{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.324012427873946,
  "eval_steps": 500,
  "global_step": 6000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 7.78931941986084,
      "epoch": 0.004438526409232135,
      "grad_norm": 1.0390625,
      "learning_rate": 2e-06,
      "loss": 7.4769,
      "mean_token_accuracy": 0.08522442281246186,
      "num_tokens": 10699.0,
      "step": 5
    },
    {
      "entropy": 7.787262725830078,
      "epoch": 0.00887705281846427,
      "grad_norm": 1.0546875,
      "learning_rate": 4.5e-06,
      "loss": 7.4824,
      "mean_token_accuracy": 0.08313449285924435,
      "num_tokens": 20868.0,
      "step": 10
    },
    {
      "entropy": 7.764151668548584,
      "epoch": 0.013315579227696404,
      "grad_norm": 1.046875,
      "learning_rate": 7e-06,
      "loss": 7.4352,
      "mean_token_accuracy": 0.08592776954174042,
      "num_tokens": 31565.0,
      "step": 15
    },
    {
      "entropy": 7.8020752429962155,
      "epoch": 0.01775410563692854,
      "grad_norm": 0.91796875,
      "learning_rate": 9.5e-06,
      "loss": 7.5052,
      "mean_token_accuracy": 0.08421143889427185,
      "num_tokens": 42396.0,
      "step": 20
    },
    {
      "entropy": 7.809156513214111,
      "epoch": 0.022192632046160676,
      "grad_norm": 1.015625,
      "learning_rate": 1.2e-05,
      "loss": 7.5029,
      "mean_token_accuracy": 0.08358340412378311,
      "num_tokens": 52422.0,
      "step": 25
    },
    {
      "entropy": 7.804666757583618,
      "epoch": 0.02663115845539281,
      "grad_norm": 1.0078125,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 7.4415,
      "mean_token_accuracy": 0.0846289798617363,
      "num_tokens": 63167.0,
      "step": 30
    },
    {
      "entropy": 7.820871639251709,
      "epoch": 0.031069684864624945,
      "grad_norm": 0.94921875,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 7.4654,
      "mean_token_accuracy": 0.08765448182821274,
      "num_tokens": 73620.0,
      "step": 35
    },
    {
      "entropy": 7.827526521682739,
      "epoch": 0.03550821127385708,
      "grad_norm": 0.9140625,
      "learning_rate": 1.95e-05,
      "loss": 7.3795,
      "mean_token_accuracy": 0.08707336336374283,
      "num_tokens": 83176.0,
      "step": 40
    },
    {
      "entropy": 7.789561748504639,
      "epoch": 0.03994673768308921,
      "grad_norm": 0.9921875,
      "learning_rate": 2.2e-05,
      "loss": 7.435,
      "mean_token_accuracy": 0.0886497862637043,
      "num_tokens": 94238.0,
      "step": 45
    },
    {
      "entropy": 7.7829310417175295,
      "epoch": 0.04438526409232135,
      "grad_norm": 0.84375,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 7.4332,
      "mean_token_accuracy": 0.08750618845224381,
      "num_tokens": 104855.0,
      "step": 50
    },
    {
      "entropy": 7.799046611785888,
      "epoch": 0.048823790501553485,
      "grad_norm": 0.90625,
      "learning_rate": 2.7e-05,
      "loss": 7.4366,
      "mean_token_accuracy": 0.09203671216964722,
      "num_tokens": 113812.0,
      "step": 55
    },
    {
      "entropy": 7.795975160598755,
      "epoch": 0.05326231691078562,
      "grad_norm": 0.84375,
      "learning_rate": 2.95e-05,
      "loss": 7.4903,
      "mean_token_accuracy": 0.08637641668319702,
      "num_tokens": 123392.0,
      "step": 60
    },
    {
      "entropy": 7.796218681335449,
      "epoch": 0.05770084332001776,
      "grad_norm": 0.8984375,
      "learning_rate": 3.2e-05,
      "loss": 7.4831,
      "mean_token_accuracy": 0.08530961498618125,
      "num_tokens": 132764.0,
      "step": 65
    },
    {
      "entropy": 7.810373401641845,
      "epoch": 0.06213936972924989,
      "grad_norm": 1.0625,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 7.434,
      "mean_token_accuracy": 0.08844730108976365,
      "num_tokens": 142965.0,
      "step": 70
    },
    {
      "entropy": 7.803947973251343,
      "epoch": 0.06657789613848203,
      "grad_norm": 0.90234375,
      "learning_rate": 3.7e-05,
      "loss": 7.4512,
      "mean_token_accuracy": 0.08455823883414268,
      "num_tokens": 152939.0,
      "step": 75
    },
    {
      "entropy": 7.8467789649963375,
      "epoch": 0.07101642254771416,
      "grad_norm": 0.82421875,
      "learning_rate": 3.95e-05,
      "loss": 7.5052,
      "mean_token_accuracy": 0.08405138589441777,
      "num_tokens": 164455.0,
      "step": 80
    },
    {
      "entropy": 7.778741502761841,
      "epoch": 0.0754549489569463,
      "grad_norm": 0.90234375,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 7.4599,
      "mean_token_accuracy": 0.08571180626749993,
      "num_tokens": 174110.0,
      "step": 85
    },
    {
      "entropy": 7.789217472076416,
      "epoch": 0.07989347536617843,
      "grad_norm": 0.8671875,
      "learning_rate": 4.45e-05,
      "loss": 7.4293,
      "mean_token_accuracy": 0.0909264013171196,
      "num_tokens": 184160.0,
      "step": 90
    },
    {
      "entropy": 7.8063740730285645,
      "epoch": 0.08433200177541056,
      "grad_norm": 0.87109375,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 7.4067,
      "mean_token_accuracy": 0.08633613213896751,
      "num_tokens": 194797.0,
      "step": 95
    },
    {
      "entropy": 7.8098077297210695,
      "epoch": 0.0887705281846427,
      "grad_norm": 0.87109375,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 7.4494,
      "mean_token_accuracy": 0.08306959643959999,
      "num_tokens": 204108.0,
      "step": 100
    },
    {
      "entropy": 7.793222093582154,
      "epoch": 0.09320905459387484,
      "grad_norm": 1.0625,
      "learning_rate": 5.2e-05,
      "loss": 7.4157,
      "mean_token_accuracy": 0.08446534052491188,
      "num_tokens": 214945.0,
      "step": 105
    },
    {
      "entropy": 7.810083961486816,
      "epoch": 0.09764758100310697,
      "grad_norm": 0.87890625,
      "learning_rate": 5.45e-05,
      "loss": 7.5058,
      "mean_token_accuracy": 0.08473781980574131,
      "num_tokens": 226787.0,
      "step": 110
    },
    {
      "entropy": 7.764282941818237,
      "epoch": 0.1020861074123391,
      "grad_norm": 0.88671875,
      "learning_rate": 5.7e-05,
      "loss": 7.3897,
      "mean_token_accuracy": 0.08418168947100639,
      "num_tokens": 237273.0,
      "step": 115
    },
    {
      "entropy": 7.864174175262451,
      "epoch": 0.10652463382157124,
      "grad_norm": 1.0078125,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 7.4156,
      "mean_token_accuracy": 0.09291890189051628,
      "num_tokens": 246793.0,
      "step": 120
    },
    {
      "entropy": 7.779165315628052,
      "epoch": 0.11096316023080337,
      "grad_norm": 0.83984375,
      "learning_rate": 6.2e-05,
      "loss": 7.4275,
      "mean_token_accuracy": 0.0859360195696354,
      "num_tokens": 258022.0,
      "step": 125
    },
    {
      "entropy": 7.7436638355255125,
      "epoch": 0.11540168664003551,
      "grad_norm": 0.90625,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.4169,
      "mean_token_accuracy": 0.08928811252117157,
      "num_tokens": 267670.0,
      "step": 130
    },
    {
      "entropy": 7.7615663528442385,
      "epoch": 0.11984021304926765,
      "grad_norm": 1.0078125,
      "learning_rate": 6.7e-05,
      "loss": 7.4734,
      "mean_token_accuracy": 0.08662445694208146,
      "num_tokens": 278441.0,
      "step": 135
    },
    {
      "entropy": 7.868949317932129,
      "epoch": 0.12427873945849978,
      "grad_norm": 0.98046875,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.44,
      "mean_token_accuracy": 0.0834168739616871,
      "num_tokens": 288501.0,
      "step": 140
    },
    {
      "entropy": 7.918918085098267,
      "epoch": 0.1287172658677319,
      "grad_norm": 1.0078125,
      "learning_rate": 7.2e-05,
      "loss": 7.5111,
      "mean_token_accuracy": 0.08065465465188026,
      "num_tokens": 299557.0,
      "step": 145
    },
    {
      "entropy": 7.7995758056640625,
      "epoch": 0.13315579227696406,
      "grad_norm": 0.8828125,
      "learning_rate": 7.45e-05,
      "loss": 7.4013,
      "mean_token_accuracy": 0.09215340986847878,
      "num_tokens": 309221.0,
      "step": 150
    },
    {
      "entropy": 7.789921569824219,
      "epoch": 0.13759431868619618,
      "grad_norm": 0.9609375,
      "learning_rate": 7.7e-05,
      "loss": 7.3926,
      "mean_token_accuracy": 0.0882001981139183,
      "num_tokens": 318890.0,
      "step": 155
    },
    {
      "entropy": 7.7831714153289795,
      "epoch": 0.14203284509542832,
      "grad_norm": 0.96484375,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.3731,
      "mean_token_accuracy": 0.08587820529937744,
      "num_tokens": 329547.0,
      "step": 160
    },
    {
      "entropy": 7.732769966125488,
      "epoch": 0.14647137150466044,
      "grad_norm": 0.8203125,
      "learning_rate": 8.2e-05,
      "loss": 7.4036,
      "mean_token_accuracy": 0.09170155450701714,
      "num_tokens": 339819.0,
      "step": 165
    },
    {
      "entropy": 7.838970375061035,
      "epoch": 0.1509098979138926,
      "grad_norm": 0.96875,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.4284,
      "mean_token_accuracy": 0.09364147260785102,
      "num_tokens": 349132.0,
      "step": 170
    },
    {
      "entropy": 7.720353078842163,
      "epoch": 0.15534842432312473,
      "grad_norm": 0.98046875,
      "learning_rate": 8.7e-05,
      "loss": 7.4269,
      "mean_token_accuracy": 0.08701886683702469,
      "num_tokens": 358950.0,
      "step": 175
    },
    {
      "entropy": 7.802455282211303,
      "epoch": 0.15978695073235685,
      "grad_norm": 0.93359375,
      "learning_rate": 8.95e-05,
      "loss": 7.3911,
      "mean_token_accuracy": 0.08798973783850669,
      "num_tokens": 368710.0,
      "step": 180
    },
    {
      "entropy": 7.7396303653717045,
      "epoch": 0.164225477141589,
      "grad_norm": 1.03125,
      "learning_rate": 9.2e-05,
      "loss": 7.4086,
      "mean_token_accuracy": 0.09115821123123169,
      "num_tokens": 377388.0,
      "step": 185
    },
    {
      "entropy": 7.784192132949829,
      "epoch": 0.16866400355082112,
      "grad_norm": 1.0078125,
      "learning_rate": 9.45e-05,
      "loss": 7.368,
      "mean_token_accuracy": 0.08906726986169815,
      "num_tokens": 387860.0,
      "step": 190
    },
    {
      "entropy": 7.78899359703064,
      "epoch": 0.17310252996005326,
      "grad_norm": 1.03125,
      "learning_rate": 9.7e-05,
      "loss": 7.3988,
      "mean_token_accuracy": 0.09347473680973054,
      "num_tokens": 397737.0,
      "step": 195
    },
    {
      "entropy": 7.7772763729095455,
      "epoch": 0.1775410563692854,
      "grad_norm": 0.93359375,
      "learning_rate": 9.95e-05,
      "loss": 7.4405,
      "mean_token_accuracy": 0.08559705466032028,
      "num_tokens": 408016.0,
      "step": 200
    },
    {
      "entropy": 7.831363105773926,
      "epoch": 0.18197958277851753,
      "grad_norm": 0.90625,
      "learning_rate": 0.000102,
      "loss": 7.4124,
      "mean_token_accuracy": 0.08368377536535263,
      "num_tokens": 417578.0,
      "step": 205
    },
    {
      "entropy": 7.737787818908691,
      "epoch": 0.18641810918774968,
      "grad_norm": 1.03125,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.3308,
      "mean_token_accuracy": 0.08822357207536698,
      "num_tokens": 428746.0,
      "step": 210
    },
    {
      "entropy": 7.837935161590576,
      "epoch": 0.1908566355969818,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000107,
      "loss": 7.39,
      "mean_token_accuracy": 0.08428716734051704,
      "num_tokens": 439500.0,
      "step": 215
    },
    {
      "entropy": 7.771363592147827,
      "epoch": 0.19529516200621394,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0001095,
      "loss": 7.4119,
      "mean_token_accuracy": 0.08383373767137528,
      "num_tokens": 450105.0,
      "step": 220
    },
    {
      "entropy": 7.8077551364898685,
      "epoch": 0.19973368841544606,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000112,
      "loss": 7.3646,
      "mean_token_accuracy": 0.09255110621452331,
      "num_tokens": 459800.0,
      "step": 225
    },
    {
      "entropy": 7.745184326171875,
      "epoch": 0.2041722148246782,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0001145,
      "loss": 7.2759,
      "mean_token_accuracy": 0.09842955991625786,
      "num_tokens": 470371.0,
      "step": 230
    },
    {
      "entropy": 7.753671360015869,
      "epoch": 0.20861074123391035,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.3007,
      "mean_token_accuracy": 0.09646429568529129,
      "num_tokens": 480347.0,
      "step": 235
    },
    {
      "entropy": 7.81769061088562,
      "epoch": 0.21304926764314247,
      "grad_norm": 0.90625,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.3843,
      "mean_token_accuracy": 0.08836668208241463,
      "num_tokens": 491058.0,
      "step": 240
    },
    {
      "entropy": 7.67605414390564,
      "epoch": 0.21748779405237462,
      "grad_norm": 1.125,
      "learning_rate": 0.000122,
      "loss": 7.3048,
      "mean_token_accuracy": 0.08849572688341141,
      "num_tokens": 500826.0,
      "step": 245
    },
    {
      "entropy": 7.790542507171631,
      "epoch": 0.22192632046160674,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001245,
      "loss": 7.284,
      "mean_token_accuracy": 0.08906695693731308,
      "num_tokens": 510952.0,
      "step": 250
    },
    {
      "entropy": 7.7136742115020756,
      "epoch": 0.22636484687083888,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000127,
      "loss": 7.2967,
      "mean_token_accuracy": 0.09439613595604897,
      "num_tokens": 520680.0,
      "step": 255
    },
    {
      "entropy": 7.686872863769532,
      "epoch": 0.23080337328007103,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0001295,
      "loss": 7.378,
      "mean_token_accuracy": 0.0834910437464714,
      "num_tokens": 531400.0,
      "step": 260
    },
    {
      "entropy": 7.752029180526733,
      "epoch": 0.23524189968930315,
      "grad_norm": 1.3125,
      "learning_rate": 0.000132,
      "loss": 7.4008,
      "mean_token_accuracy": 0.08466090187430382,
      "num_tokens": 541657.0,
      "step": 265
    },
    {
      "entropy": 7.805071783065796,
      "epoch": 0.2396804260985353,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.3264,
      "mean_token_accuracy": 0.09501602202653885,
      "num_tokens": 550609.0,
      "step": 270
    },
    {
      "entropy": 7.560798215866089,
      "epoch": 0.2441189525077674,
      "grad_norm": 1.046875,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.292,
      "mean_token_accuracy": 0.09585189595818519,
      "num_tokens": 561482.0,
      "step": 275
    },
    {
      "entropy": 7.8325278759002686,
      "epoch": 0.24855747891699956,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001395,
      "loss": 7.2829,
      "mean_token_accuracy": 0.09389769136905671,
      "num_tokens": 570582.0,
      "step": 280
    },
    {
      "entropy": 7.7413294315338135,
      "epoch": 0.2529960053262317,
      "grad_norm": 1.015625,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.392,
      "mean_token_accuracy": 0.0857204593718052,
      "num_tokens": 581206.0,
      "step": 285
    },
    {
      "entropy": 7.825753402709961,
      "epoch": 0.2574345317354638,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001445,
      "loss": 7.193,
      "mean_token_accuracy": 0.09314141124486923,
      "num_tokens": 592122.0,
      "step": 290
    },
    {
      "entropy": 7.66306438446045,
      "epoch": 0.26187305814469597,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000147,
      "loss": 7.3151,
      "mean_token_accuracy": 0.08986739963293075,
      "num_tokens": 602919.0,
      "step": 295
    },
    {
      "entropy": 7.771809720993042,
      "epoch": 0.2663115845539281,
      "grad_norm": 1.0625,
      "learning_rate": 0.0001495,
      "loss": 7.2954,
      "mean_token_accuracy": 0.09381957724690437,
      "num_tokens": 612864.0,
      "step": 300
    },
    {
      "entropy": 7.627486038208008,
      "epoch": 0.2707501109631602,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000152,
      "loss": 7.2278,
      "mean_token_accuracy": 0.09549270346760749,
      "num_tokens": 622682.0,
      "step": 305
    },
    {
      "entropy": 7.6965526103973385,
      "epoch": 0.27518863737239235,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.3782,
      "mean_token_accuracy": 0.09121644049882889,
      "num_tokens": 634017.0,
      "step": 310
    },
    {
      "entropy": 7.837769556045532,
      "epoch": 0.2796271637816245,
      "grad_norm": 1.109375,
      "learning_rate": 0.000157,
      "loss": 7.4513,
      "mean_token_accuracy": 0.08749650418758392,
      "num_tokens": 644132.0,
      "step": 315
    },
    {
      "entropy": 7.789153337478638,
      "epoch": 0.28406569019085665,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0001595,
      "loss": 7.3349,
      "mean_token_accuracy": 0.09567861631512642,
      "num_tokens": 655246.0,
      "step": 320
    },
    {
      "entropy": 7.75645751953125,
      "epoch": 0.2885042166000888,
      "grad_norm": 1.125,
      "learning_rate": 0.000162,
      "loss": 7.3554,
      "mean_token_accuracy": 0.08929399326443672,
      "num_tokens": 664899.0,
      "step": 325
    },
    {
      "entropy": 7.748797941207886,
      "epoch": 0.2929427430093209,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.375,
      "mean_token_accuracy": 0.09080832377076149,
      "num_tokens": 675989.0,
      "step": 330
    },
    {
      "entropy": 7.664048528671264,
      "epoch": 0.29738126941855303,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.3221,
      "mean_token_accuracy": 0.09584744423627853,
      "num_tokens": 687759.0,
      "step": 335
    },
    {
      "entropy": 7.7706445217132565,
      "epoch": 0.3018197958277852,
      "grad_norm": 1.015625,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.2649,
      "mean_token_accuracy": 0.09178439974784851,
      "num_tokens": 696943.0,
      "step": 340
    },
    {
      "entropy": 7.699291706085205,
      "epoch": 0.3062583222370173,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.2908,
      "mean_token_accuracy": 0.0902982622385025,
      "num_tokens": 708139.0,
      "step": 345
    },
    {
      "entropy": 7.69988489151001,
      "epoch": 0.31069684864624947,
      "grad_norm": 1.125,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.2561,
      "mean_token_accuracy": 0.09475043341517449,
      "num_tokens": 718000.0,
      "step": 350
    },
    {
      "entropy": 7.659535074234009,
      "epoch": 0.31513537505548156,
      "grad_norm": 1.140625,
      "learning_rate": 0.000177,
      "loss": 7.2602,
      "mean_token_accuracy": 0.09070580154657364,
      "num_tokens": 728188.0,
      "step": 355
    },
    {
      "entropy": 7.705199956893921,
      "epoch": 0.3195739014647137,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0001795,
      "loss": 7.291,
      "mean_token_accuracy": 0.09317795038223267,
      "num_tokens": 738590.0,
      "step": 360
    },
    {
      "entropy": 7.739383554458618,
      "epoch": 0.32401242787394585,
      "grad_norm": 1.109375,
      "learning_rate": 0.000182,
      "loss": 7.3736,
      "mean_token_accuracy": 0.09111551642417907,
      "num_tokens": 748628.0,
      "step": 365
    },
    {
      "entropy": 7.654271173477173,
      "epoch": 0.328450954283178,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001845,
      "loss": 7.2501,
      "mean_token_accuracy": 0.09355314895510673,
      "num_tokens": 760126.0,
      "step": 370
    },
    {
      "entropy": 7.799070549011231,
      "epoch": 0.33288948069241014,
      "grad_norm": 1.140625,
      "learning_rate": 0.000187,
      "loss": 7.3846,
      "mean_token_accuracy": 0.08661651834845543,
      "num_tokens": 770255.0,
      "step": 375
    },
    {
      "entropy": 7.63041615486145,
      "epoch": 0.33732800710164224,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0001895,
      "loss": 7.3517,
      "mean_token_accuracy": 0.08633785173296929,
      "num_tokens": 780744.0,
      "step": 380
    },
    {
      "entropy": 7.73739104270935,
      "epoch": 0.3417665335108744,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000192,
      "loss": 7.2372,
      "mean_token_accuracy": 0.09704277291893959,
      "num_tokens": 790617.0,
      "step": 385
    },
    {
      "entropy": 7.6727227687835695,
      "epoch": 0.34620505992010653,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0001945,
      "loss": 7.2751,
      "mean_token_accuracy": 0.09424345120787621,
      "num_tokens": 800751.0,
      "step": 390
    },
    {
      "entropy": 7.7567243576049805,
      "epoch": 0.3506435863293387,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.3627,
      "mean_token_accuracy": 0.09301683828234672,
      "num_tokens": 810596.0,
      "step": 395
    },
    {
      "entropy": 7.74293794631958,
      "epoch": 0.3550821127385708,
      "grad_norm": 1.078125,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.3311,
      "mean_token_accuracy": 0.08610923066735268,
      "num_tokens": 821001.0,
      "step": 400
    },
    {
      "entropy": 7.658763313293457,
      "epoch": 0.3595206391478029,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000202,
      "loss": 7.2313,
      "mean_token_accuracy": 0.09779016748070717,
      "num_tokens": 831653.0,
      "step": 405
    },
    {
      "entropy": 7.616789484024048,
      "epoch": 0.36395916555703506,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00020449999999999998,
      "loss": 7.1687,
      "mean_token_accuracy": 0.10131467953324318,
      "num_tokens": 842953.0,
      "step": 410
    },
    {
      "entropy": 7.563587284088134,
      "epoch": 0.3683976919662672,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000207,
      "loss": 7.2395,
      "mean_token_accuracy": 0.0917266458272934,
      "num_tokens": 853521.0,
      "step": 415
    },
    {
      "entropy": 7.62612771987915,
      "epoch": 0.37283621837549935,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0002095,
      "loss": 7.1909,
      "mean_token_accuracy": 0.09840409010648728,
      "num_tokens": 863525.0,
      "step": 420
    },
    {
      "entropy": 7.7186483383178714,
      "epoch": 0.37727474478473144,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000212,
      "loss": 7.2794,
      "mean_token_accuracy": 0.09545301347970962,
      "num_tokens": 874101.0,
      "step": 425
    },
    {
      "entropy": 7.579249715805053,
      "epoch": 0.3817132711939636,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002145,
      "loss": 7.1828,
      "mean_token_accuracy": 0.1006378948688507,
      "num_tokens": 884422.0,
      "step": 430
    },
    {
      "entropy": 7.561653470993042,
      "epoch": 0.38615179760319573,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00021700000000000002,
      "loss": 7.1764,
      "mean_token_accuracy": 0.09805170893669128,
      "num_tokens": 894787.0,
      "step": 435
    },
    {
      "entropy": 7.73306303024292,
      "epoch": 0.3905903240124279,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002195,
      "loss": 7.2928,
      "mean_token_accuracy": 0.08978268429636956,
      "num_tokens": 906295.0,
      "step": 440
    },
    {
      "entropy": 7.6030584335327145,
      "epoch": 0.39502885042166,
      "grad_norm": 1.078125,
      "learning_rate": 0.000222,
      "loss": 7.2112,
      "mean_token_accuracy": 0.09668504521250725,
      "num_tokens": 916791.0,
      "step": 445
    },
    {
      "entropy": 7.55758113861084,
      "epoch": 0.3994673768308921,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002245,
      "loss": 7.1147,
      "mean_token_accuracy": 0.09961161985993386,
      "num_tokens": 926383.0,
      "step": 450
    },
    {
      "entropy": 7.644591426849365,
      "epoch": 0.40390590324012426,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00022700000000000002,
      "loss": 7.1943,
      "mean_token_accuracy": 0.10151847004890442,
      "num_tokens": 937319.0,
      "step": 455
    },
    {
      "entropy": 7.615101528167725,
      "epoch": 0.4083444296493564,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022950000000000002,
      "loss": 7.2615,
      "mean_token_accuracy": 0.09779548048973083,
      "num_tokens": 947641.0,
      "step": 460
    },
    {
      "entropy": 7.583732938766479,
      "epoch": 0.41278295605858856,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023200000000000003,
      "loss": 7.1568,
      "mean_token_accuracy": 0.09658500477671624,
      "num_tokens": 957778.0,
      "step": 465
    },
    {
      "entropy": 7.627662897109985,
      "epoch": 0.4172214824678207,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023449999999999998,
      "loss": 7.1428,
      "mean_token_accuracy": 0.09743335545063019,
      "num_tokens": 967619.0,
      "step": 470
    },
    {
      "entropy": 7.599508905410767,
      "epoch": 0.4216600088770528,
      "grad_norm": 1.09375,
      "learning_rate": 0.000237,
      "loss": 7.2186,
      "mean_token_accuracy": 0.09613262861967087,
      "num_tokens": 978301.0,
      "step": 475
    },
    {
      "entropy": 7.5192223072052,
      "epoch": 0.42609853528628494,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002395,
      "loss": 7.2032,
      "mean_token_accuracy": 0.09714524820446968,
      "num_tokens": 989783.0,
      "step": 480
    },
    {
      "entropy": 7.552160596847534,
      "epoch": 0.4305370616955171,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000242,
      "loss": 7.0988,
      "mean_token_accuracy": 0.10628278627991676,
      "num_tokens": 999827.0,
      "step": 485
    },
    {
      "entropy": 7.5833902835845945,
      "epoch": 0.43497558810474923,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002445,
      "loss": 7.2168,
      "mean_token_accuracy": 0.09577535167336464,
      "num_tokens": 1011294.0,
      "step": 490
    },
    {
      "entropy": 7.581695985794068,
      "epoch": 0.4394141145139814,
      "grad_norm": 1.09375,
      "learning_rate": 0.000247,
      "loss": 7.1425,
      "mean_token_accuracy": 0.09786203876137733,
      "num_tokens": 1021672.0,
      "step": 495
    },
    {
      "entropy": 7.592150259017944,
      "epoch": 0.44385264092321347,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002495,
      "loss": 7.1239,
      "mean_token_accuracy": 0.10068506821990013,
      "num_tokens": 1031281.0,
      "step": 500
    },
    {
      "epoch": 0.44385264092321347,
      "eval_entropy": 7.398865165166787,
      "eval_loss": 7.180932521820068,
      "eval_mean_token_accuracy": 0.09428256430669806,
      "eval_num_tokens": 1031281.0,
      "eval_runtime": 2.1123,
      "eval_samples_per_second": 1593.963,
      "eval_steps_per_second": 199.305,
      "step": 500
    },
    {
      "entropy": 7.5125542163848875,
      "epoch": 0.4482911673324456,
      "grad_norm": 1.1875,
      "learning_rate": 0.000252,
      "loss": 7.1865,
      "mean_token_accuracy": 0.09644722416996956,
      "num_tokens": 1042458.0,
      "step": 505
    },
    {
      "entropy": 7.581930828094483,
      "epoch": 0.45272969374167776,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002545,
      "loss": 7.1654,
      "mean_token_accuracy": 0.10110429748892784,
      "num_tokens": 1051645.0,
      "step": 510
    },
    {
      "entropy": 7.556396198272705,
      "epoch": 0.4571682201509099,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000257,
      "loss": 7.1574,
      "mean_token_accuracy": 0.10311417281627655,
      "num_tokens": 1061634.0,
      "step": 515
    },
    {
      "entropy": 7.615333938598633,
      "epoch": 0.46160674656014206,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002595,
      "loss": 7.1778,
      "mean_token_accuracy": 0.10142057165503501,
      "num_tokens": 1071666.0,
      "step": 520
    },
    {
      "entropy": 7.468676328659058,
      "epoch": 0.46604527296937415,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000262,
      "loss": 7.1345,
      "mean_token_accuracy": 0.09515785649418831,
      "num_tokens": 1082652.0,
      "step": 525
    },
    {
      "entropy": 7.533456087112427,
      "epoch": 0.4704837993786063,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026450000000000003,
      "loss": 7.0998,
      "mean_token_accuracy": 0.0970863476395607,
      "num_tokens": 1092396.0,
      "step": 530
    },
    {
      "entropy": 7.495568656921387,
      "epoch": 0.47492232578783844,
      "grad_norm": 1.015625,
      "learning_rate": 0.00026700000000000004,
      "loss": 7.0605,
      "mean_token_accuracy": 0.10546803250908851,
      "num_tokens": 1102504.0,
      "step": 535
    },
    {
      "entropy": 7.440021228790283,
      "epoch": 0.4793608521970706,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026950000000000005,
      "loss": 7.112,
      "mean_token_accuracy": 0.09773821383714676,
      "num_tokens": 1112505.0,
      "step": 540
    },
    {
      "entropy": 7.558693218231201,
      "epoch": 0.48379937860630273,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027200000000000005,
      "loss": 7.1158,
      "mean_token_accuracy": 0.0986550621688366,
      "num_tokens": 1122603.0,
      "step": 545
    },
    {
      "entropy": 7.461544179916382,
      "epoch": 0.4882379050155348,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002745,
      "loss": 7.1284,
      "mean_token_accuracy": 0.09657116010785102,
      "num_tokens": 1132178.0,
      "step": 550
    },
    {
      "entropy": 7.562966060638428,
      "epoch": 0.49267643142476697,
      "grad_norm": 1.171875,
      "learning_rate": 0.000277,
      "loss": 7.0946,
      "mean_token_accuracy": 0.09613885954022408,
      "num_tokens": 1141712.0,
      "step": 555
    },
    {
      "entropy": 7.508441638946533,
      "epoch": 0.4971149578339991,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002795,
      "loss": 7.1157,
      "mean_token_accuracy": 0.0992264598608017,
      "num_tokens": 1151748.0,
      "step": 560
    },
    {
      "entropy": 7.558133125305176,
      "epoch": 0.5015534842432312,
      "grad_norm": 2.265625,
      "learning_rate": 0.00028199999999999997,
      "loss": 7.0116,
      "mean_token_accuracy": 0.10735590234398842,
      "num_tokens": 1161234.0,
      "step": 565
    },
    {
      "entropy": 7.4882372379302975,
      "epoch": 0.5059920106524634,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002845,
      "loss": 7.0915,
      "mean_token_accuracy": 0.10239496752619744,
      "num_tokens": 1171656.0,
      "step": 570
    },
    {
      "entropy": 7.4804298877716064,
      "epoch": 0.5104305370616955,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000287,
      "loss": 7.1359,
      "mean_token_accuracy": 0.09924655333161354,
      "num_tokens": 1182983.0,
      "step": 575
    },
    {
      "entropy": 7.447319459915161,
      "epoch": 0.5148690634709276,
      "grad_norm": 1.25,
      "learning_rate": 0.0002895,
      "loss": 7.1005,
      "mean_token_accuracy": 0.09991766214370727,
      "num_tokens": 1193720.0,
      "step": 580
    },
    {
      "entropy": 7.4760712623596195,
      "epoch": 0.5193075898801598,
      "grad_norm": 1.21875,
      "learning_rate": 0.000292,
      "loss": 7.0304,
      "mean_token_accuracy": 0.10363117456436158,
      "num_tokens": 1203286.0,
      "step": 585
    },
    {
      "entropy": 7.436632061004639,
      "epoch": 0.5237461162893919,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002945,
      "loss": 7.0193,
      "mean_token_accuracy": 0.10448477938771247,
      "num_tokens": 1212875.0,
      "step": 590
    },
    {
      "entropy": 7.479945039749145,
      "epoch": 0.5281846426986241,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000297,
      "loss": 7.0455,
      "mean_token_accuracy": 0.09875646382570266,
      "num_tokens": 1223629.0,
      "step": 595
    },
    {
      "entropy": 7.455782127380371,
      "epoch": 0.5326231691078562,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002995,
      "loss": 7.1663,
      "mean_token_accuracy": 0.10330844521522523,
      "num_tokens": 1233670.0,
      "step": 600
    },
    {
      "entropy": 7.540210819244384,
      "epoch": 0.5370616955170884,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000302,
      "loss": 7.0882,
      "mean_token_accuracy": 0.10033463165163994,
      "num_tokens": 1244140.0,
      "step": 605
    },
    {
      "entropy": 7.42637152671814,
      "epoch": 0.5415002219263204,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003045,
      "loss": 7.0617,
      "mean_token_accuracy": 0.10370045378804207,
      "num_tokens": 1254496.0,
      "step": 610
    },
    {
      "entropy": 7.397940731048584,
      "epoch": 0.5459387483355526,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000307,
      "loss": 7.0262,
      "mean_token_accuracy": 0.10960142239928246,
      "num_tokens": 1265391.0,
      "step": 615
    },
    {
      "entropy": 7.426065683364868,
      "epoch": 0.5503772747447847,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003095,
      "loss": 6.9633,
      "mean_token_accuracy": 0.10688221156597137,
      "num_tokens": 1275857.0,
      "step": 620
    },
    {
      "entropy": 7.255760049819946,
      "epoch": 0.5548158011540169,
      "grad_norm": 1.109375,
      "learning_rate": 0.000312,
      "loss": 6.9785,
      "mean_token_accuracy": 0.10750289410352706,
      "num_tokens": 1285122.0,
      "step": 625
    },
    {
      "entropy": 7.525947952270508,
      "epoch": 0.559254327563249,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003145,
      "loss": 7.0365,
      "mean_token_accuracy": 0.10156842395663261,
      "num_tokens": 1294476.0,
      "step": 630
    },
    {
      "entropy": 7.383149433135986,
      "epoch": 0.5636928539724811,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000317,
      "loss": 7.0386,
      "mean_token_accuracy": 0.10288441628217697,
      "num_tokens": 1304959.0,
      "step": 635
    },
    {
      "entropy": 7.389831638336181,
      "epoch": 0.5681313803817133,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003195,
      "loss": 6.9818,
      "mean_token_accuracy": 0.10231715813279152,
      "num_tokens": 1316121.0,
      "step": 640
    },
    {
      "entropy": 7.392814636230469,
      "epoch": 0.5725699067909454,
      "grad_norm": 1.21875,
      "learning_rate": 0.000322,
      "loss": 7.0793,
      "mean_token_accuracy": 0.10169301480054856,
      "num_tokens": 1326750.0,
      "step": 645
    },
    {
      "entropy": 7.383424234390259,
      "epoch": 0.5770084332001776,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.9712,
      "mean_token_accuracy": 0.10646747648715973,
      "num_tokens": 1337044.0,
      "step": 650
    },
    {
      "entropy": 7.4170256614685055,
      "epoch": 0.5814469596094097,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032700000000000003,
      "loss": 7.0334,
      "mean_token_accuracy": 0.09991655647754669,
      "num_tokens": 1347610.0,
      "step": 655
    },
    {
      "entropy": 7.2603747844696045,
      "epoch": 0.5858854860186418,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.9609,
      "mean_token_accuracy": 0.10539844930171967,
      "num_tokens": 1356521.0,
      "step": 660
    },
    {
      "entropy": 7.422995662689209,
      "epoch": 0.5903240124278739,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.9662,
      "mean_token_accuracy": 0.10543927997350692,
      "num_tokens": 1366741.0,
      "step": 665
    },
    {
      "entropy": 7.369775009155274,
      "epoch": 0.5947625388371061,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033450000000000005,
      "loss": 7.0411,
      "mean_token_accuracy": 0.10072910860180855,
      "num_tokens": 1377018.0,
      "step": 670
    },
    {
      "entropy": 7.43230504989624,
      "epoch": 0.5992010652463382,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000337,
      "loss": 6.959,
      "mean_token_accuracy": 0.10484865829348564,
      "num_tokens": 1386387.0,
      "step": 675
    },
    {
      "entropy": 7.3505573749542235,
      "epoch": 0.6036395916555704,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003395,
      "loss": 6.961,
      "mean_token_accuracy": 0.10673004984855652,
      "num_tokens": 1396049.0,
      "step": 680
    },
    {
      "entropy": 7.355041980743408,
      "epoch": 0.6080781180648025,
      "grad_norm": 1.125,
      "learning_rate": 0.000342,
      "loss": 7.0803,
      "mean_token_accuracy": 0.0983244001865387,
      "num_tokens": 1406952.0,
      "step": 685
    },
    {
      "entropy": 7.410180234909058,
      "epoch": 0.6125166444740346,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.9874,
      "mean_token_accuracy": 0.10184509009122848,
      "num_tokens": 1417146.0,
      "step": 690
    },
    {
      "entropy": 7.331110858917237,
      "epoch": 0.6169551708832668,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000347,
      "loss": 6.9654,
      "mean_token_accuracy": 0.11014403700828553,
      "num_tokens": 1426146.0,
      "step": 695
    },
    {
      "entropy": 7.430603981018066,
      "epoch": 0.6213936972924989,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003495,
      "loss": 7.0644,
      "mean_token_accuracy": 0.10093853399157524,
      "num_tokens": 1437913.0,
      "step": 700
    },
    {
      "entropy": 7.3031325340271,
      "epoch": 0.625832223701731,
      "grad_norm": 1.125,
      "learning_rate": 0.000352,
      "loss": 6.9308,
      "mean_token_accuracy": 0.11181655824184418,
      "num_tokens": 1448329.0,
      "step": 705
    },
    {
      "entropy": 7.382747077941895,
      "epoch": 0.6302707501109631,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003545,
      "loss": 6.9559,
      "mean_token_accuracy": 0.10284734219312668,
      "num_tokens": 1457896.0,
      "step": 710
    },
    {
      "entropy": 7.310059356689453,
      "epoch": 0.6347092765201953,
      "grad_norm": 1.21875,
      "learning_rate": 0.000357,
      "loss": 6.9601,
      "mean_token_accuracy": 0.1071910947561264,
      "num_tokens": 1467705.0,
      "step": 715
    },
    {
      "entropy": 7.25341272354126,
      "epoch": 0.6391478029294274,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003595,
      "loss": 6.9512,
      "mean_token_accuracy": 0.10645409822463989,
      "num_tokens": 1478132.0,
      "step": 720
    },
    {
      "entropy": 7.391849327087402,
      "epoch": 0.6435863293386596,
      "grad_norm": 1.203125,
      "learning_rate": 0.000362,
      "loss": 7.019,
      "mean_token_accuracy": 0.10602135583758354,
      "num_tokens": 1488424.0,
      "step": 725
    },
    {
      "entropy": 7.310156202316284,
      "epoch": 0.6480248557478917,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003645,
      "loss": 6.8812,
      "mean_token_accuracy": 0.11527719348669052,
      "num_tokens": 1498197.0,
      "step": 730
    },
    {
      "entropy": 7.234117460250855,
      "epoch": 0.6524633821571239,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000367,
      "loss": 6.9052,
      "mean_token_accuracy": 0.10894913673400879,
      "num_tokens": 1508242.0,
      "step": 735
    },
    {
      "entropy": 7.232631635665894,
      "epoch": 0.656901908566356,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003695,
      "loss": 6.8983,
      "mean_token_accuracy": 0.11159973964095116,
      "num_tokens": 1517652.0,
      "step": 740
    },
    {
      "entropy": 7.2952375411987305,
      "epoch": 0.6613404349755881,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000372,
      "loss": 6.9347,
      "mean_token_accuracy": 0.10787282809615135,
      "num_tokens": 1527501.0,
      "step": 745
    },
    {
      "entropy": 7.291040468215942,
      "epoch": 0.6657789613848203,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003745,
      "loss": 6.9339,
      "mean_token_accuracy": 0.11132785305380821,
      "num_tokens": 1537640.0,
      "step": 750
    },
    {
      "entropy": 7.20764536857605,
      "epoch": 0.6702174877940523,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000377,
      "loss": 6.8701,
      "mean_token_accuracy": 0.11440131813287735,
      "num_tokens": 1548049.0,
      "step": 755
    },
    {
      "entropy": 7.340287876129151,
      "epoch": 0.6746560142032845,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003795,
      "loss": 7.0474,
      "mean_token_accuracy": 0.10261607840657234,
      "num_tokens": 1558389.0,
      "step": 760
    },
    {
      "entropy": 7.222410678863525,
      "epoch": 0.6790945406125166,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000382,
      "loss": 6.8176,
      "mean_token_accuracy": 0.11613213866949082,
      "num_tokens": 1568489.0,
      "step": 765
    },
    {
      "entropy": 7.2836723804473875,
      "epoch": 0.6835330670217488,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003845,
      "loss": 6.9313,
      "mean_token_accuracy": 0.10658463388681412,
      "num_tokens": 1578430.0,
      "step": 770
    },
    {
      "entropy": 7.352204275131226,
      "epoch": 0.6879715934309809,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038700000000000003,
      "loss": 7.001,
      "mean_token_accuracy": 0.10380481109023094,
      "num_tokens": 1590307.0,
      "step": 775
    },
    {
      "entropy": 7.313673448562622,
      "epoch": 0.6924101198402131,
      "grad_norm": 1.84375,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.8585,
      "mean_token_accuracy": 0.11356580927968025,
      "num_tokens": 1601719.0,
      "step": 780
    },
    {
      "entropy": 7.149180507659912,
      "epoch": 0.6968486462494452,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.8735,
      "mean_token_accuracy": 0.11371317654848098,
      "num_tokens": 1612047.0,
      "step": 785
    },
    {
      "entropy": 7.195708179473877,
      "epoch": 0.7012871726586773,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.8651,
      "mean_token_accuracy": 0.11098539307713509,
      "num_tokens": 1623141.0,
      "step": 790
    },
    {
      "entropy": 7.241322708129883,
      "epoch": 0.7057256990679095,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.8738,
      "mean_token_accuracy": 0.1125923864543438,
      "num_tokens": 1633249.0,
      "step": 795
    },
    {
      "entropy": 7.270007610321045,
      "epoch": 0.7101642254771416,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003995,
      "loss": 6.9612,
      "mean_token_accuracy": 0.10728915557265281,
      "num_tokens": 1644451.0,
      "step": 800
    },
    {
      "entropy": 7.245293235778808,
      "epoch": 0.7146027518863737,
      "grad_norm": 1.140625,
      "learning_rate": 0.000402,
      "loss": 6.8967,
      "mean_token_accuracy": 0.11031619012355805,
      "num_tokens": 1655196.0,
      "step": 805
    },
    {
      "entropy": 7.137471008300781,
      "epoch": 0.7190412782956058,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004045,
      "loss": 6.8781,
      "mean_token_accuracy": 0.11013623625040055,
      "num_tokens": 1665613.0,
      "step": 810
    },
    {
      "entropy": 7.282591247558594,
      "epoch": 0.723479804704838,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.8188,
      "mean_token_accuracy": 0.11500794291496277,
      "num_tokens": 1675116.0,
      "step": 815
    },
    {
      "entropy": 7.219833612442017,
      "epoch": 0.7279183311140701,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004095,
      "loss": 6.9236,
      "mean_token_accuracy": 0.10631008669734002,
      "num_tokens": 1684055.0,
      "step": 820
    },
    {
      "entropy": 7.231753492355347,
      "epoch": 0.7323568575233023,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000412,
      "loss": 6.7797,
      "mean_token_accuracy": 0.12125139310956001,
      "num_tokens": 1694046.0,
      "step": 825
    },
    {
      "entropy": 7.143084859848022,
      "epoch": 0.7367953839325344,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004145,
      "loss": 6.8325,
      "mean_token_accuracy": 0.1153494082391262,
      "num_tokens": 1704133.0,
      "step": 830
    },
    {
      "entropy": 7.14149580001831,
      "epoch": 0.7412339103417666,
      "grad_norm": 1.09375,
      "learning_rate": 0.000417,
      "loss": 6.9275,
      "mean_token_accuracy": 0.11007370427250862,
      "num_tokens": 1714678.0,
      "step": 835
    },
    {
      "entropy": 7.204297304153442,
      "epoch": 0.7456724367509987,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004195,
      "loss": 6.7453,
      "mean_token_accuracy": 0.12288743630051613,
      "num_tokens": 1725145.0,
      "step": 840
    },
    {
      "entropy": 7.071165657043457,
      "epoch": 0.7501109631602308,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000422,
      "loss": 6.8729,
      "mean_token_accuracy": 0.11628773137927055,
      "num_tokens": 1735530.0,
      "step": 845
    },
    {
      "entropy": 7.21942458152771,
      "epoch": 0.7545494895694629,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004245,
      "loss": 6.8964,
      "mean_token_accuracy": 0.11181129217147827,
      "num_tokens": 1746520.0,
      "step": 850
    },
    {
      "entropy": 7.154268980026245,
      "epoch": 0.758988015978695,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000427,
      "loss": 6.8532,
      "mean_token_accuracy": 0.1179992400109768,
      "num_tokens": 1756091.0,
      "step": 855
    },
    {
      "entropy": 7.089676809310913,
      "epoch": 0.7634265423879272,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004295,
      "loss": 6.8561,
      "mean_token_accuracy": 0.10415932089090348,
      "num_tokens": 1765968.0,
      "step": 860
    },
    {
      "entropy": 7.192745923995972,
      "epoch": 0.7678650687971593,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000432,
      "loss": 6.8773,
      "mean_token_accuracy": 0.11299539059400558,
      "num_tokens": 1777409.0,
      "step": 865
    },
    {
      "entropy": 7.130894994735717,
      "epoch": 0.7723035952063915,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004345,
      "loss": 6.8824,
      "mean_token_accuracy": 0.10640934631228446,
      "num_tokens": 1788440.0,
      "step": 870
    },
    {
      "entropy": 7.20861964225769,
      "epoch": 0.7767421216156236,
      "grad_norm": 1.15625,
      "learning_rate": 0.000437,
      "loss": 6.8417,
      "mean_token_accuracy": 0.11558186262845993,
      "num_tokens": 1799145.0,
      "step": 875
    },
    {
      "entropy": 7.05075421333313,
      "epoch": 0.7811806480248558,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004395,
      "loss": 6.808,
      "mean_token_accuracy": 0.116386828571558,
      "num_tokens": 1809366.0,
      "step": 880
    },
    {
      "entropy": 7.129081392288208,
      "epoch": 0.7856191744340879,
      "grad_norm": 1.046875,
      "learning_rate": 0.000442,
      "loss": 6.8875,
      "mean_token_accuracy": 0.10899080038070678,
      "num_tokens": 1820878.0,
      "step": 885
    },
    {
      "entropy": 7.155272579193115,
      "epoch": 0.79005770084332,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004445,
      "loss": 6.8349,
      "mean_token_accuracy": 0.11343251317739486,
      "num_tokens": 1831870.0,
      "step": 890
    },
    {
      "entropy": 7.249894714355468,
      "epoch": 0.7944962272525522,
      "grad_norm": 1.15625,
      "learning_rate": 0.000447,
      "loss": 6.801,
      "mean_token_accuracy": 0.11338023543357849,
      "num_tokens": 1841482.0,
      "step": 895
    },
    {
      "entropy": 6.969785833358765,
      "epoch": 0.7989347536617842,
      "grad_norm": 1.125,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.8102,
      "mean_token_accuracy": 0.11380291059613228,
      "num_tokens": 1851748.0,
      "step": 900
    },
    {
      "entropy": 7.100828695297241,
      "epoch": 0.8033732800710164,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.7389,
      "mean_token_accuracy": 0.1162735216319561,
      "num_tokens": 1862292.0,
      "step": 905
    },
    {
      "entropy": 7.249449729919434,
      "epoch": 0.8078118064802485,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.959,
      "mean_token_accuracy": 0.10582878962159156,
      "num_tokens": 1873825.0,
      "step": 910
    },
    {
      "entropy": 7.126041460037231,
      "epoch": 0.8122503328894807,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.7795,
      "mean_token_accuracy": 0.11718677878379821,
      "num_tokens": 1883824.0,
      "step": 915
    },
    {
      "entropy": 7.029882144927979,
      "epoch": 0.8166888592987128,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.7984,
      "mean_token_accuracy": 0.10695101991295815,
      "num_tokens": 1894236.0,
      "step": 920
    },
    {
      "entropy": 7.212645196914673,
      "epoch": 0.821127385707945,
      "grad_norm": 1.21875,
      "learning_rate": 0.000462,
      "loss": 6.779,
      "mean_token_accuracy": 0.11693326756358147,
      "num_tokens": 1904083.0,
      "step": 925
    },
    {
      "entropy": 7.023255491256714,
      "epoch": 0.8255659121171771,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004645,
      "loss": 6.7907,
      "mean_token_accuracy": 0.11693133264780045,
      "num_tokens": 1914845.0,
      "step": 930
    },
    {
      "entropy": 7.073426914215088,
      "epoch": 0.8300044385264093,
      "grad_norm": 1.03125,
      "learning_rate": 0.000467,
      "loss": 6.7945,
      "mean_token_accuracy": 0.11135339140892028,
      "num_tokens": 1925999.0,
      "step": 935
    },
    {
      "entropy": 7.0789491653442385,
      "epoch": 0.8344429649356414,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004695,
      "loss": 6.694,
      "mean_token_accuracy": 0.1280534103512764,
      "num_tokens": 1935401.0,
      "step": 940
    },
    {
      "entropy": 6.9891969680786135,
      "epoch": 0.8388814913448736,
      "grad_norm": 1.25,
      "learning_rate": 0.000472,
      "loss": 6.7607,
      "mean_token_accuracy": 0.12046314924955367,
      "num_tokens": 1945510.0,
      "step": 945
    },
    {
      "entropy": 7.16357650756836,
      "epoch": 0.8433200177541056,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004745,
      "loss": 6.7473,
      "mean_token_accuracy": 0.11847912147641182,
      "num_tokens": 1956934.0,
      "step": 950
    },
    {
      "entropy": 7.054219198226929,
      "epoch": 0.8477585441633377,
      "grad_norm": 1.03125,
      "learning_rate": 0.000477,
      "loss": 6.7688,
      "mean_token_accuracy": 0.11719952076673508,
      "num_tokens": 1967032.0,
      "step": 955
    },
    {
      "entropy": 7.080799579620361,
      "epoch": 0.8521970705725699,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004795,
      "loss": 6.8584,
      "mean_token_accuracy": 0.11015522852540016,
      "num_tokens": 1978895.0,
      "step": 960
    },
    {
      "entropy": 7.09550929069519,
      "epoch": 0.856635596981802,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000482,
      "loss": 6.816,
      "mean_token_accuracy": 0.11443090364336968,
      "num_tokens": 1990156.0,
      "step": 965
    },
    {
      "entropy": 7.139660930633545,
      "epoch": 0.8610741233910342,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004845,
      "loss": 6.7035,
      "mean_token_accuracy": 0.11342604532837867,
      "num_tokens": 1999643.0,
      "step": 970
    },
    {
      "entropy": 6.957034969329834,
      "epoch": 0.8655126498002663,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000487,
      "loss": 6.7472,
      "mean_token_accuracy": 0.11747978329658508,
      "num_tokens": 2010328.0,
      "step": 975
    },
    {
      "entropy": 7.062716341018676,
      "epoch": 0.8699511762094985,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004895,
      "loss": 6.8272,
      "mean_token_accuracy": 0.11245640963315964,
      "num_tokens": 2021353.0,
      "step": 980
    },
    {
      "entropy": 7.081027889251709,
      "epoch": 0.8743897026187306,
      "grad_norm": 1.140625,
      "learning_rate": 0.000492,
      "loss": 6.7701,
      "mean_token_accuracy": 0.11547347530722618,
      "num_tokens": 2032331.0,
      "step": 985
    },
    {
      "entropy": 7.023243951797485,
      "epoch": 0.8788282290279628,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004945,
      "loss": 6.7011,
      "mean_token_accuracy": 0.1206856571137905,
      "num_tokens": 2041786.0,
      "step": 990
    },
    {
      "entropy": 6.975612211227417,
      "epoch": 0.8832667554371949,
      "grad_norm": 1.03125,
      "learning_rate": 0.000497,
      "loss": 6.6885,
      "mean_token_accuracy": 0.11817316859960555,
      "num_tokens": 2052512.0,
      "step": 995
    },
    {
      "entropy": 7.02327561378479,
      "epoch": 0.8877052818464269,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004995,
      "loss": 6.6295,
      "mean_token_accuracy": 0.12657489329576493,
      "num_tokens": 2062785.0,
      "step": 1000
    },
    {
      "epoch": 0.8877052818464269,
      "eval_entropy": 6.811686369698857,
      "eval_loss": 6.767911434173584,
      "eval_mean_token_accuracy": 0.11571601520772501,
      "eval_num_tokens": 2062785.0,
      "eval_runtime": 2.1114,
      "eval_samples_per_second": 1594.66,
      "eval_steps_per_second": 199.392,
      "step": 1000
    },
    {
      "entropy": 7.073720645904541,
      "epoch": 0.8921438082556591,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999998312369076,
      "loss": 6.7709,
      "mean_token_accuracy": 0.11314490139484405,
      "num_tokens": 2072621.0,
      "step": 1005
    },
    {
      "entropy": 7.060971355438232,
      "epoch": 0.8965823346648912,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999991456372788,
      "loss": 6.8178,
      "mean_token_accuracy": 0.10872112512588501,
      "num_tokens": 2083356.0,
      "step": 1010
    },
    {
      "entropy": 7.0227889060974125,
      "epoch": 0.9010208610741234,
      "grad_norm": 1.203125,
      "learning_rate": 0.000499997932655026,
      "loss": 6.7777,
      "mean_token_accuracy": 0.11077685430645942,
      "num_tokens": 2092787.0,
      "step": 1015
    },
    {
      "entropy": 6.9889302253723145,
      "epoch": 0.9054593874833555,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999961922929925,
      "loss": 6.6739,
      "mean_token_accuracy": 0.11871889233589172,
      "num_tokens": 2103721.0,
      "step": 1020
    },
    {
      "entropy": 6.950629568099975,
      "epoch": 0.9098979138925877,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999939245552573,
      "loss": 6.6956,
      "mean_token_accuracy": 0.12753814980387687,
      "num_tokens": 2115016.0,
      "step": 1025
    },
    {
      "entropy": 6.951031637191773,
      "epoch": 0.9143364403018198,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000499991129447136,
      "loss": 6.6594,
      "mean_token_accuracy": 0.1263865128159523,
      "num_tokens": 2126183.0,
      "step": 1030
    },
    {
      "entropy": 6.928300905227661,
      "epoch": 0.918774966711052,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999878069751803,
      "loss": 6.7069,
      "mean_token_accuracy": 0.11957084909081458,
      "num_tokens": 2135955.0,
      "step": 1035
    },
    {
      "entropy": 6.915622138977051,
      "epoch": 0.9232134931202841,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999839571471776,
      "loss": 6.6919,
      "mean_token_accuracy": 0.11919597014784813,
      "num_tokens": 2146064.0,
      "step": 1040
    },
    {
      "entropy": 6.9707104682922365,
      "epoch": 0.9276520195295161,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004999795799721517,
      "loss": 6.6444,
      "mean_token_accuracy": 0.1262632966041565,
      "num_tokens": 2155707.0,
      "step": 1045
    },
    {
      "entropy": 6.99800386428833,
      "epoch": 0.9320905459387483,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999746754603624,
      "loss": 6.6928,
      "mean_token_accuracy": 0.11961428299546242,
      "num_tokens": 2166037.0,
      "step": 1050
    },
    {
      "entropy": 6.904142761230469,
      "epoch": 0.9365290723479804,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999692436233055,
      "loss": 6.7519,
      "mean_token_accuracy": 0.11929019168019295,
      "num_tokens": 2175993.0,
      "step": 1055
    },
    {
      "entropy": 7.008247995376587,
      "epoch": 0.9409675987572126,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000499963284473713,
      "loss": 6.7099,
      "mean_token_accuracy": 0.12117866948246955,
      "num_tokens": 2186829.0,
      "step": 1060
    },
    {
      "entropy": 6.9378032207489015,
      "epoch": 0.9454061251664447,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999567980255526,
      "loss": 6.665,
      "mean_token_accuracy": 0.12237748578190803,
      "num_tokens": 2195845.0,
      "step": 1065
    },
    {
      "entropy": 7.0276360511779785,
      "epoch": 0.9498446515756769,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999497842940282,
      "loss": 6.6921,
      "mean_token_accuracy": 0.12066913396120071,
      "num_tokens": 2204937.0,
      "step": 1070
    },
    {
      "entropy": 6.900293159484863,
      "epoch": 0.954283177984909,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999422432955794,
      "loss": 6.6703,
      "mean_token_accuracy": 0.12011573910713196,
      "num_tokens": 2214582.0,
      "step": 1075
    },
    {
      "entropy": 6.922044038772583,
      "epoch": 0.9587217043941412,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999341750478818,
      "loss": 6.6229,
      "mean_token_accuracy": 0.1269613318145275,
      "num_tokens": 2225157.0,
      "step": 1080
    },
    {
      "entropy": 6.99785623550415,
      "epoch": 0.9631602308033733,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499925579569847,
      "loss": 6.6932,
      "mean_token_accuracy": 0.12152787148952485,
      "num_tokens": 2234823.0,
      "step": 1085
    },
    {
      "entropy": 6.851722621917725,
      "epoch": 0.9675987572126055,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999164568816219,
      "loss": 6.6802,
      "mean_token_accuracy": 0.119229806214571,
      "num_tokens": 2244666.0,
      "step": 1090
    },
    {
      "entropy": 7.007480764389038,
      "epoch": 0.9720372836218375,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999068070045897,
      "loss": 6.6412,
      "mean_token_accuracy": 0.117543176561594,
      "num_tokens": 2254586.0,
      "step": 1095
    },
    {
      "entropy": 6.836320638656616,
      "epoch": 0.9764758100310696,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499896629961369,
      "loss": 6.6026,
      "mean_token_accuracy": 0.12370306029915809,
      "num_tokens": 2264453.0,
      "step": 1100
    },
    {
      "entropy": 6.999147844314575,
      "epoch": 0.9809143364403018,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000499885925775814,
      "loss": 6.6656,
      "mean_token_accuracy": 0.11901217773556709,
      "num_tokens": 2275170.0,
      "step": 1105
    },
    {
      "entropy": 6.810418939590454,
      "epoch": 0.9853528628495339,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998746944730148,
      "loss": 6.6517,
      "mean_token_accuracy": 0.12251685783267022,
      "num_tokens": 2285448.0,
      "step": 1110
    },
    {
      "entropy": 6.978043413162231,
      "epoch": 0.9897913892587661,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998629360792965,
      "loss": 6.6361,
      "mean_token_accuracy": 0.1255613885819912,
      "num_tokens": 2295673.0,
      "step": 1115
    },
    {
      "entropy": 6.860985517501831,
      "epoch": 0.9942299156679982,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998506506222202,
      "loss": 6.644,
      "mean_token_accuracy": 0.12026142254471779,
      "num_tokens": 2306115.0,
      "step": 1120
    },
    {
      "entropy": 6.872910737991333,
      "epoch": 0.9986684420772304,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004998378381305821,
      "loss": 6.637,
      "mean_token_accuracy": 0.12394563928246498,
      "num_tokens": 2316495.0,
      "step": 1125
    },
    {
      "entropy": 6.93203698264228,
      "epoch": 1.0026631158455392,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004998244986344138,
      "loss": 6.5838,
      "mean_token_accuracy": 0.1257213337553872,
      "num_tokens": 2326409.0,
      "step": 1130
    },
    {
      "entropy": 6.777434873580932,
      "epoch": 1.0071016422547714,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998106321649822,
      "loss": 6.439,
      "mean_token_accuracy": 0.1350421704351902,
      "num_tokens": 2337394.0,
      "step": 1135
    },
    {
      "entropy": 6.8583564281463625,
      "epoch": 1.0115401686640035,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004997962387547893,
      "loss": 6.5089,
      "mean_token_accuracy": 0.12366606816649436,
      "num_tokens": 2347768.0,
      "step": 1140
    },
    {
      "entropy": 6.965995025634766,
      "epoch": 1.0159786950732357,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997813184375723,
      "loss": 6.5134,
      "mean_token_accuracy": 0.12577569335699082,
      "num_tokens": 2357969.0,
      "step": 1145
    },
    {
      "entropy": 6.872522068023682,
      "epoch": 1.0204172214824678,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997658712483034,
      "loss": 6.4866,
      "mean_token_accuracy": 0.1270396701991558,
      "num_tokens": 2367878.0,
      "step": 1150
    },
    {
      "entropy": 6.814556646347046,
      "epoch": 1.0248557478917,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997498972231898,
      "loss": 6.4781,
      "mean_token_accuracy": 0.12578893974423408,
      "num_tokens": 2377072.0,
      "step": 1155
    },
    {
      "entropy": 6.867713165283203,
      "epoch": 1.0292942743009321,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997333963996734,
      "loss": 6.503,
      "mean_token_accuracy": 0.13253186494112015,
      "num_tokens": 2387348.0,
      "step": 1160
    },
    {
      "entropy": 6.912645387649536,
      "epoch": 1.0337328007101643,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997163688164313,
      "loss": 6.4919,
      "mean_token_accuracy": 0.12540172412991524,
      "num_tokens": 2398113.0,
      "step": 1165
    },
    {
      "entropy": 6.738404178619385,
      "epoch": 1.0381713271193964,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996988145133745,
      "loss": 6.3882,
      "mean_token_accuracy": 0.13407822996377944,
      "num_tokens": 2408074.0,
      "step": 1170
    },
    {
      "entropy": 6.699475193023682,
      "epoch": 1.0426098535286286,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996807335316496,
      "loss": 6.3874,
      "mean_token_accuracy": 0.1336548797786236,
      "num_tokens": 2419752.0,
      "step": 1175
    },
    {
      "entropy": 6.843112182617188,
      "epoch": 1.0470483799378607,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996621259136368,
      "loss": 6.4743,
      "mean_token_accuracy": 0.1282840073108673,
      "num_tokens": 2430853.0,
      "step": 1180
    },
    {
      "entropy": 6.741962194442749,
      "epoch": 1.0514869063470929,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996429917029513,
      "loss": 6.4597,
      "mean_token_accuracy": 0.1284678265452385,
      "num_tokens": 2440686.0,
      "step": 1185
    },
    {
      "entropy": 6.87503981590271,
      "epoch": 1.055925432756325,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996233309444424,
      "loss": 6.4862,
      "mean_token_accuracy": 0.1273033231496811,
      "num_tokens": 2450849.0,
      "step": 1190
    },
    {
      "entropy": 6.7747828483581545,
      "epoch": 1.060363959165557,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996031436841934,
      "loss": 6.3842,
      "mean_token_accuracy": 0.13247138634324074,
      "num_tokens": 2460827.0,
      "step": 1195
    },
    {
      "entropy": 6.7094989776611325,
      "epoch": 1.064802485574789,
      "grad_norm": 1.125,
      "learning_rate": 0.0004995824299695219,
      "loss": 6.4723,
      "mean_token_accuracy": 0.1337954096496105,
      "num_tokens": 2471265.0,
      "step": 1200
    },
    {
      "entropy": 6.8431929588317875,
      "epoch": 1.0692410119840212,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995611898489796,
      "loss": 6.3655,
      "mean_token_accuracy": 0.13849280402064323,
      "num_tokens": 2481346.0,
      "step": 1205
    },
    {
      "entropy": 6.77330379486084,
      "epoch": 1.0736795383932534,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995394233723516,
      "loss": 6.4696,
      "mean_token_accuracy": 0.13246190845966338,
      "num_tokens": 2492281.0,
      "step": 1210
    },
    {
      "entropy": 6.771844577789307,
      "epoch": 1.0781180648024855,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995171305906574,
      "loss": 6.397,
      "mean_token_accuracy": 0.1389589823782444,
      "num_tokens": 2502094.0,
      "step": 1215
    },
    {
      "entropy": 6.706840229034424,
      "epoch": 1.0825565912117177,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004994943115561495,
      "loss": 6.3975,
      "mean_token_accuracy": 0.13392824158072472,
      "num_tokens": 2512553.0,
      "step": 1220
    },
    {
      "entropy": 6.795430517196655,
      "epoch": 1.0869951176209498,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004994709663223143,
      "loss": 6.3947,
      "mean_token_accuracy": 0.13575630336999894,
      "num_tokens": 2522350.0,
      "step": 1225
    },
    {
      "entropy": 6.717914056777954,
      "epoch": 1.091433644030182,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004994470949438712,
      "loss": 6.4571,
      "mean_token_accuracy": 0.12943646237254142,
      "num_tokens": 2532884.0,
      "step": 1230
    },
    {
      "entropy": 6.821149110794067,
      "epoch": 1.095872170439414,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994226974767734,
      "loss": 6.4947,
      "mean_token_accuracy": 0.12449371442198753,
      "num_tokens": 2543364.0,
      "step": 1235
    },
    {
      "entropy": 6.7451752662658695,
      "epoch": 1.1003106968486462,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004993977739782066,
      "loss": 6.3817,
      "mean_token_accuracy": 0.13001500964164733,
      "num_tokens": 2552281.0,
      "step": 1240
    },
    {
      "entropy": 6.721602582931519,
      "epoch": 1.1047492232578784,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00049937232450659,
      "loss": 6.446,
      "mean_token_accuracy": 0.12717969268560408,
      "num_tokens": 2561963.0,
      "step": 1245
    },
    {
      "entropy": 6.7753888130187985,
      "epoch": 1.1091877496671105,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004993463491215753,
      "loss": 6.4337,
      "mean_token_accuracy": 0.13999854624271393,
      "num_tokens": 2572951.0,
      "step": 1250
    },
    {
      "entropy": 6.816381406784058,
      "epoch": 1.1136262760763427,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499319847884047,
      "loss": 6.4535,
      "mean_token_accuracy": 0.1278873711824417,
      "num_tokens": 2583019.0,
      "step": 1255
    },
    {
      "entropy": 6.7529043674469,
      "epoch": 1.1180648024855748,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004992928208561224,
      "loss": 6.3902,
      "mean_token_accuracy": 0.13753967881202697,
      "num_tokens": 2593385.0,
      "step": 1260
    },
    {
      "entropy": 6.774140882492065,
      "epoch": 1.122503328894807,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004992652681011508,
      "loss": 6.3959,
      "mean_token_accuracy": 0.13451069965958595,
      "num_tokens": 2603679.0,
      "step": 1265
    },
    {
      "entropy": 6.745535135269165,
      "epoch": 1.1269418553040391,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000499237189683714,
      "loss": 6.458,
      "mean_token_accuracy": 0.1327129691839218,
      "num_tokens": 2614029.0,
      "step": 1270
    },
    {
      "entropy": 6.658591461181641,
      "epoch": 1.1313803817132713,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004992085856696259,
      "loss": 6.4209,
      "mean_token_accuracy": 0.1382272757589817,
      "num_tokens": 2624063.0,
      "step": 1275
    },
    {
      "entropy": 6.722748279571533,
      "epoch": 1.1358189081225034,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004991794561259325,
      "loss": 6.3691,
      "mean_token_accuracy": 0.14201630800962448,
      "num_tokens": 2633792.0,
      "step": 1280
    },
    {
      "entropy": 6.6542223453521725,
      "epoch": 1.1402574345317356,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004991498011209112,
      "loss": 6.3232,
      "mean_token_accuracy": 0.13645180314779282,
      "num_tokens": 2643401.0,
      "step": 1285
    },
    {
      "entropy": 6.773741817474365,
      "epoch": 1.1446959609409677,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004991196207240714,
      "loss": 6.4539,
      "mean_token_accuracy": 0.1282748505473137,
      "num_tokens": 2653909.0,
      "step": 1290
    },
    {
      "entropy": 6.693137884140015,
      "epoch": 1.1491344873501999,
      "grad_norm": 1.25,
      "learning_rate": 0.0004990889150061541,
      "loss": 6.4317,
      "mean_token_accuracy": 0.1349467732012272,
      "num_tokens": 2664386.0,
      "step": 1295
    },
    {
      "entropy": 6.766120195388794,
      "epoch": 1.1535730137594318,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004990576840391312,
      "loss": 6.4865,
      "mean_token_accuracy": 0.12709747180342673,
      "num_tokens": 2675398.0,
      "step": 1300
    },
    {
      "entropy": 6.749423599243164,
      "epoch": 1.158011540168664,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000499025927896206,
      "loss": 6.389,
      "mean_token_accuracy": 0.1319727636873722,
      "num_tokens": 2684895.0,
      "step": 1305
    },
    {
      "entropy": 6.622655010223388,
      "epoch": 1.162450066577896,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004989936466518127,
      "loss": 6.367,
      "mean_token_accuracy": 0.1397032156586647,
      "num_tokens": 2694706.0,
      "step": 1310
    },
    {
      "entropy": 6.733213758468628,
      "epoch": 1.1668885929871282,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004989608403816164,
      "loss": 6.3576,
      "mean_token_accuracy": 0.13502043187618257,
      "num_tokens": 2706486.0,
      "step": 1315
    },
    {
      "entropy": 6.691570997238159,
      "epoch": 1.1713271193963604,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004989275091625126,
      "loss": 6.4414,
      "mean_token_accuracy": 0.14210355877876282,
      "num_tokens": 2716613.0,
      "step": 1320
    },
    {
      "entropy": 6.886738586425781,
      "epoch": 1.1757656458055925,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004988936530726276,
      "loss": 6.5302,
      "mean_token_accuracy": 0.1268772892653942,
      "num_tokens": 2727313.0,
      "step": 1325
    },
    {
      "entropy": 6.723033380508423,
      "epoch": 1.1802041722148247,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004988592721913176,
      "loss": 6.3995,
      "mean_token_accuracy": 0.1361616112291813,
      "num_tokens": 2737826.0,
      "step": 1330
    },
    {
      "entropy": 6.63746075630188,
      "epoch": 1.1846426986240568,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004988243665991692,
      "loss": 6.3232,
      "mean_token_accuracy": 0.13523943722248077,
      "num_tokens": 2748962.0,
      "step": 1335
    },
    {
      "entropy": 6.7660548210144045,
      "epoch": 1.189081225033289,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004987889363779985,
      "loss": 6.3726,
      "mean_token_accuracy": 0.13909099400043487,
      "num_tokens": 2760116.0,
      "step": 1340
    },
    {
      "entropy": 6.642192268371582,
      "epoch": 1.193519751442521,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987529816108517,
      "loss": 6.3868,
      "mean_token_accuracy": 0.13063296377658845,
      "num_tokens": 2771664.0,
      "step": 1345
    },
    {
      "entropy": 6.735338401794434,
      "epoch": 1.1979582778517532,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004987165023820043,
      "loss": 6.3339,
      "mean_token_accuracy": 0.13763192743062974,
      "num_tokens": 2781792.0,
      "step": 1350
    },
    {
      "entropy": 6.5532543659210205,
      "epoch": 1.2023968042609854,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986794987769611,
      "loss": 6.3272,
      "mean_token_accuracy": 0.1380511462688446,
      "num_tokens": 2792203.0,
      "step": 1355
    },
    {
      "entropy": 6.720197010040283,
      "epoch": 1.2068353306702175,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000498641970882456,
      "loss": 6.4867,
      "mean_token_accuracy": 0.13111176937818528,
      "num_tokens": 2802077.0,
      "step": 1360
    },
    {
      "entropy": 6.638952589035034,
      "epoch": 1.2112738570794497,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004986039187864518,
      "loss": 6.3655,
      "mean_token_accuracy": 0.14026057049632074,
      "num_tokens": 2811993.0,
      "step": 1365
    },
    {
      "entropy": 6.676184177398682,
      "epoch": 1.2157123834886818,
      "grad_norm": 1.125,
      "learning_rate": 0.0004985653425781401,
      "loss": 6.402,
      "mean_token_accuracy": 0.13089463859796524,
      "num_tokens": 2821924.0,
      "step": 1370
    },
    {
      "entropy": 6.675514078140258,
      "epoch": 1.220150909897914,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004985262423479408,
      "loss": 6.3627,
      "mean_token_accuracy": 0.1411583587527275,
      "num_tokens": 2831521.0,
      "step": 1375
    },
    {
      "entropy": 6.686585426330566,
      "epoch": 1.2245894363071461,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004984866181875021,
      "loss": 6.3526,
      "mean_token_accuracy": 0.13705047890543937,
      "num_tokens": 2841864.0,
      "step": 1380
    },
    {
      "entropy": 6.604874324798584,
      "epoch": 1.229027962716378,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004984464701897005,
      "loss": 6.3588,
      "mean_token_accuracy": 0.13925145491957663,
      "num_tokens": 2852199.0,
      "step": 1385
    },
    {
      "entropy": 6.6554466724395756,
      "epoch": 1.2334664891256102,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004984057984486402,
      "loss": 6.3641,
      "mean_token_accuracy": 0.1341533102095127,
      "num_tokens": 2862167.0,
      "step": 1390
    },
    {
      "entropy": 6.686126089096069,
      "epoch": 1.2379050155348423,
      "grad_norm": 1.125,
      "learning_rate": 0.0004983646030596527,
      "loss": 6.4079,
      "mean_token_accuracy": 0.13984917402267455,
      "num_tokens": 2872604.0,
      "step": 1395
    },
    {
      "entropy": 6.672490835189819,
      "epoch": 1.2423435419440745,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004983228841192975,
      "loss": 6.3752,
      "mean_token_accuracy": 0.13595345988869667,
      "num_tokens": 2882401.0,
      "step": 1400
    },
    {
      "entropy": 6.631757831573486,
      "epoch": 1.2467820683533066,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004982806417253607,
      "loss": 6.3511,
      "mean_token_accuracy": 0.13876891285181045,
      "num_tokens": 2892004.0,
      "step": 1405
    },
    {
      "entropy": 6.63784236907959,
      "epoch": 1.2512205947625388,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004982378759768557,
      "loss": 6.3538,
      "mean_token_accuracy": 0.13770614489912986,
      "num_tokens": 2902369.0,
      "step": 1410
    },
    {
      "entropy": 6.656727027893067,
      "epoch": 1.255659121171771,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004981945869740225,
      "loss": 6.3934,
      "mean_token_accuracy": 0.13234489262104035,
      "num_tokens": 2912532.0,
      "step": 1415
    },
    {
      "entropy": 6.692376708984375,
      "epoch": 1.260097647581003,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004981507748183276,
      "loss": 6.3853,
      "mean_token_accuracy": 0.1314159318804741,
      "num_tokens": 2922621.0,
      "step": 1420
    },
    {
      "entropy": 6.710336446762085,
      "epoch": 1.2645361739902352,
      "grad_norm": 1.125,
      "learning_rate": 0.0004981064396124635,
      "loss": 6.3583,
      "mean_token_accuracy": 0.13584646657109262,
      "num_tokens": 2932510.0,
      "step": 1425
    },
    {
      "entropy": 6.6139873504638675,
      "epoch": 1.2689747003994674,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004980615814603492,
      "loss": 6.3846,
      "mean_token_accuracy": 0.12972714975476266,
      "num_tokens": 2942982.0,
      "step": 1430
    },
    {
      "entropy": 6.653766632080078,
      "epoch": 1.2734132268086995,
      "grad_norm": 1.25,
      "learning_rate": 0.0004980162004671288,
      "loss": 6.2817,
      "mean_token_accuracy": 0.1397496670484543,
      "num_tokens": 2953417.0,
      "step": 1435
    },
    {
      "entropy": 6.590647983551025,
      "epoch": 1.2778517532179317,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004979702967391725,
      "loss": 6.287,
      "mean_token_accuracy": 0.14722541570663453,
      "num_tokens": 2961771.0,
      "step": 1440
    },
    {
      "entropy": 6.587854480743408,
      "epoch": 1.2822902796271638,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004979238703840755,
      "loss": 6.3671,
      "mean_token_accuracy": 0.13086313456296922,
      "num_tokens": 2971980.0,
      "step": 1445
    },
    {
      "entropy": 6.69378170967102,
      "epoch": 1.286728806036396,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004978769215106579,
      "loss": 6.3792,
      "mean_token_accuracy": 0.13028891906142234,
      "num_tokens": 2982087.0,
      "step": 1450
    },
    {
      "entropy": 6.681559753417969,
      "epoch": 1.291167332445628,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004978294502289646,
      "loss": 6.4874,
      "mean_token_accuracy": 0.12749797403812407,
      "num_tokens": 2993424.0,
      "step": 1455
    },
    {
      "entropy": 6.710448789596557,
      "epoch": 1.2956058588548602,
      "grad_norm": 1.125,
      "learning_rate": 0.0004977814566502651,
      "loss": 6.3213,
      "mean_token_accuracy": 0.1379827730357647,
      "num_tokens": 3003081.0,
      "step": 1460
    },
    {
      "entropy": 6.677050495147705,
      "epoch": 1.3000443852640924,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004977329408870532,
      "loss": 6.4048,
      "mean_token_accuracy": 0.1378847599029541,
      "num_tokens": 3013829.0,
      "step": 1465
    },
    {
      "entropy": 6.576231288909912,
      "epoch": 1.3044829116733245,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004976839030530464,
      "loss": 6.3102,
      "mean_token_accuracy": 0.1419086903333664,
      "num_tokens": 3024511.0,
      "step": 1470
    },
    {
      "entropy": 6.593928623199463,
      "epoch": 1.3089214380825567,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000497634343263186,
      "loss": 6.4043,
      "mean_token_accuracy": 0.12826052978634833,
      "num_tokens": 3034503.0,
      "step": 1475
    },
    {
      "entropy": 6.648963260650635,
      "epoch": 1.3133599644917888,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004975842616336369,
      "loss": 6.3201,
      "mean_token_accuracy": 0.141230496019125,
      "num_tokens": 3044823.0,
      "step": 1480
    },
    {
      "entropy": 6.510059928894043,
      "epoch": 1.317798490901021,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004975336582817869,
      "loss": 6.33,
      "mean_token_accuracy": 0.1362045116722584,
      "num_tokens": 3056412.0,
      "step": 1485
    },
    {
      "entropy": 6.728832674026489,
      "epoch": 1.3222370173102531,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004974825333262469,
      "loss": 6.3945,
      "mean_token_accuracy": 0.12927686050534248,
      "num_tokens": 3067101.0,
      "step": 1490
    },
    {
      "entropy": 6.7084520816802975,
      "epoch": 1.3266755437194853,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004974308868868501,
      "loss": 6.3995,
      "mean_token_accuracy": 0.12630158364772798,
      "num_tokens": 3077638.0,
      "step": 1495
    },
    {
      "entropy": 6.551306915283203,
      "epoch": 1.3311140701287172,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004973787190846524,
      "loss": 6.3218,
      "mean_token_accuracy": 0.14200132787227632,
      "num_tokens": 3087970.0,
      "step": 1500
    },
    {
      "epoch": 1.3311140701287172,
      "eval_entropy": 6.368161134651891,
      "eval_loss": 6.45974588394165,
      "eval_mean_token_accuracy": 0.13263512823969623,
      "eval_num_tokens": 3087970.0,
      "eval_runtime": 2.0891,
      "eval_samples_per_second": 1611.728,
      "eval_steps_per_second": 201.526,
      "step": 1500
    },
    {
      "entropy": 6.630201387405395,
      "epoch": 1.3355525965379493,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004973260300419316,
      "loss": 6.3578,
      "mean_token_accuracy": 0.13766367435455323,
      "num_tokens": 3097895.0,
      "step": 1505
    },
    {
      "entropy": 6.590450048446655,
      "epoch": 1.3399911229471815,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004972728198821871,
      "loss": 6.3072,
      "mean_token_accuracy": 0.1387806810438633,
      "num_tokens": 3107042.0,
      "step": 1510
    },
    {
      "entropy": 6.625532007217407,
      "epoch": 1.3444296493564136,
      "grad_norm": 1.046875,
      "learning_rate": 0.00049721908873014,
      "loss": 6.315,
      "mean_token_accuracy": 0.14337668791413308,
      "num_tokens": 3117959.0,
      "step": 1515
    },
    {
      "entropy": 6.558200359344482,
      "epoch": 1.3488681757656458,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004971648367117323,
      "loss": 6.3543,
      "mean_token_accuracy": 0.13464199230074883,
      "num_tokens": 3128407.0,
      "step": 1520
    },
    {
      "entropy": 6.676009941101074,
      "epoch": 1.353306702174878,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004971100639541271,
      "loss": 6.3267,
      "mean_token_accuracy": 0.14146978706121444,
      "num_tokens": 3138265.0,
      "step": 1525
    },
    {
      "entropy": 6.57694182395935,
      "epoch": 1.35774522858411,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000497054770585708,
      "loss": 6.3353,
      "mean_token_accuracy": 0.13715606033802033,
      "num_tokens": 3148677.0,
      "step": 1530
    },
    {
      "entropy": 6.556269979476928,
      "epoch": 1.3621837549933422,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004969989567360788,
      "loss": 6.3503,
      "mean_token_accuracy": 0.13542981594800949,
      "num_tokens": 3158684.0,
      "step": 1535
    },
    {
      "entropy": 6.643629789352417,
      "epoch": 1.3666222814025744,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004969426225360635,
      "loss": 6.3543,
      "mean_token_accuracy": 0.1378885067999363,
      "num_tokens": 3169124.0,
      "step": 1540
    },
    {
      "entropy": 6.596967315673828,
      "epoch": 1.3710608078118065,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004968857681177056,
      "loss": 6.2667,
      "mean_token_accuracy": 0.1364466980099678,
      "num_tokens": 3179282.0,
      "step": 1545
    },
    {
      "entropy": 6.605694532394409,
      "epoch": 1.3754993342210386,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004968283936142679,
      "loss": 6.301,
      "mean_token_accuracy": 0.13783995881676675,
      "num_tokens": 3190144.0,
      "step": 1550
    },
    {
      "entropy": 6.484689426422119,
      "epoch": 1.3799378606302708,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004967704991602322,
      "loss": 6.3025,
      "mean_token_accuracy": 0.14239512234926224,
      "num_tokens": 3201277.0,
      "step": 1555
    },
    {
      "entropy": 6.584119367599487,
      "epoch": 1.384376387039503,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004967120848912995,
      "loss": 6.318,
      "mean_token_accuracy": 0.13924861773848535,
      "num_tokens": 3211691.0,
      "step": 1560
    },
    {
      "entropy": 6.51182188987732,
      "epoch": 1.388814913448735,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004966531509443884,
      "loss": 6.2972,
      "mean_token_accuracy": 0.1423930622637272,
      "num_tokens": 3220452.0,
      "step": 1565
    },
    {
      "entropy": 6.665063858032227,
      "epoch": 1.3932534398579672,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004965936974576363,
      "loss": 6.2572,
      "mean_token_accuracy": 0.1434251219034195,
      "num_tokens": 3230821.0,
      "step": 1570
    },
    {
      "entropy": 6.569358205795288,
      "epoch": 1.3976919662671992,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004965337245703981,
      "loss": 6.333,
      "mean_token_accuracy": 0.13867368102073668,
      "num_tokens": 3240621.0,
      "step": 1575
    },
    {
      "entropy": 6.564372253417969,
      "epoch": 1.4021304926764313,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004964732324232462,
      "loss": 6.302,
      "mean_token_accuracy": 0.13816518783569337,
      "num_tokens": 3251442.0,
      "step": 1580
    },
    {
      "entropy": 6.6255796432495115,
      "epoch": 1.4065690190856635,
      "grad_norm": 1.140625,
      "learning_rate": 0.00049641222115797,
      "loss": 6.3077,
      "mean_token_accuracy": 0.13412498235702514,
      "num_tokens": 3261164.0,
      "step": 1585
    },
    {
      "entropy": 6.539021825790405,
      "epoch": 1.4110075454948956,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004963506909175758,
      "loss": 6.3365,
      "mean_token_accuracy": 0.13258393555879594,
      "num_tokens": 3271442.0,
      "step": 1590
    },
    {
      "entropy": 6.633949136734008,
      "epoch": 1.4154460719041277,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000496288641846286,
      "loss": 6.3385,
      "mean_token_accuracy": 0.13758450075984002,
      "num_tokens": 3282159.0,
      "step": 1595
    },
    {
      "entropy": 6.485613584518433,
      "epoch": 1.41988459831336,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004962260740895398,
      "loss": 6.2791,
      "mean_token_accuracy": 0.14571996703743934,
      "num_tokens": 3291424.0,
      "step": 1600
    },
    {
      "entropy": 6.557862329483032,
      "epoch": 1.424323124722592,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004961629877939913,
      "loss": 6.3422,
      "mean_token_accuracy": 0.1376090705394745,
      "num_tokens": 3302098.0,
      "step": 1605
    },
    {
      "entropy": 6.554436111450196,
      "epoch": 1.4287616511318242,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000496099383107511,
      "loss": 6.3472,
      "mean_token_accuracy": 0.13372454419732094,
      "num_tokens": 3313292.0,
      "step": 1610
    },
    {
      "entropy": 6.600845575332642,
      "epoch": 1.4332001775410563,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004960352601791835,
      "loss": 6.2123,
      "mean_token_accuracy": 0.14182500839233397,
      "num_tokens": 3324272.0,
      "step": 1615
    },
    {
      "entropy": 6.54074878692627,
      "epoch": 1.4376387039502885,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004959706191593087,
      "loss": 6.3102,
      "mean_token_accuracy": 0.1401732988655567,
      "num_tokens": 3335299.0,
      "step": 1620
    },
    {
      "entropy": 6.616578388214111,
      "epoch": 1.4420772303595206,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004959054601994007,
      "loss": 6.2826,
      "mean_token_accuracy": 0.1450162336230278,
      "num_tokens": 3344652.0,
      "step": 1625
    },
    {
      "entropy": 6.52901258468628,
      "epoch": 1.4465157567687528,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004958397834521878,
      "loss": 6.3048,
      "mean_token_accuracy": 0.139273801445961,
      "num_tokens": 3355166.0,
      "step": 1630
    },
    {
      "entropy": 6.562482261657715,
      "epoch": 1.450954283177985,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004957735890716115,
      "loss": 6.2622,
      "mean_token_accuracy": 0.1440594784915447,
      "num_tokens": 3366330.0,
      "step": 1635
    },
    {
      "entropy": 6.549576234817505,
      "epoch": 1.455392809587217,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004957068772128273,
      "loss": 6.3315,
      "mean_token_accuracy": 0.14034282714128493,
      "num_tokens": 3375930.0,
      "step": 1640
    },
    {
      "entropy": 6.5108832836151125,
      "epoch": 1.4598313359964492,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000495639648032203,
      "loss": 6.1684,
      "mean_token_accuracy": 0.16133927404880524,
      "num_tokens": 3387310.0,
      "step": 1645
    },
    {
      "entropy": 6.47861647605896,
      "epoch": 1.4642698624056814,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004955719016873192,
      "loss": 6.2004,
      "mean_token_accuracy": 0.1491427481174469,
      "num_tokens": 3398022.0,
      "step": 1650
    },
    {
      "entropy": 6.540802431106568,
      "epoch": 1.4687083888149135,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004955036383369688,
      "loss": 6.216,
      "mean_token_accuracy": 0.14632173478603364,
      "num_tokens": 3407315.0,
      "step": 1655
    },
    {
      "entropy": 6.459036493301392,
      "epoch": 1.4731469152241456,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004954348581411564,
      "loss": 6.3148,
      "mean_token_accuracy": 0.14125865548849106,
      "num_tokens": 3418009.0,
      "step": 1660
    },
    {
      "entropy": 6.666120433807373,
      "epoch": 1.4775854416333778,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000495365561261098,
      "loss": 6.3262,
      "mean_token_accuracy": 0.13915122225880622,
      "num_tokens": 3428659.0,
      "step": 1665
    },
    {
      "entropy": 6.513209629058838,
      "epoch": 1.48202396804261,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004952957478592209,
      "loss": 6.2605,
      "mean_token_accuracy": 0.14562821611762047,
      "num_tokens": 3439102.0,
      "step": 1670
    },
    {
      "entropy": 6.495951080322266,
      "epoch": 1.486462494451842,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004952254180991628,
      "loss": 6.3143,
      "mean_token_accuracy": 0.14086345955729485,
      "num_tokens": 3450269.0,
      "step": 1675
    },
    {
      "entropy": 6.550183248519898,
      "epoch": 1.4909010208610742,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004951545721457719,
      "loss": 6.2323,
      "mean_token_accuracy": 0.15024334490299224,
      "num_tokens": 3459876.0,
      "step": 1680
    },
    {
      "entropy": 6.439149618148804,
      "epoch": 1.4953395472703064,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004950832101651062,
      "loss": 6.2879,
      "mean_token_accuracy": 0.1395568035542965,
      "num_tokens": 3469276.0,
      "step": 1685
    },
    {
      "entropy": 6.540930128097534,
      "epoch": 1.4997780736795385,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004950113323244336,
      "loss": 6.1873,
      "mean_token_accuracy": 0.14649046510457991,
      "num_tokens": 3479835.0,
      "step": 1690
    },
    {
      "entropy": 6.476472091674805,
      "epoch": 1.5042166000887707,
      "grad_norm": 1.0,
      "learning_rate": 0.0004949389387922305,
      "loss": 6.2472,
      "mean_token_accuracy": 0.14312107488512993,
      "num_tokens": 3491176.0,
      "step": 1695
    },
    {
      "entropy": 6.531066989898681,
      "epoch": 1.5086551264980028,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004948660297381826,
      "loss": 6.2262,
      "mean_token_accuracy": 0.14610032215714455,
      "num_tokens": 3501122.0,
      "step": 1700
    },
    {
      "entropy": 6.392983913421631,
      "epoch": 1.5130936529072347,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004947926053331836,
      "loss": 6.1558,
      "mean_token_accuracy": 0.1542304791510105,
      "num_tokens": 3510875.0,
      "step": 1705
    },
    {
      "entropy": 6.541310834884643,
      "epoch": 1.517532179316467,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004947186657493354,
      "loss": 6.2377,
      "mean_token_accuracy": 0.1494641751050949,
      "num_tokens": 3521798.0,
      "step": 1710
    },
    {
      "entropy": 6.438729286193848,
      "epoch": 1.521970705725699,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004946442111599473,
      "loss": 6.2043,
      "mean_token_accuracy": 0.1451006591320038,
      "num_tokens": 3531456.0,
      "step": 1715
    },
    {
      "entropy": 6.580926847457886,
      "epoch": 1.5264092321349312,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004945692417395358,
      "loss": 6.251,
      "mean_token_accuracy": 0.14176156222820283,
      "num_tokens": 3541315.0,
      "step": 1720
    },
    {
      "entropy": 6.448061084747314,
      "epoch": 1.5308477585441633,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000494493757663824,
      "loss": 6.2238,
      "mean_token_accuracy": 0.14351213127374648,
      "num_tokens": 3552876.0,
      "step": 1725
    },
    {
      "entropy": 6.4509584426879885,
      "epoch": 1.5352862849533955,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004944177591097415,
      "loss": 6.2331,
      "mean_token_accuracy": 0.14354335144162178,
      "num_tokens": 3563967.0,
      "step": 1730
    },
    {
      "entropy": 6.457042932510376,
      "epoch": 1.5397248113626276,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004943412462554236,
      "loss": 6.2159,
      "mean_token_accuracy": 0.1507117457687855,
      "num_tokens": 3573510.0,
      "step": 1735
    },
    {
      "entropy": 6.546308755874634,
      "epoch": 1.5441633377718598,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004942642192802114,
      "loss": 6.3099,
      "mean_token_accuracy": 0.13970684260129929,
      "num_tokens": 3584863.0,
      "step": 1740
    },
    {
      "entropy": 6.455177211761475,
      "epoch": 1.548601864181092,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004941866783646508,
      "loss": 6.2268,
      "mean_token_accuracy": 0.14618334472179412,
      "num_tokens": 3594489.0,
      "step": 1745
    },
    {
      "entropy": 6.556396770477295,
      "epoch": 1.553040390590324,
      "grad_norm": 1.0,
      "learning_rate": 0.0004941086236904923,
      "loss": 6.2043,
      "mean_token_accuracy": 0.14727460145950316,
      "num_tokens": 3605159.0,
      "step": 1750
    },
    {
      "entropy": 6.481395721435547,
      "epoch": 1.557478916999556,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004940300554406909,
      "loss": 6.3012,
      "mean_token_accuracy": 0.14313508197665215,
      "num_tokens": 3615254.0,
      "step": 1755
    },
    {
      "entropy": 6.445766162872315,
      "epoch": 1.5619174434087881,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000493950973799405,
      "loss": 6.1703,
      "mean_token_accuracy": 0.15018191039562226,
      "num_tokens": 3625015.0,
      "step": 1760
    },
    {
      "entropy": 6.42508454322815,
      "epoch": 1.5663559698180203,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004938713789519967,
      "loss": 6.2497,
      "mean_token_accuracy": 0.1505135826766491,
      "num_tokens": 3636083.0,
      "step": 1765
    },
    {
      "entropy": 6.512313747406006,
      "epoch": 1.5707944962272524,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004937912710850309,
      "loss": 6.166,
      "mean_token_accuracy": 0.1465531922876835,
      "num_tokens": 3646363.0,
      "step": 1770
    },
    {
      "entropy": 6.370706844329834,
      "epoch": 1.5752330226364846,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004937106503862749,
      "loss": 6.1685,
      "mean_token_accuracy": 0.1537150263786316,
      "num_tokens": 3656868.0,
      "step": 1775
    },
    {
      "entropy": 6.539106035232544,
      "epoch": 1.5796715490457167,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004936295170446981,
      "loss": 6.3492,
      "mean_token_accuracy": 0.1347063384950161,
      "num_tokens": 3668105.0,
      "step": 1780
    },
    {
      "entropy": 6.607371664047241,
      "epoch": 1.5841100754549489,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004935478712504715,
      "loss": 6.2487,
      "mean_token_accuracy": 0.1385604441165924,
      "num_tokens": 3678841.0,
      "step": 1785
    },
    {
      "entropy": 6.4086525440216064,
      "epoch": 1.588548601864181,
      "grad_norm": 1.125,
      "learning_rate": 0.0004934657131949674,
      "loss": 6.2415,
      "mean_token_accuracy": 0.14987936168909072,
      "num_tokens": 3688603.0,
      "step": 1790
    },
    {
      "entropy": 6.529016494750977,
      "epoch": 1.5929871282734132,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004933830430707585,
      "loss": 6.2542,
      "mean_token_accuracy": 0.13600879460573195,
      "num_tokens": 3698312.0,
      "step": 1795
    },
    {
      "entropy": 6.4327093124389645,
      "epoch": 1.5974256546826453,
      "grad_norm": 1.046875,
      "learning_rate": 0.000493299861071618,
      "loss": 6.1204,
      "mean_token_accuracy": 0.15229299366474153,
      "num_tokens": 3708252.0,
      "step": 1800
    },
    {
      "entropy": 6.483391284942627,
      "epoch": 1.6018641810918774,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004932161673925189,
      "loss": 6.1691,
      "mean_token_accuracy": 0.1462477594614029,
      "num_tokens": 3718431.0,
      "step": 1805
    },
    {
      "entropy": 6.433411121368408,
      "epoch": 1.6063027075011096,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004931319622296333,
      "loss": 6.0884,
      "mean_token_accuracy": 0.15784337967634202,
      "num_tokens": 3729170.0,
      "step": 1810
    },
    {
      "entropy": 6.474455261230469,
      "epoch": 1.6107412339103417,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004930472457803324,
      "loss": 6.1861,
      "mean_token_accuracy": 0.1482206791639328,
      "num_tokens": 3738885.0,
      "step": 1815
    },
    {
      "entropy": 6.340183973312378,
      "epoch": 1.6151797603195739,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004929620182431858,
      "loss": 6.1744,
      "mean_token_accuracy": 0.15203531160950662,
      "num_tokens": 3748535.0,
      "step": 1820
    },
    {
      "entropy": 6.444498491287232,
      "epoch": 1.619618286728806,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004928762798179612,
      "loss": 6.237,
      "mean_token_accuracy": 0.14974780827760698,
      "num_tokens": 3758827.0,
      "step": 1825
    },
    {
      "entropy": 6.455253171920776,
      "epoch": 1.6240568131380382,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004927900307056233,
      "loss": 6.1615,
      "mean_token_accuracy": 0.1510964810848236,
      "num_tokens": 3768632.0,
      "step": 1830
    },
    {
      "entropy": 6.441723012924195,
      "epoch": 1.6284953395472703,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004927032711083342,
      "loss": 6.214,
      "mean_token_accuracy": 0.14101684093475342,
      "num_tokens": 3779324.0,
      "step": 1835
    },
    {
      "entropy": 6.381724882125854,
      "epoch": 1.6329338659565025,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004926160012294526,
      "loss": 6.1586,
      "mean_token_accuracy": 0.15024627447128297,
      "num_tokens": 3789744.0,
      "step": 1840
    },
    {
      "entropy": 6.370701789855957,
      "epoch": 1.6373723923657346,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000492528221273533,
      "loss": 6.1104,
      "mean_token_accuracy": 0.1493580684065819,
      "num_tokens": 3799831.0,
      "step": 1845
    },
    {
      "entropy": 6.542438220977783,
      "epoch": 1.6418109187749668,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004924399314463258,
      "loss": 6.2365,
      "mean_token_accuracy": 0.14592818170785904,
      "num_tokens": 3812102.0,
      "step": 1850
    },
    {
      "entropy": 6.4679584980010985,
      "epoch": 1.646249445184199,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004923511319547761,
      "loss": 6.2914,
      "mean_token_accuracy": 0.13645976781845093,
      "num_tokens": 3822750.0,
      "step": 1855
    },
    {
      "entropy": 6.439510631561279,
      "epoch": 1.650687971593431,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000492261823007024,
      "loss": 6.2064,
      "mean_token_accuracy": 0.1489210918545723,
      "num_tokens": 3832354.0,
      "step": 1860
    },
    {
      "entropy": 6.458003902435303,
      "epoch": 1.6551264980026632,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004921720048124034,
      "loss": 6.2224,
      "mean_token_accuracy": 0.1468009516596794,
      "num_tokens": 3842930.0,
      "step": 1865
    },
    {
      "entropy": 6.5039207458496096,
      "epoch": 1.6595650244118954,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004920816775814422,
      "loss": 6.2599,
      "mean_token_accuracy": 0.14020814523100852,
      "num_tokens": 3852952.0,
      "step": 1870
    },
    {
      "entropy": 6.454715347290039,
      "epoch": 1.6640035508211275,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004919908415258612,
      "loss": 6.2067,
      "mean_token_accuracy": 0.14455220550298692,
      "num_tokens": 3863070.0,
      "step": 1875
    },
    {
      "entropy": 6.4132616996765135,
      "epoch": 1.6684420772303596,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000491899496858574,
      "loss": 6.1501,
      "mean_token_accuracy": 0.15156037211418152,
      "num_tokens": 3872767.0,
      "step": 1880
    },
    {
      "entropy": 6.3353992938995365,
      "epoch": 1.6728806036395918,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004918076437936859,
      "loss": 6.1056,
      "mean_token_accuracy": 0.15438561514019966,
      "num_tokens": 3883183.0,
      "step": 1885
    },
    {
      "entropy": 6.433599185943604,
      "epoch": 1.677319130048824,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004917152825464947,
      "loss": 6.1996,
      "mean_token_accuracy": 0.1471118450164795,
      "num_tokens": 3893304.0,
      "step": 1890
    },
    {
      "entropy": 6.44781608581543,
      "epoch": 1.681757656458056,
      "grad_norm": 1.0,
      "learning_rate": 0.0004916224133334885,
      "loss": 6.135,
      "mean_token_accuracy": 0.1504994697868824,
      "num_tokens": 3903786.0,
      "step": 1895
    },
    {
      "entropy": 6.413273906707763,
      "epoch": 1.686196182867288,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004915290363723465,
      "loss": 6.1426,
      "mean_token_accuracy": 0.14888948425650597,
      "num_tokens": 3913703.0,
      "step": 1900
    },
    {
      "entropy": 6.3861230373382565,
      "epoch": 1.6906347092765202,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004914351518819379,
      "loss": 6.197,
      "mean_token_accuracy": 0.14716671407222748,
      "num_tokens": 3923653.0,
      "step": 1905
    },
    {
      "entropy": 6.3618451118469235,
      "epoch": 1.6950732356857523,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004913407600823216,
      "loss": 6.0837,
      "mean_token_accuracy": 0.15539871603250505,
      "num_tokens": 3933915.0,
      "step": 1910
    },
    {
      "entropy": 6.429026556015015,
      "epoch": 1.6995117620949844,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004912458611947454,
      "loss": 6.1613,
      "mean_token_accuracy": 0.14607803374528885,
      "num_tokens": 3943334.0,
      "step": 1915
    },
    {
      "entropy": 6.512318658828735,
      "epoch": 1.7039502885042166,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000491150455441646,
      "loss": 6.2164,
      "mean_token_accuracy": 0.13555126562714576,
      "num_tokens": 3952608.0,
      "step": 1920
    },
    {
      "entropy": 6.345305156707764,
      "epoch": 1.7083888149134487,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004910545430466478,
      "loss": 6.123,
      "mean_token_accuracy": 0.1472136050462723,
      "num_tokens": 3963398.0,
      "step": 1925
    },
    {
      "entropy": 6.38430609703064,
      "epoch": 1.7128273413226809,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004909581242345628,
      "loss": 6.1552,
      "mean_token_accuracy": 0.14694450199604034,
      "num_tokens": 3973878.0,
      "step": 1930
    },
    {
      "entropy": 6.430195760726929,
      "epoch": 1.717265867731913,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004908611992313905,
      "loss": 6.1648,
      "mean_token_accuracy": 0.13664614632725716,
      "num_tokens": 3983681.0,
      "step": 1935
    },
    {
      "entropy": 6.39329776763916,
      "epoch": 1.7217043941411452,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004907637682643162,
      "loss": 6.201,
      "mean_token_accuracy": 0.14567800536751746,
      "num_tokens": 3995026.0,
      "step": 1940
    },
    {
      "entropy": 6.430534839630127,
      "epoch": 1.7261429205503773,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004906658315617112,
      "loss": 6.2022,
      "mean_token_accuracy": 0.14595392793416978,
      "num_tokens": 4005089.0,
      "step": 1945
    },
    {
      "entropy": 6.359121513366699,
      "epoch": 1.7305814469596092,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004905673893531331,
      "loss": 6.1076,
      "mean_token_accuracy": 0.15097568482160567,
      "num_tokens": 4014795.0,
      "step": 1950
    },
    {
      "entropy": 6.326131105422974,
      "epoch": 1.7350199733688414,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004904684418693231,
      "loss": 6.1828,
      "mean_token_accuracy": 0.14517654925584794,
      "num_tokens": 4025563.0,
      "step": 1955
    },
    {
      "entropy": 6.499860239028931,
      "epoch": 1.7394584997780735,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004903689893422077,
      "loss": 6.116,
      "mean_token_accuracy": 0.1486818477511406,
      "num_tokens": 4035050.0,
      "step": 1960
    },
    {
      "entropy": 6.305110931396484,
      "epoch": 1.7438970261873057,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004902690320048966,
      "loss": 6.1345,
      "mean_token_accuracy": 0.14928318560123444,
      "num_tokens": 4046109.0,
      "step": 1965
    },
    {
      "entropy": 6.449776601791382,
      "epoch": 1.7483355525965378,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000490168570091683,
      "loss": 6.1848,
      "mean_token_accuracy": 0.1390085846185684,
      "num_tokens": 4056776.0,
      "step": 1970
    },
    {
      "entropy": 6.319344854354858,
      "epoch": 1.75277407900577,
      "grad_norm": 1.125,
      "learning_rate": 0.0004900676038380429,
      "loss": 6.0406,
      "mean_token_accuracy": 0.16621824055910112,
      "num_tokens": 4066326.0,
      "step": 1975
    },
    {
      "entropy": 6.454939699172973,
      "epoch": 1.7572126054150021,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004899661334806342,
      "loss": 6.1759,
      "mean_token_accuracy": 0.15071096420288085,
      "num_tokens": 4077478.0,
      "step": 1980
    },
    {
      "entropy": 6.262599611282349,
      "epoch": 1.7616511318242343,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004898641592572965,
      "loss": 6.1057,
      "mean_token_accuracy": 0.15066718459129333,
      "num_tokens": 4087661.0,
      "step": 1985
    },
    {
      "entropy": 6.442129898071289,
      "epoch": 1.7660896582334664,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004897616814070505,
      "loss": 6.1694,
      "mean_token_accuracy": 0.15258410200476646,
      "num_tokens": 4097450.0,
      "step": 1990
    },
    {
      "entropy": 6.382274055480957,
      "epoch": 1.7705281846426986,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004896587001700972,
      "loss": 6.2003,
      "mean_token_accuracy": 0.14331723377108574,
      "num_tokens": 4106850.0,
      "step": 1995
    },
    {
      "entropy": 6.4386186599731445,
      "epoch": 1.7749667110519307,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004895552157878174,
      "loss": 6.2273,
      "mean_token_accuracy": 0.1446727268397808,
      "num_tokens": 4117561.0,
      "step": 2000
    },
    {
      "epoch": 1.7749667110519307,
      "eval_entropy": 6.304367476574315,
      "eval_loss": 6.241267681121826,
      "eval_mean_token_accuracy": 0.14483241421876214,
      "eval_num_tokens": 4117561.0,
      "eval_runtime": 2.0862,
      "eval_samples_per_second": 1613.972,
      "eval_steps_per_second": 201.806,
      "step": 2000
    },
    {
      "entropy": 6.513484525680542,
      "epoch": 1.7794052374611629,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004894512285027717,
      "loss": 6.1375,
      "mean_token_accuracy": 0.15294163376092912,
      "num_tokens": 4126913.0,
      "step": 2005
    },
    {
      "entropy": 6.32349796295166,
      "epoch": 1.783843763870395,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004893467385586991,
      "loss": 6.1763,
      "mean_token_accuracy": 0.14652613997459413,
      "num_tokens": 4136501.0,
      "step": 2010
    },
    {
      "entropy": 6.455650901794433,
      "epoch": 1.7882822902796272,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000489241746200517,
      "loss": 6.1304,
      "mean_token_accuracy": 0.14899376183748245,
      "num_tokens": 4147263.0,
      "step": 2015
    },
    {
      "entropy": 6.39251856803894,
      "epoch": 1.7927208166888593,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004891362516743201,
      "loss": 6.1805,
      "mean_token_accuracy": 0.1439195953309536,
      "num_tokens": 4159251.0,
      "step": 2020
    },
    {
      "entropy": 6.375892972946167,
      "epoch": 1.7971593430980914,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004890302552273805,
      "loss": 6.1605,
      "mean_token_accuracy": 0.14276045113801955,
      "num_tokens": 4170601.0,
      "step": 2025
    },
    {
      "entropy": 6.455142402648926,
      "epoch": 1.8015978695073236,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004889237571081465,
      "loss": 6.1291,
      "mean_token_accuracy": 0.1502329871058464,
      "num_tokens": 4181650.0,
      "step": 2030
    },
    {
      "entropy": 6.3142640590667725,
      "epoch": 1.8060363959165557,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004888167575662425,
      "loss": 6.0242,
      "mean_token_accuracy": 0.1540951281785965,
      "num_tokens": 4191514.0,
      "step": 2035
    },
    {
      "entropy": 6.364542388916016,
      "epoch": 1.8104749223257879,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004887092568524682,
      "loss": 6.1543,
      "mean_token_accuracy": 0.1545771673321724,
      "num_tokens": 4200909.0,
      "step": 2040
    },
    {
      "entropy": 6.418929243087769,
      "epoch": 1.81491344873502,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004886012552187979,
      "loss": 6.0958,
      "mean_token_accuracy": 0.1500440776348114,
      "num_tokens": 4211139.0,
      "step": 2045
    },
    {
      "entropy": 6.353375577926636,
      "epoch": 1.8193519751442522,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004884927529183799,
      "loss": 6.1404,
      "mean_token_accuracy": 0.1507970094680786,
      "num_tokens": 4221754.0,
      "step": 2050
    },
    {
      "entropy": 6.401797246932984,
      "epoch": 1.8237905015534843,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004883837502055363,
      "loss": 6.0354,
      "mean_token_accuracy": 0.15554228723049163,
      "num_tokens": 4232771.0,
      "step": 2055
    },
    {
      "entropy": 6.274783992767334,
      "epoch": 1.8282290279627165,
      "grad_norm": 1.0,
      "learning_rate": 0.0004882742473357619,
      "loss": 6.047,
      "mean_token_accuracy": 0.162461756169796,
      "num_tokens": 4242860.0,
      "step": 2060
    },
    {
      "entropy": 6.384546136856079,
      "epoch": 1.8326675543719486,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048816424456572403,
      "loss": 6.2057,
      "mean_token_accuracy": 0.1450534589588642,
      "num_tokens": 4253442.0,
      "step": 2065
    },
    {
      "entropy": 6.373266220092773,
      "epoch": 1.8371060807811808,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004880537421532618,
      "loss": 6.1037,
      "mean_token_accuracy": 0.14555230140686035,
      "num_tokens": 4263856.0,
      "step": 2070
    },
    {
      "entropy": 6.2938440322875975,
      "epoch": 1.841544607190413,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048794274035738515,
      "loss": 6.0456,
      "mean_token_accuracy": 0.158289635181427,
      "num_tokens": 4274251.0,
      "step": 2075
    },
    {
      "entropy": 6.333366107940674,
      "epoch": 1.845983133599645,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004878312394382747,
      "loss": 6.0724,
      "mean_token_accuracy": 0.15534539371728898,
      "num_tokens": 4283486.0,
      "step": 2080
    },
    {
      "entropy": 6.355884695053101,
      "epoch": 1.8504216600088772,
      "grad_norm": 1.03125,
      "learning_rate": 0.000487719239657281,
      "loss": 6.168,
      "mean_token_accuracy": 0.14514408335089685,
      "num_tokens": 4293908.0,
      "step": 2085
    },
    {
      "entropy": 6.315519952774048,
      "epoch": 1.8548601864181093,
      "grad_norm": 1.03125,
      "learning_rate": 0.000487606741276924,
      "loss": 6.0074,
      "mean_token_accuracy": 0.15041493326425553,
      "num_tokens": 4303255.0,
      "step": 2090
    },
    {
      "entropy": 6.367549180984497,
      "epoch": 1.8592987128273413,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004874937445608921,
      "loss": 6.222,
      "mean_token_accuracy": 0.14236995279788972,
      "num_tokens": 4313981.0,
      "step": 2095
    },
    {
      "entropy": 6.414429569244385,
      "epoch": 1.8637372392365734,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004873802497740418,
      "loss": 6.1659,
      "mean_token_accuracy": 0.1489059269428253,
      "num_tokens": 4325010.0,
      "step": 2100
    },
    {
      "entropy": 6.336979866027832,
      "epoch": 1.8681757656458056,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004872662571823973,
      "loss": 6.142,
      "mean_token_accuracy": 0.15032578706741334,
      "num_tokens": 4336279.0,
      "step": 2105
    },
    {
      "entropy": 6.276753234863281,
      "epoch": 1.8726142920550377,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048715176705314936,
      "loss": 6.0886,
      "mean_token_accuracy": 0.15055317282676697,
      "num_tokens": 4345816.0,
      "step": 2110
    },
    {
      "entropy": 6.390111589431763,
      "epoch": 1.8770528184642699,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048703677965465506,
      "loss": 6.1546,
      "mean_token_accuracy": 0.14694484174251557,
      "num_tokens": 4356520.0,
      "step": 2115
    },
    {
      "entropy": 6.38926854133606,
      "epoch": 1.881491344873502,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048692129525643694,
      "loss": 6.0455,
      "mean_token_accuracy": 0.15440051704645158,
      "num_tokens": 4367517.0,
      "step": 2120
    },
    {
      "entropy": 6.319567394256592,
      "epoch": 1.8859298712827341,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048680531412918267,
      "loss": 6.1221,
      "mean_token_accuracy": 0.14795979335904122,
      "num_tokens": 4378043.0,
      "step": 2125
    },
    {
      "entropy": 6.339055061340332,
      "epoch": 1.8903683976919663,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004866888365447439,
      "loss": 6.1116,
      "mean_token_accuracy": 0.15579188615083694,
      "num_tokens": 4388098.0,
      "step": 2130
    },
    {
      "entropy": 6.342102432250977,
      "epoch": 1.8948069241011984,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004865718627761363,
      "loss": 6.0351,
      "mean_token_accuracy": 0.16426317244768143,
      "num_tokens": 4397745.0,
      "step": 2135
    },
    {
      "entropy": 6.255016899108886,
      "epoch": 1.8992454505104306,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004864543930975383,
      "loss": 6.1125,
      "mean_token_accuracy": 0.1508719764649868,
      "num_tokens": 4407995.0,
      "step": 2140
    },
    {
      "entropy": 6.372575426101685,
      "epoch": 1.9036839769196625,
      "grad_norm": 1.125,
      "learning_rate": 0.0004863364277842908,
      "loss": 6.0864,
      "mean_token_accuracy": 0.1544354021549225,
      "num_tokens": 4417574.0,
      "step": 2145
    },
    {
      "entropy": 6.287673425674439,
      "epoch": 1.9081225033288947,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004862179671128965,
      "loss": 6.0029,
      "mean_token_accuracy": 0.1653035208582878,
      "num_tokens": 4428772.0,
      "step": 2150
    },
    {
      "entropy": 6.284837818145752,
      "epoch": 1.9125610297381268,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000486099011361019,
      "loss": 6.0921,
      "mean_token_accuracy": 0.156430184841156,
      "num_tokens": 4438971.0,
      "step": 2155
    },
    {
      "entropy": 6.439360094070435,
      "epoch": 1.916999556147359,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048597956080748264,
      "loss": 6.2184,
      "mean_token_accuracy": 0.147563187032938,
      "num_tokens": 4449685.0,
      "step": 2160
    },
    {
      "entropy": 6.310244178771972,
      "epoch": 1.921438082556591,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048585961573227116,
      "loss": 6.1103,
      "mean_token_accuracy": 0.15256588608026506,
      "num_tokens": 4460260.0,
      "step": 2165
    },
    {
      "entropy": 6.324094629287719,
      "epoch": 1.9258766089658232,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004857391764165277,
      "loss": 6.0654,
      "mean_token_accuracy": 0.15756789296865464,
      "num_tokens": 4471149.0,
      "step": 2170
    },
    {
      "entropy": 6.27415452003479,
      "epoch": 1.9303151353750554,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048561824314255383,
      "loss": 6.0969,
      "mean_token_accuracy": 0.15434358417987823,
      "num_tokens": 4481506.0,
      "step": 2175
    },
    {
      "entropy": 6.373397016525269,
      "epoch": 1.9347536617842875,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048549681619380886,
      "loss": 6.1389,
      "mean_token_accuracy": 0.14398213624954223,
      "num_tokens": 4492541.0,
      "step": 2180
    },
    {
      "entropy": 6.348176717758179,
      "epoch": 1.9391921881935197,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004853748958549092,
      "loss": 6.0879,
      "mean_token_accuracy": 0.15113953053951262,
      "num_tokens": 4502727.0,
      "step": 2185
    },
    {
      "entropy": 6.243179750442505,
      "epoch": 1.9436307146027518,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004852524824116278,
      "loss": 6.1136,
      "mean_token_accuracy": 0.14689265042543412,
      "num_tokens": 4512657.0,
      "step": 2190
    },
    {
      "entropy": 6.332787084579468,
      "epoch": 1.948069241011984,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048512957615089354,
      "loss": 5.9777,
      "mean_token_accuracy": 0.15856263190507888,
      "num_tokens": 4522083.0,
      "step": 2195
    },
    {
      "entropy": 6.226612901687622,
      "epoch": 1.9525077674212161,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004850061773607902,
      "loss": 6.0255,
      "mean_token_accuracy": 0.1586281806230545,
      "num_tokens": 4531882.0,
      "step": 2200
    },
    {
      "entropy": 6.2661069393157955,
      "epoch": 1.9569462938304483,
      "grad_norm": 1.078125,
      "learning_rate": 0.000484882286330556,
      "loss": 6.0668,
      "mean_token_accuracy": 0.15751879066228866,
      "num_tokens": 4541724.0,
      "step": 2205
    },
    {
      "entropy": 6.290356159210205,
      "epoch": 1.9613848202396804,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004847579033505833,
      "loss": 6.0823,
      "mean_token_accuracy": 0.1551389679312706,
      "num_tokens": 4551889.0,
      "step": 2210
    },
    {
      "entropy": 6.299551630020142,
      "epoch": 1.9658233466489126,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004846330287124171,
      "loss": 6.1149,
      "mean_token_accuracy": 0.1571243390440941,
      "num_tokens": 4561792.0,
      "step": 2215
    },
    {
      "entropy": 6.349295568466187,
      "epoch": 1.9702618730581447,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048450766270875513,
      "loss": 6.0104,
      "mean_token_accuracy": 0.1578826203942299,
      "num_tokens": 4572021.0,
      "step": 2220
    },
    {
      "entropy": 6.304033708572388,
      "epoch": 1.9747003994673769,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048438180563344666,
      "loss": 6.0659,
      "mean_token_accuracy": 0.15220063626766206,
      "num_tokens": 4582800.0,
      "step": 2225
    },
    {
      "entropy": 6.356626272201538,
      "epoch": 1.979138925876609,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048425545778149213,
      "loss": 6.1376,
      "mean_token_accuracy": 0.1478087991476059,
      "num_tokens": 4593035.0,
      "step": 2230
    },
    {
      "entropy": 6.293792295455932,
      "epoch": 1.9835774522858411,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004841286194490423,
      "loss": 6.0004,
      "mean_token_accuracy": 0.1586893692612648,
      "num_tokens": 4602625.0,
      "step": 2235
    },
    {
      "entropy": 6.298813009262085,
      "epoch": 1.9880159786950733,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048400129093339745,
      "loss": 6.0939,
      "mean_token_accuracy": 0.15187210515141486,
      "num_tokens": 4613419.0,
      "step": 2240
    },
    {
      "entropy": 6.2934352397918705,
      "epoch": 1.9924545051043054,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048387347253300703,
      "loss": 6.0592,
      "mean_token_accuracy": 0.15442992970347405,
      "num_tokens": 4622575.0,
      "step": 2245
    },
    {
      "entropy": 6.2924830436706545,
      "epoch": 1.9968930315135376,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004837451645474685,
      "loss": 6.0589,
      "mean_token_accuracy": 0.15284667909145355,
      "num_tokens": 4632497.0,
      "step": 2250
    },
    {
      "entropy": 6.336415661705865,
      "epoch": 2.0008877052818463,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048361636727752716,
      "loss": 6.0301,
      "mean_token_accuracy": 0.15711821946832868,
      "num_tokens": 4641305.0,
      "step": 2255
    },
    {
      "entropy": 6.320528984069824,
      "epoch": 2.0053262316910785,
      "grad_norm": 1.109375,
      "learning_rate": 0.000483487081025075,
      "loss": 5.826,
      "mean_token_accuracy": 0.16330702155828475,
      "num_tokens": 4650986.0,
      "step": 2260
    },
    {
      "entropy": 6.267225694656372,
      "epoch": 2.0097647581003106,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048335730609315,
      "loss": 5.7919,
      "mean_token_accuracy": 0.165120992064476,
      "num_tokens": 4661836.0,
      "step": 2265
    },
    {
      "entropy": 6.285591506958008,
      "epoch": 2.014203284509543,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048322704278593595,
      "loss": 5.7822,
      "mean_token_accuracy": 0.1718212589621544,
      "num_tokens": 4671325.0,
      "step": 2270
    },
    {
      "entropy": 6.2631378173828125,
      "epoch": 2.018641810918775,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048309629140876097,
      "loss": 5.7983,
      "mean_token_accuracy": 0.16487672328948974,
      "num_tokens": 4682334.0,
      "step": 2275
    },
    {
      "entropy": 6.278176546096802,
      "epoch": 2.023080337328007,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004829650522680974,
      "loss": 5.8037,
      "mean_token_accuracy": 0.16945539861917497,
      "num_tokens": 4691486.0,
      "step": 2280
    },
    {
      "entropy": 6.2450908660888675,
      "epoch": 2.027518863737239,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004828333256715609,
      "loss": 5.7808,
      "mean_token_accuracy": 0.1654707446694374,
      "num_tokens": 4700841.0,
      "step": 2285
    },
    {
      "entropy": 6.16944465637207,
      "epoch": 2.0319573901464714,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004827011119279096,
      "loss": 5.7364,
      "mean_token_accuracy": 0.17380234599113464,
      "num_tokens": 4709842.0,
      "step": 2290
    },
    {
      "entropy": 6.2213818550109865,
      "epoch": 2.0363959165557035,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048256841134704335,
      "loss": 5.7495,
      "mean_token_accuracy": 0.1679978162050247,
      "num_tokens": 4720193.0,
      "step": 2295
    },
    {
      "entropy": 6.182414245605469,
      "epoch": 2.0408344429649357,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048243522424000333,
      "loss": 5.8063,
      "mean_token_accuracy": 0.1693178117275238,
      "num_tokens": 4729503.0,
      "step": 2300
    },
    {
      "entropy": 6.248026180267334,
      "epoch": 2.045272969374168,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000482301550918971,
      "loss": 5.7874,
      "mean_token_accuracy": 0.17118435055017472,
      "num_tokens": 4740241.0,
      "step": 2305
    },
    {
      "entropy": 6.120760440826416,
      "epoch": 2.0497114957834,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004821673916972676,
      "loss": 5.676,
      "mean_token_accuracy": 0.18444487154483796,
      "num_tokens": 4749616.0,
      "step": 2310
    },
    {
      "entropy": 6.1944496631622314,
      "epoch": 2.054150022192632,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000482032746889353,
      "loss": 5.7767,
      "mean_token_accuracy": 0.16697147488594055,
      "num_tokens": 4759846.0,
      "step": 2315
    },
    {
      "entropy": 6.24187216758728,
      "epoch": 2.0585885486018642,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048189761681082557,
      "loss": 5.8296,
      "mean_token_accuracy": 0.16175905466079712,
      "num_tokens": 4769441.0,
      "step": 2320
    },
    {
      "entropy": 6.2141406536102295,
      "epoch": 2.0630270750110964,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004817620017784209,
      "loss": 5.7617,
      "mean_token_accuracy": 0.16938935965299606,
      "num_tokens": 4779962.0,
      "step": 2325
    },
    {
      "entropy": 6.242314195632934,
      "epoch": 2.0674656014203285,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048162590211001143,
      "loss": 5.8326,
      "mean_token_accuracy": 0.16407378911972045,
      "num_tokens": 4790275.0,
      "step": 2330
    },
    {
      "entropy": 6.061077022552491,
      "epoch": 2.0719041278295607,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048148931812460536,
      "loss": 5.7234,
      "mean_token_accuracy": 0.17769070863723754,
      "num_tokens": 4800799.0,
      "step": 2335
    },
    {
      "entropy": 6.300735187530518,
      "epoch": 2.076342654238793,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004813522501423464,
      "loss": 5.8311,
      "mean_token_accuracy": 0.16275566965341567,
      "num_tokens": 4811786.0,
      "step": 2340
    },
    {
      "entropy": 6.152251577377319,
      "epoch": 2.080781180648025,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004812146984845124,
      "loss": 5.7383,
      "mean_token_accuracy": 0.17451401948928832,
      "num_tokens": 4821321.0,
      "step": 2345
    },
    {
      "entropy": 6.21392674446106,
      "epoch": 2.085219707057257,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048107666347351505,
      "loss": 5.7691,
      "mean_token_accuracy": 0.17131485641002656,
      "num_tokens": 4831418.0,
      "step": 2350
    },
    {
      "entropy": 6.152781057357788,
      "epoch": 2.0896582334664893,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048093814543289894,
      "loss": 5.7544,
      "mean_token_accuracy": 0.1736764594912529,
      "num_tokens": 4843326.0,
      "step": 2355
    },
    {
      "entropy": 6.162942695617676,
      "epoch": 2.0940967598757214,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048079914468734117,
      "loss": 5.7209,
      "mean_token_accuracy": 0.17366211116313934,
      "num_tokens": 4853946.0,
      "step": 2360
    },
    {
      "entropy": 6.194116878509521,
      "epoch": 2.0985352862849536,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048065966156264964,
      "loss": 5.761,
      "mean_token_accuracy": 0.1711730867624283,
      "num_tokens": 4863494.0,
      "step": 2365
    },
    {
      "entropy": 6.145648765563965,
      "epoch": 2.1029738126941857,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048051969638576345,
      "loss": 5.749,
      "mean_token_accuracy": 0.16969802528619765,
      "num_tokens": 4872583.0,
      "step": 2370
    },
    {
      "entropy": 6.198772192001343,
      "epoch": 2.107412339103418,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048037924948475134,
      "loss": 5.6999,
      "mean_token_accuracy": 0.18126114159822465,
      "num_tokens": 4882333.0,
      "step": 2375
    },
    {
      "entropy": 6.133423471450806,
      "epoch": 2.11185086551265,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004802383211888114,
      "loss": 5.8388,
      "mean_token_accuracy": 0.16247562766075135,
      "num_tokens": 4892874.0,
      "step": 2380
    },
    {
      "entropy": 6.190569829940796,
      "epoch": 2.116289391921882,
      "grad_norm": 1.0,
      "learning_rate": 0.0004800969118282697,
      "loss": 5.7077,
      "mean_token_accuracy": 0.17178617119789125,
      "num_tokens": 4903666.0,
      "step": 2385
    },
    {
      "entropy": 6.148941993713379,
      "epoch": 2.120727918331114,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004799550217345803,
      "loss": 5.8135,
      "mean_token_accuracy": 0.16479915231466294,
      "num_tokens": 4913565.0,
      "step": 2390
    },
    {
      "entropy": 6.25219759941101,
      "epoch": 2.125166444740346,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047981265124032375,
      "loss": 5.8806,
      "mean_token_accuracy": 0.16119145303964616,
      "num_tokens": 4924664.0,
      "step": 2395
    },
    {
      "entropy": 6.1365477561950685,
      "epoch": 2.129604971149578,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047966980067920686,
      "loss": 5.787,
      "mean_token_accuracy": 0.16446771174669267,
      "num_tokens": 4935361.0,
      "step": 2400
    },
    {
      "entropy": 6.212709236145019,
      "epoch": 2.1340434975588103,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047952647038606157,
      "loss": 5.7139,
      "mean_token_accuracy": 0.1764111801981926,
      "num_tokens": 4946954.0,
      "step": 2405
    },
    {
      "entropy": 6.166495609283447,
      "epoch": 2.1384820239680424,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004793826606968443,
      "loss": 5.8285,
      "mean_token_accuracy": 0.1607503190636635,
      "num_tokens": 4957497.0,
      "step": 2410
    },
    {
      "entropy": 6.101367950439453,
      "epoch": 2.1429205503772746,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000479238371948635,
      "loss": 5.7757,
      "mean_token_accuracy": 0.16357363164424896,
      "num_tokens": 4968069.0,
      "step": 2415
    },
    {
      "entropy": 6.219218921661377,
      "epoch": 2.1473590767865067,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004790936044796369,
      "loss": 5.7844,
      "mean_token_accuracy": 0.17338328510522844,
      "num_tokens": 4978417.0,
      "step": 2420
    },
    {
      "entropy": 6.17771635055542,
      "epoch": 2.151797603195739,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047894835862917473,
      "loss": 5.7621,
      "mean_token_accuracy": 0.16970881968736648,
      "num_tokens": 4989670.0,
      "step": 2425
    },
    {
      "entropy": 6.143991899490357,
      "epoch": 2.156236129604971,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047880263473769514,
      "loss": 5.8096,
      "mean_token_accuracy": 0.16401513367891313,
      "num_tokens": 5000346.0,
      "step": 2430
    },
    {
      "entropy": 6.209756994247437,
      "epoch": 2.160674656014203,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004786564331467648,
      "loss": 5.7988,
      "mean_token_accuracy": 0.16377755850553513,
      "num_tokens": 5011878.0,
      "step": 2435
    },
    {
      "entropy": 6.224458789825439,
      "epoch": 2.1651131824234353,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047850975419907034,
      "loss": 5.8718,
      "mean_token_accuracy": 0.162304849922657,
      "num_tokens": 5022939.0,
      "step": 2440
    },
    {
      "entropy": 6.232002401351929,
      "epoch": 2.1695517088326675,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047836259823841716,
      "loss": 5.7963,
      "mean_token_accuracy": 0.1669462203979492,
      "num_tokens": 5033613.0,
      "step": 2445
    },
    {
      "entropy": 6.125792789459228,
      "epoch": 2.1739902352418996,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004782149656097287,
      "loss": 5.7654,
      "mean_token_accuracy": 0.1679580643773079,
      "num_tokens": 5043926.0,
      "step": 2450
    },
    {
      "entropy": 6.151848697662354,
      "epoch": 2.1784287616511318,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047806685665904586,
      "loss": 5.8006,
      "mean_token_accuracy": 0.17178182303905487,
      "num_tokens": 5054179.0,
      "step": 2455
    },
    {
      "entropy": 6.080358839035034,
      "epoch": 2.182867288060364,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004779182717335258,
      "loss": 5.7625,
      "mean_token_accuracy": 0.16996095776557923,
      "num_tokens": 5063654.0,
      "step": 2460
    },
    {
      "entropy": 6.169263648986816,
      "epoch": 2.187305814469596,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047776921118144154,
      "loss": 5.8695,
      "mean_token_accuracy": 0.1624024137854576,
      "num_tokens": 5073558.0,
      "step": 2465
    },
    {
      "entropy": 6.250872087478638,
      "epoch": 2.191744340878828,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047761967535218084,
      "loss": 5.8175,
      "mean_token_accuracy": 0.16334515810012817,
      "num_tokens": 5083218.0,
      "step": 2470
    },
    {
      "entropy": 6.116773986816407,
      "epoch": 2.1961828672880603,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004774696645962453,
      "loss": 5.7848,
      "mean_token_accuracy": 0.17051488161087036,
      "num_tokens": 5093037.0,
      "step": 2475
    },
    {
      "entropy": 6.085236072540283,
      "epoch": 2.2006213936972925,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004773191792652501,
      "loss": 5.7379,
      "mean_token_accuracy": 0.17253995835781097,
      "num_tokens": 5103434.0,
      "step": 2480
    },
    {
      "entropy": 6.203386831283569,
      "epoch": 2.2050599201065246,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004771682197119224,
      "loss": 5.8513,
      "mean_token_accuracy": 0.16081491261720657,
      "num_tokens": 5113325.0,
      "step": 2485
    },
    {
      "entropy": 6.135442781448364,
      "epoch": 2.2094984465157568,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047701678629010115,
      "loss": 5.812,
      "mean_token_accuracy": 0.16577334851026534,
      "num_tokens": 5124957.0,
      "step": 2490
    },
    {
      "entropy": 6.2259856224060055,
      "epoch": 2.213936972924989,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004768648793547359,
      "loss": 5.7559,
      "mean_token_accuracy": 0.17162533700466157,
      "num_tokens": 5134491.0,
      "step": 2495
    },
    {
      "entropy": 6.040803050994873,
      "epoch": 2.218375499334221,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004767124992618863,
      "loss": 5.7236,
      "mean_token_accuracy": 0.17656566202640533,
      "num_tokens": 5145063.0,
      "step": 2500
    },
    {
      "epoch": 2.218375499334221,
      "eval_entropy": 5.758825140157958,
      "eval_loss": 6.117055416107178,
      "eval_mean_token_accuracy": 0.15501565050838112,
      "eval_num_tokens": 5145063.0,
      "eval_runtime": 2.0862,
      "eval_samples_per_second": 1613.95,
      "eval_steps_per_second": 201.804,
      "step": 2500
    },
    {
      "entropy": 6.138974285125732,
      "epoch": 2.222814025743453,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004765596463687207,
      "loss": 5.839,
      "mean_token_accuracy": 0.16339409202337266,
      "num_tokens": 5155396.0,
      "step": 2505
    },
    {
      "entropy": 6.151093578338623,
      "epoch": 2.2272525521526854,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047640632103351576,
      "loss": 5.7831,
      "mean_token_accuracy": 0.16532318145036698,
      "num_tokens": 5166499.0,
      "step": 2510
    },
    {
      "entropy": 6.171177387237549,
      "epoch": 2.2316910785619175,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047625252361565586,
      "loss": 5.8143,
      "mean_token_accuracy": 0.1652636408805847,
      "num_tokens": 5177530.0,
      "step": 2515
    },
    {
      "entropy": 6.115115642547607,
      "epoch": 2.2361296049711497,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047609825447563137,
      "loss": 5.6925,
      "mean_token_accuracy": 0.1710483983159065,
      "num_tokens": 5187427.0,
      "step": 2520
    },
    {
      "entropy": 6.033501386642456,
      "epoch": 2.240568131380382,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004759435139750388,
      "loss": 5.6936,
      "mean_token_accuracy": 0.17407233715057374,
      "num_tokens": 5198888.0,
      "step": 2525
    },
    {
      "entropy": 6.088896322250366,
      "epoch": 2.245006657789614,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047578830247657915,
      "loss": 5.6968,
      "mean_token_accuracy": 0.17464374303817748,
      "num_tokens": 5208463.0,
      "step": 2530
    },
    {
      "entropy": 6.135959005355835,
      "epoch": 2.249445184198846,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047563262034405773,
      "loss": 5.8129,
      "mean_token_accuracy": 0.16065521836280822,
      "num_tokens": 5219989.0,
      "step": 2535
    },
    {
      "entropy": 6.196387863159179,
      "epoch": 2.2538837106080782,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047547646794238277,
      "loss": 5.823,
      "mean_token_accuracy": 0.16879613399505616,
      "num_tokens": 5231443.0,
      "step": 2540
    },
    {
      "entropy": 6.069067049026489,
      "epoch": 2.2583222370173104,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004753198456375647,
      "loss": 5.7687,
      "mean_token_accuracy": 0.17048330157995223,
      "num_tokens": 5241633.0,
      "step": 2545
    },
    {
      "entropy": 6.197365951538086,
      "epoch": 2.2627607634265425,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004751627537967158,
      "loss": 5.7616,
      "mean_token_accuracy": 0.17182915061712264,
      "num_tokens": 5252486.0,
      "step": 2550
    },
    {
      "entropy": 6.106606483459473,
      "epoch": 2.2671992898357747,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047500519278804835,
      "loss": 5.7842,
      "mean_token_accuracy": 0.16441592425107956,
      "num_tokens": 5262672.0,
      "step": 2555
    },
    {
      "entropy": 6.161299324035644,
      "epoch": 2.271637816245007,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004748471629808746,
      "loss": 5.7257,
      "mean_token_accuracy": 0.1692223384976387,
      "num_tokens": 5273547.0,
      "step": 2560
    },
    {
      "entropy": 6.044488430023193,
      "epoch": 2.276076342654239,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047468866474560575,
      "loss": 5.7477,
      "mean_token_accuracy": 0.1677919000387192,
      "num_tokens": 5284935.0,
      "step": 2565
    },
    {
      "entropy": 6.1366087913513185,
      "epoch": 2.280514869063471,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047452969845375074,
      "loss": 5.7731,
      "mean_token_accuracy": 0.17664981484413148,
      "num_tokens": 5295237.0,
      "step": 2570
    },
    {
      "entropy": 6.143746471405029,
      "epoch": 2.2849533954727033,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004743702644779157,
      "loss": 5.8109,
      "mean_token_accuracy": 0.16415809839963913,
      "num_tokens": 5305288.0,
      "step": 2575
    },
    {
      "entropy": 6.148390245437622,
      "epoch": 2.2893919218819354,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000474210363191803,
      "loss": 5.7359,
      "mean_token_accuracy": 0.1691213622689247,
      "num_tokens": 5316031.0,
      "step": 2580
    },
    {
      "entropy": 6.067453241348266,
      "epoch": 2.2938304482911676,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004740499949702103,
      "loss": 5.7778,
      "mean_token_accuracy": 0.16342108249664306,
      "num_tokens": 5327544.0,
      "step": 2585
    },
    {
      "entropy": 6.038386487960816,
      "epoch": 2.2982689747003997,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004738891601890298,
      "loss": 5.718,
      "mean_token_accuracy": 0.17032305002212525,
      "num_tokens": 5338282.0,
      "step": 2590
    },
    {
      "entropy": 6.144178962707519,
      "epoch": 2.302707501109632,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004737278592252473,
      "loss": 5.7589,
      "mean_token_accuracy": 0.17450494766235353,
      "num_tokens": 5348300.0,
      "step": 2595
    },
    {
      "entropy": 6.15695972442627,
      "epoch": 2.3071460275188636,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004735660924569411,
      "loss": 5.8636,
      "mean_token_accuracy": 0.16162828356027603,
      "num_tokens": 5360193.0,
      "step": 2600
    },
    {
      "entropy": 6.121260452270508,
      "epoch": 2.3115845539280957,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004734038602632815,
      "loss": 5.7297,
      "mean_token_accuracy": 0.1731833279132843,
      "num_tokens": 5369655.0,
      "step": 2605
    },
    {
      "entropy": 6.111371564865112,
      "epoch": 2.316023080337328,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047324116302452975,
      "loss": 5.792,
      "mean_token_accuracy": 0.16499196141958236,
      "num_tokens": 5379654.0,
      "step": 2610
    },
    {
      "entropy": 6.121448135375976,
      "epoch": 2.32046160674656,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004730780011220369,
      "loss": 5.7201,
      "mean_token_accuracy": 0.17435721009969712,
      "num_tokens": 5389527.0,
      "step": 2615
    },
    {
      "entropy": 6.137268733978272,
      "epoch": 2.324900133155792,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004729143749382435,
      "loss": 5.8075,
      "mean_token_accuracy": 0.1641774520277977,
      "num_tokens": 5399480.0,
      "step": 2620
    },
    {
      "entropy": 6.093860912322998,
      "epoch": 2.3293386595650243,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047275028485667793,
      "loss": 5.6667,
      "mean_token_accuracy": 0.17897211164236068,
      "num_tokens": 5410065.0,
      "step": 2625
    },
    {
      "entropy": 6.025512218475342,
      "epoch": 2.3337771859742564,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004725857312619563,
      "loss": 5.7156,
      "mean_token_accuracy": 0.17173065543174743,
      "num_tokens": 5419519.0,
      "step": 2630
    },
    {
      "entropy": 6.155104923248291,
      "epoch": 2.3382157123834886,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004724207145397809,
      "loss": 5.7943,
      "mean_token_accuracy": 0.1667987123131752,
      "num_tokens": 5429907.0,
      "step": 2635
    },
    {
      "entropy": 6.147578144073487,
      "epoch": 2.3426542387927207,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004722552350769397,
      "loss": 5.7846,
      "mean_token_accuracy": 0.1689673498272896,
      "num_tokens": 5440051.0,
      "step": 2640
    },
    {
      "entropy": 6.044529819488526,
      "epoch": 2.347092765201953,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047208929326130535,
      "loss": 5.7874,
      "mean_token_accuracy": 0.16316695511341095,
      "num_tokens": 5452596.0,
      "step": 2645
    },
    {
      "entropy": 6.1003374576568605,
      "epoch": 2.351531291611185,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047192288948183394,
      "loss": 5.8084,
      "mean_token_accuracy": 0.15916725099086762,
      "num_tokens": 5463133.0,
      "step": 2650
    },
    {
      "entropy": 6.074586248397827,
      "epoch": 2.355969818020417,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047175602412856453,
      "loss": 5.7162,
      "mean_token_accuracy": 0.17868943214416505,
      "num_tokens": 5473934.0,
      "step": 2655
    },
    {
      "entropy": 6.048851585388183,
      "epoch": 2.3604083444296493,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047158869759261843,
      "loss": 5.7626,
      "mean_token_accuracy": 0.1687912866473198,
      "num_tokens": 5483921.0,
      "step": 2660
    },
    {
      "entropy": 6.145990705490112,
      "epoch": 2.3648468708388815,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004714209102661973,
      "loss": 5.7686,
      "mean_token_accuracy": 0.16941458731889725,
      "num_tokens": 5494218.0,
      "step": 2665
    },
    {
      "entropy": 6.121583032608032,
      "epoch": 2.3692853972481136,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004712526625425832,
      "loss": 5.811,
      "mean_token_accuracy": 0.1643875315785408,
      "num_tokens": 5504942.0,
      "step": 2670
    },
    {
      "entropy": 6.102924728393555,
      "epoch": 2.3737239236573457,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047108395481613736,
      "loss": 5.8558,
      "mean_token_accuracy": 0.15613222420215606,
      "num_tokens": 5514624.0,
      "step": 2675
    },
    {
      "entropy": 6.14481372833252,
      "epoch": 2.378162450066578,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047091478748229927,
      "loss": 5.6984,
      "mean_token_accuracy": 0.17392003685235977,
      "num_tokens": 5524098.0,
      "step": 2680
    },
    {
      "entropy": 6.04828372001648,
      "epoch": 2.38260097647581,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004707451609375854,
      "loss": 5.7557,
      "mean_token_accuracy": 0.1692517578601837,
      "num_tokens": 5534585.0,
      "step": 2685
    },
    {
      "entropy": 6.10555362701416,
      "epoch": 2.387039502885042,
      "grad_norm": 1.125,
      "learning_rate": 0.0004705750755795889,
      "loss": 5.7966,
      "mean_token_accuracy": 0.168571537733078,
      "num_tokens": 5544136.0,
      "step": 2690
    },
    {
      "entropy": 6.117819595336914,
      "epoch": 2.3914780292942743,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047040453180697823,
      "loss": 5.691,
      "mean_token_accuracy": 0.1738590866327286,
      "num_tokens": 5554939.0,
      "step": 2695
    },
    {
      "entropy": 6.123975086212158,
      "epoch": 2.3959165557035065,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004702335300194963,
      "loss": 5.7762,
      "mean_token_accuracy": 0.16760747879743576,
      "num_tokens": 5564533.0,
      "step": 2700
    },
    {
      "entropy": 6.117423820495605,
      "epoch": 2.4003550821127386,
      "grad_norm": 1.0,
      "learning_rate": 0.0004700620706179596,
      "loss": 5.8196,
      "mean_token_accuracy": 0.16853979229927063,
      "num_tokens": 5575140.0,
      "step": 2705
    },
    {
      "entropy": 6.141474914550781,
      "epoch": 2.4047936085219708,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004698901540042573,
      "loss": 5.7775,
      "mean_token_accuracy": 0.16901845932006837,
      "num_tokens": 5585399.0,
      "step": 2710
    },
    {
      "entropy": 6.079065132141113,
      "epoch": 2.409232134931203,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046971778058135024,
      "loss": 5.6638,
      "mean_token_accuracy": 0.18335538804531099,
      "num_tokens": 5596194.0,
      "step": 2715
    },
    {
      "entropy": 6.042174386978149,
      "epoch": 2.413670661340435,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046954495075326986,
      "loss": 5.8439,
      "mean_token_accuracy": 0.16659024804830552,
      "num_tokens": 5608317.0,
      "step": 2720
    },
    {
      "entropy": 6.13245415687561,
      "epoch": 2.418109187749667,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046937166492511746,
      "loss": 5.725,
      "mean_token_accuracy": 0.1753625452518463,
      "num_tokens": 5617777.0,
      "step": 2725
    },
    {
      "entropy": 6.06634931564331,
      "epoch": 2.4225477141588994,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046919792350306317,
      "loss": 5.7311,
      "mean_token_accuracy": 0.1771342933177948,
      "num_tokens": 5628056.0,
      "step": 2730
    },
    {
      "entropy": 5.94120683670044,
      "epoch": 2.4269862405681315,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004690237268943451,
      "loss": 5.7222,
      "mean_token_accuracy": 0.17666102647781373,
      "num_tokens": 5638073.0,
      "step": 2735
    },
    {
      "entropy": 6.103239822387695,
      "epoch": 2.4314247669773636,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046884907550726816,
      "loss": 5.6812,
      "mean_token_accuracy": 0.1819820836186409,
      "num_tokens": 5647290.0,
      "step": 2740
    },
    {
      "entropy": 6.077079916000367,
      "epoch": 2.435863293386596,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046867396975120324,
      "loss": 5.7744,
      "mean_token_accuracy": 0.16964205503463745,
      "num_tokens": 5656882.0,
      "step": 2745
    },
    {
      "entropy": 6.1173424243927,
      "epoch": 2.440301819795828,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004684984100365863,
      "loss": 5.8316,
      "mean_token_accuracy": 0.16495220363140106,
      "num_tokens": 5667803.0,
      "step": 2750
    },
    {
      "entropy": 6.142762899398804,
      "epoch": 2.44474034620506,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046832239677491736,
      "loss": 5.7461,
      "mean_token_accuracy": 0.1722828820347786,
      "num_tokens": 5678319.0,
      "step": 2755
    },
    {
      "entropy": 6.022468090057373,
      "epoch": 2.4491788726142922,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004681459303787594,
      "loss": 5.7948,
      "mean_token_accuracy": 0.16484333127737044,
      "num_tokens": 5688677.0,
      "step": 2760
    },
    {
      "entropy": 6.150290822982788,
      "epoch": 2.4536173990235244,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004679690112617376,
      "loss": 5.8075,
      "mean_token_accuracy": 0.16610533744096756,
      "num_tokens": 5699538.0,
      "step": 2765
    },
    {
      "entropy": 6.095004320144653,
      "epoch": 2.458055925432756,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004677916398385383,
      "loss": 5.6928,
      "mean_token_accuracy": 0.18449453860521317,
      "num_tokens": 5709461.0,
      "step": 2770
    },
    {
      "entropy": 6.063209056854248,
      "epoch": 2.4624944518419882,
      "grad_norm": 1.109375,
      "learning_rate": 0.000467613816524908,
      "loss": 5.7058,
      "mean_token_accuracy": 0.17079256027936934,
      "num_tokens": 5718220.0,
      "step": 2775
    },
    {
      "entropy": 6.105566453933716,
      "epoch": 2.4669329782512204,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004674355417376524,
      "loss": 5.7678,
      "mean_token_accuracy": 0.16715284287929535,
      "num_tokens": 5728638.0,
      "step": 2780
    },
    {
      "entropy": 6.0401427268981935,
      "epoch": 2.4713715046604525,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046725681589463537,
      "loss": 5.7113,
      "mean_token_accuracy": 0.17297857254743576,
      "num_tokens": 5738854.0,
      "step": 2785
    },
    {
      "entropy": 6.040780830383301,
      "epoch": 2.4758100310696847,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046707763941477817,
      "loss": 5.6819,
      "mean_token_accuracy": 0.1767168715596199,
      "num_tokens": 5748624.0,
      "step": 2790
    },
    {
      "entropy": 6.007115411758423,
      "epoch": 2.480248557478917,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004668980127180582,
      "loss": 5.7215,
      "mean_token_accuracy": 0.17497175931930542,
      "num_tokens": 5759846.0,
      "step": 2795
    },
    {
      "entropy": 6.1033079624176025,
      "epoch": 2.484687083888149,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004667179362255081,
      "loss": 5.7512,
      "mean_token_accuracy": 0.17809866815805436,
      "num_tokens": 5769983.0,
      "step": 2800
    },
    {
      "entropy": 6.070840358734131,
      "epoch": 2.489125610297381,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004665374103592149,
      "loss": 5.7566,
      "mean_token_accuracy": 0.1682102620601654,
      "num_tokens": 5780691.0,
      "step": 2805
    },
    {
      "entropy": 6.047124242782592,
      "epoch": 2.4935641367066133,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004663564355423189,
      "loss": 5.6824,
      "mean_token_accuracy": 0.17610101848840715,
      "num_tokens": 5790690.0,
      "step": 2810
    },
    {
      "entropy": 6.104051399230957,
      "epoch": 2.4980026631158454,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004661750121990128,
      "loss": 5.7507,
      "mean_token_accuracy": 0.17295027673244476,
      "num_tokens": 5801088.0,
      "step": 2815
    },
    {
      "entropy": 5.899098825454712,
      "epoch": 2.5024411895250775,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046599314075454034,
      "loss": 5.6632,
      "mean_token_accuracy": 0.17894352227449417,
      "num_tokens": 5811451.0,
      "step": 2820
    },
    {
      "entropy": 6.042473268508911,
      "epoch": 2.5068797159343097,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046581082163519585,
      "loss": 5.6985,
      "mean_token_accuracy": 0.1734083727002144,
      "num_tokens": 5822024.0,
      "step": 2825
    },
    {
      "entropy": 5.998702621459961,
      "epoch": 2.511318242343542,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046562805526832284,
      "loss": 5.7612,
      "mean_token_accuracy": 0.1688615247607231,
      "num_tokens": 5831634.0,
      "step": 2830
    },
    {
      "entropy": 6.080638408660889,
      "epoch": 2.515756768752774,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004654448420823133,
      "loss": 5.7544,
      "mean_token_accuracy": 0.17119256556034088,
      "num_tokens": 5842422.0,
      "step": 2835
    },
    {
      "entropy": 6.030310201644897,
      "epoch": 2.520195295162006,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046526118250660636,
      "loss": 5.7404,
      "mean_token_accuracy": 0.17302993685007095,
      "num_tokens": 5851658.0,
      "step": 2840
    },
    {
      "entropy": 6.0097393035888675,
      "epoch": 2.5246338215712383,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004650770769716875,
      "loss": 5.7635,
      "mean_token_accuracy": 0.16970502585172653,
      "num_tokens": 5861900.0,
      "step": 2845
    },
    {
      "entropy": 6.13044376373291,
      "epoch": 2.5290723479804704,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004648925259090875,
      "loss": 5.8304,
      "mean_token_accuracy": 0.16660041362047195,
      "num_tokens": 5872675.0,
      "step": 2850
    },
    {
      "entropy": 6.098389196395874,
      "epoch": 2.5335108743897026,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046470752975138146,
      "loss": 5.7127,
      "mean_token_accuracy": 0.17505834698677064,
      "num_tokens": 5882768.0,
      "step": 2855
    },
    {
      "entropy": 5.994097661972046,
      "epoch": 2.5379494007989347,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046452208893218777,
      "loss": 5.7086,
      "mean_token_accuracy": 0.17897956669330597,
      "num_tokens": 5892970.0,
      "step": 2860
    },
    {
      "entropy": 6.031912326812744,
      "epoch": 2.542387927208167,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000464336203886167,
      "loss": 5.6679,
      "mean_token_accuracy": 0.17664587646722793,
      "num_tokens": 5903597.0,
      "step": 2865
    },
    {
      "entropy": 6.076654577255249,
      "epoch": 2.546826453617399,
      "grad_norm": 1.03125,
      "learning_rate": 0.000464149875049021,
      "loss": 5.7082,
      "mean_token_accuracy": 0.17591760605573653,
      "num_tokens": 5913908.0,
      "step": 2870
    },
    {
      "entropy": 6.035532999038696,
      "epoch": 2.551264980026631,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046396310285749175,
      "loss": 5.7447,
      "mean_token_accuracy": 0.17051061540842055,
      "num_tokens": 5923022.0,
      "step": 2875
    },
    {
      "entropy": 6.030104494094848,
      "epoch": 2.5557035064358633,
      "grad_norm": 1.125,
      "learning_rate": 0.00046377588774936077,
      "loss": 5.7383,
      "mean_token_accuracy": 0.17380277961492538,
      "num_tokens": 5933005.0,
      "step": 2880
    },
    {
      "entropy": 6.0437328815460205,
      "epoch": 2.5601420328450954,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046358823016344713,
      "loss": 5.7029,
      "mean_token_accuracy": 0.17715244591236115,
      "num_tokens": 5943197.0,
      "step": 2885
    },
    {
      "entropy": 6.017801523208618,
      "epoch": 2.5645805592543276,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004634001305396076,
      "loss": 5.7701,
      "mean_token_accuracy": 0.17026674896478652,
      "num_tokens": 5954788.0,
      "step": 2890
    },
    {
      "entropy": 6.168081951141358,
      "epoch": 2.5690190856635597,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046321158931873486,
      "loss": 5.7751,
      "mean_token_accuracy": 0.16409984081983567,
      "num_tokens": 5965896.0,
      "step": 2895
    },
    {
      "entropy": 5.9579003810882565,
      "epoch": 2.573457612072792,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004630226069427566,
      "loss": 5.616,
      "mean_token_accuracy": 0.18554621189832687,
      "num_tokens": 5975492.0,
      "step": 2900
    },
    {
      "entropy": 6.020355606079102,
      "epoch": 2.577896138482024,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046283318385463454,
      "loss": 5.6971,
      "mean_token_accuracy": 0.1745652511715889,
      "num_tokens": 5985489.0,
      "step": 2905
    },
    {
      "entropy": 5.985205554962159,
      "epoch": 2.582334664891256,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004626433204983636,
      "loss": 5.647,
      "mean_token_accuracy": 0.174534772336483,
      "num_tokens": 5995031.0,
      "step": 2910
    },
    {
      "entropy": 6.006907415390015,
      "epoch": 2.5867731913004883,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046245301731897024,
      "loss": 5.7595,
      "mean_token_accuracy": 0.16590401232242585,
      "num_tokens": 6005010.0,
      "step": 2915
    },
    {
      "entropy": 6.114031076431274,
      "epoch": 2.5912117177097205,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046226227476251256,
      "loss": 5.6742,
      "mean_token_accuracy": 0.1767223834991455,
      "num_tokens": 6016403.0,
      "step": 2920
    },
    {
      "entropy": 5.938989210128784,
      "epoch": 2.5956502441189526,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004620710932760776,
      "loss": 5.6682,
      "mean_token_accuracy": 0.18038018196821212,
      "num_tokens": 6026312.0,
      "step": 2925
    },
    {
      "entropy": 6.016516923904419,
      "epoch": 2.6000887705281848,
      "grad_norm": 1.078125,
      "learning_rate": 0.000461879473307782,
      "loss": 5.867,
      "mean_token_accuracy": 0.16858119815587996,
      "num_tokens": 6036857.0,
      "step": 2930
    },
    {
      "entropy": 6.142938184738159,
      "epoch": 2.604527296937417,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004616874153067698,
      "loss": 5.7728,
      "mean_token_accuracy": 0.16579627990722656,
      "num_tokens": 6048270.0,
      "step": 2935
    },
    {
      "entropy": 6.05487551689148,
      "epoch": 2.608965823346649,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004614949197232118,
      "loss": 5.696,
      "mean_token_accuracy": 0.17658320367336272,
      "num_tokens": 6059351.0,
      "step": 2940
    },
    {
      "entropy": 5.983553838729859,
      "epoch": 2.613404349755881,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004613019870083045,
      "loss": 5.7345,
      "mean_token_accuracy": 0.1750687301158905,
      "num_tokens": 6069639.0,
      "step": 2945
    },
    {
      "entropy": 6.009236431121826,
      "epoch": 2.6178428761651134,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046110861761426903,
      "loss": 5.7475,
      "mean_token_accuracy": 0.16875652074813843,
      "num_tokens": 6081393.0,
      "step": 2950
    },
    {
      "entropy": 6.063278436660767,
      "epoch": 2.6222814025743455,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046091481199435005,
      "loss": 5.7469,
      "mean_token_accuracy": 0.1720976859331131,
      "num_tokens": 6091043.0,
      "step": 2955
    },
    {
      "entropy": 6.051684427261352,
      "epoch": 2.6267199289835776,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004607205706028147,
      "loss": 5.733,
      "mean_token_accuracy": 0.1665470317006111,
      "num_tokens": 6101502.0,
      "step": 2960
    },
    {
      "entropy": 5.959883069992065,
      "epoch": 2.63115845539281,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004605258938949514,
      "loss": 5.6537,
      "mean_token_accuracy": 0.1800918236374855,
      "num_tokens": 6110908.0,
      "step": 2965
    },
    {
      "entropy": 6.070306491851807,
      "epoch": 2.635596981802042,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046033078232706923,
      "loss": 5.7551,
      "mean_token_accuracy": 0.17454806268215178,
      "num_tokens": 6120189.0,
      "step": 2970
    },
    {
      "entropy": 6.073927783966065,
      "epoch": 2.640035508211274,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046013523635649625,
      "loss": 5.7497,
      "mean_token_accuracy": 0.16859546005725862,
      "num_tokens": 6129095.0,
      "step": 2975
    },
    {
      "entropy": 6.0158473491668705,
      "epoch": 2.6444740346205062,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00045993925644157883,
      "loss": 5.7132,
      "mean_token_accuracy": 0.1755117155611515,
      "num_tokens": 6139129.0,
      "step": 2980
    },
    {
      "entropy": 5.979988050460816,
      "epoch": 2.6489125610297384,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004597428430416807,
      "loss": 5.7295,
      "mean_token_accuracy": 0.17120434045791627,
      "num_tokens": 6148966.0,
      "step": 2985
    },
    {
      "entropy": 6.136995887756347,
      "epoch": 2.6533510874389705,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045954599661718126,
      "loss": 5.8193,
      "mean_token_accuracy": 0.1676889553666115,
      "num_tokens": 6158254.0,
      "step": 2990
    },
    {
      "entropy": 6.045438051223755,
      "epoch": 2.6577896138482027,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00045934871762947504,
      "loss": 5.6964,
      "mean_token_accuracy": 0.17370334416627883,
      "num_tokens": 6167922.0,
      "step": 2995
    },
    {
      "entropy": 5.940775346755982,
      "epoch": 2.6622281402574344,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045915100654097076,
      "loss": 5.7092,
      "mean_token_accuracy": 0.17816627770662308,
      "num_tokens": 6179494.0,
      "step": 3000
    },
    {
      "epoch": 2.6622281402574344,
      "eval_entropy": 5.789337054001181,
      "eval_loss": 6.018830299377441,
      "eval_mean_token_accuracy": 0.16076800524480553,
      "eval_num_tokens": 6179494.0,
      "eval_runtime": 2.0923,
      "eval_samples_per_second": 1609.206,
      "eval_steps_per_second": 201.211,
      "step": 3000
    },
    {
      "entropy": 6.011846494674683,
      "epoch": 2.6666666666666665,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045895286381508944,
      "loss": 5.7312,
      "mean_token_accuracy": 0.1753581464290619,
      "num_tokens": 6189732.0,
      "step": 3005
    },
    {
      "entropy": 6.029772520065308,
      "epoch": 2.6711051930758987,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004587542899162642,
      "loss": 5.7683,
      "mean_token_accuracy": 0.16387217342853547,
      "num_tokens": 6200582.0,
      "step": 3010
    },
    {
      "entropy": 6.066657304763794,
      "epoch": 2.675543719485131,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045855528530993874,
      "loss": 5.7476,
      "mean_token_accuracy": 0.17216766774654388,
      "num_tokens": 6210009.0,
      "step": 3015
    },
    {
      "entropy": 6.039933919906616,
      "epoch": 2.679982245894363,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004583558504625661,
      "loss": 5.7558,
      "mean_token_accuracy": 0.1696187898516655,
      "num_tokens": 6220178.0,
      "step": 3020
    },
    {
      "entropy": 5.955683088302612,
      "epoch": 2.684420772303595,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045815598584160824,
      "loss": 5.6471,
      "mean_token_accuracy": 0.1844172552227974,
      "num_tokens": 6230147.0,
      "step": 3025
    },
    {
      "entropy": 6.097689056396485,
      "epoch": 2.6888592987128273,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00045795569191553384,
      "loss": 5.7985,
      "mean_token_accuracy": 0.16727579087018968,
      "num_tokens": 6241561.0,
      "step": 3030
    },
    {
      "entropy": 5.979752445220948,
      "epoch": 2.6932978251220594,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004577549691538183,
      "loss": 5.7252,
      "mean_token_accuracy": 0.18227578103542327,
      "num_tokens": 6251360.0,
      "step": 3035
    },
    {
      "entropy": 6.113001489639283,
      "epoch": 2.6977363515312915,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045755381802694206,
      "loss": 5.7266,
      "mean_token_accuracy": 0.1760123386979103,
      "num_tokens": 6261947.0,
      "step": 3040
    },
    {
      "entropy": 6.00572943687439,
      "epoch": 2.7021748779405237,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045735223900638967,
      "loss": 5.7118,
      "mean_token_accuracy": 0.16842464506626129,
      "num_tokens": 6272247.0,
      "step": 3045
    },
    {
      "entropy": 6.02584981918335,
      "epoch": 2.706613404349756,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045715023256464855,
      "loss": 5.7608,
      "mean_token_accuracy": 0.1715293511748314,
      "num_tokens": 6282171.0,
      "step": 3050
    },
    {
      "entropy": 6.107793283462525,
      "epoch": 2.711051930758988,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045694779917520797,
      "loss": 5.7451,
      "mean_token_accuracy": 0.17290424555540085,
      "num_tokens": 6290993.0,
      "step": 3055
    },
    {
      "entropy": 5.978008699417114,
      "epoch": 2.71549045716822,
      "grad_norm": 1.09375,
      "learning_rate": 0.000456744939312558,
      "loss": 5.6471,
      "mean_token_accuracy": 0.18025294095277786,
      "num_tokens": 6301039.0,
      "step": 3060
    },
    {
      "entropy": 5.9843508243560795,
      "epoch": 2.7199289835774523,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004565416534521883,
      "loss": 5.7301,
      "mean_token_accuracy": 0.16947097778320314,
      "num_tokens": 6312095.0,
      "step": 3065
    },
    {
      "entropy": 5.932244157791137,
      "epoch": 2.7243675099866844,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000456337942070587,
      "loss": 5.6472,
      "mean_token_accuracy": 0.18285827040672303,
      "num_tokens": 6322621.0,
      "step": 3070
    },
    {
      "entropy": 6.0621102809906,
      "epoch": 2.7288060363959166,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004561338056452396,
      "loss": 5.7749,
      "mean_token_accuracy": 0.17220864966511726,
      "num_tokens": 6332733.0,
      "step": 3075
    },
    {
      "entropy": 5.973612880706787,
      "epoch": 2.7332445628051487,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004559292446546281,
      "loss": 5.7143,
      "mean_token_accuracy": 0.17332967817783357,
      "num_tokens": 6343162.0,
      "step": 3080
    },
    {
      "entropy": 6.0895998001098635,
      "epoch": 2.737683089214381,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004557242595782293,
      "loss": 5.742,
      "mean_token_accuracy": 0.17037878483533858,
      "num_tokens": 6354283.0,
      "step": 3085
    },
    {
      "entropy": 6.039672994613648,
      "epoch": 2.742121615623613,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004555188508965144,
      "loss": 5.6795,
      "mean_token_accuracy": 0.17761039584875107,
      "num_tokens": 6364645.0,
      "step": 3090
    },
    {
      "entropy": 5.966477537155152,
      "epoch": 2.746560142032845,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045531301909094724,
      "loss": 5.6686,
      "mean_token_accuracy": 0.17619081139564513,
      "num_tokens": 6375156.0,
      "step": 3095
    },
    {
      "entropy": 6.012517929077148,
      "epoch": 2.7509986684420773,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004551067646439834,
      "loss": 5.6824,
      "mean_token_accuracy": 0.17580184042453767,
      "num_tokens": 6385105.0,
      "step": 3100
    },
    {
      "entropy": 6.031947231292724,
      "epoch": 2.7554371948513094,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045490008803906936,
      "loss": 5.7849,
      "mean_token_accuracy": 0.1666751965880394,
      "num_tokens": 6396122.0,
      "step": 3105
    },
    {
      "entropy": 6.084775066375732,
      "epoch": 2.7598757212605416,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004546929897606409,
      "loss": 5.6284,
      "mean_token_accuracy": 0.18060048222541808,
      "num_tokens": 6405668.0,
      "step": 3110
    },
    {
      "entropy": 5.962884092330933,
      "epoch": 2.7643142476697737,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045448547029412225,
      "loss": 5.7133,
      "mean_token_accuracy": 0.17631541192531586,
      "num_tokens": 6415845.0,
      "step": 3115
    },
    {
      "entropy": 5.991909742355347,
      "epoch": 2.768752774079006,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045427753012592477,
      "loss": 5.6724,
      "mean_token_accuracy": 0.1697130471467972,
      "num_tokens": 6424945.0,
      "step": 3120
    },
    {
      "entropy": 6.0469557762146,
      "epoch": 2.773191300488238,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045406916974344617,
      "loss": 5.7563,
      "mean_token_accuracy": 0.17011816203594207,
      "num_tokens": 6435224.0,
      "step": 3125
    },
    {
      "entropy": 6.046081399917602,
      "epoch": 2.77762982689747,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045386038963506883,
      "loss": 5.7334,
      "mean_token_accuracy": 0.1699560046195984,
      "num_tokens": 6444519.0,
      "step": 3130
    },
    {
      "entropy": 5.950239896774292,
      "epoch": 2.7820683533067023,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004536511902901591,
      "loss": 5.7693,
      "mean_token_accuracy": 0.17263632714748384,
      "num_tokens": 6454377.0,
      "step": 3135
    },
    {
      "entropy": 5.952496862411499,
      "epoch": 2.7865068797159345,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004534415721990659,
      "loss": 5.5402,
      "mean_token_accuracy": 0.1998279348015785,
      "num_tokens": 6464073.0,
      "step": 3140
    },
    {
      "entropy": 6.071445178985596,
      "epoch": 2.790945406125166,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045323153585311975,
      "loss": 5.7179,
      "mean_token_accuracy": 0.1706698343157768,
      "num_tokens": 6473517.0,
      "step": 3145
    },
    {
      "entropy": 6.029016780853271,
      "epoch": 2.7953839325343983,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004530210817446316,
      "loss": 5.6796,
      "mean_token_accuracy": 0.17350984811782838,
      "num_tokens": 6482465.0,
      "step": 3150
    },
    {
      "entropy": 6.018514966964721,
      "epoch": 2.7998224589436305,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004528102103668913,
      "loss": 5.7618,
      "mean_token_accuracy": 0.17125869393348694,
      "num_tokens": 6493420.0,
      "step": 3155
    },
    {
      "entropy": 6.028185033798218,
      "epoch": 2.8042609853528626,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004525989222141671,
      "loss": 5.719,
      "mean_token_accuracy": 0.1788898825645447,
      "num_tokens": 6503395.0,
      "step": 3160
    },
    {
      "entropy": 6.030786037445068,
      "epoch": 2.8086995117620948,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045238721778170386,
      "loss": 5.7436,
      "mean_token_accuracy": 0.17074056565761567,
      "num_tokens": 6514353.0,
      "step": 3165
    },
    {
      "entropy": 6.061694574356079,
      "epoch": 2.813138038171327,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045217509756572256,
      "loss": 5.6871,
      "mean_token_accuracy": 0.17090310156345367,
      "num_tokens": 6523501.0,
      "step": 3170
    },
    {
      "entropy": 5.97159743309021,
      "epoch": 2.817576564580559,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004519625620634182,
      "loss": 5.7318,
      "mean_token_accuracy": 0.16962333917617797,
      "num_tokens": 6535135.0,
      "step": 3175
    },
    {
      "entropy": 6.10450177192688,
      "epoch": 2.822015090989791,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00045174961177295964,
      "loss": 5.7209,
      "mean_token_accuracy": 0.17412717789411544,
      "num_tokens": 6546100.0,
      "step": 3180
    },
    {
      "entropy": 5.934656810760498,
      "epoch": 2.8264536173990233,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045153624719348773,
      "loss": 5.6484,
      "mean_token_accuracy": 0.17352755069732667,
      "num_tokens": 6557091.0,
      "step": 3185
    },
    {
      "entropy": 5.935671329498291,
      "epoch": 2.8308921438082555,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045132246882511457,
      "loss": 5.6754,
      "mean_token_accuracy": 0.17148020267486572,
      "num_tokens": 6567469.0,
      "step": 3190
    },
    {
      "entropy": 6.015770435333252,
      "epoch": 2.8353306702174876,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004511082771689219,
      "loss": 5.7007,
      "mean_token_accuracy": 0.17414820641279222,
      "num_tokens": 6577442.0,
      "step": 3195
    },
    {
      "entropy": 6.07279109954834,
      "epoch": 2.83976919662672,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045089367272696046,
      "loss": 5.6299,
      "mean_token_accuracy": 0.1816476598381996,
      "num_tokens": 6587321.0,
      "step": 3200
    },
    {
      "entropy": 6.001161098480225,
      "epoch": 2.844207723035952,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045067865600224833,
      "loss": 5.7098,
      "mean_token_accuracy": 0.17837129086256026,
      "num_tokens": 6597971.0,
      "step": 3205
    },
    {
      "entropy": 5.96258053779602,
      "epoch": 2.848646249445184,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045046322749877005,
      "loss": 5.6859,
      "mean_token_accuracy": 0.1785979002714157,
      "num_tokens": 6608251.0,
      "step": 3210
    },
    {
      "entropy": 6.015719032287597,
      "epoch": 2.853084775854416,
      "grad_norm": 0.96875,
      "learning_rate": 0.00045024738772147534,
      "loss": 5.6568,
      "mean_token_accuracy": 0.1777540847659111,
      "num_tokens": 6618045.0,
      "step": 3215
    },
    {
      "entropy": 6.079815149307251,
      "epoch": 2.8575233022636484,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004500311371762778,
      "loss": 5.723,
      "mean_token_accuracy": 0.1762673020362854,
      "num_tokens": 6628162.0,
      "step": 3220
    },
    {
      "entropy": 6.04972767829895,
      "epoch": 2.8619618286728805,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00044981447637005396,
      "loss": 5.7761,
      "mean_token_accuracy": 0.16095962822437287,
      "num_tokens": 6639225.0,
      "step": 3225
    },
    {
      "entropy": 5.983947801589966,
      "epoch": 2.8664003550821127,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000449597405810642,
      "loss": 5.6133,
      "mean_token_accuracy": 0.17978738248348236,
      "num_tokens": 6649248.0,
      "step": 3230
    },
    {
      "entropy": 5.941468191146851,
      "epoch": 2.870838881491345,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004493799260068405,
      "loss": 5.7351,
      "mean_token_accuracy": 0.17311612367630005,
      "num_tokens": 6660458.0,
      "step": 3235
    },
    {
      "entropy": 6.038877630233765,
      "epoch": 2.875277407900577,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004491620374684072,
      "loss": 5.6762,
      "mean_token_accuracy": 0.17014368772506713,
      "num_tokens": 6669951.0,
      "step": 3240
    },
    {
      "entropy": 6.017770910263062,
      "epoch": 2.879715934309809,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044894374070605795,
      "loss": 5.743,
      "mean_token_accuracy": 0.17674416303634644,
      "num_tokens": 6680472.0,
      "step": 3245
    },
    {
      "entropy": 6.084974384307861,
      "epoch": 2.8841544607190412,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044872503623146544,
      "loss": 5.7117,
      "mean_token_accuracy": 0.16619863510131835,
      "num_tokens": 6690939.0,
      "step": 3250
    },
    {
      "entropy": 5.94278769493103,
      "epoch": 2.8885929871282734,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044850592455725804,
      "loss": 5.7764,
      "mean_token_accuracy": 0.17108806669712068,
      "num_tokens": 6701087.0,
      "step": 3255
    },
    {
      "entropy": 6.017983531951904,
      "epoch": 2.8930315135375055,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004482864061970185,
      "loss": 5.6862,
      "mean_token_accuracy": 0.18754952698945998,
      "num_tokens": 6712292.0,
      "step": 3260
    },
    {
      "entropy": 6.036191987991333,
      "epoch": 2.8974700399467377,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044806648166528286,
      "loss": 5.659,
      "mean_token_accuracy": 0.17132395058870314,
      "num_tokens": 6722818.0,
      "step": 3265
    },
    {
      "entropy": 5.893555545806885,
      "epoch": 2.90190856635597,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044784615147753934,
      "loss": 5.6816,
      "mean_token_accuracy": 0.1757792666554451,
      "num_tokens": 6733303.0,
      "step": 3270
    },
    {
      "entropy": 5.996296119689942,
      "epoch": 2.906347092765202,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044762541615022664,
      "loss": 5.7015,
      "mean_token_accuracy": 0.17875064462423323,
      "num_tokens": 6742483.0,
      "step": 3275
    },
    {
      "entropy": 5.979998683929443,
      "epoch": 2.910785619174434,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044740427620073344,
      "loss": 5.6603,
      "mean_token_accuracy": 0.17813819795846939,
      "num_tokens": 6752836.0,
      "step": 3280
    },
    {
      "entropy": 5.954287004470825,
      "epoch": 2.9152241455836663,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044718273214739654,
      "loss": 5.7308,
      "mean_token_accuracy": 0.17084962874650955,
      "num_tokens": 6764321.0,
      "step": 3285
    },
    {
      "entropy": 6.067618417739868,
      "epoch": 2.9196626719928984,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004469607845095002,
      "loss": 5.6775,
      "mean_token_accuracy": 0.1735297128558159,
      "num_tokens": 6774034.0,
      "step": 3290
    },
    {
      "entropy": 5.988525152206421,
      "epoch": 2.9241011984021306,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004467384338072744,
      "loss": 5.7772,
      "mean_token_accuracy": 0.1673002153635025,
      "num_tokens": 6786182.0,
      "step": 3295
    },
    {
      "entropy": 6.02162880897522,
      "epoch": 2.9285397248113627,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004465156805618941,
      "loss": 5.6475,
      "mean_token_accuracy": 0.17642576694488527,
      "num_tokens": 6795631.0,
      "step": 3300
    },
    {
      "entropy": 5.964630270004273,
      "epoch": 2.932978251220595,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004462925252954775,
      "loss": 5.7392,
      "mean_token_accuracy": 0.16975196450948715,
      "num_tokens": 6805212.0,
      "step": 3305
    },
    {
      "entropy": 6.031642150878906,
      "epoch": 2.937416777629827,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004460689685310855,
      "loss": 5.7476,
      "mean_token_accuracy": 0.1703239858150482,
      "num_tokens": 6815680.0,
      "step": 3310
    },
    {
      "entropy": 6.008225917816162,
      "epoch": 2.941855304039059,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004458450107927199,
      "loss": 5.6538,
      "mean_token_accuracy": 0.180711767077446,
      "num_tokens": 6824694.0,
      "step": 3315
    },
    {
      "entropy": 5.845587253570557,
      "epoch": 2.9462938304482913,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044562065260532214,
      "loss": 5.621,
      "mean_token_accuracy": 0.17849345952272416,
      "num_tokens": 6835219.0,
      "step": 3320
    },
    {
      "entropy": 6.04475736618042,
      "epoch": 2.9507323568575234,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044539589449477265,
      "loss": 5.7213,
      "mean_token_accuracy": 0.1738622322678566,
      "num_tokens": 6845992.0,
      "step": 3325
    },
    {
      "entropy": 6.052994966506958,
      "epoch": 2.9551708832667556,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000445170736987889,
      "loss": 5.6658,
      "mean_token_accuracy": 0.18287943601608275,
      "num_tokens": 6855287.0,
      "step": 3330
    },
    {
      "entropy": 6.029227447509766,
      "epoch": 2.9596094096759877,
      "grad_norm": 1.046875,
      "learning_rate": 0.000444945180612425,
      "loss": 5.775,
      "mean_token_accuracy": 0.17114163041114808,
      "num_tokens": 6865841.0,
      "step": 3335
    },
    {
      "entropy": 6.034116411209107,
      "epoch": 2.96404793608522,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044471922589706944,
      "loss": 5.8053,
      "mean_token_accuracy": 0.16967750638723372,
      "num_tokens": 6875931.0,
      "step": 3340
    },
    {
      "entropy": 6.023122358322143,
      "epoch": 2.968486462494452,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004444928733714446,
      "loss": 5.7363,
      "mean_token_accuracy": 0.17056947499513625,
      "num_tokens": 6886092.0,
      "step": 3345
    },
    {
      "entropy": 6.0595800399780275,
      "epoch": 2.972924988903684,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044426612356610555,
      "loss": 5.7344,
      "mean_token_accuracy": 0.16704397201538085,
      "num_tokens": 6897312.0,
      "step": 3350
    },
    {
      "entropy": 5.955773544311524,
      "epoch": 2.9773635153129163,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004440389770125382,
      "loss": 5.6457,
      "mean_token_accuracy": 0.18342512249946594,
      "num_tokens": 6907838.0,
      "step": 3355
    },
    {
      "entropy": 5.930391263961792,
      "epoch": 2.9818020417221485,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004438114342431586,
      "loss": 5.6485,
      "mean_token_accuracy": 0.17502613514661788,
      "num_tokens": 6916729.0,
      "step": 3360
    },
    {
      "entropy": 5.983162069320679,
      "epoch": 2.9862405681313806,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044358349579131143,
      "loss": 5.5849,
      "mean_token_accuracy": 0.1869617834687233,
      "num_tokens": 6927074.0,
      "step": 3365
    },
    {
      "entropy": 5.962836027145386,
      "epoch": 2.9906790945406128,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044335516219126876,
      "loss": 5.6287,
      "mean_token_accuracy": 0.17767293900251388,
      "num_tokens": 6938437.0,
      "step": 3370
    },
    {
      "entropy": 5.964249944686889,
      "epoch": 2.995117620949845,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044312643397822907,
      "loss": 5.6745,
      "mean_token_accuracy": 0.18153784573078155,
      "num_tokens": 6949036.0,
      "step": 3375
    },
    {
      "entropy": 6.030031394958496,
      "epoch": 2.999556147359077,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044289731168831566,
      "loss": 5.5911,
      "mean_token_accuracy": 0.18498829305171965,
      "num_tokens": 6958459.0,
      "step": 3380
    },
    {
      "entropy": 5.8808572557237415,
      "epoch": 3.003550821127386,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004426677958585755,
      "loss": 5.4157,
      "mean_token_accuracy": 0.19034291472699907,
      "num_tokens": 6967264.0,
      "step": 3385
    },
    {
      "entropy": 5.903432178497314,
      "epoch": 3.007989347536618,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00044243788702697797,
      "loss": 5.3709,
      "mean_token_accuracy": 0.19530602395534516,
      "num_tokens": 6977465.0,
      "step": 3390
    },
    {
      "entropy": 5.998833322525025,
      "epoch": 3.01242787394585,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004422075857324137,
      "loss": 5.3929,
      "mean_token_accuracy": 0.19290464073419572,
      "num_tokens": 6987583.0,
      "step": 3395
    },
    {
      "entropy": 5.954317235946656,
      "epoch": 3.0168664003550822,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044197689251469324,
      "loss": 5.38,
      "mean_token_accuracy": 0.1891101285815239,
      "num_tokens": 6997894.0,
      "step": 3400
    },
    {
      "entropy": 5.9602972030639645,
      "epoch": 3.0213049267643144,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044174580791454555,
      "loss": 5.3515,
      "mean_token_accuracy": 0.19597329050302506,
      "num_tokens": 7008019.0,
      "step": 3405
    },
    {
      "entropy": 5.9017480373382565,
      "epoch": 3.0257434531735465,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004415143324736174,
      "loss": 5.3916,
      "mean_token_accuracy": 0.19494948536157608,
      "num_tokens": 7018272.0,
      "step": 3410
    },
    {
      "entropy": 5.968024396896363,
      "epoch": 3.0301819795827787,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004412824667344712,
      "loss": 5.439,
      "mean_token_accuracy": 0.1884671375155449,
      "num_tokens": 7028185.0,
      "step": 3415
    },
    {
      "entropy": 6.047925806045532,
      "epoch": 3.034620505992011,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044105021124058455,
      "loss": 5.4481,
      "mean_token_accuracy": 0.18440557271242142,
      "num_tokens": 7038123.0,
      "step": 3420
    },
    {
      "entropy": 5.86896710395813,
      "epoch": 3.039059032401243,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044081756653634825,
      "loss": 5.342,
      "mean_token_accuracy": 0.1948542922735214,
      "num_tokens": 7048379.0,
      "step": 3425
    },
    {
      "entropy": 5.908081436157227,
      "epoch": 3.043497558810475,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004405845331670659,
      "loss": 5.323,
      "mean_token_accuracy": 0.20013415217399597,
      "num_tokens": 7057878.0,
      "step": 3430
    },
    {
      "entropy": 5.90395884513855,
      "epoch": 3.0479360852197073,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004403511116789514,
      "loss": 5.4602,
      "mean_token_accuracy": 0.18673462122678758,
      "num_tokens": 7067998.0,
      "step": 3435
    },
    {
      "entropy": 5.890441370010376,
      "epoch": 3.052374611628939,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044011730261912915,
      "loss": 5.3764,
      "mean_token_accuracy": 0.19622484892606734,
      "num_tokens": 7079236.0,
      "step": 3440
    },
    {
      "entropy": 5.89832239151001,
      "epoch": 3.056813138038171,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004398831065356314,
      "loss": 5.3428,
      "mean_token_accuracy": 0.19831295162439347,
      "num_tokens": 7088947.0,
      "step": 3445
    },
    {
      "entropy": 5.946053075790405,
      "epoch": 3.0612516644474033,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043964852397739793,
      "loss": 5.4557,
      "mean_token_accuracy": 0.18364953398704528,
      "num_tokens": 7099480.0,
      "step": 3450
    },
    {
      "entropy": 5.869796800613403,
      "epoch": 3.0656901908566354,
      "grad_norm": 1.3125,
      "learning_rate": 0.000439413555494274,
      "loss": 5.3686,
      "mean_token_accuracy": 0.1966594323515892,
      "num_tokens": 7109471.0,
      "step": 3455
    },
    {
      "entropy": 5.870002937316895,
      "epoch": 3.0701287172658676,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004391782016370098,
      "loss": 5.4148,
      "mean_token_accuracy": 0.1896324947476387,
      "num_tokens": 7119790.0,
      "step": 3460
    },
    {
      "entropy": 5.988854026794433,
      "epoch": 3.0745672436750997,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004389424629572586,
      "loss": 5.3878,
      "mean_token_accuracy": 0.19112442284822465,
      "num_tokens": 7129736.0,
      "step": 3465
    },
    {
      "entropy": 5.899620294570923,
      "epoch": 3.079005770084332,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043870634000757605,
      "loss": 5.3816,
      "mean_token_accuracy": 0.18703049421310425,
      "num_tokens": 7139786.0,
      "step": 3470
    },
    {
      "entropy": 5.943547010421753,
      "epoch": 3.083444296493564,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004384698333414179,
      "loss": 5.387,
      "mean_token_accuracy": 0.1897724226117134,
      "num_tokens": 7150742.0,
      "step": 3475
    },
    {
      "entropy": 5.958523893356324,
      "epoch": 3.087882822902796,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004382329435131396,
      "loss": 5.4457,
      "mean_token_accuracy": 0.18277957141399384,
      "num_tokens": 7161740.0,
      "step": 3480
    },
    {
      "entropy": 5.954104804992676,
      "epoch": 3.0923213493120283,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00043799567107799504,
      "loss": 5.423,
      "mean_token_accuracy": 0.18841141164302827,
      "num_tokens": 7172615.0,
      "step": 3485
    },
    {
      "entropy": 5.788148641586304,
      "epoch": 3.0967598757212604,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043775801659213425,
      "loss": 5.3323,
      "mean_token_accuracy": 0.20167517215013503,
      "num_tokens": 7183037.0,
      "step": 3490
    },
    {
      "entropy": 5.911949014663696,
      "epoch": 3.1011984021304926,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004375199806126034,
      "loss": 5.4289,
      "mean_token_accuracy": 0.19386280030012132,
      "num_tokens": 7193621.0,
      "step": 3495
    },
    {
      "entropy": 5.8650144100189205,
      "epoch": 3.1056369285397247,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043728156369734236,
      "loss": 5.313,
      "mean_token_accuracy": 0.20414462238550185,
      "num_tokens": 7204419.0,
      "step": 3500
    },
    {
      "epoch": 3.1056369285397247,
      "eval_entropy": 5.6460988130818635,
      "eval_loss": 5.959334373474121,
      "eval_mean_token_accuracy": 0.16493839260872073,
      "eval_num_tokens": 7204419.0,
      "eval_runtime": 2.0846,
      "eval_samples_per_second": 1615.155,
      "eval_steps_per_second": 201.954,
      "step": 3500
    },
    {
      "entropy": 5.834872102737426,
      "epoch": 3.110075454948957,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004370427664051841,
      "loss": 5.3529,
      "mean_token_accuracy": 0.19269932806491852,
      "num_tokens": 7214372.0,
      "step": 3505
    },
    {
      "entropy": 5.863707590103149,
      "epoch": 3.114513981358189,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004368035892958533,
      "loss": 5.3994,
      "mean_token_accuracy": 0.19355032294988633,
      "num_tokens": 7224990.0,
      "step": 3510
    },
    {
      "entropy": 5.937055206298828,
      "epoch": 3.118952507767421,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043656403292996454,
      "loss": 5.4193,
      "mean_token_accuracy": 0.1840432032942772,
      "num_tokens": 7235487.0,
      "step": 3515
    },
    {
      "entropy": 5.885832786560059,
      "epoch": 3.1233910341766533,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004363240978690218,
      "loss": 5.4026,
      "mean_token_accuracy": 0.19053976088762284,
      "num_tokens": 7245942.0,
      "step": 3520
    },
    {
      "entropy": 5.951666831970215,
      "epoch": 3.1278295605858855,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043608378467541626,
      "loss": 5.4028,
      "mean_token_accuracy": 0.18414484113454818,
      "num_tokens": 7255323.0,
      "step": 3525
    },
    {
      "entropy": 5.941895866394043,
      "epoch": 3.1322680869951176,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043584309391242584,
      "loss": 5.4639,
      "mean_token_accuracy": 0.1837260752916336,
      "num_tokens": 7266339.0,
      "step": 3530
    },
    {
      "entropy": 5.955119895935058,
      "epoch": 3.1367066134043498,
      "grad_norm": 1.125,
      "learning_rate": 0.0004356020261442131,
      "loss": 5.4029,
      "mean_token_accuracy": 0.1911745175719261,
      "num_tokens": 7277501.0,
      "step": 3535
    },
    {
      "entropy": 5.9398706436157225,
      "epoch": 3.141145139813582,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043536058193582443,
      "loss": 5.455,
      "mean_token_accuracy": 0.1892775923013687,
      "num_tokens": 7287639.0,
      "step": 3540
    },
    {
      "entropy": 5.849388504028321,
      "epoch": 3.145583666222814,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004351187618531886,
      "loss": 5.3415,
      "mean_token_accuracy": 0.19347691684961318,
      "num_tokens": 7297333.0,
      "step": 3545
    },
    {
      "entropy": 5.835960865020752,
      "epoch": 3.150022192632046,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043487656646311535,
      "loss": 5.4584,
      "mean_token_accuracy": 0.18551218509674072,
      "num_tokens": 7307750.0,
      "step": 3550
    },
    {
      "entropy": 5.860294580459595,
      "epoch": 3.1544607190412783,
      "grad_norm": 1.125,
      "learning_rate": 0.0004346339963332941,
      "loss": 5.3436,
      "mean_token_accuracy": 0.20199633687734603,
      "num_tokens": 7317128.0,
      "step": 3555
    },
    {
      "entropy": 5.871670961380005,
      "epoch": 3.1588992454505105,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004343910520322927,
      "loss": 5.4169,
      "mean_token_accuracy": 0.1937512531876564,
      "num_tokens": 7327574.0,
      "step": 3560
    },
    {
      "entropy": 5.894437789916992,
      "epoch": 3.1633377718597426,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000434147734129556,
      "loss": 5.4207,
      "mean_token_accuracy": 0.192408949136734,
      "num_tokens": 7337776.0,
      "step": 3565
    },
    {
      "entropy": 5.906270456314087,
      "epoch": 3.1677762982689748,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043390404319540443,
      "loss": 5.4257,
      "mean_token_accuracy": 0.19216728061437607,
      "num_tokens": 7347691.0,
      "step": 3570
    },
    {
      "entropy": 5.894905376434326,
      "epoch": 3.172214824678207,
      "grad_norm": 1.125,
      "learning_rate": 0.00043365997980103304,
      "loss": 5.4454,
      "mean_token_accuracy": 0.18370102643966674,
      "num_tokens": 7358152.0,
      "step": 3575
    },
    {
      "entropy": 5.903047180175781,
      "epoch": 3.176653351087439,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043341554451850964,
      "loss": 5.3993,
      "mean_token_accuracy": 0.19171175360679626,
      "num_tokens": 7369017.0,
      "step": 3580
    },
    {
      "entropy": 5.9143860816955565,
      "epoch": 3.181091877496671,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043317073792077394,
      "loss": 5.4409,
      "mean_token_accuracy": 0.19202805012464524,
      "num_tokens": 7378589.0,
      "step": 3585
    },
    {
      "entropy": 5.907209157943726,
      "epoch": 3.1855304039059034,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004329255605816357,
      "loss": 5.3684,
      "mean_token_accuracy": 0.19171188175678253,
      "num_tokens": 7388053.0,
      "step": 3590
    },
    {
      "entropy": 5.889437580108643,
      "epoch": 3.1899689303151355,
      "grad_norm": 1.078125,
      "learning_rate": 0.000432680013075774,
      "loss": 5.4492,
      "mean_token_accuracy": 0.18419877737760543,
      "num_tokens": 7398200.0,
      "step": 3595
    },
    {
      "entropy": 5.893877220153809,
      "epoch": 3.1944074567243677,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004324340959787354,
      "loss": 5.3605,
      "mean_token_accuracy": 0.20110263228416442,
      "num_tokens": 7408378.0,
      "step": 3600
    },
    {
      "entropy": 5.872954082489014,
      "epoch": 3.1988459831336,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043218780986693274,
      "loss": 5.4328,
      "mean_token_accuracy": 0.1992975264787674,
      "num_tokens": 7418961.0,
      "step": 3605
    },
    {
      "entropy": 5.846698236465454,
      "epoch": 3.203284509542832,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004319411553176438,
      "loss": 5.5195,
      "mean_token_accuracy": 0.18346895426511764,
      "num_tokens": 7429844.0,
      "step": 3610
    },
    {
      "entropy": 5.879422616958618,
      "epoch": 3.207723035952064,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004316941329090101,
      "loss": 5.3498,
      "mean_token_accuracy": 0.20174553096294404,
      "num_tokens": 7440824.0,
      "step": 3615
    },
    {
      "entropy": 5.9176287174224855,
      "epoch": 3.2121615623612962,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043144674322003525,
      "loss": 5.4666,
      "mean_token_accuracy": 0.1915660098195076,
      "num_tokens": 7451666.0,
      "step": 3620
    },
    {
      "entropy": 5.904341077804565,
      "epoch": 3.2166000887705284,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004311989868305837,
      "loss": 5.4375,
      "mean_token_accuracy": 0.1913430780172348,
      "num_tokens": 7462476.0,
      "step": 3625
    },
    {
      "entropy": 5.900774383544922,
      "epoch": 3.2210386151797605,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043095086432137954,
      "loss": 5.4534,
      "mean_token_accuracy": 0.18185227811336518,
      "num_tokens": 7472388.0,
      "step": 3630
    },
    {
      "entropy": 5.887851858139038,
      "epoch": 3.2254771415889927,
      "grad_norm": 1.125,
      "learning_rate": 0.000430702376274005,
      "loss": 5.401,
      "mean_token_accuracy": 0.19130902141332626,
      "num_tokens": 7482770.0,
      "step": 3635
    },
    {
      "entropy": 5.906753969192505,
      "epoch": 3.2299156679982244,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043045352327089907,
      "loss": 5.468,
      "mean_token_accuracy": 0.1859576016664505,
      "num_tokens": 7492107.0,
      "step": 3640
    },
    {
      "entropy": 5.898250150680542,
      "epoch": 3.2343541944074565,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043020430589535625,
      "loss": 5.3594,
      "mean_token_accuracy": 0.19287962913513185,
      "num_tokens": 7501597.0,
      "step": 3645
    },
    {
      "entropy": 5.905869817733764,
      "epoch": 3.2387927208166887,
      "grad_norm": 1.109375,
      "learning_rate": 0.000429954724731525,
      "loss": 5.4058,
      "mean_token_accuracy": 0.19971251338720322,
      "num_tokens": 7513200.0,
      "step": 3650
    },
    {
      "entropy": 5.91200532913208,
      "epoch": 3.243231247225921,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004297047803644063,
      "loss": 5.3459,
      "mean_token_accuracy": 0.19347363710403442,
      "num_tokens": 7522807.0,
      "step": 3655
    },
    {
      "entropy": 5.8334047317504885,
      "epoch": 3.247669773635153,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000429454473379853,
      "loss": 5.3477,
      "mean_token_accuracy": 0.1945561319589615,
      "num_tokens": 7533577.0,
      "step": 3660
    },
    {
      "entropy": 5.883318090438843,
      "epoch": 3.252108300044385,
      "grad_norm": 1.125,
      "learning_rate": 0.0004292038043645675,
      "loss": 5.4106,
      "mean_token_accuracy": 0.18551634699106218,
      "num_tokens": 7543640.0,
      "step": 3665
    },
    {
      "entropy": 5.794455099105835,
      "epoch": 3.2565468264536173,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00042895277390610074,
      "loss": 5.4428,
      "mean_token_accuracy": 0.19491562098264695,
      "num_tokens": 7554692.0,
      "step": 3670
    },
    {
      "entropy": 5.990721559524536,
      "epoch": 3.2609853528628494,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004287013825928509,
      "loss": 5.385,
      "mean_token_accuracy": 0.18947930932044982,
      "num_tokens": 7563925.0,
      "step": 3675
    },
    {
      "entropy": 5.733651828765869,
      "epoch": 3.2654238792720816,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004284496310140622,
      "loss": 5.2984,
      "mean_token_accuracy": 0.20197273343801497,
      "num_tokens": 7573484.0,
      "step": 3680
    },
    {
      "entropy": 5.799650478363037,
      "epoch": 3.2698624056813137,
      "grad_norm": 1.125,
      "learning_rate": 0.0004281975197598231,
      "loss": 5.3955,
      "mean_token_accuracy": 0.19549006968736649,
      "num_tokens": 7582808.0,
      "step": 3685
    },
    {
      "entropy": 5.88456883430481,
      "epoch": 3.274300932090546,
      "grad_norm": 1.125,
      "learning_rate": 0.0004279450494210651,
      "loss": 5.4827,
      "mean_token_accuracy": 0.18646293729543686,
      "num_tokens": 7593790.0,
      "step": 3690
    },
    {
      "entropy": 5.911808729171753,
      "epoch": 3.278739458499778,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004276922205895614,
      "loss": 5.48,
      "mean_token_accuracy": 0.1881803423166275,
      "num_tokens": 7603756.0,
      "step": 3695
    },
    {
      "entropy": 5.939221954345703,
      "epoch": 3.28317798490901,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004274390338579257,
      "loss": 5.4207,
      "mean_token_accuracy": 0.19052503407001495,
      "num_tokens": 7614126.0,
      "step": 3700
    },
    {
      "entropy": 5.749071359634399,
      "epoch": 3.2876165113182423,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004271854898196102,
      "loss": 5.3746,
      "mean_token_accuracy": 0.19580249339342118,
      "num_tokens": 7624573.0,
      "step": 3705
    },
    {
      "entropy": 5.878935194015503,
      "epoch": 3.2920550377274744,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004269315890689049,
      "loss": 5.4431,
      "mean_token_accuracy": 0.1863309696316719,
      "num_tokens": 7634715.0,
      "step": 3710
    },
    {
      "entropy": 5.853589487075806,
      "epoch": 3.2964935641367066,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042667733220093574,
      "loss": 5.3239,
      "mean_token_accuracy": 0.20386914610862733,
      "num_tokens": 7644241.0,
      "step": 3715
    },
    {
      "entropy": 5.874784135818482,
      "epoch": 3.3009320905459387,
      "grad_norm": 1.125,
      "learning_rate": 0.0004264227198116635,
      "loss": 5.4795,
      "mean_token_accuracy": 0.18705491870641708,
      "num_tokens": 7655655.0,
      "step": 3720
    },
    {
      "entropy": 5.848001718521118,
      "epoch": 3.305370616955171,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042616775249788223,
      "loss": 5.4973,
      "mean_token_accuracy": 0.18835359066724777,
      "num_tokens": 7665232.0,
      "step": 3725
    },
    {
      "entropy": 5.913939046859741,
      "epoch": 3.309809143364403,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004259124308572178,
      "loss": 5.4316,
      "mean_token_accuracy": 0.1916329488158226,
      "num_tokens": 7675178.0,
      "step": 3730
    },
    {
      "entropy": 5.899500942230224,
      "epoch": 3.314247669773635,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004256567554881268,
      "loss": 5.3627,
      "mean_token_accuracy": 0.1938821107149124,
      "num_tokens": 7685835.0,
      "step": 3735
    },
    {
      "entropy": 5.912274503707886,
      "epoch": 3.3186861961828673,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004254007269898948,
      "loss": 5.3577,
      "mean_token_accuracy": 0.1939299672842026,
      "num_tokens": 7694828.0,
      "step": 3740
    },
    {
      "entropy": 5.82884202003479,
      "epoch": 3.3231247225920995,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042514434596263513,
      "loss": 5.4271,
      "mean_token_accuracy": 0.1924677923321724,
      "num_tokens": 7706055.0,
      "step": 3745
    },
    {
      "entropy": 5.830545425415039,
      "epoch": 3.3275632490013316,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004248876130072873,
      "loss": 5.4581,
      "mean_token_accuracy": 0.1842326655983925,
      "num_tokens": 7716220.0,
      "step": 3750
    },
    {
      "entropy": 5.860052013397217,
      "epoch": 3.3320017754105637,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00042463052872561587,
      "loss": 5.4372,
      "mean_token_accuracy": 0.19057605415582657,
      "num_tokens": 7726709.0,
      "step": 3755
    },
    {
      "entropy": 5.963085699081421,
      "epoch": 3.336440301819796,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042437309372020886,
      "loss": 5.4012,
      "mean_token_accuracy": 0.1977565497159958,
      "num_tokens": 7736507.0,
      "step": 3760
    },
    {
      "entropy": 5.872720527648926,
      "epoch": 3.340878828229028,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042411530859447634,
      "loss": 5.4912,
      "mean_token_accuracy": 0.1859763205051422,
      "num_tokens": 7747031.0,
      "step": 3765
    },
    {
      "entropy": 5.914382410049439,
      "epoch": 3.34531735463826,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004238571739526489,
      "loss": 5.5078,
      "mean_token_accuracy": 0.18674080073833466,
      "num_tokens": 7756896.0,
      "step": 3770
    },
    {
      "entropy": 5.829860639572144,
      "epoch": 3.3497558810474923,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004235986903997767,
      "loss": 5.3098,
      "mean_token_accuracy": 0.1985625222325325,
      "num_tokens": 7768243.0,
      "step": 3775
    },
    {
      "entropy": 5.909512329101562,
      "epoch": 3.3541944074567245,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004233398585417275,
      "loss": 5.3886,
      "mean_token_accuracy": 0.1947034165263176,
      "num_tokens": 7778017.0,
      "step": 3780
    },
    {
      "entropy": 5.905759811401367,
      "epoch": 3.3586329338659566,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004230806789851854,
      "loss": 5.4766,
      "mean_token_accuracy": 0.18501238375902176,
      "num_tokens": 7788648.0,
      "step": 3785
    },
    {
      "entropy": 5.870203638076783,
      "epoch": 3.3630714602751888,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042282115233764953,
      "loss": 5.4228,
      "mean_token_accuracy": 0.1922216847538948,
      "num_tokens": 7799444.0,
      "step": 3790
    },
    {
      "entropy": 5.915548610687256,
      "epoch": 3.367509986684421,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004225612792074326,
      "loss": 5.4095,
      "mean_token_accuracy": 0.18869893848896027,
      "num_tokens": 7809324.0,
      "step": 3795
    },
    {
      "entropy": 5.847271156311035,
      "epoch": 3.371948513093653,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042230106020365964,
      "loss": 5.4219,
      "mean_token_accuracy": 0.18740808665752412,
      "num_tokens": 7819835.0,
      "step": 3800
    },
    {
      "entropy": 5.823662281036377,
      "epoch": 3.376387039502885,
      "grad_norm": 1.125,
      "learning_rate": 0.00042204049593626617,
      "loss": 5.3148,
      "mean_token_accuracy": 0.19989162534475327,
      "num_tokens": 7829978.0,
      "step": 3805
    },
    {
      "entropy": 5.8751567840576175,
      "epoch": 3.3808255659121174,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042177958701599696,
      "loss": 5.466,
      "mean_token_accuracy": 0.1890934631228447,
      "num_tokens": 7840638.0,
      "step": 3810
    },
    {
      "entropy": 5.777163314819336,
      "epoch": 3.3852640923213495,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004215183340544047,
      "loss": 5.2881,
      "mean_token_accuracy": 0.20688982158899308,
      "num_tokens": 7852025.0,
      "step": 3815
    },
    {
      "entropy": 5.833953237533569,
      "epoch": 3.389702618730581,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004212567376638485,
      "loss": 5.3775,
      "mean_token_accuracy": 0.19608416706323623,
      "num_tokens": 7861446.0,
      "step": 3820
    },
    {
      "entropy": 5.823704862594605,
      "epoch": 3.3941411451398134,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042099479845749256,
      "loss": 5.4895,
      "mean_token_accuracy": 0.18594682961702347,
      "num_tokens": 7871521.0,
      "step": 3825
    },
    {
      "entropy": 5.859103012084961,
      "epoch": 3.3985796715490455,
      "grad_norm": 1.125,
      "learning_rate": 0.00042073251704930414,
      "loss": 5.4215,
      "mean_token_accuracy": 0.19555966407060624,
      "num_tokens": 7882352.0,
      "step": 3830
    },
    {
      "entropy": 5.895108985900879,
      "epoch": 3.4030181979582776,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042046989405405326,
      "loss": 5.3998,
      "mean_token_accuracy": 0.19245795011520386,
      "num_tokens": 7892487.0,
      "step": 3835
    },
    {
      "entropy": 5.786870241165161,
      "epoch": 3.40745672436751,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004202069300873102,
      "loss": 5.3629,
      "mean_token_accuracy": 0.19259919673204423,
      "num_tokens": 7902262.0,
      "step": 3840
    },
    {
      "entropy": 5.887244415283203,
      "epoch": 3.411895250776742,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004199436257654445,
      "loss": 5.3403,
      "mean_token_accuracy": 0.19992637187242507,
      "num_tokens": 7913415.0,
      "step": 3845
    },
    {
      "entropy": 5.800942850112915,
      "epoch": 3.416333777185974,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004196799817056234,
      "loss": 5.3702,
      "mean_token_accuracy": 0.1944957345724106,
      "num_tokens": 7923378.0,
      "step": 3850
    },
    {
      "entropy": 5.833743000030518,
      "epoch": 3.4207723035952062,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041941599852581067,
      "loss": 5.4005,
      "mean_token_accuracy": 0.192107592523098,
      "num_tokens": 7932155.0,
      "step": 3855
    },
    {
      "entropy": 5.866676568984985,
      "epoch": 3.4252108300044384,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041915167684476495,
      "loss": 5.4391,
      "mean_token_accuracy": 0.1913757935166359,
      "num_tokens": 7942236.0,
      "step": 3860
    },
    {
      "entropy": 5.916124677658081,
      "epoch": 3.4296493564136705,
      "grad_norm": 1.171875,
      "learning_rate": 0.000418887017282038,
      "loss": 5.4685,
      "mean_token_accuracy": 0.19052421748638154,
      "num_tokens": 7952067.0,
      "step": 3865
    },
    {
      "entropy": 5.809151649475098,
      "epoch": 3.4340878828229027,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041862202045797386,
      "loss": 5.3092,
      "mean_token_accuracy": 0.19900714755058288,
      "num_tokens": 7961456.0,
      "step": 3870
    },
    {
      "entropy": 5.823811960220337,
      "epoch": 3.438526409232135,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004183566869937069,
      "loss": 5.3782,
      "mean_token_accuracy": 0.20037556439638138,
      "num_tokens": 7971601.0,
      "step": 3875
    },
    {
      "entropy": 5.87151517868042,
      "epoch": 3.442964935641367,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004180910175111608,
      "loss": 5.4483,
      "mean_token_accuracy": 0.18243330270051955,
      "num_tokens": 7982732.0,
      "step": 3880
    },
    {
      "entropy": 5.812599134445191,
      "epoch": 3.447403462050599,
      "grad_norm": 1.34375,
      "learning_rate": 0.00041782501263304655,
      "loss": 5.4182,
      "mean_token_accuracy": 0.18896383941173553,
      "num_tokens": 7993550.0,
      "step": 3885
    },
    {
      "entropy": 5.85041913986206,
      "epoch": 3.4518419884598313,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004175586729828614,
      "loss": 5.3866,
      "mean_token_accuracy": 0.1983974426984787,
      "num_tokens": 8003753.0,
      "step": 3890
    },
    {
      "entropy": 5.848913478851318,
      "epoch": 3.4562805148690634,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041729199918488725,
      "loss": 5.367,
      "mean_token_accuracy": 0.1954442948102951,
      "num_tokens": 8014516.0,
      "step": 3895
    },
    {
      "entropy": 5.776366281509399,
      "epoch": 3.4607190412782955,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041702499186418936,
      "loss": 5.4013,
      "mean_token_accuracy": 0.191869093477726,
      "num_tokens": 8024734.0,
      "step": 3900
    },
    {
      "entropy": 5.825012636184693,
      "epoch": 3.4651575676875277,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041675765164661473,
      "loss": 5.2844,
      "mean_token_accuracy": 0.20244546234607697,
      "num_tokens": 8033681.0,
      "step": 3905
    },
    {
      "entropy": 5.812342643737793,
      "epoch": 3.46959609409676,
      "grad_norm": 1.25,
      "learning_rate": 0.0004164899791587903,
      "loss": 5.4335,
      "mean_token_accuracy": 0.18668857365846633,
      "num_tokens": 8043202.0,
      "step": 3910
    },
    {
      "entropy": 5.874458456039429,
      "epoch": 3.474034620505992,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041622197502812224,
      "loss": 5.4737,
      "mean_token_accuracy": 0.18350510895252228,
      "num_tokens": 8052964.0,
      "step": 3915
    },
    {
      "entropy": 5.868040704727173,
      "epoch": 3.478473146915224,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000415953639882794,
      "loss": 5.4358,
      "mean_token_accuracy": 0.191118323802948,
      "num_tokens": 8063964.0,
      "step": 3920
    },
    {
      "entropy": 5.817435216903687,
      "epoch": 3.4829116733244563,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041568497435176473,
      "loss": 5.3394,
      "mean_token_accuracy": 0.1965772658586502,
      "num_tokens": 8073659.0,
      "step": 3925
    },
    {
      "entropy": 5.775618934631348,
      "epoch": 3.4873501997336884,
      "grad_norm": 1.25,
      "learning_rate": 0.0004154159790647681,
      "loss": 5.4096,
      "mean_token_accuracy": 0.1935155361890793,
      "num_tokens": 8083850.0,
      "step": 3930
    },
    {
      "entropy": 5.857674884796142,
      "epoch": 3.4917887261429206,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041514665465231063,
      "loss": 5.3836,
      "mean_token_accuracy": 0.19599000364542007,
      "num_tokens": 8094308.0,
      "step": 3935
    },
    {
      "entropy": 5.884661388397217,
      "epoch": 3.4962272525521527,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00041487700174567036,
      "loss": 5.4342,
      "mean_token_accuracy": 0.1878661945462227,
      "num_tokens": 8105383.0,
      "step": 3940
    },
    {
      "entropy": 5.894474458694458,
      "epoch": 3.500665778961385,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041460702097689535,
      "loss": 5.4853,
      "mean_token_accuracy": 0.1829699918627739,
      "num_tokens": 8116888.0,
      "step": 3945
    },
    {
      "entropy": 5.815571308135986,
      "epoch": 3.505104305370617,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041433671297880204,
      "loss": 5.3644,
      "mean_token_accuracy": 0.2002154141664505,
      "num_tokens": 8126419.0,
      "step": 3950
    },
    {
      "entropy": 5.856853008270264,
      "epoch": 3.509542831779849,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004140660783849739,
      "loss": 5.4585,
      "mean_token_accuracy": 0.18545352667570114,
      "num_tokens": 8136734.0,
      "step": 3955
    },
    {
      "entropy": 5.7611549377441404,
      "epoch": 3.5139813581890813,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041379511782975995,
      "loss": 5.4275,
      "mean_token_accuracy": 0.18803369849920273,
      "num_tokens": 8146362.0,
      "step": 3960
    },
    {
      "entropy": 5.867103481292725,
      "epoch": 3.5184198845983135,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004135238319482731,
      "loss": 5.3396,
      "mean_token_accuracy": 0.19985446333885193,
      "num_tokens": 8156493.0,
      "step": 3965
    },
    {
      "entropy": 5.827406597137451,
      "epoch": 3.5228584110075456,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041325222137638914,
      "loss": 5.3822,
      "mean_token_accuracy": 0.198279732465744,
      "num_tokens": 8166693.0,
      "step": 3970
    },
    {
      "entropy": 5.8414324760437015,
      "epoch": 3.5272969374167777,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004129802867507444,
      "loss": 5.5056,
      "mean_token_accuracy": 0.18893716484308243,
      "num_tokens": 8177819.0,
      "step": 3975
    },
    {
      "entropy": 5.8395007133483885,
      "epoch": 3.53173546382601,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004127080287087354,
      "loss": 5.4563,
      "mean_token_accuracy": 0.19080454111099243,
      "num_tokens": 8187559.0,
      "step": 3980
    },
    {
      "entropy": 5.878630113601685,
      "epoch": 3.536173990235242,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041243544788851616,
      "loss": 5.4661,
      "mean_token_accuracy": 0.18955991715192794,
      "num_tokens": 8198051.0,
      "step": 3985
    },
    {
      "entropy": 5.788226985931397,
      "epoch": 3.540612516644474,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041216254492899753,
      "loss": 5.2939,
      "mean_token_accuracy": 0.20163479596376419,
      "num_tokens": 8207969.0,
      "step": 3990
    },
    {
      "entropy": 5.814974308013916,
      "epoch": 3.5450510430537063,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004118893204698454,
      "loss": 5.3902,
      "mean_token_accuracy": 0.18923236578702926,
      "num_tokens": 8219057.0,
      "step": 3995
    },
    {
      "entropy": 5.7541250705719,
      "epoch": 3.5494895694629385,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004116157751514793,
      "loss": 5.3578,
      "mean_token_accuracy": 0.19907362014055252,
      "num_tokens": 8228156.0,
      "step": 4000
    },
    {
      "epoch": 3.5494895694629385,
      "eval_entropy": 5.581153839047901,
      "eval_loss": 5.913293838500977,
      "eval_mean_token_accuracy": 0.16851918041847663,
      "eval_num_tokens": 8228156.0,
      "eval_runtime": 2.0884,
      "eval_samples_per_second": 1612.224,
      "eval_steps_per_second": 201.588,
      "step": 4000
    },
    {
      "entropy": 5.824972915649414,
      "epoch": 3.5539280958721706,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041134190961507073,
      "loss": 5.4659,
      "mean_token_accuracy": 0.19036727249622346,
      "num_tokens": 8238707.0,
      "step": 4005
    },
    {
      "entropy": 5.870899343490601,
      "epoch": 3.5583666222814028,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041106772450254177,
      "loss": 5.4352,
      "mean_token_accuracy": 0.19054232090711593,
      "num_tokens": 8248026.0,
      "step": 4010
    },
    {
      "entropy": 5.803697156906128,
      "epoch": 3.562805148690635,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041079322045656366,
      "loss": 5.3887,
      "mean_token_accuracy": 0.2025946080684662,
      "num_tokens": 8258615.0,
      "step": 4015
    },
    {
      "entropy": 5.885553455352783,
      "epoch": 3.567243675099867,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000410518398120555,
      "loss": 5.4536,
      "mean_token_accuracy": 0.1920897349715233,
      "num_tokens": 8268912.0,
      "step": 4020
    },
    {
      "entropy": 5.87697319984436,
      "epoch": 3.571682201509099,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041024325813868065,
      "loss": 5.4202,
      "mean_token_accuracy": 0.19221669733524321,
      "num_tokens": 8278503.0,
      "step": 4025
    },
    {
      "entropy": 5.786037254333496,
      "epoch": 3.5761207279183314,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040996780115584995,
      "loss": 5.4195,
      "mean_token_accuracy": 0.1956888884305954,
      "num_tokens": 8289725.0,
      "step": 4030
    },
    {
      "entropy": 5.849604320526123,
      "epoch": 3.5805592543275635,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004096920278177153,
      "loss": 5.4054,
      "mean_token_accuracy": 0.19833215177059174,
      "num_tokens": 8300391.0,
      "step": 4035
    },
    {
      "entropy": 5.870030736923217,
      "epoch": 3.5849977807367956,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004094159387706703,
      "loss": 5.4031,
      "mean_token_accuracy": 0.19163014888763427,
      "num_tokens": 8310656.0,
      "step": 4040
    },
    {
      "entropy": 5.793809986114502,
      "epoch": 3.589436307146028,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004091395346618491,
      "loss": 5.3533,
      "mean_token_accuracy": 0.2000315383076668,
      "num_tokens": 8320605.0,
      "step": 4045
    },
    {
      "entropy": 5.88753571510315,
      "epoch": 3.5938748335552595,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040886281613912396,
      "loss": 5.4909,
      "mean_token_accuracy": 0.18160921484231948,
      "num_tokens": 8331683.0,
      "step": 4050
    },
    {
      "entropy": 5.859517431259155,
      "epoch": 3.5983133599644916,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004085857838511042,
      "loss": 5.494,
      "mean_token_accuracy": 0.18673501014709473,
      "num_tokens": 8342132.0,
      "step": 4055
    },
    {
      "entropy": 5.813091182708741,
      "epoch": 3.602751886373724,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040830843844713447,
      "loss": 5.4091,
      "mean_token_accuracy": 0.1943575844168663,
      "num_tokens": 8352202.0,
      "step": 4060
    },
    {
      "entropy": 5.824704217910766,
      "epoch": 3.607190412782956,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040803078057729354,
      "loss": 5.4583,
      "mean_token_accuracy": 0.18807061463594438,
      "num_tokens": 8362204.0,
      "step": 4065
    },
    {
      "entropy": 5.858654499053955,
      "epoch": 3.611628939192188,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004077528108923924,
      "loss": 5.4523,
      "mean_token_accuracy": 0.19497880935668946,
      "num_tokens": 8372525.0,
      "step": 4070
    },
    {
      "entropy": 5.830624771118164,
      "epoch": 3.6160674656014202,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004074745300439732,
      "loss": 5.4152,
      "mean_token_accuracy": 0.1913478270173073,
      "num_tokens": 8383007.0,
      "step": 4075
    },
    {
      "entropy": 5.777306318283081,
      "epoch": 3.6205059920106524,
      "grad_norm": 1.078125,
      "learning_rate": 0.000407195938684307,
      "loss": 5.3941,
      "mean_token_accuracy": 0.19022576361894608,
      "num_tokens": 8392928.0,
      "step": 4080
    },
    {
      "entropy": 5.886033535003662,
      "epoch": 3.6249445184198845,
      "grad_norm": 1.078125,
      "learning_rate": 0.000406917037466393,
      "loss": 5.4263,
      "mean_token_accuracy": 0.19012691229581832,
      "num_tokens": 8403008.0,
      "step": 4085
    },
    {
      "entropy": 5.789344787597656,
      "epoch": 3.6293830448291167,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004066378270439567,
      "loss": 5.4782,
      "mean_token_accuracy": 0.17886057496070862,
      "num_tokens": 8413513.0,
      "step": 4090
    },
    {
      "entropy": 5.841081666946411,
      "epoch": 3.633821571238349,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004063583080714481,
      "loss": 5.3738,
      "mean_token_accuracy": 0.1967576861381531,
      "num_tokens": 8423894.0,
      "step": 4095
    },
    {
      "entropy": 5.8004848003387455,
      "epoch": 3.638260097647581,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040607848120404063,
      "loss": 5.413,
      "mean_token_accuracy": 0.19520466923713684,
      "num_tokens": 8434467.0,
      "step": 4100
    },
    {
      "entropy": 5.803511476516723,
      "epoch": 3.642698624056813,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004057983470976293,
      "loss": 5.357,
      "mean_token_accuracy": 0.2016320988535881,
      "num_tokens": 8446037.0,
      "step": 4105
    },
    {
      "entropy": 5.77181224822998,
      "epoch": 3.6471371504660453,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040551790640882954,
      "loss": 5.4094,
      "mean_token_accuracy": 0.20021369606256484,
      "num_tokens": 8457032.0,
      "step": 4110
    },
    {
      "entropy": 5.854924535751342,
      "epoch": 3.6515756768752774,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040523715979497486,
      "loss": 5.4118,
      "mean_token_accuracy": 0.19359598606824874,
      "num_tokens": 8466088.0,
      "step": 4115
    },
    {
      "entropy": 5.865506505966186,
      "epoch": 3.6560142032845095,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004049561079141163,
      "loss": 5.4117,
      "mean_token_accuracy": 0.19178108721971512,
      "num_tokens": 8477214.0,
      "step": 4120
    },
    {
      "entropy": 5.798425579071045,
      "epoch": 3.6604527296937417,
      "grad_norm": 1.125,
      "learning_rate": 0.0004046747514250202,
      "loss": 5.4725,
      "mean_token_accuracy": 0.19099220335483552,
      "num_tokens": 8488031.0,
      "step": 4125
    },
    {
      "entropy": 5.95139889717102,
      "epoch": 3.664891256102974,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004043930909871671,
      "loss": 5.5991,
      "mean_token_accuracy": 0.18338969349861145,
      "num_tokens": 8500214.0,
      "step": 4130
    },
    {
      "entropy": 5.890401697158813,
      "epoch": 3.669329782512206,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040411112726074954,
      "loss": 5.439,
      "mean_token_accuracy": 0.19255967885255815,
      "num_tokens": 8509768.0,
      "step": 4135
    },
    {
      "entropy": 5.854125690460205,
      "epoch": 3.673768308921438,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040382886090667154,
      "loss": 5.4488,
      "mean_token_accuracy": 0.19439015686511993,
      "num_tokens": 8519967.0,
      "step": 4140
    },
    {
      "entropy": 5.899671220779419,
      "epoch": 3.6782068353306703,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004035462925865459,
      "loss": 5.4697,
      "mean_token_accuracy": 0.18602120280265808,
      "num_tokens": 8531624.0,
      "step": 4145
    },
    {
      "entropy": 5.917951536178589,
      "epoch": 3.6826453617399024,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040326342296269363,
      "loss": 5.4044,
      "mean_token_accuracy": 0.19721968472003937,
      "num_tokens": 8542130.0,
      "step": 4150
    },
    {
      "entropy": 5.776456356048584,
      "epoch": 3.6870838881491346,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040298025269814165,
      "loss": 5.4315,
      "mean_token_accuracy": 0.19611175805330278,
      "num_tokens": 8552155.0,
      "step": 4155
    },
    {
      "entropy": 5.807200622558594,
      "epoch": 3.6915224145583667,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004026967824566218,
      "loss": 5.3751,
      "mean_token_accuracy": 0.19862941205501555,
      "num_tokens": 8561602.0,
      "step": 4160
    },
    {
      "entropy": 5.829988574981689,
      "epoch": 3.695960940967599,
      "grad_norm": 1.15625,
      "learning_rate": 0.000402413012902569,
      "loss": 5.3549,
      "mean_token_accuracy": 0.1946346491575241,
      "num_tokens": 8571957.0,
      "step": 4165
    },
    {
      "entropy": 5.757293558120727,
      "epoch": 3.700399467376831,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040212894470111966,
      "loss": 5.3879,
      "mean_token_accuracy": 0.19661612063646317,
      "num_tokens": 8582550.0,
      "step": 4170
    },
    {
      "entropy": 5.827823162078857,
      "epoch": 3.704837993786063,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004018445785181102,
      "loss": 5.422,
      "mean_token_accuracy": 0.19695309102535247,
      "num_tokens": 8592353.0,
      "step": 4175
    },
    {
      "entropy": 5.832696485519409,
      "epoch": 3.7092765201952953,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004015599150200755,
      "loss": 5.4116,
      "mean_token_accuracy": 0.1886480987071991,
      "num_tokens": 8602629.0,
      "step": 4180
    },
    {
      "entropy": 5.811392307281494,
      "epoch": 3.7137150466045274,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040127495487424735,
      "loss": 5.4699,
      "mean_token_accuracy": 0.19141082167625428,
      "num_tokens": 8612308.0,
      "step": 4185
    },
    {
      "entropy": 5.834705829620361,
      "epoch": 3.7181535730137596,
      "grad_norm": 1.125,
      "learning_rate": 0.0004009896987485531,
      "loss": 5.366,
      "mean_token_accuracy": 0.20177424997091292,
      "num_tokens": 8622177.0,
      "step": 4190
    },
    {
      "entropy": 5.809908723831176,
      "epoch": 3.7225920994229913,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040070414731161326,
      "loss": 5.443,
      "mean_token_accuracy": 0.19458999037742614,
      "num_tokens": 8632435.0,
      "step": 4195
    },
    {
      "entropy": 5.747843647003174,
      "epoch": 3.7270306258322234,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040041830123274105,
      "loss": 5.2288,
      "mean_token_accuracy": 0.20880044400691986,
      "num_tokens": 8642331.0,
      "step": 4200
    },
    {
      "entropy": 5.903952407836914,
      "epoch": 3.7314691522414556,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004001321611819401,
      "loss": 5.3903,
      "mean_token_accuracy": 0.18388656824827193,
      "num_tokens": 8652838.0,
      "step": 4205
    },
    {
      "entropy": 5.80471625328064,
      "epoch": 3.7359076786506877,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003998457278299033,
      "loss": 5.4137,
      "mean_token_accuracy": 0.19091238379478453,
      "num_tokens": 8662908.0,
      "step": 4210
    },
    {
      "entropy": 5.801855611801147,
      "epoch": 3.74034620505992,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003995590018480107,
      "loss": 5.4813,
      "mean_token_accuracy": 0.18728871196508406,
      "num_tokens": 8673015.0,
      "step": 4215
    },
    {
      "entropy": 5.828760576248169,
      "epoch": 3.744784731469152,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039927198390832843,
      "loss": 5.3716,
      "mean_token_accuracy": 0.20105525702238083,
      "num_tokens": 8682149.0,
      "step": 4220
    },
    {
      "entropy": 5.803042364120484,
      "epoch": 3.749223257878384,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000398984674683607,
      "loss": 5.3658,
      "mean_token_accuracy": 0.19573144614696503,
      "num_tokens": 8692715.0,
      "step": 4225
    },
    {
      "entropy": 5.784009027481079,
      "epoch": 3.7536617842876163,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003986970748472795,
      "loss": 5.4067,
      "mean_token_accuracy": 0.20018333494663237,
      "num_tokens": 8702379.0,
      "step": 4230
    },
    {
      "entropy": 5.8258195400238035,
      "epoch": 3.7581003106968485,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003984091850734604,
      "loss": 5.3853,
      "mean_token_accuracy": 0.19552432000637054,
      "num_tokens": 8712653.0,
      "step": 4235
    },
    {
      "entropy": 5.8420146942138675,
      "epoch": 3.7625388371060806,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003981210060369435,
      "loss": 5.4715,
      "mean_token_accuracy": 0.18399242907762528,
      "num_tokens": 8724054.0,
      "step": 4240
    },
    {
      "entropy": 5.857115030288696,
      "epoch": 3.7669773635153128,
      "grad_norm": 1.078125,
      "learning_rate": 0.000397832538413201,
      "loss": 5.5001,
      "mean_token_accuracy": 0.18325462341308593,
      "num_tokens": 8734629.0,
      "step": 4245
    },
    {
      "entropy": 5.845765590667725,
      "epoch": 3.771415889924545,
      "grad_norm": 1.0,
      "learning_rate": 0.0003975437828783811,
      "loss": 5.4329,
      "mean_token_accuracy": 0.1934712439775467,
      "num_tokens": 8746198.0,
      "step": 4250
    },
    {
      "entropy": 5.730560302734375,
      "epoch": 3.775854416333777,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039725474010930716,
      "loss": 5.3907,
      "mean_token_accuracy": 0.20444272756576537,
      "num_tokens": 8756604.0,
      "step": 4255
    },
    {
      "entropy": 5.827708339691162,
      "epoch": 3.780292942743009,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003969654107834756,
      "loss": 5.4509,
      "mean_token_accuracy": 0.1945435181260109,
      "num_tokens": 8768088.0,
      "step": 4260
    },
    {
      "entropy": 5.82703127861023,
      "epoch": 3.7847314691522413,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003966757955790547,
      "loss": 5.3874,
      "mean_token_accuracy": 0.19351006299257278,
      "num_tokens": 8778318.0,
      "step": 4265
    },
    {
      "entropy": 5.848085165023804,
      "epoch": 3.7891699955614735,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003963858951748826,
      "loss": 5.4017,
      "mean_token_accuracy": 0.1907930001616478,
      "num_tokens": 8788659.0,
      "step": 4270
    },
    {
      "entropy": 5.796998500823975,
      "epoch": 3.7936085219707056,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003960957102504661,
      "loss": 5.4293,
      "mean_token_accuracy": 0.18963894993066788,
      "num_tokens": 8799052.0,
      "step": 4275
    },
    {
      "entropy": 5.907032918930054,
      "epoch": 3.798047048379938,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003958052414859788,
      "loss": 5.4359,
      "mean_token_accuracy": 0.1835399091243744,
      "num_tokens": 8808534.0,
      "step": 4280
    },
    {
      "entropy": 5.759076499938965,
      "epoch": 3.80248557478917,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003955144895622597,
      "loss": 5.3888,
      "mean_token_accuracy": 0.1968761757016182,
      "num_tokens": 8818802.0,
      "step": 4285
    },
    {
      "entropy": 5.813445806503296,
      "epoch": 3.806924101198402,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003952234551608114,
      "loss": 5.3661,
      "mean_token_accuracy": 0.18817414045333863,
      "num_tokens": 8829123.0,
      "step": 4290
    },
    {
      "entropy": 5.800790023803711,
      "epoch": 3.8113626276076342,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003949321389637986,
      "loss": 5.4292,
      "mean_token_accuracy": 0.19044290333986283,
      "num_tokens": 8839771.0,
      "step": 4295
    },
    {
      "entropy": 5.825014877319336,
      "epoch": 3.8158011540168664,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003946405416540466,
      "loss": 5.4491,
      "mean_token_accuracy": 0.18945735543966294,
      "num_tokens": 8849621.0,
      "step": 4300
    },
    {
      "entropy": 5.940211343765259,
      "epoch": 3.8202396804260985,
      "grad_norm": 1.125,
      "learning_rate": 0.00039434866391503963,
      "loss": 5.4527,
      "mean_token_accuracy": 0.1803130567073822,
      "num_tokens": 8860672.0,
      "step": 4305
    },
    {
      "entropy": 5.789125680923462,
      "epoch": 3.8246782068353307,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039405650643091917,
      "loss": 5.3998,
      "mean_token_accuracy": 0.19494441598653794,
      "num_tokens": 8870721.0,
      "step": 4310
    },
    {
      "entropy": 5.757489824295044,
      "epoch": 3.829116733244563,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003937640698864823,
      "loss": 5.4515,
      "mean_token_accuracy": 0.19027765691280366,
      "num_tokens": 8882142.0,
      "step": 4315
    },
    {
      "entropy": 5.861147117614746,
      "epoch": 3.833555259653795,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039347135496718027,
      "loss": 5.4136,
      "mean_token_accuracy": 0.1945791110396385,
      "num_tokens": 8892573.0,
      "step": 4320
    },
    {
      "entropy": 5.772451877593994,
      "epoch": 3.837993786063027,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039317836235911705,
      "loss": 5.4017,
      "mean_token_accuracy": 0.195092111825943,
      "num_tokens": 8902942.0,
      "step": 4325
    },
    {
      "entropy": 5.806214046478272,
      "epoch": 3.8424323124722592,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003928850927490472,
      "loss": 5.4216,
      "mean_token_accuracy": 0.18946156948804854,
      "num_tokens": 8913208.0,
      "step": 4330
    },
    {
      "entropy": 5.805463075637817,
      "epoch": 3.8468708388814914,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003925915468243746,
      "loss": 5.4276,
      "mean_token_accuracy": 0.1883111536502838,
      "num_tokens": 8923080.0,
      "step": 4335
    },
    {
      "entropy": 5.89415135383606,
      "epoch": 3.8513093652907235,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039229772527315073,
      "loss": 5.4224,
      "mean_token_accuracy": 0.18533381819725037,
      "num_tokens": 8933823.0,
      "step": 4340
    },
    {
      "entropy": 5.795059251785278,
      "epoch": 3.8557478916999557,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003920036287840734,
      "loss": 5.3395,
      "mean_token_accuracy": 0.19785182178020477,
      "num_tokens": 8943412.0,
      "step": 4345
    },
    {
      "entropy": 5.793986701965332,
      "epoch": 3.860186418109188,
      "grad_norm": 1.125,
      "learning_rate": 0.00039170925804648486,
      "loss": 5.4513,
      "mean_token_accuracy": 0.19680401533842087,
      "num_tokens": 8953616.0,
      "step": 4350
    },
    {
      "entropy": 5.841963052749634,
      "epoch": 3.86462494451842,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039141461375036957,
      "loss": 5.4467,
      "mean_token_accuracy": 0.18167138546705247,
      "num_tokens": 8964790.0,
      "step": 4355
    },
    {
      "entropy": 5.803447532653808,
      "epoch": 3.869063470927652,
      "grad_norm": 1.125,
      "learning_rate": 0.0003911196965863539,
      "loss": 5.3662,
      "mean_token_accuracy": 0.1910330817103386,
      "num_tokens": 8975476.0,
      "step": 4360
    },
    {
      "entropy": 5.852850914001465,
      "epoch": 3.8735019973368843,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039082450724570357,
      "loss": 5.4888,
      "mean_token_accuracy": 0.18827887177467345,
      "num_tokens": 8986132.0,
      "step": 4365
    },
    {
      "entropy": 5.80005111694336,
      "epoch": 3.8779405237461164,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003905290464203221,
      "loss": 5.4303,
      "mean_token_accuracy": 0.19234416782855987,
      "num_tokens": 8996384.0,
      "step": 4370
    },
    {
      "entropy": 5.816908454895019,
      "epoch": 3.8823790501553486,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003902333148027495,
      "loss": 5.391,
      "mean_token_accuracy": 0.20051574110984802,
      "num_tokens": 9006288.0,
      "step": 4375
    },
    {
      "entropy": 5.857726764678955,
      "epoch": 3.8868175765645807,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003899373130861605,
      "loss": 5.4102,
      "mean_token_accuracy": 0.19304264485836028,
      "num_tokens": 9016713.0,
      "step": 4380
    },
    {
      "entropy": 5.917016649246216,
      "epoch": 3.891256102973813,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038964104196436284,
      "loss": 5.3835,
      "mean_token_accuracy": 0.19150962233543395,
      "num_tokens": 9026524.0,
      "step": 4385
    },
    {
      "entropy": 5.777474451065063,
      "epoch": 3.895694629383045,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038934450213179596,
      "loss": 5.4493,
      "mean_token_accuracy": 0.1875316545367241,
      "num_tokens": 9037063.0,
      "step": 4390
    },
    {
      "entropy": 5.824544095993042,
      "epoch": 3.900133155792277,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038904769428352873,
      "loss": 5.4017,
      "mean_token_accuracy": 0.1963329166173935,
      "num_tokens": 9047685.0,
      "step": 4395
    },
    {
      "entropy": 5.849619722366333,
      "epoch": 3.9045716822015093,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038875061911525856,
      "loss": 5.3666,
      "mean_token_accuracy": 0.1921242356300354,
      "num_tokens": 9058084.0,
      "step": 4400
    },
    {
      "entropy": 5.744299793243409,
      "epoch": 3.9090102086107414,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003884532773233094,
      "loss": 5.3635,
      "mean_token_accuracy": 0.20264142900705337,
      "num_tokens": 9068384.0,
      "step": 4405
    },
    {
      "entropy": 5.833960247039795,
      "epoch": 3.9134487350199736,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038815566960463015,
      "loss": 5.4482,
      "mean_token_accuracy": 0.18840712010860444,
      "num_tokens": 9079222.0,
      "step": 4410
    },
    {
      "entropy": 5.850077724456787,
      "epoch": 3.9178872614292057,
      "grad_norm": 1.125,
      "learning_rate": 0.00038785779665679275,
      "loss": 5.4202,
      "mean_token_accuracy": 0.18839995115995406,
      "num_tokens": 9089526.0,
      "step": 4415
    },
    {
      "entropy": 5.751740980148315,
      "epoch": 3.922325787838438,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003875596591779913,
      "loss": 5.3186,
      "mean_token_accuracy": 0.2056492105126381,
      "num_tokens": 9099054.0,
      "step": 4420
    },
    {
      "entropy": 5.793323278427124,
      "epoch": 3.92676431424767,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003872612578670395,
      "loss": 5.4754,
      "mean_token_accuracy": 0.18841795325279237,
      "num_tokens": 9109290.0,
      "step": 4425
    },
    {
      "entropy": 5.885016775131225,
      "epoch": 3.931202840656902,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038696259342336966,
      "loss": 5.499,
      "mean_token_accuracy": 0.18841444998979567,
      "num_tokens": 9120796.0,
      "step": 4430
    },
    {
      "entropy": 5.861675357818603,
      "epoch": 3.935641367066134,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038666366654703077,
      "loss": 5.4846,
      "mean_token_accuracy": 0.1862151712179184,
      "num_tokens": 9131306.0,
      "step": 4435
    },
    {
      "entropy": 5.840978145599365,
      "epoch": 3.940079893475366,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038636447793868715,
      "loss": 5.4734,
      "mean_token_accuracy": 0.18990519195795058,
      "num_tokens": 9140278.0,
      "step": 4440
    },
    {
      "entropy": 5.851602220535279,
      "epoch": 3.944518419884598,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038606502829961645,
      "loss": 5.4267,
      "mean_token_accuracy": 0.19005220383405685,
      "num_tokens": 9150742.0,
      "step": 4445
    },
    {
      "entropy": 5.7608050346374515,
      "epoch": 3.9489569462938303,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003857653183317081,
      "loss": 5.3303,
      "mean_token_accuracy": 0.19899591207504272,
      "num_tokens": 9160227.0,
      "step": 4450
    },
    {
      "entropy": 5.791858434677124,
      "epoch": 3.9533954727030625,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003854653487374617,
      "loss": 5.4306,
      "mean_token_accuracy": 0.19147539585828782,
      "num_tokens": 9170733.0,
      "step": 4455
    },
    {
      "entropy": 5.8443633079528805,
      "epoch": 3.9578339991122946,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003851651202199856,
      "loss": 5.413,
      "mean_token_accuracy": 0.19143973886966706,
      "num_tokens": 9180582.0,
      "step": 4460
    },
    {
      "entropy": 5.821952390670776,
      "epoch": 3.9622725255215268,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003848646334829949,
      "loss": 5.4525,
      "mean_token_accuracy": 0.1855350062251091,
      "num_tokens": 9191245.0,
      "step": 4465
    },
    {
      "entropy": 5.840106630325318,
      "epoch": 3.966711051930759,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038456388923081006,
      "loss": 5.4459,
      "mean_token_accuracy": 0.18592941164970397,
      "num_tokens": 9201657.0,
      "step": 4470
    },
    {
      "entropy": 5.830752325057984,
      "epoch": 3.971149578339991,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038426288816835485,
      "loss": 5.4103,
      "mean_token_accuracy": 0.18891609460115433,
      "num_tokens": 9212102.0,
      "step": 4475
    },
    {
      "entropy": 5.847698926925659,
      "epoch": 3.975588104749223,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003839616310011554,
      "loss": 5.3934,
      "mean_token_accuracy": 0.18960051685571672,
      "num_tokens": 9221022.0,
      "step": 4480
    },
    {
      "entropy": 5.806353616714477,
      "epoch": 3.9800266311584553,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003836601184353379,
      "loss": 5.4564,
      "mean_token_accuracy": 0.18732759356498718,
      "num_tokens": 9232105.0,
      "step": 4485
    },
    {
      "entropy": 5.804685926437378,
      "epoch": 3.9844651575676875,
      "grad_norm": 0.90625,
      "learning_rate": 0.00038335835117762706,
      "loss": 5.3984,
      "mean_token_accuracy": 0.1938449651002884,
      "num_tokens": 9242338.0,
      "step": 4490
    },
    {
      "entropy": 5.864504861831665,
      "epoch": 3.9889036839769196,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038305632993534483,
      "loss": 5.4276,
      "mean_token_accuracy": 0.1886606141924858,
      "num_tokens": 9252551.0,
      "step": 4495
    },
    {
      "entropy": 5.843533515930176,
      "epoch": 3.993342210386152,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038275405541640835,
      "loss": 5.403,
      "mean_token_accuracy": 0.19167436361312867,
      "num_tokens": 9263424.0,
      "step": 4500
    },
    {
      "epoch": 3.993342210386152,
      "eval_entropy": 5.580333083372501,
      "eval_loss": 5.84522008895874,
      "eval_mean_token_accuracy": 0.17348938688388063,
      "eval_num_tokens": 9263424.0,
      "eval_runtime": 2.2779,
      "eval_samples_per_second": 1478.121,
      "eval_steps_per_second": 184.82,
      "step": 4500
    },
    {
      "entropy": 5.849137592315674,
      "epoch": 3.997780736795384,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038245152832932843,
      "loss": 5.4953,
      "mean_token_accuracy": 0.1837543785572052,
      "num_tokens": 9274355.0,
      "step": 4505
    },
    {
      "entropy": 5.780615064832899,
      "epoch": 4.001775410563693,
      "grad_norm": 0.984375,
      "learning_rate": 0.00038214874938320795,
      "loss": 5.2503,
      "mean_token_accuracy": 0.205965217616823,
      "num_tokens": 9283179.0,
      "step": 4510
    },
    {
      "entropy": 5.817928743362427,
      "epoch": 4.006213936972925,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003818457192877399,
      "loss": 5.1978,
      "mean_token_accuracy": 0.208414126932621,
      "num_tokens": 9293709.0,
      "step": 4515
    },
    {
      "entropy": 5.824892950057984,
      "epoch": 4.010652463382157,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003815424387532063,
      "loss": 5.2238,
      "mean_token_accuracy": 0.21198973655700684,
      "num_tokens": 9303971.0,
      "step": 4520
    },
    {
      "entropy": 5.810022640228271,
      "epoch": 4.015090989791389,
      "grad_norm": 1.109375,
      "learning_rate": 0.000381238908490476,
      "loss": 5.1717,
      "mean_token_accuracy": 0.208732508122921,
      "num_tokens": 9313578.0,
      "step": 4525
    },
    {
      "entropy": 5.752050828933716,
      "epoch": 4.019529516200621,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038093512921100306,
      "loss": 5.1907,
      "mean_token_accuracy": 0.2054088294506073,
      "num_tokens": 9323538.0,
      "step": 4530
    },
    {
      "entropy": 5.838475942611694,
      "epoch": 4.023968042609853,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003806311016268254,
      "loss": 5.1886,
      "mean_token_accuracy": 0.20790643393993377,
      "num_tokens": 9335266.0,
      "step": 4535
    },
    {
      "entropy": 5.79393162727356,
      "epoch": 4.028406569019086,
      "grad_norm": 1.15625,
      "learning_rate": 0.000380326826450563,
      "loss": 5.1481,
      "mean_token_accuracy": 0.20686888098716735,
      "num_tokens": 9345247.0,
      "step": 4540
    },
    {
      "entropy": 5.778472805023194,
      "epoch": 4.032845095428318,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038002230439541603,
      "loss": 5.1838,
      "mean_token_accuracy": 0.20136410593986512,
      "num_tokens": 9355435.0,
      "step": 4545
    },
    {
      "entropy": 5.719447374343872,
      "epoch": 4.03728362183755,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037971753617516336,
      "loss": 5.109,
      "mean_token_accuracy": 0.2196057543158531,
      "num_tokens": 9365503.0,
      "step": 4550
    },
    {
      "entropy": 5.798331022262573,
      "epoch": 4.041722148246782,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037941252250416074,
      "loss": 5.0692,
      "mean_token_accuracy": 0.21513248682022096,
      "num_tokens": 9374361.0,
      "step": 4555
    },
    {
      "entropy": 5.722120237350464,
      "epoch": 4.046160674656014,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037910726409733965,
      "loss": 5.0786,
      "mean_token_accuracy": 0.22050419300794602,
      "num_tokens": 9383992.0,
      "step": 4560
    },
    {
      "entropy": 5.756671810150147,
      "epoch": 4.050599201065246,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003788017616702048,
      "loss": 5.2069,
      "mean_token_accuracy": 0.20192554146051406,
      "num_tokens": 9394270.0,
      "step": 4565
    },
    {
      "entropy": 5.837602424621582,
      "epoch": 4.055037727474478,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037849601593883297,
      "loss": 5.1378,
      "mean_token_accuracy": 0.20951651483774186,
      "num_tokens": 9404913.0,
      "step": 4570
    },
    {
      "entropy": 5.830431222915649,
      "epoch": 4.059476253883711,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037819002761987127,
      "loss": 5.2469,
      "mean_token_accuracy": 0.20075388699769975,
      "num_tokens": 9415718.0,
      "step": 4575
    },
    {
      "entropy": 5.777271699905396,
      "epoch": 4.063914780292943,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003778837974305355,
      "loss": 5.1694,
      "mean_token_accuracy": 0.20675463527441024,
      "num_tokens": 9425761.0,
      "step": 4580
    },
    {
      "entropy": 5.696642255783081,
      "epoch": 4.068353306702175,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037757732608860824,
      "loss": 5.1757,
      "mean_token_accuracy": 0.2074501171708107,
      "num_tokens": 9437519.0,
      "step": 4585
    },
    {
      "entropy": 5.773201513290405,
      "epoch": 4.072791833111407,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037727061431243737,
      "loss": 5.0117,
      "mean_token_accuracy": 0.22399133294820786,
      "num_tokens": 9447452.0,
      "step": 4590
    },
    {
      "entropy": 5.802689504623413,
      "epoch": 4.077230359520639,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037696366282093433,
      "loss": 5.2194,
      "mean_token_accuracy": 0.20754450112581252,
      "num_tokens": 9458155.0,
      "step": 4595
    },
    {
      "entropy": 5.665275192260742,
      "epoch": 4.081668885929871,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037665647233357243,
      "loss": 5.035,
      "mean_token_accuracy": 0.22337092012166976,
      "num_tokens": 9467478.0,
      "step": 4600
    },
    {
      "entropy": 5.747597599029541,
      "epoch": 4.0861074123391035,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003763490435703853,
      "loss": 5.1954,
      "mean_token_accuracy": 0.2060795918107033,
      "num_tokens": 9478636.0,
      "step": 4605
    },
    {
      "entropy": 5.818109655380249,
      "epoch": 4.090545938748336,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003760413772519648,
      "loss": 5.1343,
      "mean_token_accuracy": 0.21224167048931122,
      "num_tokens": 9488626.0,
      "step": 4610
    },
    {
      "entropy": 5.805790519714355,
      "epoch": 4.094984465157568,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037573347409945983,
      "loss": 5.2094,
      "mean_token_accuracy": 0.20401569008827208,
      "num_tokens": 9501119.0,
      "step": 4615
    },
    {
      "entropy": 5.736809253692627,
      "epoch": 4.0994229915668,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003754253348345743,
      "loss": 5.0526,
      "mean_token_accuracy": 0.21723955422639846,
      "num_tokens": 9511075.0,
      "step": 4620
    },
    {
      "entropy": 5.720714664459228,
      "epoch": 4.103861517976032,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003751169601795657,
      "loss": 5.1149,
      "mean_token_accuracy": 0.2152253657579422,
      "num_tokens": 9520956.0,
      "step": 4625
    },
    {
      "entropy": 5.765715742111206,
      "epoch": 4.108300044385264,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037480835085724313,
      "loss": 5.1922,
      "mean_token_accuracy": 0.20740145891904832,
      "num_tokens": 9531020.0,
      "step": 4630
    },
    {
      "entropy": 5.826910781860351,
      "epoch": 4.112738570794496,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003744995075909656,
      "loss": 5.1995,
      "mean_token_accuracy": 0.20915032178163528,
      "num_tokens": 9541230.0,
      "step": 4635
    },
    {
      "entropy": 5.742929315567016,
      "epoch": 4.1171770972037285,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003741904311046408,
      "loss": 5.1937,
      "mean_token_accuracy": 0.20653378814458848,
      "num_tokens": 9551055.0,
      "step": 4640
    },
    {
      "entropy": 5.820126247406006,
      "epoch": 4.121615623612961,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003738811221227228,
      "loss": 5.2131,
      "mean_token_accuracy": 0.2067299783229828,
      "num_tokens": 9560982.0,
      "step": 4645
    },
    {
      "entropy": 5.5898134231567385,
      "epoch": 4.126054150022193,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037357158137021077,
      "loss": 5.0703,
      "mean_token_accuracy": 0.21925579160451888,
      "num_tokens": 9571950.0,
      "step": 4650
    },
    {
      "entropy": 5.7765885353088375,
      "epoch": 4.130492676431425,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000373261809572647,
      "loss": 5.192,
      "mean_token_accuracy": 0.2030755400657654,
      "num_tokens": 9581670.0,
      "step": 4655
    },
    {
      "entropy": 5.8400060653686525,
      "epoch": 4.134931202840657,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037295180745611553,
      "loss": 5.2075,
      "mean_token_accuracy": 0.2056944653391838,
      "num_tokens": 9592282.0,
      "step": 4660
    },
    {
      "entropy": 5.656895732879638,
      "epoch": 4.139369729249889,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003726415757472401,
      "loss": 5.1455,
      "mean_token_accuracy": 0.21399735808372497,
      "num_tokens": 9602530.0,
      "step": 4665
    },
    {
      "entropy": 5.784669542312622,
      "epoch": 4.143808255659121,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037233111517318257,
      "loss": 5.2154,
      "mean_token_accuracy": 0.20880706161260604,
      "num_tokens": 9612454.0,
      "step": 4670
    },
    {
      "entropy": 5.761039924621582,
      "epoch": 4.1482467820683535,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003720204264616414,
      "loss": 5.1301,
      "mean_token_accuracy": 0.21270381510257722,
      "num_tokens": 9623461.0,
      "step": 4675
    },
    {
      "entropy": 5.727607536315918,
      "epoch": 4.152685308477586,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003717095103408497,
      "loss": 5.1241,
      "mean_token_accuracy": 0.21046421229839324,
      "num_tokens": 9632502.0,
      "step": 4680
    },
    {
      "entropy": 5.74509973526001,
      "epoch": 4.157123834886818,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037139836753957353,
      "loss": 5.1327,
      "mean_token_accuracy": 0.2110206052660942,
      "num_tokens": 9642801.0,
      "step": 4685
    },
    {
      "entropy": 5.717509031295776,
      "epoch": 4.16156236129605,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037108699878711044,
      "loss": 5.1467,
      "mean_token_accuracy": 0.22023138552904128,
      "num_tokens": 9652638.0,
      "step": 4690
    },
    {
      "entropy": 5.7005359649658205,
      "epoch": 4.166000887705282,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037077540481328744,
      "loss": 5.1091,
      "mean_token_accuracy": 0.21125866025686263,
      "num_tokens": 9662480.0,
      "step": 4695
    },
    {
      "entropy": 5.737263202667236,
      "epoch": 4.170439414114514,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003704635863484596,
      "loss": 5.2501,
      "mean_token_accuracy": 0.20748572200536727,
      "num_tokens": 9673269.0,
      "step": 4700
    },
    {
      "entropy": 5.7536533832550045,
      "epoch": 4.174877940523746,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037015154412350806,
      "loss": 5.0887,
      "mean_token_accuracy": 0.20911131352186202,
      "num_tokens": 9682905.0,
      "step": 4705
    },
    {
      "entropy": 5.684925746917725,
      "epoch": 4.1793164669329785,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036983927886983847,
      "loss": 5.1556,
      "mean_token_accuracy": 0.212339323759079,
      "num_tokens": 9693609.0,
      "step": 4710
    },
    {
      "entropy": 5.720796203613281,
      "epoch": 4.183754993342211,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036952679131937923,
      "loss": 5.1308,
      "mean_token_accuracy": 0.21655481159687043,
      "num_tokens": 9703927.0,
      "step": 4715
    },
    {
      "entropy": 5.760318803787231,
      "epoch": 4.188193519751443,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003692140822045798,
      "loss": 5.1275,
      "mean_token_accuracy": 0.21760178804397584,
      "num_tokens": 9714440.0,
      "step": 4720
    },
    {
      "entropy": 5.6793900489807125,
      "epoch": 4.192632046160675,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036890115225840904,
      "loss": 5.1579,
      "mean_token_accuracy": 0.21766061335802078,
      "num_tokens": 9724586.0,
      "step": 4725
    },
    {
      "entropy": 5.717368078231812,
      "epoch": 4.197070572569907,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003685880022143533,
      "loss": 5.1252,
      "mean_token_accuracy": 0.2100989669561386,
      "num_tokens": 9734602.0,
      "step": 4730
    },
    {
      "entropy": 5.812760972976685,
      "epoch": 4.201509098979139,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036827463280641494,
      "loss": 5.2243,
      "mean_token_accuracy": 0.19821423143148423,
      "num_tokens": 9744970.0,
      "step": 4735
    },
    {
      "entropy": 5.766703367233276,
      "epoch": 4.205947625388371,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036796104476911033,
      "loss": 5.1653,
      "mean_token_accuracy": 0.2106243133544922,
      "num_tokens": 9754529.0,
      "step": 4740
    },
    {
      "entropy": 5.708454179763794,
      "epoch": 4.210386151797604,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036764723883746865,
      "loss": 5.1962,
      "mean_token_accuracy": 0.20117447674274444,
      "num_tokens": 9765742.0,
      "step": 4745
    },
    {
      "entropy": 5.742970275878906,
      "epoch": 4.214824678206836,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003673332157470293,
      "loss": 5.1156,
      "mean_token_accuracy": 0.2144193544983864,
      "num_tokens": 9776563.0,
      "step": 4750
    },
    {
      "entropy": 5.772315835952758,
      "epoch": 4.219263204616068,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000367018976233841,
      "loss": 5.137,
      "mean_token_accuracy": 0.2113969936966896,
      "num_tokens": 9786663.0,
      "step": 4755
    },
    {
      "entropy": 5.725389003753662,
      "epoch": 4.2237017310253,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003667045210344598,
      "loss": 5.1432,
      "mean_token_accuracy": 0.2056875169277191,
      "num_tokens": 9796499.0,
      "step": 4760
    },
    {
      "entropy": 5.7241839408874515,
      "epoch": 4.228140257434532,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003663898508859471,
      "loss": 5.1622,
      "mean_token_accuracy": 0.21654269099235535,
      "num_tokens": 9807305.0,
      "step": 4765
    },
    {
      "entropy": 5.753515100479126,
      "epoch": 4.232578783843764,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003660749665258684,
      "loss": 5.1908,
      "mean_token_accuracy": 0.20672854781150818,
      "num_tokens": 9818259.0,
      "step": 4770
    },
    {
      "entropy": 5.696164798736572,
      "epoch": 4.237017310252996,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003657598686922909,
      "loss": 5.1094,
      "mean_token_accuracy": 0.21507840007543563,
      "num_tokens": 9830146.0,
      "step": 4775
    },
    {
      "entropy": 5.7545092582702635,
      "epoch": 4.241455836662228,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003654445581237824,
      "loss": 5.1845,
      "mean_token_accuracy": 0.20769521594047546,
      "num_tokens": 9840847.0,
      "step": 4780
    },
    {
      "entropy": 5.662838459014893,
      "epoch": 4.245894363071461,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003651290355594096,
      "loss": 5.1024,
      "mean_token_accuracy": 0.2179262727499008,
      "num_tokens": 9850948.0,
      "step": 4785
    },
    {
      "entropy": 5.748239183425904,
      "epoch": 4.250332889480692,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003648133017387356,
      "loss": 5.1987,
      "mean_token_accuracy": 0.20759087502956391,
      "num_tokens": 9861205.0,
      "step": 4790
    },
    {
      "entropy": 5.666525506973267,
      "epoch": 4.254771415889924,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036449735740181884,
      "loss": 5.1803,
      "mean_token_accuracy": 0.205241397023201,
      "num_tokens": 9871660.0,
      "step": 4795
    },
    {
      "entropy": 5.73541841506958,
      "epoch": 4.259209942299156,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036418120328921146,
      "loss": 5.0938,
      "mean_token_accuracy": 0.22027103304862977,
      "num_tokens": 9881533.0,
      "step": 4800
    },
    {
      "entropy": 5.766236066818237,
      "epoch": 4.263648468708388,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036386484014195696,
      "loss": 5.2506,
      "mean_token_accuracy": 0.20312998443841934,
      "num_tokens": 9892707.0,
      "step": 4805
    },
    {
      "entropy": 5.696821928024292,
      "epoch": 4.268086995117621,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003635482687015891,
      "loss": 5.0745,
      "mean_token_accuracy": 0.21657687425613403,
      "num_tokens": 9901562.0,
      "step": 4810
    },
    {
      "entropy": 5.70862922668457,
      "epoch": 4.272525521526853,
      "grad_norm": 1.25,
      "learning_rate": 0.00036323148971012954,
      "loss": 5.1427,
      "mean_token_accuracy": 0.2149751827120781,
      "num_tokens": 9910742.0,
      "step": 4815
    },
    {
      "entropy": 5.784573650360107,
      "epoch": 4.276964047936085,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036291450391008655,
      "loss": 5.2066,
      "mean_token_accuracy": 0.20904852747917174,
      "num_tokens": 9922322.0,
      "step": 4820
    },
    {
      "entropy": 5.728090095520019,
      "epoch": 4.281402574345317,
      "grad_norm": 1.25,
      "learning_rate": 0.00036259731204445347,
      "loss": 5.117,
      "mean_token_accuracy": 0.20740775316953658,
      "num_tokens": 9931868.0,
      "step": 4825
    },
    {
      "entropy": 5.731380271911621,
      "epoch": 4.285841100754549,
      "grad_norm": 1.125,
      "learning_rate": 0.0003622799148567061,
      "loss": 5.179,
      "mean_token_accuracy": 0.21309688240289687,
      "num_tokens": 9942198.0,
      "step": 4830
    },
    {
      "entropy": 5.804520463943481,
      "epoch": 4.290279627163781,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003619623130908017,
      "loss": 5.209,
      "mean_token_accuracy": 0.2023901730775833,
      "num_tokens": 9953094.0,
      "step": 4835
    },
    {
      "entropy": 5.745091295242309,
      "epoch": 4.2947181535730135,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003616445074911774,
      "loss": 5.1656,
      "mean_token_accuracy": 0.2081662744283676,
      "num_tokens": 9964759.0,
      "step": 4840
    },
    {
      "entropy": 5.738615322113037,
      "epoch": 4.299156679982246,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003613264988027477,
      "loss": 5.1629,
      "mean_token_accuracy": 0.2116822764277458,
      "num_tokens": 9973916.0,
      "step": 4845
    },
    {
      "entropy": 5.72474102973938,
      "epoch": 4.303595206391478,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003610082877709031,
      "loss": 5.2008,
      "mean_token_accuracy": 0.20751210302114487,
      "num_tokens": 9985388.0,
      "step": 4850
    },
    {
      "entropy": 5.754292154312134,
      "epoch": 4.30803373280071,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036068987514150866,
      "loss": 5.1662,
      "mean_token_accuracy": 0.21000789403915404,
      "num_tokens": 9996143.0,
      "step": 4855
    },
    {
      "entropy": 5.7305224418640135,
      "epoch": 4.312472259209942,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036037126166090167,
      "loss": 5.2454,
      "mean_token_accuracy": 0.1990845561027527,
      "num_tokens": 10007376.0,
      "step": 4860
    },
    {
      "entropy": 5.722129201889038,
      "epoch": 4.316910785619174,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003600524480758906,
      "loss": 5.1317,
      "mean_token_accuracy": 0.2125830441713333,
      "num_tokens": 10016435.0,
      "step": 4865
    },
    {
      "entropy": 5.771578645706176,
      "epoch": 4.321349312028406,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035973343513375254,
      "loss": 5.2502,
      "mean_token_accuracy": 0.20185581594705582,
      "num_tokens": 10027577.0,
      "step": 4870
    },
    {
      "entropy": 5.819239234924316,
      "epoch": 4.3257878384376385,
      "grad_norm": 1.171875,
      "learning_rate": 0.000359414223582232,
      "loss": 5.1953,
      "mean_token_accuracy": 0.1998763546347618,
      "num_tokens": 10037099.0,
      "step": 4875
    },
    {
      "entropy": 5.746256351470947,
      "epoch": 4.330226364846871,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003590948141695392,
      "loss": 5.1537,
      "mean_token_accuracy": 0.20699663162231446,
      "num_tokens": 10047718.0,
      "step": 4880
    },
    {
      "entropy": 5.696747159957885,
      "epoch": 4.334664891256103,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003587752076443479,
      "loss": 5.1114,
      "mean_token_accuracy": 0.2158285990357399,
      "num_tokens": 10057507.0,
      "step": 4885
    },
    {
      "entropy": 5.7871781349182125,
      "epoch": 4.339103417665335,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003584554047557939,
      "loss": 5.2184,
      "mean_token_accuracy": 0.20748222768306732,
      "num_tokens": 10067643.0,
      "step": 4890
    },
    {
      "entropy": 5.658500385284424,
      "epoch": 4.343541944074567,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035813540625347334,
      "loss": 5.0818,
      "mean_token_accuracy": 0.21400361955165864,
      "num_tokens": 10077296.0,
      "step": 4895
    },
    {
      "entropy": 5.799658584594726,
      "epoch": 4.347980470483799,
      "grad_norm": 1.375,
      "learning_rate": 0.0003578152128874409,
      "loss": 5.2096,
      "mean_token_accuracy": 0.20976602435112,
      "num_tokens": 10087218.0,
      "step": 4900
    },
    {
      "entropy": 5.7649839401245115,
      "epoch": 4.352418996893031,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035749482540820796,
      "loss": 5.2046,
      "mean_token_accuracy": 0.2050032064318657,
      "num_tokens": 10096915.0,
      "step": 4905
    },
    {
      "entropy": 5.669079971313477,
      "epoch": 4.3568575233022635,
      "grad_norm": 1.25,
      "learning_rate": 0.00035717424456674086,
      "loss": 5.1605,
      "mean_token_accuracy": 0.20748622417449952,
      "num_tokens": 10108096.0,
      "step": 4910
    },
    {
      "entropy": 5.729409265518188,
      "epoch": 4.361296049711496,
      "grad_norm": 1.25,
      "learning_rate": 0.0003568534711144591,
      "loss": 5.1616,
      "mean_token_accuracy": 0.20923524498939514,
      "num_tokens": 10119131.0,
      "step": 4915
    },
    {
      "entropy": 5.727017021179199,
      "epoch": 4.365734576120728,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035653250580323383,
      "loss": 5.2125,
      "mean_token_accuracy": 0.2000671833753586,
      "num_tokens": 10129490.0,
      "step": 4920
    },
    {
      "entropy": 5.716195392608642,
      "epoch": 4.37017310252996,
      "grad_norm": 1.125,
      "learning_rate": 0.00035621134938538585,
      "loss": 5.0923,
      "mean_token_accuracy": 0.2212134048342705,
      "num_tokens": 10140282.0,
      "step": 4925
    },
    {
      "entropy": 5.7920060634613035,
      "epoch": 4.374611628939192,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003558900026136838,
      "loss": 5.2556,
      "mean_token_accuracy": 0.20267445594072342,
      "num_tokens": 10150256.0,
      "step": 4930
    },
    {
      "entropy": 5.661780214309692,
      "epoch": 4.379050155348424,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003555684662413424,
      "loss": 5.1197,
      "mean_token_accuracy": 0.21318988204002381,
      "num_tokens": 10160993.0,
      "step": 4935
    },
    {
      "entropy": 5.699134492874146,
      "epoch": 4.383488681757656,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003552467410220212,
      "loss": 5.1816,
      "mean_token_accuracy": 0.20566583573818206,
      "num_tokens": 10171585.0,
      "step": 4940
    },
    {
      "entropy": 5.622805070877075,
      "epoch": 4.3879272081668885,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003549248277098221,
      "loss": 5.1294,
      "mean_token_accuracy": 0.21171367913484573,
      "num_tokens": 10180923.0,
      "step": 4945
    },
    {
      "entropy": 5.71085786819458,
      "epoch": 4.392365734576121,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003546027270592878,
      "loss": 5.1071,
      "mean_token_accuracy": 0.21543533205986024,
      "num_tokens": 10190365.0,
      "step": 4950
    },
    {
      "entropy": 5.789462852478027,
      "epoch": 4.396804260985353,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035428043982540017,
      "loss": 5.2529,
      "mean_token_accuracy": 0.2044300004839897,
      "num_tokens": 10200400.0,
      "step": 4955
    },
    {
      "entropy": 5.729992198944092,
      "epoch": 4.401242787394585,
      "grad_norm": 1.125,
      "learning_rate": 0.00035395796676357855,
      "loss": 5.174,
      "mean_token_accuracy": 0.21109428107738495,
      "num_tokens": 10210489.0,
      "step": 4960
    },
    {
      "entropy": 5.7173525333404545,
      "epoch": 4.405681313803817,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003536353086296778,
      "loss": 5.1718,
      "mean_token_accuracy": 0.20589128136634827,
      "num_tokens": 10221141.0,
      "step": 4965
    },
    {
      "entropy": 5.723089933395386,
      "epoch": 4.410119840213049,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035331246617998654,
      "loss": 5.1542,
      "mean_token_accuracy": 0.2141341060400009,
      "num_tokens": 10230776.0,
      "step": 4970
    },
    {
      "entropy": 5.744500684738159,
      "epoch": 4.414558366622281,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003529894401712253,
      "loss": 5.1449,
      "mean_token_accuracy": 0.21146791875362397,
      "num_tokens": 10240366.0,
      "step": 4975
    },
    {
      "entropy": 5.6858978271484375,
      "epoch": 4.4189968930315136,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035266623136054505,
      "loss": 5.1778,
      "mean_token_accuracy": 0.20696759223937988,
      "num_tokens": 10250833.0,
      "step": 4980
    },
    {
      "entropy": 5.750640869140625,
      "epoch": 4.423435419440746,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035234284050552516,
      "loss": 5.2571,
      "mean_token_accuracy": 0.2038218379020691,
      "num_tokens": 10260242.0,
      "step": 4985
    },
    {
      "entropy": 5.765140056610107,
      "epoch": 4.427873945849978,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003520192683641718,
      "loss": 5.2297,
      "mean_token_accuracy": 0.2027687519788742,
      "num_tokens": 10272362.0,
      "step": 4990
    },
    {
      "entropy": 5.744220781326294,
      "epoch": 4.43231247225921,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003516955156949157,
      "loss": 5.1941,
      "mean_token_accuracy": 0.20152689814567565,
      "num_tokens": 10283176.0,
      "step": 4995
    },
    {
      "entropy": 5.7464361667633055,
      "epoch": 4.436750998668442,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035137158325661115,
      "loss": 5.2608,
      "mean_token_accuracy": 0.20259124040603638,
      "num_tokens": 10293630.0,
      "step": 5000
    },
    {
      "epoch": 4.436750998668442,
      "eval_entropy": 5.5769944247610495,
      "eval_loss": 5.8343658447265625,
      "eval_mean_token_accuracy": 0.17468414655632192,
      "eval_num_tokens": 10293630.0,
      "eval_runtime": 2.0854,
      "eval_samples_per_second": 1614.524,
      "eval_steps_per_second": 201.875,
      "step": 5000
    },
    {
      "entropy": 5.790251922607422,
      "epoch": 4.441189525077674,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035104747180853376,
      "loss": 5.1822,
      "mean_token_accuracy": 0.20639875233173371,
      "num_tokens": 10303775.0,
      "step": 5005
    },
    {
      "entropy": 5.715227222442627,
      "epoch": 4.445628051486906,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003507231821103785,
      "loss": 5.1551,
      "mean_token_accuracy": 0.21106622070074083,
      "num_tokens": 10314142.0,
      "step": 5010
    },
    {
      "entropy": 5.747024440765381,
      "epoch": 4.450066577896139,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035039871492225814,
      "loss": 5.1535,
      "mean_token_accuracy": 0.2024983510375023,
      "num_tokens": 10324401.0,
      "step": 5015
    },
    {
      "entropy": 5.675720167160034,
      "epoch": 4.454505104305371,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035007407100470187,
      "loss": 5.1739,
      "mean_token_accuracy": 0.21270538568496705,
      "num_tokens": 10334872.0,
      "step": 5020
    },
    {
      "entropy": 5.7632105350494385,
      "epoch": 4.458943630714603,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003497492511186527,
      "loss": 5.253,
      "mean_token_accuracy": 0.1968903511762619,
      "num_tokens": 10345504.0,
      "step": 5025
    },
    {
      "entropy": 5.672514867782593,
      "epoch": 4.463382157123835,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034942425602546616,
      "loss": 5.1775,
      "mean_token_accuracy": 0.2216365560889244,
      "num_tokens": 10355191.0,
      "step": 5030
    },
    {
      "entropy": 5.778588008880615,
      "epoch": 4.467820683533067,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034909908648690874,
      "loss": 5.2407,
      "mean_token_accuracy": 0.20244300067424775,
      "num_tokens": 10365185.0,
      "step": 5035
    },
    {
      "entropy": 5.789299154281617,
      "epoch": 4.472259209942299,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034877374326515556,
      "loss": 5.1825,
      "mean_token_accuracy": 0.207652448117733,
      "num_tokens": 10374977.0,
      "step": 5040
    },
    {
      "entropy": 5.714522647857666,
      "epoch": 4.4766977363515315,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034844822712278874,
      "loss": 5.2086,
      "mean_token_accuracy": 0.21069382429122924,
      "num_tokens": 10385182.0,
      "step": 5045
    },
    {
      "entropy": 5.704091644287109,
      "epoch": 4.481136262760764,
      "grad_norm": 1.125,
      "learning_rate": 0.0003481225388227961,
      "loss": 5.1767,
      "mean_token_accuracy": 0.21677854061126708,
      "num_tokens": 10395685.0,
      "step": 5050
    },
    {
      "entropy": 5.69912896156311,
      "epoch": 4.485574789169996,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034779667912856864,
      "loss": 5.1712,
      "mean_token_accuracy": 0.2076883152127266,
      "num_tokens": 10405920.0,
      "step": 5055
    },
    {
      "entropy": 5.758427858352661,
      "epoch": 4.490013315579228,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003474706488038993,
      "loss": 5.2679,
      "mean_token_accuracy": 0.19982607960700988,
      "num_tokens": 10417815.0,
      "step": 5060
    },
    {
      "entropy": 5.775757932662964,
      "epoch": 4.49445184198846,
      "grad_norm": 1.125,
      "learning_rate": 0.00034714444861298077,
      "loss": 5.2791,
      "mean_token_accuracy": 0.20297911316156386,
      "num_tokens": 10428603.0,
      "step": 5065
    },
    {
      "entropy": 5.725766277313232,
      "epoch": 4.498890368397692,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034681807932040407,
      "loss": 5.1524,
      "mean_token_accuracy": 0.2125113993883133,
      "num_tokens": 10438053.0,
      "step": 5070
    },
    {
      "entropy": 5.751907110214233,
      "epoch": 4.503328894806924,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003464915416911565,
      "loss": 5.2354,
      "mean_token_accuracy": 0.2079993560910225,
      "num_tokens": 10447750.0,
      "step": 5075
    },
    {
      "entropy": 5.751536750793457,
      "epoch": 4.5077674212161565,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034616483649062003,
      "loss": 5.2334,
      "mean_token_accuracy": 0.19996216744184495,
      "num_tokens": 10458168.0,
      "step": 5080
    },
    {
      "entropy": 5.77265772819519,
      "epoch": 4.512205947625389,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003458379644845693,
      "loss": 5.2019,
      "mean_token_accuracy": 0.20402003526687623,
      "num_tokens": 10468537.0,
      "step": 5085
    },
    {
      "entropy": 5.749958181381226,
      "epoch": 4.516644474034621,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003455109264391699,
      "loss": 5.2142,
      "mean_token_accuracy": 0.20577894747257233,
      "num_tokens": 10478549.0,
      "step": 5090
    },
    {
      "entropy": 5.679034996032715,
      "epoch": 4.521083000443853,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003451837231209766,
      "loss": 5.1651,
      "mean_token_accuracy": 0.22028762549161912,
      "num_tokens": 10488579.0,
      "step": 5095
    },
    {
      "entropy": 5.727813005447388,
      "epoch": 4.525521526853085,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034485635529693173,
      "loss": 5.2168,
      "mean_token_accuracy": 0.21334245204925537,
      "num_tokens": 10498869.0,
      "step": 5100
    },
    {
      "entropy": 5.774104833602905,
      "epoch": 4.529960053262317,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003445288237343632,
      "loss": 5.1222,
      "mean_token_accuracy": 0.21038362979888917,
      "num_tokens": 10509994.0,
      "step": 5105
    },
    {
      "entropy": 5.692390060424804,
      "epoch": 4.534398579671549,
      "grad_norm": 1.25,
      "learning_rate": 0.00034420112920098247,
      "loss": 5.1689,
      "mean_token_accuracy": 0.20680999457836152,
      "num_tokens": 10520386.0,
      "step": 5110
    },
    {
      "entropy": 5.761635255813599,
      "epoch": 4.5388371060807815,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003438732724648831,
      "loss": 5.3269,
      "mean_token_accuracy": 0.19650017619132995,
      "num_tokens": 10530802.0,
      "step": 5115
    },
    {
      "entropy": 5.7955528736114506,
      "epoch": 4.543275632490014,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034354525429453894,
      "loss": 5.1972,
      "mean_token_accuracy": 0.2062854290008545,
      "num_tokens": 10540581.0,
      "step": 5120
    },
    {
      "entropy": 5.708793306350708,
      "epoch": 4.547714158899246,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034321707545880223,
      "loss": 5.1542,
      "mean_token_accuracy": 0.21162179708480836,
      "num_tokens": 10550908.0,
      "step": 5125
    },
    {
      "entropy": 5.64053750038147,
      "epoch": 4.552152685308478,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034288873672690167,
      "loss": 5.1124,
      "mean_token_accuracy": 0.21572550535202026,
      "num_tokens": 10561710.0,
      "step": 5130
    },
    {
      "entropy": 5.748101234436035,
      "epoch": 4.55659121171771,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034256023886844076,
      "loss": 5.1799,
      "mean_token_accuracy": 0.20977407246828078,
      "num_tokens": 10571469.0,
      "step": 5135
    },
    {
      "entropy": 5.795603084564209,
      "epoch": 4.561029738126942,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034223158265339615,
      "loss": 5.1568,
      "mean_token_accuracy": 0.2124243423342705,
      "num_tokens": 10580968.0,
      "step": 5140
    },
    {
      "entropy": 5.732140684127808,
      "epoch": 4.5654682645361735,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034190276885211554,
      "loss": 5.1957,
      "mean_token_accuracy": 0.20838963836431504,
      "num_tokens": 10590848.0,
      "step": 5145
    },
    {
      "entropy": 5.73495044708252,
      "epoch": 4.5699067909454065,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003415737982353159,
      "loss": 5.1829,
      "mean_token_accuracy": 0.202427276968956,
      "num_tokens": 10600341.0,
      "step": 5150
    },
    {
      "entropy": 5.7257547855377195,
      "epoch": 4.574345317354638,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000341244671574082,
      "loss": 5.1245,
      "mean_token_accuracy": 0.21159932017326355,
      "num_tokens": 10611733.0,
      "step": 5155
    },
    {
      "entropy": 5.703883457183838,
      "epoch": 4.578783843763871,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003409153896398641,
      "loss": 5.1376,
      "mean_token_accuracy": 0.20912213623523712,
      "num_tokens": 10621570.0,
      "step": 5160
    },
    {
      "entropy": 5.7086564064025875,
      "epoch": 4.583222370173102,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034058595320447685,
      "loss": 5.1727,
      "mean_token_accuracy": 0.20659724175930022,
      "num_tokens": 10632438.0,
      "step": 5165
    },
    {
      "entropy": 5.724616813659668,
      "epoch": 4.587660896582335,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034025636304009646,
      "loss": 5.0619,
      "mean_token_accuracy": 0.21662326902151108,
      "num_tokens": 10641738.0,
      "step": 5170
    },
    {
      "entropy": 5.72975525856018,
      "epoch": 4.592099422991566,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003399266199192597,
      "loss": 5.2308,
      "mean_token_accuracy": 0.20958700329065322,
      "num_tokens": 10652854.0,
      "step": 5175
    },
    {
      "entropy": 5.65272216796875,
      "epoch": 4.596537949400799,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003395967246148622,
      "loss": 5.1518,
      "mean_token_accuracy": 0.2112141489982605,
      "num_tokens": 10663401.0,
      "step": 5180
    },
    {
      "entropy": 5.7544811248779295,
      "epoch": 4.600976475810031,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033926667790015584,
      "loss": 5.2547,
      "mean_token_accuracy": 0.20084319710731507,
      "num_tokens": 10675126.0,
      "step": 5185
    },
    {
      "entropy": 5.77255654335022,
      "epoch": 4.605415002219264,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003389364805487476,
      "loss": 5.2229,
      "mean_token_accuracy": 0.20229778736829757,
      "num_tokens": 10685212.0,
      "step": 5190
    },
    {
      "entropy": 5.691144752502441,
      "epoch": 4.609853528628495,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033860613333459757,
      "loss": 5.1348,
      "mean_token_accuracy": 0.2177197515964508,
      "num_tokens": 10695467.0,
      "step": 5195
    },
    {
      "entropy": 5.67885947227478,
      "epoch": 4.614292055037727,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003382756370320169,
      "loss": 5.1713,
      "mean_token_accuracy": 0.20744576752185823,
      "num_tokens": 10706089.0,
      "step": 5200
    },
    {
      "entropy": 5.803925657272339,
      "epoch": 4.618730581446959,
      "grad_norm": 1.125,
      "learning_rate": 0.0003379449924156664,
      "loss": 5.1694,
      "mean_token_accuracy": 0.20873753428459169,
      "num_tokens": 10716593.0,
      "step": 5205
    },
    {
      "entropy": 5.753027248382568,
      "epoch": 4.623169107856191,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003376142002605547,
      "loss": 5.1973,
      "mean_token_accuracy": 0.20896549969911576,
      "num_tokens": 10727642.0,
      "step": 5210
    },
    {
      "entropy": 5.683163452148437,
      "epoch": 4.6276076342654235,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003372832613420356,
      "loss": 5.0631,
      "mean_token_accuracy": 0.21866176128387452,
      "num_tokens": 10737135.0,
      "step": 5215
    },
    {
      "entropy": 5.693540334701538,
      "epoch": 4.632046160674656,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003369521764358075,
      "loss": 5.2051,
      "mean_token_accuracy": 0.20459542274475098,
      "num_tokens": 10747527.0,
      "step": 5220
    },
    {
      "entropy": 5.647522068023681,
      "epoch": 4.636484687083888,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003366209463179109,
      "loss": 5.072,
      "mean_token_accuracy": 0.21823544502258302,
      "num_tokens": 10757082.0,
      "step": 5225
    },
    {
      "entropy": 5.7527947425842285,
      "epoch": 4.64092321349312,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003362895717647265,
      "loss": 5.2071,
      "mean_token_accuracy": 0.20612578988075256,
      "num_tokens": 10768280.0,
      "step": 5230
    },
    {
      "entropy": 5.79077115058899,
      "epoch": 4.645361739902352,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003359580535529735,
      "loss": 5.1666,
      "mean_token_accuracy": 0.20676536858081818,
      "num_tokens": 10778276.0,
      "step": 5235
    },
    {
      "entropy": 5.723053407669068,
      "epoch": 4.649800266311584,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033562639245970807,
      "loss": 5.1456,
      "mean_token_accuracy": 0.206996351480484,
      "num_tokens": 10788004.0,
      "step": 5240
    },
    {
      "entropy": 5.712559366226197,
      "epoch": 4.654238792720816,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033529458926232105,
      "loss": 5.2224,
      "mean_token_accuracy": 0.2089390218257904,
      "num_tokens": 10797918.0,
      "step": 5245
    },
    {
      "entropy": 5.691585636138916,
      "epoch": 4.658677319130049,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003349626447385366,
      "loss": 5.157,
      "mean_token_accuracy": 0.2116892606019974,
      "num_tokens": 10808545.0,
      "step": 5250
    },
    {
      "entropy": 5.678885221481323,
      "epoch": 4.663115845539281,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003346305596664099,
      "loss": 5.2123,
      "mean_token_accuracy": 0.20954641252756118,
      "num_tokens": 10818941.0,
      "step": 5255
    },
    {
      "entropy": 5.76534628868103,
      "epoch": 4.667554371948513,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033429833482432567,
      "loss": 5.2666,
      "mean_token_accuracy": 0.2046421304345131,
      "num_tokens": 10829980.0,
      "step": 5260
    },
    {
      "entropy": 5.735799407958984,
      "epoch": 4.671992898357745,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033396597099099624,
      "loss": 5.1517,
      "mean_token_accuracy": 0.212375408411026,
      "num_tokens": 10840512.0,
      "step": 5265
    },
    {
      "entropy": 5.748387145996094,
      "epoch": 4.676431424766977,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033363346894545984,
      "loss": 5.1999,
      "mean_token_accuracy": 0.2036992847919464,
      "num_tokens": 10850837.0,
      "step": 5270
    },
    {
      "entropy": 5.699161005020142,
      "epoch": 4.680869951176209,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033330082946707827,
      "loss": 5.1828,
      "mean_token_accuracy": 0.21137236356735228,
      "num_tokens": 10862036.0,
      "step": 5275
    },
    {
      "entropy": 5.781758213043213,
      "epoch": 4.6853084775854414,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000332968053335536,
      "loss": 5.1708,
      "mean_token_accuracy": 0.211983323097229,
      "num_tokens": 10870900.0,
      "step": 5280
    },
    {
      "entropy": 5.730520582199096,
      "epoch": 4.689747003994674,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033263514133083757,
      "loss": 5.2581,
      "mean_token_accuracy": 0.20092154294252396,
      "num_tokens": 10880989.0,
      "step": 5285
    },
    {
      "entropy": 5.704589605331421,
      "epoch": 4.694185530403906,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033230209423330587,
      "loss": 5.178,
      "mean_token_accuracy": 0.2079184263944626,
      "num_tokens": 10890759.0,
      "step": 5290
    },
    {
      "entropy": 5.781274271011353,
      "epoch": 4.698624056813138,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003319689128235804,
      "loss": 5.1725,
      "mean_token_accuracy": 0.20617727935314178,
      "num_tokens": 10900700.0,
      "step": 5295
    },
    {
      "entropy": 5.6957298755645756,
      "epoch": 4.70306258322237,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033163559788261575,
      "loss": 5.1661,
      "mean_token_accuracy": 0.20840034782886505,
      "num_tokens": 10910378.0,
      "step": 5300
    },
    {
      "entropy": 5.7365562915802,
      "epoch": 4.707501109631602,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003313021501916795,
      "loss": 5.1849,
      "mean_token_accuracy": 0.20706552118062974,
      "num_tokens": 10920605.0,
      "step": 5305
    },
    {
      "entropy": 5.672279596328735,
      "epoch": 4.711939636040834,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033096857053235016,
      "loss": 5.14,
      "mean_token_accuracy": 0.21455953121185303,
      "num_tokens": 10931063.0,
      "step": 5310
    },
    {
      "entropy": 5.710047006607056,
      "epoch": 4.7163781624500665,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033063485968651545,
      "loss": 5.1996,
      "mean_token_accuracy": 0.21062558740377427,
      "num_tokens": 10941418.0,
      "step": 5315
    },
    {
      "entropy": 5.646402835845947,
      "epoch": 4.720816688859299,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003303010184363711,
      "loss": 5.1237,
      "mean_token_accuracy": 0.21142874658107758,
      "num_tokens": 10951664.0,
      "step": 5320
    },
    {
      "entropy": 5.749710512161255,
      "epoch": 4.725255215268531,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032996704756441803,
      "loss": 5.2223,
      "mean_token_accuracy": 0.20792291164398194,
      "num_tokens": 10962060.0,
      "step": 5325
    },
    {
      "entropy": 5.758749008178711,
      "epoch": 4.729693741677763,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003296329478534612,
      "loss": 5.2622,
      "mean_token_accuracy": 0.20276703983545302,
      "num_tokens": 10973630.0,
      "step": 5330
    },
    {
      "entropy": 5.694210529327393,
      "epoch": 4.734132268086995,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003292987200866075,
      "loss": 5.2161,
      "mean_token_accuracy": 0.2052842602133751,
      "num_tokens": 10984016.0,
      "step": 5335
    },
    {
      "entropy": 5.766045379638672,
      "epoch": 4.738570794496227,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003289643650472639,
      "loss": 5.1933,
      "mean_token_accuracy": 0.217355939745903,
      "num_tokens": 10994531.0,
      "step": 5340
    },
    {
      "entropy": 5.7393327236175535,
      "epoch": 4.743009320905459,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003286298835191358,
      "loss": 5.1673,
      "mean_token_accuracy": 0.2125002473592758,
      "num_tokens": 11004498.0,
      "step": 5345
    },
    {
      "entropy": 5.7455676078796385,
      "epoch": 4.7474478473146915,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032829527628622517,
      "loss": 5.1947,
      "mean_token_accuracy": 0.20545327067375183,
      "num_tokens": 11014460.0,
      "step": 5350
    },
    {
      "entropy": 5.711310625076294,
      "epoch": 4.751886373723924,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032796054413282834,
      "loss": 5.1853,
      "mean_token_accuracy": 0.2101360887289047,
      "num_tokens": 11024488.0,
      "step": 5355
    },
    {
      "entropy": 5.7556007385253904,
      "epoch": 4.756324900133156,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003276256878435347,
      "loss": 5.2054,
      "mean_token_accuracy": 0.21297023296356202,
      "num_tokens": 11033652.0,
      "step": 5360
    },
    {
      "entropy": 5.807051658630371,
      "epoch": 4.760763426542388,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003272907082032244,
      "loss": 5.2588,
      "mean_token_accuracy": 0.20238023847341538,
      "num_tokens": 11044587.0,
      "step": 5365
    },
    {
      "entropy": 5.66930661201477,
      "epoch": 4.76520195295162,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032695560599706706,
      "loss": 5.1097,
      "mean_token_accuracy": 0.21576271057128907,
      "num_tokens": 11053836.0,
      "step": 5370
    },
    {
      "entropy": 5.657267951965332,
      "epoch": 4.769640479360852,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032662038201051915,
      "loss": 5.1985,
      "mean_token_accuracy": 0.2085062623023987,
      "num_tokens": 11064250.0,
      "step": 5375
    },
    {
      "entropy": 5.730874490737915,
      "epoch": 4.774079005770084,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032628503702932275,
      "loss": 5.2485,
      "mean_token_accuracy": 0.2009250283241272,
      "num_tokens": 11074516.0,
      "step": 5380
    },
    {
      "entropy": 5.77521390914917,
      "epoch": 4.7785175321793165,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003259495718395038,
      "loss": 5.1925,
      "mean_token_accuracy": 0.21358498334884643,
      "num_tokens": 11084519.0,
      "step": 5385
    },
    {
      "entropy": 5.704379510879517,
      "epoch": 4.782956058588549,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003256139872273696,
      "loss": 5.2315,
      "mean_token_accuracy": 0.20225883275270462,
      "num_tokens": 11094492.0,
      "step": 5390
    },
    {
      "entropy": 5.678598880767822,
      "epoch": 4.787394584997781,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032527828397950763,
      "loss": 5.1649,
      "mean_token_accuracy": 0.2148972600698471,
      "num_tokens": 11104196.0,
      "step": 5395
    },
    {
      "entropy": 5.673313236236572,
      "epoch": 4.791833111407013,
      "grad_norm": 1.25,
      "learning_rate": 0.0003249424628827834,
      "loss": 5.071,
      "mean_token_accuracy": 0.2209952265024185,
      "num_tokens": 11114436.0,
      "step": 5400
    },
    {
      "entropy": 5.779289436340332,
      "epoch": 4.796271637816245,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032460652472433854,
      "loss": 5.2209,
      "mean_token_accuracy": 0.20202610343694688,
      "num_tokens": 11123866.0,
      "step": 5405
    },
    {
      "entropy": 5.7611284255981445,
      "epoch": 4.800710164225477,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032427047029158924,
      "loss": 5.2148,
      "mean_token_accuracy": 0.21345822662115096,
      "num_tokens": 11134817.0,
      "step": 5410
    },
    {
      "entropy": 5.715837097167968,
      "epoch": 4.805148690634709,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003239343003722239,
      "loss": 5.1417,
      "mean_token_accuracy": 0.2122260719537735,
      "num_tokens": 11145252.0,
      "step": 5415
    },
    {
      "entropy": 5.738485956192017,
      "epoch": 4.8095872170439415,
      "grad_norm": 1.15625,
      "learning_rate": 0.000323598015754202,
      "loss": 5.2536,
      "mean_token_accuracy": 0.20232891887426377,
      "num_tokens": 11154957.0,
      "step": 5420
    },
    {
      "entropy": 5.721615982055664,
      "epoch": 4.814025743453174,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003232616172257518,
      "loss": 5.1721,
      "mean_token_accuracy": 0.21200567334890366,
      "num_tokens": 11166565.0,
      "step": 5425
    },
    {
      "entropy": 5.751074361801147,
      "epoch": 4.818464269862406,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032292510557536844,
      "loss": 5.2772,
      "mean_token_accuracy": 0.20485566854476928,
      "num_tokens": 11176922.0,
      "step": 5430
    },
    {
      "entropy": 5.765995121002197,
      "epoch": 4.822902796271638,
      "grad_norm": 1.21875,
      "learning_rate": 0.000322588481591812,
      "loss": 5.1212,
      "mean_token_accuracy": 0.21359312385320664,
      "num_tokens": 11187526.0,
      "step": 5435
    },
    {
      "entropy": 5.726685285568237,
      "epoch": 4.82734132268087,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032225174606410634,
      "loss": 5.1969,
      "mean_token_accuracy": 0.20894990414381026,
      "num_tokens": 11199138.0,
      "step": 5440
    },
    {
      "entropy": 5.756986904144287,
      "epoch": 4.831779849090102,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032191489978153646,
      "loss": 5.2013,
      "mean_token_accuracy": 0.20594589412212372,
      "num_tokens": 11209158.0,
      "step": 5445
    },
    {
      "entropy": 5.692794561386108,
      "epoch": 4.836218375499334,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003215779435336471,
      "loss": 5.11,
      "mean_token_accuracy": 0.22132737040519715,
      "num_tokens": 11218202.0,
      "step": 5450
    },
    {
      "entropy": 5.732383728027344,
      "epoch": 4.840656901908567,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003212408781102404,
      "loss": 5.1896,
      "mean_token_accuracy": 0.20523899495601655,
      "num_tokens": 11227885.0,
      "step": 5455
    },
    {
      "entropy": 5.655904531478882,
      "epoch": 4.845095428317799,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003209037043013751,
      "loss": 5.2079,
      "mean_token_accuracy": 0.20618089586496352,
      "num_tokens": 11237686.0,
      "step": 5460
    },
    {
      "entropy": 5.712638187408447,
      "epoch": 4.849533954727031,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003205664228973632,
      "loss": 5.1826,
      "mean_token_accuracy": 0.20865253657102584,
      "num_tokens": 11248287.0,
      "step": 5465
    },
    {
      "entropy": 5.777940988540649,
      "epoch": 4.853972481136263,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003202290346887696,
      "loss": 5.1814,
      "mean_token_accuracy": 0.20437095165252686,
      "num_tokens": 11258376.0,
      "step": 5470
    },
    {
      "entropy": 5.766065216064453,
      "epoch": 4.858411007545495,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003198915404664088,
      "loss": 5.2219,
      "mean_token_accuracy": 0.19672314822673798,
      "num_tokens": 11268966.0,
      "step": 5475
    },
    {
      "entropy": 5.700806474685669,
      "epoch": 4.862849533954727,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031955394102134454,
      "loss": 5.2355,
      "mean_token_accuracy": 0.2040538489818573,
      "num_tokens": 11279836.0,
      "step": 5480
    },
    {
      "entropy": 5.7688816547393795,
      "epoch": 4.8672880603639594,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003192162371448868,
      "loss": 5.2809,
      "mean_token_accuracy": 0.20273662358522415,
      "num_tokens": 11290009.0,
      "step": 5485
    },
    {
      "entropy": 5.706856298446655,
      "epoch": 4.871726586773192,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031887842962859033,
      "loss": 5.1465,
      "mean_token_accuracy": 0.2068696990609169,
      "num_tokens": 11300460.0,
      "step": 5490
    },
    {
      "entropy": 5.7477126121521,
      "epoch": 4.876165113182424,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031854051926425277,
      "loss": 5.2111,
      "mean_token_accuracy": 0.20210069715976714,
      "num_tokens": 11309995.0,
      "step": 5495
    },
    {
      "entropy": 5.722295808792114,
      "epoch": 4.880603639591656,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031820250684391304,
      "loss": 5.1839,
      "mean_token_accuracy": 0.21283525675535203,
      "num_tokens": 11319682.0,
      "step": 5500
    },
    {
      "epoch": 4.880603639591656,
      "eval_entropy": 5.523329989643957,
      "eval_loss": 5.800832271575928,
      "eval_mean_token_accuracy": 0.17711487331044928,
      "eval_num_tokens": 11319682.0,
      "eval_runtime": 2.2641,
      "eval_samples_per_second": 1487.117,
      "eval_steps_per_second": 185.945,
      "step": 5500
    },
    {
      "entropy": 5.73695330619812,
      "epoch": 4.885042166000888,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031786439315984925,
      "loss": 5.2303,
      "mean_token_accuracy": 0.20797490924596787,
      "num_tokens": 11330186.0,
      "step": 5505
    },
    {
      "entropy": 5.770789432525635,
      "epoch": 4.88948069241012,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031752617900457656,
      "loss": 5.2255,
      "mean_token_accuracy": 0.21493835896253585,
      "num_tokens": 11340284.0,
      "step": 5510
    },
    {
      "entropy": 5.741439962387085,
      "epoch": 4.893919218819352,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000317187865170846,
      "loss": 5.2033,
      "mean_token_accuracy": 0.2078318029642105,
      "num_tokens": 11351098.0,
      "step": 5515
    },
    {
      "entropy": 5.707396221160889,
      "epoch": 4.8983577452285845,
      "grad_norm": 1.125,
      "learning_rate": 0.000316849452451642,
      "loss": 5.2368,
      "mean_token_accuracy": 0.20806288421154023,
      "num_tokens": 11361783.0,
      "step": 5520
    },
    {
      "entropy": 5.690252161026001,
      "epoch": 4.902796271637817,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031651094164018097,
      "loss": 5.1346,
      "mean_token_accuracy": 0.21844119280576707,
      "num_tokens": 11372840.0,
      "step": 5525
    },
    {
      "entropy": 5.7909657001495365,
      "epoch": 4.907234798047049,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003161723335299089,
      "loss": 5.1128,
      "mean_token_accuracy": 0.2176864340901375,
      "num_tokens": 11382785.0,
      "step": 5530
    },
    {
      "entropy": 5.6382709503173825,
      "epoch": 4.911673324456281,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003158336289145003,
      "loss": 5.1838,
      "mean_token_accuracy": 0.21205998063087464,
      "num_tokens": 11392714.0,
      "step": 5535
    },
    {
      "entropy": 5.741606950759888,
      "epoch": 4.916111850865512,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031549482858785554,
      "loss": 5.2685,
      "mean_token_accuracy": 0.19965731650590895,
      "num_tokens": 11404755.0,
      "step": 5540
    },
    {
      "entropy": 5.803937005996704,
      "epoch": 4.920550377274745,
      "grad_norm": 1.25,
      "learning_rate": 0.00031515593334409934,
      "loss": 5.2045,
      "mean_token_accuracy": 0.20873982608318328,
      "num_tokens": 11414938.0,
      "step": 5545
    },
    {
      "entropy": 5.686682176589966,
      "epoch": 4.9249889036839765,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003148169439775792,
      "loss": 5.1479,
      "mean_token_accuracy": 0.21138025522232057,
      "num_tokens": 11424962.0,
      "step": 5550
    },
    {
      "entropy": 5.696095275878906,
      "epoch": 4.9294274300932095,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00031447786128286286,
      "loss": 5.1778,
      "mean_token_accuracy": 0.21060777753591536,
      "num_tokens": 11436416.0,
      "step": 5555
    },
    {
      "entropy": 5.765445613861084,
      "epoch": 4.933865956502441,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003141386860547371,
      "loss": 5.2256,
      "mean_token_accuracy": 0.20483573526144028,
      "num_tokens": 11446820.0,
      "step": 5560
    },
    {
      "entropy": 5.775624513626099,
      "epoch": 4.938304482911674,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003137994190882054,
      "loss": 5.1784,
      "mean_token_accuracy": 0.2152741551399231,
      "num_tokens": 11456880.0,
      "step": 5565
    },
    {
      "entropy": 5.679759883880616,
      "epoch": 4.942743009320905,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003134600611784865,
      "loss": 5.1812,
      "mean_token_accuracy": 0.20349172353744507,
      "num_tokens": 11465758.0,
      "step": 5570
    },
    {
      "entropy": 5.664295959472656,
      "epoch": 4.947181535730138,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003131206131210119,
      "loss": 5.0953,
      "mean_token_accuracy": 0.21441334635019302,
      "num_tokens": 11476209.0,
      "step": 5575
    },
    {
      "entropy": 5.703896141052246,
      "epoch": 4.951620062139369,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003127810757114249,
      "loss": 5.1239,
      "mean_token_accuracy": 0.21405192017555236,
      "num_tokens": 11486510.0,
      "step": 5580
    },
    {
      "entropy": 5.755063581466675,
      "epoch": 4.9560585885486015,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031244144974557775,
      "loss": 5.1462,
      "mean_token_accuracy": 0.2157666265964508,
      "num_tokens": 11496910.0,
      "step": 5585
    },
    {
      "entropy": 5.703655195236206,
      "epoch": 4.960497114957834,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003121017360195308,
      "loss": 5.1891,
      "mean_token_accuracy": 0.20666308254003524,
      "num_tokens": 11507488.0,
      "step": 5590
    },
    {
      "entropy": 5.749307441711426,
      "epoch": 4.964935641367066,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031176193532954957,
      "loss": 5.2232,
      "mean_token_accuracy": 0.20651409029960632,
      "num_tokens": 11518398.0,
      "step": 5595
    },
    {
      "entropy": 5.7297381401062015,
      "epoch": 4.969374167776298,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003114220484721038,
      "loss": 5.1564,
      "mean_token_accuracy": 0.20705897808074952,
      "num_tokens": 11528470.0,
      "step": 5600
    },
    {
      "entropy": 5.670311784744262,
      "epoch": 4.97381269418553,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031108207624386486,
      "loss": 5.1299,
      "mean_token_accuracy": 0.2155166521668434,
      "num_tokens": 11537765.0,
      "step": 5605
    },
    {
      "entropy": 5.7339622497558596,
      "epoch": 4.978251220594762,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003107420194417047,
      "loss": 5.249,
      "mean_token_accuracy": 0.2012795925140381,
      "num_tokens": 11549065.0,
      "step": 5610
    },
    {
      "entropy": 5.769830179214478,
      "epoch": 4.982689747003994,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000310401878862693,
      "loss": 5.2641,
      "mean_token_accuracy": 0.20485369861125946,
      "num_tokens": 11559010.0,
      "step": 5615
    },
    {
      "entropy": 5.698846435546875,
      "epoch": 4.9871282734132265,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003100616553040962,
      "loss": 5.1158,
      "mean_token_accuracy": 0.21705816090106964,
      "num_tokens": 11567978.0,
      "step": 5620
    },
    {
      "entropy": 5.703730916976928,
      "epoch": 4.991566799822459,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030972134956337493,
      "loss": 5.1565,
      "mean_token_accuracy": 0.20373631715774537,
      "num_tokens": 11579433.0,
      "step": 5625
    },
    {
      "entropy": 5.69541335105896,
      "epoch": 4.996005326231691,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030938096243818275,
      "loss": 5.1387,
      "mean_token_accuracy": 0.21119639426469802,
      "num_tokens": 11589882.0,
      "step": 5630
    },
    {
      "entropy": 5.7001016404893665,
      "epoch": 5.0,
      "grad_norm": 2.296875,
      "learning_rate": 0.00030904049472636367,
      "loss": 5.1901,
      "mean_token_accuracy": 0.21247350341743892,
      "num_tokens": 11598630.0,
      "step": 5635
    },
    {
      "entropy": 5.739415073394776,
      "epoch": 5.004438526409232,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030869994722595095,
      "loss": 5.008,
      "mean_token_accuracy": 0.22126824110746385,
      "num_tokens": 11608797.0,
      "step": 5640
    },
    {
      "entropy": 5.663621139526367,
      "epoch": 5.008877052818464,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003083593207351644,
      "loss": 5.0175,
      "mean_token_accuracy": 0.22501287162303923,
      "num_tokens": 11619095.0,
      "step": 5645
    },
    {
      "entropy": 5.700268888473511,
      "epoch": 5.013315579227696,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003080186160524095,
      "loss": 5.0311,
      "mean_token_accuracy": 0.22601672559976577,
      "num_tokens": 11629770.0,
      "step": 5650
    },
    {
      "entropy": 5.718568563461304,
      "epoch": 5.017754105636929,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003076778339762745,
      "loss": 4.948,
      "mean_token_accuracy": 0.23284862637519838,
      "num_tokens": 11640785.0,
      "step": 5655
    },
    {
      "entropy": 5.695920658111572,
      "epoch": 5.022192632046161,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030733697530552955,
      "loss": 4.9336,
      "mean_token_accuracy": 0.22667457312345504,
      "num_tokens": 11651995.0,
      "step": 5660
    },
    {
      "entropy": 5.738033103942871,
      "epoch": 5.026631158455393,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003069960408391239,
      "loss": 5.0168,
      "mean_token_accuracy": 0.21783950328826904,
      "num_tokens": 11662821.0,
      "step": 5665
    },
    {
      "entropy": 5.666689348220825,
      "epoch": 5.031069684864625,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030665503137618466,
      "loss": 4.8584,
      "mean_token_accuracy": 0.2404223471879959,
      "num_tokens": 11672423.0,
      "step": 5670
    },
    {
      "entropy": 5.6299999237060545,
      "epoch": 5.035508211273857,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003063139477160147,
      "loss": 4.918,
      "mean_token_accuracy": 0.23635224550962447,
      "num_tokens": 11683010.0,
      "step": 5675
    },
    {
      "entropy": 5.682493114471436,
      "epoch": 5.039946737683089,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030597279065809103,
      "loss": 4.9942,
      "mean_token_accuracy": 0.22202890664339064,
      "num_tokens": 11693887.0,
      "step": 5680
    },
    {
      "entropy": 5.711723518371582,
      "epoch": 5.0443852640923215,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003056315610020622,
      "loss": 5.0131,
      "mean_token_accuracy": 0.22168050855398178,
      "num_tokens": 11705152.0,
      "step": 5685
    },
    {
      "entropy": 5.658341979980468,
      "epoch": 5.048823790501554,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003052902595477474,
      "loss": 4.9349,
      "mean_token_accuracy": 0.22618422508239747,
      "num_tokens": 11714953.0,
      "step": 5690
    },
    {
      "entropy": 5.796881008148193,
      "epoch": 5.053262316910786,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00030494888709513377,
      "loss": 5.048,
      "mean_token_accuracy": 0.2163314253091812,
      "num_tokens": 11725273.0,
      "step": 5695
    },
    {
      "entropy": 5.722124481201172,
      "epoch": 5.057700843320018,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003046074444443751,
      "loss": 5.0219,
      "mean_token_accuracy": 0.21710677593946456,
      "num_tokens": 11735757.0,
      "step": 5700
    },
    {
      "entropy": 5.642864036560058,
      "epoch": 5.06213936972925,
      "grad_norm": 1.328125,
      "learning_rate": 0.00030426593239578966,
      "loss": 5.0246,
      "mean_token_accuracy": 0.22806545048952104,
      "num_tokens": 11746190.0,
      "step": 5705
    },
    {
      "entropy": 5.654903078079224,
      "epoch": 5.066577896138482,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003039243517498583,
      "loss": 4.9854,
      "mean_token_accuracy": 0.2298506259918213,
      "num_tokens": 11756692.0,
      "step": 5710
    },
    {
      "entropy": 5.702650833129883,
      "epoch": 5.071016422547714,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003035827033072228,
      "loss": 4.9319,
      "mean_token_accuracy": 0.22459085732698442,
      "num_tokens": 11768071.0,
      "step": 5715
    },
    {
      "entropy": 5.720613718032837,
      "epoch": 5.0754549489569465,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030324098786868364,
      "loss": 5.0182,
      "mean_token_accuracy": 0.22162954807281493,
      "num_tokens": 11778685.0,
      "step": 5720
    },
    {
      "entropy": 5.6652062892913815,
      "epoch": 5.079893475366179,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030289920623519854,
      "loss": 4.9676,
      "mean_token_accuracy": 0.2229399025440216,
      "num_tokens": 11788078.0,
      "step": 5725
    },
    {
      "entropy": 5.676460599899292,
      "epoch": 5.084332001775411,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003025573592078803,
      "loss": 4.9833,
      "mean_token_accuracy": 0.22160238772630692,
      "num_tokens": 11796440.0,
      "step": 5730
    },
    {
      "entropy": 5.745248508453369,
      "epoch": 5.088770528184643,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000302215447587995,
      "loss": 5.0358,
      "mean_token_accuracy": 0.21714013665914536,
      "num_tokens": 11806263.0,
      "step": 5735
    },
    {
      "entropy": 5.646965360641479,
      "epoch": 5.093209054593875,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030187347217696005,
      "loss": 4.9496,
      "mean_token_accuracy": 0.22523796260356904,
      "num_tokens": 11816368.0,
      "step": 5740
    },
    {
      "entropy": 5.5966925621032715,
      "epoch": 5.097647581003107,
      "grad_norm": 1.125,
      "learning_rate": 0.00030153143377634244,
      "loss": 4.9276,
      "mean_token_accuracy": 0.22945380359888076,
      "num_tokens": 11826319.0,
      "step": 5745
    },
    {
      "entropy": 5.676280736923218,
      "epoch": 5.102086107412339,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003011893331878569,
      "loss": 4.9664,
      "mean_token_accuracy": 0.23140579015016555,
      "num_tokens": 11835971.0,
      "step": 5750
    },
    {
      "entropy": 5.701886701583862,
      "epoch": 5.1065246338215715,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030084717121336383,
      "loss": 4.9618,
      "mean_token_accuracy": 0.22556509524583818,
      "num_tokens": 11845506.0,
      "step": 5755
    },
    {
      "entropy": 5.715242481231689,
      "epoch": 5.110963160230804,
      "grad_norm": 1.125,
      "learning_rate": 0.00030050494865486744,
      "loss": 5.0605,
      "mean_token_accuracy": 0.2205240473151207,
      "num_tokens": 11856083.0,
      "step": 5760
    },
    {
      "entropy": 5.595279550552368,
      "epoch": 5.115401686640036,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003001626663145141,
      "loss": 4.9238,
      "mean_token_accuracy": 0.23146225064992904,
      "num_tokens": 11867061.0,
      "step": 5765
    },
    {
      "entropy": 5.720956182479858,
      "epoch": 5.119840213049268,
      "grad_norm": 1.25,
      "learning_rate": 0.00029982032499459017,
      "loss": 5.0122,
      "mean_token_accuracy": 0.22077643722295762,
      "num_tokens": 11877751.0,
      "step": 5770
    },
    {
      "entropy": 5.651248216629028,
      "epoch": 5.1242787394585,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029947792549752034,
      "loss": 4.9893,
      "mean_token_accuracy": 0.22719825357198714,
      "num_tokens": 11887649.0,
      "step": 5775
    },
    {
      "entropy": 5.659734487533569,
      "epoch": 5.128717265867732,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029913546862586567,
      "loss": 5.0319,
      "mean_token_accuracy": 0.223132161796093,
      "num_tokens": 11898833.0,
      "step": 5780
    },
    {
      "entropy": 5.735756158828735,
      "epoch": 5.133155792276964,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002987929551823215,
      "loss": 5.0595,
      "mean_token_accuracy": 0.22174521237611772,
      "num_tokens": 11910341.0,
      "step": 5785
    },
    {
      "entropy": 5.6615043640136715,
      "epoch": 5.1375943186861965,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002984503859697162,
      "loss": 4.9508,
      "mean_token_accuracy": 0.23014761209487916,
      "num_tokens": 11919393.0,
      "step": 5790
    },
    {
      "entropy": 5.6714434146881105,
      "epoch": 5.142032845095429,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002981077617910085,
      "loss": 4.9194,
      "mean_token_accuracy": 0.23738617449998856,
      "num_tokens": 11930765.0,
      "step": 5795
    },
    {
      "entropy": 5.629246377944947,
      "epoch": 5.146471371504661,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029776508344928597,
      "loss": 4.9687,
      "mean_token_accuracy": 0.22703385949134827,
      "num_tokens": 11940054.0,
      "step": 5800
    },
    {
      "entropy": 5.650801801681519,
      "epoch": 5.150909897913893,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002974223517477633,
      "loss": 4.9464,
      "mean_token_accuracy": 0.23225471675395964,
      "num_tokens": 11950756.0,
      "step": 5805
    },
    {
      "entropy": 5.657662773132325,
      "epoch": 5.155348424323125,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002970795674897802,
      "loss": 4.9582,
      "mean_token_accuracy": 0.22137288451194764,
      "num_tokens": 11961062.0,
      "step": 5810
    },
    {
      "entropy": 5.690600299835205,
      "epoch": 5.159786950732357,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002967367314787995,
      "loss": 4.9235,
      "mean_token_accuracy": 0.22654454112052919,
      "num_tokens": 11970557.0,
      "step": 5815
    },
    {
      "entropy": 5.638317346572876,
      "epoch": 5.164225477141589,
      "grad_norm": 1.125,
      "learning_rate": 0.00029639384451840545,
      "loss": 4.9384,
      "mean_token_accuracy": 0.22700047940015794,
      "num_tokens": 11981331.0,
      "step": 5820
    },
    {
      "entropy": 5.6244978427886965,
      "epoch": 5.168664003550822,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002960509074123015,
      "loss": 4.9891,
      "mean_token_accuracy": 0.21644605994224547,
      "num_tokens": 11992159.0,
      "step": 5825
    },
    {
      "entropy": 5.716595268249511,
      "epoch": 5.173102529960053,
      "grad_norm": 1.234375,
      "learning_rate": 0.000295707920964309,
      "loss": 4.9931,
      "mean_token_accuracy": 0.22334617823362352,
      "num_tokens": 12002603.0,
      "step": 5830
    },
    {
      "entropy": 5.7139030456542965,
      "epoch": 5.177541056369286,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002953648859783646,
      "loss": 5.1011,
      "mean_token_accuracy": 0.21508881002664565,
      "num_tokens": 12012623.0,
      "step": 5835
    },
    {
      "entropy": 5.665186405181885,
      "epoch": 5.181979582778517,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002950218032585191,
      "loss": 4.9749,
      "mean_token_accuracy": 0.22413384765386582,
      "num_tokens": 12023017.0,
      "step": 5840
    },
    {
      "entropy": 5.671549224853516,
      "epoch": 5.186418109187749,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002946786736089346,
      "loss": 4.9773,
      "mean_token_accuracy": 0.2214719161391258,
      "num_tokens": 12033032.0,
      "step": 5845
    },
    {
      "entropy": 5.677174139022827,
      "epoch": 5.190856635596981,
      "grad_norm": 1.125,
      "learning_rate": 0.0002943354978338838,
      "loss": 5.0024,
      "mean_token_accuracy": 0.2283272624015808,
      "num_tokens": 12043609.0,
      "step": 5850
    },
    {
      "entropy": 5.730323934555054,
      "epoch": 5.1952951620062136,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002939922767377472,
      "loss": 5.0104,
      "mean_token_accuracy": 0.22389555275440215,
      "num_tokens": 12053282.0,
      "step": 5855
    },
    {
      "entropy": 5.7337881565094,
      "epoch": 5.199733688415446,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002936490111250116,
      "loss": 5.0561,
      "mean_token_accuracy": 0.217279052734375,
      "num_tokens": 12063672.0,
      "step": 5860
    },
    {
      "entropy": 5.663741397857666,
      "epoch": 5.204172214824678,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002933057018002681,
      "loss": 4.9316,
      "mean_token_accuracy": 0.22680189162492753,
      "num_tokens": 12073921.0,
      "step": 5865
    },
    {
      "entropy": 5.718992376327515,
      "epoch": 5.20861074123391,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029296234956821044,
      "loss": 4.9402,
      "mean_token_accuracy": 0.22692192792892457,
      "num_tokens": 12085149.0,
      "step": 5870
    },
    {
      "entropy": 5.621643352508545,
      "epoch": 5.213049267643142,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002926189552336326,
      "loss": 4.9647,
      "mean_token_accuracy": 0.2331915631890297,
      "num_tokens": 12096515.0,
      "step": 5875
    },
    {
      "entropy": 5.674287557601929,
      "epoch": 5.217487794052374,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002922755196014277,
      "loss": 4.9135,
      "mean_token_accuracy": 0.22644905298948287,
      "num_tokens": 12106352.0,
      "step": 5880
    },
    {
      "entropy": 5.652184581756591,
      "epoch": 5.221926320461606,
      "grad_norm": 1.3125,
      "learning_rate": 0.000291932043476585,
      "loss": 4.9627,
      "mean_token_accuracy": 0.2293543979525566,
      "num_tokens": 12116344.0,
      "step": 5885
    },
    {
      "entropy": 5.67815523147583,
      "epoch": 5.226364846870839,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002915885276641895,
      "loss": 5.1006,
      "mean_token_accuracy": 0.21285598427057267,
      "num_tokens": 12127402.0,
      "step": 5890
    },
    {
      "entropy": 5.687961149215698,
      "epoch": 5.230803373280071,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029124497296941843,
      "loss": 4.9267,
      "mean_token_accuracy": 0.22959166169166564,
      "num_tokens": 12138065.0,
      "step": 5895
    },
    {
      "entropy": 5.636407089233399,
      "epoch": 5.235241899689303,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029090138019754075,
      "loss": 4.9783,
      "mean_token_accuracy": 0.22571838796138763,
      "num_tokens": 12148450.0,
      "step": 5900
    },
    {
      "entropy": 5.608906221389771,
      "epoch": 5.239680426098535,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002905577501539144,
      "loss": 4.9822,
      "mean_token_accuracy": 0.21735644042491914,
      "num_tokens": 12158587.0,
      "step": 5905
    },
    {
      "entropy": 5.650953483581543,
      "epoch": 5.244118952507767,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002902140836439847,
      "loss": 4.9359,
      "mean_token_accuracy": 0.22076532393693923,
      "num_tokens": 12169139.0,
      "step": 5910
    },
    {
      "entropy": 5.664229774475098,
      "epoch": 5.248557478916999,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002898703814732824,
      "loss": 4.9963,
      "mean_token_accuracy": 0.22586210668087006,
      "num_tokens": 12179178.0,
      "step": 5915
    },
    {
      "entropy": 5.747469329833985,
      "epoch": 5.2529960053262315,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028952664444742204,
      "loss": 5.0667,
      "mean_token_accuracy": 0.21449797302484513,
      "num_tokens": 12189018.0,
      "step": 5920
    },
    {
      "entropy": 5.725044107437133,
      "epoch": 5.257434531735464,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002891828733720996,
      "loss": 5.0464,
      "mean_token_accuracy": 0.22428497821092605,
      "num_tokens": 12199250.0,
      "step": 5925
    },
    {
      "entropy": 5.674435091018677,
      "epoch": 5.261873058144696,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028883906905309103,
      "loss": 5.0602,
      "mean_token_accuracy": 0.21731068789958954,
      "num_tokens": 12209572.0,
      "step": 5930
    },
    {
      "entropy": 5.678584957122803,
      "epoch": 5.266311584553928,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002884952322962502,
      "loss": 5.0185,
      "mean_token_accuracy": 0.22419251948595048,
      "num_tokens": 12220168.0,
      "step": 5935
    },
    {
      "entropy": 5.725101184844971,
      "epoch": 5.27075011096316,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028815136390750694,
      "loss": 4.967,
      "mean_token_accuracy": 0.2251395106315613,
      "num_tokens": 12229765.0,
      "step": 5940
    },
    {
      "entropy": 5.604526662826538,
      "epoch": 5.275188637372392,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002878074646928653,
      "loss": 4.9616,
      "mean_token_accuracy": 0.22482145875692366,
      "num_tokens": 12240087.0,
      "step": 5945
    },
    {
      "entropy": 5.676482009887695,
      "epoch": 5.279627163781624,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002874635354584015,
      "loss": 4.9899,
      "mean_token_accuracy": 0.22442464679479598,
      "num_tokens": 12251489.0,
      "step": 5950
    },
    {
      "entropy": 5.718767786026001,
      "epoch": 5.2840656901908565,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002871195770102621,
      "loss": 5.0116,
      "mean_token_accuracy": 0.22521375119686127,
      "num_tokens": 12261114.0,
      "step": 5955
    },
    {
      "entropy": 5.649609994888306,
      "epoch": 5.288504216600089,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002867755901546624,
      "loss": 4.9961,
      "mean_token_accuracy": 0.21924589872360228,
      "num_tokens": 12271840.0,
      "step": 5960
    },
    {
      "entropy": 5.661839437484741,
      "epoch": 5.292942743009321,
      "grad_norm": 1.25,
      "learning_rate": 0.00028643157569788386,
      "loss": 4.9661,
      "mean_token_accuracy": 0.22587821185588836,
      "num_tokens": 12281407.0,
      "step": 5965
    },
    {
      "entropy": 5.667834806442261,
      "epoch": 5.297381269418553,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028608753444627307,
      "loss": 4.9471,
      "mean_token_accuracy": 0.23786964565515517,
      "num_tokens": 12293087.0,
      "step": 5970
    },
    {
      "entropy": 5.657839250564575,
      "epoch": 5.301819795827785,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000285743467206239,
      "loss": 4.9241,
      "mean_token_accuracy": 0.23313162177801133,
      "num_tokens": 12303638.0,
      "step": 5975
    },
    {
      "entropy": 5.723939514160156,
      "epoch": 5.306258322237017,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028539937478425196,
      "loss": 5.004,
      "mean_token_accuracy": 0.22194170355796813,
      "num_tokens": 12314059.0,
      "step": 5980
    },
    {
      "entropy": 5.626681756973267,
      "epoch": 5.310696848646249,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002850552579868409,
      "loss": 4.9439,
      "mean_token_accuracy": 0.23021697402000427,
      "num_tokens": 12323447.0,
      "step": 5985
    },
    {
      "entropy": 5.587870359420776,
      "epoch": 5.3151353750554815,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002847111176205922,
      "loss": 4.9479,
      "mean_token_accuracy": 0.23014504015445708,
      "num_tokens": 12333676.0,
      "step": 5990
    },
    {
      "entropy": 5.688661241531372,
      "epoch": 5.319573901464714,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002843669544921473,
      "loss": 5.0272,
      "mean_token_accuracy": 0.22250010669231415,
      "num_tokens": 12344188.0,
      "step": 5995
    },
    {
      "entropy": 5.607497310638427,
      "epoch": 5.324012427873946,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002840227694082011,
      "loss": 4.8953,
      "mean_token_accuracy": 0.2363758459687233,
      "num_tokens": 12353898.0,
      "step": 6000
    },
    {
      "epoch": 5.324012427873946,
      "eval_entropy": 5.461305147112124,
      "eval_loss": 5.801708221435547,
      "eval_mean_token_accuracy": 0.1778349627579193,
      "eval_num_tokens": 12353898.0,
      "eval_runtime": 2.0862,
      "eval_samples_per_second": 1613.94,
      "eval_steps_per_second": 201.802,
      "step": 6000
    }
  ],
  "logging_steps": 5,
  "max_steps": 11260,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2779609975511040.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}