2047847494482994,
      "num_tokens": 54558071.0,
      "step": 31440
    },
    {
      "entropy": 5.745472240447998,
      "epoch": 2.446527912857421,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000439626812211215,
      "loss": 5.0436,
      "mean_token_accuracy": 0.1970282807946205,
      "num_tokens": 54565814.0,
      "step": 31445
    },
    {
      "entropy": 5.692371273040772,
      "epoch": 2.446916942229138,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043960791852259735,
      "loss": 4.9961,
      "mean_token_accuracy": 0.19384474009275438,
      "num_tokens": 54574911.0,
      "step": 31450
    },
    {
      "entropy": 5.6977245807647705,
      "epoch": 2.447305971600856,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004395890223363165,
      "loss": 4.9605,
      "mean_token_accuracy": 0.2079338774085045,
      "num_tokens": 54584129.0,
      "step": 31455
    },
    {
      "entropy": 5.71160569190979,
      "epoch": 2.4476950009725735,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043957012365265915,
      "loss": 4.9213,
      "mean_token_accuracy": 0.20982561111450196,
      "num_tokens": 54593203.0,
      "step": 31460
    },
    {
      "entropy": 5.702391147613525,
      "epoch": 2.4480840303442912,
      "grad_norm": 1.25,
      "learning_rate": 0.00043955122247191214,
      "loss": 5.0574,
      "mean_token_accuracy": 0.1994591921567917,
      "num_tokens": 54601436.0,
      "step": 31465
    },
    {
      "entropy": 5.653574419021607,
      "epoch": 2.4484730597160085,
      "grad_norm": 1.125,
      "learning_rate": 0.0004395323187943623,
      "loss": 5.0082,
      "mean_token_accuracy": 0.20017025470733643,
      "num_tokens": 54610550.0,
      "step": 31470
    },
    {
      "entropy": 5.673337650299072,
      "epoch": 2.448862089087726,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004395134126202964,
      "loss": 4.9202,
      "mean_token_accuracy": 0.20830218344926835,
      "num_tokens": 54619176.0,
      "step": 31475
    },
    {
      "entropy": 5.741445875167846,
      "epoch": 2.449251118459444,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004394945039500012,
      "loss": 4.9766,
      "mean_token_accuracy": 0.2049320325255394,
      "num_tokens": 54627259.0,
      "step": 31480
    },
    {
      "entropy": 5.681425666809082,
      "epoch": 2.449640147831161,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043947559278376385,
      "loss": 4.9692,
      "mean_token_accuracy": 0.19679538607597352,
      "num_tokens": 54635892.0,
      "step": 31485
    },
    {
      "entropy": 5.764542007446289,
      "epoch": 2.450029177202879,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043945667912187117,
      "loss": 5.0498,
      "mean_token_accuracy": 0.19391198009252547,
      "num_tokens": 54645360.0,
      "step": 31490
    },
    {
      "entropy": 5.71007719039917,
      "epoch": 2.4504182065745965,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004394377629646101,
      "loss": 4.966,
      "mean_token_accuracy": 0.20584850162267684,
      "num_tokens": 54654968.0,
      "step": 31495
    },
    {
      "entropy": 5.6735528945922855,
      "epoch": 2.4508072359463138,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043941884431226763,
      "loss": 4.9157,
      "mean_token_accuracy": 0.21346940994262695,
      "num_tokens": 54663347.0,
      "step": 31500
    },
    {
      "entropy": 5.7577108383178714,
      "epoch": 2.4511962653180315,
      "grad_norm": 1.25,
      "learning_rate": 0.000439399923165131,
      "loss": 5.0431,
      "mean_token_accuracy": 0.20293481796979904,
      "num_tokens": 54671744.0,
      "step": 31505
    },
    {
      "entropy": 5.723678684234619,
      "epoch": 2.451585294689749,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004393809995234872,
      "loss": 4.9532,
      "mean_token_accuracy": 0.20153075605630874,
      "num_tokens": 54680098.0,
      "step": 31510
    },
    {
      "entropy": 5.706916570663452,
      "epoch": 2.4519743240614664,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004393620733876233,
      "loss": 5.0486,
      "mean_token_accuracy": 0.2013229787349701,
      "num_tokens": 54689513.0,
      "step": 31515
    },
    {
      "entropy": 5.738279008865357,
      "epoch": 2.452363353433184,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004393431447578267,
      "loss": 4.9879,
      "mean_token_accuracy": 0.21006645858287812,
      "num_tokens": 54698079.0,
      "step": 31520
    },
    {
      "entropy": 5.71909465789795,
      "epoch": 2.452752382804902,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043932421363438427,
      "loss": 4.937,
      "mean_token_accuracy": 0.21364226341247558,
      "num_tokens": 54706956.0,
      "step": 31525
    },
    {
      "entropy": 5.630348825454712,
      "epoch": 2.4531414121766195,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004393052800175835,
      "loss": 4.9967,
      "mean_token_accuracy": 0.2034238874912262,
      "num_tokens": 54716131.0,
      "step": 31530
    },
    {
      "entropy": 5.715518665313721,
      "epoch": 2.4535304415483368,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043928634390771166,
      "loss": 4.9305,
      "mean_token_accuracy": 0.2039777472615242,
      "num_tokens": 54725080.0,
      "step": 31535
    },
    {
      "entropy": 5.750002098083496,
      "epoch": 2.4539194709200545,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043926740530505603,
      "loss": 4.9618,
      "mean_token_accuracy": 0.20628543198108673,
      "num_tokens": 54733273.0,
      "step": 31540
    },
    {
      "entropy": 5.72439661026001,
      "epoch": 2.454308500291772,
      "grad_norm": 1.140625,
      "learning_rate": 0.000439248464209904,
      "loss": 4.995,
      "mean_token_accuracy": 0.2047957718372345,
      "num_tokens": 54742359.0,
      "step": 31545
    },
    {
      "entropy": 5.7102861404418945,
      "epoch": 2.4546975296634894,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000439229520622543,
      "loss": 4.9891,
      "mean_token_accuracy": 0.19966392815113068,
      "num_tokens": 54750360.0,
      "step": 31550
    },
    {
      "entropy": 5.699611806869507,
      "epoch": 2.455086559035207,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004392105745432603,
      "loss": 4.9668,
      "mean_token_accuracy": 0.20473674237728118,
      "num_tokens": 54758663.0,
      "step": 31555
    },
    {
      "entropy": 5.747363567352295,
      "epoch": 2.455475588406925,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004391916259723436,
      "loss": 4.9252,
      "mean_token_accuracy": 0.21641176491975783,
      "num_tokens": 54767578.0,
      "step": 31560
    },
    {
      "entropy": 5.7250946998596195,
      "epoch": 2.4558646177786425,
      "grad_norm": 1.359375,
      "learning_rate": 0.00043917267491008023,
      "loss": 5.096,
      "mean_token_accuracy": 0.1988742768764496,
      "num_tokens": 54776651.0,
      "step": 31565
    },
    {
      "entropy": 5.749834299087524,
      "epoch": 2.4562536471503598,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004391537213567579,
      "loss": 4.986,
      "mean_token_accuracy": 0.20456638485193251,
      "num_tokens": 54784865.0,
      "step": 31570
    },
    {
      "entropy": 5.696139097213745,
      "epoch": 2.4566426765220775,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004391347653126641,
      "loss": 4.8412,
      "mean_token_accuracy": 0.21593325287103654,
      "num_tokens": 54794367.0,
      "step": 31575
    },
    {
      "entropy": 5.624767637252807,
      "epoch": 2.457031705893795,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043911580677808646,
      "loss": 4.87,
      "mean_token_accuracy": 0.21529455184936525,
      "num_tokens": 54802606.0,
      "step": 31580
    },
    {
      "entropy": 5.654270219802856,
      "epoch": 2.4574207352655124,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043909684575331264,
      "loss": 4.9122,
      "mean_token_accuracy": 0.21618134379386902,
      "num_tokens": 54812207.0,
      "step": 31585
    },
    {
      "entropy": 5.745765495300293,
      "epoch": 2.45780976463723,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004390778822386304,
      "loss": 4.9626,
      "mean_token_accuracy": 0.2117878556251526,
      "num_tokens": 54821047.0,
      "step": 31590
    },
    {
      "entropy": 5.862848997116089,
      "epoch": 2.458198794008948,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043905891623432754,
      "loss": 5.1185,
      "mean_token_accuracy": 0.20013767927885057,
      "num_tokens": 54829849.0,
      "step": 31595
    },
    {
      "entropy": 5.722316741943359,
      "epoch": 2.458587823380665,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043903994774069165,
      "loss": 5.0624,
      "mean_token_accuracy": 0.19554626047611237,
      "num_tokens": 54838690.0,
      "step": 31600
    },
    {
      "entropy": 5.70993447303772,
      "epoch": 2.4589768527523828,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004390209767580107,
      "loss": 5.0076,
      "mean_token_accuracy": 0.20605431348085404,
      "num_tokens": 54847795.0,
      "step": 31605
    },
    {
      "entropy": 5.734815645217895,
      "epoch": 2.4593658821241005,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004390020032865725,
      "loss": 4.9133,
      "mean_token_accuracy": 0.21225012391805648,
      "num_tokens": 54856676.0,
      "step": 31610
    },
    {
      "entropy": 5.705066156387329,
      "epoch": 2.4597549114958177,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004389830273266649,
      "loss": 4.9486,
      "mean_token_accuracy": 0.20221970081329346,
      "num_tokens": 54865177.0,
      "step": 31615
    },
    {
      "entropy": 5.690595579147339,
      "epoch": 2.4601439408675354,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043896404887857596,
      "loss": 4.935,
      "mean_token_accuracy": 0.20192739218473435,
      "num_tokens": 54872921.0,
      "step": 31620
    },
    {
      "entropy": 5.680861949920654,
      "epoch": 2.460532970239253,
      "grad_norm": 1.25,
      "learning_rate": 0.0004389450679425935,
      "loss": 5.0506,
      "mean_token_accuracy": 0.1967609256505966,
      "num_tokens": 54882045.0,
      "step": 31625
    },
    {
      "entropy": 5.805809259414673,
      "epoch": 2.460921999610971,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004389260845190055,
      "loss": 4.9953,
      "mean_token_accuracy": 0.208863665163517,
      "num_tokens": 54890647.0,
      "step": 31630
    },
    {
      "entropy": 5.744081163406372,
      "epoch": 2.461311028982688,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043890709860810014,
      "loss": 5.0081,
      "mean_token_accuracy": 0.2034199133515358,
      "num_tokens": 54898958.0,
      "step": 31635
    },
    {
      "entropy": 5.725208139419555,
      "epoch": 2.4617000583544058,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043888811021016545,
      "loss": 5.0238,
      "mean_token_accuracy": 0.19682927131652833,
      "num_tokens": 54907329.0,
      "step": 31640
    },
    {
      "entropy": 5.708975839614868,
      "epoch": 2.4620890877261234,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043886911932548953,
      "loss": 4.9851,
      "mean_token_accuracy": 0.2140280470252037,
      "num_tokens": 54916093.0,
      "step": 31645
    },
    {
      "entropy": 5.7385646343231205,
      "epoch": 2.4624781170978407,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004388501259543605,
      "loss": 5.0156,
      "mean_token_accuracy": 0.1991320863366127,
      "num_tokens": 54925030.0,
      "step": 31650
    },
    {
      "entropy": 5.742335796356201,
      "epoch": 2.4628671464695584,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004388311300970667,
      "loss": 4.9672,
      "mean_token_accuracy": 0.20399895459413528,
      "num_tokens": 54933719.0,
      "step": 31655
    },
    {
      "entropy": 5.759457111358643,
      "epoch": 2.463256175841276,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004388121317538962,
      "loss": 5.0803,
      "mean_token_accuracy": 0.19380418509244918,
      "num_tokens": 54941533.0,
      "step": 31660
    },
    {
      "entropy": 5.627963829040527,
      "epoch": 2.463645205212994,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004387931309251374,
      "loss": 4.9222,
      "mean_token_accuracy": 0.2139589622616768,
      "num_tokens": 54949819.0,
      "step": 31665
    },
    {
      "entropy": 5.729830408096314,
      "epoch": 2.464034234584711,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004387741276110785,
      "loss": 5.028,
      "mean_token_accuracy": 0.1999380111694336,
      "num_tokens": 54959199.0,
      "step": 31670
    },
    {
      "entropy": 5.759858512878418,
      "epoch": 2.4644232639564287,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043875512181200784,
      "loss": 5.0333,
      "mean_token_accuracy": 0.20584970861673355,
      "num_tokens": 54967742.0,
      "step": 31675
    },
    {
      "entropy": 5.798473167419433,
      "epoch": 2.4648122933281464,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004387361135282139,
      "loss": 5.1134,
      "mean_token_accuracy": 0.19341319650411606,
      "num_tokens": 54976419.0,
      "step": 31680
    },
    {
      "entropy": 5.704580163955688,
      "epoch": 2.4652013226998637,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043871710275998507,
      "loss": 4.9687,
      "mean_token_accuracy": 0.2057668849825859,
      "num_tokens": 54984975.0,
      "step": 31685
    },
    {
      "entropy": 5.725417137145996,
      "epoch": 2.4655903520715814,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004386980895076097,
      "loss": 4.9938,
      "mean_token_accuracy": 0.2034144729375839,
      "num_tokens": 54993746.0,
      "step": 31690
    },
    {
      "entropy": 5.720940256118775,
      "epoch": 2.465979381443299,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043867907377137646,
      "loss": 4.9047,
      "mean_token_accuracy": 0.2087928906083107,
      "num_tokens": 55002130.0,
      "step": 31695
    },
    {
      "entropy": 5.694974660873413,
      "epoch": 2.4663684108150163,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043866005555157376,
      "loss": 5.0369,
      "mean_token_accuracy": 0.20344994962215424,
      "num_tokens": 55010790.0,
      "step": 31700
    },
    {
      "entropy": 5.68558292388916,
      "epoch": 2.466757440186734,
      "grad_norm": 1.25,
      "learning_rate": 0.0004386410348484903,
      "loss": 4.9906,
      "mean_token_accuracy": 0.19897642135620117,
      "num_tokens": 55019542.0,
      "step": 31705
    },
    {
      "entropy": 5.671811771392822,
      "epoch": 2.4671464695584517,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004386220116624145,
      "loss": 4.8463,
      "mean_token_accuracy": 0.2133927196264267,
      "num_tokens": 55029230.0,
      "step": 31710
    },
    {
      "entropy": 5.7228179454803465,
      "epoch": 2.467535498930169,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043860298599363506,
      "loss": 4.9546,
      "mean_token_accuracy": 0.20828036069869996,
      "num_tokens": 55037371.0,
      "step": 31715
    },
    {
      "entropy": 5.6627613544464115,
      "epoch": 2.4679245283018867,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004385839578424408,
      "loss": 4.9721,
      "mean_token_accuracy": 0.2041296988725662,
      "num_tokens": 55046022.0,
      "step": 31720
    },
    {
      "entropy": 5.732797574996948,
      "epoch": 2.4683135576736044,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043856492720912037,
      "loss": 4.9113,
      "mean_token_accuracy": 0.2164381206035614,
      "num_tokens": 55054713.0,
      "step": 31725
    },
    {
      "entropy": 5.60374231338501,
      "epoch": 2.468702587045322,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004385458940939624,
      "loss": 4.9043,
      "mean_token_accuracy": 0.217742919921875,
      "num_tokens": 55064317.0,
      "step": 31730
    },
    {
      "entropy": 5.668994140625,
      "epoch": 2.4690916164170393,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004385268584972559,
      "loss": 4.9861,
      "mean_token_accuracy": 0.2116192728281021,
      "num_tokens": 55072700.0,
      "step": 31735
    },
    {
      "entropy": 5.755138969421386,
      "epoch": 2.469480645788757,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004385078204192896,
      "loss": 5.0972,
      "mean_token_accuracy": 0.1957418516278267,
      "num_tokens": 55081794.0,
      "step": 31740
    },
    {
      "entropy": 5.794542074203491,
      "epoch": 2.4698696751604747,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004384887798603523,
      "loss": 5.0755,
      "mean_token_accuracy": 0.20222117900848388,
      "num_tokens": 55090665.0,
      "step": 31745
    },
    {
      "entropy": 5.748948383331299,
      "epoch": 2.470258704532192,
      "grad_norm": 1.234375,
      "learning_rate": 0.000438469736820733,
      "loss": 4.9878,
      "mean_token_accuracy": 0.20395275801420212,
      "num_tokens": 55099306.0,
      "step": 31750
    },
    {
      "entropy": 5.725245237350464,
      "epoch": 2.4706477339039097,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043845069130072067,
      "loss": 5.0424,
      "mean_token_accuracy": 0.19820895195007324,
      "num_tokens": 55108420.0,
      "step": 31755
    },
    {
      "entropy": 5.765846252441406,
      "epoch": 2.4710367632756274,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004384316433006043,
      "loss": 4.9788,
      "mean_token_accuracy": 0.20367041677236558,
      "num_tokens": 55116629.0,
      "step": 31760
    },
    {
      "entropy": 5.743185138702392,
      "epoch": 2.471425792647345,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043841259282067276,
      "loss": 5.0001,
      "mean_token_accuracy": 0.2037462368607521,
      "num_tokens": 55125469.0,
      "step": 31765
    },
    {
      "entropy": 5.704171657562256,
      "epoch": 2.4718148220190623,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043839353986121533,
      "loss": 4.9886,
      "mean_token_accuracy": 0.20464126765727997,
      "num_tokens": 55134194.0,
      "step": 31770
    },
    {
      "entropy": 5.750198698043823,
      "epoch": 2.47220385139078,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004383744844225209,
      "loss": 5.0242,
      "mean_token_accuracy": 0.2025087520480156,
      "num_tokens": 55142931.0,
      "step": 31775
    },
    {
      "entropy": 5.7638637065887455,
      "epoch": 2.4725928807624977,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043835542650487875,
      "loss": 5.0191,
      "mean_token_accuracy": 0.20524778962135315,
      "num_tokens": 55152331.0,
      "step": 31780
    },
    {
      "entropy": 5.734733915328979,
      "epoch": 2.472981910134215,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004383363661085781,
      "loss": 4.9884,
      "mean_token_accuracy": 0.20543016791343688,
      "num_tokens": 55160860.0,
      "step": 31785
    },
    {
      "entropy": 5.760632801055908,
      "epoch": 2.4733709395059327,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004383173032339079,
      "loss": 5.0737,
      "mean_token_accuracy": 0.19823102355003358,
      "num_tokens": 55169891.0,
      "step": 31790
    },
    {
      "entropy": 5.69638295173645,
      "epoch": 2.4737599688776504,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004382982378811577,
      "loss": 4.9723,
      "mean_token_accuracy": 0.209269417822361,
      "num_tokens": 55178283.0,
      "step": 31795
    },
    {
      "entropy": 5.737106561660767,
      "epoch": 2.474148998249368,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004382791700506166,
      "loss": 5.0175,
      "mean_token_accuracy": 0.20122259855270386,
      "num_tokens": 55186683.0,
      "step": 31800
    },
    {
      "entropy": 5.676229047775268,
      "epoch": 2.4745380276210853,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000438260099742574,
      "loss": 4.8911,
      "mean_token_accuracy": 0.2153608724474907,
      "num_tokens": 55194758.0,
      "step": 31805
    },
    {
      "entropy": 5.689475631713867,
      "epoch": 2.474927056992803,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004382410269573192,
      "loss": 4.9359,
      "mean_token_accuracy": 0.21000287532806397,
      "num_tokens": 55203171.0,
      "step": 31810
    },
    {
      "entropy": 5.74536657333374,
      "epoch": 2.4753160863645203,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043822195169514167,
      "loss": 4.9872,
      "mean_token_accuracy": 0.20264419615268708,
      "num_tokens": 55210942.0,
      "step": 31815
    },
    {
      "entropy": 5.750460290908814,
      "epoch": 2.475705115736238,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043820287395633086,
      "loss": 5.0724,
      "mean_token_accuracy": 0.19789913594722747,
      "num_tokens": 55219937.0,
      "step": 31820
    },
    {
      "entropy": 5.741223382949829,
      "epoch": 2.4760941451079557,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004381837937411761,
      "loss": 5.0299,
      "mean_token_accuracy": 0.20495262742042542,
      "num_tokens": 55229461.0,
      "step": 31825
    },
    {
      "entropy": 5.662536668777466,
      "epoch": 2.4764831744796734,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043816471104996706,
      "loss": 4.9017,
      "mean_token_accuracy": 0.20790998488664628,
      "num_tokens": 55237910.0,
      "step": 31830
    },
    {
      "entropy": 5.687501668930054,
      "epoch": 2.4768722038513906,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004381456258829933,
      "loss": 4.9622,
      "mean_token_accuracy": 0.2038867563009262,
      "num_tokens": 55246131.0,
      "step": 31835
    },
    {
      "entropy": 5.703996372222901,
      "epoch": 2.4772612332231083,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043812653824054426,
      "loss": 4.9625,
      "mean_token_accuracy": 0.21161567121744157,
      "num_tokens": 55254677.0,
      "step": 31840
    },
    {
      "entropy": 5.732633018493653,
      "epoch": 2.477650262594826,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004381074481229097,
      "loss": 4.9777,
      "mean_token_accuracy": 0.1992129549384117,
      "num_tokens": 55263496.0,
      "step": 31845
    },
    {
      "entropy": 5.658323335647583,
      "epoch": 2.4780392919665433,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004380883555303791,
      "loss": 4.909,
      "mean_token_accuracy": 0.20670702308416367,
      "num_tokens": 55272614.0,
      "step": 31850
    },
    {
      "entropy": 5.683746337890625,
      "epoch": 2.478428321338261,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004380692604632424,
      "loss": 4.9525,
      "mean_token_accuracy": 0.20904795825481415,
      "num_tokens": 55281595.0,
      "step": 31855
    },
    {
      "entropy": 5.777969455718994,
      "epoch": 2.4788173507099787,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043805016292178923,
      "loss": 5.0574,
      "mean_token_accuracy": 0.20904883444309236,
      "num_tokens": 55291089.0,
      "step": 31860
    },
    {
      "entropy": 5.6704264163970945,
      "epoch": 2.4792063800816964,
      "grad_norm": 1.25,
      "learning_rate": 0.0004380310629063094,
      "loss": 4.8502,
      "mean_token_accuracy": 0.21935873329639435,
      "num_tokens": 55299505.0,
      "step": 31865
    },
    {
      "entropy": 5.710926532745361,
      "epoch": 2.4795954094534136,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004380119604170925,
      "loss": 5.0459,
      "mean_token_accuracy": 0.19802145212888717,
      "num_tokens": 55307383.0,
      "step": 31870
    },
    {
      "entropy": 5.75420389175415,
      "epoch": 2.4799844388251313,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043799285545442876,
      "loss": 5.0591,
      "mean_token_accuracy": 0.19179680496454238,
      "num_tokens": 55316312.0,
      "step": 31875
    },
    {
      "entropy": 5.744991970062256,
      "epoch": 2.480373468196849,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004379737480186078,
      "loss": 5.0742,
      "mean_token_accuracy": 0.20198495388031007,
      "num_tokens": 55325818.0,
      "step": 31880
    },
    {
      "entropy": 5.695103883743286,
      "epoch": 2.4807624975685663,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043795463810991956,
      "loss": 5.0201,
      "mean_token_accuracy": 0.19727768003940582,
      "num_tokens": 55334641.0,
      "step": 31885
    },
    {
      "entropy": 5.752845716476441,
      "epoch": 2.481151526940284,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004379355257286541,
      "loss": 4.9975,
      "mean_token_accuracy": 0.20458939671516418,
      "num_tokens": 55343109.0,
      "step": 31890
    },
    {
      "entropy": 5.712238836288452,
      "epoch": 2.4815405563120017,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043791641087510136,
      "loss": 4.9591,
      "mean_token_accuracy": 0.2126741960644722,
      "num_tokens": 55351832.0,
      "step": 31895
    },
    {
      "entropy": 5.7133111476898195,
      "epoch": 2.4819295856837194,
      "grad_norm": 1.25,
      "learning_rate": 0.0004378972935495514,
      "loss": 5.0028,
      "mean_token_accuracy": 0.20555525422096252,
      "num_tokens": 55361688.0,
      "step": 31900
    },
    {
      "entropy": 5.750016593933106,
      "epoch": 2.4823186150554366,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004378781737522943,
      "loss": 4.9951,
      "mean_token_accuracy": 0.19938890784978866,
      "num_tokens": 55370572.0,
      "step": 31905
    },
    {
      "entropy": 5.725487518310547,
      "epoch": 2.4827076444271543,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043785905148362007,
      "loss": 4.9721,
      "mean_token_accuracy": 0.2114108085632324,
      "num_tokens": 55379134.0,
      "step": 31910
    },
    {
      "entropy": 5.6971229076385494,
      "epoch": 2.483096673798872,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043783992674381903,
      "loss": 4.9524,
      "mean_token_accuracy": 0.20730831772089003,
      "num_tokens": 55387312.0,
      "step": 31915
    },
    {
      "entropy": 5.633147764205932,
      "epoch": 2.4834857031705893,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043782079953318126,
      "loss": 4.8526,
      "mean_token_accuracy": 0.21258982568979262,
      "num_tokens": 55396208.0,
      "step": 31920
    },
    {
      "entropy": 5.723811101913452,
      "epoch": 2.483874732542307,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004378016698519971,
      "loss": 5.0195,
      "mean_token_accuracy": 0.20523508191108703,
      "num_tokens": 55404142.0,
      "step": 31925
    },
    {
      "entropy": 5.692132186889649,
      "epoch": 2.4842637619140246,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043778253770055666,
      "loss": 4.922,
      "mean_token_accuracy": 0.21246977150440216,
      "num_tokens": 55412227.0,
      "step": 31930
    },
    {
      "entropy": 5.66039252281189,
      "epoch": 2.484652791285742,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043776340307915033,
      "loss": 4.9215,
      "mean_token_accuracy": 0.20812310576438903,
      "num_tokens": 55420566.0,
      "step": 31935
    },
    {
      "entropy": 5.698721647262573,
      "epoch": 2.4850418206574596,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004377442659880684,
      "loss": 4.9095,
      "mean_token_accuracy": 0.21190724819898604,
      "num_tokens": 55429456.0,
      "step": 31940
    },
    {
      "entropy": 5.663396120071411,
      "epoch": 2.4854308500291773,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004377251264276013,
      "loss": 4.9045,
      "mean_token_accuracy": 0.21155412644147872,
      "num_tokens": 55437607.0,
      "step": 31945
    },
    {
      "entropy": 5.627484464645386,
      "epoch": 2.4858198794008945,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043770598439803936,
      "loss": 4.9321,
      "mean_token_accuracy": 0.21017150729894638,
      "num_tokens": 55446654.0,
      "step": 31950
    },
    {
      "entropy": 5.661251878738403,
      "epoch": 2.4862089087726122,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043768683989967304,
      "loss": 5.0068,
      "mean_token_accuracy": 0.21182485520839692,
      "num_tokens": 55454963.0,
      "step": 31955
    },
    {
      "entropy": 5.701824045181274,
      "epoch": 2.48659793814433,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043766769293279294,
      "loss": 4.9473,
      "mean_token_accuracy": 0.19996298402547835,
      "num_tokens": 55463259.0,
      "step": 31960
    },
    {
      "entropy": 5.647969055175781,
      "epoch": 2.4869869675160476,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043764854349768957,
      "loss": 4.9472,
      "mean_token_accuracy": 0.20149911642074586,
      "num_tokens": 55472390.0,
      "step": 31965
    },
    {
      "entropy": 5.763763093948365,
      "epoch": 2.487375996887765,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043762939159465333,
      "loss": 5.0206,
      "mean_token_accuracy": 0.19603514969348906,
      "num_tokens": 55480821.0,
      "step": 31970
    },
    {
      "entropy": 5.8180993556976315,
      "epoch": 2.4877650262594826,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043761023722397495,
      "loss": 5.0074,
      "mean_token_accuracy": 0.20506039410829544,
      "num_tokens": 55489270.0,
      "step": 31975
    },
    {
      "entropy": 5.660141134262085,
      "epoch": 2.4881540556312003,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000437591080385945,
      "loss": 4.9349,
      "mean_token_accuracy": 0.21390817910432816,
      "num_tokens": 55497811.0,
      "step": 31980
    },
    {
      "entropy": 5.670733499526977,
      "epoch": 2.4885430850029175,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004375719210808543,
      "loss": 4.9077,
      "mean_token_accuracy": 0.204403792321682,
      "num_tokens": 55507068.0,
      "step": 31985
    },
    {
      "entropy": 5.679367256164551,
      "epoch": 2.4889321143746352,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004375527593089934,
      "loss": 4.8868,
      "mean_token_accuracy": 0.21107714027166366,
      "num_tokens": 55515922.0,
      "step": 31990
    },
    {
      "entropy": 5.690799760818481,
      "epoch": 2.489321143746353,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004375335950706531,
      "loss": 4.9552,
      "mean_token_accuracy": 0.20800403207540513,
      "num_tokens": 55524839.0,
      "step": 31995
    },
    {
      "entropy": 5.561902141571045,
      "epoch": 2.4897101731180706,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004375144283661242,
      "loss": 4.858,
      "mean_token_accuracy": 0.20913346111774445,
      "num_tokens": 55534426.0,
      "step": 32000
    },
    {
      "entropy": 5.6797041416168215,
      "epoch": 2.490099202489788,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004374952591956975,
      "loss": 4.9854,
      "mean_token_accuracy": 0.2033844694495201,
      "num_tokens": 55542773.0,
      "step": 32005
    },
    {
      "entropy": 5.7466654777526855,
      "epoch": 2.4904882318615056,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043747608755966396,
      "loss": 4.9502,
      "mean_token_accuracy": 0.20740076452493666,
      "num_tokens": 55551582.0,
      "step": 32010
    },
    {
      "entropy": 5.816459798812867,
      "epoch": 2.4908772612332233,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004374569134583143,
      "loss": 5.0744,
      "mean_token_accuracy": 0.20035319477319719,
      "num_tokens": 55561067.0,
      "step": 32015
    },
    {
      "entropy": 5.659238815307617,
      "epoch": 2.4912662906049405,
      "grad_norm": 1.25,
      "learning_rate": 0.0004374377368919396,
      "loss": 4.9299,
      "mean_token_accuracy": 0.2087470233440399,
      "num_tokens": 55569677.0,
      "step": 32020
    },
    {
      "entropy": 5.755471038818359,
      "epoch": 2.4916553199766582,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043741855786083085,
      "loss": 5.1238,
      "mean_token_accuracy": 0.19786030054092407,
      "num_tokens": 55578391.0,
      "step": 32025
    },
    {
      "entropy": 5.7636699199676515,
      "epoch": 2.492044349348376,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043739937636527894,
      "loss": 5.0274,
      "mean_token_accuracy": 0.20690529495477678,
      "num_tokens": 55587074.0,
      "step": 32030
    },
    {
      "entropy": 5.7338334083557125,
      "epoch": 2.492433378720093,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043738019240557494,
      "loss": 4.949,
      "mean_token_accuracy": 0.20173933655023574,
      "num_tokens": 55596944.0,
      "step": 32035
    },
    {
      "entropy": 5.758412456512451,
      "epoch": 2.492822408091811,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043736100598201,
      "loss": 4.954,
      "mean_token_accuracy": 0.20482483208179475,
      "num_tokens": 55605763.0,
      "step": 32040
    },
    {
      "entropy": 5.650083446502686,
      "epoch": 2.4932114374635286,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004373418170948752,
      "loss": 4.9905,
      "mean_token_accuracy": 0.20380522757768632,
      "num_tokens": 55614693.0,
      "step": 32045
    },
    {
      "entropy": 5.780364656448365,
      "epoch": 2.493600466835246,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004373226257444617,
      "loss": 5.138,
      "mean_token_accuracy": 0.19959172159433364,
      "num_tokens": 55623520.0,
      "step": 32050
    },
    {
      "entropy": 5.746345663070679,
      "epoch": 2.4939894962069635,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004373034319310608,
      "loss": 4.9317,
      "mean_token_accuracy": 0.21149479895830153,
      "num_tokens": 55631818.0,
      "step": 32055
    },
    {
      "entropy": 5.785731840133667,
      "epoch": 2.4943785255786812,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043728423565496367,
      "loss": 5.09,
      "mean_token_accuracy": 0.19993118792772294,
      "num_tokens": 55641388.0,
      "step": 32060
    },
    {
      "entropy": 5.662295532226563,
      "epoch": 2.494767554950399,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004372650369164615,
      "loss": 4.9653,
      "mean_token_accuracy": 0.209219329059124,
      "num_tokens": 55650374.0,
      "step": 32065
    },
    {
      "entropy": 5.735369253158569,
      "epoch": 2.495156584322116,
      "grad_norm": 1.25,
      "learning_rate": 0.00043724583571584565,
      "loss": 4.9953,
      "mean_token_accuracy": 0.20849401354789734,
      "num_tokens": 55658598.0,
      "step": 32070
    },
    {
      "entropy": 5.73172607421875,
      "epoch": 2.495545613693834,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043722663205340753,
      "loss": 5.0436,
      "mean_token_accuracy": 0.1948054850101471,
      "num_tokens": 55666954.0,
      "step": 32075
    },
    {
      "entropy": 5.735919427871704,
      "epoch": 2.4959346430655516,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004372074259294384,
      "loss": 4.9748,
      "mean_token_accuracy": 0.20781266838312148,
      "num_tokens": 55675451.0,
      "step": 32080
    },
    {
      "entropy": 5.72340145111084,
      "epoch": 2.496323672437269,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004371882173442299,
      "loss": 5.095,
      "mean_token_accuracy": 0.195270399749279,
      "num_tokens": 55683778.0,
      "step": 32085
    },
    {
      "entropy": 5.658077430725098,
      "epoch": 2.4967127018089865,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043716900629807323,
      "loss": 4.9277,
      "mean_token_accuracy": 0.21355827748775483,
      "num_tokens": 55692440.0,
      "step": 32090
    },
    {
      "entropy": 5.695153331756591,
      "epoch": 2.497101731180704,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043714979279126006,
      "loss": 4.9536,
      "mean_token_accuracy": 0.20435247868299483,
      "num_tokens": 55700648.0,
      "step": 32095
    },
    {
      "entropy": 5.697821998596192,
      "epoch": 2.497490760552422,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004371305768240818,
      "loss": 4.93,
      "mean_token_accuracy": 0.2100788712501526,
      "num_tokens": 55709515.0,
      "step": 32100
    },
    {
      "entropy": 5.715228033065796,
      "epoch": 2.497879789924139,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004371113583968302,
      "loss": 5.0457,
      "mean_token_accuracy": 0.2022640436887741,
      "num_tokens": 55717876.0,
      "step": 32105
    },
    {
      "entropy": 5.712648725509643,
      "epoch": 2.498268819295857,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004370921375097967,
      "loss": 4.9668,
      "mean_token_accuracy": 0.204514579474926,
      "num_tokens": 55726330.0,
      "step": 32110
    },
    {
      "entropy": 5.735370779037476,
      "epoch": 2.4986578486675746,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043707291416327307,
      "loss": 5.0535,
      "mean_token_accuracy": 0.20366884022951126,
      "num_tokens": 55736102.0,
      "step": 32115
    },
    {
      "entropy": 5.772699213027954,
      "epoch": 2.499046878039292,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004370536883575509,
      "loss": 5.0632,
      "mean_token_accuracy": 0.20349960327148436,
      "num_tokens": 55744022.0,
      "step": 32120
    },
    {
      "entropy": 5.721553754806519,
      "epoch": 2.4994359074110095,
      "grad_norm": 1.25,
      "learning_rate": 0.0004370344600929219,
      "loss": 4.9899,
      "mean_token_accuracy": 0.1990184247493744,
      "num_tokens": 55752613.0,
      "step": 32125
    },
    {
      "entropy": 5.727171277999878,
      "epoch": 2.499824936782727,
      "grad_norm": 1.125,
      "learning_rate": 0.0004370152293696779,
      "loss": 4.9446,
      "mean_token_accuracy": 0.2108341559767723,
      "num_tokens": 55761402.0,
      "step": 32130
    },
    {
      "entropy": 5.704691171646118,
      "epoch": 2.500213966154445,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004369959961881107,
      "loss": 4.8998,
      "mean_token_accuracy": 0.20949224531650543,
      "num_tokens": 55769630.0,
      "step": 32135
    },
    {
      "entropy": 5.6444531917572025,
      "epoch": 2.500602995526162,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000436976760548512,
      "loss": 4.9847,
      "mean_token_accuracy": 0.2065363869071007,
      "num_tokens": 55777928.0,
      "step": 32140
    },
    {
      "entropy": 5.695544910430908,
      "epoch": 2.50099202489788,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043695752245117393,
      "loss": 4.9936,
      "mean_token_accuracy": 0.2044360712170601,
      "num_tokens": 55786585.0,
      "step": 32145
    },
    {
      "entropy": 5.70560941696167,
      "epoch": 2.501381054269597,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004369382818963882,
      "loss": 4.877,
      "mean_token_accuracy": 0.21774827986955642,
      "num_tokens": 55795565.0,
      "step": 32150
    },
    {
      "entropy": 5.7134768009185795,
      "epoch": 2.501770083641315,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004369190388844467,
      "loss": 4.9735,
      "mean_token_accuracy": 0.20383271872997283,
      "num_tokens": 55804211.0,
      "step": 32155
    },
    {
      "entropy": 5.672632312774658,
      "epoch": 2.5021591130130325,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043689979341564154,
      "loss": 4.9938,
      "mean_token_accuracy": 0.20692790895700455,
      "num_tokens": 55812604.0,
      "step": 32160
    },
    {
      "entropy": 5.667633390426635,
      "epoch": 2.50254814238475,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004368805454902648,
      "loss": 4.9392,
      "mean_token_accuracy": 0.20593280494213104,
      "num_tokens": 55821293.0,
      "step": 32165
    },
    {
      "entropy": 5.668863964080811,
      "epoch": 2.5029371717564675,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043686129510860826,
      "loss": 4.9136,
      "mean_token_accuracy": 0.20429967045783998,
      "num_tokens": 55830423.0,
      "step": 32170
    },
    {
      "entropy": 5.746794605255127,
      "epoch": 2.503326201128185,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043684204227096435,
      "loss": 4.9934,
      "mean_token_accuracy": 0.2002092108130455,
      "num_tokens": 55838889.0,
      "step": 32175
    },
    {
      "entropy": 5.750613069534301,
      "epoch": 2.503715230499903,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000436822786977625,
      "loss": 5.0084,
      "mean_token_accuracy": 0.20229347795248032,
      "num_tokens": 55846670.0,
      "step": 32180
    },
    {
      "entropy": 5.7446081161499025,
      "epoch": 2.50410425987162,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043680352922888244,
      "loss": 5.0224,
      "mean_token_accuracy": 0.20572800189256668,
      "num_tokens": 55855341.0,
      "step": 32185
    },
    {
      "entropy": 5.762192678451538,
      "epoch": 2.504493289243338,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004367842690250288,
      "loss": 4.971,
      "mean_token_accuracy": 0.20543303191661835,
      "num_tokens": 55863491.0,
      "step": 32190
    },
    {
      "entropy": 5.667235660552978,
      "epoch": 2.5048823186150555,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004367650063663565,
      "loss": 4.8787,
      "mean_token_accuracy": 0.21335530728101731,
      "num_tokens": 55872346.0,
      "step": 32195
    },
    {
      "entropy": 5.660463094711304,
      "epoch": 2.505271347986773,
      "grad_norm": 1.25,
      "learning_rate": 0.0004367457412531576,
      "loss": 4.9056,
      "mean_token_accuracy": 0.2109124705195427,
      "num_tokens": 55880672.0,
      "step": 32200
    },
    {
      "entropy": 5.775030183792114,
      "epoch": 2.5056603773584905,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004367264736857245,
      "loss": 5.0212,
      "mean_token_accuracy": 0.20965417921543122,
      "num_tokens": 55888938.0,
      "step": 32205
    },
    {
      "entropy": 5.752799606323242,
      "epoch": 2.506049406730208,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004367072036643497,
      "loss": 4.9218,
      "mean_token_accuracy": 0.2099924325942993,
      "num_tokens": 55897522.0,
      "step": 32210
    },
    {
      "entropy": 5.72102575302124,
      "epoch": 2.506438436101926,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043668793118932536,
      "loss": 4.9508,
      "mean_token_accuracy": 0.20431464314460754,
      "num_tokens": 55905425.0,
      "step": 32215
    },
    {
      "entropy": 5.676226568222046,
      "epoch": 2.506827465473643,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043666865626094405,
      "loss": 4.9662,
      "mean_token_accuracy": 0.2090769425034523,
      "num_tokens": 55913990.0,
      "step": 32220
    },
    {
      "entropy": 5.812540054321289,
      "epoch": 2.507216494845361,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043664937887949825,
      "loss": 5.1067,
      "mean_token_accuracy": 0.19190953075885772,
      "num_tokens": 55922899.0,
      "step": 32225
    },
    {
      "entropy": 5.729031944274903,
      "epoch": 2.5076055242170785,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004366300990452803,
      "loss": 4.965,
      "mean_token_accuracy": 0.20764295309782027,
      "num_tokens": 55930935.0,
      "step": 32230
    },
    {
      "entropy": 5.767800569534302,
      "epoch": 2.507994553588796,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043661081675858295,
      "loss": 5.0032,
      "mean_token_accuracy": 0.20917452573776246,
      "num_tokens": 55939258.0,
      "step": 32235
    },
    {
      "entropy": 5.711412620544434,
      "epoch": 2.5083835829605134,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043659153201969865,
      "loss": 5.0005,
      "mean_token_accuracy": 0.20354715287685393,
      "num_tokens": 55947672.0,
      "step": 32240
    },
    {
      "entropy": 5.6939311027526855,
      "epoch": 2.508772612332231,
      "grad_norm": 1.25,
      "learning_rate": 0.0004365722448289201,
      "loss": 5.0135,
      "mean_token_accuracy": 0.20896376967430114,
      "num_tokens": 55956251.0,
      "step": 32245
    },
    {
      "entropy": 5.673779392242432,
      "epoch": 2.5091616417039484,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004365529551865399,
      "loss": 4.8466,
      "mean_token_accuracy": 0.21313344389200212,
      "num_tokens": 55965038.0,
      "step": 32250
    },
    {
      "entropy": 5.781704473495483,
      "epoch": 2.509550671075666,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043653366309285066,
      "loss": 5.0591,
      "mean_token_accuracy": 0.19696590900421143,
      "num_tokens": 55973140.0,
      "step": 32255
    },
    {
      "entropy": 5.7035298347473145,
      "epoch": 2.509939700447384,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004365143685481452,
      "loss": 5.0104,
      "mean_token_accuracy": 0.20032477974891663,
      "num_tokens": 55981097.0,
      "step": 32260
    },
    {
      "entropy": 5.637684965133667,
      "epoch": 2.5103287298191015,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004364950715527164,
      "loss": 4.8866,
      "mean_token_accuracy": 0.21138738393783568,
      "num_tokens": 55990077.0,
      "step": 32265
    },
    {
      "entropy": 5.703327369689942,
      "epoch": 2.5107177591908187,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004364757721068568,
      "loss": 4.9887,
      "mean_token_accuracy": 0.21135185956954955,
      "num_tokens": 55999289.0,
      "step": 32270
    },
    {
      "entropy": 5.693033027648926,
      "epoch": 2.5111067885625364,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043645647021085945,
      "loss": 4.9149,
      "mean_token_accuracy": 0.21232777535915376,
      "num_tokens": 56007736.0,
      "step": 32275
    },
    {
      "entropy": 5.696954154968262,
      "epoch": 2.511495817934254,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043643716586501706,
      "loss": 4.9135,
      "mean_token_accuracy": 0.21127343326807022,
      "num_tokens": 56016623.0,
      "step": 32280
    },
    {
      "entropy": 5.713492488861084,
      "epoch": 2.5118848473059714,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043641785906962276,
      "loss": 5.0068,
      "mean_token_accuracy": 0.20217975974082947,
      "num_tokens": 56024536.0,
      "step": 32285
    },
    {
      "entropy": 5.69862208366394,
      "epoch": 2.512273876677689,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004363985498249693,
      "loss": 5.0072,
      "mean_token_accuracy": 0.20458639413118362,
      "num_tokens": 56033618.0,
      "step": 32290
    },
    {
      "entropy": 5.7400007247924805,
      "epoch": 2.512662906049407,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043637923813134974,
      "loss": 5.0322,
      "mean_token_accuracy": 0.20439385771751403,
      "num_tokens": 56042836.0,
      "step": 32295
    },
    {
      "entropy": 5.722143697738647,
      "epoch": 2.5130519354211245,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043635992398905713,
      "loss": 4.9651,
      "mean_token_accuracy": 0.1997453138232231,
      "num_tokens": 56051296.0,
      "step": 32300
    },
    {
      "entropy": 5.692016744613648,
      "epoch": 2.5134409647928417,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004363406073983844,
      "loss": 4.943,
      "mean_token_accuracy": 0.20739489942789077,
      "num_tokens": 56060093.0,
      "step": 32305
    },
    {
      "entropy": 5.762143087387085,
      "epoch": 2.5138299941645594,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043632128835962485,
      "loss": 4.9313,
      "mean_token_accuracy": 0.2064829871058464,
      "num_tokens": 56068441.0,
      "step": 32310
    },
    {
      "entropy": 5.682105970382691,
      "epoch": 2.514219023536277,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043630196687307156,
      "loss": 4.8964,
      "mean_token_accuracy": 0.20389014184474946,
      "num_tokens": 56076592.0,
      "step": 32315
    },
    {
      "entropy": 5.621613597869873,
      "epoch": 2.5146080529079944,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004362826429390176,
      "loss": 4.8975,
      "mean_token_accuracy": 0.21033428460359574,
      "num_tokens": 56085299.0,
      "step": 32320
    },
    {
      "entropy": 5.690698671340942,
      "epoch": 2.514997082279712,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043626331655775624,
      "loss": 4.9306,
      "mean_token_accuracy": 0.20827299654483794,
      "num_tokens": 56093891.0,
      "step": 32325
    },
    {
      "entropy": 5.701816368103027,
      "epoch": 2.51538611165143,
      "grad_norm": 1.25,
      "learning_rate": 0.00043624398772958075,
      "loss": 5.0427,
      "mean_token_accuracy": 0.20374808460474014,
      "num_tokens": 56102767.0,
      "step": 32330
    },
    {
      "entropy": 5.674157190322876,
      "epoch": 2.5157751410231475,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004362246564547844,
      "loss": 4.947,
      "mean_token_accuracy": 0.20889146476984025,
      "num_tokens": 56110861.0,
      "step": 32335
    },
    {
      "entropy": 5.740803480148315,
      "epoch": 2.5161641703948647,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043620532273366044,
      "loss": 5.061,
      "mean_token_accuracy": 0.19930058121681213,
      "num_tokens": 56119842.0,
      "step": 32340
    },
    {
      "entropy": 5.7867364406585695,
      "epoch": 2.5165531997665824,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004361859865665024,
      "loss": 5.0345,
      "mean_token_accuracy": 0.20164532512426375,
      "num_tokens": 56128246.0,
      "step": 32345
    },
    {
      "entropy": 5.748019599914551,
      "epoch": 2.5169422291382997,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043616664795360346,
      "loss": 5.0123,
      "mean_token_accuracy": 0.20736071318387986,
      "num_tokens": 56137320.0,
      "step": 32350
    },
    {
      "entropy": 5.697682952880859,
      "epoch": 2.5173312585100174,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043614730689525716,
      "loss": 4.9575,
      "mean_token_accuracy": 0.21178145557641984,
      "num_tokens": 56145739.0,
      "step": 32355
    },
    {
      "entropy": 5.675608062744141,
      "epoch": 2.517720287881735,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043612796339175707,
      "loss": 4.9505,
      "mean_token_accuracy": 0.21417725533246995,
      "num_tokens": 56154526.0,
      "step": 32360
    },
    {
      "entropy": 5.793307256698609,
      "epoch": 2.5181093172534528,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043610861744339653,
      "loss": 5.1371,
      "mean_token_accuracy": 0.19573514312505721,
      "num_tokens": 56163209.0,
      "step": 32365
    },
    {
      "entropy": 5.755225658416748,
      "epoch": 2.5184983466251705,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004360892690504692,
      "loss": 4.9622,
      "mean_token_accuracy": 0.2139703392982483,
      "num_tokens": 56171541.0,
      "step": 32370
    },
    {
      "entropy": 5.679674673080444,
      "epoch": 2.5188873759968877,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004360699182132685,
      "loss": 4.9527,
      "mean_token_accuracy": 0.20583407133817672,
      "num_tokens": 56180309.0,
      "step": 32375
    },
    {
      "entropy": 5.75362606048584,
      "epoch": 2.5192764053686054,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043605056493208825,
      "loss": 5.0456,
      "mean_token_accuracy": 0.20934692174196243,
      "num_tokens": 56188916.0,
      "step": 32380
    },
    {
      "entropy": 5.688563823699951,
      "epoch": 2.5196654347403227,
      "grad_norm": 1.15625,
      "learning_rate": 0.000436031209207222,
      "loss": 4.8603,
      "mean_token_accuracy": 0.21684483140707017,
      "num_tokens": 56197358.0,
      "step": 32385
    },
    {
      "entropy": 5.67838683128357,
      "epoch": 2.5200544641120404,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043601185103896346,
      "loss": 5.0007,
      "mean_token_accuracy": 0.19860987812280656,
      "num_tokens": 56206060.0,
      "step": 32390
    },
    {
      "entropy": 5.648096704483033,
      "epoch": 2.520443493483758,
      "grad_norm": 1.34375,
      "learning_rate": 0.00043599249042760636,
      "loss": 4.927,
      "mean_token_accuracy": 0.20682421773672105,
      "num_tokens": 56214539.0,
      "step": 32395
    },
    {
      "entropy": 5.611673593521118,
      "epoch": 2.5208325228554758,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043597312737344446,
      "loss": 4.8383,
      "mean_token_accuracy": 0.2143823742866516,
      "num_tokens": 56222386.0,
      "step": 32400
    },
    {
      "entropy": 5.612220048904419,
      "epoch": 2.521221552227193,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043595376187677157,
      "loss": 4.9206,
      "mean_token_accuracy": 0.2042781874537468,
      "num_tokens": 56231036.0,
      "step": 32405
    },
    {
      "entropy": 5.710822629928589,
      "epoch": 2.5216105815989107,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043593439393788147,
      "loss": 4.9974,
      "mean_token_accuracy": 0.20774319171905517,
      "num_tokens": 56239494.0,
      "step": 32410
    },
    {
      "entropy": 5.754711151123047,
      "epoch": 2.5219996109706284,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004359150235570682,
      "loss": 5.0131,
      "mean_token_accuracy": 0.20264858156442642,
      "num_tokens": 56248075.0,
      "step": 32415
    },
    {
      "entropy": 5.683545112609863,
      "epoch": 2.5223886403423457,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004358956507346255,
      "loss": 5.007,
      "mean_token_accuracy": 0.20088429003953934,
      "num_tokens": 56256630.0,
      "step": 32420
    },
    {
      "entropy": 5.652667093276977,
      "epoch": 2.5227776697140634,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004358762754708474,
      "loss": 4.8518,
      "mean_token_accuracy": 0.2120782181620598,
      "num_tokens": 56264922.0,
      "step": 32425
    },
    {
      "entropy": 5.753397655487061,
      "epoch": 2.523166699085781,
      "grad_norm": 1.25,
      "learning_rate": 0.00043585689776602795,
      "loss": 5.0909,
      "mean_token_accuracy": 0.20110637396574021,
      "num_tokens": 56273996.0,
      "step": 32430
    },
    {
      "entropy": 5.761503839492798,
      "epoch": 2.5235557284574988,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043583751762046104,
      "loss": 5.0258,
      "mean_token_accuracy": 0.2060214549303055,
      "num_tokens": 56283583.0,
      "step": 32435
    },
    {
      "entropy": 5.735718727111816,
      "epoch": 2.523944757829216,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004358181350344408,
      "loss": 4.9686,
      "mean_token_accuracy": 0.2112908571958542,
      "num_tokens": 56292197.0,
      "step": 32440
    },
    {
      "entropy": 5.735447359085083,
      "epoch": 2.5243337872009337,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004357987500082613,
      "loss": 5.031,
      "mean_token_accuracy": 0.20029715448617935,
      "num_tokens": 56301177.0,
      "step": 32445
    },
    {
      "entropy": 5.74889121055603,
      "epoch": 2.524722816572651,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043577936254221675,
      "loss": 4.9815,
      "mean_token_accuracy": 0.20018882304430008,
      "num_tokens": 56310024.0,
      "step": 32450
    },
    {
      "entropy": 5.762160110473633,
      "epoch": 2.5251118459443687,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004357599726366013,
      "loss": 4.9902,
      "mean_token_accuracy": 0.20021328181028367,
      "num_tokens": 56318190.0,
      "step": 32455
    },
    {
      "entropy": 5.636119413375854,
      "epoch": 2.5255008753160864,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004357405802917091,
      "loss": 4.8461,
      "mean_token_accuracy": 0.2185628041625023,
      "num_tokens": 56326443.0,
      "step": 32460
    },
    {
      "entropy": 5.648688507080078,
      "epoch": 2.525889904687804,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004357211855078345,
      "loss": 4.9602,
      "mean_token_accuracy": 0.20393080860376359,
      "num_tokens": 56335263.0,
      "step": 32465
    },
    {
      "entropy": 5.753473806381225,
      "epoch": 2.5262789340595218,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00043570178828527157,
      "loss": 5.0015,
      "mean_token_accuracy": 0.20627198815345765,
      "num_tokens": 56343202.0,
      "step": 32470
    },
    {
      "entropy": 5.692942762374878,
      "epoch": 2.526667963431239,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004356823886243149,
      "loss": 4.897,
      "mean_token_accuracy": 0.2102028489112854,
      "num_tokens": 56352150.0,
      "step": 32475
    },
    {
      "entropy": 5.6208490371704105,
      "epoch": 2.5270569928029567,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043566298652525875,
      "loss": 4.863,
      "mean_token_accuracy": 0.2176535338163376,
      "num_tokens": 56360460.0,
      "step": 32480
    },
    {
      "entropy": 5.643869066238404,
      "epoch": 2.527446022174674,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043564358198839746,
      "loss": 5.0422,
      "mean_token_accuracy": 0.20829371213912964,
      "num_tokens": 56369889.0,
      "step": 32485
    },
    {
      "entropy": 5.7295105934143065,
      "epoch": 2.5278350515463917,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004356241750140256,
      "loss": 4.8801,
      "mean_token_accuracy": 0.20854940861463547,
      "num_tokens": 56378500.0,
      "step": 32490
    },
    {
      "entropy": 5.776626539230347,
      "epoch": 2.5282240809181094,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043560476560243743,
      "loss": 5.0412,
      "mean_token_accuracy": 0.20312431901693345,
      "num_tokens": 56387390.0,
      "step": 32495
    },
    {
      "entropy": 5.782455587387085,
      "epoch": 2.528613110289827,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004355853537539276,
      "loss": 5.0312,
      "mean_token_accuracy": 0.20380175858736038,
      "num_tokens": 56395782.0,
      "step": 32500
    },
    {
      "entropy": 5.7282756805419925,
      "epoch": 2.5290021396615443,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004355659394687906,
      "loss": 4.9239,
      "mean_token_accuracy": 0.21438917964696885,
      "num_tokens": 56405023.0,
      "step": 32505
    },
    {
      "entropy": 5.791355228424072,
      "epoch": 2.529391169033262,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004355465227473211,
      "loss": 5.0496,
      "mean_token_accuracy": 0.2037491738796234,
      "num_tokens": 56413669.0,
      "step": 32510
    },
    {
      "entropy": 5.703431558609009,
      "epoch": 2.5297801984049797,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004355271035898136,
      "loss": 4.9882,
      "mean_token_accuracy": 0.20518281310796738,
      "num_tokens": 56422286.0,
      "step": 32515
    },
    {
      "entropy": 5.730549335479736,
      "epoch": 2.530169227776697,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004355076819965628,
      "loss": 5.0098,
      "mean_token_accuracy": 0.2091906875371933,
      "num_tokens": 56430891.0,
      "step": 32520
    },
    {
      "entropy": 5.74372034072876,
      "epoch": 2.5305582571484146,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043548825796786353,
      "loss": 5.0036,
      "mean_token_accuracy": 0.20668756663799287,
      "num_tokens": 56439231.0,
      "step": 32525
    },
    {
      "entropy": 5.628321504592895,
      "epoch": 2.5309472865201323,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043546883150401023,
      "loss": 4.761,
      "mean_token_accuracy": 0.22700072228908538,
      "num_tokens": 56447508.0,
      "step": 32530
    },
    {
      "entropy": 5.670591688156128,
      "epoch": 2.53133631589185,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043544940260529787,
      "loss": 4.9562,
      "mean_token_accuracy": 0.20679677724838258,
      "num_tokens": 56456049.0,
      "step": 32535
    },
    {
      "entropy": 5.694138860702514,
      "epoch": 2.5317253452635673,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004354299712720213,
      "loss": 5.0049,
      "mean_token_accuracy": 0.20608452707529068,
      "num_tokens": 56464775.0,
      "step": 32540
    },
    {
      "entropy": 5.71428484916687,
      "epoch": 2.532114374635285,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004354105375044752,
      "loss": 4.9685,
      "mean_token_accuracy": 0.20670738071203232,
      "num_tokens": 56473835.0,
      "step": 32545
    },
    {
      "entropy": 5.755975914001465,
      "epoch": 2.5325034040070027,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043539110130295445,
      "loss": 4.9713,
      "mean_token_accuracy": 0.20677268654108047,
      "num_tokens": 56482488.0,
      "step": 32550
    },
    {
      "entropy": 5.708358860015869,
      "epoch": 2.53289243337872,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043537166266775405,
      "loss": 4.9174,
      "mean_token_accuracy": 0.20758798122406005,
      "num_tokens": 56491928.0,
      "step": 32555
    },
    {
      "entropy": 5.65998420715332,
      "epoch": 2.5332814627504376,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000435352221599169,
      "loss": 4.8802,
      "mean_token_accuracy": 0.21052341908216476,
      "num_tokens": 56500247.0,
      "step": 32560
    },
    {
      "entropy": 5.711657333374023,
      "epoch": 2.5336704921221553,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004353327780974941,
      "loss": 5.0048,
      "mean_token_accuracy": 0.20473283231258393,
      "num_tokens": 56509507.0,
      "step": 32565
    },
    {
      "entropy": 5.789334917068482,
      "epoch": 2.534059521493873,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004353133321630246,
      "loss": 5.0273,
      "mean_token_accuracy": 0.20635245591402054,
      "num_tokens": 56517895.0,
      "step": 32570
    },
    {
      "entropy": 5.671020269393921,
      "epoch": 2.5344485508655903,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004352938837960554,
      "loss": 4.9122,
      "mean_token_accuracy": 0.21317447870969772,
      "num_tokens": 56526894.0,
      "step": 32575
    },
    {
      "entropy": 5.684700155258179,
      "epoch": 2.534837580237308,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043527443299688166,
      "loss": 4.974,
      "mean_token_accuracy": 0.2075178250670433,
      "num_tokens": 56535970.0,
      "step": 32580
    },
    {
      "entropy": 5.75758581161499,
      "epoch": 2.5352266096090252,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043525497976579843,
      "loss": 4.9767,
      "mean_token_accuracy": 0.21011517196893692,
      "num_tokens": 56544385.0,
      "step": 32585
    },
    {
      "entropy": 5.676788854598999,
      "epoch": 2.535615638980743,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000435235524103101,
      "loss": 4.9795,
      "mean_token_accuracy": 0.19947906583547592,
      "num_tokens": 56552645.0,
      "step": 32590
    },
    {
      "entropy": 5.805925273895264,
      "epoch": 2.5360046683524606,
      "grad_norm": 1.25,
      "learning_rate": 0.0004352160660090846,
      "loss": 5.0689,
      "mean_token_accuracy": 0.1996101751923561,
      "num_tokens": 56562223.0,
      "step": 32595
    },
    {
      "entropy": 5.730615091323853,
      "epoch": 2.5363936977241783,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004351966054840443,
      "loss": 4.9309,
      "mean_token_accuracy": 0.20985871702432632,
      "num_tokens": 56571015.0,
      "step": 32600
    },
    {
      "entropy": 5.649417686462402,
      "epoch": 2.5367827270958956,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043517714252827556,
      "loss": 4.8936,
      "mean_token_accuracy": 0.21775211691856383,
      "num_tokens": 56579566.0,
      "step": 32605
    },
    {
      "entropy": 5.752285766601562,
      "epoch": 2.5371717564676133,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004351576771420737,
      "loss": 4.989,
      "mean_token_accuracy": 0.20706076622009278,
      "num_tokens": 56588231.0,
      "step": 32610
    },
    {
      "entropy": 5.722892236709595,
      "epoch": 2.537560785839331,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004351382093257339,
      "loss": 4.967,
      "mean_token_accuracy": 0.21211074143648148,
      "num_tokens": 56596581.0,
      "step": 32615
    },
    {
      "entropy": 5.655333852767944,
      "epoch": 2.5379498152110482,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004351187390795517,
      "loss": 4.893,
      "mean_token_accuracy": 0.21630424708127977,
      "num_tokens": 56605440.0,
      "step": 32620
    },
    {
      "entropy": 5.70293550491333,
      "epoch": 2.538338844582766,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043509926640382244,
      "loss": 5.0685,
      "mean_token_accuracy": 0.20204764008522033,
      "num_tokens": 56614819.0,
      "step": 32625
    },
    {
      "entropy": 5.703183269500732,
      "epoch": 2.5387278739544836,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004350797912988417,
      "loss": 5.0173,
      "mean_token_accuracy": 0.2049080029129982,
      "num_tokens": 56623372.0,
      "step": 32630
    },
    {
      "entropy": 5.651555728912354,
      "epoch": 2.5391169033262013,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004350603137649049,
      "loss": 4.9184,
      "mean_token_accuracy": 0.2106804922223091,
      "num_tokens": 56631768.0,
      "step": 32635
    },
    {
      "entropy": 5.674939823150635,
      "epoch": 2.5395059326979186,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004350408338023075,
      "loss": 4.946,
      "mean_token_accuracy": 0.20357927680015564,
      "num_tokens": 56640135.0,
      "step": 32640
    },
    {
      "entropy": 5.661525774002075,
      "epoch": 2.5398949620696363,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004350213514113453,
      "loss": 4.863,
      "mean_token_accuracy": 0.20919623970985413,
      "num_tokens": 56648007.0,
      "step": 32645
    },
    {
      "entropy": 5.760555553436279,
      "epoch": 2.540283991441354,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043500186659231377,
      "loss": 5.0418,
      "mean_token_accuracy": 0.2018701419234276,
      "num_tokens": 56656926.0,
      "step": 32650
    },
    {
      "entropy": 5.7797376155853275,
      "epoch": 2.5406730208130712,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004349823793455086,
      "loss": 5.0781,
      "mean_token_accuracy": 0.20385206937789918,
      "num_tokens": 56665404.0,
      "step": 32655
    },
    {
      "entropy": 5.725810384750366,
      "epoch": 2.541062050184789,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043496288967122547,
      "loss": 4.978,
      "mean_token_accuracy": 0.2070690795779228,
      "num_tokens": 56673922.0,
      "step": 32660
    },
    {
      "entropy": 5.692542839050293,
      "epoch": 2.5414510795565066,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043494339756976007,
      "loss": 4.9383,
      "mean_token_accuracy": 0.2087753340601921,
      "num_tokens": 56682062.0,
      "step": 32665
    },
    {
      "entropy": 5.744929361343384,
      "epoch": 2.5418401089282243,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004349239030414082,
      "loss": 5.0751,
      "mean_token_accuracy": 0.20303717255592346,
      "num_tokens": 56690571.0,
      "step": 32670
    },
    {
      "entropy": 5.729385757446289,
      "epoch": 2.5422291382999416,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043490440608646557,
      "loss": 4.9859,
      "mean_token_accuracy": 0.2011660650372505,
      "num_tokens": 56699599.0,
      "step": 32675
    },
    {
      "entropy": 5.719655752182007,
      "epoch": 2.5426181676716593,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043488490670522817,
      "loss": 4.9284,
      "mean_token_accuracy": 0.2111671820282936,
      "num_tokens": 56708120.0,
      "step": 32680
    },
    {
      "entropy": 5.680696773529053,
      "epoch": 2.5430071970433765,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004348654048979918,
      "loss": 4.9363,
      "mean_token_accuracy": 0.20661073327064514,
      "num_tokens": 56716339.0,
      "step": 32685
    },
    {
      "entropy": 5.675560235977173,
      "epoch": 2.543396226415094,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004348459006650523,
      "loss": 4.8559,
      "mean_token_accuracy": 0.22328649014234542,
      "num_tokens": 56724157.0,
      "step": 32690
    },
    {
      "entropy": 5.725537347793579,
      "epoch": 2.543785255786812,
      "grad_norm": 1.25,
      "learning_rate": 0.00043482639400670576,
      "loss": 4.9999,
      "mean_token_accuracy": 0.2042088359594345,
      "num_tokens": 56733075.0,
      "step": 32695
    },
    {
      "entropy": 5.634770298004151,
      "epoch": 2.5441742851585296,
      "grad_norm": 1.203125,
      "learning_rate": 0.000434806884923248,
      "loss": 4.8962,
      "mean_token_accuracy": 0.21420707553625107,
      "num_tokens": 56742184.0,
      "step": 32700
    },
    {
      "entropy": 5.680974292755127,
      "epoch": 2.5445633145302473,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004347873734149752,
      "loss": 4.8891,
      "mean_token_accuracy": 0.20550039112567903,
      "num_tokens": 56750559.0,
      "step": 32705
    },
    {
      "entropy": 5.749752426147461,
      "epoch": 2.5449523439019646,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043476785948218325,
      "loss": 5.0419,
      "mean_token_accuracy": 0.199990214407444,
      "num_tokens": 56760031.0,
      "step": 32710
    },
    {
      "entropy": 5.71320686340332,
      "epoch": 2.5453413732736823,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043474834312516835,
      "loss": 5.0046,
      "mean_token_accuracy": 0.20961831361055375,
      "num_tokens": 56768496.0,
      "step": 32715
    },
    {
      "entropy": 5.718040227890015,
      "epoch": 2.5457304026453995,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004347288243442266,
      "loss": 4.9832,
      "mean_token_accuracy": 0.20774815529584884,
      "num_tokens": 56776876.0,
      "step": 32720
    },
    {
      "entropy": 5.701543855667114,
      "epoch": 2.546119432017117,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004347093031396543,
      "loss": 4.9428,
      "mean_token_accuracy": 0.20958171933889388,
      "num_tokens": 56785661.0,
      "step": 32725
    },
    {
      "entropy": 5.675418901443481,
      "epoch": 2.546508461388835,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004346897795117474,
      "loss": 4.9481,
      "mean_token_accuracy": 0.2063525587320328,
      "num_tokens": 56794818.0,
      "step": 32730
    },
    {
      "entropy": 5.695310878753662,
      "epoch": 2.5468974907605526,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004346702534608023,
      "loss": 4.9845,
      "mean_token_accuracy": 0.20333731770515442,
      "num_tokens": 56804528.0,
      "step": 32735
    },
    {
      "entropy": 5.734045791625976,
      "epoch": 2.54728652013227,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004346507249871153,
      "loss": 4.9555,
      "mean_token_accuracy": 0.2121738761663437,
      "num_tokens": 56812766.0,
      "step": 32740
    },
    {
      "entropy": 5.705202341079712,
      "epoch": 2.5476755495039876,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004346311940909826,
      "loss": 4.9173,
      "mean_token_accuracy": 0.21172067373991013,
      "num_tokens": 56821108.0,
      "step": 32745
    },
    {
      "entropy": 5.739062213897705,
      "epoch": 2.5480645788757053,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043461166077270056,
      "loss": 5.0413,
      "mean_token_accuracy": 0.19954050928354264,
      "num_tokens": 56829060.0,
      "step": 32750
    },
    {
      "entropy": 5.753063774108886,
      "epoch": 2.5484536082474225,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043459212503256566,
      "loss": 5.0143,
      "mean_token_accuracy": 0.20793634802103042,
      "num_tokens": 56837440.0,
      "step": 32755
    },
    {
      "entropy": 5.736642932891845,
      "epoch": 2.54884263761914,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004345725868708743,
      "loss": 4.9677,
      "mean_token_accuracy": 0.20921308994293214,
      "num_tokens": 56845485.0,
      "step": 32760
    },
    {
      "entropy": 5.739649915695191,
      "epoch": 2.549231666990858,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043455304628792294,
      "loss": 4.9577,
      "mean_token_accuracy": 0.21059300154447555,
      "num_tokens": 56854498.0,
      "step": 32765
    },
    {
      "entropy": 5.6710960388183596,
      "epoch": 2.5496206963625756,
      "grad_norm": 1.203125,
      "learning_rate": 0.000434533503284008,
      "loss": 4.9542,
      "mean_token_accuracy": 0.21101408898830415,
      "num_tokens": 56862910.0,
      "step": 32770
    },
    {
      "entropy": 5.705244398117065,
      "epoch": 2.550009725734293,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004345139578594261,
      "loss": 4.9516,
      "mean_token_accuracy": 0.2060338154435158,
      "num_tokens": 56872235.0,
      "step": 32775
    },
    {
      "entropy": 5.72784390449524,
      "epoch": 2.5503987551060106,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004344944100144738,
      "loss": 4.9794,
      "mean_token_accuracy": 0.2040374055504799,
      "num_tokens": 56881506.0,
      "step": 32780
    },
    {
      "entropy": 5.710367155075073,
      "epoch": 2.550787784477728,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043447485974944765,
      "loss": 4.9399,
      "mean_token_accuracy": 0.21292935758829118,
      "num_tokens": 56890218.0,
      "step": 32785
    },
    {
      "entropy": 5.707914733886719,
      "epoch": 2.5511768138494455,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043445530706464437,
      "loss": 4.974,
      "mean_token_accuracy": 0.20519142597913742,
      "num_tokens": 56898037.0,
      "step": 32790
    },
    {
      "entropy": 5.635830450057983,
      "epoch": 2.551565843221163,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004344357519603606,
      "loss": 4.8726,
      "mean_token_accuracy": 0.2113183245062828,
      "num_tokens": 56906892.0,
      "step": 32795
    },
    {
      "entropy": 5.658908033370972,
      "epoch": 2.551954872592881,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004344161944368931,
      "loss": 4.8093,
      "mean_token_accuracy": 0.2193207174539566,
      "num_tokens": 56915286.0,
      "step": 32800
    },
    {
      "entropy": 5.66134557723999,
      "epoch": 2.5523439019645986,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004343966344945385,
      "loss": 4.923,
      "mean_token_accuracy": 0.21246115416288375,
      "num_tokens": 56923670.0,
      "step": 32805
    },
    {
      "entropy": 5.767597436904907,
      "epoch": 2.552732931336316,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043437707213359375,
      "loss": 5.0659,
      "mean_token_accuracy": 0.1997691884636879,
      "num_tokens": 56932092.0,
      "step": 32810
    },
    {
      "entropy": 5.732899236679077,
      "epoch": 2.5531219607080335,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043435750735435554,
      "loss": 5.0287,
      "mean_token_accuracy": 0.20661164224147796,
      "num_tokens": 56940938.0,
      "step": 32815
    },
    {
      "entropy": 5.750022745132446,
      "epoch": 2.553510990079751,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043433794015712085,
      "loss": 4.9922,
      "mean_token_accuracy": 0.20382631421089173,
      "num_tokens": 56950001.0,
      "step": 32820
    },
    {
      "entropy": 5.747639131546021,
      "epoch": 2.5539000194514685,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004343183705421865,
      "loss": 5.0381,
      "mean_token_accuracy": 0.204196035861969,
      "num_tokens": 56958898.0,
      "step": 32825
    },
    {
      "entropy": 5.737569522857666,
      "epoch": 2.554289048823186,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004342987985098495,
      "loss": 4.9493,
      "mean_token_accuracy": 0.21158340126276015,
      "num_tokens": 56967586.0,
      "step": 32830
    },
    {
      "entropy": 5.647342538833618,
      "epoch": 2.554678078194904,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043427922406040676,
      "loss": 4.8767,
      "mean_token_accuracy": 0.20882456749677658,
      "num_tokens": 56975781.0,
      "step": 32835
    },
    {
      "entropy": 5.75801191329956,
      "epoch": 2.555067107566621,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043425964719415525,
      "loss": 5.0013,
      "mean_token_accuracy": 0.20718843191862107,
      "num_tokens": 56984230.0,
      "step": 32840
    },
    {
      "entropy": 5.736223030090332,
      "epoch": 2.555456136938339,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043424006791139213,
      "loss": 4.9482,
      "mean_token_accuracy": 0.21074491143226623,
      "num_tokens": 56992100.0,
      "step": 32845
    },
    {
      "entropy": 5.736366510391235,
      "epoch": 2.5558451663100565,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043422048621241443,
      "loss": 5.0438,
      "mean_token_accuracy": 0.2033479019999504,
      "num_tokens": 57000407.0,
      "step": 32850
    },
    {
      "entropy": 5.751066827774048,
      "epoch": 2.556234195681774,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043420090209751933,
      "loss": 5.0299,
      "mean_token_accuracy": 0.20587778687477112,
      "num_tokens": 57008781.0,
      "step": 32855
    },
    {
      "entropy": 5.705770587921142,
      "epoch": 2.5566232250534915,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004341813155670039,
      "loss": 4.9103,
      "mean_token_accuracy": 0.2176198422908783,
      "num_tokens": 57017568.0,
      "step": 32860
    },
    {
      "entropy": 5.709763431549073,
      "epoch": 2.557012254425209,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004341617266211654,
      "loss": 4.8683,
      "mean_token_accuracy": 0.21652187407016754,
      "num_tokens": 57025674.0,
      "step": 32865
    },
    {
      "entropy": 5.717769527435303,
      "epoch": 2.557401283796927,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000434142135260301,
      "loss": 4.9558,
      "mean_token_accuracy": 0.20812390893697738,
      "num_tokens": 57033657.0,
      "step": 32870
    },
    {
      "entropy": 5.734671783447266,
      "epoch": 2.557790313168644,
      "grad_norm": 1.234375,
      "learning_rate": 0.000434122541484708,
      "loss": 5.0432,
      "mean_token_accuracy": 0.2031599685549736,
      "num_tokens": 57043057.0,
      "step": 32875
    },
    {
      "entropy": 5.746889305114746,
      "epoch": 2.558179342540362,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004341029452946837,
      "loss": 4.9855,
      "mean_token_accuracy": 0.20213370770215988,
      "num_tokens": 57051721.0,
      "step": 32880
    },
    {
      "entropy": 5.7522646427154545,
      "epoch": 2.558568371912079,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004340833466905254,
      "loss": 4.9909,
      "mean_token_accuracy": 0.20609702318906784,
      "num_tokens": 57060223.0,
      "step": 32885
    },
    {
      "entropy": 5.752629041671753,
      "epoch": 2.558957401283797,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043406374567253055,
      "loss": 4.9849,
      "mean_token_accuracy": 0.2062977984547615,
      "num_tokens": 57068408.0,
      "step": 32890
    },
    {
      "entropy": 5.747454500198364,
      "epoch": 2.5593464306555145,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004340441422409965,
      "loss": 5.0495,
      "mean_token_accuracy": 0.19594773054122924,
      "num_tokens": 57076784.0,
      "step": 32895
    },
    {
      "entropy": 5.769601964950562,
      "epoch": 2.559735460027232,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004340245363962209,
      "loss": 5.0065,
      "mean_token_accuracy": 0.20425455868244172,
      "num_tokens": 57084745.0,
      "step": 32900
    },
    {
      "entropy": 5.680600309371949,
      "epoch": 2.56012448939895,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043400492813850083,
      "loss": 4.9452,
      "mean_token_accuracy": 0.2084403946995735,
      "num_tokens": 57093395.0,
      "step": 32905
    },
    {
      "entropy": 5.68587098121643,
      "epoch": 2.560513518770667,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043398531746813426,
      "loss": 4.916,
      "mean_token_accuracy": 0.20208196043968202,
      "num_tokens": 57101665.0,
      "step": 32910
    },
    {
      "entropy": 5.6256715774536135,
      "epoch": 2.560902548142385,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043396570438541845,
      "loss": 4.885,
      "mean_token_accuracy": 0.2043420046567917,
      "num_tokens": 57110328.0,
      "step": 32915
    },
    {
      "entropy": 5.718499183654785,
      "epoch": 2.561291577514102,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004339460888906512,
      "loss": 5.0109,
      "mean_token_accuracy": 0.2003476142883301,
      "num_tokens": 57118834.0,
      "step": 32920
    },
    {
      "entropy": 5.796879434585572,
      "epoch": 2.5616806068858198,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004339264709841299,
      "loss": 5.0796,
      "mean_token_accuracy": 0.1950897142291069,
      "num_tokens": 57127672.0,
      "step": 32925
    },
    {
      "entropy": 5.692834138870239,
      "epoch": 2.5620696362575375,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043390685066615244,
      "loss": 4.9154,
      "mean_token_accuracy": 0.2142313927412033,
      "num_tokens": 57136209.0,
      "step": 32930
    },
    {
      "entropy": 5.706555557250977,
      "epoch": 2.562458665629255,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004338872279370164,
      "loss": 4.9723,
      "mean_token_accuracy": 0.20826450288295745,
      "num_tokens": 57144427.0,
      "step": 32935
    },
    {
      "entropy": 5.714497375488281,
      "epoch": 2.5628476950009724,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004338676027970196,
      "loss": 5.0799,
      "mean_token_accuracy": 0.19217242449522018,
      "num_tokens": 57152860.0,
      "step": 32940
    },
    {
      "entropy": 5.739664459228516,
      "epoch": 2.56323672437269,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043384797524645977,
      "loss": 4.9651,
      "mean_token_accuracy": 0.21088458150625228,
      "num_tokens": 57162397.0,
      "step": 32945
    },
    {
      "entropy": 5.715655326843262,
      "epoch": 2.563625753744408,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004338283452856348,
      "loss": 4.9463,
      "mean_token_accuracy": 0.21446819454431534,
      "num_tokens": 57171178.0,
      "step": 32950
    },
    {
      "entropy": 5.7391314029693605,
      "epoch": 2.564014783116125,
      "grad_norm": 1.25,
      "learning_rate": 0.0004338087129148425,
      "loss": 4.9614,
      "mean_token_accuracy": 0.20871428698301314,
      "num_tokens": 57178715.0,
      "step": 32955
    },
    {
      "entropy": 5.788486099243164,
      "epoch": 2.5644038124878428,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043378907813438066,
      "loss": 5.0361,
      "mean_token_accuracy": 0.19637156426906585,
      "num_tokens": 57187239.0,
      "step": 32960
    },
    {
      "entropy": 5.735196781158447,
      "epoch": 2.5647928418595605,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043376944094454734,
      "loss": 4.9694,
      "mean_token_accuracy": 0.2023715481162071,
      "num_tokens": 57195655.0,
      "step": 32965
    },
    {
      "entropy": 5.685262155532837,
      "epoch": 2.565181871231278,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00043374980134564044,
      "loss": 4.9974,
      "mean_token_accuracy": 0.20280040353536605,
      "num_tokens": 57205026.0,
      "step": 32970
    },
    {
      "entropy": 5.6915048122406,
      "epoch": 2.5655709006029954,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043373015933795804,
      "loss": 4.9645,
      "mean_token_accuracy": 0.2075129821896553,
      "num_tokens": 57213344.0,
      "step": 32975
    },
    {
      "entropy": 5.733766269683838,
      "epoch": 2.565959929974713,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004337105149217981,
      "loss": 4.9587,
      "mean_token_accuracy": 0.20439160168170928,
      "num_tokens": 57222174.0,
      "step": 32980
    },
    {
      "entropy": 5.675614643096924,
      "epoch": 2.566348959346431,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043369086809745875,
      "loss": 4.9545,
      "mean_token_accuracy": 0.20837066173553467,
      "num_tokens": 57231099.0,
      "step": 32985
    },
    {
      "entropy": 5.722344636917114,
      "epoch": 2.566737988718148,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043367121886523795,
      "loss": 5.0072,
      "mean_token_accuracy": 0.19933224767446517,
      "num_tokens": 57239143.0,
      "step": 32990
    },
    {
      "entropy": 5.663932943344117,
      "epoch": 2.5671270180898658,
      "grad_norm": 1.1875,
      "learning_rate": 0.000433651567225434,
      "loss": 4.9831,
      "mean_token_accuracy": 0.20635493993759155,
      "num_tokens": 57247971.0,
      "step": 32995
    },
    {
      "entropy": 5.703922080993652,
      "epoch": 2.5675160474615835,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004336319131783451,
      "loss": 4.9698,
      "mean_token_accuracy": 0.20716686248779298,
      "num_tokens": 57256257.0,
      "step": 33000
    },
    {
      "epoch": 2.5675160474615835,
      "eval_entropy": 5.478959092368749,
      "eval_loss": 5.059050559997559,
      "eval_mean_token_accuracy": 0.2121637366733606,
      "eval_num_tokens": 57256257.0,
      "eval_runtime": 21.655,
      "eval_samples_per_second": 1919.094,
      "eval_steps_per_second": 239.898,
      "step": 33000
    }
  ],
  "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": 1.153513868166144e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}