eus-latn-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407/checkpoint-11500/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.880103137086378,
  "eval_steps": 500,
  "global_step": 11500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 4.90788049697876,
      "epoch": 0.004297378599054577,
      "grad_norm": 23.0,
      "learning_rate": 2e-06,
      "loss": 13.9846,
      "mean_token_accuracy": 0.0,
      "num_tokens": 10107.0,
      "step": 5
    },
    {
      "entropy": 4.913597154617309,
      "epoch": 0.008594757198109154,
      "grad_norm": 23.875,
      "learning_rate": 4.5e-06,
      "loss": 13.7937,
      "mean_token_accuracy": 0.0,
      "num_tokens": 18391.0,
      "step": 10
    },
    {
      "entropy": 4.986179208755493,
      "epoch": 0.01289213579716373,
      "grad_norm": 25.25,
      "learning_rate": 7e-06,
      "loss": 13.7057,
      "mean_token_accuracy": 7.022471982054412e-05,
      "num_tokens": 27061.0,
      "step": 15
    },
    {
      "entropy": 5.105097055435181,
      "epoch": 0.017189514396218308,
      "grad_norm": 31.25,
      "learning_rate": 9.5e-06,
      "loss": 13.4704,
      "mean_token_accuracy": 0.0,
      "num_tokens": 36339.0,
      "step": 20
    },
    {
      "entropy": 5.69900426864624,
      "epoch": 0.021486892995272882,
      "grad_norm": 38.5,
      "learning_rate": 1.2e-05,
      "loss": 12.6357,
      "mean_token_accuracy": 0.0002869559801183641,
      "num_tokens": 45770.0,
      "step": 25
    },
    {
      "entropy": 7.67990927696228,
      "epoch": 0.02578427159432746,
      "grad_norm": 20.875,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 11.573,
      "mean_token_accuracy": 0.0,
      "num_tokens": 54575.0,
      "step": 30
    },
    {
      "entropy": 10.355410289764404,
      "epoch": 0.030081650193382038,
      "grad_norm": 3.453125,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.8476,
      "mean_token_accuracy": 0.0001408907352015376,
      "num_tokens": 66403.0,
      "step": 35
    },
    {
      "entropy": 10.715928554534912,
      "epoch": 0.034379028792436615,
      "grad_norm": 3.53125,
      "learning_rate": 1.95e-05,
      "loss": 10.6808,
      "mean_token_accuracy": 0.001065203151665628,
      "num_tokens": 76510.0,
      "step": 40
    },
    {
      "entropy": 10.736311149597167,
      "epoch": 0.03867640739149119,
      "grad_norm": 3.328125,
      "learning_rate": 2.2e-05,
      "loss": 10.4612,
      "mean_token_accuracy": 0.014692968456074595,
      "num_tokens": 84836.0,
      "step": 45
    },
    {
      "entropy": 10.713325881958008,
      "epoch": 0.042973785990545764,
      "grad_norm": 2.359375,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 10.2528,
      "mean_token_accuracy": 0.021226395200937987,
      "num_tokens": 93197.0,
      "step": 50
    },
    {
      "entropy": 10.619150924682618,
      "epoch": 0.047271164589600345,
      "grad_norm": 2.90625,
      "learning_rate": 2.7e-05,
      "loss": 10.0647,
      "mean_token_accuracy": 0.019917360693216323,
      "num_tokens": 101546.0,
      "step": 55
    },
    {
      "entropy": 10.670361804962159,
      "epoch": 0.05156854318865492,
      "grad_norm": 2.046875,
      "learning_rate": 2.95e-05,
      "loss": 9.9914,
      "mean_token_accuracy": 0.023917235806584357,
      "num_tokens": 111703.0,
      "step": 60
    },
    {
      "entropy": 10.706762981414794,
      "epoch": 0.055865921787709494,
      "grad_norm": 2.0,
      "learning_rate": 3.2e-05,
      "loss": 9.8119,
      "mean_token_accuracy": 0.05865457281470299,
      "num_tokens": 119894.0,
      "step": 65
    },
    {
      "entropy": 10.680897426605224,
      "epoch": 0.060163300386764075,
      "grad_norm": 1.96875,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.7551,
      "mean_token_accuracy": 0.060981281101703644,
      "num_tokens": 128885.0,
      "step": 70
    },
    {
      "entropy": 10.6582462310791,
      "epoch": 0.06446067898581866,
      "grad_norm": 1.9609375,
      "learning_rate": 3.7e-05,
      "loss": 9.7084,
      "mean_token_accuracy": 0.05785200446844101,
      "num_tokens": 138106.0,
      "step": 75
    },
    {
      "entropy": 10.663964080810548,
      "epoch": 0.06875805758487323,
      "grad_norm": 1.8671875,
      "learning_rate": 3.95e-05,
      "loss": 9.649,
      "mean_token_accuracy": 0.060528215020895004,
      "num_tokens": 146691.0,
      "step": 80
    },
    {
      "entropy": 10.653559303283691,
      "epoch": 0.0730554361839278,
      "grad_norm": 1.875,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.5699,
      "mean_token_accuracy": 0.060284944623708724,
      "num_tokens": 155792.0,
      "step": 85
    },
    {
      "entropy": 10.636913108825684,
      "epoch": 0.07735281478298238,
      "grad_norm": 1.7578125,
      "learning_rate": 4.45e-05,
      "loss": 9.5761,
      "mean_token_accuracy": 0.05633340328931809,
      "num_tokens": 166944.0,
      "step": 90
    },
    {
      "entropy": 10.626116371154785,
      "epoch": 0.08165019338203695,
      "grad_norm": 1.9375,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 9.3779,
      "mean_token_accuracy": 0.06080182269215584,
      "num_tokens": 175303.0,
      "step": 95
    },
    {
      "entropy": 10.605600547790527,
      "epoch": 0.08594757198109153,
      "grad_norm": 1.78125,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 9.3452,
      "mean_token_accuracy": 0.05839457102119923,
      "num_tokens": 184708.0,
      "step": 100
    },
    {
      "entropy": 10.575385189056396,
      "epoch": 0.09024495058014612,
      "grad_norm": 1.75,
      "learning_rate": 5.2e-05,
      "loss": 9.2591,
      "mean_token_accuracy": 0.06349388286471366,
      "num_tokens": 193835.0,
      "step": 105
    },
    {
      "entropy": 10.532122993469239,
      "epoch": 0.09454232917920069,
      "grad_norm": 1.8515625,
      "learning_rate": 5.45e-05,
      "loss": 9.1605,
      "mean_token_accuracy": 0.07098993584513665,
      "num_tokens": 203344.0,
      "step": 110
    },
    {
      "entropy": 10.480678367614747,
      "epoch": 0.09883970777825526,
      "grad_norm": 1.5703125,
      "learning_rate": 5.7e-05,
      "loss": 9.1509,
      "mean_token_accuracy": 0.05849636979401111,
      "num_tokens": 213048.0,
      "step": 115
    },
    {
      "entropy": 10.444736099243164,
      "epoch": 0.10313708637730984,
      "grad_norm": 1.5546875,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.9602,
      "mean_token_accuracy": 0.07322230078279972,
      "num_tokens": 221784.0,
      "step": 120
    },
    {
      "entropy": 10.375800514221192,
      "epoch": 0.10743446497636441,
      "grad_norm": 1.5234375,
      "learning_rate": 6.2e-05,
      "loss": 8.8772,
      "mean_token_accuracy": 0.07452203780412674,
      "num_tokens": 230971.0,
      "step": 125
    },
    {
      "entropy": 10.297348403930664,
      "epoch": 0.11173184357541899,
      "grad_norm": 1.4453125,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.7682,
      "mean_token_accuracy": 0.07423729747533798,
      "num_tokens": 240524.0,
      "step": 130
    },
    {
      "entropy": 10.169759750366211,
      "epoch": 0.11602922217447358,
      "grad_norm": 1.421875,
      "learning_rate": 6.7e-05,
      "loss": 8.6555,
      "mean_token_accuracy": 0.07146260924637318,
      "num_tokens": 249220.0,
      "step": 135
    },
    {
      "entropy": 10.120108795166015,
      "epoch": 0.12032660077352815,
      "grad_norm": 1.34375,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.6406,
      "mean_token_accuracy": 0.0681857068091631,
      "num_tokens": 258934.0,
      "step": 140
    },
    {
      "entropy": 9.981133079528808,
      "epoch": 0.12462397937258272,
      "grad_norm": 1.4765625,
      "learning_rate": 7.2e-05,
      "loss": 8.5023,
      "mean_token_accuracy": 0.06928090676665306,
      "num_tokens": 267680.0,
      "step": 145
    },
    {
      "entropy": 9.886031436920167,
      "epoch": 0.1289213579716373,
      "grad_norm": 1.234375,
      "learning_rate": 7.45e-05,
      "loss": 8.4198,
      "mean_token_accuracy": 0.07501428201794624,
      "num_tokens": 276227.0,
      "step": 150
    },
    {
      "entropy": 9.753608226776123,
      "epoch": 0.1332187365706919,
      "grad_norm": 1.1875,
      "learning_rate": 7.7e-05,
      "loss": 8.2874,
      "mean_token_accuracy": 0.07755625247955322,
      "num_tokens": 286342.0,
      "step": 155
    },
    {
      "entropy": 9.609795665740966,
      "epoch": 0.13751611516974646,
      "grad_norm": 1.0,
      "learning_rate": 7.950000000000001e-05,
      "loss": 8.2601,
      "mean_token_accuracy": 0.06969402097165585,
      "num_tokens": 294994.0,
      "step": 160
    },
    {
      "entropy": 9.492193508148194,
      "epoch": 0.14181349376880104,
      "grad_norm": 0.94140625,
      "learning_rate": 8.2e-05,
      "loss": 8.1727,
      "mean_token_accuracy": 0.07241848222911358,
      "num_tokens": 303882.0,
      "step": 165
    },
    {
      "entropy": 9.254251098632812,
      "epoch": 0.1461108723678556,
      "grad_norm": 0.86328125,
      "learning_rate": 8.450000000000001e-05,
      "loss": 8.0965,
      "mean_token_accuracy": 0.07132276222109794,
      "num_tokens": 312515.0,
      "step": 170
    },
    {
      "entropy": 9.1218976020813,
      "epoch": 0.15040825096691018,
      "grad_norm": 0.94140625,
      "learning_rate": 8.7e-05,
      "loss": 8.0532,
      "mean_token_accuracy": 0.06978616863489151,
      "num_tokens": 320801.0,
      "step": 175
    },
    {
      "entropy": 8.900997734069824,
      "epoch": 0.15470562956596476,
      "grad_norm": 0.78125,
      "learning_rate": 8.95e-05,
      "loss": 8.0224,
      "mean_token_accuracy": 0.07444634772837162,
      "num_tokens": 329382.0,
      "step": 180
    },
    {
      "entropy": 8.709702205657958,
      "epoch": 0.15900300816501933,
      "grad_norm": 0.90234375,
      "learning_rate": 9.2e-05,
      "loss": 7.9836,
      "mean_token_accuracy": 0.06757781356573105,
      "num_tokens": 337894.0,
      "step": 185
    },
    {
      "entropy": 8.654694271087646,
      "epoch": 0.1633003867640739,
      "grad_norm": 0.8046875,
      "learning_rate": 9.45e-05,
      "loss": 7.9562,
      "mean_token_accuracy": 0.06648631133139134,
      "num_tokens": 346380.0,
      "step": 190
    },
    {
      "entropy": 8.521081924438477,
      "epoch": 0.16759776536312848,
      "grad_norm": 0.65625,
      "learning_rate": 9.7e-05,
      "loss": 7.9991,
      "mean_token_accuracy": 0.06590078994631768,
      "num_tokens": 356305.0,
      "step": 195
    },
    {
      "entropy": 8.442760276794434,
      "epoch": 0.17189514396218306,
      "grad_norm": 1.0390625,
      "learning_rate": 9.95e-05,
      "loss": 7.9449,
      "mean_token_accuracy": 0.06936337985098362,
      "num_tokens": 364899.0,
      "step": 200
    },
    {
      "entropy": 8.41897268295288,
      "epoch": 0.17619252256123766,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000102,
      "loss": 7.9196,
      "mean_token_accuracy": 0.0691080518066883,
      "num_tokens": 373663.0,
      "step": 205
    },
    {
      "entropy": 8.352293586730957,
      "epoch": 0.18048990116029223,
      "grad_norm": 0.83203125,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.8931,
      "mean_token_accuracy": 0.07045749872922898,
      "num_tokens": 382730.0,
      "step": 210
    },
    {
      "entropy": 8.23891839981079,
      "epoch": 0.1847872797593468,
      "grad_norm": 0.67578125,
      "learning_rate": 0.000107,
      "loss": 7.9378,
      "mean_token_accuracy": 0.06828119307756424,
      "num_tokens": 392676.0,
      "step": 215
    },
    {
      "entropy": 8.277576446533203,
      "epoch": 0.18908465835840138,
      "grad_norm": 0.828125,
      "learning_rate": 0.0001095,
      "loss": 7.9179,
      "mean_token_accuracy": 0.07761286348104476,
      "num_tokens": 401050.0,
      "step": 220
    },
    {
      "entropy": 8.313580322265626,
      "epoch": 0.19338203695745596,
      "grad_norm": 0.77734375,
      "learning_rate": 0.000112,
      "loss": 7.8686,
      "mean_token_accuracy": 0.0736297857016325,
      "num_tokens": 410009.0,
      "step": 225
    },
    {
      "entropy": 8.305255031585693,
      "epoch": 0.19767941555651053,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0001145,
      "loss": 7.8585,
      "mean_token_accuracy": 0.07423750944435596,
      "num_tokens": 419302.0,
      "step": 230
    },
    {
      "entropy": 8.293313884735108,
      "epoch": 0.2019767941555651,
      "grad_norm": 0.828125,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.7711,
      "mean_token_accuracy": 0.08085694760084153,
      "num_tokens": 427296.0,
      "step": 235
    },
    {
      "entropy": 8.249619960784912,
      "epoch": 0.20627417275461968,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.8993,
      "mean_token_accuracy": 0.07239912934601307,
      "num_tokens": 436368.0,
      "step": 240
    },
    {
      "entropy": 8.27509069442749,
      "epoch": 0.21057155135367425,
      "grad_norm": 0.8125,
      "learning_rate": 0.000122,
      "loss": 7.8229,
      "mean_token_accuracy": 0.0778175387531519,
      "num_tokens": 445535.0,
      "step": 245
    },
    {
      "entropy": 8.245413112640382,
      "epoch": 0.21486892995272883,
      "grad_norm": 0.9375,
      "learning_rate": 0.0001245,
      "loss": 7.7971,
      "mean_token_accuracy": 0.07436932176351548,
      "num_tokens": 454769.0,
      "step": 250
    },
    {
      "entropy": 8.240999603271485,
      "epoch": 0.2191663085517834,
      "grad_norm": 0.84375,
      "learning_rate": 0.000127,
      "loss": 7.8138,
      "mean_token_accuracy": 0.07600181177258492,
      "num_tokens": 463975.0,
      "step": 255
    },
    {
      "entropy": 8.211737537384034,
      "epoch": 0.22346368715083798,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001295,
      "loss": 7.8027,
      "mean_token_accuracy": 0.08170078098773956,
      "num_tokens": 472899.0,
      "step": 260
    },
    {
      "entropy": 8.172839450836182,
      "epoch": 0.22776106574989258,
      "grad_norm": 1.09375,
      "learning_rate": 0.000132,
      "loss": 7.8171,
      "mean_token_accuracy": 0.07734721302986144,
      "num_tokens": 481556.0,
      "step": 265
    },
    {
      "entropy": 8.150885677337646,
      "epoch": 0.23205844434894715,
      "grad_norm": 1.0625,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.7804,
      "mean_token_accuracy": 0.07963913902640343,
      "num_tokens": 490253.0,
      "step": 270
    },
    {
      "entropy": 8.222941017150879,
      "epoch": 0.23635582294800173,
      "grad_norm": 1.140625,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.785,
      "mean_token_accuracy": 0.07519972808659077,
      "num_tokens": 498444.0,
      "step": 275
    },
    {
      "entropy": 8.142861270904541,
      "epoch": 0.2406532015470563,
      "grad_norm": 0.73046875,
      "learning_rate": 0.0001395,
      "loss": 7.7267,
      "mean_token_accuracy": 0.08277813419699669,
      "num_tokens": 508330.0,
      "step": 280
    },
    {
      "entropy": 8.140101432800293,
      "epoch": 0.24495058014611087,
      "grad_norm": 0.859375,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.8112,
      "mean_token_accuracy": 0.0772799663245678,
      "num_tokens": 517900.0,
      "step": 285
    },
    {
      "entropy": 8.134088563919068,
      "epoch": 0.24924795874516545,
      "grad_norm": 1.125,
      "learning_rate": 0.0001445,
      "loss": 7.7764,
      "mean_token_accuracy": 0.0777011513710022,
      "num_tokens": 527808.0,
      "step": 290
    },
    {
      "entropy": 8.083810663223266,
      "epoch": 0.25354533734422,
      "grad_norm": 0.81640625,
      "learning_rate": 0.000147,
      "loss": 7.7652,
      "mean_token_accuracy": 0.07280438467860222,
      "num_tokens": 536931.0,
      "step": 295
    },
    {
      "entropy": 8.169271469116211,
      "epoch": 0.2578427159432746,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001495,
      "loss": 7.7957,
      "mean_token_accuracy": 0.0796648234128952,
      "num_tokens": 545758.0,
      "step": 300
    },
    {
      "entropy": 8.125834083557129,
      "epoch": 0.26214009454232917,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000152,
      "loss": 7.7996,
      "mean_token_accuracy": 0.07962435148656369,
      "num_tokens": 555165.0,
      "step": 305
    },
    {
      "entropy": 8.13117265701294,
      "epoch": 0.2664374731413838,
      "grad_norm": 0.859375,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.7089,
      "mean_token_accuracy": 0.07958239540457726,
      "num_tokens": 564719.0,
      "step": 310
    },
    {
      "entropy": 8.057837867736817,
      "epoch": 0.2707348517404383,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000157,
      "loss": 7.6928,
      "mean_token_accuracy": 0.0828514739871025,
      "num_tokens": 573572.0,
      "step": 315
    },
    {
      "entropy": 8.223217296600343,
      "epoch": 0.2750322303394929,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001595,
      "loss": 7.8692,
      "mean_token_accuracy": 0.07304716818034648,
      "num_tokens": 581497.0,
      "step": 320
    },
    {
      "entropy": 8.053431844711303,
      "epoch": 0.27932960893854747,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000162,
      "loss": 7.7017,
      "mean_token_accuracy": 0.08142868839204312,
      "num_tokens": 591107.0,
      "step": 325
    },
    {
      "entropy": 8.063822364807129,
      "epoch": 0.28362698753760207,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.7406,
      "mean_token_accuracy": 0.0793286319822073,
      "num_tokens": 600241.0,
      "step": 330
    },
    {
      "entropy": 8.13500518798828,
      "epoch": 0.2879243661366566,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.6735,
      "mean_token_accuracy": 0.07961011826992034,
      "num_tokens": 608697.0,
      "step": 335
    },
    {
      "entropy": 8.040066146850586,
      "epoch": 0.2922217447357112,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.6584,
      "mean_token_accuracy": 0.08232032433152199,
      "num_tokens": 617275.0,
      "step": 340
    },
    {
      "entropy": 8.135028553009032,
      "epoch": 0.29651912333476577,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.7489,
      "mean_token_accuracy": 0.08271522969007492,
      "num_tokens": 626644.0,
      "step": 345
    },
    {
      "entropy": 8.011529541015625,
      "epoch": 0.30081650193382037,
      "grad_norm": 0.890625,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.6696,
      "mean_token_accuracy": 0.0852140061557293,
      "num_tokens": 635110.0,
      "step": 350
    },
    {
      "entropy": 8.075648546218872,
      "epoch": 0.30511388053287497,
      "grad_norm": 1.046875,
      "learning_rate": 0.000177,
      "loss": 7.8072,
      "mean_token_accuracy": 0.07943284660577773,
      "num_tokens": 644746.0,
      "step": 355
    },
    {
      "entropy": 8.112467098236085,
      "epoch": 0.3094112591319295,
      "grad_norm": 1.375,
      "learning_rate": 0.0001795,
      "loss": 7.6965,
      "mean_token_accuracy": 0.08117582276463509,
      "num_tokens": 654281.0,
      "step": 360
    },
    {
      "entropy": 8.078112173080445,
      "epoch": 0.3137086377309841,
      "grad_norm": 1.375,
      "learning_rate": 0.000182,
      "loss": 7.6883,
      "mean_token_accuracy": 0.08609185442328453,
      "num_tokens": 663174.0,
      "step": 365
    },
    {
      "entropy": 7.986557722091675,
      "epoch": 0.31800601633003867,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001845,
      "loss": 7.6259,
      "mean_token_accuracy": 0.08559975251555443,
      "num_tokens": 672178.0,
      "step": 370
    },
    {
      "entropy": 8.04137225151062,
      "epoch": 0.32230339492909327,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000187,
      "loss": 7.6371,
      "mean_token_accuracy": 0.08901203349232674,
      "num_tokens": 681323.0,
      "step": 375
    },
    {
      "entropy": 8.031937456130981,
      "epoch": 0.3266007735281478,
      "grad_norm": 1.03125,
      "learning_rate": 0.0001895,
      "loss": 7.5447,
      "mean_token_accuracy": 0.08979653343558311,
      "num_tokens": 690461.0,
      "step": 380
    },
    {
      "entropy": 7.969046926498413,
      "epoch": 0.3308981521272024,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000192,
      "loss": 7.61,
      "mean_token_accuracy": 0.07953116819262504,
      "num_tokens": 699199.0,
      "step": 385
    },
    {
      "entropy": 8.007894515991211,
      "epoch": 0.33519553072625696,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0001945,
      "loss": 7.5729,
      "mean_token_accuracy": 0.08715594932436943,
      "num_tokens": 707949.0,
      "step": 390
    },
    {
      "entropy": 8.037422609329223,
      "epoch": 0.33949290932531156,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.6302,
      "mean_token_accuracy": 0.08520551696419716,
      "num_tokens": 715752.0,
      "step": 395
    },
    {
      "entropy": 8.013258504867554,
      "epoch": 0.3437902879243661,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.5717,
      "mean_token_accuracy": 0.08992563635110855,
      "num_tokens": 724416.0,
      "step": 400
    },
    {
      "entropy": 8.055727624893189,
      "epoch": 0.3480876665234207,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000202,
      "loss": 7.5987,
      "mean_token_accuracy": 0.08299638405442238,
      "num_tokens": 733116.0,
      "step": 405
    },
    {
      "entropy": 7.958417224884033,
      "epoch": 0.3523850451224753,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00020449999999999998,
      "loss": 7.5847,
      "mean_token_accuracy": 0.08687192872166634,
      "num_tokens": 742093.0,
      "step": 410
    },
    {
      "entropy": 7.976228284835815,
      "epoch": 0.35668242372152986,
      "grad_norm": 0.953125,
      "learning_rate": 0.000207,
      "loss": 7.5894,
      "mean_token_accuracy": 0.0836369551718235,
      "num_tokens": 750402.0,
      "step": 415
    },
    {
      "entropy": 7.9206877708435055,
      "epoch": 0.36097980232058446,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002095,
      "loss": 7.5472,
      "mean_token_accuracy": 0.08791278377175331,
      "num_tokens": 760961.0,
      "step": 420
    },
    {
      "entropy": 7.975084543228149,
      "epoch": 0.365277180919639,
      "grad_norm": 1.03125,
      "learning_rate": 0.000212,
      "loss": 7.6033,
      "mean_token_accuracy": 0.08895924240350724,
      "num_tokens": 770554.0,
      "step": 425
    },
    {
      "entropy": 7.908215951919556,
      "epoch": 0.3695745595186936,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002145,
      "loss": 7.5822,
      "mean_token_accuracy": 0.08788935765624047,
      "num_tokens": 779172.0,
      "step": 430
    },
    {
      "entropy": 8.003646516799927,
      "epoch": 0.37387193811774816,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021700000000000002,
      "loss": 7.6269,
      "mean_token_accuracy": 0.0811423234641552,
      "num_tokens": 788040.0,
      "step": 435
    },
    {
      "entropy": 8.020764684677124,
      "epoch": 0.37816931671680276,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002195,
      "loss": 7.5218,
      "mean_token_accuracy": 0.08682032898068429,
      "num_tokens": 796786.0,
      "step": 440
    },
    {
      "entropy": 7.909510374069214,
      "epoch": 0.3824666953158573,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000222,
      "loss": 7.55,
      "mean_token_accuracy": 0.08918910399079323,
      "num_tokens": 805520.0,
      "step": 445
    },
    {
      "entropy": 7.937164402008056,
      "epoch": 0.3867640739149119,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002245,
      "loss": 7.4594,
      "mean_token_accuracy": 0.08978109285235406,
      "num_tokens": 814939.0,
      "step": 450
    },
    {
      "entropy": 7.91705961227417,
      "epoch": 0.39106145251396646,
      "grad_norm": 1.109375,
      "learning_rate": 0.00022700000000000002,
      "loss": 7.4827,
      "mean_token_accuracy": 0.08797861486673356,
      "num_tokens": 823862.0,
      "step": 455
    },
    {
      "entropy": 7.884686851501465,
      "epoch": 0.39535883111302106,
      "grad_norm": 1.078125,
      "learning_rate": 0.00022950000000000002,
      "loss": 7.499,
      "mean_token_accuracy": 0.08911096453666686,
      "num_tokens": 832820.0,
      "step": 460
    },
    {
      "entropy": 7.961242198944092,
      "epoch": 0.39965620971207566,
      "grad_norm": 1.375,
      "learning_rate": 0.00023200000000000003,
      "loss": 7.463,
      "mean_token_accuracy": 0.08996951058506966,
      "num_tokens": 841538.0,
      "step": 465
    },
    {
      "entropy": 7.896987247467041,
      "epoch": 0.4039535883111302,
      "grad_norm": 0.96875,
      "learning_rate": 0.00023449999999999998,
      "loss": 7.5646,
      "mean_token_accuracy": 0.0861456498503685,
      "num_tokens": 851123.0,
      "step": 470
    },
    {
      "entropy": 8.01444125175476,
      "epoch": 0.4082509669101848,
      "grad_norm": 1.234375,
      "learning_rate": 0.000237,
      "loss": 7.5439,
      "mean_token_accuracy": 0.0893861047923565,
      "num_tokens": 860357.0,
      "step": 475
    },
    {
      "entropy": 7.83814754486084,
      "epoch": 0.41254834550923936,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002395,
      "loss": 7.458,
      "mean_token_accuracy": 0.09221402034163476,
      "num_tokens": 869980.0,
      "step": 480
    },
    {
      "entropy": 7.830886650085449,
      "epoch": 0.41684572410829396,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000242,
      "loss": 7.4055,
      "mean_token_accuracy": 0.08723795711994171,
      "num_tokens": 878250.0,
      "step": 485
    },
    {
      "entropy": 7.888704490661621,
      "epoch": 0.4211431027073485,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0002445,
      "loss": 7.5116,
      "mean_token_accuracy": 0.08683100938796998,
      "num_tokens": 887624.0,
      "step": 490
    },
    {
      "entropy": 7.814441156387329,
      "epoch": 0.4254404813064031,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000247,
      "loss": 7.4596,
      "mean_token_accuracy": 0.08961054012179374,
      "num_tokens": 897120.0,
      "step": 495
    },
    {
      "entropy": 7.8445289611816404,
      "epoch": 0.42973785990545765,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0002495,
      "loss": 7.4306,
      "mean_token_accuracy": 0.09552830010652542,
      "num_tokens": 906215.0,
      "step": 500
    },
    {
      "epoch": 0.42973785990545765,
      "eval_entropy": 7.63128979463835,
      "eval_loss": 7.479809284210205,
      "eval_mean_token_accuracy": 0.0912898702228123,
      "eval_num_tokens": 906215.0,
      "eval_runtime": 2.0631,
      "eval_samples_per_second": 1720.253,
      "eval_steps_per_second": 215.213,
      "step": 500
    },
    {
      "entropy": 7.809079313278199,
      "epoch": 0.43403523850451226,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000252,
      "loss": 7.4118,
      "mean_token_accuracy": 0.08965951427817345,
      "num_tokens": 915181.0,
      "step": 505
    },
    {
      "entropy": 7.847093725204468,
      "epoch": 0.4383326171035668,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002545,
      "loss": 7.429,
      "mean_token_accuracy": 0.09137469455599785,
      "num_tokens": 924377.0,
      "step": 510
    },
    {
      "entropy": 7.866995763778687,
      "epoch": 0.4426299957026214,
      "grad_norm": 1.25,
      "learning_rate": 0.000257,
      "loss": 7.5053,
      "mean_token_accuracy": 0.09116312190890312,
      "num_tokens": 933114.0,
      "step": 515
    },
    {
      "entropy": 7.865507459640503,
      "epoch": 0.44692737430167595,
      "grad_norm": 0.984375,
      "learning_rate": 0.0002595,
      "loss": 7.4689,
      "mean_token_accuracy": 0.08555961102247238,
      "num_tokens": 943306.0,
      "step": 520
    },
    {
      "entropy": 7.781367206573487,
      "epoch": 0.45122475290073055,
      "grad_norm": 1.0,
      "learning_rate": 0.000262,
      "loss": 7.3577,
      "mean_token_accuracy": 0.09929761290550232,
      "num_tokens": 951515.0,
      "step": 525
    },
    {
      "entropy": 7.766856384277344,
      "epoch": 0.45552213149978515,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026450000000000003,
      "loss": 7.4694,
      "mean_token_accuracy": 0.09030876457691192,
      "num_tokens": 962686.0,
      "step": 530
    },
    {
      "entropy": 7.901469707489014,
      "epoch": 0.4598195100988397,
      "grad_norm": 1.046875,
      "learning_rate": 0.00026700000000000004,
      "loss": 7.4753,
      "mean_token_accuracy": 0.09142492190003396,
      "num_tokens": 972136.0,
      "step": 535
    },
    {
      "entropy": 7.716486024856567,
      "epoch": 0.4641168886978943,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026950000000000005,
      "loss": 7.3638,
      "mean_token_accuracy": 0.09268842414021491,
      "num_tokens": 981301.0,
      "step": 540
    },
    {
      "entropy": 7.767406940460205,
      "epoch": 0.46841426729694885,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027200000000000005,
      "loss": 7.3949,
      "mean_token_accuracy": 0.09462741911411285,
      "num_tokens": 990360.0,
      "step": 545
    },
    {
      "entropy": 7.815031290054321,
      "epoch": 0.47271164589600345,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002745,
      "loss": 7.4369,
      "mean_token_accuracy": 0.09093201607465744,
      "num_tokens": 999415.0,
      "step": 550
    },
    {
      "entropy": 7.7446294784545895,
      "epoch": 0.477009024495058,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000277,
      "loss": 7.3662,
      "mean_token_accuracy": 0.09483416303992272,
      "num_tokens": 1008762.0,
      "step": 555
    },
    {
      "entropy": 7.768563890457154,
      "epoch": 0.4813064030941126,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002795,
      "loss": 7.3624,
      "mean_token_accuracy": 0.09757203981280327,
      "num_tokens": 1017704.0,
      "step": 560
    },
    {
      "entropy": 7.691549444198609,
      "epoch": 0.48560378169316715,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028199999999999997,
      "loss": 7.3155,
      "mean_token_accuracy": 0.0982676163315773,
      "num_tokens": 1026251.0,
      "step": 565
    },
    {
      "entropy": 7.725941944122314,
      "epoch": 0.48990116029222175,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002845,
      "loss": 7.2996,
      "mean_token_accuracy": 0.09966658651828766,
      "num_tokens": 1036191.0,
      "step": 570
    },
    {
      "entropy": 7.717142486572266,
      "epoch": 0.4941985388912763,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000287,
      "loss": 7.3328,
      "mean_token_accuracy": 0.09679939597845078,
      "num_tokens": 1044936.0,
      "step": 575
    },
    {
      "entropy": 7.726046752929688,
      "epoch": 0.4984959174903309,
      "grad_norm": 1.6875,
      "learning_rate": 0.0002895,
      "loss": 7.3382,
      "mean_token_accuracy": 0.09426090195775032,
      "num_tokens": 1053683.0,
      "step": 580
    },
    {
      "entropy": 7.633957862854004,
      "epoch": 0.5027932960893855,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000292,
      "loss": 7.3198,
      "mean_token_accuracy": 0.09904149472713471,
      "num_tokens": 1062932.0,
      "step": 585
    },
    {
      "entropy": 7.770825815200806,
      "epoch": 0.50709067468844,
      "grad_norm": 1.25,
      "learning_rate": 0.0002945,
      "loss": 7.3476,
      "mean_token_accuracy": 0.09258887320756912,
      "num_tokens": 1072313.0,
      "step": 590
    },
    {
      "entropy": 7.691164636611939,
      "epoch": 0.5113880532874946,
      "grad_norm": 1.375,
      "learning_rate": 0.000297,
      "loss": 7.3104,
      "mean_token_accuracy": 0.09508624523878098,
      "num_tokens": 1081675.0,
      "step": 595
    },
    {
      "entropy": 7.71315016746521,
      "epoch": 0.5156854318865493,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002995,
      "loss": 7.3272,
      "mean_token_accuracy": 0.09484435245394707,
      "num_tokens": 1091541.0,
      "step": 600
    },
    {
      "entropy": 7.773202800750733,
      "epoch": 0.5199828104856038,
      "grad_norm": 1.078125,
      "learning_rate": 0.000302,
      "loss": 7.3362,
      "mean_token_accuracy": 0.10049614533782006,
      "num_tokens": 1100724.0,
      "step": 605
    },
    {
      "entropy": 7.529764127731323,
      "epoch": 0.5242801890846583,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003045,
      "loss": 7.2799,
      "mean_token_accuracy": 0.10042180344462395,
      "num_tokens": 1108869.0,
      "step": 610
    },
    {
      "entropy": 7.751347589492798,
      "epoch": 0.5285775676837129,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000307,
      "loss": 7.3109,
      "mean_token_accuracy": 0.09996724799275399,
      "num_tokens": 1117314.0,
      "step": 615
    },
    {
      "entropy": 7.551737356185913,
      "epoch": 0.5328749462827675,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003095,
      "loss": 7.2409,
      "mean_token_accuracy": 0.09862213358283042,
      "num_tokens": 1126786.0,
      "step": 620
    },
    {
      "entropy": 7.609012699127197,
      "epoch": 0.5371723248818221,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000312,
      "loss": 7.2349,
      "mean_token_accuracy": 0.09517790451645851,
      "num_tokens": 1136013.0,
      "step": 625
    },
    {
      "entropy": 7.7104791641235355,
      "epoch": 0.5414697034808766,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003145,
      "loss": 7.2709,
      "mean_token_accuracy": 0.10074363201856613,
      "num_tokens": 1144970.0,
      "step": 630
    },
    {
      "entropy": 7.587553977966309,
      "epoch": 0.5457670820799312,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000317,
      "loss": 7.2408,
      "mean_token_accuracy": 0.1027376800775528,
      "num_tokens": 1153810.0,
      "step": 635
    },
    {
      "entropy": 7.597120857238769,
      "epoch": 0.5500644606789858,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003195,
      "loss": 7.2171,
      "mean_token_accuracy": 0.10400616675615311,
      "num_tokens": 1162498.0,
      "step": 640
    },
    {
      "entropy": 7.627674579620361,
      "epoch": 0.5543618392780404,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000322,
      "loss": 7.3008,
      "mean_token_accuracy": 0.09583014771342277,
      "num_tokens": 1172091.0,
      "step": 645
    },
    {
      "entropy": 7.639283657073975,
      "epoch": 0.5586592178770949,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032450000000000003,
      "loss": 7.2766,
      "mean_token_accuracy": 0.09886594414710999,
      "num_tokens": 1181400.0,
      "step": 650
    },
    {
      "entropy": 7.654122257232666,
      "epoch": 0.5629565964761496,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00032700000000000003,
      "loss": 7.2462,
      "mean_token_accuracy": 0.10131431221961976,
      "num_tokens": 1189780.0,
      "step": 655
    },
    {
      "entropy": 7.497343397140503,
      "epoch": 0.5672539750752041,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032950000000000004,
      "loss": 7.2029,
      "mean_token_accuracy": 0.10030499249696731,
      "num_tokens": 1198671.0,
      "step": 660
    },
    {
      "entropy": 7.620815324783325,
      "epoch": 0.5715513536742587,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033200000000000005,
      "loss": 7.2489,
      "mean_token_accuracy": 0.0975674532353878,
      "num_tokens": 1207173.0,
      "step": 665
    },
    {
      "entropy": 7.556894588470459,
      "epoch": 0.5758487322733132,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033450000000000005,
      "loss": 7.1783,
      "mean_token_accuracy": 0.10579388439655305,
      "num_tokens": 1216387.0,
      "step": 670
    },
    {
      "entropy": 7.638325643539429,
      "epoch": 0.5801461108723679,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000337,
      "loss": 7.1487,
      "mean_token_accuracy": 0.10557905286550522,
      "num_tokens": 1224461.0,
      "step": 675
    },
    {
      "entropy": 7.557331895828247,
      "epoch": 0.5844434894714224,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003395,
      "loss": 7.2067,
      "mean_token_accuracy": 0.10470035746693611,
      "num_tokens": 1233774.0,
      "step": 680
    },
    {
      "entropy": 7.48933973312378,
      "epoch": 0.588740868070477,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000342,
      "loss": 7.173,
      "mean_token_accuracy": 0.1077287070453167,
      "num_tokens": 1242812.0,
      "step": 685
    },
    {
      "entropy": 7.698177099227905,
      "epoch": 0.5930382466695315,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034449999999999997,
      "loss": 7.2652,
      "mean_token_accuracy": 0.09988318458199501,
      "num_tokens": 1252872.0,
      "step": 690
    },
    {
      "entropy": 7.376710081100464,
      "epoch": 0.5973356252685862,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000347,
      "loss": 7.0588,
      "mean_token_accuracy": 0.10656947568058968,
      "num_tokens": 1260852.0,
      "step": 695
    },
    {
      "entropy": 7.442113256454467,
      "epoch": 0.6016330038676407,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003495,
      "loss": 7.1521,
      "mean_token_accuracy": 0.10797703564167023,
      "num_tokens": 1268925.0,
      "step": 700
    },
    {
      "entropy": 7.575676298141479,
      "epoch": 0.6059303824666953,
      "grad_norm": 0.921875,
      "learning_rate": 0.000352,
      "loss": 7.2053,
      "mean_token_accuracy": 0.1014741227030754,
      "num_tokens": 1278994.0,
      "step": 705
    },
    {
      "entropy": 7.616156387329101,
      "epoch": 0.6102277610657499,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003545,
      "loss": 7.1986,
      "mean_token_accuracy": 0.1000370778143406,
      "num_tokens": 1287698.0,
      "step": 710
    },
    {
      "entropy": 7.500000095367431,
      "epoch": 0.6145251396648045,
      "grad_norm": 1.234375,
      "learning_rate": 0.000357,
      "loss": 7.162,
      "mean_token_accuracy": 0.10414053946733474,
      "num_tokens": 1297475.0,
      "step": 715
    },
    {
      "entropy": 7.507882452011108,
      "epoch": 0.618822518263859,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003595,
      "loss": 7.1852,
      "mean_token_accuracy": 0.10390809029340745,
      "num_tokens": 1306836.0,
      "step": 720
    },
    {
      "entropy": 7.507648563385009,
      "epoch": 0.6231198968629136,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000362,
      "loss": 7.2148,
      "mean_token_accuracy": 0.10065656825900078,
      "num_tokens": 1315872.0,
      "step": 725
    },
    {
      "entropy": 7.490116739273072,
      "epoch": 0.6274172754619682,
      "grad_norm": 1.5,
      "learning_rate": 0.0003645,
      "loss": 7.163,
      "mean_token_accuracy": 0.10862415507435799,
      "num_tokens": 1324624.0,
      "step": 730
    },
    {
      "entropy": 7.495247173309326,
      "epoch": 0.6317146540610228,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000367,
      "loss": 7.1357,
      "mean_token_accuracy": 0.11479571759700775,
      "num_tokens": 1333058.0,
      "step": 735
    },
    {
      "entropy": 7.333670282363892,
      "epoch": 0.6360120326600773,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003695,
      "loss": 7.0924,
      "mean_token_accuracy": 0.10395059809088707,
      "num_tokens": 1342376.0,
      "step": 740
    },
    {
      "entropy": 7.5272722244262695,
      "epoch": 0.6403094112591319,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000372,
      "loss": 7.1365,
      "mean_token_accuracy": 0.09796565026044846,
      "num_tokens": 1351386.0,
      "step": 745
    },
    {
      "entropy": 7.43572735786438,
      "epoch": 0.6446067898581865,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003745,
      "loss": 7.0538,
      "mean_token_accuracy": 0.11305336132645608,
      "num_tokens": 1358958.0,
      "step": 750
    },
    {
      "entropy": 7.471644115447998,
      "epoch": 0.6489041684572411,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000377,
      "loss": 7.1846,
      "mean_token_accuracy": 0.10423881933093071,
      "num_tokens": 1368599.0,
      "step": 755
    },
    {
      "entropy": 7.529570055007935,
      "epoch": 0.6532015470562956,
      "grad_norm": 1.921875,
      "learning_rate": 0.0003795,
      "loss": 7.2099,
      "mean_token_accuracy": 0.10423211976885796,
      "num_tokens": 1378529.0,
      "step": 760
    },
    {
      "entropy": 7.447458553314209,
      "epoch": 0.6574989256553503,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000382,
      "loss": 6.995,
      "mean_token_accuracy": 0.11015288382768632,
      "num_tokens": 1386993.0,
      "step": 765
    },
    {
      "entropy": 7.396367502212525,
      "epoch": 0.6617963042544048,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003845,
      "loss": 7.0909,
      "mean_token_accuracy": 0.11230141893029214,
      "num_tokens": 1395790.0,
      "step": 770
    },
    {
      "entropy": 7.3504656791687015,
      "epoch": 0.6660936828534594,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038700000000000003,
      "loss": 7.1044,
      "mean_token_accuracy": 0.11002744212746621,
      "num_tokens": 1405587.0,
      "step": 775
    },
    {
      "entropy": 7.394146966934204,
      "epoch": 0.6703910614525139,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038950000000000003,
      "loss": 7.0762,
      "mean_token_accuracy": 0.11049489751458168,
      "num_tokens": 1414478.0,
      "step": 780
    },
    {
      "entropy": 7.4956427097320555,
      "epoch": 0.6746884400515686,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00039200000000000004,
      "loss": 7.1402,
      "mean_token_accuracy": 0.11158580034971237,
      "num_tokens": 1423791.0,
      "step": 785
    },
    {
      "entropy": 7.476742601394653,
      "epoch": 0.6789858186506231,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039450000000000005,
      "loss": 7.1666,
      "mean_token_accuracy": 0.10097377747297287,
      "num_tokens": 1432955.0,
      "step": 790
    },
    {
      "entropy": 7.437243223190308,
      "epoch": 0.6832831972496777,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039700000000000005,
      "loss": 7.142,
      "mean_token_accuracy": 0.10785873606801033,
      "num_tokens": 1441907.0,
      "step": 795
    },
    {
      "entropy": 7.319718980789185,
      "epoch": 0.6875805758487322,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003995,
      "loss": 7.009,
      "mean_token_accuracy": 0.11284941285848618,
      "num_tokens": 1451062.0,
      "step": 800
    },
    {
      "entropy": 7.428938245773315,
      "epoch": 0.6918779544477869,
      "grad_norm": 1.125,
      "learning_rate": 0.000402,
      "loss": 7.0856,
      "mean_token_accuracy": 0.1089414782822132,
      "num_tokens": 1460132.0,
      "step": 805
    },
    {
      "entropy": 7.379511928558349,
      "epoch": 0.6961753330468414,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004045,
      "loss": 6.9721,
      "mean_token_accuracy": 0.11648677214980126,
      "num_tokens": 1469582.0,
      "step": 810
    },
    {
      "entropy": 7.305799245834351,
      "epoch": 0.700472711645896,
      "grad_norm": 1.359375,
      "learning_rate": 0.00040699999999999997,
      "loss": 7.0538,
      "mean_token_accuracy": 0.10682568848133087,
      "num_tokens": 1479053.0,
      "step": 815
    },
    {
      "entropy": 7.325813055038452,
      "epoch": 0.7047700902449506,
      "grad_norm": 1.125,
      "learning_rate": 0.0004095,
      "loss": 7.0643,
      "mean_token_accuracy": 0.10892463400959969,
      "num_tokens": 1488189.0,
      "step": 820
    },
    {
      "entropy": 7.319435358047485,
      "epoch": 0.7090674688440052,
      "grad_norm": 1.25,
      "learning_rate": 0.000412,
      "loss": 6.9276,
      "mean_token_accuracy": 0.11442206054925919,
      "num_tokens": 1497324.0,
      "step": 825
    },
    {
      "entropy": 7.347770023345947,
      "epoch": 0.7133648474430597,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004145,
      "loss": 7.0315,
      "mean_token_accuracy": 0.11255533620715141,
      "num_tokens": 1506543.0,
      "step": 830
    },
    {
      "entropy": 7.312428903579712,
      "epoch": 0.7176622260421143,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000417,
      "loss": 7.022,
      "mean_token_accuracy": 0.11886205449700356,
      "num_tokens": 1516737.0,
      "step": 835
    },
    {
      "entropy": 7.217544221878052,
      "epoch": 0.7219596046411689,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004195,
      "loss": 6.8992,
      "mean_token_accuracy": 0.1128052294254303,
      "num_tokens": 1525561.0,
      "step": 840
    },
    {
      "entropy": 7.309549617767334,
      "epoch": 0.7262569832402235,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000422,
      "loss": 6.9283,
      "mean_token_accuracy": 0.11749697029590607,
      "num_tokens": 1533323.0,
      "step": 845
    },
    {
      "entropy": 7.349371337890625,
      "epoch": 0.730554361839278,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004245,
      "loss": 6.9859,
      "mean_token_accuracy": 0.11753521487116814,
      "num_tokens": 1542632.0,
      "step": 850
    },
    {
      "entropy": 7.286138200759888,
      "epoch": 0.7348517404383326,
      "grad_norm": 1.390625,
      "learning_rate": 0.000427,
      "loss": 6.9421,
      "mean_token_accuracy": 0.11514310464262963,
      "num_tokens": 1551236.0,
      "step": 855
    },
    {
      "entropy": 7.193848180770874,
      "epoch": 0.7391491190373872,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004295,
      "loss": 6.9406,
      "mean_token_accuracy": 0.11694173738360406,
      "num_tokens": 1559674.0,
      "step": 860
    },
    {
      "entropy": 7.279930353164673,
      "epoch": 0.7434464976364418,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000432,
      "loss": 7.0219,
      "mean_token_accuracy": 0.11416142880916595,
      "num_tokens": 1569481.0,
      "step": 865
    },
    {
      "entropy": 7.322668170928955,
      "epoch": 0.7477438762354963,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004345,
      "loss": 7.0371,
      "mean_token_accuracy": 0.11332832500338555,
      "num_tokens": 1578488.0,
      "step": 870
    },
    {
      "entropy": 7.236435031890869,
      "epoch": 0.752041254834551,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000437,
      "loss": 6.9403,
      "mean_token_accuracy": 0.11505360826849938,
      "num_tokens": 1586675.0,
      "step": 875
    },
    {
      "entropy": 7.316900205612183,
      "epoch": 0.7563386334336055,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004395,
      "loss": 7.048,
      "mean_token_accuracy": 0.10665819644927979,
      "num_tokens": 1595411.0,
      "step": 880
    },
    {
      "entropy": 7.344276475906372,
      "epoch": 0.7606360120326601,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000442,
      "loss": 7.0296,
      "mean_token_accuracy": 0.11478703245520591,
      "num_tokens": 1604046.0,
      "step": 885
    },
    {
      "entropy": 7.300757169723511,
      "epoch": 0.7649333906317146,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004445,
      "loss": 7.0139,
      "mean_token_accuracy": 0.10667152628302574,
      "num_tokens": 1613759.0,
      "step": 890
    },
    {
      "entropy": 7.2238458633422855,
      "epoch": 0.7692307692307693,
      "grad_norm": 1.0625,
      "learning_rate": 0.000447,
      "loss": 6.9705,
      "mean_token_accuracy": 0.11665266454219818,
      "num_tokens": 1623323.0,
      "step": 895
    },
    {
      "entropy": 7.301079988479614,
      "epoch": 0.7735281478298238,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.9563,
      "mean_token_accuracy": 0.11468923836946487,
      "num_tokens": 1631727.0,
      "step": 900
    },
    {
      "entropy": 7.2134482860565186,
      "epoch": 0.7778255264288784,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.9011,
      "mean_token_accuracy": 0.11652599200606346,
      "num_tokens": 1639544.0,
      "step": 905
    },
    {
      "entropy": 7.3395082473754885,
      "epoch": 0.7821229050279329,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045450000000000004,
      "loss": 7.0207,
      "mean_token_accuracy": 0.10909055322408676,
      "num_tokens": 1648931.0,
      "step": 910
    },
    {
      "entropy": 7.182226657867432,
      "epoch": 0.7864202836269876,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.8925,
      "mean_token_accuracy": 0.11583453565835952,
      "num_tokens": 1657688.0,
      "step": 915
    },
    {
      "entropy": 7.263977575302124,
      "epoch": 0.7907176622260421,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.961,
      "mean_token_accuracy": 0.11562951728701591,
      "num_tokens": 1666879.0,
      "step": 920
    },
    {
      "entropy": 7.198910236358643,
      "epoch": 0.7950150408250967,
      "grad_norm": 1.109375,
      "learning_rate": 0.000462,
      "loss": 6.9431,
      "mean_token_accuracy": 0.11777721717953682,
      "num_tokens": 1676773.0,
      "step": 925
    },
    {
      "entropy": 7.280461740493775,
      "epoch": 0.7993124194241513,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004645,
      "loss": 7.0482,
      "mean_token_accuracy": 0.10834663063287735,
      "num_tokens": 1686144.0,
      "step": 930
    },
    {
      "entropy": 7.22996711730957,
      "epoch": 0.8036097980232059,
      "grad_norm": 1.15625,
      "learning_rate": 0.000467,
      "loss": 7.019,
      "mean_token_accuracy": 0.10943668559193612,
      "num_tokens": 1695476.0,
      "step": 935
    },
    {
      "entropy": 7.193253135681152,
      "epoch": 0.8079071766222604,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004695,
      "loss": 6.9077,
      "mean_token_accuracy": 0.11359465345740319,
      "num_tokens": 1704907.0,
      "step": 940
    },
    {
      "entropy": 7.2548192024230955,
      "epoch": 0.812204555221315,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000472,
      "loss": 6.9424,
      "mean_token_accuracy": 0.11716259941458702,
      "num_tokens": 1714564.0,
      "step": 945
    },
    {
      "entropy": 7.272028970718384,
      "epoch": 0.8165019338203696,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004745,
      "loss": 6.9794,
      "mean_token_accuracy": 0.11456828936934471,
      "num_tokens": 1725285.0,
      "step": 950
    },
    {
      "entropy": 7.167596483230591,
      "epoch": 0.8207993124194242,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000477,
      "loss": 6.9522,
      "mean_token_accuracy": 0.11233177706599236,
      "num_tokens": 1734331.0,
      "step": 955
    },
    {
      "entropy": 7.214958715438843,
      "epoch": 0.8250966910184787,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004795,
      "loss": 6.8938,
      "mean_token_accuracy": 0.11346546709537506,
      "num_tokens": 1742340.0,
      "step": 960
    },
    {
      "entropy": 7.167982864379883,
      "epoch": 0.8293940696175333,
      "grad_norm": 1.15625,
      "learning_rate": 0.000482,
      "loss": 6.8975,
      "mean_token_accuracy": 0.11846677288413048,
      "num_tokens": 1751725.0,
      "step": 965
    },
    {
      "entropy": 7.095029211044311,
      "epoch": 0.8336914482165879,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004845,
      "loss": 6.8141,
      "mean_token_accuracy": 0.12262827306985855,
      "num_tokens": 1760294.0,
      "step": 970
    },
    {
      "entropy": 7.099491214752197,
      "epoch": 0.8379888268156425,
      "grad_norm": 1.140625,
      "learning_rate": 0.000487,
      "loss": 6.8854,
      "mean_token_accuracy": 0.11621206700801849,
      "num_tokens": 1768912.0,
      "step": 975
    },
    {
      "entropy": 7.235454845428467,
      "epoch": 0.842286205414697,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004895,
      "loss": 7.0682,
      "mean_token_accuracy": 0.11861302182078362,
      "num_tokens": 1778633.0,
      "step": 980
    },
    {
      "entropy": 7.143100643157959,
      "epoch": 0.8465835840137517,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000492,
      "loss": 6.9304,
      "mean_token_accuracy": 0.11692859306931495,
      "num_tokens": 1787275.0,
      "step": 985
    },
    {
      "entropy": 7.2316286087036135,
      "epoch": 0.8508809626128062,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004945,
      "loss": 6.9348,
      "mean_token_accuracy": 0.11131151169538497,
      "num_tokens": 1795994.0,
      "step": 990
    },
    {
      "entropy": 7.094430828094483,
      "epoch": 0.8551783412118608,
      "grad_norm": 0.90625,
      "learning_rate": 0.000497,
      "loss": 6.8575,
      "mean_token_accuracy": 0.11684277504682541,
      "num_tokens": 1806379.0,
      "step": 995
    },
    {
      "entropy": 7.199878978729248,
      "epoch": 0.8594757198109153,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004995,
      "loss": 6.878,
      "mean_token_accuracy": 0.1225502647459507,
      "num_tokens": 1816135.0,
      "step": 1000
    },
    {
      "epoch": 0.8594757198109153,
      "eval_entropy": 6.741199510591525,
      "eval_loss": 6.911599636077881,
      "eval_mean_token_accuracy": 0.11959127177324919,
      "eval_num_tokens": 1816135.0,
      "eval_runtime": 2.0596,
      "eval_samples_per_second": 1723.145,
      "eval_steps_per_second": 215.575,
      "step": 1000
    },
    {
      "entropy": 7.135749387741089,
      "epoch": 0.86377309840997,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004999998427807679,
      "loss": 6.88,
      "mean_token_accuracy": 0.11578838974237442,
      "num_tokens": 1824777.0,
      "step": 1005
    },
    {
      "entropy": 7.0467671871185305,
      "epoch": 0.8680704770090245,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004999992040780138,
      "loss": 6.9483,
      "mean_token_accuracy": 0.11809242218732834,
      "num_tokens": 1833807.0,
      "step": 1010
    },
    {
      "entropy": 7.2159933090209964,
      "epoch": 0.8723678556080791,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999980740669294,
      "loss": 6.8907,
      "mean_token_accuracy": 0.11748233065009117,
      "num_tokens": 1843375.0,
      "step": 1015
    },
    {
      "entropy": 7.1540666103363035,
      "epoch": 0.8766652342071336,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999964527499823,
      "loss": 6.948,
      "mean_token_accuracy": 0.11102593019604683,
      "num_tokens": 1853036.0,
      "step": 1020
    },
    {
      "entropy": 7.12061505317688,
      "epoch": 0.8809626128061883,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999943401307127,
      "loss": 6.9196,
      "mean_token_accuracy": 0.11150210499763488,
      "num_tokens": 1862041.0,
      "step": 1025
    },
    {
      "entropy": 7.122325229644775,
      "epoch": 0.8852599914052428,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999917362137337,
      "loss": 6.8248,
      "mean_token_accuracy": 0.12204656526446342,
      "num_tokens": 1870707.0,
      "step": 1030
    },
    {
      "entropy": 7.1019820213317875,
      "epoch": 0.8895573700042974,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999886410047312,
      "loss": 6.815,
      "mean_token_accuracy": 0.12252610921859741,
      "num_tokens": 1879787.0,
      "step": 1035
    },
    {
      "entropy": 7.131076002120972,
      "epoch": 0.8938547486033519,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999850545104638,
      "loss": 6.8854,
      "mean_token_accuracy": 0.11829843446612358,
      "num_tokens": 1889413.0,
      "step": 1040
    },
    {
      "entropy": 7.105505514144897,
      "epoch": 0.8981521272024066,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999809767387633,
      "loss": 6.8543,
      "mean_token_accuracy": 0.11870720088481904,
      "num_tokens": 1898283.0,
      "step": 1045
    },
    {
      "entropy": 7.1863609790802006,
      "epoch": 0.9024495058014611,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999764076985337,
      "loss": 6.8726,
      "mean_token_accuracy": 0.11925134509801864,
      "num_tokens": 1907175.0,
      "step": 1050
    },
    {
      "entropy": 7.012548351287842,
      "epoch": 0.9067468844005157,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999713473997519,
      "loss": 6.9285,
      "mean_token_accuracy": 0.11801080331206322,
      "num_tokens": 1918223.0,
      "step": 1055
    },
    {
      "entropy": 7.097743034362793,
      "epoch": 0.9110442629995703,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999657958534677,
      "loss": 6.8618,
      "mean_token_accuracy": 0.11624032855033875,
      "num_tokens": 1928801.0,
      "step": 1060
    },
    {
      "entropy": 7.046412801742553,
      "epoch": 0.9153416415986249,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999597530718034,
      "loss": 6.8171,
      "mean_token_accuracy": 0.12201576456427574,
      "num_tokens": 1937406.0,
      "step": 1065
    },
    {
      "entropy": 7.076975727081299,
      "epoch": 0.9196390201976794,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499953219067954,
      "loss": 6.829,
      "mean_token_accuracy": 0.11706628054380416,
      "num_tokens": 1947184.0,
      "step": 1070
    },
    {
      "entropy": 7.0976677417755125,
      "epoch": 0.923936398796734,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004999461938561873,
      "loss": 6.8518,
      "mean_token_accuracy": 0.12058016434311866,
      "num_tokens": 1956293.0,
      "step": 1075
    },
    {
      "entropy": 7.072698259353638,
      "epoch": 0.9282337773957886,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999386774518432,
      "loss": 6.8318,
      "mean_token_accuracy": 0.11745714917778968,
      "num_tokens": 1964791.0,
      "step": 1080
    },
    {
      "entropy": 7.01435604095459,
      "epoch": 0.9325311559948432,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999306698713349,
      "loss": 6.7357,
      "mean_token_accuracy": 0.12115372568368912,
      "num_tokens": 1973754.0,
      "step": 1085
    },
    {
      "entropy": 7.059027576446534,
      "epoch": 0.9368285345938977,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999221711321477,
      "loss": 6.8225,
      "mean_token_accuracy": 0.12251915633678437,
      "num_tokens": 1983035.0,
      "step": 1090
    },
    {
      "entropy": 6.931616544723511,
      "epoch": 0.9411259131929522,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999131812528393,
      "loss": 6.8285,
      "mean_token_accuracy": 0.11646857559680938,
      "num_tokens": 1992584.0,
      "step": 1095
    },
    {
      "entropy": 7.093084859848022,
      "epoch": 0.9454232917920069,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00049990370025304,
      "loss": 6.8713,
      "mean_token_accuracy": 0.11829434335231781,
      "num_tokens": 2001876.0,
      "step": 1100
    },
    {
      "entropy": 7.065115928649902,
      "epoch": 0.9497206703910615,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998937281534526,
      "loss": 6.7715,
      "mean_token_accuracy": 0.12478053271770477,
      "num_tokens": 2011067.0,
      "step": 1105
    },
    {
      "entropy": 7.09222378730774,
      "epoch": 0.954018048990116,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998832649758521,
      "loss": 6.847,
      "mean_token_accuracy": 0.11962009072303773,
      "num_tokens": 2020763.0,
      "step": 1110
    },
    {
      "entropy": 6.995938444137574,
      "epoch": 0.9583154275891707,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004998723107430862,
      "loss": 6.8303,
      "mean_token_accuracy": 0.1202974759042263,
      "num_tokens": 2029534.0,
      "step": 1115
    },
    {
      "entropy": 7.090177202224732,
      "epoch": 0.9626128061882252,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998608654790741,
      "loss": 6.7773,
      "mean_token_accuracy": 0.11798964366316796,
      "num_tokens": 2039143.0,
      "step": 1120
    },
    {
      "entropy": 6.939780521392822,
      "epoch": 0.9669101847872797,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000499848929208808,
      "loss": 6.7368,
      "mean_token_accuracy": 0.12413144260644912,
      "num_tokens": 2048253.0,
      "step": 1125
    },
    {
      "entropy": 6.968171167373657,
      "epoch": 0.9712075633863343,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004998365019583519,
      "loss": 6.7792,
      "mean_token_accuracy": 0.12741474211215972,
      "num_tokens": 2057234.0,
      "step": 1130
    },
    {
      "entropy": 7.055770301818848,
      "epoch": 0.975504941985389,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998235837548417,
      "loss": 6.8277,
      "mean_token_accuracy": 0.12360714524984359,
      "num_tokens": 2065431.0,
      "step": 1135
    },
    {
      "entropy": 7.023305130004883,
      "epoch": 0.9798023205844435,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499810174626486,
      "loss": 6.8226,
      "mean_token_accuracy": 0.12461819276213645,
      "num_tokens": 2074723.0,
      "step": 1140
    },
    {
      "entropy": 6.984486246109009,
      "epoch": 0.984099699183498,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004997962746025646,
      "loss": 6.7019,
      "mean_token_accuracy": 0.13492951169610023,
      "num_tokens": 2084509.0,
      "step": 1145
    },
    {
      "entropy": 7.0700774669647215,
      "epoch": 0.9883970777825526,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004997818837134298,
      "loss": 6.849,
      "mean_token_accuracy": 0.12504945918917656,
      "num_tokens": 2093110.0,
      "step": 1150
    },
    {
      "entropy": 6.870955991744995,
      "epoch": 0.9926944563816072,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997670019905057,
      "loss": 6.702,
      "mean_token_accuracy": 0.1213143840432167,
      "num_tokens": 2102355.0,
      "step": 1155
    },
    {
      "entropy": 7.033004474639893,
      "epoch": 0.9969918349806618,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997516294662876,
      "loss": 6.7378,
      "mean_token_accuracy": 0.12335852757096291,
      "num_tokens": 2110418.0,
      "step": 1160
    },
    {
      "entropy": 6.9259838528103295,
      "epoch": 1.0008594757198108,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997357661743433,
      "loss": 6.717,
      "mean_token_accuracy": 0.128744567433993,
      "num_tokens": 2117866.0,
      "step": 1165
    },
    {
      "entropy": 6.995123481750488,
      "epoch": 1.0051568543188656,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997194121493118,
      "loss": 6.5734,
      "mean_token_accuracy": 0.12787071764469146,
      "num_tokens": 2126082.0,
      "step": 1170
    },
    {
      "entropy": 6.919755554199218,
      "epoch": 1.0094542329179201,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997025674269037,
      "loss": 6.5449,
      "mean_token_accuracy": 0.1354580543935299,
      "num_tokens": 2134042.0,
      "step": 1175
    },
    {
      "entropy": 6.861895275115967,
      "epoch": 1.0137516115169747,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996852320439013,
      "loss": 6.6139,
      "mean_token_accuracy": 0.12945729792118071,
      "num_tokens": 2142570.0,
      "step": 1180
    },
    {
      "entropy": 6.911503505706787,
      "epoch": 1.0180489901160292,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996674060381578,
      "loss": 6.5633,
      "mean_token_accuracy": 0.13328466117382048,
      "num_tokens": 2151310.0,
      "step": 1185
    },
    {
      "entropy": 6.907685422897339,
      "epoch": 1.0223463687150838,
      "grad_norm": 1.125,
      "learning_rate": 0.0004996490894485985,
      "loss": 6.6286,
      "mean_token_accuracy": 0.12506390661001204,
      "num_tokens": 2160662.0,
      "step": 1190
    },
    {
      "entropy": 6.932943487167359,
      "epoch": 1.0266437473141383,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996302823152193,
      "loss": 6.566,
      "mean_token_accuracy": 0.1292807199060917,
      "num_tokens": 2170067.0,
      "step": 1195
    },
    {
      "entropy": 6.858935070037842,
      "epoch": 1.0309411259131929,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996109846790873,
      "loss": 6.5326,
      "mean_token_accuracy": 0.1338893659412861,
      "num_tokens": 2178850.0,
      "step": 1200
    },
    {
      "entropy": 6.8632307052612305,
      "epoch": 1.0352385045122476,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004995911965823412,
      "loss": 6.5605,
      "mean_token_accuracy": 0.13701220229268074,
      "num_tokens": 2188307.0,
      "step": 1205
    },
    {
      "entropy": 6.856139659881592,
      "epoch": 1.0395358831113022,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004995709180681899,
      "loss": 6.5326,
      "mean_token_accuracy": 0.140198864787817,
      "num_tokens": 2197026.0,
      "step": 1210
    },
    {
      "entropy": 6.900337648391724,
      "epoch": 1.0438332617103567,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499550149180914,
      "loss": 6.5314,
      "mean_token_accuracy": 0.13545763865113258,
      "num_tokens": 2205537.0,
      "step": 1215
    },
    {
      "entropy": 6.889223337173462,
      "epoch": 1.0481306403094113,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004995288899658641,
      "loss": 6.5543,
      "mean_token_accuracy": 0.13490064293146134,
      "num_tokens": 2214508.0,
      "step": 1220
    },
    {
      "entropy": 6.821092844009399,
      "epoch": 1.0524280189084658,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995071404694619,
      "loss": 6.6698,
      "mean_token_accuracy": 0.1264920301735401,
      "num_tokens": 2223084.0,
      "step": 1225
    },
    {
      "entropy": 6.998721551895142,
      "epoch": 1.0567253975075204,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994849007391996,
      "loss": 6.5995,
      "mean_token_accuracy": 0.1271670438349247,
      "num_tokens": 2231406.0,
      "step": 1230
    },
    {
      "entropy": 6.76812891960144,
      "epoch": 1.061022776106575,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994621708236401,
      "loss": 6.5119,
      "mean_token_accuracy": 0.131594667583704,
      "num_tokens": 2239867.0,
      "step": 1235
    },
    {
      "entropy": 6.892703628540039,
      "epoch": 1.0653201547056295,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499438950772416,
      "loss": 6.5595,
      "mean_token_accuracy": 0.13020255863666536,
      "num_tokens": 2248844.0,
      "step": 1240
    },
    {
      "entropy": 6.80609540939331,
      "epoch": 1.0696175333046842,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004994152406362311,
      "loss": 6.4868,
      "mean_token_accuracy": 0.1369902640581131,
      "num_tokens": 2257599.0,
      "step": 1245
    },
    {
      "entropy": 6.888293266296387,
      "epoch": 1.0739149119037388,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004993910404668586,
      "loss": 6.5419,
      "mean_token_accuracy": 0.12804562970995903,
      "num_tokens": 2266510.0,
      "step": 1250
    },
    {
      "entropy": 6.807245254516602,
      "epoch": 1.0782122905027933,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499366350317142,
      "loss": 6.5445,
      "mean_token_accuracy": 0.13409808352589608,
      "num_tokens": 2275462.0,
      "step": 1255
    },
    {
      "entropy": 6.824594402313233,
      "epoch": 1.0825096691018479,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004993411702409948,
      "loss": 6.5053,
      "mean_token_accuracy": 0.13172662556171416,
      "num_tokens": 2283826.0,
      "step": 1260
    },
    {
      "entropy": 6.823530054092407,
      "epoch": 1.0868070477009024,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993155002934002,
      "loss": 6.5131,
      "mean_token_accuracy": 0.1332196593284607,
      "num_tokens": 2292967.0,
      "step": 1265
    },
    {
      "entropy": 6.906051445007324,
      "epoch": 1.091104426299957,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004992893405304111,
      "loss": 6.6504,
      "mean_token_accuracy": 0.13221597969532012,
      "num_tokens": 2302336.0,
      "step": 1270
    },
    {
      "entropy": 6.785560846328735,
      "epoch": 1.0954018048990115,
      "grad_norm": 1.109375,
      "learning_rate": 0.00049926269100915,
      "loss": 6.5522,
      "mean_token_accuracy": 0.13710481822490692,
      "num_tokens": 2311465.0,
      "step": 1275
    },
    {
      "entropy": 6.9224625587463375,
      "epoch": 1.0996991834980663,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004992355517878087,
      "loss": 6.6554,
      "mean_token_accuracy": 0.12584567815065384,
      "num_tokens": 2320281.0,
      "step": 1280
    },
    {
      "entropy": 6.836811113357544,
      "epoch": 1.1039965620971208,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004992079229256484,
      "loss": 6.5584,
      "mean_token_accuracy": 0.12910671755671502,
      "num_tokens": 2329755.0,
      "step": 1285
    },
    {
      "entropy": 6.704578638076782,
      "epoch": 1.1082939406961754,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991798044829996,
      "loss": 6.5024,
      "mean_token_accuracy": 0.12955976650118828,
      "num_tokens": 2338807.0,
      "step": 1290
    },
    {
      "entropy": 6.910696506500244,
      "epoch": 1.11259131929523,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004991511965212618,
      "loss": 6.5914,
      "mean_token_accuracy": 0.13309080600738527,
      "num_tokens": 2348056.0,
      "step": 1295
    },
    {
      "entropy": 6.757680082321167,
      "epoch": 1.1168886978942845,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004991220991029032,
      "loss": 6.5917,
      "mean_token_accuracy": 0.12701491191983222,
      "num_tokens": 2357780.0,
      "step": 1300
    },
    {
      "entropy": 6.866291856765747,
      "epoch": 1.121186076493339,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000499092512291461,
      "loss": 6.5742,
      "mean_token_accuracy": 0.1293620027601719,
      "num_tokens": 2367060.0,
      "step": 1305
    },
    {
      "entropy": 6.836187791824341,
      "epoch": 1.1254834550923936,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499062436151541,
      "loss": 6.5736,
      "mean_token_accuracy": 0.1331700198352337,
      "num_tokens": 2375751.0,
      "step": 1310
    },
    {
      "entropy": 6.893719673156738,
      "epoch": 1.129780833691448,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004990318707488173,
      "loss": 6.6178,
      "mean_token_accuracy": 0.1281808391213417,
      "num_tokens": 2385013.0,
      "step": 1315
    },
    {
      "entropy": 6.797501945495606,
      "epoch": 1.1340782122905029,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004990008161500327,
      "loss": 6.5368,
      "mean_token_accuracy": 0.1343330167233944,
      "num_tokens": 2392935.0,
      "step": 1320
    },
    {
      "entropy": 6.829013967514038,
      "epoch": 1.1383755908895574,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000498969272422998,
      "loss": 6.532,
      "mean_token_accuracy": 0.1358281135559082,
      "num_tokens": 2401560.0,
      "step": 1325
    },
    {
      "entropy": 6.755909633636475,
      "epoch": 1.142672969488612,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004989372396365921,
      "loss": 6.4518,
      "mean_token_accuracy": 0.13816773220896722,
      "num_tokens": 2410050.0,
      "step": 1330
    },
    {
      "entropy": 6.890218067169189,
      "epoch": 1.1469703480876665,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004989047178607618,
      "loss": 6.547,
      "mean_token_accuracy": 0.1365761749446392,
      "num_tokens": 2418980.0,
      "step": 1335
    },
    {
      "entropy": 6.804721832275391,
      "epoch": 1.151267726686721,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004988717071665215,
      "loss": 6.5574,
      "mean_token_accuracy": 0.1326664350926876,
      "num_tokens": 2427992.0,
      "step": 1340
    },
    {
      "entropy": 6.721598196029663,
      "epoch": 1.1555651052857756,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004988382076259537,
      "loss": 6.4609,
      "mean_token_accuracy": 0.138286504894495,
      "num_tokens": 2436368.0,
      "step": 1345
    },
    {
      "entropy": 6.781888771057129,
      "epoch": 1.1598624838848304,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004988042193122077,
      "loss": 6.4618,
      "mean_token_accuracy": 0.13687749654054643,
      "num_tokens": 2445499.0,
      "step": 1350
    },
    {
      "entropy": 6.851740026473999,
      "epoch": 1.164159862483885,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004987697422995005,
      "loss": 6.4922,
      "mean_token_accuracy": 0.13173471242189408,
      "num_tokens": 2454312.0,
      "step": 1355
    },
    {
      "entropy": 6.713117170333862,
      "epoch": 1.1684572410829395,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004987347766631161,
      "loss": 6.548,
      "mean_token_accuracy": 0.1326726868748665,
      "num_tokens": 2462922.0,
      "step": 1360
    },
    {
      "entropy": 6.791418552398682,
      "epoch": 1.172754619681994,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986993224794055,
      "loss": 6.5917,
      "mean_token_accuracy": 0.12623094841837884,
      "num_tokens": 2472195.0,
      "step": 1365
    },
    {
      "entropy": 6.79559736251831,
      "epoch": 1.1770519982810486,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004986633798257865,
      "loss": 6.4975,
      "mean_token_accuracy": 0.1346569150686264,
      "num_tokens": 2481021.0,
      "step": 1370
    },
    {
      "entropy": 6.720452213287354,
      "epoch": 1.181349376880103,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004986269487807434,
      "loss": 6.5027,
      "mean_token_accuracy": 0.13570783659815788,
      "num_tokens": 2490250.0,
      "step": 1375
    },
    {
      "entropy": 6.770303201675415,
      "epoch": 1.1856467554791577,
      "grad_norm": 1.03125,
      "learning_rate": 0.000498590029423827,
      "loss": 6.5757,
      "mean_token_accuracy": 0.1362602196633816,
      "num_tokens": 2499122.0,
      "step": 1380
    },
    {
      "entropy": 6.838917875289917,
      "epoch": 1.1899441340782122,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004985526218356546,
      "loss": 6.5512,
      "mean_token_accuracy": 0.1287482999265194,
      "num_tokens": 2508454.0,
      "step": 1385
    },
    {
      "entropy": 6.7279455184936525,
      "epoch": 1.1942415126772667,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004985147260979093,
      "loss": 6.4843,
      "mean_token_accuracy": 0.14064208790659904,
      "num_tokens": 2517353.0,
      "step": 1390
    },
    {
      "entropy": 6.755608987808228,
      "epoch": 1.1985388912763215,
      "grad_norm": 1.125,
      "learning_rate": 0.0004984763422933402,
      "loss": 6.4984,
      "mean_token_accuracy": 0.13640860766172408,
      "num_tokens": 2526321.0,
      "step": 1395
    },
    {
      "entropy": 6.755157899856568,
      "epoch": 1.202836269875376,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004984374705057623,
      "loss": 6.5168,
      "mean_token_accuracy": 0.13545396104454993,
      "num_tokens": 2535924.0,
      "step": 1400
    },
    {
      "entropy": 6.818363904953003,
      "epoch": 1.2071336484744306,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004983981108200561,
      "loss": 6.5065,
      "mean_token_accuracy": 0.1336505174636841,
      "num_tokens": 2545606.0,
      "step": 1405
    },
    {
      "entropy": 6.692107009887695,
      "epoch": 1.2114310270734852,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004983582633221672,
      "loss": 6.5062,
      "mean_token_accuracy": 0.1383553020656109,
      "num_tokens": 2554947.0,
      "step": 1410
    },
    {
      "entropy": 6.83019061088562,
      "epoch": 1.2157284056725397,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004983179280991068,
      "loss": 6.6534,
      "mean_token_accuracy": 0.12812817469239235,
      "num_tokens": 2564462.0,
      "step": 1415
    },
    {
      "entropy": 6.796614646911621,
      "epoch": 1.2200257842715942,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004982771052389508,
      "loss": 6.4877,
      "mean_token_accuracy": 0.1379811257123947,
      "num_tokens": 2573124.0,
      "step": 1420
    },
    {
      "entropy": 6.79015884399414,
      "epoch": 1.224323162870649,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004982357948308401,
      "loss": 6.59,
      "mean_token_accuracy": 0.12893555760383607,
      "num_tokens": 2581829.0,
      "step": 1425
    },
    {
      "entropy": 6.829104900360107,
      "epoch": 1.2286205414697036,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004981939969649799,
      "loss": 6.4392,
      "mean_token_accuracy": 0.13909578174352646,
      "num_tokens": 2590631.0,
      "step": 1430
    },
    {
      "entropy": 6.743937397003174,
      "epoch": 1.232917920068758,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004981517117326404,
      "loss": 6.5519,
      "mean_token_accuracy": 0.13750746101140976,
      "num_tokens": 2600684.0,
      "step": 1435
    },
    {
      "entropy": 6.769179630279541,
      "epoch": 1.2372152986678127,
      "grad_norm": 1.0,
      "learning_rate": 0.0004981089392261553,
      "loss": 6.4703,
      "mean_token_accuracy": 0.14099505171179771,
      "num_tokens": 2609667.0,
      "step": 1440
    },
    {
      "entropy": 6.69663987159729,
      "epoch": 1.2415126772668672,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000498065679538923,
      "loss": 6.5504,
      "mean_token_accuracy": 0.13748323917388916,
      "num_tokens": 2620025.0,
      "step": 1445
    },
    {
      "entropy": 6.7704236030578615,
      "epoch": 1.2458100558659218,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004980219327654049,
      "loss": 6.4502,
      "mean_token_accuracy": 0.1409187763929367,
      "num_tokens": 2629032.0,
      "step": 1450
    },
    {
      "entropy": 6.728482866287232,
      "epoch": 1.2501074344649763,
      "grad_norm": 1.109375,
      "learning_rate": 0.000497977699001127,
      "loss": 6.4341,
      "mean_token_accuracy": 0.1414632149040699,
      "num_tokens": 2638303.0,
      "step": 1455
    },
    {
      "entropy": 6.760922050476074,
      "epoch": 1.2544048130640308,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004979329783426778,
      "loss": 6.4682,
      "mean_token_accuracy": 0.13974226340651513,
      "num_tokens": 2647902.0,
      "step": 1460
    },
    {
      "entropy": 6.739649820327759,
      "epoch": 1.2587021916630854,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004978877708877094,
      "loss": 6.5202,
      "mean_token_accuracy": 0.1345860943198204,
      "num_tokens": 2657902.0,
      "step": 1465
    },
    {
      "entropy": 6.732415056228637,
      "epoch": 1.2629995702621402,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004978420767349368,
      "loss": 6.4395,
      "mean_token_accuracy": 0.1351854085922241,
      "num_tokens": 2667082.0,
      "step": 1470
    },
    {
      "entropy": 6.715655040740967,
      "epoch": 1.2672969488611947,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004977958959841379,
      "loss": 6.5146,
      "mean_token_accuracy": 0.1359639957547188,
      "num_tokens": 2676855.0,
      "step": 1475
    },
    {
      "entropy": 6.778389596939087,
      "epoch": 1.2715943274602493,
      "grad_norm": 1.09375,
      "learning_rate": 0.000497749228736153,
      "loss": 6.4591,
      "mean_token_accuracy": 0.13977851048111917,
      "num_tokens": 2685750.0,
      "step": 1480
    },
    {
      "entropy": 6.708815574645996,
      "epoch": 1.2758917060593038,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004977020750928845,
      "loss": 6.5176,
      "mean_token_accuracy": 0.14020123705267906,
      "num_tokens": 2695272.0,
      "step": 1485
    },
    {
      "entropy": 6.7224524974822994,
      "epoch": 1.2801890846583583,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004976544351572973,
      "loss": 6.4449,
      "mean_token_accuracy": 0.14301179200410843,
      "num_tokens": 2704806.0,
      "step": 1490
    },
    {
      "entropy": 6.649525260925293,
      "epoch": 1.2844864632574131,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004976063090334179,
      "loss": 6.4976,
      "mean_token_accuracy": 0.13876865953207015,
      "num_tokens": 2713521.0,
      "step": 1495
    },
    {
      "entropy": 6.740707063674927,
      "epoch": 1.2887838418564677,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004975576968263346,
      "loss": 6.502,
      "mean_token_accuracy": 0.12958571687340736,
      "num_tokens": 2721848.0,
      "step": 1500
    },
    {
      "epoch": 1.2887838418564677,
      "eval_entropy": 6.470501623712145,
      "eval_loss": 6.560330390930176,
      "eval_mean_token_accuracy": 0.13803270239349422,
      "eval_num_tokens": 2721848.0,
      "eval_runtime": 2.0451,
      "eval_samples_per_second": 1735.35,
      "eval_steps_per_second": 217.102,
      "step": 1500
    },
    {
      "entropy": 6.670271635055542,
      "epoch": 1.2930812204555222,
      "grad_norm": 0.921875,
      "learning_rate": 0.000497508598642197,
      "loss": 6.4776,
      "mean_token_accuracy": 0.1379353679716587,
      "num_tokens": 2731473.0,
      "step": 1505
    },
    {
      "entropy": 6.75262417793274,
      "epoch": 1.2973785990545768,
      "grad_norm": 1.015625,
      "learning_rate": 0.000497459014588216,
      "loss": 6.5328,
      "mean_token_accuracy": 0.13872059807181358,
      "num_tokens": 2739867.0,
      "step": 1510
    },
    {
      "entropy": 6.689033460617066,
      "epoch": 1.3016759776536313,
      "grad_norm": 1.03125,
      "learning_rate": 0.000497408944772663,
      "loss": 6.4493,
      "mean_token_accuracy": 0.13728251829743385,
      "num_tokens": 2748903.0,
      "step": 1515
    },
    {
      "entropy": 6.651050281524658,
      "epoch": 1.3059733562526858,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004973583893048707,
      "loss": 6.4194,
      "mean_token_accuracy": 0.1359269514679909,
      "num_tokens": 2757711.0,
      "step": 1520
    },
    {
      "entropy": 6.832587623596192,
      "epoch": 1.3102707348517404,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004973073482952321,
      "loss": 6.4476,
      "mean_token_accuracy": 0.1389494240283966,
      "num_tokens": 2765633.0,
      "step": 1525
    },
    {
      "entropy": 6.610323524475097,
      "epoch": 1.314568113450795,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004972558218552004,
      "loss": 6.4865,
      "mean_token_accuracy": 0.1345442496240139,
      "num_tokens": 2774495.0,
      "step": 1530
    },
    {
      "entropy": 6.781571102142334,
      "epoch": 1.3188654920498495,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004972038100972885,
      "loss": 6.511,
      "mean_token_accuracy": 0.13351112902164458,
      "num_tokens": 2782665.0,
      "step": 1535
    },
    {
      "entropy": 6.640726566314697,
      "epoch": 1.323162870648904,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004971513131350697,
      "loss": 6.4315,
      "mean_token_accuracy": 0.13331406638026239,
      "num_tokens": 2791394.0,
      "step": 1540
    },
    {
      "entropy": 6.637244319915771,
      "epoch": 1.3274602492479588,
      "grad_norm": 1.125,
      "learning_rate": 0.0004970983310831759,
      "loss": 6.4216,
      "mean_token_accuracy": 0.13528204038739206,
      "num_tokens": 2800488.0,
      "step": 1545
    },
    {
      "entropy": 6.73633337020874,
      "epoch": 1.3317576278470133,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004970448640572989,
      "loss": 6.554,
      "mean_token_accuracy": 0.13508356660604476,
      "num_tokens": 2810116.0,
      "step": 1550
    },
    {
      "entropy": 6.68259482383728,
      "epoch": 1.336055006446068,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004969909121741895,
      "loss": 6.362,
      "mean_token_accuracy": 0.14365303963422776,
      "num_tokens": 2819205.0,
      "step": 1555
    },
    {
      "entropy": 6.660860872268676,
      "epoch": 1.3403523850451224,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004969364755516569,
      "loss": 6.4237,
      "mean_token_accuracy": 0.13722610473632812,
      "num_tokens": 2828017.0,
      "step": 1560
    },
    {
      "entropy": 6.719718647003174,
      "epoch": 1.344649763644177,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004968815543085689,
      "loss": 6.4716,
      "mean_token_accuracy": 0.1409387357532978,
      "num_tokens": 2837125.0,
      "step": 1565
    },
    {
      "entropy": 6.677382087707519,
      "epoch": 1.3489471422432318,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004968261485648516,
      "loss": 6.4702,
      "mean_token_accuracy": 0.1383634626865387,
      "num_tokens": 2845438.0,
      "step": 1570
    },
    {
      "entropy": 6.7003734588623045,
      "epoch": 1.3532445208422863,
      "grad_norm": 1.0,
      "learning_rate": 0.000496770258441489,
      "loss": 6.4578,
      "mean_token_accuracy": 0.14429745227098464,
      "num_tokens": 2854389.0,
      "step": 1575
    },
    {
      "entropy": 6.630741024017334,
      "epoch": 1.3575418994413408,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004967138840605228,
      "loss": 6.4274,
      "mean_token_accuracy": 0.14213711023330688,
      "num_tokens": 2863654.0,
      "step": 1580
    },
    {
      "entropy": 6.580131101608276,
      "epoch": 1.3618392780403954,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000496657025545052,
      "loss": 6.3333,
      "mean_token_accuracy": 0.14458294957876205,
      "num_tokens": 2872871.0,
      "step": 1585
    },
    {
      "entropy": 6.652856779098511,
      "epoch": 1.36613665663945,
      "grad_norm": 1.109375,
      "learning_rate": 0.000496599683019233,
      "loss": 6.4385,
      "mean_token_accuracy": 0.1378958448767662,
      "num_tokens": 2881140.0,
      "step": 1590
    },
    {
      "entropy": 6.743864488601685,
      "epoch": 1.3704340352385045,
      "grad_norm": 1.0625,
      "learning_rate": 0.000496541856608279,
      "loss": 6.4138,
      "mean_token_accuracy": 0.13684486821293831,
      "num_tokens": 2889945.0,
      "step": 1595
    },
    {
      "entropy": 6.543468856811524,
      "epoch": 1.374731413837559,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004964835464384595,
      "loss": 6.3442,
      "mean_token_accuracy": 0.14644223153591157,
      "num_tokens": 2898897.0,
      "step": 1600
    },
    {
      "entropy": 6.6516578674316404,
      "epoch": 1.3790287924366136,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000496424752637101,
      "loss": 6.3765,
      "mean_token_accuracy": 0.14313278645277022,
      "num_tokens": 2907717.0,
      "step": 1605
    },
    {
      "entropy": 6.6023597240448,
      "epoch": 1.3833261710356681,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004963654753325853,
      "loss": 6.3608,
      "mean_token_accuracy": 0.142365200817585,
      "num_tokens": 2916213.0,
      "step": 1610
    },
    {
      "entropy": 6.675694608688355,
      "epoch": 1.387623549634723,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004963057146543505,
      "loss": 6.5232,
      "mean_token_accuracy": 0.13410560339689254,
      "num_tokens": 2925706.0,
      "step": 1615
    },
    {
      "entropy": 6.640337657928467,
      "epoch": 1.3919209282337774,
      "grad_norm": 0.96875,
      "learning_rate": 0.00049624547073289,
      "loss": 6.4341,
      "mean_token_accuracy": 0.1372315354645252,
      "num_tokens": 2934464.0,
      "step": 1620
    },
    {
      "entropy": 6.644095182418823,
      "epoch": 1.396218306832832,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004961847436997526,
      "loss": 6.3392,
      "mean_token_accuracy": 0.1437459670007229,
      "num_tokens": 2944095.0,
      "step": 1625
    },
    {
      "entropy": 6.544746780395508,
      "epoch": 1.4005156854318865,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004961235336875416,
      "loss": 6.3337,
      "mean_token_accuracy": 0.14818646609783173,
      "num_tokens": 2953357.0,
      "step": 1630
    },
    {
      "entropy": 6.590663099288941,
      "epoch": 1.404813064030941,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004960618408299154,
      "loss": 6.4913,
      "mean_token_accuracy": 0.13292545676231385,
      "num_tokens": 2963020.0,
      "step": 1635
    },
    {
      "entropy": 6.676125764846802,
      "epoch": 1.4091104426299956,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004959996652615865,
      "loss": 6.3291,
      "mean_token_accuracy": 0.1426067478954792,
      "num_tokens": 2971955.0,
      "step": 1640
    },
    {
      "entropy": 6.682071399688721,
      "epoch": 1.4134078212290504,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004959370071183216,
      "loss": 6.3941,
      "mean_token_accuracy": 0.1417133815586567,
      "num_tokens": 2980662.0,
      "step": 1645
    },
    {
      "entropy": 6.630775785446167,
      "epoch": 1.417705199828105,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004958738665369407,
      "loss": 6.5253,
      "mean_token_accuracy": 0.12776071205735207,
      "num_tokens": 2990038.0,
      "step": 1650
    },
    {
      "entropy": 6.679714679718018,
      "epoch": 1.4220025784271595,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004958102436553179,
      "loss": 6.4334,
      "mean_token_accuracy": 0.1367899753153324,
      "num_tokens": 2999835.0,
      "step": 1655
    },
    {
      "entropy": 6.633715724945068,
      "epoch": 1.426299957026214,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00049574613861238,
      "loss": 6.428,
      "mean_token_accuracy": 0.13732998967170715,
      "num_tokens": 3009593.0,
      "step": 1660
    },
    {
      "entropy": 6.719851779937744,
      "epoch": 1.4305973356252686,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004956815515481069,
      "loss": 6.4586,
      "mean_token_accuracy": 0.14100154265761375,
      "num_tokens": 3019187.0,
      "step": 1665
    },
    {
      "entropy": 6.548164558410645,
      "epoch": 1.4348947142243231,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004956164826035309,
      "loss": 6.3723,
      "mean_token_accuracy": 0.1403822347521782,
      "num_tokens": 3027875.0,
      "step": 1670
    },
    {
      "entropy": 6.590739965438843,
      "epoch": 1.4391920928233777,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004955509319207363,
      "loss": 6.3917,
      "mean_token_accuracy": 0.139241461455822,
      "num_tokens": 3036902.0,
      "step": 1675
    },
    {
      "entropy": 6.564586973190307,
      "epoch": 1.4434894714224322,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004954848996428601,
      "loss": 6.37,
      "mean_token_accuracy": 0.1485365867614746,
      "num_tokens": 3046653.0,
      "step": 1680
    },
    {
      "entropy": 6.688973379135132,
      "epoch": 1.4477868500214868,
      "grad_norm": 1.21875,
      "learning_rate": 0.00049541838591409,
      "loss": 6.4727,
      "mean_token_accuracy": 0.13771305233240128,
      "num_tokens": 3056273.0,
      "step": 1685
    },
    {
      "entropy": 6.601505422592163,
      "epoch": 1.4520842286205415,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004953513908796657,
      "loss": 6.3754,
      "mean_token_accuracy": 0.142202128469944,
      "num_tokens": 3065662.0,
      "step": 1690
    },
    {
      "entropy": 6.688344573974609,
      "epoch": 1.456381607219596,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004952839146858773,
      "loss": 6.4125,
      "mean_token_accuracy": 0.14289402961730957,
      "num_tokens": 3073970.0,
      "step": 1695
    },
    {
      "entropy": 6.557283878326416,
      "epoch": 1.4606789858186506,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004952159574800658,
      "loss": 6.4149,
      "mean_token_accuracy": 0.1368069365620613,
      "num_tokens": 3082500.0,
      "step": 1700
    },
    {
      "entropy": 6.6616050720214846,
      "epoch": 1.4649763644177052,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004951475194106229,
      "loss": 6.3624,
      "mean_token_accuracy": 0.14598480761051177,
      "num_tokens": 3091574.0,
      "step": 1705
    },
    {
      "entropy": 6.564044332504272,
      "epoch": 1.4692737430167597,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004950786006269898,
      "loss": 6.4712,
      "mean_token_accuracy": 0.13445298001170158,
      "num_tokens": 3102402.0,
      "step": 1710
    },
    {
      "entropy": 6.630729532241821,
      "epoch": 1.4735711216158143,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004950092012796576,
      "loss": 6.2883,
      "mean_token_accuracy": 0.14547964110970496,
      "num_tokens": 3111347.0,
      "step": 1715
    },
    {
      "entropy": 6.57949857711792,
      "epoch": 1.477868500214869,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004949393215201666,
      "loss": 6.3576,
      "mean_token_accuracy": 0.14195869863033295,
      "num_tokens": 3120018.0,
      "step": 1720
    },
    {
      "entropy": 6.621220111846924,
      "epoch": 1.4821658788139236,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004948689615011065,
      "loss": 6.4303,
      "mean_token_accuracy": 0.13502154499292374,
      "num_tokens": 3129669.0,
      "step": 1725
    },
    {
      "entropy": 6.622284650802612,
      "epoch": 1.4864632574129781,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004947981213761154,
      "loss": 6.3626,
      "mean_token_accuracy": 0.14655405953526496,
      "num_tokens": 3139112.0,
      "step": 1730
    },
    {
      "entropy": 6.566265106201172,
      "epoch": 1.4907606360120327,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004947268012998797,
      "loss": 6.314,
      "mean_token_accuracy": 0.15282158702611923,
      "num_tokens": 3148437.0,
      "step": 1735
    },
    {
      "entropy": 6.555867433547974,
      "epoch": 1.4950580146110872,
      "grad_norm": 0.9375,
      "learning_rate": 0.000494655001428134,
      "loss": 6.3163,
      "mean_token_accuracy": 0.14129877462983131,
      "num_tokens": 3158165.0,
      "step": 1740
    },
    {
      "entropy": 6.610077142715454,
      "epoch": 1.4993553932101418,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004945827219176604,
      "loss": 6.3806,
      "mean_token_accuracy": 0.1494217187166214,
      "num_tokens": 3167262.0,
      "step": 1745
    },
    {
      "entropy": 6.497924327850342,
      "epoch": 1.5036527718091963,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004945099629262888,
      "loss": 6.365,
      "mean_token_accuracy": 0.14257355630397797,
      "num_tokens": 3176696.0,
      "step": 1750
    },
    {
      "entropy": 6.763267278671265,
      "epoch": 1.5079501504082509,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004944367246128954,
      "loss": 6.4414,
      "mean_token_accuracy": 0.13902869820594788,
      "num_tokens": 3185857.0,
      "step": 1755
    },
    {
      "entropy": 6.499178218841553,
      "epoch": 1.5122475290073054,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004943630071374036,
      "loss": 6.2842,
      "mean_token_accuracy": 0.15046660900115966,
      "num_tokens": 3194687.0,
      "step": 1760
    },
    {
      "entropy": 6.597637414932251,
      "epoch": 1.51654490760636,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004942888106607828,
      "loss": 6.3408,
      "mean_token_accuracy": 0.14210622310638427,
      "num_tokens": 3204913.0,
      "step": 1765
    },
    {
      "entropy": 6.629096126556396,
      "epoch": 1.5208422862054147,
      "grad_norm": 1.0,
      "learning_rate": 0.0004942141353450486,
      "loss": 6.3319,
      "mean_token_accuracy": 0.14792249947786332,
      "num_tokens": 3213312.0,
      "step": 1770
    },
    {
      "entropy": 6.488542318344116,
      "epoch": 1.5251396648044693,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004941389813532619,
      "loss": 6.254,
      "mean_token_accuracy": 0.15323352664709092,
      "num_tokens": 3222992.0,
      "step": 1775
    },
    {
      "entropy": 6.508915090560913,
      "epoch": 1.5294370434035238,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000494063348849529,
      "loss": 6.3006,
      "mean_token_accuracy": 0.15017192512750627,
      "num_tokens": 3232836.0,
      "step": 1780
    },
    {
      "entropy": 6.618602657318116,
      "epoch": 1.5337344220025786,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004939872379990011,
      "loss": 6.4568,
      "mean_token_accuracy": 0.1389564484357834,
      "num_tokens": 3243171.0,
      "step": 1785
    },
    {
      "entropy": 6.68247127532959,
      "epoch": 1.5380318006016331,
      "grad_norm": 1.125,
      "learning_rate": 0.0004939106489678739,
      "loss": 6.3675,
      "mean_token_accuracy": 0.1462523892521858,
      "num_tokens": 3251995.0,
      "step": 1790
    },
    {
      "entropy": 6.5040504932403564,
      "epoch": 1.5423291792006877,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000493833581923387,
      "loss": 6.3162,
      "mean_token_accuracy": 0.14888131469488144,
      "num_tokens": 3260841.0,
      "step": 1795
    },
    {
      "entropy": 6.610679483413696,
      "epoch": 1.5466265577997422,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004937560370338244,
      "loss": 6.4592,
      "mean_token_accuracy": 0.13561476543545722,
      "num_tokens": 3270979.0,
      "step": 1800
    },
    {
      "entropy": 6.649268531799317,
      "epoch": 1.5509239363987968,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000493678014468513,
      "loss": 6.3782,
      "mean_token_accuracy": 0.15145477950572966,
      "num_tokens": 3279848.0,
      "step": 1805
    },
    {
      "entropy": 6.525541353225708,
      "epoch": 1.5552213149978513,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004935995143978227,
      "loss": 6.3828,
      "mean_token_accuracy": 0.14420599341392518,
      "num_tokens": 3289172.0,
      "step": 1810
    },
    {
      "entropy": 6.548158836364746,
      "epoch": 1.5595186935969059,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004935205369931664,
      "loss": 6.2838,
      "mean_token_accuracy": 0.1488450989127159,
      "num_tokens": 3297432.0,
      "step": 1815
    },
    {
      "entropy": 6.629633808135987,
      "epoch": 1.5638160721959604,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004934410824269992,
      "loss": 6.307,
      "mean_token_accuracy": 0.1468965768814087,
      "num_tokens": 3307486.0,
      "step": 1820
    },
    {
      "entropy": 6.505390453338623,
      "epoch": 1.568113450795015,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004933611508728182,
      "loss": 6.277,
      "mean_token_accuracy": 0.15054729878902434,
      "num_tokens": 3316296.0,
      "step": 1825
    },
    {
      "entropy": 6.567506504058838,
      "epoch": 1.5724108293940695,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000493280742505162,
      "loss": 6.3093,
      "mean_token_accuracy": 0.1449831858277321,
      "num_tokens": 3326080.0,
      "step": 1830
    },
    {
      "entropy": 6.486201763153076,
      "epoch": 1.576708207993124,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004931998574996102,
      "loss": 6.2364,
      "mean_token_accuracy": 0.14821842014789582,
      "num_tokens": 3334826.0,
      "step": 1835
    },
    {
      "entropy": 6.458436775207519,
      "epoch": 1.5810055865921788,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004931184960327832,
      "loss": 6.243,
      "mean_token_accuracy": 0.15286262780427934,
      "num_tokens": 3343261.0,
      "step": 1840
    },
    {
      "entropy": 6.5017963409423825,
      "epoch": 1.5853029651912334,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004930366582823421,
      "loss": 6.2846,
      "mean_token_accuracy": 0.1460932895541191,
      "num_tokens": 3352513.0,
      "step": 1845
    },
    {
      "entropy": 6.565137147903442,
      "epoch": 1.589600343790288,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004929543444269879,
      "loss": 6.3357,
      "mean_token_accuracy": 0.14929510578513144,
      "num_tokens": 3361577.0,
      "step": 1850
    },
    {
      "entropy": 6.485863208770752,
      "epoch": 1.5938977223893425,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000492871554646461,
      "loss": 6.3948,
      "mean_token_accuracy": 0.14030372574925423,
      "num_tokens": 3370591.0,
      "step": 1855
    },
    {
      "entropy": 6.554273700714111,
      "epoch": 1.5981951009883972,
      "grad_norm": 1.0,
      "learning_rate": 0.0004927882891215413,
      "loss": 6.306,
      "mean_token_accuracy": 0.14827967956662177,
      "num_tokens": 3379761.0,
      "step": 1860
    },
    {
      "entropy": 6.592674922943115,
      "epoch": 1.6024924795874518,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004927045480340475,
      "loss": 6.3969,
      "mean_token_accuracy": 0.1377456396818161,
      "num_tokens": 3388974.0,
      "step": 1865
    },
    {
      "entropy": 6.534068393707275,
      "epoch": 1.6067898581865063,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004926203315668363,
      "loss": 6.3089,
      "mean_token_accuracy": 0.14855122640728952,
      "num_tokens": 3398339.0,
      "step": 1870
    },
    {
      "entropy": 6.51072096824646,
      "epoch": 1.6110872367855609,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004925356399038032,
      "loss": 6.2784,
      "mean_token_accuracy": 0.1444276086986065,
      "num_tokens": 3408292.0,
      "step": 1875
    },
    {
      "entropy": 6.546713066101074,
      "epoch": 1.6153846153846154,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004924504732298808,
      "loss": 6.25,
      "mean_token_accuracy": 0.15252432748675346,
      "num_tokens": 3417057.0,
      "step": 1880
    },
    {
      "entropy": 6.566928911209106,
      "epoch": 1.61968199398367,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004923648317310391,
      "loss": 6.3645,
      "mean_token_accuracy": 0.14241925179958342,
      "num_tokens": 3425830.0,
      "step": 1885
    },
    {
      "entropy": 6.581201171875,
      "epoch": 1.6239793725827245,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004922787155942849,
      "loss": 6.3913,
      "mean_token_accuracy": 0.13747936934232713,
      "num_tokens": 3435513.0,
      "step": 1890
    },
    {
      "entropy": 6.493123197555542,
      "epoch": 1.628276751181779,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004921921250076611,
      "loss": 6.3081,
      "mean_token_accuracy": 0.1506146050989628,
      "num_tokens": 3444684.0,
      "step": 1895
    },
    {
      "entropy": 6.519382381439209,
      "epoch": 1.6325741297808336,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004921050601602475,
      "loss": 6.358,
      "mean_token_accuracy": 0.14751976430416108,
      "num_tokens": 3453454.0,
      "step": 1900
    },
    {
      "entropy": 6.561795377731324,
      "epoch": 1.6368715083798882,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004920175212421587,
      "loss": 6.2876,
      "mean_token_accuracy": 0.1480035036802292,
      "num_tokens": 3463228.0,
      "step": 1905
    },
    {
      "entropy": 6.427498245239258,
      "epoch": 1.6411688869789427,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004919295084445445,
      "loss": 6.1795,
      "mean_token_accuracy": 0.15354781299829484,
      "num_tokens": 3472131.0,
      "step": 1910
    },
    {
      "entropy": 6.453372526168823,
      "epoch": 1.6454662655779975,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004918410219595899,
      "loss": 6.262,
      "mean_token_accuracy": 0.1537986733019352,
      "num_tokens": 3480642.0,
      "step": 1915
    },
    {
      "entropy": 6.673685932159424,
      "epoch": 1.649763644177052,
      "grad_norm": 0.9375,
      "learning_rate": 0.000491752061980514,
      "loss": 6.2464,
      "mean_token_accuracy": 0.15058573931455613,
      "num_tokens": 3489346.0,
      "step": 1920
    },
    {
      "entropy": 6.364662742614746,
      "epoch": 1.6540610227761066,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004916626287015697,
      "loss": 6.2797,
      "mean_token_accuracy": 0.14565936028957366,
      "num_tokens": 3498473.0,
      "step": 1925
    },
    {
      "entropy": 6.596463584899903,
      "epoch": 1.658358401375161,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004915727223180436,
      "loss": 6.2889,
      "mean_token_accuracy": 0.14369736313819886,
      "num_tokens": 3507415.0,
      "step": 1930
    },
    {
      "entropy": 6.553056049346924,
      "epoch": 1.6626557799742159,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004914823430262554,
      "loss": 6.3913,
      "mean_token_accuracy": 0.13537019938230516,
      "num_tokens": 3516873.0,
      "step": 1935
    },
    {
      "entropy": 6.495272874832153,
      "epoch": 1.6669531585732704,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004913914910235573,
      "loss": 6.2558,
      "mean_token_accuracy": 0.1485748916864395,
      "num_tokens": 3525047.0,
      "step": 1940
    },
    {
      "entropy": 6.465020227432251,
      "epoch": 1.671250537172325,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004913001665083337,
      "loss": 6.2614,
      "mean_token_accuracy": 0.1493809461593628,
      "num_tokens": 3534354.0,
      "step": 1945
    },
    {
      "entropy": 6.480109977722168,
      "epoch": 1.6755479157713795,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004912083696800008,
      "loss": 6.3061,
      "mean_token_accuracy": 0.14220771193504333,
      "num_tokens": 3543830.0,
      "step": 1950
    },
    {
      "entropy": 6.481950473785401,
      "epoch": 1.679845294370434,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004911161007390063,
      "loss": 6.214,
      "mean_token_accuracy": 0.1464408040046692,
      "num_tokens": 3552314.0,
      "step": 1955
    },
    {
      "entropy": 6.433442115783691,
      "epoch": 1.6841426729694886,
      "grad_norm": 1.125,
      "learning_rate": 0.0004910233598868287,
      "loss": 6.2733,
      "mean_token_accuracy": 0.14608966633677484,
      "num_tokens": 3561656.0,
      "step": 1960
    },
    {
      "entropy": 6.5319255828857425,
      "epoch": 1.6884400515685432,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004909301473259769,
      "loss": 6.2782,
      "mean_token_accuracy": 0.14630958884954454,
      "num_tokens": 3571784.0,
      "step": 1965
    },
    {
      "entropy": 6.498413991928101,
      "epoch": 1.6927374301675977,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004908364632599899,
      "loss": 6.2466,
      "mean_token_accuracy": 0.14925297126173973,
      "num_tokens": 3580626.0,
      "step": 1970
    },
    {
      "entropy": 6.409846687316895,
      "epoch": 1.6970348087666522,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004907423078934362,
      "loss": 6.2432,
      "mean_token_accuracy": 0.14702003747224807,
      "num_tokens": 3589916.0,
      "step": 1975
    },
    {
      "entropy": 6.491791534423828,
      "epoch": 1.7013321873657068,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004906476814319134,
      "loss": 6.2665,
      "mean_token_accuracy": 0.1481113463640213,
      "num_tokens": 3599128.0,
      "step": 1980
    },
    {
      "entropy": 6.418525266647339,
      "epoch": 1.7056295659647613,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004905525840820481,
      "loss": 6.2822,
      "mean_token_accuracy": 0.14321094751358032,
      "num_tokens": 3608764.0,
      "step": 1985
    },
    {
      "entropy": 6.584860706329346,
      "epoch": 1.709926944563816,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004904570160514948,
      "loss": 6.318,
      "mean_token_accuracy": 0.14112781435251237,
      "num_tokens": 3619082.0,
      "step": 1990
    },
    {
      "entropy": 6.488425922393799,
      "epoch": 1.7142243231628707,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004903609775489358,
      "loss": 6.2898,
      "mean_token_accuracy": 0.1483103021979332,
      "num_tokens": 3628695.0,
      "step": 1995
    },
    {
      "entropy": 6.527116823196411,
      "epoch": 1.7185217017619252,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004902644687840809,
      "loss": 6.2518,
      "mean_token_accuracy": 0.14549941420555115,
      "num_tokens": 3637599.0,
      "step": 2000
    },
    {
      "epoch": 1.7185217017619252,
      "eval_entropy": 6.19347925014324,
      "eval_loss": 6.329723358154297,
      "eval_mean_token_accuracy": 0.1496910023091881,
      "eval_num_tokens": 3637599.0,
      "eval_runtime": 2.05,
      "eval_samples_per_second": 1731.192,
      "eval_steps_per_second": 216.582,
      "step": 2000
    },
    {
      "entropy": 6.444890356063842,
      "epoch": 1.7228190803609797,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004901674899676667,
      "loss": 6.2616,
      "mean_token_accuracy": 0.1490693584084511,
      "num_tokens": 3647406.0,
      "step": 2005
    },
    {
      "entropy": 6.4421497821807865,
      "epoch": 1.7271164589600345,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004900700413114561,
      "loss": 6.1411,
      "mean_token_accuracy": 0.1515228644013405,
      "num_tokens": 3656531.0,
      "step": 2010
    },
    {
      "entropy": 6.438943338394165,
      "epoch": 1.731413837559089,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000489972123028238,
      "loss": 6.2278,
      "mean_token_accuracy": 0.1497970812022686,
      "num_tokens": 3664922.0,
      "step": 2015
    },
    {
      "entropy": 6.477704477310181,
      "epoch": 1.7357112161581436,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004898737353318268,
      "loss": 6.1731,
      "mean_token_accuracy": 0.1531771868467331,
      "num_tokens": 3673283.0,
      "step": 2020
    },
    {
      "entropy": 6.417043018341064,
      "epoch": 1.7400085947571982,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000489774878437062,
      "loss": 6.3189,
      "mean_token_accuracy": 0.1491183653473854,
      "num_tokens": 3681760.0,
      "step": 2025
    },
    {
      "entropy": 6.496932172775269,
      "epoch": 1.7443059733562527,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004896755525598074,
      "loss": 6.137,
      "mean_token_accuracy": 0.15091845840215684,
      "num_tokens": 3689408.0,
      "step": 2030
    },
    {
      "entropy": 6.424662256240845,
      "epoch": 1.7486033519553073,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004895757579169511,
      "loss": 6.2482,
      "mean_token_accuracy": 0.14746748507022858,
      "num_tokens": 3697904.0,
      "step": 2035
    },
    {
      "entropy": 6.576484441757202,
      "epoch": 1.7529007305543618,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004894754947264047,
      "loss": 6.2665,
      "mean_token_accuracy": 0.14906733110547066,
      "num_tokens": 3706704.0,
      "step": 2040
    },
    {
      "entropy": 6.423267459869384,
      "epoch": 1.7571981091534163,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000489374763207103,
      "loss": 6.3263,
      "mean_token_accuracy": 0.14524614065885544,
      "num_tokens": 3715690.0,
      "step": 2045
    },
    {
      "entropy": 6.511725521087646,
      "epoch": 1.761495487752471,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004892735635790033,
      "loss": 6.1332,
      "mean_token_accuracy": 0.1547706514596939,
      "num_tokens": 3724835.0,
      "step": 2050
    },
    {
      "entropy": 6.357218599319458,
      "epoch": 1.7657928663515254,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000489171896063085,
      "loss": 6.1786,
      "mean_token_accuracy": 0.15807082131505013,
      "num_tokens": 3733977.0,
      "step": 2055
    },
    {
      "entropy": 6.503004789352417,
      "epoch": 1.77009024495058,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004890697608813495,
      "loss": 6.2892,
      "mean_token_accuracy": 0.14620151743292809,
      "num_tokens": 3742665.0,
      "step": 2060
    },
    {
      "entropy": 6.575605297088623,
      "epoch": 1.7743876235496348,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004889671582568193,
      "loss": 6.3394,
      "mean_token_accuracy": 0.14444702491164207,
      "num_tokens": 3751647.0,
      "step": 2065
    },
    {
      "entropy": 6.386636257171631,
      "epoch": 1.7786850021486893,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004888640884135374,
      "loss": 6.2572,
      "mean_token_accuracy": 0.14602783024311067,
      "num_tokens": 3760852.0,
      "step": 2070
    },
    {
      "entropy": 6.443797779083252,
      "epoch": 1.7829823807477438,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004887605515765671,
      "loss": 6.1999,
      "mean_token_accuracy": 0.1500324159860611,
      "num_tokens": 3768640.0,
      "step": 2075
    },
    {
      "entropy": 6.453762483596802,
      "epoch": 1.7872797593467986,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004886565479719914,
      "loss": 6.2371,
      "mean_token_accuracy": 0.14520688354969025,
      "num_tokens": 3776504.0,
      "step": 2080
    },
    {
      "entropy": 6.546750164031982,
      "epoch": 1.7915771379458532,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004885520778269128,
      "loss": 6.2633,
      "mean_token_accuracy": 0.1508520394563675,
      "num_tokens": 3786353.0,
      "step": 2085
    },
    {
      "entropy": 6.421859884262085,
      "epoch": 1.7958745165449077,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004884471413694523,
      "loss": 6.2927,
      "mean_token_accuracy": 0.14654150381684303,
      "num_tokens": 3795902.0,
      "step": 2090
    },
    {
      "entropy": 6.475854921340942,
      "epoch": 1.8001718951439623,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004883417388287491,
      "loss": 6.2152,
      "mean_token_accuracy": 0.14346302896738053,
      "num_tokens": 3805986.0,
      "step": 2095
    },
    {
      "entropy": 6.390176630020141,
      "epoch": 1.8044692737430168,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004882358704349603,
      "loss": 6.3309,
      "mean_token_accuracy": 0.15342760533094407,
      "num_tokens": 3814915.0,
      "step": 2100
    },
    {
      "entropy": 6.5035735130310055,
      "epoch": 1.8087666523420713,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004881295364192601,
      "loss": 6.2208,
      "mean_token_accuracy": 0.15399999171495438,
      "num_tokens": 3823966.0,
      "step": 2105
    },
    {
      "entropy": 6.512830305099487,
      "epoch": 1.813064030941126,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004880227370138394,
      "loss": 6.2733,
      "mean_token_accuracy": 0.141741144657135,
      "num_tokens": 3832775.0,
      "step": 2110
    },
    {
      "entropy": 6.412053823471069,
      "epoch": 1.8173614095401804,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004879154724519057,
      "loss": 6.1883,
      "mean_token_accuracy": 0.15187056958675385,
      "num_tokens": 3842808.0,
      "step": 2115
    },
    {
      "entropy": 6.525836372375489,
      "epoch": 1.821658788139235,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004878077429676816,
      "loss": 6.3298,
      "mean_token_accuracy": 0.14646431654691697,
      "num_tokens": 3853303.0,
      "step": 2120
    },
    {
      "entropy": 6.404772138595581,
      "epoch": 1.8259561667382895,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004876995487964054,
      "loss": 6.2232,
      "mean_token_accuracy": 0.14396868348121644,
      "num_tokens": 3862462.0,
      "step": 2125
    },
    {
      "entropy": 6.504109525680542,
      "epoch": 1.830253545337344,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048759089017432996,
      "loss": 6.3474,
      "mean_token_accuracy": 0.1436037801206112,
      "num_tokens": 3871596.0,
      "step": 2130
    },
    {
      "entropy": 6.516725015640259,
      "epoch": 1.8345509239363988,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004874817673387222,
      "loss": 6.2488,
      "mean_token_accuracy": 0.150234591960907,
      "num_tokens": 3881276.0,
      "step": 2135
    },
    {
      "entropy": 6.40357723236084,
      "epoch": 1.8388483025354534,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048737218052786275,
      "loss": 6.3306,
      "mean_token_accuracy": 0.1420318104326725,
      "num_tokens": 3891610.0,
      "step": 2140
    },
    {
      "entropy": 6.566394519805908,
      "epoch": 1.843145681134508,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048726212998104554,
      "loss": 6.2537,
      "mean_token_accuracy": 0.14613947346806527,
      "num_tokens": 3900584.0,
      "step": 2145
    },
    {
      "entropy": 6.378025674819947,
      "epoch": 1.8474430597335625,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004871516159385768,
      "loss": 6.2048,
      "mean_token_accuracy": 0.15661944225430488,
      "num_tokens": 3910208.0,
      "step": 2150
    },
    {
      "entropy": 6.331248712539673,
      "epoch": 1.8517404383326173,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004870406386417752,
      "loss": 6.1133,
      "mean_token_accuracy": 0.15902349054813386,
      "num_tokens": 3918424.0,
      "step": 2155
    },
    {
      "entropy": 6.3689392566680905,
      "epoch": 1.8560378169316718,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004869291983329707,
      "loss": 6.0464,
      "mean_token_accuracy": 0.16715734601020812,
      "num_tokens": 3926206.0,
      "step": 2160
    },
    {
      "entropy": 6.45319676399231,
      "epoch": 1.8603351955307263,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004868172952555044,
      "loss": 6.1646,
      "mean_token_accuracy": 0.14661258310079575,
      "num_tokens": 3935769.0,
      "step": 2165
    },
    {
      "entropy": 6.394322872161865,
      "epoch": 1.864632574129781,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004867049296537278,
      "loss": 6.1415,
      "mean_token_accuracy": 0.15269692093133927,
      "num_tokens": 3945118.0,
      "step": 2170
    },
    {
      "entropy": 6.474433708190918,
      "epoch": 1.8689299527288354,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004865921017730027,
      "loss": 6.2316,
      "mean_token_accuracy": 0.15484376698732377,
      "num_tokens": 3954012.0,
      "step": 2175
    },
    {
      "entropy": 6.439454174041748,
      "epoch": 1.87322733132789,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048647881185969995,
      "loss": 6.2453,
      "mean_token_accuracy": 0.14685868918895723,
      "num_tokens": 3964239.0,
      "step": 2180
    },
    {
      "entropy": 6.393148756027221,
      "epoch": 1.8775247099269445,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004863650601611994,
      "loss": 6.1549,
      "mean_token_accuracy": 0.15369305908679962,
      "num_tokens": 3973694.0,
      "step": 2185
    },
    {
      "entropy": 6.382004022598267,
      "epoch": 1.881822088525999,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048625084692588937,
      "loss": 6.1903,
      "mean_token_accuracy": 0.15866641253232955,
      "num_tokens": 3982706.0,
      "step": 2190
    },
    {
      "entropy": 6.365177822113037,
      "epoch": 1.8861194671250536,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048613617240316593,
      "loss": 6.1493,
      "mean_token_accuracy": 0.15633954405784606,
      "num_tokens": 3990934.0,
      "step": 2195
    },
    {
      "entropy": 6.4114241123199465,
      "epoch": 1.8904168457241082,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004860210368434323,
      "loss": 6.202,
      "mean_token_accuracy": 0.15539102777838706,
      "num_tokens": 3999864.0,
      "step": 2200
    },
    {
      "entropy": 6.44605655670166,
      "epoch": 1.8947142243231627,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048590544049809857,
      "loss": 6.2143,
      "mean_token_accuracy": 0.15502872616052626,
      "num_tokens": 4008273.0,
      "step": 2205
    },
    {
      "entropy": 6.396330881118774,
      "epoch": 1.8990116029222175,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000485789383619581,
      "loss": 6.2196,
      "mean_token_accuracy": 0.1552313044667244,
      "num_tokens": 4017697.0,
      "step": 2210
    },
    {
      "entropy": 6.393840074539185,
      "epoch": 1.903308981521272,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004856728664613015,
      "loss": 6.2381,
      "mean_token_accuracy": 0.1517387181520462,
      "num_tokens": 4026775.0,
      "step": 2215
    },
    {
      "entropy": 6.4121527671813965,
      "epoch": 1.9076063601203266,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048555588927768674,
      "loss": 6.2065,
      "mean_token_accuracy": 0.15306443050503732,
      "num_tokens": 4036476.0,
      "step": 2220
    },
    {
      "entropy": 6.4631853103637695,
      "epoch": 1.9119037387193811,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004854384523241683,
      "loss": 6.1853,
      "mean_token_accuracy": 0.14910603538155556,
      "num_tokens": 4045221.0,
      "step": 2225
    },
    {
      "entropy": 6.358554410934448,
      "epoch": 1.916201117318436,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048532055585718143,
      "loss": 6.1143,
      "mean_token_accuracy": 0.1558242991566658,
      "num_tokens": 4053754.0,
      "step": 2230
    },
    {
      "entropy": 6.419343090057373,
      "epoch": 1.9204984959174904,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048520220013416505,
      "loss": 6.1398,
      "mean_token_accuracy": 0.15855558216571808,
      "num_tokens": 4061730.0,
      "step": 2235
    },
    {
      "entropy": 6.371563529968261,
      "epoch": 1.924795874516545,
      "grad_norm": 1.0,
      "learning_rate": 0.0004850833854135607,
      "loss": 6.2178,
      "mean_token_accuracy": 0.15214873254299163,
      "num_tokens": 4070501.0,
      "step": 2240
    },
    {
      "entropy": 6.432424831390381,
      "epoch": 1.9290932531155995,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004849641119548122,
      "loss": 6.284,
      "mean_token_accuracy": 0.14980236291885377,
      "num_tokens": 4079621.0,
      "step": 2245
    },
    {
      "entropy": 6.53593487739563,
      "epoch": 1.933390631714654,
      "grad_norm": 1.0,
      "learning_rate": 0.000484844380018365,
      "loss": 6.2792,
      "mean_token_accuracy": 0.14946389943361282,
      "num_tokens": 4090106.0,
      "step": 2250
    },
    {
      "entropy": 6.445489692687988,
      "epoch": 1.9376880103137086,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000484724189865666,
      "loss": 6.2137,
      "mean_token_accuracy": 0.15263792276382446,
      "num_tokens": 4099269.0,
      "step": 2255
    },
    {
      "entropy": 6.2474573135375975,
      "epoch": 1.9419853889127632,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004846035417591624,
      "loss": 6.14,
      "mean_token_accuracy": 0.15635560899972917,
      "num_tokens": 4108414.0,
      "step": 2260
    },
    {
      "entropy": 6.473322486877441,
      "epoch": 1.9462827675118177,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004844824359623014,
      "loss": 6.2727,
      "mean_token_accuracy": 0.14365370869636535,
      "num_tokens": 4117731.0,
      "step": 2265
    },
    {
      "entropy": 6.42958698272705,
      "epoch": 1.9505801461108723,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048436087273952966,
      "loss": 6.2456,
      "mean_token_accuracy": 0.14321349710226058,
      "num_tokens": 4127194.0,
      "step": 2270
    },
    {
      "entropy": 6.395676183700561,
      "epoch": 1.9548775247099268,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048423885235629265,
      "loss": 6.2082,
      "mean_token_accuracy": 0.1566771924495697,
      "num_tokens": 4135594.0,
      "step": 2275
    },
    {
      "entropy": 6.3766546726226805,
      "epoch": 1.9591749033089814,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004841163750790342,
      "loss": 6.2387,
      "mean_token_accuracy": 0.1482451669871807,
      "num_tokens": 4145027.0,
      "step": 2280
    },
    {
      "entropy": 6.411153888702392,
      "epoch": 1.9634722819080361,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048399344117519555,
      "loss": 6.0918,
      "mean_token_accuracy": 0.15616220831871033,
      "num_tokens": 4153754.0,
      "step": 2285
    },
    {
      "entropy": 6.349991703033448,
      "epoch": 1.9677696605070907,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048387005091321544,
      "loss": 6.1779,
      "mean_token_accuracy": 0.15535371601581574,
      "num_tokens": 4162765.0,
      "step": 2290
    },
    {
      "entropy": 6.42883620262146,
      "epoch": 1.9720670391061452,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048374620456252877,
      "loss": 6.1885,
      "mean_token_accuracy": 0.1536056511104107,
      "num_tokens": 4171589.0,
      "step": 2295
    },
    {
      "entropy": 6.385624837875366,
      "epoch": 1.9763644177052,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048362190239356644,
      "loss": 6.1974,
      "mean_token_accuracy": 0.15068082362413407,
      "num_tokens": 4181817.0,
      "step": 2300
    },
    {
      "entropy": 6.353290271759033,
      "epoch": 1.9806617963042545,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00048349714467775474,
      "loss": 6.1594,
      "mean_token_accuracy": 0.1520315356552601,
      "num_tokens": 4191350.0,
      "step": 2305
    },
    {
      "entropy": 6.391085815429688,
      "epoch": 1.984959174903309,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048337193168751464,
      "loss": 6.2041,
      "mean_token_accuracy": 0.1448718011379242,
      "num_tokens": 4199888.0,
      "step": 2310
    },
    {
      "entropy": 6.419165563583374,
      "epoch": 1.9892565535023636,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004832462636962613,
      "loss": 6.1675,
      "mean_token_accuracy": 0.15215626806020738,
      "num_tokens": 4209509.0,
      "step": 2315
    },
    {
      "entropy": 6.369716596603394,
      "epoch": 1.9935539321014182,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004831201409784034,
      "loss": 6.1396,
      "mean_token_accuracy": 0.154638309776783,
      "num_tokens": 4218496.0,
      "step": 2320
    },
    {
      "entropy": 6.423878908157349,
      "epoch": 1.9978513107004727,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004829935638093424,
      "loss": 6.1503,
      "mean_token_accuracy": 0.154294952750206,
      "num_tokens": 4227504.0,
      "step": 2325
    },
    {
      "entropy": 6.347577359941271,
      "epoch": 2.0017189514396216,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004828665324654724,
      "loss": 6.0631,
      "mean_token_accuracy": 0.15955217679341635,
      "num_tokens": 4235338.0,
      "step": 2330
    },
    {
      "entropy": 6.4371147632598875,
      "epoch": 2.006016330038676,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004827390472241791,
      "loss": 5.8518,
      "mean_token_accuracy": 0.1650273993611336,
      "num_tokens": 4244905.0,
      "step": 2335
    },
    {
      "entropy": 6.327361726760865,
      "epoch": 2.010313708637731,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004826111083638392,
      "loss": 5.9297,
      "mean_token_accuracy": 0.16078748032450677,
      "num_tokens": 4254533.0,
      "step": 2340
    },
    {
      "entropy": 6.365427732467651,
      "epoch": 2.0146110872367857,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048248271616382,
      "loss": 5.8909,
      "mean_token_accuracy": 0.16574428230524063,
      "num_tokens": 4264023.0,
      "step": 2345
    },
    {
      "entropy": 6.342710208892822,
      "epoch": 2.0189084658358403,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048235387090447894,
      "loss": 5.9334,
      "mean_token_accuracy": 0.16069018244743347,
      "num_tokens": 4273298.0,
      "step": 2350
    },
    {
      "entropy": 6.364804077148437,
      "epoch": 2.023205844434895,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048222457286716235,
      "loss": 5.8764,
      "mean_token_accuracy": 0.1686857357621193,
      "num_tokens": 4283244.0,
      "step": 2355
    },
    {
      "entropy": 6.2878947257995605,
      "epoch": 2.0275032230339494,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048209482233420564,
      "loss": 5.8282,
      "mean_token_accuracy": 0.1770843207836151,
      "num_tokens": 4291677.0,
      "step": 2360
    },
    {
      "entropy": 6.3038060665130615,
      "epoch": 2.031800601633004,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000481964619588932,
      "loss": 5.8865,
      "mean_token_accuracy": 0.16635389626026154,
      "num_tokens": 4300822.0,
      "step": 2365
    },
    {
      "entropy": 6.329894876480102,
      "epoch": 2.0360979802320585,
      "grad_norm": 1.0,
      "learning_rate": 0.0004818339649156523,
      "loss": 5.9031,
      "mean_token_accuracy": 0.16931673139333725,
      "num_tokens": 4310149.0,
      "step": 2370
    },
    {
      "entropy": 6.202824544906616,
      "epoch": 2.040395358831113,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048170285859966395,
      "loss": 5.8119,
      "mean_token_accuracy": 0.1753219798207283,
      "num_tokens": 4319109.0,
      "step": 2375
    },
    {
      "entropy": 6.33490400314331,
      "epoch": 2.0446927374301676,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048157130092725087,
      "loss": 5.7905,
      "mean_token_accuracy": 0.17381954342126846,
      "num_tokens": 4327921.0,
      "step": 2380
    },
    {
      "entropy": 6.351769590377808,
      "epoch": 2.048990116029222,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004814392921856824,
      "loss": 5.9396,
      "mean_token_accuracy": 0.16599524021148682,
      "num_tokens": 4338026.0,
      "step": 2385
    },
    {
      "entropy": 6.274995040893555,
      "epoch": 2.0532874946282766,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004813068326632128,
      "loss": 5.7941,
      "mean_token_accuracy": 0.1739266350865364,
      "num_tokens": 4347794.0,
      "step": 2390
    },
    {
      "entropy": 6.335586738586426,
      "epoch": 2.057584873227331,
      "grad_norm": 1.0,
      "learning_rate": 0.0004811739226490809,
      "loss": 5.9723,
      "mean_token_accuracy": 0.1657353535294533,
      "num_tokens": 4357249.0,
      "step": 2395
    },
    {
      "entropy": 6.2787346839904785,
      "epoch": 2.0618822518263857,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048104056243350896,
      "loss": 5.9151,
      "mean_token_accuracy": 0.1622727930545807,
      "num_tokens": 4366053.0,
      "step": 2400
    },
    {
      "entropy": 6.314475345611572,
      "epoch": 2.0661796304254403,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004809067523077023,
      "loss": 5.9368,
      "mean_token_accuracy": 0.16686248034238815,
      "num_tokens": 4375543.0,
      "step": 2405
    },
    {
      "entropy": 6.281144666671753,
      "epoch": 2.0704770090244953,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048077249256384884,
      "loss": 5.8062,
      "mean_token_accuracy": 0.17436677366495132,
      "num_tokens": 4384332.0,
      "step": 2410
    },
    {
      "entropy": 6.224628639221192,
      "epoch": 2.07477438762355,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004806377834951182,
      "loss": 5.9153,
      "mean_token_accuracy": 0.16276209503412248,
      "num_tokens": 4393670.0,
      "step": 2415
    },
    {
      "entropy": 6.381396484375,
      "epoch": 2.0790717662226044,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048050262539566104,
      "loss": 5.9074,
      "mean_token_accuracy": 0.1704580768942833,
      "num_tokens": 4402763.0,
      "step": 2420
    },
    {
      "entropy": 6.253005933761597,
      "epoch": 2.083369144821659,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004803670185606087,
      "loss": 5.8316,
      "mean_token_accuracy": 0.17470465898513793,
      "num_tokens": 4411863.0,
      "step": 2425
    },
    {
      "entropy": 6.290292263031006,
      "epoch": 2.0876665234207135,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004802309632860724,
      "loss": 5.9153,
      "mean_token_accuracy": 0.16799798905849456,
      "num_tokens": 4421110.0,
      "step": 2430
    },
    {
      "entropy": 6.328047180175782,
      "epoch": 2.091963902019768,
      "grad_norm": 1.0,
      "learning_rate": 0.00048009445986914236,
      "loss": 5.9037,
      "mean_token_accuracy": 0.16361760944128037,
      "num_tokens": 4430249.0,
      "step": 2435
    },
    {
      "entropy": 6.251105260848999,
      "epoch": 2.0962612806188226,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047995750860788756,
      "loss": 5.8763,
      "mean_token_accuracy": 0.16064124852418898,
      "num_tokens": 4439686.0,
      "step": 2440
    },
    {
      "entropy": 6.269289493560791,
      "epoch": 2.100558659217877,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004798201098013547,
      "loss": 5.8784,
      "mean_token_accuracy": 0.17077162116765976,
      "num_tokens": 4448645.0,
      "step": 2445
    },
    {
      "entropy": 6.216794633865357,
      "epoch": 2.1048560378169316,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047968226374956797,
      "loss": 5.8564,
      "mean_token_accuracy": 0.16725987941026688,
      "num_tokens": 4456870.0,
      "step": 2450
    },
    {
      "entropy": 6.243628740310669,
      "epoch": 2.109153416415986,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047954397075352794,
      "loss": 5.8711,
      "mean_token_accuracy": 0.16993128061294555,
      "num_tokens": 4466287.0,
      "step": 2455
    },
    {
      "entropy": 6.267875003814697,
      "epoch": 2.1134507950150407,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047940523111521136,
      "loss": 5.7674,
      "mean_token_accuracy": 0.17551021575927733,
      "num_tokens": 4474461.0,
      "step": 2460
    },
    {
      "entropy": 6.209959268569946,
      "epoch": 2.1177481736140953,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004792660451375701,
      "loss": 5.8378,
      "mean_token_accuracy": 0.1700508713722229,
      "num_tokens": 4483002.0,
      "step": 2465
    },
    {
      "entropy": 6.252925443649292,
      "epoch": 2.12204555221315,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047912641312453064,
      "loss": 5.8445,
      "mean_token_accuracy": 0.16888150274753572,
      "num_tokens": 4492061.0,
      "step": 2470
    },
    {
      "entropy": 6.2727385520935055,
      "epoch": 2.1263429308122044,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00047898633538099363,
      "loss": 5.9077,
      "mean_token_accuracy": 0.16311377584934234,
      "num_tokens": 4501829.0,
      "step": 2475
    },
    {
      "entropy": 6.240498495101929,
      "epoch": 2.130640309411259,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004788458122128327,
      "loss": 5.926,
      "mean_token_accuracy": 0.16604274660348892,
      "num_tokens": 4511539.0,
      "step": 2480
    },
    {
      "entropy": 6.272474479675293,
      "epoch": 2.134937688010314,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047870484392689434,
      "loss": 5.7914,
      "mean_token_accuracy": 0.1672845244407654,
      "num_tokens": 4520425.0,
      "step": 2485
    },
    {
      "entropy": 6.2719934463500975,
      "epoch": 2.1392350666093685,
      "grad_norm": 1.03125,
      "learning_rate": 0.000478563430830997,
      "loss": 5.8803,
      "mean_token_accuracy": 0.16633028984069825,
      "num_tokens": 4529474.0,
      "step": 2490
    },
    {
      "entropy": 6.207609510421753,
      "epoch": 2.143532445208423,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047842157323393035,
      "loss": 5.8296,
      "mean_token_accuracy": 0.1706731289625168,
      "num_tokens": 4538082.0,
      "step": 2495
    },
    {
      "entropy": 6.286458158493042,
      "epoch": 2.1478298238074776,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004782792714454547,
      "loss": 6.0015,
      "mean_token_accuracy": 0.16442835330963135,
      "num_tokens": 4547340.0,
      "step": 2500
    },
    {
      "epoch": 2.1478298238074776,
      "eval_entropy": 6.109945370270325,
      "eval_loss": 6.212513446807861,
      "eval_mean_token_accuracy": 0.15539610120761502,
      "eval_num_tokens": 4547340.0,
      "eval_runtime": 2.0504,
      "eval_samples_per_second": 1730.921,
      "eval_steps_per_second": 216.548,
      "step": 2500
    },
    {
      "entropy": 6.2429650783538815,
      "epoch": 2.152127202406532,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004781365257763002,
      "loss": 5.8456,
      "mean_token_accuracy": 0.16931284964084625,
      "num_tokens": 4556415.0,
      "step": 2505
    },
    {
      "entropy": 6.164100790023804,
      "epoch": 2.1564245810055866,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047799333653816633,
      "loss": 5.7458,
      "mean_token_accuracy": 0.1765029177069664,
      "num_tokens": 4565156.0,
      "step": 2510
    },
    {
      "entropy": 6.2381916522979735,
      "epoch": 2.160721959604641,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00047784970404372124,
      "loss": 5.8464,
      "mean_token_accuracy": 0.16380368024110795,
      "num_tokens": 4574678.0,
      "step": 2515
    },
    {
      "entropy": 6.172519302368164,
      "epoch": 2.1650193382036957,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047770562860660083,
      "loss": 5.8538,
      "mean_token_accuracy": 0.16628692597150802,
      "num_tokens": 4583253.0,
      "step": 2520
    },
    {
      "entropy": 6.255974197387696,
      "epoch": 2.1693167168027503,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004775611105414083,
      "loss": 5.9376,
      "mean_token_accuracy": 0.1599043294787407,
      "num_tokens": 4594042.0,
      "step": 2525
    },
    {
      "entropy": 6.199368286132812,
      "epoch": 2.173614095401805,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004774161501637133,
      "loss": 5.8834,
      "mean_token_accuracy": 0.1693019151687622,
      "num_tokens": 4603128.0,
      "step": 2530
    },
    {
      "entropy": 6.238662958145142,
      "epoch": 2.1779114740008594,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004772707477900514,
      "loss": 5.8539,
      "mean_token_accuracy": 0.16817047595977783,
      "num_tokens": 4611537.0,
      "step": 2535
    },
    {
      "entropy": 6.345929908752441,
      "epoch": 2.182208852599914,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004771249037379232,
      "loss": 5.968,
      "mean_token_accuracy": 0.16030408143997193,
      "num_tokens": 4622481.0,
      "step": 2540
    },
    {
      "entropy": 6.233392953872681,
      "epoch": 2.1865062311989685,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004769786183257939,
      "loss": 5.8709,
      "mean_token_accuracy": 0.17347402423620223,
      "num_tokens": 4631259.0,
      "step": 2545
    },
    {
      "entropy": 6.184610652923584,
      "epoch": 2.190803609798023,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004768318918730924,
      "loss": 5.7984,
      "mean_token_accuracy": 0.17670518308877944,
      "num_tokens": 4640266.0,
      "step": 2550
    },
    {
      "entropy": 6.218139505386352,
      "epoch": 2.195100988397078,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047668472470021044,
      "loss": 5.8713,
      "mean_token_accuracy": 0.16367379873991011,
      "num_tokens": 4649520.0,
      "step": 2555
    },
    {
      "entropy": 6.239636707305908,
      "epoch": 2.1993983669961326,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004765371171285025,
      "loss": 5.8163,
      "mean_token_accuracy": 0.17504389137029647,
      "num_tokens": 4658501.0,
      "step": 2560
    },
    {
      "entropy": 6.096165037155151,
      "epoch": 2.203695745595187,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047638906948028445,
      "loss": 5.8729,
      "mean_token_accuracy": 0.16270434856414795,
      "num_tokens": 4667567.0,
      "step": 2565
    },
    {
      "entropy": 6.269977855682373,
      "epoch": 2.2079931241942417,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047624058207883317,
      "loss": 5.8604,
      "mean_token_accuracy": 0.17035209834575654,
      "num_tokens": 4676618.0,
      "step": 2570
    },
    {
      "entropy": 6.308097791671753,
      "epoch": 2.212290502793296,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047609165524838576,
      "loss": 5.9352,
      "mean_token_accuracy": 0.1621068611741066,
      "num_tokens": 4685967.0,
      "step": 2575
    },
    {
      "entropy": 6.127339649200439,
      "epoch": 2.2165878813923507,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004759422893141389,
      "loss": 5.8236,
      "mean_token_accuracy": 0.1692316710948944,
      "num_tokens": 4694568.0,
      "step": 2580
    },
    {
      "entropy": 6.216268539428711,
      "epoch": 2.2208852599914053,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004757924846022482,
      "loss": 5.89,
      "mean_token_accuracy": 0.1616070792078972,
      "num_tokens": 4703648.0,
      "step": 2585
    },
    {
      "entropy": 6.20239634513855,
      "epoch": 2.22518263859046,
      "grad_norm": 1.125,
      "learning_rate": 0.00047564224143982714,
      "loss": 5.7428,
      "mean_token_accuracy": 0.17824842780828476,
      "num_tokens": 4712444.0,
      "step": 2590
    },
    {
      "entropy": 6.250929498672486,
      "epoch": 2.2294800171895144,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047549156015494676,
      "loss": 5.9075,
      "mean_token_accuracy": 0.16583409011363984,
      "num_tokens": 4722034.0,
      "step": 2595
    },
    {
      "entropy": 6.108888292312622,
      "epoch": 2.233777395788569,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047534044107663484,
      "loss": 5.9105,
      "mean_token_accuracy": 0.1638724222779274,
      "num_tokens": 4731344.0,
      "step": 2600
    },
    {
      "entropy": 6.311524677276611,
      "epoch": 2.2380747743876235,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047518888453487496,
      "loss": 5.817,
      "mean_token_accuracy": 0.1758144274353981,
      "num_tokens": 4739302.0,
      "step": 2605
    },
    {
      "entropy": 6.171684503555298,
      "epoch": 2.242372152986678,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004750368908606061,
      "loss": 5.9379,
      "mean_token_accuracy": 0.16401576846837998,
      "num_tokens": 4748848.0,
      "step": 2610
    },
    {
      "entropy": 6.278733396530152,
      "epoch": 2.2466695315857326,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047488446038572164,
      "loss": 5.988,
      "mean_token_accuracy": 0.158454030752182,
      "num_tokens": 4758194.0,
      "step": 2615
    },
    {
      "entropy": 6.271483707427978,
      "epoch": 2.250966910184787,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004747315934430688,
      "loss": 5.8964,
      "mean_token_accuracy": 0.16533373892307282,
      "num_tokens": 4768081.0,
      "step": 2620
    },
    {
      "entropy": 6.122837829589844,
      "epoch": 2.2552642887838417,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000474578290366448,
      "loss": 5.8258,
      "mean_token_accuracy": 0.16424619704484938,
      "num_tokens": 4776471.0,
      "step": 2625
    },
    {
      "entropy": 6.242071962356567,
      "epoch": 2.259561667382896,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004744245514906117,
      "loss": 5.8387,
      "mean_token_accuracy": 0.17140334099531174,
      "num_tokens": 4784403.0,
      "step": 2630
    },
    {
      "entropy": 6.109820699691772,
      "epoch": 2.263859045981951,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047427037715126426,
      "loss": 5.8071,
      "mean_token_accuracy": 0.16793861836194993,
      "num_tokens": 4792779.0,
      "step": 2635
    },
    {
      "entropy": 6.13071641921997,
      "epoch": 2.2681564245810057,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004741157676850608,
      "loss": 5.7959,
      "mean_token_accuracy": 0.1719123378396034,
      "num_tokens": 4801426.0,
      "step": 2640
    },
    {
      "entropy": 6.219759845733643,
      "epoch": 2.2724538031800603,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047396072342960663,
      "loss": 5.8436,
      "mean_token_accuracy": 0.16473171710968018,
      "num_tokens": 4810329.0,
      "step": 2645
    },
    {
      "entropy": 6.172075414657593,
      "epoch": 2.276751181779115,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047380524472345645,
      "loss": 5.8926,
      "mean_token_accuracy": 0.16456550508737564,
      "num_tokens": 4819544.0,
      "step": 2650
    },
    {
      "entropy": 6.204276752471924,
      "epoch": 2.2810485603781694,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004736493319061134,
      "loss": 5.9046,
      "mean_token_accuracy": 0.16881225854158402,
      "num_tokens": 4828113.0,
      "step": 2655
    },
    {
      "entropy": 6.19776291847229,
      "epoch": 2.285345938977224,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004734929853180291,
      "loss": 5.8878,
      "mean_token_accuracy": 0.16559435576200485,
      "num_tokens": 4836989.0,
      "step": 2660
    },
    {
      "entropy": 6.220638990402222,
      "epoch": 2.2896433175762785,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00047333620530060175,
      "loss": 5.9317,
      "mean_token_accuracy": 0.1637803792953491,
      "num_tokens": 4847103.0,
      "step": 2665
    },
    {
      "entropy": 6.251469230651855,
      "epoch": 2.293940696175333,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004731789921961764,
      "loss": 5.9358,
      "mean_token_accuracy": 0.164635768532753,
      "num_tokens": 4856238.0,
      "step": 2670
    },
    {
      "entropy": 6.212742853164673,
      "epoch": 2.2982380747743876,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004730213463480434,
      "loss": 5.8428,
      "mean_token_accuracy": 0.17302587628364563,
      "num_tokens": 4864608.0,
      "step": 2675
    },
    {
      "entropy": 6.159517288208008,
      "epoch": 2.302535453373442,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047286326810043857,
      "loss": 5.7962,
      "mean_token_accuracy": 0.16983986347913743,
      "num_tokens": 4873889.0,
      "step": 2680
    },
    {
      "entropy": 6.198332214355469,
      "epoch": 2.3068328319724967,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047270475779854137,
      "loss": 5.829,
      "mean_token_accuracy": 0.1737384930253029,
      "num_tokens": 4882902.0,
      "step": 2685
    },
    {
      "entropy": 6.295279121398925,
      "epoch": 2.311130210571551,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047254581578847507,
      "loss": 5.8707,
      "mean_token_accuracy": 0.16589304506778718,
      "num_tokens": 4892390.0,
      "step": 2690
    },
    {
      "entropy": 6.145449829101563,
      "epoch": 2.3154275891706058,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004723864424173055,
      "loss": 5.9872,
      "mean_token_accuracy": 0.1639627918601036,
      "num_tokens": 4901625.0,
      "step": 2695
    },
    {
      "entropy": 6.2244713306427,
      "epoch": 2.3197249677696608,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004722266380330403,
      "loss": 5.7716,
      "mean_token_accuracy": 0.17777814120054244,
      "num_tokens": 4910804.0,
      "step": 2700
    },
    {
      "entropy": 6.205074501037598,
      "epoch": 2.3240223463687153,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047206640298462857,
      "loss": 5.8554,
      "mean_token_accuracy": 0.17233493477106093,
      "num_tokens": 4920441.0,
      "step": 2705
    },
    {
      "entropy": 6.170949125289917,
      "epoch": 2.32831972496777,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047190573762195945,
      "loss": 5.8997,
      "mean_token_accuracy": 0.1642414078116417,
      "num_tokens": 4930204.0,
      "step": 2710
    },
    {
      "entropy": 6.167499303817749,
      "epoch": 2.3326171035668244,
      "grad_norm": 0.875,
      "learning_rate": 0.00047174464229586186,
      "loss": 6.007,
      "mean_token_accuracy": 0.15407418310642243,
      "num_tokens": 4941191.0,
      "step": 2715
    },
    {
      "entropy": 6.3167987823486325,
      "epoch": 2.336914482165879,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004715831173581036,
      "loss": 5.9632,
      "mean_token_accuracy": 0.16583859026432038,
      "num_tokens": 4951825.0,
      "step": 2720
    },
    {
      "entropy": 6.154605293273926,
      "epoch": 2.3412118607649335,
      "grad_norm": 0.90625,
      "learning_rate": 0.00047142116316139073,
      "loss": 5.9062,
      "mean_token_accuracy": 0.16728972047567367,
      "num_tokens": 4960632.0,
      "step": 2725
    },
    {
      "entropy": 6.263886308670044,
      "epoch": 2.345509239363988,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004712587800593663,
      "loss": 5.935,
      "mean_token_accuracy": 0.16565471589565278,
      "num_tokens": 4969455.0,
      "step": 2730
    },
    {
      "entropy": 6.191297912597657,
      "epoch": 2.3498066179630426,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004710959684066102,
      "loss": 5.8388,
      "mean_token_accuracy": 0.17293576747179032,
      "num_tokens": 4978997.0,
      "step": 2735
    },
    {
      "entropy": 6.195657110214233,
      "epoch": 2.354103996562097,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047093272855863803,
      "loss": 5.9072,
      "mean_token_accuracy": 0.16806580722332,
      "num_tokens": 4988305.0,
      "step": 2740
    },
    {
      "entropy": 6.159595584869384,
      "epoch": 2.3584013751611517,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004707690608719003,
      "loss": 5.8381,
      "mean_token_accuracy": 0.17171266674995422,
      "num_tokens": 4997022.0,
      "step": 2745
    },
    {
      "entropy": 6.204063940048218,
      "epoch": 2.362698753760206,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004706049657037818,
      "loss": 5.8984,
      "mean_token_accuracy": 0.16405170410871506,
      "num_tokens": 5005664.0,
      "step": 2750
    },
    {
      "entropy": 6.20027117729187,
      "epoch": 2.3669961323592608,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004704404434126009,
      "loss": 5.8461,
      "mean_token_accuracy": 0.16249971985816955,
      "num_tokens": 5014769.0,
      "step": 2755
    },
    {
      "entropy": 6.224915170669556,
      "epoch": 2.3712935109583153,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047027549435760843,
      "loss": 5.9216,
      "mean_token_accuracy": 0.1661619648337364,
      "num_tokens": 5024060.0,
      "step": 2760
    },
    {
      "entropy": 6.301250553131103,
      "epoch": 2.37559088955737,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004701101188989872,
      "loss": 5.9655,
      "mean_token_accuracy": 0.15823127031326295,
      "num_tokens": 5033046.0,
      "step": 2765
    },
    {
      "entropy": 6.196396017074585,
      "epoch": 2.3798882681564244,
      "grad_norm": 1.125,
      "learning_rate": 0.00046994431739785114,
      "loss": 5.8026,
      "mean_token_accuracy": 0.1763622686266899,
      "num_tokens": 5040894.0,
      "step": 2770
    },
    {
      "entropy": 6.190552377700806,
      "epoch": 2.384185646755479,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046977809021624454,
      "loss": 5.9674,
      "mean_token_accuracy": 0.16489340364933014,
      "num_tokens": 5050961.0,
      "step": 2775
    },
    {
      "entropy": 6.236513423919678,
      "epoch": 2.3884830253545335,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004696114377171409,
      "loss": 5.9022,
      "mean_token_accuracy": 0.1675196796655655,
      "num_tokens": 5060226.0,
      "step": 2780
    },
    {
      "entropy": 6.134419250488281,
      "epoch": 2.3927804039535885,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004694443602644429,
      "loss": 5.8574,
      "mean_token_accuracy": 0.1658073604106903,
      "num_tokens": 5069225.0,
      "step": 2785
    },
    {
      "entropy": 6.230590677261352,
      "epoch": 2.397077782552643,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004692768582229808,
      "loss": 5.8537,
      "mean_token_accuracy": 0.16909998953342437,
      "num_tokens": 5078386.0,
      "step": 2790
    },
    {
      "entropy": 6.158745527267456,
      "epoch": 2.4013751611516976,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00046910893195851213,
      "loss": 5.7812,
      "mean_token_accuracy": 0.16957359611988068,
      "num_tokens": 5087161.0,
      "step": 2795
    },
    {
      "entropy": 6.144252824783325,
      "epoch": 2.405672539750752,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046894058183772074,
      "loss": 5.9448,
      "mean_token_accuracy": 0.16215559244155883,
      "num_tokens": 5096613.0,
      "step": 2800
    },
    {
      "entropy": 6.177387857437134,
      "epoch": 2.4099699183498067,
      "grad_norm": 1.0625,
      "learning_rate": 0.000468771808228216,
      "loss": 5.9009,
      "mean_token_accuracy": 0.16563187092542647,
      "num_tokens": 5106534.0,
      "step": 2805
    },
    {
      "entropy": 6.154799270629883,
      "epoch": 2.414267296948861,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046860261149853197,
      "loss": 5.9281,
      "mean_token_accuracy": 0.16382392197847367,
      "num_tokens": 5115975.0,
      "step": 2810
    },
    {
      "entropy": 6.148438119888306,
      "epoch": 2.4185646755479158,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004684329920181268,
      "loss": 5.8216,
      "mean_token_accuracy": 0.17049064636230468,
      "num_tokens": 5124635.0,
      "step": 2815
    },
    {
      "entropy": 6.1651019096374515,
      "epoch": 2.4228620541469703,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046826295015738154,
      "loss": 5.7785,
      "mean_token_accuracy": 0.17763484865427018,
      "num_tokens": 5133226.0,
      "step": 2820
    },
    {
      "entropy": 6.099349737167358,
      "epoch": 2.427159432746025,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004680924862875996,
      "loss": 5.8788,
      "mean_token_accuracy": 0.170813749730587,
      "num_tokens": 5142257.0,
      "step": 2825
    },
    {
      "entropy": 6.231349277496338,
      "epoch": 2.4314568113450794,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046792160078100605,
      "loss": 5.8826,
      "mean_token_accuracy": 0.16604618430137635,
      "num_tokens": 5150752.0,
      "step": 2830
    },
    {
      "entropy": 6.136086511611938,
      "epoch": 2.435754189944134,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046775029401074653,
      "loss": 5.7844,
      "mean_token_accuracy": 0.178993821144104,
      "num_tokens": 5160237.0,
      "step": 2835
    },
    {
      "entropy": 6.202201557159424,
      "epoch": 2.4400515685431885,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046757856635088645,
      "loss": 5.8537,
      "mean_token_accuracy": 0.1712223544716835,
      "num_tokens": 5169752.0,
      "step": 2840
    },
    {
      "entropy": 6.172966527938843,
      "epoch": 2.444348947142243,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004674064181764105,
      "loss": 5.9056,
      "mean_token_accuracy": 0.1733956053853035,
      "num_tokens": 5178892.0,
      "step": 2845
    },
    {
      "entropy": 6.177707719802856,
      "epoch": 2.448646325741298,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046723384986322147,
      "loss": 5.8777,
      "mean_token_accuracy": 0.16896003782749175,
      "num_tokens": 5188468.0,
      "step": 2850
    },
    {
      "entropy": 6.167446327209473,
      "epoch": 2.4529437043403526,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004670608617881395,
      "loss": 5.8142,
      "mean_token_accuracy": 0.17425985783338546,
      "num_tokens": 5197565.0,
      "step": 2855
    },
    {
      "entropy": 6.108637952804566,
      "epoch": 2.457241082939407,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004668874543289014,
      "loss": 5.7986,
      "mean_token_accuracy": 0.18018365651369095,
      "num_tokens": 5205791.0,
      "step": 2860
    },
    {
      "entropy": 6.141421508789063,
      "epoch": 2.4615384615384617,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046671362786415986,
      "loss": 5.8011,
      "mean_token_accuracy": 0.18125482499599457,
      "num_tokens": 5214773.0,
      "step": 2865
    },
    {
      "entropy": 6.088820695877075,
      "epoch": 2.465835840137516,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046653938277348237,
      "loss": 5.8487,
      "mean_token_accuracy": 0.17472687363624573,
      "num_tokens": 5223734.0,
      "step": 2870
    },
    {
      "entropy": 6.261274719238282,
      "epoch": 2.4701332187365708,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004663647194373505,
      "loss": 5.9156,
      "mean_token_accuracy": 0.16283378154039382,
      "num_tokens": 5231742.0,
      "step": 2875
    },
    {
      "entropy": 6.124586343765259,
      "epoch": 2.4744305973356253,
      "grad_norm": 0.9375,
      "learning_rate": 0.00046618963823715913,
      "loss": 5.8653,
      "mean_token_accuracy": 0.1686026558279991,
      "num_tokens": 5241673.0,
      "step": 2880
    },
    {
      "entropy": 6.198731851577759,
      "epoch": 2.47872797593468,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046601413955521575,
      "loss": 5.8336,
      "mean_token_accuracy": 0.1733954593539238,
      "num_tokens": 5250082.0,
      "step": 2885
    },
    {
      "entropy": 6.167904710769653,
      "epoch": 2.4830253545337344,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004658382237747393,
      "loss": 5.894,
      "mean_token_accuracy": 0.16233221739530562,
      "num_tokens": 5259680.0,
      "step": 2890
    },
    {
      "entropy": 6.210194540023804,
      "epoch": 2.487322733132789,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00046566189127985946,
      "loss": 5.8836,
      "mean_token_accuracy": 0.17433235198259353,
      "num_tokens": 5269561.0,
      "step": 2895
    },
    {
      "entropy": 6.169331502914429,
      "epoch": 2.4916201117318435,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000465485142455616,
      "loss": 5.8388,
      "mean_token_accuracy": 0.1740201309323311,
      "num_tokens": 5278659.0,
      "step": 2900
    },
    {
      "entropy": 6.07768177986145,
      "epoch": 2.495917490330898,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046530797768795765,
      "loss": 5.801,
      "mean_token_accuracy": 0.17944221645593644,
      "num_tokens": 5287619.0,
      "step": 2905
    },
    {
      "entropy": 6.17024712562561,
      "epoch": 2.5002148689299526,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00046513039736374153,
      "loss": 5.9382,
      "mean_token_accuracy": 0.16346622705459596,
      "num_tokens": 5297334.0,
      "step": 2910
    },
    {
      "entropy": 6.215419769287109,
      "epoch": 2.504512247529007,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004649524018707319,
      "loss": 5.8424,
      "mean_token_accuracy": 0.16681308597326278,
      "num_tokens": 5306208.0,
      "step": 2915
    },
    {
      "entropy": 6.142744636535644,
      "epoch": 2.5088096261280617,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046477399159759996,
      "loss": 5.7948,
      "mean_token_accuracy": 0.17883567363023758,
      "num_tokens": 5314754.0,
      "step": 2920
    },
    {
      "entropy": 6.037441778182983,
      "epoch": 2.5131070047271162,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046459516693392246,
      "loss": 5.8052,
      "mean_token_accuracy": 0.17655965834856033,
      "num_tokens": 5324000.0,
      "step": 2925
    },
    {
      "entropy": 6.213449382781983,
      "epoch": 2.517404383326171,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004644159282701808,
      "loss": 5.8495,
      "mean_token_accuracy": 0.16930186077952386,
      "num_tokens": 5332478.0,
      "step": 2930
    },
    {
      "entropy": 6.226251840591431,
      "epoch": 2.5217017619252258,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046423627599776076,
      "loss": 5.9213,
      "mean_token_accuracy": 0.15651765167713166,
      "num_tokens": 5341635.0,
      "step": 2935
    },
    {
      "entropy": 6.143741273880005,
      "epoch": 2.5259991405242803,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000464056210508951,
      "loss": 5.9163,
      "mean_token_accuracy": 0.16428655087947847,
      "num_tokens": 5350144.0,
      "step": 2940
    },
    {
      "entropy": 6.176555585861206,
      "epoch": 2.530296519123335,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004638757321969426,
      "loss": 5.8319,
      "mean_token_accuracy": 0.16960337311029433,
      "num_tokens": 5358788.0,
      "step": 2945
    },
    {
      "entropy": 6.169870615005493,
      "epoch": 2.5345938977223894,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046369484145582815,
      "loss": 5.9296,
      "mean_token_accuracy": 0.16125051379203797,
      "num_tokens": 5368057.0,
      "step": 2950
    },
    {
      "entropy": 6.0688502311706545,
      "epoch": 2.538891276321444,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046351353868060054,
      "loss": 5.7804,
      "mean_token_accuracy": 0.17502057552337646,
      "num_tokens": 5376739.0,
      "step": 2955
    },
    {
      "entropy": 6.1720726013183596,
      "epoch": 2.5431886549204985,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046333182426715273,
      "loss": 5.8841,
      "mean_token_accuracy": 0.16769777685403825,
      "num_tokens": 5385967.0,
      "step": 2960
    },
    {
      "entropy": 6.232366323471069,
      "epoch": 2.547486033519553,
      "grad_norm": 1.0,
      "learning_rate": 0.00046314969861227626,
      "loss": 5.9187,
      "mean_token_accuracy": 0.16190613955259323,
      "num_tokens": 5395192.0,
      "step": 2965
    },
    {
      "entropy": 6.1295037269592285,
      "epoch": 2.5517834121186076,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004629671621136608,
      "loss": 5.876,
      "mean_token_accuracy": 0.17085160166025162,
      "num_tokens": 5404694.0,
      "step": 2970
    },
    {
      "entropy": 6.156940174102783,
      "epoch": 2.556080790717662,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004627842151698931,
      "loss": 5.8791,
      "mean_token_accuracy": 0.16839216351509095,
      "num_tokens": 5413102.0,
      "step": 2975
    },
    {
      "entropy": 6.141358661651611,
      "epoch": 2.5603781693167167,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046260085818045625,
      "loss": 5.9057,
      "mean_token_accuracy": 0.16761772185564042,
      "num_tokens": 5423339.0,
      "step": 2980
    },
    {
      "entropy": 6.211736726760864,
      "epoch": 2.5646755479157712,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004624170915457284,
      "loss": 5.8683,
      "mean_token_accuracy": 0.17020333409309388,
      "num_tokens": 5432377.0,
      "step": 2985
    },
    {
      "entropy": 6.145984411239624,
      "epoch": 2.5689729265148262,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046223291566698264,
      "loss": 5.787,
      "mean_token_accuracy": 0.17505069077014923,
      "num_tokens": 5441038.0,
      "step": 2990
    },
    {
      "entropy": 6.096357202529907,
      "epoch": 2.5732703051138808,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004620483309463855,
      "loss": 5.8135,
      "mean_token_accuracy": 0.17221614569425583,
      "num_tokens": 5449557.0,
      "step": 2995
    },
    {
      "entropy": 6.217533254623413,
      "epoch": 2.5775676837129353,
      "grad_norm": 1.0,
      "learning_rate": 0.0004618633377869961,
      "loss": 5.9446,
      "mean_token_accuracy": 0.15857684314250947,
      "num_tokens": 5458931.0,
      "step": 3000
    },
    {
      "epoch": 2.5775676837129353,
      "eval_entropy": 6.002345256977253,
      "eval_loss": 6.1212897300720215,
      "eval_mean_token_accuracy": 0.16277532707396392,
      "eval_num_tokens": 5458931.0,
      "eval_runtime": 2.0558,
      "eval_samples_per_second": 1726.341,
      "eval_steps_per_second": 215.975,
      "step": 3000
    },
    {
      "entropy": 6.1347846508026125,
      "epoch": 2.58186506231199,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004616779365927656,
      "loss": 5.7734,
      "mean_token_accuracy": 0.1822523444890976,
      "num_tokens": 5468539.0,
      "step": 3005
    },
    {
      "entropy": 6.024096488952637,
      "epoch": 2.5861624409110444,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004614921277685361,
      "loss": 5.7085,
      "mean_token_accuracy": 0.17936478108167647,
      "num_tokens": 5475710.0,
      "step": 3010
    },
    {
      "entropy": 6.1361793041229244,
      "epoch": 2.590459819510099,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046130591172003976,
      "loss": 5.8527,
      "mean_token_accuracy": 0.17134219259023667,
      "num_tokens": 5484597.0,
      "step": 3015
    },
    {
      "entropy": 6.120989418029785,
      "epoch": 2.5947571981091535,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004611192888538981,
      "loss": 5.9411,
      "mean_token_accuracy": 0.1608140453696251,
      "num_tokens": 5493213.0,
      "step": 3020
    },
    {
      "entropy": 6.2168285846710205,
      "epoch": 2.599054576708208,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046093225957762084,
      "loss": 5.9308,
      "mean_token_accuracy": 0.16593084782361983,
      "num_tokens": 5502556.0,
      "step": 3025
    },
    {
      "entropy": 6.197446489334107,
      "epoch": 2.6033519553072626,
      "grad_norm": 1.0,
      "learning_rate": 0.0004607448242996051,
      "loss": 5.8398,
      "mean_token_accuracy": 0.17084067910909653,
      "num_tokens": 5511779.0,
      "step": 3030
    },
    {
      "entropy": 6.1454957008361815,
      "epoch": 2.607649333906317,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004605569834291347,
      "loss": 5.8161,
      "mean_token_accuracy": 0.17526462078094482,
      "num_tokens": 5520836.0,
      "step": 3035
    },
    {
      "entropy": 6.083783388137817,
      "epoch": 2.6119467125053717,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046036873737637904,
      "loss": 5.8457,
      "mean_token_accuracy": 0.17209848016500473,
      "num_tokens": 5529285.0,
      "step": 3040
    },
    {
      "entropy": 6.101385116577148,
      "epoch": 2.6162440911044262,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004601800865523921,
      "loss": 5.8519,
      "mean_token_accuracy": 0.16869369894266129,
      "num_tokens": 5538160.0,
      "step": 3045
    },
    {
      "entropy": 6.142069292068482,
      "epoch": 2.620541469703481,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045999103136911204,
      "loss": 5.8559,
      "mean_token_accuracy": 0.16590654104948044,
      "num_tokens": 5547355.0,
      "step": 3050
    },
    {
      "entropy": 6.143155193328857,
      "epoch": 2.6248388483025353,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00045980157223935965,
      "loss": 5.8708,
      "mean_token_accuracy": 0.17009032517671585,
      "num_tokens": 5557299.0,
      "step": 3055
    },
    {
      "entropy": 6.0996771335601805,
      "epoch": 2.62913622690159,
      "grad_norm": 0.96875,
      "learning_rate": 0.00045961170957683806,
      "loss": 5.7957,
      "mean_token_accuracy": 0.1701590210199356,
      "num_tokens": 5565469.0,
      "step": 3060
    },
    {
      "entropy": 6.1573389053344725,
      "epoch": 2.6334336055006444,
      "grad_norm": 0.984375,
      "learning_rate": 0.00045942144379613147,
      "loss": 5.9159,
      "mean_token_accuracy": 0.1658922776579857,
      "num_tokens": 5574740.0,
      "step": 3065
    },
    {
      "entropy": 6.166552925109864,
      "epoch": 2.637730984099699,
      "grad_norm": 1.0,
      "learning_rate": 0.00045923077531270426,
      "loss": 5.8854,
      "mean_token_accuracy": 0.16845837086439133,
      "num_tokens": 5583438.0,
      "step": 3070
    },
    {
      "entropy": 6.17145938873291,
      "epoch": 2.6420283626987535,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004590397045429001,
      "loss": 5.857,
      "mean_token_accuracy": 0.17479295134544373,
      "num_tokens": 5592389.0,
      "step": 3075
    },
    {
      "entropy": 6.079704666137696,
      "epoch": 2.646325741297808,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00045884823190394134,
      "loss": 5.7566,
      "mean_token_accuracy": 0.1772332966327667,
      "num_tokens": 5601598.0,
      "step": 3080
    },
    {
      "entropy": 6.118323564529419,
      "epoch": 2.650623119896863,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004586563578139275,
      "loss": 5.8446,
      "mean_token_accuracy": 0.16692771315574645,
      "num_tokens": 5610498.0,
      "step": 3085
    },
    {
      "entropy": 6.109970378875732,
      "epoch": 2.6549204984959176,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045846408269183505,
      "loss": 5.7654,
      "mean_token_accuracy": 0.1782268226146698,
      "num_tokens": 5620082.0,
      "step": 3090
    },
    {
      "entropy": 6.152325868606567,
      "epoch": 2.659217877094972,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00045827140695751603,
      "loss": 5.8366,
      "mean_token_accuracy": 0.1716820165514946,
      "num_tokens": 5630291.0,
      "step": 3095
    },
    {
      "entropy": 6.097200155258179,
      "epoch": 2.6635152556940267,
      "grad_norm": 1.125,
      "learning_rate": 0.0004580783310316971,
      "loss": 5.8345,
      "mean_token_accuracy": 0.17504153102636338,
      "num_tokens": 5638784.0,
      "step": 3100
    },
    {
      "entropy": 6.027546215057373,
      "epoch": 2.6678126342930812,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045788485533597895,
      "loss": 5.6879,
      "mean_token_accuracy": 0.18603624403476715,
      "num_tokens": 5647968.0,
      "step": 3105
    },
    {
      "entropy": 6.150567054748535,
      "epoch": 2.672110012892136,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045769098029283526,
      "loss": 5.9347,
      "mean_token_accuracy": 0.16201953887939452,
      "num_tokens": 5657543.0,
      "step": 3110
    },
    {
      "entropy": 6.15388331413269,
      "epoch": 2.6764073914911903,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004574967063256115,
      "loss": 5.8353,
      "mean_token_accuracy": 0.17699069976806642,
      "num_tokens": 5666535.0,
      "step": 3115
    },
    {
      "entropy": 6.084828329086304,
      "epoch": 2.680704770090245,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045730203385852447,
      "loss": 5.9196,
      "mean_token_accuracy": 0.16706804931163788,
      "num_tokens": 5676273.0,
      "step": 3120
    },
    {
      "entropy": 6.041564130783081,
      "epoch": 2.6850021486892994,
      "grad_norm": 1.0,
      "learning_rate": 0.000457106963316661,
      "loss": 5.8215,
      "mean_token_accuracy": 0.17681703120470046,
      "num_tokens": 5684888.0,
      "step": 3125
    },
    {
      "entropy": 6.1651557922363285,
      "epoch": 2.689299527288354,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045691149512597717,
      "loss": 5.8817,
      "mean_token_accuracy": 0.16579522490501403,
      "num_tokens": 5693626.0,
      "step": 3130
    },
    {
      "entropy": 6.171338653564453,
      "epoch": 2.6935969058874085,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00045671562971329736,
      "loss": 5.7997,
      "mean_token_accuracy": 0.17142064571380616,
      "num_tokens": 5702542.0,
      "step": 3135
    },
    {
      "entropy": 6.036250162124634,
      "epoch": 2.6978942844864635,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045651936750631337,
      "loss": 5.8332,
      "mean_token_accuracy": 0.17233078181743622,
      "num_tokens": 5711440.0,
      "step": 3140
    },
    {
      "entropy": 6.216920471191406,
      "epoch": 2.702191663085518,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045632270893358333,
      "loss": 5.8818,
      "mean_token_accuracy": 0.16918585151433946,
      "num_tokens": 5721495.0,
      "step": 3145
    },
    {
      "entropy": 6.1993273258209225,
      "epoch": 2.7064890416845726,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004561256544245312,
      "loss": 5.9189,
      "mean_token_accuracy": 0.1607155531644821,
      "num_tokens": 5730664.0,
      "step": 3150
    },
    {
      "entropy": 6.047014474868774,
      "epoch": 2.710786420283627,
      "grad_norm": 1.046875,
      "learning_rate": 0.000455928204409445,
      "loss": 5.8044,
      "mean_token_accuracy": 0.17973511815071105,
      "num_tokens": 5740229.0,
      "step": 3155
    },
    {
      "entropy": 6.113893842697143,
      "epoch": 2.7150837988826817,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045573035931947684,
      "loss": 5.802,
      "mean_token_accuracy": 0.17425440698862077,
      "num_tokens": 5748549.0,
      "step": 3160
    },
    {
      "entropy": 6.111674499511719,
      "epoch": 2.7193811774817362,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004555321195866411,
      "loss": 5.7391,
      "mean_token_accuracy": 0.17396723330020905,
      "num_tokens": 5757603.0,
      "step": 3165
    },
    {
      "entropy": 6.137631750106811,
      "epoch": 2.723678556080791,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004553334856438143,
      "loss": 5.9224,
      "mean_token_accuracy": 0.16666966825723648,
      "num_tokens": 5767520.0,
      "step": 3170
    },
    {
      "entropy": 6.1654016971588135,
      "epoch": 2.7279759346798453,
      "grad_norm": 0.9375,
      "learning_rate": 0.00045513445792473356,
      "loss": 5.911,
      "mean_token_accuracy": 0.16542696654796601,
      "num_tokens": 5776778.0,
      "step": 3175
    },
    {
      "entropy": 6.174024152755737,
      "epoch": 2.7322733132789,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004549350368639958,
      "loss": 5.9451,
      "mean_token_accuracy": 0.16035651415586472,
      "num_tokens": 5785652.0,
      "step": 3180
    },
    {
      "entropy": 6.2061694145202635,
      "epoch": 2.7365706918779544,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045473522289705693,
      "loss": 5.891,
      "mean_token_accuracy": 0.1694064900279045,
      "num_tokens": 5795766.0,
      "step": 3185
    },
    {
      "entropy": 6.0452875137329105,
      "epoch": 2.740868070477009,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045453501646023085,
      "loss": 5.9372,
      "mean_token_accuracy": 0.16121204793453217,
      "num_tokens": 5804504.0,
      "step": 3190
    },
    {
      "entropy": 6.124906301498413,
      "epoch": 2.7451654490760635,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00045433441799068837,
      "loss": 5.84,
      "mean_token_accuracy": 0.17111808210611343,
      "num_tokens": 5814161.0,
      "step": 3195
    },
    {
      "entropy": 6.147184801101685,
      "epoch": 2.749462827675118,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004541334279264562,
      "loss": 5.7407,
      "mean_token_accuracy": 0.18340681195259095,
      "num_tokens": 5822235.0,
      "step": 3200
    },
    {
      "entropy": 6.072987508773804,
      "epoch": 2.7537602062741726,
      "grad_norm": 1.125,
      "learning_rate": 0.00045393204670641656,
      "loss": 5.7547,
      "mean_token_accuracy": 0.1745822474360466,
      "num_tokens": 5831572.0,
      "step": 3205
    },
    {
      "entropy": 6.005992603302002,
      "epoch": 2.758057584873227,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004537302747703055,
      "loss": 5.7829,
      "mean_token_accuracy": 0.17934428602457048,
      "num_tokens": 5839694.0,
      "step": 3210
    },
    {
      "entropy": 6.185523939132691,
      "epoch": 2.7623549634722817,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045352811255871216,
      "loss": 5.9086,
      "mean_token_accuracy": 0.16498518288135527,
      "num_tokens": 5849131.0,
      "step": 3215
    },
    {
      "entropy": 6.20290732383728,
      "epoch": 2.7666523420713363,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00045332556051307804,
      "loss": 5.8267,
      "mean_token_accuracy": 0.16665343046188355,
      "num_tokens": 5858861.0,
      "step": 3220
    },
    {
      "entropy": 6.123448705673217,
      "epoch": 2.770949720670391,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045312261907569585,
      "loss": 5.8263,
      "mean_token_accuracy": 0.17210643440485002,
      "num_tokens": 5867585.0,
      "step": 3225
    },
    {
      "entropy": 6.09133882522583,
      "epoch": 2.775247099269446,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045291928868970867,
      "loss": 5.8557,
      "mean_token_accuracy": 0.16730334907770156,
      "num_tokens": 5876256.0,
      "step": 3230
    },
    {
      "entropy": 6.103731679916382,
      "epoch": 2.7795444778685003,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004527155697991087,
      "loss": 5.9124,
      "mean_token_accuracy": 0.16498502045869828,
      "num_tokens": 5885302.0,
      "step": 3235
    },
    {
      "entropy": 6.116323661804199,
      "epoch": 2.783841856467555,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004525114628487365,
      "loss": 5.928,
      "mean_token_accuracy": 0.1643183633685112,
      "num_tokens": 5895066.0,
      "step": 3240
    },
    {
      "entropy": 6.18259449005127,
      "epoch": 2.7881392350666094,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045230696828428026,
      "loss": 5.9034,
      "mean_token_accuracy": 0.16625797897577285,
      "num_tokens": 5903258.0,
      "step": 3245
    },
    {
      "entropy": 6.126916885375977,
      "epoch": 2.792436613665664,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004521020865522742,
      "loss": 5.7972,
      "mean_token_accuracy": 0.171038818359375,
      "num_tokens": 5911714.0,
      "step": 3250
    },
    {
      "entropy": 6.091484212875367,
      "epoch": 2.7967339922647185,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045189681810009827,
      "loss": 5.8604,
      "mean_token_accuracy": 0.1673845484852791,
      "num_tokens": 5920432.0,
      "step": 3255
    },
    {
      "entropy": 6.245140838623047,
      "epoch": 2.801031370863773,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045169116337597653,
      "loss": 5.8792,
      "mean_token_accuracy": 0.16876531690359114,
      "num_tokens": 5929202.0,
      "step": 3260
    },
    {
      "entropy": 6.171068525314331,
      "epoch": 2.8053287494628276,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000451485122828977,
      "loss": 5.9096,
      "mean_token_accuracy": 0.1670824408531189,
      "num_tokens": 5938034.0,
      "step": 3265
    },
    {
      "entropy": 6.007605361938476,
      "epoch": 2.809626128061882,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00045127869690900956,
      "loss": 5.7687,
      "mean_token_accuracy": 0.17781976610422134,
      "num_tokens": 5946944.0,
      "step": 3270
    },
    {
      "entropy": 6.040731048583984,
      "epoch": 2.8139235066609367,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045107188606682613,
      "loss": 5.8725,
      "mean_token_accuracy": 0.17112763077020646,
      "num_tokens": 5956475.0,
      "step": 3275
    },
    {
      "entropy": 6.213654851913452,
      "epoch": 2.8182208852599913,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004508646907540188,
      "loss": 5.8253,
      "mean_token_accuracy": 0.16726296544075012,
      "num_tokens": 5965814.0,
      "step": 3280
    },
    {
      "entropy": 6.115692520141602,
      "epoch": 2.8225182638590463,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004506571114230195,
      "loss": 5.8797,
      "mean_token_accuracy": 0.16483918130397796,
      "num_tokens": 5973850.0,
      "step": 3285
    },
    {
      "entropy": 6.044183731079102,
      "epoch": 2.826815642458101,
      "grad_norm": 0.984375,
      "learning_rate": 0.00045044914852709824,
      "loss": 5.8337,
      "mean_token_accuracy": 0.16715002954006195,
      "num_tokens": 5982987.0,
      "step": 3290
    },
    {
      "entropy": 6.176396608352661,
      "epoch": 2.8311130210571553,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004502408025203631,
      "loss": 5.8088,
      "mean_token_accuracy": 0.17391630709171296,
      "num_tokens": 5992227.0,
      "step": 3295
    },
    {
      "entropy": 6.108564949035644,
      "epoch": 2.83541039965621,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004500320738577584,
      "loss": 5.8006,
      "mean_token_accuracy": 0.17633765786886216,
      "num_tokens": 6000243.0,
      "step": 3300
    },
    {
      "entropy": 6.053935432434082,
      "epoch": 2.8397077782552644,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044982296299506407,
      "loss": 5.7997,
      "mean_token_accuracy": 0.17893847674131394,
      "num_tokens": 6009771.0,
      "step": 3305
    },
    {
      "entropy": 6.1164305210113525,
      "epoch": 2.844005156854319,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004496134703888948,
      "loss": 5.875,
      "mean_token_accuracy": 0.1690862849354744,
      "num_tokens": 6018683.0,
      "step": 3310
    },
    {
      "entropy": 6.122239971160889,
      "epoch": 2.8483025354533735,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044940359649669846,
      "loss": 5.7361,
      "mean_token_accuracy": 0.18076154738664627,
      "num_tokens": 6027422.0,
      "step": 3315
    },
    {
      "entropy": 6.093537425994873,
      "epoch": 2.852599914052428,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044919334177675595,
      "loss": 5.8232,
      "mean_token_accuracy": 0.16410151720046998,
      "num_tokens": 6035670.0,
      "step": 3320
    },
    {
      "entropy": 6.088628053665161,
      "epoch": 2.8568972926514826,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044898270668817955,
      "loss": 5.7633,
      "mean_token_accuracy": 0.177126844227314,
      "num_tokens": 6044092.0,
      "step": 3325
    },
    {
      "entropy": 6.090199422836304,
      "epoch": 2.861194671250537,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000448771691690912,
      "loss": 5.8325,
      "mean_token_accuracy": 0.16831086575984955,
      "num_tokens": 6053970.0,
      "step": 3330
    },
    {
      "entropy": 6.140215587615967,
      "epoch": 2.8654920498495917,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004485602972457257,
      "loss": 5.7898,
      "mean_token_accuracy": 0.1754717782139778,
      "num_tokens": 6062965.0,
      "step": 3335
    },
    {
      "entropy": 6.150375843048096,
      "epoch": 2.8697894284486463,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044834852381422165,
      "loss": 5.8695,
      "mean_token_accuracy": 0.17155641317367554,
      "num_tokens": 6072420.0,
      "step": 3340
    },
    {
      "entropy": 6.0596644401550295,
      "epoch": 2.874086807047701,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044813637185882836,
      "loss": 5.7734,
      "mean_token_accuracy": 0.1735791265964508,
      "num_tokens": 6080915.0,
      "step": 3345
    },
    {
      "entropy": 6.116873121261596,
      "epoch": 2.8783841856467554,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044792384184280106,
      "loss": 5.8954,
      "mean_token_accuracy": 0.16623952835798264,
      "num_tokens": 6090453.0,
      "step": 3350
    },
    {
      "entropy": 6.090284633636474,
      "epoch": 2.88268156424581,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044771093423022013,
      "loss": 5.9298,
      "mean_token_accuracy": 0.15913961827754974,
      "num_tokens": 6099390.0,
      "step": 3355
    },
    {
      "entropy": 6.114023637771607,
      "epoch": 2.8869789428448644,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004474976494859909,
      "loss": 5.8486,
      "mean_token_accuracy": 0.17675736397504807,
      "num_tokens": 6108677.0,
      "step": 3360
    },
    {
      "entropy": 6.098732328414917,
      "epoch": 2.891276321443919,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004472839880758419,
      "loss": 5.7711,
      "mean_token_accuracy": 0.17180029749870301,
      "num_tokens": 6117151.0,
      "step": 3365
    },
    {
      "entropy": 6.195354080200195,
      "epoch": 2.8955737000429735,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004470699504663242,
      "loss": 5.8942,
      "mean_token_accuracy": 0.16719547659158707,
      "num_tokens": 6127167.0,
      "step": 3370
    },
    {
      "entropy": 6.082501411437988,
      "epoch": 2.899871078642028,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004468555371248104,
      "loss": 5.7776,
      "mean_token_accuracy": 0.17165370136499405,
      "num_tokens": 6136487.0,
      "step": 3375
    },
    {
      "entropy": 6.114503192901611,
      "epoch": 2.904168457241083,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004466407485194937,
      "loss": 5.8846,
      "mean_token_accuracy": 0.16588175743818284,
      "num_tokens": 6145334.0,
      "step": 3380
    },
    {
      "entropy": 6.101830673217774,
      "epoch": 2.9084658358401376,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004464255851193864,
      "loss": 5.8134,
      "mean_token_accuracy": 0.16928868889808654,
      "num_tokens": 6155062.0,
      "step": 3385
    },
    {
      "entropy": 6.060112714767456,
      "epoch": 2.912763214439192,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004462100473943194,
      "loss": 5.779,
      "mean_token_accuracy": 0.17902038842439652,
      "num_tokens": 6164313.0,
      "step": 3390
    },
    {
      "entropy": 6.066328811645508,
      "epoch": 2.9170605930382467,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000445994135814941,
      "loss": 5.8031,
      "mean_token_accuracy": 0.17068170607089997,
      "num_tokens": 6173513.0,
      "step": 3395
    },
    {
      "entropy": 6.1107556343078615,
      "epoch": 2.9213579716373013,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044577785085271566,
      "loss": 5.8222,
      "mean_token_accuracy": 0.16928994953632354,
      "num_tokens": 6182000.0,
      "step": 3400
    },
    {
      "entropy": 6.177258014678955,
      "epoch": 2.925655350236356,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004455611929799235,
      "loss": 5.8653,
      "mean_token_accuracy": 0.15872504711151122,
      "num_tokens": 6191887.0,
      "step": 3405
    },
    {
      "entropy": 6.03788800239563,
      "epoch": 2.9299527288354104,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004453441626696585,
      "loss": 5.8994,
      "mean_token_accuracy": 0.15877071171998977,
      "num_tokens": 6202897.0,
      "step": 3410
    },
    {
      "entropy": 6.142374467849732,
      "epoch": 2.934250107434465,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044512676039582823,
      "loss": 5.7983,
      "mean_token_accuracy": 0.17737043499946595,
      "num_tokens": 6211811.0,
      "step": 3415
    },
    {
      "entropy": 6.133689117431641,
      "epoch": 2.9385474860335195,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004449089866331524,
      "loss": 5.807,
      "mean_token_accuracy": 0.17649049013853074,
      "num_tokens": 6219896.0,
      "step": 3420
    },
    {
      "entropy": 6.010902643203735,
      "epoch": 2.942844864632574,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004446908418571617,
      "loss": 5.7923,
      "mean_token_accuracy": 0.17107276171445845,
      "num_tokens": 6228212.0,
      "step": 3425
    },
    {
      "entropy": 6.108887815475464,
      "epoch": 2.9471422432316285,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004444723265441973,
      "loss": 5.9385,
      "mean_token_accuracy": 0.16253667175769806,
      "num_tokens": 6238133.0,
      "step": 3430
    },
    {
      "entropy": 6.079803562164306,
      "epoch": 2.9514396218306835,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004442534411714092,
      "loss": 5.8493,
      "mean_token_accuracy": 0.1624876856803894,
      "num_tokens": 6247331.0,
      "step": 3435
    },
    {
      "entropy": 6.170568227767944,
      "epoch": 2.955737000429738,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044403418621675555,
      "loss": 5.8456,
      "mean_token_accuracy": 0.1654752418398857,
      "num_tokens": 6255280.0,
      "step": 3440
    },
    {
      "entropy": 6.140178108215332,
      "epoch": 2.9600343790287926,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004438145621590017,
      "loss": 5.8073,
      "mean_token_accuracy": 0.17526279389858246,
      "num_tokens": 6264752.0,
      "step": 3445
    },
    {
      "entropy": 6.028224134445191,
      "epoch": 2.964331757627847,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044359456947771857,
      "loss": 5.7614,
      "mean_token_accuracy": 0.1661372035741806,
      "num_tokens": 6273258.0,
      "step": 3450
    },
    {
      "entropy": 5.952836751937866,
      "epoch": 2.9686291362269017,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004433742086532824,
      "loss": 5.6712,
      "mean_token_accuracy": 0.18851299732923507,
      "num_tokens": 6281584.0,
      "step": 3455
    },
    {
      "entropy": 6.098459720611572,
      "epoch": 2.9729265148259563,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044315348016687317,
      "loss": 5.8038,
      "mean_token_accuracy": 0.17163048237562178,
      "num_tokens": 6290016.0,
      "step": 3460
    },
    {
      "entropy": 6.039625978469848,
      "epoch": 2.977223893425011,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004429323845004736,
      "loss": 5.7066,
      "mean_token_accuracy": 0.17723663598299028,
      "num_tokens": 6298569.0,
      "step": 3465
    },
    {
      "entropy": 6.064740991592407,
      "epoch": 2.9815212720240654,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00044271092213686824,
      "loss": 5.7313,
      "mean_token_accuracy": 0.17582760155200958,
      "num_tokens": 6307684.0,
      "step": 3470
    },
    {
      "entropy": 6.119738626480102,
      "epoch": 2.98581865062312,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00044248909355964247,
      "loss": 5.8843,
      "mean_token_accuracy": 0.16779844611883163,
      "num_tokens": 6317767.0,
      "step": 3475
    },
    {
      "entropy": 6.1805627822875975,
      "epoch": 2.9901160292221745,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044226689925318117,
      "loss": 5.8958,
      "mean_token_accuracy": 0.16598885059356688,
      "num_tokens": 6327457.0,
      "step": 3480
    },
    {
      "entropy": 6.034802579879761,
      "epoch": 2.994413407821229,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044204433970266785,
      "loss": 5.6993,
      "mean_token_accuracy": 0.18884851187467575,
      "num_tokens": 6335747.0,
      "step": 3485
    },
    {
      "entropy": 6.0412510395050045,
      "epoch": 2.9987107864202835,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004418214153940837,
      "loss": 5.7915,
      "mean_token_accuracy": 0.1731286823749542,
      "num_tokens": 6344750.0,
      "step": 3490
    },
    {
      "entropy": 6.094671514299181,
      "epoch": 3.002578427159433,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00044159812681420624,
      "loss": 5.7345,
      "mean_token_accuracy": 0.1751474059290356,
      "num_tokens": 6354779.0,
      "step": 3495
    },
    {
      "entropy": 6.1476243019104,
      "epoch": 3.0068758057584875,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004413744744506086,
      "loss": 5.5077,
      "mean_token_accuracy": 0.1835872769355774,
      "num_tokens": 6363809.0,
      "step": 3500
    },
    {
      "epoch": 3.0068758057584875,
      "eval_entropy": 5.855848801028621,
      "eval_loss": 6.045755863189697,
      "eval_mean_token_accuracy": 0.16792396449290956,
      "eval_num_tokens": 6363809.0,
      "eval_runtime": 2.0563,
      "eval_samples_per_second": 1725.906,
      "eval_steps_per_second": 215.921,
      "step": 3500
    },
    {
      "entropy": 6.019393825531006,
      "epoch": 3.011173184357542,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044115045879165806,
      "loss": 5.5828,
      "mean_token_accuracy": 0.18049468696117402,
      "num_tokens": 6373082.0,
      "step": 3505
    },
    {
      "entropy": 6.042645025253296,
      "epoch": 3.0154705629565965,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044092608032651515,
      "loss": 5.5465,
      "mean_token_accuracy": 0.18014225363731384,
      "num_tokens": 6381286.0,
      "step": 3510
    },
    {
      "entropy": 6.056296157836914,
      "epoch": 3.019767941555651,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00044070133954513305,
      "loss": 5.4969,
      "mean_token_accuracy": 0.1933508113026619,
      "num_tokens": 6390217.0,
      "step": 3515
    },
    {
      "entropy": 6.078204870223999,
      "epoch": 3.0240653201547056,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004404762369382555,
      "loss": 5.526,
      "mean_token_accuracy": 0.18391967862844466,
      "num_tokens": 6399276.0,
      "step": 3520
    },
    {
      "entropy": 6.0515869140625,
      "epoch": 3.02836269875376,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044025077299741683,
      "loss": 5.5016,
      "mean_token_accuracy": 0.19308611005544662,
      "num_tokens": 6407981.0,
      "step": 3525
    },
    {
      "entropy": 6.015412378311157,
      "epoch": 3.0326600773528147,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044002494821494007,
      "loss": 5.4997,
      "mean_token_accuracy": 0.18891025483608245,
      "num_tokens": 6416159.0,
      "step": 3530
    },
    {
      "entropy": 5.964804553985596,
      "epoch": 3.0369574559518693,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043979876308393635,
      "loss": 5.5364,
      "mean_token_accuracy": 0.18845383673906327,
      "num_tokens": 6424564.0,
      "step": 3535
    },
    {
      "entropy": 6.114687299728393,
      "epoch": 3.041254834550924,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004395722180983036,
      "loss": 5.5876,
      "mean_token_accuracy": 0.18196578919887543,
      "num_tokens": 6434163.0,
      "step": 3540
    },
    {
      "entropy": 5.9505010604858395,
      "epoch": 3.0455522131499784,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00043934531375272535,
      "loss": 5.3994,
      "mean_token_accuracy": 0.20380107909440995,
      "num_tokens": 6443372.0,
      "step": 3545
    },
    {
      "entropy": 5.993015623092651,
      "epoch": 3.049849591749033,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00043911805054267015,
      "loss": 5.5175,
      "mean_token_accuracy": 0.18979651629924774,
      "num_tokens": 6452638.0,
      "step": 3550
    },
    {
      "entropy": 6.1458171844482425,
      "epoch": 3.0541469703480875,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043889042896439004,
      "loss": 5.5275,
      "mean_token_accuracy": 0.18725358247756957,
      "num_tokens": 6461319.0,
      "step": 3555
    },
    {
      "entropy": 6.006018257141113,
      "epoch": 3.0584443489471425,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043866244951491946,
      "loss": 5.4399,
      "mean_token_accuracy": 0.20162247568368913,
      "num_tokens": 6469506.0,
      "step": 3560
    },
    {
      "entropy": 6.016952753067017,
      "epoch": 3.062741727546197,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043843411269207445,
      "loss": 5.4942,
      "mean_token_accuracy": 0.1971167653799057,
      "num_tokens": 6478404.0,
      "step": 3565
    },
    {
      "entropy": 5.9416725635528564,
      "epoch": 3.0670391061452515,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004382054189944514,
      "loss": 5.4333,
      "mean_token_accuracy": 0.19149247109889983,
      "num_tokens": 6487447.0,
      "step": 3570
    },
    {
      "entropy": 5.9900328636169435,
      "epoch": 3.071336484744306,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004379763689214259,
      "loss": 5.4947,
      "mean_token_accuracy": 0.18396613895893096,
      "num_tokens": 6496738.0,
      "step": 3575
    },
    {
      "entropy": 6.03482666015625,
      "epoch": 3.0756338633433606,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004377469629731518,
      "loss": 5.4889,
      "mean_token_accuracy": 0.18621663898229598,
      "num_tokens": 6505848.0,
      "step": 3580
    },
    {
      "entropy": 6.009872531890869,
      "epoch": 3.079931241942415,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004375172016505599,
      "loss": 5.4722,
      "mean_token_accuracy": 0.18729298561811447,
      "num_tokens": 6515731.0,
      "step": 3585
    },
    {
      "entropy": 6.015996217727661,
      "epoch": 3.0842286205414697,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004372870854553572,
      "loss": 5.535,
      "mean_token_accuracy": 0.19139505177736282,
      "num_tokens": 6524914.0,
      "step": 3590
    },
    {
      "entropy": 6.040276765823364,
      "epoch": 3.0885259991405243,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004370566148900255,
      "loss": 5.5266,
      "mean_token_accuracy": 0.19199742376804352,
      "num_tokens": 6533712.0,
      "step": 3595
    },
    {
      "entropy": 6.044233226776123,
      "epoch": 3.092823377739579,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043682579045782024,
      "loss": 5.5934,
      "mean_token_accuracy": 0.1833044856786728,
      "num_tokens": 6543313.0,
      "step": 3600
    },
    {
      "entropy": 5.956815242767334,
      "epoch": 3.0971207563386334,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004365946126627699,
      "loss": 5.4899,
      "mean_token_accuracy": 0.19520556181669235,
      "num_tokens": 6551634.0,
      "step": 3605
    },
    {
      "entropy": 6.048838329315186,
      "epoch": 3.101418134937688,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043636308200967433,
      "loss": 5.4857,
      "mean_token_accuracy": 0.19389465898275376,
      "num_tokens": 6560695.0,
      "step": 3610
    },
    {
      "entropy": 5.896581077575684,
      "epoch": 3.1057155135367425,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004361311990041039,
      "loss": 5.3895,
      "mean_token_accuracy": 0.19544197469949723,
      "num_tokens": 6569086.0,
      "step": 3615
    },
    {
      "entropy": 5.937729692459106,
      "epoch": 3.110012892135797,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043589896415239843,
      "loss": 5.4733,
      "mean_token_accuracy": 0.19641588479280472,
      "num_tokens": 6578287.0,
      "step": 3620
    },
    {
      "entropy": 6.008415412902832,
      "epoch": 3.1143102707348516,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00043566637796166595,
      "loss": 5.5318,
      "mean_token_accuracy": 0.18539339154958726,
      "num_tokens": 6587015.0,
      "step": 3625
    },
    {
      "entropy": 6.041479587554932,
      "epoch": 3.118607649333906,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043543344093978186,
      "loss": 5.5769,
      "mean_token_accuracy": 0.18712035566568375,
      "num_tokens": 6596187.0,
      "step": 3630
    },
    {
      "entropy": 5.977596187591553,
      "epoch": 3.122905027932961,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043520015359538745,
      "loss": 5.438,
      "mean_token_accuracy": 0.1976032465696335,
      "num_tokens": 6605226.0,
      "step": 3635
    },
    {
      "entropy": 5.904521131515503,
      "epoch": 3.1272024065320156,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004349665164378891,
      "loss": 5.4965,
      "mean_token_accuracy": 0.18549832254648208,
      "num_tokens": 6613232.0,
      "step": 3640
    },
    {
      "entropy": 5.961996459960938,
      "epoch": 3.13149978513107,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043473252997745684,
      "loss": 5.5089,
      "mean_token_accuracy": 0.18430595695972443,
      "num_tokens": 6622247.0,
      "step": 3645
    },
    {
      "entropy": 6.0388407707214355,
      "epoch": 3.1357971637301247,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00043449819472502366,
      "loss": 5.4487,
      "mean_token_accuracy": 0.18688230067491532,
      "num_tokens": 6630883.0,
      "step": 3650
    },
    {
      "entropy": 5.955649662017822,
      "epoch": 3.1400945423291793,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004342635111922841,
      "loss": 5.5868,
      "mean_token_accuracy": 0.18630024939775466,
      "num_tokens": 6639399.0,
      "step": 3655
    },
    {
      "entropy": 5.97322793006897,
      "epoch": 3.144391920928234,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004340284798916931,
      "loss": 5.4973,
      "mean_token_accuracy": 0.18773317337036133,
      "num_tokens": 6649288.0,
      "step": 3660
    },
    {
      "entropy": 5.958513021469116,
      "epoch": 3.1486892995272884,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004337931013364653,
      "loss": 5.4246,
      "mean_token_accuracy": 0.19373546093702315,
      "num_tokens": 6658670.0,
      "step": 3665
    },
    {
      "entropy": 5.95462064743042,
      "epoch": 3.152986678126343,
      "grad_norm": 1.0625,
      "learning_rate": 0.000433557376040573,
      "loss": 5.4994,
      "mean_token_accuracy": 0.1918413907289505,
      "num_tokens": 6667302.0,
      "step": 3670
    },
    {
      "entropy": 6.042338705062866,
      "epoch": 3.1572840567253975,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043332130451874645,
      "loss": 5.5563,
      "mean_token_accuracy": 0.1914373368024826,
      "num_tokens": 6677393.0,
      "step": 3675
    },
    {
      "entropy": 5.98822922706604,
      "epoch": 3.161581435324452,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00043308488728647127,
      "loss": 5.54,
      "mean_token_accuracy": 0.18344997316598893,
      "num_tokens": 6686727.0,
      "step": 3680
    },
    {
      "entropy": 5.9039342403411865,
      "epoch": 3.1658788139235066,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004328481248599882,
      "loss": 5.4334,
      "mean_token_accuracy": 0.19382288604974746,
      "num_tokens": 6696116.0,
      "step": 3685
    },
    {
      "entropy": 6.034891176223755,
      "epoch": 3.170176192522561,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004326110177562918,
      "loss": 5.5577,
      "mean_token_accuracy": 0.18678940385580062,
      "num_tokens": 6704640.0,
      "step": 3690
    },
    {
      "entropy": 5.893950605392456,
      "epoch": 3.1744735711216157,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043237356649312926,
      "loss": 5.4122,
      "mean_token_accuracy": 0.19891896545886995,
      "num_tokens": 6713663.0,
      "step": 3695
    },
    {
      "entropy": 5.947980260848999,
      "epoch": 3.17877094972067,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004321357715889991,
      "loss": 5.5514,
      "mean_token_accuracy": 0.1866046443581581,
      "num_tokens": 6722965.0,
      "step": 3700
    },
    {
      "entropy": 5.995095252990723,
      "epoch": 3.1830683283197247,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004318976335631505,
      "loss": 5.4921,
      "mean_token_accuracy": 0.19499906301498413,
      "num_tokens": 6732776.0,
      "step": 3705
    },
    {
      "entropy": 5.996333026885987,
      "epoch": 3.1873657069187797,
      "grad_norm": 1.0,
      "learning_rate": 0.00043165915293558155,
      "loss": 5.4857,
      "mean_token_accuracy": 0.1897098481655121,
      "num_tokens": 6741309.0,
      "step": 3710
    },
    {
      "entropy": 5.911564064025879,
      "epoch": 3.1916630855178343,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004314203302270388,
      "loss": 5.552,
      "mean_token_accuracy": 0.18683515042066573,
      "num_tokens": 6750584.0,
      "step": 3715
    },
    {
      "entropy": 6.021612453460693,
      "epoch": 3.195960464116889,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004311811659590154,
      "loss": 5.5092,
      "mean_token_accuracy": 0.191811640560627,
      "num_tokens": 6759344.0,
      "step": 3720
    },
    {
      "entropy": 6.040820455551147,
      "epoch": 3.2002578427159434,
      "grad_norm": 0.875,
      "learning_rate": 0.0004309416606537507,
      "loss": 5.6735,
      "mean_token_accuracy": 0.18092238456010817,
      "num_tokens": 6770345.0,
      "step": 3725
    },
    {
      "entropy": 5.995995426177979,
      "epoch": 3.204555221314998,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043070181483422843,
      "loss": 5.5563,
      "mean_token_accuracy": 0.18278944492340088,
      "num_tokens": 6779991.0,
      "step": 3730
    },
    {
      "entropy": 5.931873416900634,
      "epoch": 3.2088525999140525,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000430461629024176,
      "loss": 5.4941,
      "mean_token_accuracy": 0.1947342798113823,
      "num_tokens": 6788972.0,
      "step": 3735
    },
    {
      "entropy": 5.919456720352173,
      "epoch": 3.213149978513107,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004302211037480634,
      "loss": 5.432,
      "mean_token_accuracy": 0.19142798632383345,
      "num_tokens": 6796967.0,
      "step": 3740
    },
    {
      "entropy": 5.901779747009277,
      "epoch": 3.2174473571121616,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004299802395311015,
      "loss": 5.5273,
      "mean_token_accuracy": 0.19364550411701204,
      "num_tokens": 6805961.0,
      "step": 3745
    },
    {
      "entropy": 5.916573190689087,
      "epoch": 3.221744735711216,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004297390368992414,
      "loss": 5.4426,
      "mean_token_accuracy": 0.1955329716205597,
      "num_tokens": 6814657.0,
      "step": 3750
    },
    {
      "entropy": 5.99988751411438,
      "epoch": 3.2260421143102707,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042949749637917353,
      "loss": 5.487,
      "mean_token_accuracy": 0.1926707699894905,
      "num_tokens": 6823095.0,
      "step": 3755
    },
    {
      "entropy": 5.944211053848266,
      "epoch": 3.230339492909325,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004292556184983256,
      "loss": 5.5001,
      "mean_token_accuracy": 0.18788901716470718,
      "num_tokens": 6832195.0,
      "step": 3760
    },
    {
      "entropy": 6.045601272583008,
      "epoch": 3.2346368715083798,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004290134037848623,
      "loss": 5.6363,
      "mean_token_accuracy": 0.1813679426908493,
      "num_tokens": 6840922.0,
      "step": 3765
    },
    {
      "entropy": 5.955678081512451,
      "epoch": 3.2389342501074343,
      "grad_norm": 1.125,
      "learning_rate": 0.00042877085276768386,
      "loss": 5.4867,
      "mean_token_accuracy": 0.19231866747140886,
      "num_tokens": 6849182.0,
      "step": 3770
    },
    {
      "entropy": 5.957758140563965,
      "epoch": 3.243231628706489,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042852796597642455,
      "loss": 5.4781,
      "mean_token_accuracy": 0.1923646554350853,
      "num_tokens": 6857932.0,
      "step": 3775
    },
    {
      "entropy": 6.057449293136597,
      "epoch": 3.247529007305544,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004282847439414522,
      "loss": 5.6176,
      "mean_token_accuracy": 0.17261463701725005,
      "num_tokens": 6867283.0,
      "step": 3780
    },
    {
      "entropy": 6.008208608627319,
      "epoch": 3.2518263859045984,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004280411871938664,
      "loss": 5.593,
      "mean_token_accuracy": 0.1863952949643135,
      "num_tokens": 6876123.0,
      "step": 3785
    },
    {
      "entropy": 6.041609096527099,
      "epoch": 3.256123764503653,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004277972962654979,
      "loss": 5.5183,
      "mean_token_accuracy": 0.18469562530517578,
      "num_tokens": 6885239.0,
      "step": 3790
    },
    {
      "entropy": 5.943878412246704,
      "epoch": 3.2604211431027075,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004275530716889069,
      "loss": 5.5652,
      "mean_token_accuracy": 0.18427515923976898,
      "num_tokens": 6895061.0,
      "step": 3795
    },
    {
      "entropy": 6.032454538345337,
      "epoch": 3.264718521701762,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004273085139973822,
      "loss": 5.6198,
      "mean_token_accuracy": 0.17869567573070527,
      "num_tokens": 6903828.0,
      "step": 3800
    },
    {
      "entropy": 6.017523956298828,
      "epoch": 3.2690159003008166,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004270636237249401,
      "loss": 5.5322,
      "mean_token_accuracy": 0.18812069296836853,
      "num_tokens": 6912805.0,
      "step": 3805
    },
    {
      "entropy": 5.979595184326172,
      "epoch": 3.273313278899871,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042681840140632314,
      "loss": 5.5722,
      "mean_token_accuracy": 0.18031454235315322,
      "num_tokens": 6922165.0,
      "step": 3810
    },
    {
      "entropy": 6.027545738220215,
      "epoch": 3.2776106574989257,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004265728475769989,
      "loss": 5.5313,
      "mean_token_accuracy": 0.18603000342845916,
      "num_tokens": 6931677.0,
      "step": 3815
    },
    {
      "entropy": 5.958409976959229,
      "epoch": 3.28190803609798,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004263269627731586,
      "loss": 5.5149,
      "mean_token_accuracy": 0.1968382939696312,
      "num_tokens": 6940486.0,
      "step": 3820
    },
    {
      "entropy": 5.907626962661743,
      "epoch": 3.2862054146970348,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004260807475317164,
      "loss": 5.533,
      "mean_token_accuracy": 0.1850649744272232,
      "num_tokens": 6948990.0,
      "step": 3825
    },
    {
      "entropy": 6.06142144203186,
      "epoch": 3.2905027932960893,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004258342023903081,
      "loss": 5.6501,
      "mean_token_accuracy": 0.17591869384050368,
      "num_tokens": 6959311.0,
      "step": 3830
    },
    {
      "entropy": 5.99991660118103,
      "epoch": 3.294800171895144,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042558732788728975,
      "loss": 5.4251,
      "mean_token_accuracy": 0.19917479604482652,
      "num_tokens": 6968619.0,
      "step": 3835
    },
    {
      "entropy": 5.907351541519165,
      "epoch": 3.2990975504941984,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042534012456173643,
      "loss": 5.4955,
      "mean_token_accuracy": 0.19041050225496292,
      "num_tokens": 6977469.0,
      "step": 3840
    },
    {
      "entropy": 5.923426151275635,
      "epoch": 3.303394929093253,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042509259295344157,
      "loss": 5.4776,
      "mean_token_accuracy": 0.18866734802722931,
      "num_tokens": 6986772.0,
      "step": 3845
    },
    {
      "entropy": 5.954880619049073,
      "epoch": 3.3076923076923075,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042484473360291514,
      "loss": 5.4812,
      "mean_token_accuracy": 0.18148696422576904,
      "num_tokens": 6993937.0,
      "step": 3850
    },
    {
      "entropy": 5.907461214065552,
      "epoch": 3.311989686291362,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042459654705138294,
      "loss": 5.5487,
      "mean_token_accuracy": 0.1898924693465233,
      "num_tokens": 7003222.0,
      "step": 3855
    },
    {
      "entropy": 5.9601836681365965,
      "epoch": 3.316287064890417,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004243480338407853,
      "loss": 5.5086,
      "mean_token_accuracy": 0.1929486498236656,
      "num_tokens": 7012055.0,
      "step": 3860
    },
    {
      "entropy": 5.963890552520752,
      "epoch": 3.3205844434894716,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004240991945137755,
      "loss": 5.5054,
      "mean_token_accuracy": 0.18731537461280823,
      "num_tokens": 7021036.0,
      "step": 3865
    },
    {
      "entropy": 5.950737476348877,
      "epoch": 3.324881822088526,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042385002961371944,
      "loss": 5.4952,
      "mean_token_accuracy": 0.19253836423158646,
      "num_tokens": 7030450.0,
      "step": 3870
    },
    {
      "entropy": 6.046236276626587,
      "epoch": 3.3291792006875807,
      "grad_norm": 1.125,
      "learning_rate": 0.0004236005396846935,
      "loss": 5.6098,
      "mean_token_accuracy": 0.18545194268226622,
      "num_tokens": 7039740.0,
      "step": 3875
    },
    {
      "entropy": 6.006177759170532,
      "epoch": 3.333476579286635,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042335072527148406,
      "loss": 5.5791,
      "mean_token_accuracy": 0.1877056986093521,
      "num_tokens": 7050430.0,
      "step": 3880
    },
    {
      "entropy": 5.879164409637451,
      "epoch": 3.3377739578856898,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004231005869195859,
      "loss": 5.5594,
      "mean_token_accuracy": 0.18650871366262436,
      "num_tokens": 7059477.0,
      "step": 3885
    },
    {
      "entropy": 5.980384731292725,
      "epoch": 3.3420713364847443,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004228501251752011,
      "loss": 5.5078,
      "mean_token_accuracy": 0.1914823979139328,
      "num_tokens": 7067805.0,
      "step": 3890
    },
    {
      "entropy": 5.989449071884155,
      "epoch": 3.346368715083799,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042259934058523814,
      "loss": 5.5087,
      "mean_token_accuracy": 0.18367088735103607,
      "num_tokens": 7077606.0,
      "step": 3895
    },
    {
      "entropy": 5.979048013687134,
      "epoch": 3.3506660936828534,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042234823369731027,
      "loss": 5.4773,
      "mean_token_accuracy": 0.1893484875559807,
      "num_tokens": 7085647.0,
      "step": 3900
    },
    {
      "entropy": 5.909920740127563,
      "epoch": 3.354963472281908,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042209680505973465,
      "loss": 5.5042,
      "mean_token_accuracy": 0.1931456595659256,
      "num_tokens": 7095298.0,
      "step": 3905
    },
    {
      "entropy": 5.919304323196411,
      "epoch": 3.3592608508809625,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004218450552215308,
      "loss": 5.574,
      "mean_token_accuracy": 0.1899503216147423,
      "num_tokens": 7105207.0,
      "step": 3910
    },
    {
      "entropy": 5.992146730422974,
      "epoch": 3.363558229480017,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004215929847324199,
      "loss": 5.6349,
      "mean_token_accuracy": 0.17933368384838105,
      "num_tokens": 7114833.0,
      "step": 3915
    },
    {
      "entropy": 6.022987604141235,
      "epoch": 3.3678556080790716,
      "grad_norm": 1.078125,
      "learning_rate": 0.000421340594142823,
      "loss": 5.4385,
      "mean_token_accuracy": 0.19864116609096527,
      "num_tokens": 7123608.0,
      "step": 3920
    },
    {
      "entropy": 5.8856823444366455,
      "epoch": 3.3721529866781266,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042108788400386035,
      "loss": 5.5106,
      "mean_token_accuracy": 0.18838870525360107,
      "num_tokens": 7132250.0,
      "step": 3925
    },
    {
      "entropy": 5.962529850006104,
      "epoch": 3.376450365277181,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004208348548673498,
      "loss": 5.594,
      "mean_token_accuracy": 0.19215150624513627,
      "num_tokens": 7142086.0,
      "step": 3930
    },
    {
      "entropy": 6.050545597076416,
      "epoch": 3.3807477438762357,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000420581507285806,
      "loss": 5.5323,
      "mean_token_accuracy": 0.17952318042516707,
      "num_tokens": 7152434.0,
      "step": 3935
    },
    {
      "entropy": 5.88720850944519,
      "epoch": 3.38504512247529,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004203278418124386,
      "loss": 5.4662,
      "mean_token_accuracy": 0.20225470662117004,
      "num_tokens": 7163041.0,
      "step": 3940
    },
    {
      "entropy": 5.8799656391143795,
      "epoch": 3.3893425010743448,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004200738590011518,
      "loss": 5.4692,
      "mean_token_accuracy": 0.1973979264497757,
      "num_tokens": 7171875.0,
      "step": 3945
    },
    {
      "entropy": 5.9314147472381595,
      "epoch": 3.3936398796733993,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041981955940654245,
      "loss": 5.5841,
      "mean_token_accuracy": 0.1857954144477844,
      "num_tokens": 7180803.0,
      "step": 3950
    },
    {
      "entropy": 5.935078239440918,
      "epoch": 3.397937258272454,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004195649435838992,
      "loss": 5.5974,
      "mean_token_accuracy": 0.175727042555809,
      "num_tokens": 7190661.0,
      "step": 3955
    },
    {
      "entropy": 5.895944023132325,
      "epoch": 3.4022346368715084,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004193100120892013,
      "loss": 5.44,
      "mean_token_accuracy": 0.19905115962028502,
      "num_tokens": 7199357.0,
      "step": 3960
    },
    {
      "entropy": 5.981336355209351,
      "epoch": 3.406532015470563,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004190547654791172,
      "loss": 5.5999,
      "mean_token_accuracy": 0.18368346244096756,
      "num_tokens": 7209856.0,
      "step": 3965
    },
    {
      "entropy": 5.998037719726563,
      "epoch": 3.4108293940696175,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041879920431100347,
      "loss": 5.5623,
      "mean_token_accuracy": 0.18161152303218842,
      "num_tokens": 7218778.0,
      "step": 3970
    },
    {
      "entropy": 5.957395601272583,
      "epoch": 3.415126772668672,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004185433291429036,
      "loss": 5.5985,
      "mean_token_accuracy": 0.1867300182580948,
      "num_tokens": 7228442.0,
      "step": 3975
    },
    {
      "entropy": 5.984077501296997,
      "epoch": 3.4194241512677266,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041828714053354665,
      "loss": 5.5695,
      "mean_token_accuracy": 0.18295224159955978,
      "num_tokens": 7238724.0,
      "step": 3980
    },
    {
      "entropy": 5.846696996688843,
      "epoch": 3.423721529866781,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004180306390423462,
      "loss": 5.5286,
      "mean_token_accuracy": 0.18906760662794114,
      "num_tokens": 7247844.0,
      "step": 3985
    },
    {
      "entropy": 5.9538819789886475,
      "epoch": 3.4280189084658357,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041777382522939884,
      "loss": 5.5898,
      "mean_token_accuracy": 0.18908908665180207,
      "num_tokens": 7257260.0,
      "step": 3990
    },
    {
      "entropy": 5.988106775283813,
      "epoch": 3.4323162870648902,
      "grad_norm": 0.953125,
      "learning_rate": 0.00041751669965548344,
      "loss": 5.6046,
      "mean_token_accuracy": 0.1803540512919426,
      "num_tokens": 7266890.0,
      "step": 3995
    },
    {
      "entropy": 5.9721503257751465,
      "epoch": 3.4366136656639448,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041725926288205945,
      "loss": 5.6267,
      "mean_token_accuracy": 0.17775601744651795,
      "num_tokens": 7276114.0,
      "step": 4000
    },
    {
      "epoch": 3.4366136656639448,
      "eval_entropy": 5.739501470918054,
      "eval_loss": 6.0194196701049805,
      "eval_mean_token_accuracy": 0.16980059876165413,
      "eval_num_tokens": 7276114.0,
      "eval_runtime": 2.0562,
      "eval_samples_per_second": 1725.976,
      "eval_steps_per_second": 215.929,
      "step": 4000
    },
    {
      "entropy": 5.989626407623291,
      "epoch": 3.4409110442629998,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004170015154712658,
      "loss": 5.5685,
      "mean_token_accuracy": 0.18304531574249266,
      "num_tokens": 7284426.0,
      "step": 4005
    },
    {
      "entropy": 5.950127267837525,
      "epoch": 3.4452084228620543,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00041674345798591993,
      "loss": 5.5915,
      "mean_token_accuracy": 0.18336501866579055,
      "num_tokens": 7294813.0,
      "step": 4010
    },
    {
      "entropy": 6.0112659454345705,
      "epoch": 3.449505801461109,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004164850909895161,
      "loss": 5.5799,
      "mean_token_accuracy": 0.1913402944803238,
      "num_tokens": 7304655.0,
      "step": 4015
    },
    {
      "entropy": 5.903563404083252,
      "epoch": 3.4538031800601634,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004162264150462247,
      "loss": 5.5439,
      "mean_token_accuracy": 0.18234843164682388,
      "num_tokens": 7313610.0,
      "step": 4020
    },
    {
      "entropy": 6.008734226226807,
      "epoch": 3.458100558659218,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041596743072089065,
      "loss": 5.5741,
      "mean_token_accuracy": 0.18414125293493272,
      "num_tokens": 7322243.0,
      "step": 4025
    },
    {
      "entropy": 6.049357938766479,
      "epoch": 3.4623979372582725,
      "grad_norm": 1.140625,
      "learning_rate": 0.000415708138579032,
      "loss": 5.5484,
      "mean_token_accuracy": 0.17788545042276382,
      "num_tokens": 7331040.0,
      "step": 4030
    },
    {
      "entropy": 5.88419828414917,
      "epoch": 3.466695315857327,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041544853918683923,
      "loss": 5.603,
      "mean_token_accuracy": 0.17848607450723647,
      "num_tokens": 7340771.0,
      "step": 4035
    },
    {
      "entropy": 5.972914791107177,
      "epoch": 3.4709926944563816,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004151886331111737,
      "loss": 5.6585,
      "mean_token_accuracy": 0.17960806488990783,
      "num_tokens": 7349960.0,
      "step": 4040
    },
    {
      "entropy": 5.9350839138031,
      "epoch": 3.475290073055436,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041492842091956646,
      "loss": 5.4909,
      "mean_token_accuracy": 0.19007912576198577,
      "num_tokens": 7357983.0,
      "step": 4045
    },
    {
      "entropy": 6.046515941619873,
      "epoch": 3.4795874516544907,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004146679031802167,
      "loss": 5.6127,
      "mean_token_accuracy": 0.18493092954158782,
      "num_tokens": 7366814.0,
      "step": 4050
    },
    {
      "entropy": 5.903052806854248,
      "epoch": 3.4838848302535452,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041440708046199123,
      "loss": 5.4767,
      "mean_token_accuracy": 0.19377027302980424,
      "num_tokens": 7374773.0,
      "step": 4055
    },
    {
      "entropy": 5.926623153686523,
      "epoch": 3.4881822088525998,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004141459533344226,
      "loss": 5.5713,
      "mean_token_accuracy": 0.18479437232017518,
      "num_tokens": 7383937.0,
      "step": 4060
    },
    {
      "entropy": 5.973228168487549,
      "epoch": 3.4924795874516543,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041388452236770795,
      "loss": 5.5426,
      "mean_token_accuracy": 0.18190130293369294,
      "num_tokens": 7392577.0,
      "step": 4065
    },
    {
      "entropy": 5.880434799194336,
      "epoch": 3.4967769660507093,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041362278813270823,
      "loss": 5.403,
      "mean_token_accuracy": 0.20291406512260438,
      "num_tokens": 7401473.0,
      "step": 4070
    },
    {
      "entropy": 5.973664808273315,
      "epoch": 3.501074344649764,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00041336075120094616,
      "loss": 5.6507,
      "mean_token_accuracy": 0.17217092961072922,
      "num_tokens": 7410831.0,
      "step": 4075
    },
    {
      "entropy": 6.015513610839844,
      "epoch": 3.5053717232488184,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041309841214460586,
      "loss": 5.6251,
      "mean_token_accuracy": 0.18172844350337983,
      "num_tokens": 7421563.0,
      "step": 4080
    },
    {
      "entropy": 5.9306933879852295,
      "epoch": 3.509669101847873,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004128357715365309,
      "loss": 5.5403,
      "mean_token_accuracy": 0.18941463977098466,
      "num_tokens": 7430174.0,
      "step": 4085
    },
    {
      "entropy": 5.9411945819854735,
      "epoch": 3.5139664804469275,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041257282995022345,
      "loss": 5.5287,
      "mean_token_accuracy": 0.19064364582300186,
      "num_tokens": 7439034.0,
      "step": 4090
    },
    {
      "entropy": 5.91850061416626,
      "epoch": 3.518263859045982,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004123095879598426,
      "loss": 5.5376,
      "mean_token_accuracy": 0.1826656237244606,
      "num_tokens": 7447663.0,
      "step": 4095
    },
    {
      "entropy": 5.951149988174438,
      "epoch": 3.5225612376450366,
      "grad_norm": 1.015625,
      "learning_rate": 0.00041204604614020397,
      "loss": 5.6242,
      "mean_token_accuracy": 0.1754978582262993,
      "num_tokens": 7456615.0,
      "step": 4100
    },
    {
      "entropy": 6.022586059570313,
      "epoch": 3.526858616244091,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004117822050667773,
      "loss": 5.6609,
      "mean_token_accuracy": 0.1797046795487404,
      "num_tokens": 7466203.0,
      "step": 4105
    },
    {
      "entropy": 5.999763250350952,
      "epoch": 3.5311559948431457,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041151806531568617,
      "loss": 5.594,
      "mean_token_accuracy": 0.1838526800274849,
      "num_tokens": 7475411.0,
      "step": 4110
    },
    {
      "entropy": 5.927717494964599,
      "epoch": 3.5354533734422002,
      "grad_norm": 0.984375,
      "learning_rate": 0.00041125362746370625,
      "loss": 5.5954,
      "mean_token_accuracy": 0.18438049405813217,
      "num_tokens": 7484965.0,
      "step": 4115
    },
    {
      "entropy": 6.04001750946045,
      "epoch": 3.5397507520412548,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004109888920882639,
      "loss": 5.5471,
      "mean_token_accuracy": 0.1918780490756035,
      "num_tokens": 7494240.0,
      "step": 4120
    },
    {
      "entropy": 5.938482570648193,
      "epoch": 3.5440481306403093,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004107238597674356,
      "loss": 5.5585,
      "mean_token_accuracy": 0.1836171880364418,
      "num_tokens": 7503560.0,
      "step": 4125
    },
    {
      "entropy": 5.898692846298218,
      "epoch": 3.548345509239364,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000410458531079946,
      "loss": 5.5004,
      "mean_token_accuracy": 0.19434000551700592,
      "num_tokens": 7512650.0,
      "step": 4130
    },
    {
      "entropy": 5.958231353759766,
      "epoch": 3.5526428878384184,
      "grad_norm": 1.125,
      "learning_rate": 0.0004101929066051668,
      "loss": 5.6095,
      "mean_token_accuracy": 0.1837508738040924,
      "num_tokens": 7521864.0,
      "step": 4135
    },
    {
      "entropy": 5.962691593170166,
      "epoch": 3.556940266437473,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004099269869231157,
      "loss": 5.511,
      "mean_token_accuracy": 0.1855240523815155,
      "num_tokens": 7531013.0,
      "step": 4140
    },
    {
      "entropy": 5.934520578384399,
      "epoch": 3.5612376450365275,
      "grad_norm": 1.0,
      "learning_rate": 0.00040966077261445495,
      "loss": 5.5316,
      "mean_token_accuracy": 0.1795806884765625,
      "num_tokens": 7539959.0,
      "step": 4145
    },
    {
      "entropy": 5.998489856719971,
      "epoch": 3.565535023635582,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004093942642604904,
      "loss": 5.4965,
      "mean_token_accuracy": 0.18766231685876847,
      "num_tokens": 7548354.0,
      "step": 4150
    },
    {
      "entropy": 5.9573204040527346,
      "epoch": 3.5698324022346366,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040912746244316944,
      "loss": 5.634,
      "mean_token_accuracy": 0.18089795708656312,
      "num_tokens": 7558321.0,
      "step": 4155
    },
    {
      "entropy": 5.945523595809936,
      "epoch": 3.5741297808336916,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00040886036774508095,
      "loss": 5.5152,
      "mean_token_accuracy": 0.1863349363207817,
      "num_tokens": 7567889.0,
      "step": 4160
    },
    {
      "entropy": 5.989820671081543,
      "epoch": 3.578427159432746,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004085929807494527,
      "loss": 5.5613,
      "mean_token_accuracy": 0.1877506971359253,
      "num_tokens": 7576752.0,
      "step": 4165
    },
    {
      "entropy": 5.926433086395264,
      "epoch": 3.5827245380318007,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004083253020401512,
      "loss": 5.4728,
      "mean_token_accuracy": 0.1965028926730156,
      "num_tokens": 7585413.0,
      "step": 4170
    },
    {
      "entropy": 5.923580551147461,
      "epoch": 3.5870219166308552,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004080573322016797,
      "loss": 5.4256,
      "mean_token_accuracy": 0.19330667108297347,
      "num_tokens": 7593966.0,
      "step": 4175
    },
    {
      "entropy": 5.883316516876221,
      "epoch": 3.59131929522991,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004077890718191773,
      "loss": 5.4469,
      "mean_token_accuracy": 0.1911528155207634,
      "num_tokens": 7602746.0,
      "step": 4180
    },
    {
      "entropy": 5.9072150707244875,
      "epoch": 3.5956166738289643,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040752052147841733,
      "loss": 5.5042,
      "mean_token_accuracy": 0.19261341094970702,
      "num_tokens": 7611245.0,
      "step": 4185
    },
    {
      "entropy": 5.955031156539917,
      "epoch": 3.599914052428019,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004072516817658065,
      "loss": 5.5253,
      "mean_token_accuracy": 0.18929073065519333,
      "num_tokens": 7620234.0,
      "step": 4190
    },
    {
      "entropy": 5.982736396789551,
      "epoch": 3.6042114310270734,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004069825532683831,
      "loss": 5.5501,
      "mean_token_accuracy": 0.18622905462980271,
      "num_tokens": 7629794.0,
      "step": 4195
    },
    {
      "entropy": 5.894542169570923,
      "epoch": 3.608508809626128,
      "grad_norm": 0.984375,
      "learning_rate": 0.00040671313657381645,
      "loss": 5.4871,
      "mean_token_accuracy": 0.1976424276828766,
      "num_tokens": 7639497.0,
      "step": 4200
    },
    {
      "entropy": 5.854848337173462,
      "epoch": 3.6128061882251825,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040644343227040473,
      "loss": 5.4499,
      "mean_token_accuracy": 0.19116452783346177,
      "num_tokens": 7647647.0,
      "step": 4205
    },
    {
      "entropy": 5.921643161773682,
      "epoch": 3.617103566824237,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004061734409470745,
      "loss": 5.6281,
      "mean_token_accuracy": 0.1854175865650177,
      "num_tokens": 7657988.0,
      "step": 4210
    },
    {
      "entropy": 5.979614973068237,
      "epoch": 3.621400945423292,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004059031631933788,
      "loss": 5.5326,
      "mean_token_accuracy": 0.1888209328055382,
      "num_tokens": 7667498.0,
      "step": 4215
    },
    {
      "entropy": 5.956178855895996,
      "epoch": 3.6256983240223466,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040563259959949615,
      "loss": 5.6886,
      "mean_token_accuracy": 0.17732578068971633,
      "num_tokens": 7677386.0,
      "step": 4220
    },
    {
      "entropy": 6.071782970428467,
      "epoch": 3.629995702621401,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004053617507562295,
      "loss": 5.5146,
      "mean_token_accuracy": 0.1861334875226021,
      "num_tokens": 7686643.0,
      "step": 4225
    },
    {
      "entropy": 5.960998058319092,
      "epoch": 3.6342930812204557,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040509061725500426,
      "loss": 5.544,
      "mean_token_accuracy": 0.18653534203767777,
      "num_tokens": 7695089.0,
      "step": 4230
    },
    {
      "entropy": 5.873150777816773,
      "epoch": 3.6385904598195102,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004048191996878677,
      "loss": 5.5444,
      "mean_token_accuracy": 0.1873533010482788,
      "num_tokens": 7703854.0,
      "step": 4235
    },
    {
      "entropy": 5.888010787963867,
      "epoch": 3.642887838418565,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040454749864748734,
      "loss": 5.4855,
      "mean_token_accuracy": 0.19170088917016984,
      "num_tokens": 7712903.0,
      "step": 4240
    },
    {
      "entropy": 5.952953052520752,
      "epoch": 3.6471852170176193,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004042755147271496,
      "loss": 5.424,
      "mean_token_accuracy": 0.1897352635860443,
      "num_tokens": 7721701.0,
      "step": 4245
    },
    {
      "entropy": 5.813195466995239,
      "epoch": 3.651482595616674,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004040032485207587,
      "loss": 5.5401,
      "mean_token_accuracy": 0.18509043604135514,
      "num_tokens": 7731318.0,
      "step": 4250
    },
    {
      "entropy": 5.967707347869873,
      "epoch": 3.6557799742157284,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004037307006228352,
      "loss": 5.4716,
      "mean_token_accuracy": 0.19402939975261688,
      "num_tokens": 7740413.0,
      "step": 4255
    },
    {
      "entropy": 5.913791751861572,
      "epoch": 3.660077352814783,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004034578716285147,
      "loss": 5.4536,
      "mean_token_accuracy": 0.19273238033056259,
      "num_tokens": 7749054.0,
      "step": 4260
    },
    {
      "entropy": 5.868339681625367,
      "epoch": 3.6643747314138375,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004031847621335467,
      "loss": 5.4793,
      "mean_token_accuracy": 0.1922049939632416,
      "num_tokens": 7757366.0,
      "step": 4265
    },
    {
      "entropy": 5.939763927459717,
      "epoch": 3.668672110012892,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004029113727342933,
      "loss": 5.5074,
      "mean_token_accuracy": 0.19062846302986144,
      "num_tokens": 7766471.0,
      "step": 4270
    },
    {
      "entropy": 5.878051853179931,
      "epoch": 3.6729694886119466,
      "grad_norm": 1.0,
      "learning_rate": 0.00040263770402772746,
      "loss": 5.4912,
      "mean_token_accuracy": 0.1884067326784134,
      "num_tokens": 7775920.0,
      "step": 4275
    },
    {
      "entropy": 5.922876453399658,
      "epoch": 3.677266867211001,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004023637566114325,
      "loss": 5.5455,
      "mean_token_accuracy": 0.18928359299898148,
      "num_tokens": 7784530.0,
      "step": 4280
    },
    {
      "entropy": 5.91483097076416,
      "epoch": 3.6815642458100557,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004020895310835999,
      "loss": 5.4927,
      "mean_token_accuracy": 0.18310107737779618,
      "num_tokens": 7793656.0,
      "step": 4285
    },
    {
      "entropy": 5.965200185775757,
      "epoch": 3.6858616244091102,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00040181502804302865,
      "loss": 5.5224,
      "mean_token_accuracy": 0.1860280752182007,
      "num_tokens": 7802185.0,
      "step": 4290
    },
    {
      "entropy": 5.891399002075195,
      "epoch": 3.690159003008165,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040154024808912377,
      "loss": 5.5087,
      "mean_token_accuracy": 0.1911735564470291,
      "num_tokens": 7810345.0,
      "step": 4295
    },
    {
      "entropy": 5.912160873413086,
      "epoch": 3.6944563816072193,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004012651918218947,
      "loss": 5.5471,
      "mean_token_accuracy": 0.17879761159420013,
      "num_tokens": 7818998.0,
      "step": 4300
    },
    {
      "entropy": 5.985769987106323,
      "epoch": 3.6987537602062743,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004009898598419544,
      "loss": 5.6809,
      "mean_token_accuracy": 0.1700455442070961,
      "num_tokens": 7828638.0,
      "step": 4305
    },
    {
      "entropy": 5.999387121200561,
      "epoch": 3.703051138805329,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000400714252750518,
      "loss": 5.6425,
      "mean_token_accuracy": 0.1807677462697029,
      "num_tokens": 7838812.0,
      "step": 4310
    },
    {
      "entropy": 5.981132459640503,
      "epoch": 3.7073485174043834,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004004383711494011,
      "loss": 5.5426,
      "mean_token_accuracy": 0.19087505489587783,
      "num_tokens": 7847458.0,
      "step": 4315
    },
    {
      "entropy": 5.959196376800537,
      "epoch": 3.711645896003438,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004001622156410189,
      "loss": 5.5691,
      "mean_token_accuracy": 0.1856666773557663,
      "num_tokens": 7856553.0,
      "step": 4320
    },
    {
      "entropy": 5.884228134155274,
      "epoch": 3.7159432746024925,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039988578682838467,
      "loss": 5.526,
      "mean_token_accuracy": 0.18373223543167114,
      "num_tokens": 7864788.0,
      "step": 4325
    },
    {
      "entropy": 5.928623342514038,
      "epoch": 3.720240653201547,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039960908531510843,
      "loss": 5.5012,
      "mean_token_accuracy": 0.1894019216299057,
      "num_tokens": 7873850.0,
      "step": 4330
    },
    {
      "entropy": 5.9985411167144775,
      "epoch": 3.7245380318006016,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003993321117053956,
      "loss": 5.6197,
      "mean_token_accuracy": 0.17805870920419692,
      "num_tokens": 7882775.0,
      "step": 4335
    },
    {
      "entropy": 6.02107458114624,
      "epoch": 3.728835410399656,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039905486660404604,
      "loss": 5.5654,
      "mean_token_accuracy": 0.1820268526673317,
      "num_tokens": 7890570.0,
      "step": 4340
    },
    {
      "entropy": 5.889177703857422,
      "epoch": 3.7331327889987107,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00039877735061645206,
      "loss": 5.513,
      "mean_token_accuracy": 0.196246138215065,
      "num_tokens": 7900090.0,
      "step": 4345
    },
    {
      "entropy": 5.958146762847901,
      "epoch": 3.7374301675977653,
      "grad_norm": 1.125,
      "learning_rate": 0.0003984995643485977,
      "loss": 5.5566,
      "mean_token_accuracy": 0.18338315188884735,
      "num_tokens": 7908077.0,
      "step": 4350
    },
    {
      "entropy": 6.022622728347779,
      "epoch": 3.74172754619682,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00039822150840705716,
      "loss": 5.5562,
      "mean_token_accuracy": 0.18958212733268737,
      "num_tokens": 7916290.0,
      "step": 4355
    },
    {
      "entropy": 5.994214487075806,
      "epoch": 3.746024924795875,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039794318339899347,
      "loss": 5.6467,
      "mean_token_accuracy": 0.1848492443561554,
      "num_tokens": 7925835.0,
      "step": 4360
    },
    {
      "entropy": 5.949737405776977,
      "epoch": 3.7503223033949293,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039766458993215726,
      "loss": 5.5959,
      "mean_token_accuracy": 0.1823095202445984,
      "num_tokens": 7935076.0,
      "step": 4365
    },
    {
      "entropy": 5.898332262039185,
      "epoch": 3.754619681993984,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039738572861488527,
      "loss": 5.4942,
      "mean_token_accuracy": 0.19010726660490035,
      "num_tokens": 7943958.0,
      "step": 4370
    },
    {
      "entropy": 5.946810245513916,
      "epoch": 3.7589170605930384,
      "grad_norm": 0.984375,
      "learning_rate": 0.000397106600056099,
      "loss": 5.5605,
      "mean_token_accuracy": 0.18583073914051057,
      "num_tokens": 7953189.0,
      "step": 4375
    },
    {
      "entropy": 5.899966716766357,
      "epoch": 3.763214439192093,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003968272048653039,
      "loss": 5.4529,
      "mean_token_accuracy": 0.19802996814250945,
      "num_tokens": 7962927.0,
      "step": 4380
    },
    {
      "entropy": 5.831211233139038,
      "epoch": 3.7675118177911475,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003965475436525873,
      "loss": 5.4713,
      "mean_token_accuracy": 0.19392775595188141,
      "num_tokens": 7973087.0,
      "step": 4385
    },
    {
      "entropy": 5.872487020492554,
      "epoch": 3.771809196390202,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003962676170286174,
      "loss": 5.4495,
      "mean_token_accuracy": 0.18794787377119065,
      "num_tokens": 7982535.0,
      "step": 4390
    },
    {
      "entropy": 5.9592584609985355,
      "epoch": 3.7761065749892566,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039598742560464223,
      "loss": 5.5228,
      "mean_token_accuracy": 0.18831065595149993,
      "num_tokens": 7990740.0,
      "step": 4395
    },
    {
      "entropy": 5.957420301437378,
      "epoch": 3.780403953588311,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003957069699924877,
      "loss": 5.5413,
      "mean_token_accuracy": 0.18056829422712325,
      "num_tokens": 7999349.0,
      "step": 4400
    },
    {
      "entropy": 5.942235231399536,
      "epoch": 3.7847013321873657,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000395426250804557,
      "loss": 5.5209,
      "mean_token_accuracy": 0.1986977517604828,
      "num_tokens": 8007615.0,
      "step": 4405
    },
    {
      "entropy": 5.904413080215454,
      "epoch": 3.7889987107864203,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00039514526865382847,
      "loss": 5.5195,
      "mean_token_accuracy": 0.19274225682020188,
      "num_tokens": 8017545.0,
      "step": 4410
    },
    {
      "entropy": 5.900614929199219,
      "epoch": 3.793296089385475,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003948640241538548,
      "loss": 5.4445,
      "mean_token_accuracy": 0.19299122989177703,
      "num_tokens": 8026381.0,
      "step": 4415
    },
    {
      "entropy": 5.911832284927368,
      "epoch": 3.7975934679845293,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003945825179187617,
      "loss": 5.5518,
      "mean_token_accuracy": 0.1861799567937851,
      "num_tokens": 8034745.0,
      "step": 4420
    },
    {
      "entropy": 5.990826416015625,
      "epoch": 3.801890846583584,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039430075056324604,
      "loss": 5.4964,
      "mean_token_accuracy": 0.19839415252208709,
      "num_tokens": 8043995.0,
      "step": 4425
    },
    {
      "entropy": 5.95226559638977,
      "epoch": 3.8061882251826384,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039401872270257546,
      "loss": 5.5897,
      "mean_token_accuracy": 0.18173896372318268,
      "num_tokens": 8053059.0,
      "step": 4430
    },
    {
      "entropy": 5.9306341171264645,
      "epoch": 3.810485603781693,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039373643495258567,
      "loss": 5.6137,
      "mean_token_accuracy": 0.18779614865779876,
      "num_tokens": 8062160.0,
      "step": 4435
    },
    {
      "entropy": 5.903824520111084,
      "epoch": 3.8147829823807475,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039345388792968056,
      "loss": 5.5284,
      "mean_token_accuracy": 0.19737044125795364,
      "num_tokens": 8071260.0,
      "step": 4440
    },
    {
      "entropy": 5.98727502822876,
      "epoch": 3.819080360979802,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039317108225082984,
      "loss": 5.6364,
      "mean_token_accuracy": 0.1810362756252289,
      "num_tokens": 8081540.0,
      "step": 4445
    },
    {
      "entropy": 5.89173469543457,
      "epoch": 3.8233777395788566,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039288801853356806,
      "loss": 5.5944,
      "mean_token_accuracy": 0.18381374925374985,
      "num_tokens": 8089785.0,
      "step": 4450
    },
    {
      "entropy": 5.9480987071990965,
      "epoch": 3.8276751181779116,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003926046973959932,
      "loss": 5.4312,
      "mean_token_accuracy": 0.19819699674844743,
      "num_tokens": 8098097.0,
      "step": 4455
    },
    {
      "entropy": 5.908696889877319,
      "epoch": 3.831972496776966,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003923211194567654,
      "loss": 5.6599,
      "mean_token_accuracy": 0.18412704020738602,
      "num_tokens": 8108693.0,
      "step": 4460
    },
    {
      "entropy": 5.9504638671875,
      "epoch": 3.8362698753760207,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039203728533510556,
      "loss": 5.5082,
      "mean_token_accuracy": 0.188208931684494,
      "num_tokens": 8117181.0,
      "step": 4465
    },
    {
      "entropy": 5.88679313659668,
      "epoch": 3.8405672539750753,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000391753195650794,
      "loss": 5.5426,
      "mean_token_accuracy": 0.18362973332405091,
      "num_tokens": 8125398.0,
      "step": 4470
    },
    {
      "entropy": 5.902078151702881,
      "epoch": 3.84486463257413,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039146885102416895,
      "loss": 5.539,
      "mean_token_accuracy": 0.18678168654441835,
      "num_tokens": 8135320.0,
      "step": 4475
    },
    {
      "entropy": 5.935784244537354,
      "epoch": 3.8491620111731844,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039118425207612553,
      "loss": 5.6157,
      "mean_token_accuracy": 0.1845711052417755,
      "num_tokens": 8144320.0,
      "step": 4480
    },
    {
      "entropy": 5.868662357330322,
      "epoch": 3.853459389772239,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039089939942811396,
      "loss": 5.482,
      "mean_token_accuracy": 0.19268393218517305,
      "num_tokens": 8153653.0,
      "step": 4485
    },
    {
      "entropy": 5.988348579406738,
      "epoch": 3.8577567683712934,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039061429370213863,
      "loss": 5.5312,
      "mean_token_accuracy": 0.18928302526474,
      "num_tokens": 8162741.0,
      "step": 4490
    },
    {
      "entropy": 5.818091154098511,
      "epoch": 3.862054146970348,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039032893552075646,
      "loss": 5.4338,
      "mean_token_accuracy": 0.1974863141775131,
      "num_tokens": 8171078.0,
      "step": 4495
    },
    {
      "entropy": 5.820233726501465,
      "epoch": 3.8663515255694025,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003900433255070758,
      "loss": 5.5124,
      "mean_token_accuracy": 0.19318781048059464,
      "num_tokens": 8179968.0,
      "step": 4500
    },
    {
      "epoch": 3.8663515255694025,
      "eval_entropy": 5.745309154192607,
      "eval_loss": 5.964412689208984,
      "eval_mean_token_accuracy": 0.17343190230160682,
      "eval_num_tokens": 8179968.0,
      "eval_runtime": 2.2641,
      "eval_samples_per_second": 1567.53,
      "eval_steps_per_second": 196.107,
      "step": 4500
    },
    {
      "entropy": 5.962471675872803,
      "epoch": 3.870648904168457,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038975746428475454,
      "loss": 5.4774,
      "mean_token_accuracy": 0.19070976674556733,
      "num_tokens": 8189261.0,
      "step": 4505
    },
    {
      "entropy": 5.985160446166992,
      "epoch": 3.874946282767512,
      "grad_norm": 1.125,
      "learning_rate": 0.00038947135247799955,
      "loss": 5.4989,
      "mean_token_accuracy": 0.1912393316626549,
      "num_tokens": 8198302.0,
      "step": 4510
    },
    {
      "entropy": 5.9148375511169435,
      "epoch": 3.8792436613665666,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038918499071156443,
      "loss": 5.4861,
      "mean_token_accuracy": 0.19183243215084075,
      "num_tokens": 8207098.0,
      "step": 4515
    },
    {
      "entropy": 5.884652423858642,
      "epoch": 3.883541039965621,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000388898379610749,
      "loss": 5.5336,
      "mean_token_accuracy": 0.18514965176582338,
      "num_tokens": 8216831.0,
      "step": 4520
    },
    {
      "entropy": 5.899249696731568,
      "epoch": 3.8878384185646757,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003886115198013973,
      "loss": 5.5341,
      "mean_token_accuracy": 0.19364315569400786,
      "num_tokens": 8225369.0,
      "step": 4525
    },
    {
      "entropy": 5.935558700561524,
      "epoch": 3.8921357971637303,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003883244119098965,
      "loss": 5.6566,
      "mean_token_accuracy": 0.17857892513275148,
      "num_tokens": 8234440.0,
      "step": 4530
    },
    {
      "entropy": 5.923916244506836,
      "epoch": 3.896433175762785,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003880370565631754,
      "loss": 5.4709,
      "mean_token_accuracy": 0.19344521462917327,
      "num_tokens": 8243707.0,
      "step": 4535
    },
    {
      "entropy": 5.9532630920410154,
      "epoch": 3.9007305543618394,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038774945438870337,
      "loss": 5.6209,
      "mean_token_accuracy": 0.1822944015264511,
      "num_tokens": 8254223.0,
      "step": 4540
    },
    {
      "entropy": 5.920027780532837,
      "epoch": 3.905027932960894,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038746160601448845,
      "loss": 5.473,
      "mean_token_accuracy": 0.1911543294787407,
      "num_tokens": 8263105.0,
      "step": 4545
    },
    {
      "entropy": 5.858027553558349,
      "epoch": 3.9093253115599484,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003871735120690766,
      "loss": 5.5276,
      "mean_token_accuracy": 0.19188787639141083,
      "num_tokens": 8271478.0,
      "step": 4550
    },
    {
      "entropy": 5.968813753128051,
      "epoch": 3.913622690159003,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003868851731815497,
      "loss": 5.5723,
      "mean_token_accuracy": 0.18731185644865037,
      "num_tokens": 8280396.0,
      "step": 4555
    },
    {
      "entropy": 5.961661005020142,
      "epoch": 3.9179200687580575,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003865965899815247,
      "loss": 5.5714,
      "mean_token_accuracy": 0.18448397666215896,
      "num_tokens": 8290371.0,
      "step": 4560
    },
    {
      "entropy": 5.880784559249878,
      "epoch": 3.922217447357112,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003863077630991518,
      "loss": 5.4645,
      "mean_token_accuracy": 0.1960570365190506,
      "num_tokens": 8298976.0,
      "step": 4565
    },
    {
      "entropy": 5.83483772277832,
      "epoch": 3.9265148259561666,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003860186931651139,
      "loss": 5.5223,
      "mean_token_accuracy": 0.18814983665943147,
      "num_tokens": 8308752.0,
      "step": 4570
    },
    {
      "entropy": 5.941116905212402,
      "epoch": 3.930812204555221,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003857293808106238,
      "loss": 5.5861,
      "mean_token_accuracy": 0.18480976521968842,
      "num_tokens": 8317343.0,
      "step": 4575
    },
    {
      "entropy": 5.940814352035522,
      "epoch": 3.9351095831542757,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003854398266674241,
      "loss": 5.4419,
      "mean_token_accuracy": 0.19702621549367905,
      "num_tokens": 8326956.0,
      "step": 4580
    },
    {
      "entropy": 5.90321455001831,
      "epoch": 3.9394069617533303,
      "grad_norm": 1.125,
      "learning_rate": 0.00038515003136778544,
      "loss": 5.553,
      "mean_token_accuracy": 0.18494745641946791,
      "num_tokens": 8335589.0,
      "step": 4585
    },
    {
      "entropy": 5.907242012023926,
      "epoch": 3.943704340352385,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038485999554450483,
      "loss": 5.5331,
      "mean_token_accuracy": 0.18324975073337554,
      "num_tokens": 8345517.0,
      "step": 4590
    },
    {
      "entropy": 5.821186304092407,
      "epoch": 3.9480017189514394,
      "grad_norm": 1.125,
      "learning_rate": 0.00038456971983090454,
      "loss": 5.453,
      "mean_token_accuracy": 0.19641250371932983,
      "num_tokens": 8354702.0,
      "step": 4595
    },
    {
      "entropy": 5.882057619094849,
      "epoch": 3.9522990975504944,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003842792048608309,
      "loss": 5.4862,
      "mean_token_accuracy": 0.1973177045583725,
      "num_tokens": 8362940.0,
      "step": 4600
    },
    {
      "entropy": 5.970648622512817,
      "epoch": 3.956596476149549,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003839884512686523,
      "loss": 5.5322,
      "mean_token_accuracy": 0.1901967242360115,
      "num_tokens": 8372034.0,
      "step": 4605
    },
    {
      "entropy": 5.885459852218628,
      "epoch": 3.9608938547486034,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038369745968925846,
      "loss": 5.5725,
      "mean_token_accuracy": 0.18121236711740493,
      "num_tokens": 8381673.0,
      "step": 4610
    },
    {
      "entropy": 5.966006803512573,
      "epoch": 3.965191233347658,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00038340623075805875,
      "loss": 5.512,
      "mean_token_accuracy": 0.18635776340961457,
      "num_tokens": 8390804.0,
      "step": 4615
    },
    {
      "entropy": 5.978103303909302,
      "epoch": 3.9694886119467125,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038311476511098053,
      "loss": 5.5444,
      "mean_token_accuracy": 0.1931615486741066,
      "num_tokens": 8399644.0,
      "step": 4620
    },
    {
      "entropy": 5.88284740447998,
      "epoch": 3.973785990545767,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003828230633844685,
      "loss": 5.5642,
      "mean_token_accuracy": 0.1885887861251831,
      "num_tokens": 8409264.0,
      "step": 4625
    },
    {
      "entropy": 5.928213977813721,
      "epoch": 3.9780833691448216,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038253112621548243,
      "loss": 5.5183,
      "mean_token_accuracy": 0.18407279402017593,
      "num_tokens": 8418383.0,
      "step": 4630
    },
    {
      "entropy": 5.955349254608154,
      "epoch": 3.982380747743876,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003822389542414966,
      "loss": 5.541,
      "mean_token_accuracy": 0.1865914523601532,
      "num_tokens": 8427411.0,
      "step": 4635
    },
    {
      "entropy": 5.922438764572144,
      "epoch": 3.9866781263429307,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038194654810049775,
      "loss": 5.4861,
      "mean_token_accuracy": 0.18929156064987182,
      "num_tokens": 8435537.0,
      "step": 4640
    },
    {
      "entropy": 5.887250232696533,
      "epoch": 3.9909755049419853,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000381653908430984,
      "loss": 5.5575,
      "mean_token_accuracy": 0.18391502797603607,
      "num_tokens": 8444400.0,
      "step": 4645
    },
    {
      "entropy": 5.968876838684082,
      "epoch": 3.99527288354104,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003813610358719634,
      "loss": 5.5495,
      "mean_token_accuracy": 0.1834430143237114,
      "num_tokens": 8453830.0,
      "step": 4650
    },
    {
      "entropy": 5.918107986450195,
      "epoch": 3.999570262140095,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00038106793106295266,
      "loss": 5.5068,
      "mean_token_accuracy": 0.19659096896648406,
      "num_tokens": 8463033.0,
      "step": 4655
    },
    {
      "entropy": 5.86531400680542,
      "epoch": 4.003437902879243,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003807745946439754,
      "loss": 5.2898,
      "mean_token_accuracy": 0.20586958858701918,
      "num_tokens": 8470740.0,
      "step": 4660
    },
    {
      "entropy": 5.880648136138916,
      "epoch": 4.007735281478298,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003804810272555612,
      "loss": 5.2641,
      "mean_token_accuracy": 0.20031324923038482,
      "num_tokens": 8480480.0,
      "step": 4665
    },
    {
      "entropy": 5.8310387134552,
      "epoch": 4.012032660077352,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003801872295387439,
      "loss": 5.226,
      "mean_token_accuracy": 0.2162238046526909,
      "num_tokens": 8489047.0,
      "step": 4670
    },
    {
      "entropy": 5.917058992385864,
      "epoch": 4.016330038676408,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003798932021350603,
      "loss": 5.3044,
      "mean_token_accuracy": 0.20220148265361787,
      "num_tokens": 8497763.0,
      "step": 4675
    },
    {
      "entropy": 5.9035172939300535,
      "epoch": 4.020627417275462,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037959894568654864,
      "loss": 5.2805,
      "mean_token_accuracy": 0.20567054748535157,
      "num_tokens": 8506814.0,
      "step": 4680
    },
    {
      "entropy": 6.014539337158203,
      "epoch": 4.024924795874517,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003793044608357474,
      "loss": 5.3835,
      "mean_token_accuracy": 0.1940508082509041,
      "num_tokens": 8516384.0,
      "step": 4685
    },
    {
      "entropy": 5.931362104415894,
      "epoch": 4.0292221744735714,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003790097482256939,
      "loss": 5.2328,
      "mean_token_accuracy": 0.2043454810976982,
      "num_tokens": 8524822.0,
      "step": 4690
    },
    {
      "entropy": 5.896431541442871,
      "epoch": 4.033519553072626,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003787148084999225,
      "loss": 5.2819,
      "mean_token_accuracy": 0.2060570999979973,
      "num_tokens": 8534129.0,
      "step": 4695
    },
    {
      "entropy": 5.873203897476197,
      "epoch": 4.0378169316716805,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037841964230246394,
      "loss": 5.3343,
      "mean_token_accuracy": 0.19927902966737748,
      "num_tokens": 8543235.0,
      "step": 4700
    },
    {
      "entropy": 5.831689119338989,
      "epoch": 4.042114310270735,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003781242502778429,
      "loss": 5.2313,
      "mean_token_accuracy": 0.21126292049884796,
      "num_tokens": 8551903.0,
      "step": 4705
    },
    {
      "entropy": 5.867668962478637,
      "epoch": 4.04641168886979,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037782863307107785,
      "loss": 5.3087,
      "mean_token_accuracy": 0.20029529184103012,
      "num_tokens": 8561173.0,
      "step": 4710
    },
    {
      "entropy": 5.9396202087402346,
      "epoch": 4.050709067468844,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037753279132767833,
      "loss": 5.2143,
      "mean_token_accuracy": 0.21614595949649812,
      "num_tokens": 8569789.0,
      "step": 4715
    },
    {
      "entropy": 5.84978985786438,
      "epoch": 4.055006446067899,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037723672569364453,
      "loss": 5.2249,
      "mean_token_accuracy": 0.20887062549591065,
      "num_tokens": 8577971.0,
      "step": 4720
    },
    {
      "entropy": 5.905433416366577,
      "epoch": 4.059303824666953,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037694043681546545,
      "loss": 5.3015,
      "mean_token_accuracy": 0.2034707173705101,
      "num_tokens": 8587299.0,
      "step": 4725
    },
    {
      "entropy": 5.830616998672485,
      "epoch": 4.063601203266008,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003766439253401177,
      "loss": 5.2617,
      "mean_token_accuracy": 0.20167354196310044,
      "num_tokens": 8595813.0,
      "step": 4730
    },
    {
      "entropy": 5.858702373504639,
      "epoch": 4.067898581865062,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037634719191506367,
      "loss": 5.2766,
      "mean_token_accuracy": 0.20875433683395386,
      "num_tokens": 8604552.0,
      "step": 4735
    },
    {
      "entropy": 5.813738679885864,
      "epoch": 4.072195960464117,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037605023718825065,
      "loss": 5.2166,
      "mean_token_accuracy": 0.20645739883184433,
      "num_tokens": 8612701.0,
      "step": 4740
    },
    {
      "entropy": 5.895049142837524,
      "epoch": 4.0764933390631715,
      "grad_norm": 0.96875,
      "learning_rate": 0.000375753061808109,
      "loss": 5.2796,
      "mean_token_accuracy": 0.20517786592245102,
      "num_tokens": 8622699.0,
      "step": 4745
    },
    {
      "entropy": 5.849208974838257,
      "epoch": 4.080790717662226,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037545566642355107,
      "loss": 5.2542,
      "mean_token_accuracy": 0.2037628948688507,
      "num_tokens": 8631821.0,
      "step": 4750
    },
    {
      "entropy": 5.8466846466064455,
      "epoch": 4.0850880962612806,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003751580516839695,
      "loss": 5.2253,
      "mean_token_accuracy": 0.20077935308218003,
      "num_tokens": 8641814.0,
      "step": 4755
    },
    {
      "entropy": 5.9084583759307865,
      "epoch": 4.089385474860335,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037486021823923574,
      "loss": 5.3032,
      "mean_token_accuracy": 0.20137091279029845,
      "num_tokens": 8649649.0,
      "step": 4760
    },
    {
      "entropy": 5.785796022415161,
      "epoch": 4.09368285345939,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037456216673969925,
      "loss": 5.2229,
      "mean_token_accuracy": 0.211810964345932,
      "num_tokens": 8658216.0,
      "step": 4765
    },
    {
      "entropy": 5.8987915992736815,
      "epoch": 4.097980232058444,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003742638978361851,
      "loss": 5.3158,
      "mean_token_accuracy": 0.19829458594322205,
      "num_tokens": 8667725.0,
      "step": 4770
    },
    {
      "entropy": 5.802492475509643,
      "epoch": 4.102277610657499,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037396541217999367,
      "loss": 5.1885,
      "mean_token_accuracy": 0.208291956782341,
      "num_tokens": 8675739.0,
      "step": 4775
    },
    {
      "entropy": 5.843210458755493,
      "epoch": 4.106574989256553,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003736667104228981,
      "loss": 5.2383,
      "mean_token_accuracy": 0.20806714445352553,
      "num_tokens": 8685764.0,
      "step": 4780
    },
    {
      "entropy": 5.875133657455445,
      "epoch": 4.110872367855608,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037336779321714376,
      "loss": 5.2339,
      "mean_token_accuracy": 0.2111602544784546,
      "num_tokens": 8695476.0,
      "step": 4785
    },
    {
      "entropy": 5.784117746353149,
      "epoch": 4.115169746454662,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037306866121544633,
      "loss": 5.3148,
      "mean_token_accuracy": 0.20060547441244125,
      "num_tokens": 8705544.0,
      "step": 4790
    },
    {
      "entropy": 5.895699596405029,
      "epoch": 4.119467125053717,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003727693150709904,
      "loss": 5.2852,
      "mean_token_accuracy": 0.20969280004501342,
      "num_tokens": 8714883.0,
      "step": 4795
    },
    {
      "entropy": 5.927401828765869,
      "epoch": 4.1237645036527715,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037246975543742843,
      "loss": 5.3251,
      "mean_token_accuracy": 0.2008032977581024,
      "num_tokens": 8724589.0,
      "step": 4800
    },
    {
      "entropy": 5.776608228683472,
      "epoch": 4.128061882251826,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000372169982968879,
      "loss": 5.209,
      "mean_token_accuracy": 0.2107146978378296,
      "num_tokens": 8733771.0,
      "step": 4805
    },
    {
      "entropy": 5.852010917663574,
      "epoch": 4.132359260850881,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003718699983199252,
      "loss": 5.2926,
      "mean_token_accuracy": 0.20044343173503876,
      "num_tokens": 8742348.0,
      "step": 4810
    },
    {
      "entropy": 5.804159498214721,
      "epoch": 4.136656639449935,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003715698021456137,
      "loss": 5.2355,
      "mean_token_accuracy": 0.21283494979143142,
      "num_tokens": 8751357.0,
      "step": 4815
    },
    {
      "entropy": 5.81705379486084,
      "epoch": 4.1409540180489905,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037126939510145294,
      "loss": 5.2994,
      "mean_token_accuracy": 0.20053811222314835,
      "num_tokens": 8760813.0,
      "step": 4820
    },
    {
      "entropy": 5.915247869491577,
      "epoch": 4.145251396648045,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003709687778434118,
      "loss": 5.327,
      "mean_token_accuracy": 0.2032524213194847,
      "num_tokens": 8770228.0,
      "step": 4825
    },
    {
      "entropy": 5.801777935028076,
      "epoch": 4.1495487752471,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003706679510279183,
      "loss": 5.1618,
      "mean_token_accuracy": 0.20846515148878098,
      "num_tokens": 8779351.0,
      "step": 4830
    },
    {
      "entropy": 5.876266050338745,
      "epoch": 4.153846153846154,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003703669153118578,
      "loss": 5.3271,
      "mean_token_accuracy": 0.19882068186998367,
      "num_tokens": 8789116.0,
      "step": 4835
    },
    {
      "entropy": 5.814975690841675,
      "epoch": 4.158143532445209,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037006567135257216,
      "loss": 5.2921,
      "mean_token_accuracy": 0.20319987535476686,
      "num_tokens": 8797790.0,
      "step": 4840
    },
    {
      "entropy": 5.877647066116333,
      "epoch": 4.162440911044263,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036976421980785764,
      "loss": 5.3344,
      "mean_token_accuracy": 0.20252140313386918,
      "num_tokens": 8808067.0,
      "step": 4845
    },
    {
      "entropy": 5.839996242523194,
      "epoch": 4.166738289643318,
      "grad_norm": 1.125,
      "learning_rate": 0.0003694625613359641,
      "loss": 5.2295,
      "mean_token_accuracy": 0.21065318286418916,
      "num_tokens": 8816587.0,
      "step": 4850
    },
    {
      "entropy": 5.878728818893433,
      "epoch": 4.171035668242372,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003691606965955929,
      "loss": 5.2976,
      "mean_token_accuracy": 0.20241713225841523,
      "num_tokens": 8826045.0,
      "step": 4855
    },
    {
      "entropy": 5.773685550689697,
      "epoch": 4.175333046841427,
      "grad_norm": 1.03125,
      "learning_rate": 0.000368858626245896,
      "loss": 5.2789,
      "mean_token_accuracy": 0.20736795663833618,
      "num_tokens": 8835427.0,
      "step": 4860
    },
    {
      "entropy": 5.827939033508301,
      "epoch": 4.1796304254404815,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003685563509464744,
      "loss": 5.2206,
      "mean_token_accuracy": 0.2094802051782608,
      "num_tokens": 8845167.0,
      "step": 4865
    },
    {
      "entropy": 5.888571691513062,
      "epoch": 4.183927804039536,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036825387135737647,
      "loss": 5.2419,
      "mean_token_accuracy": 0.20915429890155793,
      "num_tokens": 8853591.0,
      "step": 4870
    },
    {
      "entropy": 5.8387387752532955,
      "epoch": 4.188225182638591,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036795118813909674,
      "loss": 5.344,
      "mean_token_accuracy": 0.19490066319704055,
      "num_tokens": 8863647.0,
      "step": 4875
    },
    {
      "entropy": 5.900177431106568,
      "epoch": 4.192522561237645,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036764830195257437,
      "loss": 5.2694,
      "mean_token_accuracy": 0.20699672698974608,
      "num_tokens": 8872911.0,
      "step": 4880
    },
    {
      "entropy": 5.868921232223511,
      "epoch": 4.1968199398367,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003673452134591918,
      "loss": 5.3109,
      "mean_token_accuracy": 0.20538694262504578,
      "num_tokens": 8881001.0,
      "step": 4885
    },
    {
      "entropy": 5.80009765625,
      "epoch": 4.201117318435754,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000367041923320773,
      "loss": 5.2187,
      "mean_token_accuracy": 0.21138573586940765,
      "num_tokens": 8890323.0,
      "step": 4890
    },
    {
      "entropy": 5.85071029663086,
      "epoch": 4.205414697034809,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036673843219958257,
      "loss": 5.2568,
      "mean_token_accuracy": 0.206742163002491,
      "num_tokens": 8900471.0,
      "step": 4895
    },
    {
      "entropy": 5.9010358333587645,
      "epoch": 4.209712075633863,
      "grad_norm": 1.125,
      "learning_rate": 0.0003664347407583238,
      "loss": 5.3197,
      "mean_token_accuracy": 0.2024794563651085,
      "num_tokens": 8909320.0,
      "step": 4900
    },
    {
      "entropy": 5.80710015296936,
      "epoch": 4.214009454232918,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003661308496601373,
      "loss": 5.2378,
      "mean_token_accuracy": 0.21013146489858628,
      "num_tokens": 8917453.0,
      "step": 4905
    },
    {
      "entropy": 5.791521167755127,
      "epoch": 4.218306832831972,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036582675956859983,
      "loss": 5.2993,
      "mean_token_accuracy": 0.20273903012275696,
      "num_tokens": 8925737.0,
      "step": 4910
    },
    {
      "entropy": 5.754849910736084,
      "epoch": 4.222604211431027,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036552247114772263,
      "loss": 5.2318,
      "mean_token_accuracy": 0.20654798150062562,
      "num_tokens": 8935475.0,
      "step": 4915
    },
    {
      "entropy": 5.871701717376709,
      "epoch": 4.2269015900300815,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036521798506194996,
      "loss": 5.2637,
      "mean_token_accuracy": 0.21133471578359603,
      "num_tokens": 8944683.0,
      "step": 4920
    },
    {
      "entropy": 5.832123899459839,
      "epoch": 4.231198968629136,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036491330197615775,
      "loss": 5.3073,
      "mean_token_accuracy": 0.19945307672023774,
      "num_tokens": 8953837.0,
      "step": 4925
    },
    {
      "entropy": 5.831123399734497,
      "epoch": 4.235496347228191,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00036460842255565197,
      "loss": 5.3348,
      "mean_token_accuracy": 0.20013371407985686,
      "num_tokens": 8964822.0,
      "step": 4930
    },
    {
      "entropy": 5.883769750595093,
      "epoch": 4.239793725827245,
      "grad_norm": 1.375,
      "learning_rate": 0.0003643033474661676,
      "loss": 5.3194,
      "mean_token_accuracy": 0.20097711086273193,
      "num_tokens": 8974363.0,
      "step": 4935
    },
    {
      "entropy": 5.845435619354248,
      "epoch": 4.2440911044263,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036399807737386657,
      "loss": 5.241,
      "mean_token_accuracy": 0.20937079191207886,
      "num_tokens": 8983122.0,
      "step": 4940
    },
    {
      "entropy": 5.903400945663452,
      "epoch": 4.248388483025354,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003636926129453368,
      "loss": 5.3464,
      "mean_token_accuracy": 0.19763177782297134,
      "num_tokens": 8991618.0,
      "step": 4945
    },
    {
      "entropy": 5.878796339035034,
      "epoch": 4.252685861624409,
      "grad_norm": 1.125,
      "learning_rate": 0.0003633869548475904,
      "loss": 5.2559,
      "mean_token_accuracy": 0.20431189835071564,
      "num_tokens": 9000128.0,
      "step": 4950
    },
    {
      "entropy": 5.777056169509888,
      "epoch": 4.256983240223463,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003630811037480627,
      "loss": 5.2543,
      "mean_token_accuracy": 0.20079392343759536,
      "num_tokens": 9008951.0,
      "step": 4955
    },
    {
      "entropy": 5.850741004943847,
      "epoch": 4.261280618822518,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003627750603146101,
      "loss": 5.2832,
      "mean_token_accuracy": 0.20368402004241942,
      "num_tokens": 9018949.0,
      "step": 4960
    },
    {
      "entropy": 5.921943330764771,
      "epoch": 4.265577997421573,
      "grad_norm": 1.0,
      "learning_rate": 0.0003624688252155091,
      "loss": 5.2951,
      "mean_token_accuracy": 0.2009932816028595,
      "num_tokens": 9028910.0,
      "step": 4965
    },
    {
      "entropy": 5.803747653961182,
      "epoch": 4.269875376020628,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003621623991194549,
      "loss": 5.3558,
      "mean_token_accuracy": 0.19536107927560806,
      "num_tokens": 9039012.0,
      "step": 4970
    },
    {
      "entropy": 5.890829372406006,
      "epoch": 4.274172754619682,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003618557826955594,
      "loss": 5.316,
      "mean_token_accuracy": 0.20139972418546676,
      "num_tokens": 9048639.0,
      "step": 4975
    },
    {
      "entropy": 5.8294120788574215,
      "epoch": 4.278470133218737,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036154897661335063,
      "loss": 5.2675,
      "mean_token_accuracy": 0.20398803800344467,
      "num_tokens": 9057453.0,
      "step": 4980
    },
    {
      "entropy": 5.815646457672119,
      "epoch": 4.2827675118177915,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003612419815427702,
      "loss": 5.3112,
      "mean_token_accuracy": 0.19857749342918396,
      "num_tokens": 9066761.0,
      "step": 4985
    },
    {
      "entropy": 5.903983926773071,
      "epoch": 4.287064890416846,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003609347981541726,
      "loss": 5.3668,
      "mean_token_accuracy": 0.1965751200914383,
      "num_tokens": 9075535.0,
      "step": 4990
    },
    {
      "entropy": 5.863292837142945,
      "epoch": 4.291362269015901,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036062742711832376,
      "loss": 5.2677,
      "mean_token_accuracy": 0.2066899359226227,
      "num_tokens": 9084559.0,
      "step": 4995
    },
    {
      "entropy": 5.843481111526489,
      "epoch": 4.295659647614955,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003603198691063991,
      "loss": 5.2527,
      "mean_token_accuracy": 0.2051399454474449,
      "num_tokens": 9093069.0,
      "step": 5000
    },
    {
      "epoch": 4.295659647614955,
      "eval_entropy": 5.614789016074963,
      "eval_loss": 5.965892314910889,
      "eval_mean_token_accuracy": 0.17482336706204996,
      "eval_num_tokens": 9093069.0,
      "eval_runtime": 2.0568,
      "eval_samples_per_second": 1725.527,
      "eval_steps_per_second": 215.873,
      "step": 5000
    },
    {
      "entropy": 5.814894962310791,
      "epoch": 4.29995702621401,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003600121247899824,
      "loss": 5.2394,
      "mean_token_accuracy": 0.2070020854473114,
      "num_tokens": 9101914.0,
      "step": 5005
    },
    {
      "entropy": 5.834060192108154,
      "epoch": 4.304254404813064,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035970419484106404,
      "loss": 5.3231,
      "mean_token_accuracy": 0.1985424891114235,
      "num_tokens": 9110967.0,
      "step": 5010
    },
    {
      "entropy": 5.967706966400146,
      "epoch": 4.308551783412119,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003593960799320402,
      "loss": 5.4054,
      "mean_token_accuracy": 0.20113543570041656,
      "num_tokens": 9119774.0,
      "step": 5015
    },
    {
      "entropy": 5.8776092529296875,
      "epoch": 4.312849162011173,
      "grad_norm": 1.25,
      "learning_rate": 0.0003590877807357107,
      "loss": 5.3236,
      "mean_token_accuracy": 0.19780130684375763,
      "num_tokens": 9127738.0,
      "step": 5020
    },
    {
      "entropy": 5.796004724502564,
      "epoch": 4.317146540610228,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003587792979252776,
      "loss": 5.3027,
      "mean_token_accuracy": 0.20867908447980882,
      "num_tokens": 9137060.0,
      "step": 5025
    },
    {
      "entropy": 5.813083219528198,
      "epoch": 4.321443919209282,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003584706321743442,
      "loss": 5.2058,
      "mean_token_accuracy": 0.20791620910167694,
      "num_tokens": 9145169.0,
      "step": 5030
    },
    {
      "entropy": 5.838433027267456,
      "epoch": 4.325741297808337,
      "grad_norm": 1.09375,
      "learning_rate": 0.000358161784156913,
      "loss": 5.2557,
      "mean_token_accuracy": 0.2082034319639206,
      "num_tokens": 9154092.0,
      "step": 5035
    },
    {
      "entropy": 5.872574281692505,
      "epoch": 4.3300386764073915,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035785275454738456,
      "loss": 5.303,
      "mean_token_accuracy": 0.19724357575178147,
      "num_tokens": 9162824.0,
      "step": 5040
    },
    {
      "entropy": 5.770854759216308,
      "epoch": 4.334336055006446,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035754354402055635,
      "loss": 5.2083,
      "mean_token_accuracy": 0.212600913643837,
      "num_tokens": 9170977.0,
      "step": 5045
    },
    {
      "entropy": 5.752213048934936,
      "epoch": 4.338633433605501,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003572341532516202,
      "loss": 5.2475,
      "mean_token_accuracy": 0.20813672840595246,
      "num_tokens": 9179539.0,
      "step": 5050
    },
    {
      "entropy": 5.801317834854126,
      "epoch": 4.342930812204555,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003569245829161622,
      "loss": 5.3502,
      "mean_token_accuracy": 0.1998053953051567,
      "num_tokens": 9188861.0,
      "step": 5055
    },
    {
      "entropy": 5.84930191040039,
      "epoch": 4.34722819080361,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035661483369016004,
      "loss": 5.2837,
      "mean_token_accuracy": 0.1959220364689827,
      "num_tokens": 9197724.0,
      "step": 5060
    },
    {
      "entropy": 5.84346284866333,
      "epoch": 4.351525569402664,
      "grad_norm": 1.125,
      "learning_rate": 0.0003563049062499822,
      "loss": 5.3157,
      "mean_token_accuracy": 0.20674706101417542,
      "num_tokens": 9206375.0,
      "step": 5065
    },
    {
      "entropy": 5.787398624420166,
      "epoch": 4.355822948001719,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003559948012723865,
      "loss": 5.2601,
      "mean_token_accuracy": 0.20788746029138566,
      "num_tokens": 9214675.0,
      "step": 5070
    },
    {
      "entropy": 5.825500440597534,
      "epoch": 4.360120326600773,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003556845194345181,
      "loss": 5.2695,
      "mean_token_accuracy": 0.2058370053768158,
      "num_tokens": 9224128.0,
      "step": 5075
    },
    {
      "entropy": 5.749603271484375,
      "epoch": 4.364417705199828,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003553740614139086,
      "loss": 5.1964,
      "mean_token_accuracy": 0.20808056592941285,
      "num_tokens": 9232568.0,
      "step": 5080
    },
    {
      "entropy": 5.83976263999939,
      "epoch": 4.368715083798882,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003550634278884742,
      "loss": 5.3091,
      "mean_token_accuracy": 0.20686423778533936,
      "num_tokens": 9241809.0,
      "step": 5085
    },
    {
      "entropy": 5.814407444000244,
      "epoch": 4.373012462397937,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035475261953651433,
      "loss": 5.293,
      "mean_token_accuracy": 0.20587006211280823,
      "num_tokens": 9250845.0,
      "step": 5090
    },
    {
      "entropy": 5.7717420101165775,
      "epoch": 4.3773098409969915,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035444163703671026,
      "loss": 5.2598,
      "mean_token_accuracy": 0.21119688749313353,
      "num_tokens": 9259465.0,
      "step": 5095
    },
    {
      "entropy": 5.800244855880737,
      "epoch": 4.381607219596046,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00035413048106812357,
      "loss": 5.235,
      "mean_token_accuracy": 0.21408589631319047,
      "num_tokens": 9267853.0,
      "step": 5100
    },
    {
      "entropy": 5.919088697433471,
      "epoch": 4.385904598195101,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035381915231019425,
      "loss": 5.4326,
      "mean_token_accuracy": 0.18999635130167009,
      "num_tokens": 9276664.0,
      "step": 5105
    },
    {
      "entropy": 5.865285396575928,
      "epoch": 4.390201976794156,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003535076514427401,
      "loss": 5.2447,
      "mean_token_accuracy": 0.2045974314212799,
      "num_tokens": 9285482.0,
      "step": 5110
    },
    {
      "entropy": 5.868390226364136,
      "epoch": 4.39449935539321,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035319597914595436,
      "loss": 5.354,
      "mean_token_accuracy": 0.19328553676605226,
      "num_tokens": 9293936.0,
      "step": 5115
    },
    {
      "entropy": 5.828657436370849,
      "epoch": 4.398796733992265,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003528841361004049,
      "loss": 5.3605,
      "mean_token_accuracy": 0.1871650382876396,
      "num_tokens": 9303998.0,
      "step": 5120
    },
    {
      "entropy": 5.7928454875946045,
      "epoch": 4.40309411259132,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003525721229870323,
      "loss": 5.3003,
      "mean_token_accuracy": 0.2077128380537033,
      "num_tokens": 9313117.0,
      "step": 5125
    },
    {
      "entropy": 5.904635334014893,
      "epoch": 4.407391491190374,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035225994048714823,
      "loss": 5.311,
      "mean_token_accuracy": 0.20455257445573807,
      "num_tokens": 9321446.0,
      "step": 5130
    },
    {
      "entropy": 5.800194549560547,
      "epoch": 4.411688869789429,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003519475892824348,
      "loss": 5.2937,
      "mean_token_accuracy": 0.19881632030010224,
      "num_tokens": 9330752.0,
      "step": 5135
    },
    {
      "entropy": 5.801384353637696,
      "epoch": 4.415986248388483,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003516350700549419,
      "loss": 5.3344,
      "mean_token_accuracy": 0.20338008254766465,
      "num_tokens": 9339322.0,
      "step": 5140
    },
    {
      "entropy": 5.8580115795135494,
      "epoch": 4.420283626987538,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035132238348708697,
      "loss": 5.347,
      "mean_token_accuracy": 0.1988745167851448,
      "num_tokens": 9349024.0,
      "step": 5145
    },
    {
      "entropy": 5.965443801879883,
      "epoch": 4.424581005586592,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035100953026165224,
      "loss": 5.439,
      "mean_token_accuracy": 0.19176600128412247,
      "num_tokens": 9358833.0,
      "step": 5150
    },
    {
      "entropy": 5.869590854644775,
      "epoch": 4.428878384185647,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003506965110617841,
      "loss": 5.2808,
      "mean_token_accuracy": 0.208394593000412,
      "num_tokens": 9368276.0,
      "step": 5155
    },
    {
      "entropy": 5.892330980300903,
      "epoch": 4.4331757627847015,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003503833265709915,
      "loss": 5.3683,
      "mean_token_accuracy": 0.19563932716846466,
      "num_tokens": 9378501.0,
      "step": 5160
    },
    {
      "entropy": 5.908583688735962,
      "epoch": 4.437473141383756,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035006997747314404,
      "loss": 5.3235,
      "mean_token_accuracy": 0.19323734641075135,
      "num_tokens": 9387789.0,
      "step": 5165
    },
    {
      "entropy": 5.875285339355469,
      "epoch": 4.441770519982811,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034975646445247106,
      "loss": 5.3894,
      "mean_token_accuracy": 0.19741396009922027,
      "num_tokens": 9397041.0,
      "step": 5170
    },
    {
      "entropy": 5.823545503616333,
      "epoch": 4.446067898581865,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003494427881935596,
      "loss": 5.3172,
      "mean_token_accuracy": 0.2029438406229019,
      "num_tokens": 9405393.0,
      "step": 5175
    },
    {
      "entropy": 5.808364534378052,
      "epoch": 4.45036527718092,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034912894938135325,
      "loss": 5.2864,
      "mean_token_accuracy": 0.2051060155034065,
      "num_tokens": 9415127.0,
      "step": 5180
    },
    {
      "entropy": 5.826566696166992,
      "epoch": 4.454662655779974,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003488149487011506,
      "loss": 5.3995,
      "mean_token_accuracy": 0.20431725531816483,
      "num_tokens": 9424416.0,
      "step": 5185
    },
    {
      "entropy": 5.93192777633667,
      "epoch": 4.458960034379029,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034850078683860346,
      "loss": 5.3621,
      "mean_token_accuracy": 0.19914545714855195,
      "num_tokens": 9434523.0,
      "step": 5190
    },
    {
      "entropy": 5.8803558349609375,
      "epoch": 4.463257412978083,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003481864644797159,
      "loss": 5.343,
      "mean_token_accuracy": 0.2089121863245964,
      "num_tokens": 9443605.0,
      "step": 5195
    },
    {
      "entropy": 5.805825328826904,
      "epoch": 4.467554791577138,
      "grad_norm": 1.125,
      "learning_rate": 0.0003478719823108424,
      "loss": 5.3538,
      "mean_token_accuracy": 0.1973356932401657,
      "num_tokens": 9453268.0,
      "step": 5200
    },
    {
      "entropy": 5.84867639541626,
      "epoch": 4.471852170176192,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034755734101868613,
      "loss": 5.2371,
      "mean_token_accuracy": 0.2077445864677429,
      "num_tokens": 9461578.0,
      "step": 5205
    },
    {
      "entropy": 5.7984278202056885,
      "epoch": 4.476149548775247,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034724254129029795,
      "loss": 5.2669,
      "mean_token_accuracy": 0.20249382108449937,
      "num_tokens": 9470722.0,
      "step": 5210
    },
    {
      "entropy": 5.826017284393311,
      "epoch": 4.4804469273743015,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003469275838130748,
      "loss": 5.3588,
      "mean_token_accuracy": 0.19659536480903625,
      "num_tokens": 9479695.0,
      "step": 5215
    },
    {
      "entropy": 5.893008232116699,
      "epoch": 4.484744305973356,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003466124692747577,
      "loss": 5.2917,
      "mean_token_accuracy": 0.20469566136598588,
      "num_tokens": 9488444.0,
      "step": 5220
    },
    {
      "entropy": 5.802812480926514,
      "epoch": 4.489041684572411,
      "grad_norm": 1.125,
      "learning_rate": 0.00034629719836343106,
      "loss": 5.2358,
      "mean_token_accuracy": 0.2110652819275856,
      "num_tokens": 9497413.0,
      "step": 5225
    },
    {
      "entropy": 5.821634435653687,
      "epoch": 4.493339063171465,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003459817717675203,
      "loss": 5.2796,
      "mean_token_accuracy": 0.2101682260632515,
      "num_tokens": 9506135.0,
      "step": 5230
    },
    {
      "entropy": 5.847463178634643,
      "epoch": 4.49763644177052,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003456661901757913,
      "loss": 5.3575,
      "mean_token_accuracy": 0.19278268665075302,
      "num_tokens": 9516918.0,
      "step": 5235
    },
    {
      "entropy": 5.917444944381714,
      "epoch": 4.501933820369574,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034535045427734796,
      "loss": 5.3141,
      "mean_token_accuracy": 0.20803753286600113,
      "num_tokens": 9526052.0,
      "step": 5240
    },
    {
      "entropy": 5.796197938919067,
      "epoch": 4.506231198968629,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003450345647616313,
      "loss": 5.3677,
      "mean_token_accuracy": 0.2010870099067688,
      "num_tokens": 9535200.0,
      "step": 5245
    },
    {
      "entropy": 5.795945453643799,
      "epoch": 4.510528577567683,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003447185223184177,
      "loss": 5.3205,
      "mean_token_accuracy": 0.2039222776889801,
      "num_tokens": 9544786.0,
      "step": 5250
    },
    {
      "entropy": 5.844666004180908,
      "epoch": 4.514825956166739,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034440232763781765,
      "loss": 5.2781,
      "mean_token_accuracy": 0.20815713405609132,
      "num_tokens": 9553694.0,
      "step": 5255
    },
    {
      "entropy": 5.7694135189056395,
      "epoch": 4.519123334765792,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000344085981410274,
      "loss": 5.3284,
      "mean_token_accuracy": 0.20543634742498398,
      "num_tokens": 9563332.0,
      "step": 5260
    },
    {
      "entropy": 5.762286853790283,
      "epoch": 4.523420713364848,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034376948432656036,
      "loss": 5.2519,
      "mean_token_accuracy": 0.21355798244476318,
      "num_tokens": 9572367.0,
      "step": 5265
    },
    {
      "entropy": 5.908457088470459,
      "epoch": 4.527718091963902,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003434528370777798,
      "loss": 5.338,
      "mean_token_accuracy": 0.1996489316225052,
      "num_tokens": 9582535.0,
      "step": 5270
    },
    {
      "entropy": 5.8559120178222654,
      "epoch": 4.532015470562957,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034313604035536344,
      "loss": 5.3032,
      "mean_token_accuracy": 0.20419186502695083,
      "num_tokens": 9590688.0,
      "step": 5275
    },
    {
      "entropy": 5.795069074630737,
      "epoch": 4.5363128491620115,
      "grad_norm": 1.125,
      "learning_rate": 0.0003428190948510687,
      "loss": 5.3354,
      "mean_token_accuracy": 0.2021968275308609,
      "num_tokens": 9599209.0,
      "step": 5280
    },
    {
      "entropy": 5.866246700286865,
      "epoch": 4.540610227761066,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003425020012569778,
      "loss": 5.3723,
      "mean_token_accuracy": 0.1955377459526062,
      "num_tokens": 9608575.0,
      "step": 5285
    },
    {
      "entropy": 5.913688898086548,
      "epoch": 4.544907606360121,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034218476026549665,
      "loss": 5.326,
      "mean_token_accuracy": 0.20012779533863068,
      "num_tokens": 9617312.0,
      "step": 5290
    },
    {
      "entropy": 5.8284680366516115,
      "epoch": 4.549204984959175,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003418673725693524,
      "loss": 5.3094,
      "mean_token_accuracy": 0.20884560644626618,
      "num_tokens": 9626398.0,
      "step": 5295
    },
    {
      "entropy": 5.806579303741455,
      "epoch": 4.55350236355823,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003415498388615932,
      "loss": 5.2941,
      "mean_token_accuracy": 0.19823911041021347,
      "num_tokens": 9635470.0,
      "step": 5300
    },
    {
      "entropy": 5.854410552978516,
      "epoch": 4.557799742157284,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003412321598355857,
      "loss": 5.255,
      "mean_token_accuracy": 0.20601650178432465,
      "num_tokens": 9644728.0,
      "step": 5305
    },
    {
      "entropy": 5.7743758201599125,
      "epoch": 4.562097120756339,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003409143361850139,
      "loss": 5.3091,
      "mean_token_accuracy": 0.2028186500072479,
      "num_tokens": 9654129.0,
      "step": 5310
    },
    {
      "entropy": 5.799901103973388,
      "epoch": 4.566394499355393,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003405963686038775,
      "loss": 5.3917,
      "mean_token_accuracy": 0.19845977276563645,
      "num_tokens": 9662648.0,
      "step": 5315
    },
    {
      "entropy": 5.876940774917602,
      "epoch": 4.570691877954448,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003402782577864908,
      "loss": 5.3587,
      "mean_token_accuracy": 0.203027606010437,
      "num_tokens": 9672082.0,
      "step": 5320
    },
    {
      "entropy": 5.888137006759644,
      "epoch": 4.574989256553502,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033996000442748056,
      "loss": 5.2729,
      "mean_token_accuracy": 0.20779376924037934,
      "num_tokens": 9681422.0,
      "step": 5325
    },
    {
      "entropy": 5.831095361709595,
      "epoch": 4.579286635152557,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033964160922178495,
      "loss": 5.3151,
      "mean_token_accuracy": 0.19961357861757278,
      "num_tokens": 9690675.0,
      "step": 5330
    },
    {
      "entropy": 5.843279647827148,
      "epoch": 4.5835840137516115,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003393230728646518,
      "loss": 5.2947,
      "mean_token_accuracy": 0.2076662614941597,
      "num_tokens": 9700200.0,
      "step": 5335
    },
    {
      "entropy": 5.791910076141358,
      "epoch": 4.587881392350666,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033900439605163724,
      "loss": 5.3031,
      "mean_token_accuracy": 0.19897290766239167,
      "num_tokens": 9709533.0,
      "step": 5340
    },
    {
      "entropy": 5.786693143844604,
      "epoch": 4.592178770949721,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033868557947860407,
      "loss": 5.3277,
      "mean_token_accuracy": 0.20447840094566344,
      "num_tokens": 9719250.0,
      "step": 5345
    },
    {
      "entropy": 5.854731130599975,
      "epoch": 4.596476149548775,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033836662384172014,
      "loss": 5.2547,
      "mean_token_accuracy": 0.21334317177534104,
      "num_tokens": 9727837.0,
      "step": 5350
    },
    {
      "entropy": 5.809204912185669,
      "epoch": 4.60077352814783,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003380475298374573,
      "loss": 5.3575,
      "mean_token_accuracy": 0.2006920963525772,
      "num_tokens": 9737125.0,
      "step": 5355
    },
    {
      "entropy": 5.838610601425171,
      "epoch": 4.605070906746884,
      "grad_norm": 1.078125,
      "learning_rate": 0.000337728298162589,
      "loss": 5.3661,
      "mean_token_accuracy": 0.19502367824316025,
      "num_tokens": 9746309.0,
      "step": 5360
    },
    {
      "entropy": 5.8516277313232425,
      "epoch": 4.609368285345939,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033740892951418993,
      "loss": 5.2468,
      "mean_token_accuracy": 0.20967283546924592,
      "num_tokens": 9755633.0,
      "step": 5365
    },
    {
      "entropy": 5.862628698348999,
      "epoch": 4.613665663944993,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003370894245896333,
      "loss": 5.2997,
      "mean_token_accuracy": 0.19413063079118728,
      "num_tokens": 9765179.0,
      "step": 5370
    },
    {
      "entropy": 5.904676723480224,
      "epoch": 4.617963042544048,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033676978408659047,
      "loss": 5.3394,
      "mean_token_accuracy": 0.19273030310869216,
      "num_tokens": 9774085.0,
      "step": 5375
    },
    {
      "entropy": 5.841152763366699,
      "epoch": 4.622260421143102,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003364500087030283,
      "loss": 5.4193,
      "mean_token_accuracy": 0.1930931881070137,
      "num_tokens": 9784650.0,
      "step": 5380
    },
    {
      "entropy": 5.88223671913147,
      "epoch": 4.626557799742157,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033613009913720845,
      "loss": 5.285,
      "mean_token_accuracy": 0.20238686352968216,
      "num_tokens": 9793947.0,
      "step": 5385
    },
    {
      "entropy": 5.776455116271973,
      "epoch": 4.6308551783412115,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033581005608768563,
      "loss": 5.2685,
      "mean_token_accuracy": 0.20969274193048476,
      "num_tokens": 9803593.0,
      "step": 5390
    },
    {
      "entropy": 5.781027317047119,
      "epoch": 4.635152556940266,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003354898802533058,
      "loss": 5.2996,
      "mean_token_accuracy": 0.2022281602025032,
      "num_tokens": 9812295.0,
      "step": 5395
    },
    {
      "entropy": 5.817134523391724,
      "epoch": 4.6394499355393215,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003351695723332051,
      "loss": 5.3112,
      "mean_token_accuracy": 0.20822529643774032,
      "num_tokens": 9820586.0,
      "step": 5400
    },
    {
      "entropy": 5.817539119720459,
      "epoch": 4.643747314138375,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033484913302680807,
      "loss": 5.2541,
      "mean_token_accuracy": 0.20944181829690933,
      "num_tokens": 9829080.0,
      "step": 5405
    },
    {
      "entropy": 5.83193678855896,
      "epoch": 4.648044692737431,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033452856303382595,
      "loss": 5.2627,
      "mean_token_accuracy": 0.201676806807518,
      "num_tokens": 9838421.0,
      "step": 5410
    },
    {
      "entropy": 5.831003665924072,
      "epoch": 4.652342071336484,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003342078630542555,
      "loss": 5.2663,
      "mean_token_accuracy": 0.209694604575634,
      "num_tokens": 9847151.0,
      "step": 5415
    },
    {
      "entropy": 5.848418092727661,
      "epoch": 4.65663944993554,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033388703378837737,
      "loss": 5.2884,
      "mean_token_accuracy": 0.2060145139694214,
      "num_tokens": 9856803.0,
      "step": 5420
    },
    {
      "entropy": 5.795956420898437,
      "epoch": 4.660936828534594,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003335660759367544,
      "loss": 5.2041,
      "mean_token_accuracy": 0.21581413298845292,
      "num_tokens": 9865617.0,
      "step": 5425
    },
    {
      "entropy": 5.771609735488892,
      "epoch": 4.665234207133649,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033324499020023025,
      "loss": 5.276,
      "mean_token_accuracy": 0.21252951323986052,
      "num_tokens": 9875454.0,
      "step": 5430
    },
    {
      "entropy": 5.854887962341309,
      "epoch": 4.669531585732703,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003329237772799277,
      "loss": 5.2692,
      "mean_token_accuracy": 0.20226241499185563,
      "num_tokens": 9884770.0,
      "step": 5435
    },
    {
      "entropy": 5.834545612335205,
      "epoch": 4.673828964331758,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003326024378772477,
      "loss": 5.2703,
      "mean_token_accuracy": 0.2073190376162529,
      "num_tokens": 9893594.0,
      "step": 5440
    },
    {
      "entropy": 5.85920524597168,
      "epoch": 4.678126342930812,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003322809726938667,
      "loss": 5.3803,
      "mean_token_accuracy": 0.19288230538368226,
      "num_tokens": 9902260.0,
      "step": 5445
    },
    {
      "entropy": 5.826963615417481,
      "epoch": 4.682423721529867,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033195938243173645,
      "loss": 5.2868,
      "mean_token_accuracy": 0.2081781357526779,
      "num_tokens": 9911020.0,
      "step": 5450
    },
    {
      "entropy": 5.853608512878418,
      "epoch": 4.6867211001289215,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003316376677930814,
      "loss": 5.3074,
      "mean_token_accuracy": 0.2044661432504654,
      "num_tokens": 9918696.0,
      "step": 5455
    },
    {
      "entropy": 5.791952800750733,
      "epoch": 4.691018478727976,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003313158294803977,
      "loss": 5.3308,
      "mean_token_accuracy": 0.19685716181993484,
      "num_tokens": 9927638.0,
      "step": 5460
    },
    {
      "entropy": 5.853991556167602,
      "epoch": 4.695315857327031,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033099386819645176,
      "loss": 5.3095,
      "mean_token_accuracy": 0.19899636507034302,
      "num_tokens": 9936969.0,
      "step": 5465
    },
    {
      "entropy": 5.822479629516602,
      "epoch": 4.699613235926085,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003306717846442782,
      "loss": 5.2259,
      "mean_token_accuracy": 0.2068831965327263,
      "num_tokens": 9945229.0,
      "step": 5470
    },
    {
      "entropy": 5.791026258468628,
      "epoch": 4.70391061452514,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003303495795271788,
      "loss": 5.2049,
      "mean_token_accuracy": 0.20886743813753128,
      "num_tokens": 9953759.0,
      "step": 5475
    },
    {
      "entropy": 5.823954963684082,
      "epoch": 4.708207993124194,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033002725354872075,
      "loss": 5.3268,
      "mean_token_accuracy": 0.19850837290287018,
      "num_tokens": 9962771.0,
      "step": 5480
    },
    {
      "entropy": 5.844640493392944,
      "epoch": 4.712505371723249,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032970480741273514,
      "loss": 5.3256,
      "mean_token_accuracy": 0.18878505975008011,
      "num_tokens": 9972481.0,
      "step": 5485
    },
    {
      "entropy": 5.88434624671936,
      "epoch": 4.716802750322303,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003293822418233155,
      "loss": 5.2746,
      "mean_token_accuracy": 0.20000507831573486,
      "num_tokens": 9980773.0,
      "step": 5490
    },
    {
      "entropy": 5.885302734375,
      "epoch": 4.721100128921358,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003290595574848161,
      "loss": 5.3655,
      "mean_token_accuracy": 0.1920745849609375,
      "num_tokens": 9989830.0,
      "step": 5495
    },
    {
      "entropy": 5.776205539703369,
      "epoch": 4.725397507520412,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003287367551018505,
      "loss": 5.286,
      "mean_token_accuracy": 0.205222225189209,
      "num_tokens": 9999234.0,
      "step": 5500
    },
    {
      "epoch": 4.725397507520412,
      "eval_entropy": 5.63656607619277,
      "eval_loss": 5.927424907684326,
      "eval_mean_token_accuracy": 0.17751722368369768,
      "eval_num_tokens": 9999234.0,
      "eval_runtime": 2.249,
      "eval_samples_per_second": 1578.01,
      "eval_steps_per_second": 197.418,
      "step": 5500
    },
    {
      "entropy": 5.829825210571289,
      "epoch": 4.729694886119467,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003284138353792903,
      "loss": 5.3415,
      "mean_token_accuracy": 0.19969913810491563,
      "num_tokens": 10008671.0,
      "step": 5505
    },
    {
      "entropy": 5.80418119430542,
      "epoch": 4.7339922647185215,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003280907990222628,
      "loss": 5.321,
      "mean_token_accuracy": 0.1996894508600235,
      "num_tokens": 10017170.0,
      "step": 5510
    },
    {
      "entropy": 5.7827003479003904,
      "epoch": 4.738289643317576,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032776764673615055,
      "loss": 5.3449,
      "mean_token_accuracy": 0.19481876939535142,
      "num_tokens": 10025712.0,
      "step": 5515
    },
    {
      "entropy": 5.837228584289551,
      "epoch": 4.742587021916631,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003274443792265888,
      "loss": 5.3326,
      "mean_token_accuracy": 0.20928302854299546,
      "num_tokens": 10035297.0,
      "step": 5520
    },
    {
      "entropy": 5.864207553863525,
      "epoch": 4.746884400515685,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032712099719946474,
      "loss": 5.3095,
      "mean_token_accuracy": 0.2033199653029442,
      "num_tokens": 10043903.0,
      "step": 5525
    },
    {
      "entropy": 5.811628532409668,
      "epoch": 4.75118177911474,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032679750136091533,
      "loss": 5.3235,
      "mean_token_accuracy": 0.19918979555368424,
      "num_tokens": 10053035.0,
      "step": 5530
    },
    {
      "entropy": 5.733118009567261,
      "epoch": 4.755479157713794,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003264738924173262,
      "loss": 5.2886,
      "mean_token_accuracy": 0.20847785323858262,
      "num_tokens": 10061911.0,
      "step": 5535
    },
    {
      "entropy": 5.798403263092041,
      "epoch": 4.759776536312849,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032615017107533,
      "loss": 5.2955,
      "mean_token_accuracy": 0.2083728313446045,
      "num_tokens": 10070738.0,
      "step": 5540
    },
    {
      "entropy": 5.812806463241577,
      "epoch": 4.764073914911903,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003258263380418047,
      "loss": 5.3061,
      "mean_token_accuracy": 0.20582663267850876,
      "num_tokens": 10080638.0,
      "step": 5545
    },
    {
      "entropy": 5.949069786071777,
      "epoch": 4.768371293510958,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032550239402387226,
      "loss": 5.3432,
      "mean_token_accuracy": 0.19543366134166718,
      "num_tokens": 10089429.0,
      "step": 5550
    },
    {
      "entropy": 5.771684646606445,
      "epoch": 4.772668672110013,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032517833972889695,
      "loss": 5.2367,
      "mean_token_accuracy": 0.20548324584960936,
      "num_tokens": 10098109.0,
      "step": 5555
    },
    {
      "entropy": 5.854096174240112,
      "epoch": 4.776966050709067,
      "grad_norm": 1.125,
      "learning_rate": 0.00032485417586448375,
      "loss": 5.3186,
      "mean_token_accuracy": 0.20433283299207688,
      "num_tokens": 10106808.0,
      "step": 5560
    },
    {
      "entropy": 5.865534257888794,
      "epoch": 4.781263429308122,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000324529903138477,
      "loss": 5.3383,
      "mean_token_accuracy": 0.19944437593221664,
      "num_tokens": 10116372.0,
      "step": 5565
    },
    {
      "entropy": 5.804335594177246,
      "epoch": 4.785560807907177,
      "grad_norm": 1.125,
      "learning_rate": 0.0003242055222589587,
      "loss": 5.2334,
      "mean_token_accuracy": 0.21369995176792145,
      "num_tokens": 10125256.0,
      "step": 5570
    },
    {
      "entropy": 5.84380407333374,
      "epoch": 4.7898581865062315,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000323881033934247,
      "loss": 5.383,
      "mean_token_accuracy": 0.19162997752428054,
      "num_tokens": 10134784.0,
      "step": 5575
    },
    {
      "entropy": 5.883758211135865,
      "epoch": 4.794155565105286,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032355643887289486,
      "loss": 5.319,
      "mean_token_accuracy": 0.20510610938072205,
      "num_tokens": 10144324.0,
      "step": 5580
    },
    {
      "entropy": 5.8348699569702145,
      "epoch": 4.798452943704341,
      "grad_norm": 1.125,
      "learning_rate": 0.0003232317377836881,
      "loss": 5.3516,
      "mean_token_accuracy": 0.19644383192062378,
      "num_tokens": 10152866.0,
      "step": 5585
    },
    {
      "entropy": 5.7965476512908936,
      "epoch": 4.802750322303395,
      "grad_norm": 1.125,
      "learning_rate": 0.000322906931375644,
      "loss": 5.2767,
      "mean_token_accuracy": 0.20827958583831788,
      "num_tokens": 10162457.0,
      "step": 5590
    },
    {
      "entropy": 5.807642841339112,
      "epoch": 4.80704770090245,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032258202035801,
      "loss": 5.3475,
      "mean_token_accuracy": 0.1958251029253006,
      "num_tokens": 10171604.0,
      "step": 5595
    },
    {
      "entropy": 5.893746852874756,
      "epoch": 4.811345079501504,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000322257005440262,
      "loss": 5.3107,
      "mean_token_accuracy": 0.2053775817155838,
      "num_tokens": 10180762.0,
      "step": 5600
    },
    {
      "entropy": 5.770109701156616,
      "epoch": 4.815642458100559,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003219318873321025,
      "loss": 5.237,
      "mean_token_accuracy": 0.22254517376422883,
      "num_tokens": 10189122.0,
      "step": 5605
    },
    {
      "entropy": 5.815287780761719,
      "epoch": 4.819939836699613,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032160666674345954,
      "loss": 5.3277,
      "mean_token_accuracy": 0.19555671364068986,
      "num_tokens": 10197280.0,
      "step": 5610
    },
    {
      "entropy": 5.8346967697143555,
      "epoch": 4.824237215298668,
      "grad_norm": 1.03125,
      "learning_rate": 0.00032128134438448504,
      "loss": 5.3607,
      "mean_token_accuracy": 0.1988934874534607,
      "num_tokens": 10207507.0,
      "step": 5615
    },
    {
      "entropy": 5.854104518890381,
      "epoch": 4.828534593897722,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032095592096555284,
      "loss": 5.3357,
      "mean_token_accuracy": 0.20055959671735762,
      "num_tokens": 10217584.0,
      "step": 5620
    },
    {
      "entropy": 5.826172304153443,
      "epoch": 4.832831972496777,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003206303971972577,
      "loss": 5.2797,
      "mean_token_accuracy": 0.2096318557858467,
      "num_tokens": 10226388.0,
      "step": 5625
    },
    {
      "entropy": 5.772659730911255,
      "epoch": 4.8371293510958315,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003203047737904134,
      "loss": 5.3084,
      "mean_token_accuracy": 0.20517098158597946,
      "num_tokens": 10235333.0,
      "step": 5630
    },
    {
      "entropy": 5.863222885131836,
      "epoch": 4.841426729694886,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031997905145605135,
      "loss": 5.3488,
      "mean_token_accuracy": 0.19873626977205278,
      "num_tokens": 10243985.0,
      "step": 5635
    },
    {
      "entropy": 5.8777539253234865,
      "epoch": 4.845724108293941,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031965323090541874,
      "loss": 5.3422,
      "mean_token_accuracy": 0.19388414025306702,
      "num_tokens": 10252968.0,
      "step": 5640
    },
    {
      "entropy": 5.848870658874512,
      "epoch": 4.850021486892995,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003193273128499777,
      "loss": 5.2251,
      "mean_token_accuracy": 0.2032216265797615,
      "num_tokens": 10261890.0,
      "step": 5645
    },
    {
      "entropy": 5.78770604133606,
      "epoch": 4.85431886549205,
      "grad_norm": 1.125,
      "learning_rate": 0.00031900129800140287,
      "loss": 5.3196,
      "mean_token_accuracy": 0.19879245907068252,
      "num_tokens": 10271363.0,
      "step": 5650
    },
    {
      "entropy": 5.808873653411865,
      "epoch": 4.858616244091104,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031867518707158027,
      "loss": 5.3474,
      "mean_token_accuracy": 0.19671654999256133,
      "num_tokens": 10280608.0,
      "step": 5655
    },
    {
      "entropy": 5.815101957321167,
      "epoch": 4.862913622690159,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000318348980772606,
      "loss": 5.2815,
      "mean_token_accuracy": 0.20676587522029877,
      "num_tokens": 10289972.0,
      "step": 5660
    },
    {
      "entropy": 5.911508226394654,
      "epoch": 4.867211001289213,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031802267981678414,
      "loss": 5.3421,
      "mean_token_accuracy": 0.20263206511735915,
      "num_tokens": 10298740.0,
      "step": 5665
    },
    {
      "entropy": 5.834376287460327,
      "epoch": 4.871508379888268,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031769628491662563,
      "loss": 5.3047,
      "mean_token_accuracy": 0.20320590883493422,
      "num_tokens": 10307706.0,
      "step": 5670
    },
    {
      "entropy": 5.801552486419678,
      "epoch": 4.8758057584873224,
      "grad_norm": 1.015625,
      "learning_rate": 0.00031736979678484634,
      "loss": 5.3467,
      "mean_token_accuracy": 0.2049715057015419,
      "num_tokens": 10317549.0,
      "step": 5675
    },
    {
      "entropy": 5.790601015090942,
      "epoch": 4.880103137086377,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031704321613436597,
      "loss": 5.3813,
      "mean_token_accuracy": 0.19461576640605927,
      "num_tokens": 10327681.0,
      "step": 5680
    },
    {
      "entropy": 5.770719242095947,
      "epoch": 4.8844005156854315,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003167165436783061,
      "loss": 5.3222,
      "mean_token_accuracy": 0.2054004117846489,
      "num_tokens": 10336261.0,
      "step": 5685
    },
    {
      "entropy": 5.797205829620362,
      "epoch": 4.888697894284486,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031638978012998875,
      "loss": 5.2221,
      "mean_token_accuracy": 0.21439259350299836,
      "num_tokens": 10344770.0,
      "step": 5690
    },
    {
      "entropy": 5.86727466583252,
      "epoch": 4.892995272883541,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000316062926202935,
      "loss": 5.384,
      "mean_token_accuracy": 0.19244286566972732,
      "num_tokens": 10354246.0,
      "step": 5695
    },
    {
      "entropy": 5.8602556705474855,
      "epoch": 4.897292651482596,
      "grad_norm": 1.25,
      "learning_rate": 0.0003157359826108632,
      "loss": 5.2944,
      "mean_token_accuracy": 0.20203630179166793,
      "num_tokens": 10362693.0,
      "step": 5700
    },
    {
      "entropy": 5.8388330936431885,
      "epoch": 4.90159003008165,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031540895006768727,
      "loss": 5.3099,
      "mean_token_accuracy": 0.20483107417821883,
      "num_tokens": 10371639.0,
      "step": 5705
    },
    {
      "entropy": 5.827498531341552,
      "epoch": 4.905887408680705,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003150818292875158,
      "loss": 5.3257,
      "mean_token_accuracy": 0.20185545533895494,
      "num_tokens": 10381237.0,
      "step": 5710
    },
    {
      "entropy": 5.847358322143554,
      "epoch": 4.91018478727976,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003147546209846497,
      "loss": 5.2955,
      "mean_token_accuracy": 0.20283959358930587,
      "num_tokens": 10389932.0,
      "step": 5715
    },
    {
      "entropy": 5.739686775207519,
      "epoch": 4.914482165878814,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003144273258735812,
      "loss": 5.1855,
      "mean_token_accuracy": 0.20870466232299806,
      "num_tokens": 10398938.0,
      "step": 5720
    },
    {
      "entropy": 5.784277296066284,
      "epoch": 4.918779544477869,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003140999446689919,
      "loss": 5.295,
      "mean_token_accuracy": 0.2015138253569603,
      "num_tokens": 10407980.0,
      "step": 5725
    },
    {
      "entropy": 5.807836294174194,
      "epoch": 4.923076923076923,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003137724780857516,
      "loss": 5.3637,
      "mean_token_accuracy": 0.20311774164438248,
      "num_tokens": 10416990.0,
      "step": 5730
    },
    {
      "entropy": 5.88627667427063,
      "epoch": 4.927374301675978,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031344492683891634,
      "loss": 5.3789,
      "mean_token_accuracy": 0.1995622456073761,
      "num_tokens": 10426573.0,
      "step": 5735
    },
    {
      "entropy": 5.865926218032837,
      "epoch": 4.931671680275032,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003131172916437272,
      "loss": 5.3262,
      "mean_token_accuracy": 0.20199921876192092,
      "num_tokens": 10435162.0,
      "step": 5740
    },
    {
      "entropy": 5.825812768936157,
      "epoch": 4.935969058874087,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031278957321560845,
      "loss": 5.355,
      "mean_token_accuracy": 0.20129600614309312,
      "num_tokens": 10444374.0,
      "step": 5745
    },
    {
      "entropy": 5.926651048660278,
      "epoch": 4.9402664374731415,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031246177227016615,
      "loss": 5.3684,
      "mean_token_accuracy": 0.19039739072322845,
      "num_tokens": 10452679.0,
      "step": 5750
    },
    {
      "entropy": 5.812946605682373,
      "epoch": 4.944563816072196,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031213388952318653,
      "loss": 5.3155,
      "mean_token_accuracy": 0.20289614647626877,
      "num_tokens": 10461801.0,
      "step": 5755
    },
    {
      "entropy": 5.817221832275391,
      "epoch": 4.948861194671251,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003118059256906345,
      "loss": 5.3004,
      "mean_token_accuracy": 0.1964358150959015,
      "num_tokens": 10471176.0,
      "step": 5760
    },
    {
      "entropy": 5.867912340164184,
      "epoch": 4.953158573270305,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031147788148865204,
      "loss": 5.356,
      "mean_token_accuracy": 0.19638251066207885,
      "num_tokens": 10480403.0,
      "step": 5765
    },
    {
      "entropy": 5.85248556137085,
      "epoch": 4.95745595186936,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003111497576335564,
      "loss": 5.2968,
      "mean_token_accuracy": 0.20374597758054733,
      "num_tokens": 10489574.0,
      "step": 5770
    },
    {
      "entropy": 5.826297807693481,
      "epoch": 4.961753330468414,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003108215548418391,
      "loss": 5.3081,
      "mean_token_accuracy": 0.2055655375123024,
      "num_tokens": 10499631.0,
      "step": 5775
    },
    {
      "entropy": 5.815798139572143,
      "epoch": 4.966050709067469,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003104932738301637,
      "loss": 5.2897,
      "mean_token_accuracy": 0.20031124353408813,
      "num_tokens": 10508128.0,
      "step": 5780
    },
    {
      "entropy": 5.799650478363037,
      "epoch": 4.970348087666523,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031016491531536477,
      "loss": 5.2696,
      "mean_token_accuracy": 0.20488461554050447,
      "num_tokens": 10517544.0,
      "step": 5785
    },
    {
      "entropy": 5.78414101600647,
      "epoch": 4.974645466265578,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003098364800144462,
      "loss": 5.3256,
      "mean_token_accuracy": 0.2010160431265831,
      "num_tokens": 10526244.0,
      "step": 5790
    },
    {
      "entropy": 5.8842973709106445,
      "epoch": 4.9789428448646325,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003095079686445792,
      "loss": 5.3975,
      "mean_token_accuracy": 0.20257108062505721,
      "num_tokens": 10535887.0,
      "step": 5795
    },
    {
      "entropy": 5.848398447036743,
      "epoch": 4.983240223463687,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030917938192310146,
      "loss": 5.2497,
      "mean_token_accuracy": 0.20553108304738998,
      "num_tokens": 10544420.0,
      "step": 5800
    },
    {
      "entropy": 5.8361001968383786,
      "epoch": 4.9875376020627415,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030885072056751494,
      "loss": 5.3212,
      "mean_token_accuracy": 0.19935197979211808,
      "num_tokens": 10553114.0,
      "step": 5805
    },
    {
      "entropy": 5.785122728347778,
      "epoch": 4.991834980661796,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030852198529548476,
      "loss": 5.3645,
      "mean_token_accuracy": 0.2018916442990303,
      "num_tokens": 10562272.0,
      "step": 5810
    },
    {
      "entropy": 5.8573511123657225,
      "epoch": 4.996132359260851,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003081931768248373,
      "loss": 5.2916,
      "mean_token_accuracy": 0.20492337495088578,
      "num_tokens": 10571757.0,
      "step": 5815
    },
    {
      "entropy": 5.811619228786892,
      "epoch": 5.0,
      "grad_norm": 1.75,
      "learning_rate": 0.0003078642958735588,
      "loss": 5.2701,
      "mean_token_accuracy": 0.21092700295978123,
      "num_tokens": 10579660.0,
      "step": 5820
    },
    {
      "entropy": 5.830132055282593,
      "epoch": 5.0042973785990545,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030753534315979393,
      "loss": 5.1523,
      "mean_token_accuracy": 0.21071099638938903,
      "num_tokens": 10589139.0,
      "step": 5825
    },
    {
      "entropy": 5.826620626449585,
      "epoch": 5.008594757198109,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003072063194018438,
      "loss": 4.9691,
      "mean_token_accuracy": 0.22948768883943557,
      "num_tokens": 10597915.0,
      "step": 5830
    },
    {
      "entropy": 5.834732007980347,
      "epoch": 5.012892135797164,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003068772253181648,
      "loss": 5.1533,
      "mean_token_accuracy": 0.208507040143013,
      "num_tokens": 10606491.0,
      "step": 5835
    },
    {
      "entropy": 5.867878246307373,
      "epoch": 5.017189514396218,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003065480616273671,
      "loss": 5.1404,
      "mean_token_accuracy": 0.20878320187330246,
      "num_tokens": 10615852.0,
      "step": 5840
    },
    {
      "entropy": 5.929982137680054,
      "epoch": 5.021486892995273,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003062188290482123,
      "loss": 5.1237,
      "mean_token_accuracy": 0.21033858209848405,
      "num_tokens": 10625442.0,
      "step": 5845
    },
    {
      "entropy": 5.79777193069458,
      "epoch": 5.025784271594327,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00030588952829961304,
      "loss": 5.0442,
      "mean_token_accuracy": 0.22793061137199402,
      "num_tokens": 10634972.0,
      "step": 5850
    },
    {
      "entropy": 5.818366861343383,
      "epoch": 5.030081650193382,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003055601601006303,
      "loss": 5.0398,
      "mean_token_accuracy": 0.21814830154180526,
      "num_tokens": 10644487.0,
      "step": 5855
    },
    {
      "entropy": 5.821985292434692,
      "epoch": 5.034379028792436,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003052307251704728,
      "loss": 5.0798,
      "mean_token_accuracy": 0.22706313729286193,
      "num_tokens": 10654144.0,
      "step": 5860
    },
    {
      "entropy": 5.770719528198242,
      "epoch": 5.038676407391491,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003049012242284946,
      "loss": 5.1071,
      "mean_token_accuracy": 0.21176830530166627,
      "num_tokens": 10663023.0,
      "step": 5865
    },
    {
      "entropy": 5.842636013031006,
      "epoch": 5.0429737859905455,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003045716579941941,
      "loss": 5.1561,
      "mean_token_accuracy": 0.21283052414655684,
      "num_tokens": 10672001.0,
      "step": 5870
    },
    {
      "entropy": 5.800144195556641,
      "epoch": 5.0472711645896,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00030424202718721215,
      "loss": 5.1079,
      "mean_token_accuracy": 0.2135323241353035,
      "num_tokens": 10682654.0,
      "step": 5875
    },
    {
      "entropy": 5.809390497207642,
      "epoch": 5.051568543188655,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030391233252733085,
      "loss": 5.1159,
      "mean_token_accuracy": 0.21603565365076066,
      "num_tokens": 10691429.0,
      "step": 5880
    },
    {
      "entropy": 5.7686162948608395,
      "epoch": 5.055865921787709,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030358257473447144,
      "loss": 5.0934,
      "mean_token_accuracy": 0.21573030948638916,
      "num_tokens": 10701130.0,
      "step": 5885
    },
    {
      "entropy": 5.783536434173584,
      "epoch": 5.060163300386764,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030325275452869316,
      "loss": 5.0587,
      "mean_token_accuracy": 0.21876554638147355,
      "num_tokens": 10709779.0,
      "step": 5890
    },
    {
      "entropy": 5.72818398475647,
      "epoch": 5.064460678985819,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030292287263019153,
      "loss": 5.1101,
      "mean_token_accuracy": 0.22073507010936738,
      "num_tokens": 10718795.0,
      "step": 5895
    },
    {
      "entropy": 5.687624406814575,
      "epoch": 5.068758057584874,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030259292975929675,
      "loss": 5.0109,
      "mean_token_accuracy": 0.22956797927618028,
      "num_tokens": 10728202.0,
      "step": 5900
    },
    {
      "entropy": 5.79244966506958,
      "epoch": 5.073055436183928,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003022629266364723,
      "loss": 5.0609,
      "mean_token_accuracy": 0.21821176558732985,
      "num_tokens": 10737050.0,
      "step": 5905
    },
    {
      "entropy": 5.826407146453858,
      "epoch": 5.077352814782983,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030193286398231276,
      "loss": 5.0761,
      "mean_token_accuracy": 0.22239672094583512,
      "num_tokens": 10745261.0,
      "step": 5910
    },
    {
      "entropy": 5.807696723937989,
      "epoch": 5.081650193382037,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030160274251754337,
      "loss": 5.1337,
      "mean_token_accuracy": 0.21703209429979325,
      "num_tokens": 10755008.0,
      "step": 5915
    },
    {
      "entropy": 5.786716270446777,
      "epoch": 5.085947571981092,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030127256296301724,
      "loss": 5.148,
      "mean_token_accuracy": 0.20591583997011184,
      "num_tokens": 10763951.0,
      "step": 5920
    },
    {
      "entropy": 5.782669925689698,
      "epoch": 5.090244950580146,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003009423260397148,
      "loss": 5.0689,
      "mean_token_accuracy": 0.2212012901902199,
      "num_tokens": 10772770.0,
      "step": 5925
    },
    {
      "entropy": 5.8064656257629395,
      "epoch": 5.094542329179201,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030061203246874125,
      "loss": 5.1618,
      "mean_token_accuracy": 0.20938030183315276,
      "num_tokens": 10781827.0,
      "step": 5930
    },
    {
      "entropy": 5.82456750869751,
      "epoch": 5.0988397077782555,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030028168297132593,
      "loss": 5.2033,
      "mean_token_accuracy": 0.2137736588716507,
      "num_tokens": 10792321.0,
      "step": 5935
    },
    {
      "entropy": 5.780242776870727,
      "epoch": 5.10313708637731,
      "grad_norm": 1.25,
      "learning_rate": 0.0002999512782688199,
      "loss": 5.1682,
      "mean_token_accuracy": 0.2179598942399025,
      "num_tokens": 10801689.0,
      "step": 5940
    },
    {
      "entropy": 5.819448375701905,
      "epoch": 5.1074344649763646,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002996208190826951,
      "loss": 5.1073,
      "mean_token_accuracy": 0.21035155802965164,
      "num_tokens": 10810513.0,
      "step": 5945
    },
    {
      "entropy": 5.8022904872894285,
      "epoch": 5.111731843575419,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029929030613454227,
      "loss": 5.0734,
      "mean_token_accuracy": 0.2216214805841446,
      "num_tokens": 10819581.0,
      "step": 5950
    },
    {
      "entropy": 5.73983941078186,
      "epoch": 5.116029222174474,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002989597401460697,
      "loss": 5.0576,
      "mean_token_accuracy": 0.2205493226647377,
      "num_tokens": 10828139.0,
      "step": 5955
    },
    {
      "entropy": 5.789935064315796,
      "epoch": 5.120326600773528,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029862912183910105,
      "loss": 5.0828,
      "mean_token_accuracy": 0.2225214958190918,
      "num_tokens": 10836256.0,
      "step": 5960
    },
    {
      "entropy": 5.782462453842163,
      "epoch": 5.124623979372583,
      "grad_norm": 1.125,
      "learning_rate": 0.00029829845193557496,
      "loss": 5.1055,
      "mean_token_accuracy": 0.2224409133195877,
      "num_tokens": 10846255.0,
      "step": 5965
    },
    {
      "entropy": 5.778302240371704,
      "epoch": 5.128921357971637,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002979677311575421,
      "loss": 5.0935,
      "mean_token_accuracy": 0.22012680172920226,
      "num_tokens": 10855546.0,
      "step": 5970
    },
    {
      "entropy": 5.829771327972412,
      "epoch": 5.133218736570692,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002976369602271646,
      "loss": 5.1676,
      "mean_token_accuracy": 0.2150162413716316,
      "num_tokens": 10864417.0,
      "step": 5975
    },
    {
      "entropy": 5.834624910354615,
      "epoch": 5.137516115169746,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002973061398667138,
      "loss": 5.1343,
      "mean_token_accuracy": 0.21512973457574844,
      "num_tokens": 10874527.0,
      "step": 5980
    },
    {
      "entropy": 5.797853422164917,
      "epoch": 5.141813493768801,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029697527079856916,
      "loss": 5.1681,
      "mean_token_accuracy": 0.20807291567325592,
      "num_tokens": 10883486.0,
      "step": 5985
    },
    {
      "entropy": 5.7741375923156735,
      "epoch": 5.1461108723678555,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029664435374521665,
      "loss": 5.045,
      "mean_token_accuracy": 0.22194089591503144,
      "num_tokens": 10891972.0,
      "step": 5990
    },
    {
      "entropy": 5.79967098236084,
      "epoch": 5.15040825096691,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029631338942924664,
      "loss": 5.0673,
      "mean_token_accuracy": 0.21982951015233992,
      "num_tokens": 10901350.0,
      "step": 5995
    },
    {
      "entropy": 5.787040090560913,
      "epoch": 5.154705629565965,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002959823785733531,
      "loss": 5.0619,
      "mean_token_accuracy": 0.21710459440946578,
      "num_tokens": 10910114.0,
      "step": 6000
    },
    {
      "epoch": 5.154705629565965,
      "eval_entropy": 5.57793817434225,
      "eval_loss": 5.929503917694092,
      "eval_mean_token_accuracy": 0.178135054085303,
      "eval_num_tokens": 10910114.0,
      "eval_runtime": 2.0603,
      "eval_samples_per_second": 1722.584,
      "eval_steps_per_second": 215.505,
      "step": 6000
    },
    {
      "entropy": 5.742149400711059,
      "epoch": 5.159003008165019,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002956513219003312,
      "loss": 5.1367,
      "mean_token_accuracy": 0.21954900473356248,
      "num_tokens": 10919781.0,
      "step": 6005
    },
    {
      "entropy": 5.783298921585083,
      "epoch": 5.163300386764074,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029532022013307666,
      "loss": 5.1365,
      "mean_token_accuracy": 0.21482574939727783,
      "num_tokens": 10929561.0,
      "step": 6010
    },
    {
      "entropy": 5.795925092697144,
      "epoch": 5.167597765363128,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00029498907399458325,
      "loss": 5.0954,
      "mean_token_accuracy": 0.21783551722764968,
      "num_tokens": 10939123.0,
      "step": 6015
    },
    {
      "entropy": 5.840067148208618,
      "epoch": 5.171895143962183,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002946578842079418,
      "loss": 5.1365,
      "mean_token_accuracy": 0.2198743224143982,
      "num_tokens": 10947990.0,
      "step": 6020
    },
    {
      "entropy": 5.807727861404419,
      "epoch": 5.176192522561237,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002943266514963384,
      "loss": 5.0567,
      "mean_token_accuracy": 0.23506541252136232,
      "num_tokens": 10956569.0,
      "step": 6025
    },
    {
      "entropy": 5.747990560531616,
      "epoch": 5.180489901160292,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002939953765830529,
      "loss": 5.1173,
      "mean_token_accuracy": 0.21218268424272538,
      "num_tokens": 10965466.0,
      "step": 6030
    },
    {
      "entropy": 5.7549018383026125,
      "epoch": 5.184787279759346,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029366406019145735,
      "loss": 5.1244,
      "mean_token_accuracy": 0.21184627264738082,
      "num_tokens": 10975051.0,
      "step": 6035
    },
    {
      "entropy": 5.724986934661866,
      "epoch": 5.189084658358402,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002933327030450143,
      "loss": 5.0276,
      "mean_token_accuracy": 0.22142817080020905,
      "num_tokens": 10983940.0,
      "step": 6040
    },
    {
      "entropy": 5.849178886413574,
      "epoch": 5.193382036957456,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029300130586727545,
      "loss": 5.1395,
      "mean_token_accuracy": 0.2128952518105507,
      "num_tokens": 10994029.0,
      "step": 6045
    },
    {
      "entropy": 5.825051498413086,
      "epoch": 5.197679415556511,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029266986938187943,
      "loss": 5.1831,
      "mean_token_accuracy": 0.20702893286943436,
      "num_tokens": 11003616.0,
      "step": 6050
    },
    {
      "entropy": 5.735121870040894,
      "epoch": 5.2019767941555655,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002923383943125514,
      "loss": 5.0587,
      "mean_token_accuracy": 0.21676131337881088,
      "num_tokens": 11012068.0,
      "step": 6055
    },
    {
      "entropy": 5.760170888900757,
      "epoch": 5.20627417275462,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002920068813831002,
      "loss": 5.1019,
      "mean_token_accuracy": 0.21917937994003295,
      "num_tokens": 11020510.0,
      "step": 6060
    },
    {
      "entropy": 5.757324075698852,
      "epoch": 5.210571551353675,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002916753313174178,
      "loss": 5.144,
      "mean_token_accuracy": 0.21361536532640457,
      "num_tokens": 11029804.0,
      "step": 6065
    },
    {
      "entropy": 5.8071925163269045,
      "epoch": 5.214868929952729,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002913437448394768,
      "loss": 5.1207,
      "mean_token_accuracy": 0.22017257511615754,
      "num_tokens": 11038586.0,
      "step": 6070
    },
    {
      "entropy": 5.873847913742066,
      "epoch": 5.219166308551784,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029101212267332955,
      "loss": 5.1813,
      "mean_token_accuracy": 0.21376842260360718,
      "num_tokens": 11048240.0,
      "step": 6075
    },
    {
      "entropy": 5.778180503845215,
      "epoch": 5.223463687150838,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00029068046554310637,
      "loss": 5.0754,
      "mean_token_accuracy": 0.21744104623794555,
      "num_tokens": 11056703.0,
      "step": 6080
    },
    {
      "entropy": 5.738321208953858,
      "epoch": 5.227761065749893,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002903487741730139,
      "loss": 5.1102,
      "mean_token_accuracy": 0.21128816306591033,
      "num_tokens": 11066246.0,
      "step": 6085
    },
    {
      "entropy": 5.8017333984375,
      "epoch": 5.232058444348947,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029001704928733354,
      "loss": 5.1573,
      "mean_token_accuracy": 0.2133406952023506,
      "num_tokens": 11075277.0,
      "step": 6090
    },
    {
      "entropy": 5.7818268775939945,
      "epoch": 5.236355822948002,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002896852916104198,
      "loss": 5.0464,
      "mean_token_accuracy": 0.22696022540330887,
      "num_tokens": 11083759.0,
      "step": 6095
    },
    {
      "entropy": 5.810663747787475,
      "epoch": 5.240653201547056,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002893535018666988,
      "loss": 5.0293,
      "mean_token_accuracy": 0.22478870898485184,
      "num_tokens": 11091960.0,
      "step": 6100
    },
    {
      "entropy": 5.7300464630126955,
      "epoch": 5.244950580146111,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028902168078066674,
      "loss": 5.1507,
      "mean_token_accuracy": 0.20951681435108185,
      "num_tokens": 11101134.0,
      "step": 6105
    },
    {
      "entropy": 5.7556839942932125,
      "epoch": 5.2492479587451655,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002886898290768883,
      "loss": 5.0882,
      "mean_token_accuracy": 0.21937238723039626,
      "num_tokens": 11110282.0,
      "step": 6110
    },
    {
      "entropy": 5.815136528015136,
      "epoch": 5.25354533734422,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000288357947479995,
      "loss": 5.1771,
      "mean_token_accuracy": 0.21402793973684311,
      "num_tokens": 11119591.0,
      "step": 6115
    },
    {
      "entropy": 5.836182308197022,
      "epoch": 5.257842715943275,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028802603671468347,
      "loss": 5.1534,
      "mean_token_accuracy": 0.21754036992788314,
      "num_tokens": 11129164.0,
      "step": 6120
    },
    {
      "entropy": 5.809143018722534,
      "epoch": 5.262140094542329,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028769409750571413,
      "loss": 5.1374,
      "mean_token_accuracy": 0.21217462569475173,
      "num_tokens": 11137973.0,
      "step": 6125
    },
    {
      "entropy": 5.764298391342163,
      "epoch": 5.266437473141384,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028736213057790975,
      "loss": 5.19,
      "mean_token_accuracy": 0.21339703351259232,
      "num_tokens": 11147285.0,
      "step": 6130
    },
    {
      "entropy": 5.8401704788208,
      "epoch": 5.270734851740438,
      "grad_norm": 1.25,
      "learning_rate": 0.0002870301366561533,
      "loss": 5.1595,
      "mean_token_accuracy": 0.2137187123298645,
      "num_tokens": 11155303.0,
      "step": 6135
    },
    {
      "entropy": 5.786964178085327,
      "epoch": 5.275032230339493,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002866981164653867,
      "loss": 5.0347,
      "mean_token_accuracy": 0.22384295612573624,
      "num_tokens": 11163553.0,
      "step": 6140
    },
    {
      "entropy": 5.697514867782592,
      "epoch": 5.279329608938547,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002863660707306095,
      "loss": 4.9706,
      "mean_token_accuracy": 0.23047677278518677,
      "num_tokens": 11171865.0,
      "step": 6145
    },
    {
      "entropy": 5.7564788341522215,
      "epoch": 5.283626987537602,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028603400017687675,
      "loss": 5.1556,
      "mean_token_accuracy": 0.21275610327720643,
      "num_tokens": 11181137.0,
      "step": 6150
    },
    {
      "entropy": 5.6765399932861325,
      "epoch": 5.287924366136656,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028570190552929794,
      "loss": 5.0212,
      "mean_token_accuracy": 0.2336950942873955,
      "num_tokens": 11190174.0,
      "step": 6155
    },
    {
      "entropy": 5.736397409439087,
      "epoch": 5.292221744735711,
      "grad_norm": 1.28125,
      "learning_rate": 0.000285369787513035,
      "loss": 5.054,
      "mean_token_accuracy": 0.22187479585409164,
      "num_tokens": 11197964.0,
      "step": 6160
    },
    {
      "entropy": 5.84683780670166,
      "epoch": 5.2965191233347655,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00028503764685330077,
      "loss": 5.1708,
      "mean_token_accuracy": 0.21154396533966063,
      "num_tokens": 11207974.0,
      "step": 6165
    },
    {
      "entropy": 5.7679667472839355,
      "epoch": 5.30081650193382,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028470548427535794,
      "loss": 5.1142,
      "mean_token_accuracy": 0.21926384270191193,
      "num_tokens": 11216430.0,
      "step": 6170
    },
    {
      "entropy": 5.803475618362427,
      "epoch": 5.305113880532875,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028437330050451654,
      "loss": 5.1931,
      "mean_token_accuracy": 0.2082819387316704,
      "num_tokens": 11226189.0,
      "step": 6175
    },
    {
      "entropy": 5.782622385025024,
      "epoch": 5.309411259131929,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002840410962661334,
      "loss": 5.157,
      "mean_token_accuracy": 0.21044335067272185,
      "num_tokens": 11234691.0,
      "step": 6180
    },
    {
      "entropy": 5.708439731597901,
      "epoch": 5.313708637730985,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002837088722856098,
      "loss": 4.9886,
      "mean_token_accuracy": 0.22940880805253983,
      "num_tokens": 11243852.0,
      "step": 6185
    },
    {
      "entropy": 5.773501396179199,
      "epoch": 5.318006016330038,
      "grad_norm": 1.125,
      "learning_rate": 0.00028337662928838996,
      "loss": 5.1732,
      "mean_token_accuracy": 0.2154297411441803,
      "num_tokens": 11253416.0,
      "step": 6190
    },
    {
      "entropy": 5.738881921768188,
      "epoch": 5.322303394929094,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028304436799995986,
      "loss": 5.0464,
      "mean_token_accuracy": 0.23141929656267166,
      "num_tokens": 11262869.0,
      "step": 6195
    },
    {
      "entropy": 5.817911863327026,
      "epoch": 5.326600773528148,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028271208914584534,
      "loss": 5.2057,
      "mean_token_accuracy": 0.20787570029497146,
      "num_tokens": 11272386.0,
      "step": 6200
    },
    {
      "entropy": 5.823677492141724,
      "epoch": 5.330898152127203,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028237979345161065,
      "loss": 5.0555,
      "mean_token_accuracy": 0.22227236330509187,
      "num_tokens": 11281590.0,
      "step": 6205
    },
    {
      "entropy": 5.747840785980225,
      "epoch": 5.335195530726257,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002820474816428568,
      "loss": 5.1006,
      "mean_token_accuracy": 0.2157945975661278,
      "num_tokens": 11290873.0,
      "step": 6210
    },
    {
      "entropy": 5.796749877929687,
      "epoch": 5.339492909325312,
      "grad_norm": 1.25,
      "learning_rate": 0.0002817151544452198,
      "loss": 5.0784,
      "mean_token_accuracy": 0.21691201776266097,
      "num_tokens": 11299064.0,
      "step": 6215
    },
    {
      "entropy": 5.7705799579620365,
      "epoch": 5.343790287924366,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028138281258436947,
      "loss": 5.114,
      "mean_token_accuracy": 0.21452804952859877,
      "num_tokens": 11307390.0,
      "step": 6220
    },
    {
      "entropy": 5.725601530075073,
      "epoch": 5.348087666523421,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002810504567860078,
      "loss": 5.0575,
      "mean_token_accuracy": 0.22290477752685547,
      "num_tokens": 11315606.0,
      "step": 6225
    },
    {
      "entropy": 5.850395154953003,
      "epoch": 5.3523850451224755,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002807180877758667,
      "loss": 5.1794,
      "mean_token_accuracy": 0.20834580212831497,
      "num_tokens": 11323821.0,
      "step": 6230
    },
    {
      "entropy": 5.775689506530762,
      "epoch": 5.35668242372153,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028038570627970754,
      "loss": 5.1194,
      "mean_token_accuracy": 0.21776863038539887,
      "num_tokens": 11331850.0,
      "step": 6235
    },
    {
      "entropy": 5.717955589294434,
      "epoch": 5.360979802320585,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002800533130233184,
      "loss": 5.0739,
      "mean_token_accuracy": 0.21858684420585633,
      "num_tokens": 11340125.0,
      "step": 6240
    },
    {
      "entropy": 5.789701652526856,
      "epoch": 5.365277180919639,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002797209087325135,
      "loss": 5.144,
      "mean_token_accuracy": 0.2092101663351059,
      "num_tokens": 11349184.0,
      "step": 6245
    },
    {
      "entropy": 5.839946126937866,
      "epoch": 5.369574559518694,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027938849413313083,
      "loss": 5.1887,
      "mean_token_accuracy": 0.20530815720558165,
      "num_tokens": 11357536.0,
      "step": 6250
    },
    {
      "entropy": 5.7973836898803714,
      "epoch": 5.373871938117748,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000279056069951031,
      "loss": 5.1359,
      "mean_token_accuracy": 0.21271687299013137,
      "num_tokens": 11367242.0,
      "step": 6255
    },
    {
      "entropy": 5.7870954990386965,
      "epoch": 5.378169316716803,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027872363691209564,
      "loss": 5.1234,
      "mean_token_accuracy": 0.2152712032198906,
      "num_tokens": 11374932.0,
      "step": 6260
    },
    {
      "entropy": 5.753720331192016,
      "epoch": 5.382466695315857,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002783911957422256,
      "loss": 5.1096,
      "mean_token_accuracy": 0.21259453445672988,
      "num_tokens": 11383575.0,
      "step": 6265
    },
    {
      "entropy": 5.824054479598999,
      "epoch": 5.386764073914912,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002780587471673394,
      "loss": 5.1515,
      "mean_token_accuracy": 0.21363489627838134,
      "num_tokens": 11392285.0,
      "step": 6270
    },
    {
      "entropy": 5.766599225997925,
      "epoch": 5.391061452513966,
      "grad_norm": 1.140625,
      "learning_rate": 0.00027772629191337206,
      "loss": 5.1163,
      "mean_token_accuracy": 0.22057569175958633,
      "num_tokens": 11401054.0,
      "step": 6275
    },
    {
      "entropy": 5.738688850402832,
      "epoch": 5.395358831113021,
      "grad_norm": 1.265625,
      "learning_rate": 0.00027739383070627283,
      "loss": 5.1316,
      "mean_token_accuracy": 0.21595446169376373,
      "num_tokens": 11410529.0,
      "step": 6280
    },
    {
      "entropy": 5.75838303565979,
      "epoch": 5.3996562097120755,
      "grad_norm": 1.125,
      "learning_rate": 0.0002770613642720041,
      "loss": 5.0966,
      "mean_token_accuracy": 0.21665699034929276,
      "num_tokens": 11419961.0,
      "step": 6285
    },
    {
      "entropy": 5.838625240325928,
      "epoch": 5.40395358831113,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00027672889333653984,
      "loss": 5.228,
      "mean_token_accuracy": 0.20249445587396622,
      "num_tokens": 11429529.0,
      "step": 6290
    },
    {
      "entropy": 5.79939227104187,
      "epoch": 5.408250966910185,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002763964186258635,
      "loss": 5.0924,
      "mean_token_accuracy": 0.220710127055645,
      "num_tokens": 11438254.0,
      "step": 6295
    },
    {
      "entropy": 5.827362585067749,
      "epoch": 5.412548345509239,
      "grad_norm": 1.125,
      "learning_rate": 0.0002760639408659671,
      "loss": 5.1714,
      "mean_token_accuracy": 0.20810000002384185,
      "num_tokens": 11447587.0,
      "step": 6300
    },
    {
      "entropy": 5.7800661563873295,
      "epoch": 5.416845724108294,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002757314607828489,
      "loss": 5.1167,
      "mean_token_accuracy": 0.21862143725156785,
      "num_tokens": 11455493.0,
      "step": 6305
    },
    {
      "entropy": 5.73545651435852,
      "epoch": 5.421143102707348,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027539897910251293,
      "loss": 5.079,
      "mean_token_accuracy": 0.22295478582382203,
      "num_tokens": 11464143.0,
      "step": 6310
    },
    {
      "entropy": 5.764443969726562,
      "epoch": 5.425440481306403,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027506649655096595,
      "loss": 5.0322,
      "mean_token_accuracy": 0.21607312858104705,
      "num_tokens": 11471813.0,
      "step": 6315
    },
    {
      "entropy": 5.766522550582886,
      "epoch": 5.429737859905457,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002747340138542171,
      "loss": 5.082,
      "mean_token_accuracy": 0.211651112139225,
      "num_tokens": 11481374.0,
      "step": 6320
    },
    {
      "entropy": 5.8527501106262205,
      "epoch": 5.434035238504512,
      "grad_norm": 1.125,
      "learning_rate": 0.0002744015317382757,
      "loss": 5.2407,
      "mean_token_accuracy": 0.20978412181138992,
      "num_tokens": 11490575.0,
      "step": 6325
    },
    {
      "entropy": 5.799367380142212,
      "epoch": 5.438332617103566,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002740690509291498,
      "loss": 5.1368,
      "mean_token_accuracy": 0.21918633878231047,
      "num_tokens": 11499898.0,
      "step": 6330
    },
    {
      "entropy": 5.785244274139404,
      "epoch": 5.442629995702621,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002737365721528445,
      "loss": 5.1236,
      "mean_token_accuracy": 0.21274089068174362,
      "num_tokens": 11508544.0,
      "step": 6335
    },
    {
      "entropy": 5.7810187339782715,
      "epoch": 5.446927374301676,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002734040961353607,
      "loss": 5.1159,
      "mean_token_accuracy": 0.21834465563297273,
      "num_tokens": 11519239.0,
      "step": 6340
    },
    {
      "entropy": 5.774866771697998,
      "epoch": 5.451224752900731,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000273071623602693,
      "loss": 5.0848,
      "mean_token_accuracy": 0.22297502309083939,
      "num_tokens": 11529014.0,
      "step": 6345
    },
    {
      "entropy": 5.764535140991211,
      "epoch": 5.4555221314997855,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027273915528082865,
      "loss": 5.0866,
      "mean_token_accuracy": 0.21357513070106507,
      "num_tokens": 11538367.0,
      "step": 6350
    },
    {
      "entropy": 5.832653617858886,
      "epoch": 5.45981951009884,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002724066918957455,
      "loss": 5.244,
      "mean_token_accuracy": 0.20616474747657776,
      "num_tokens": 11548166.0,
      "step": 6355
    },
    {
      "entropy": 5.75134687423706,
      "epoch": 5.464116888697895,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002720742341734107,
      "loss": 5.0965,
      "mean_token_accuracy": 0.2268800511956215,
      "num_tokens": 11557187.0,
      "step": 6360
    },
    {
      "entropy": 5.841009187698364,
      "epoch": 5.468414267296949,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027174178283977904,
      "loss": 5.1381,
      "mean_token_accuracy": 0.2110605239868164,
      "num_tokens": 11566181.0,
      "step": 6365
    },
    {
      "entropy": 5.748985385894775,
      "epoch": 5.472711645896004,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027140933862079136,
      "loss": 5.101,
      "mean_token_accuracy": 0.2198827251791954,
      "num_tokens": 11576157.0,
      "step": 6370
    },
    {
      "entropy": 5.729598569869995,
      "epoch": 5.477009024495058,
      "grad_norm": 1.21875,
      "learning_rate": 0.000271076902242373,
      "loss": 5.0683,
      "mean_token_accuracy": 0.21918342411518096,
      "num_tokens": 11585325.0,
      "step": 6375
    },
    {
      "entropy": 5.765842914581299,
      "epoch": 5.481306403094113,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000270744474430432,
      "loss": 5.0643,
      "mean_token_accuracy": 0.22224877923727035,
      "num_tokens": 11594623.0,
      "step": 6380
    },
    {
      "entropy": 5.819550085067749,
      "epoch": 5.485603781693167,
      "grad_norm": 1.1875,
      "learning_rate": 0.000270412055910858,
      "loss": 5.2292,
      "mean_token_accuracy": 0.2084754228591919,
      "num_tokens": 11604370.0,
      "step": 6385
    },
    {
      "entropy": 5.727860975265503,
      "epoch": 5.489901160292222,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002700796474095201,
      "loss": 5.0449,
      "mean_token_accuracy": 0.23166752904653548,
      "num_tokens": 11613779.0,
      "step": 6390
    },
    {
      "entropy": 5.796416425704956,
      "epoch": 5.494198538891276,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002697472496522656,
      "loss": 5.133,
      "mean_token_accuracy": 0.21775419861078263,
      "num_tokens": 11623037.0,
      "step": 6395
    },
    {
      "entropy": 5.824549913406372,
      "epoch": 5.498495917490331,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002694148633649184,
      "loss": 5.1594,
      "mean_token_accuracy": 0.21549758613109588,
      "num_tokens": 11631640.0,
      "step": 6400
    },
    {
      "entropy": 5.772566461563111,
      "epoch": 5.5027932960893855,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002690824892732772,
      "loss": 5.1401,
      "mean_token_accuracy": 0.2165399119257927,
      "num_tokens": 11640500.0,
      "step": 6405
    },
    {
      "entropy": 5.810577011108398,
      "epoch": 5.50709067468844,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002687501281031142,
      "loss": 5.1688,
      "mean_token_accuracy": 0.21251920014619827,
      "num_tokens": 11649173.0,
      "step": 6410
    },
    {
      "entropy": 5.655430603027344,
      "epoch": 5.511388053287495,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002684177805801734,
      "loss": 5.008,
      "mean_token_accuracy": 0.2310524970293045,
      "num_tokens": 11658808.0,
      "step": 6415
    },
    {
      "entropy": 5.803665399551392,
      "epoch": 5.515685431886549,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026808544743016886,
      "loss": 5.1174,
      "mean_token_accuracy": 0.21396234184503554,
      "num_tokens": 11667600.0,
      "step": 6420
    },
    {
      "entropy": 5.751873636245728,
      "epoch": 5.519982810485604,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002677531293787835,
      "loss": 5.1214,
      "mean_token_accuracy": 0.21226371973752975,
      "num_tokens": 11675597.0,
      "step": 6425
    },
    {
      "entropy": 5.676282644271851,
      "epoch": 5.524280189084658,
      "grad_norm": 1.15625,
      "learning_rate": 0.000267420827151667,
      "loss": 5.0496,
      "mean_token_accuracy": 0.22589787393808364,
      "num_tokens": 11684549.0,
      "step": 6430
    },
    {
      "entropy": 5.773661279678345,
      "epoch": 5.528577567683713,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002670885414744347,
      "loss": 5.1328,
      "mean_token_accuracy": 0.22293749451637268,
      "num_tokens": 11693043.0,
      "step": 6435
    },
    {
      "entropy": 5.885936498641968,
      "epoch": 5.532874946282767,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002667562730726655,
      "loss": 5.2241,
      "mean_token_accuracy": 0.20997274518013,
      "num_tokens": 11702982.0,
      "step": 6440
    },
    {
      "entropy": 5.824616193771362,
      "epoch": 5.537172324881822,
      "grad_norm": 1.09375,
      "learning_rate": 0.00026642402267190095,
      "loss": 5.2132,
      "mean_token_accuracy": 0.20879042893648148,
      "num_tokens": 11711994.0,
      "step": 6445
    },
    {
      "entropy": 5.837517738342285,
      "epoch": 5.541469703480876,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026609179099764313,
      "loss": 5.1651,
      "mean_token_accuracy": 0.21265248954296112,
      "num_tokens": 11722165.0,
      "step": 6450
    },
    {
      "entropy": 5.760106945037842,
      "epoch": 5.545767082079931,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026575957877535323,
      "loss": 5.1336,
      "mean_token_accuracy": 0.21656470894813537,
      "num_tokens": 11731265.0,
      "step": 6455
    },
    {
      "entropy": 5.820511150360107,
      "epoch": 5.5500644606789855,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026542738673044985,
      "loss": 5.1399,
      "mean_token_accuracy": 0.21291715651750565,
      "num_tokens": 11741779.0,
      "step": 6460
    },
    {
      "entropy": 5.779645776748657,
      "epoch": 5.55436183927804,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002650952155883077,
      "loss": 5.1135,
      "mean_token_accuracy": 0.21529941111803055,
      "num_tokens": 11749976.0,
      "step": 6465
    },
    {
      "entropy": 5.730343055725098,
      "epoch": 5.558659217877095,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002647630660742559,
      "loss": 5.1101,
      "mean_token_accuracy": 0.21454885900020598,
      "num_tokens": 11759781.0,
      "step": 6470
    },
    {
      "entropy": 5.749731779098511,
      "epoch": 5.56295659647615,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000264430938913576,
      "loss": 5.0849,
      "mean_token_accuracy": 0.21852831095457076,
      "num_tokens": 11769544.0,
      "step": 6475
    },
    {
      "entropy": 5.798023128509522,
      "epoch": 5.567253975075204,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026409883483150123,
      "loss": 5.0912,
      "mean_token_accuracy": 0.21736977249383926,
      "num_tokens": 11778831.0,
      "step": 6480
    },
    {
      "entropy": 5.767594051361084,
      "epoch": 5.571551353674259,
      "grad_norm": 1.1875,
      "learning_rate": 0.000263766754553214,
      "loss": 5.1532,
      "mean_token_accuracy": 0.20626595616340637,
      "num_tokens": 11788813.0,
      "step": 6485
    },
    {
      "entropy": 5.813692712783814,
      "epoch": 5.575848732273313,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002634346988038448,
      "loss": 5.0973,
      "mean_token_accuracy": 0.21707940846681595,
      "num_tokens": 11797335.0,
      "step": 6490
    },
    {
      "entropy": 5.771366548538208,
      "epoch": 5.580146110872368,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026310266830847093,
      "loss": 5.1268,
      "mean_token_accuracy": 0.2160950183868408,
      "num_tokens": 11806741.0,
      "step": 6495
    },
    {
      "entropy": 5.851240634918213,
      "epoch": 5.584443489471423,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026277066379211406,
      "loss": 5.1666,
      "mean_token_accuracy": 0.21153911352157592,
      "num_tokens": 11815551.0,
      "step": 6500
    },
    {
      "epoch": 5.584443489471423,
      "eval_entropy": 5.587484500966631,
      "eval_loss": 5.9088215827941895,
      "eval_mean_token_accuracy": 0.1789141287104235,
      "eval_num_tokens": 11815551.0,
      "eval_runtime": 2.2415,
      "eval_samples_per_second": 1583.307,
      "eval_steps_per_second": 198.081,
      "step": 6500
    },
    {
      "entropy": 5.810330581665039,
      "epoch": 5.588740868070477,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002624386859797396,
      "loss": 5.2016,
      "mean_token_accuracy": 0.20850666165351867,
      "num_tokens": 11824483.0,
      "step": 6505
    },
    {
      "entropy": 5.691527891159057,
      "epoch": 5.593038246669532,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026210673559625406,
      "loss": 4.9896,
      "mean_token_accuracy": 0.22865322530269622,
      "num_tokens": 11832383.0,
      "step": 6510
    },
    {
      "entropy": 5.7894144535064695,
      "epoch": 5.597335625268586,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002617748133665047,
      "loss": 5.2118,
      "mean_token_accuracy": 0.2122812032699585,
      "num_tokens": 11841430.0,
      "step": 6515
    },
    {
      "entropy": 5.763946056365967,
      "epoch": 5.601633003867641,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002614429200152768,
      "loss": 5.183,
      "mean_token_accuracy": 0.21466665118932723,
      "num_tokens": 11850863.0,
      "step": 6520
    },
    {
      "entropy": 5.779318046569824,
      "epoch": 5.6059303824666955,
      "grad_norm": 1.265625,
      "learning_rate": 0.000261111056267293,
      "loss": 5.0385,
      "mean_token_accuracy": 0.22099136859178542,
      "num_tokens": 11859392.0,
      "step": 6525
    },
    {
      "entropy": 5.7794242858886715,
      "epoch": 5.61022776106575,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026077922284721084,
      "loss": 5.1003,
      "mean_token_accuracy": 0.22238964438438416,
      "num_tokens": 11868762.0,
      "step": 6530
    },
    {
      "entropy": 5.727440547943115,
      "epoch": 5.614525139664805,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026044742047962206,
      "loss": 5.0703,
      "mean_token_accuracy": 0.22508458644151688,
      "num_tokens": 11876722.0,
      "step": 6535
    },
    {
      "entropy": 5.738032865524292,
      "epoch": 5.618822518263859,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026011564988905023,
      "loss": 5.2065,
      "mean_token_accuracy": 0.21719919443130492,
      "num_tokens": 11885614.0,
      "step": 6540
    },
    {
      "entropy": 5.8498131275177006,
      "epoch": 5.623119896862914,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002597839117999499,
      "loss": 5.201,
      "mean_token_accuracy": 0.2126613289117813,
      "num_tokens": 11894702.0,
      "step": 6545
    },
    {
      "entropy": 5.789801359176636,
      "epoch": 5.627417275461968,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002594522069367044,
      "loss": 5.0842,
      "mean_token_accuracy": 0.2244862824678421,
      "num_tokens": 11902829.0,
      "step": 6550
    },
    {
      "entropy": 5.747990703582763,
      "epoch": 5.631714654061023,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002591205360236245,
      "loss": 5.1154,
      "mean_token_accuracy": 0.21725053191184998,
      "num_tokens": 11912377.0,
      "step": 6555
    },
    {
      "entropy": 5.767636060714722,
      "epoch": 5.636012032660077,
      "grad_norm": 1.265625,
      "learning_rate": 0.000258788899784947,
      "loss": 5.1205,
      "mean_token_accuracy": 0.21221498697996138,
      "num_tokens": 11920563.0,
      "step": 6560
    },
    {
      "entropy": 5.810602903366089,
      "epoch": 5.640309411259132,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00025845729894483283,
      "loss": 5.1613,
      "mean_token_accuracy": 0.209673647582531,
      "num_tokens": 11930190.0,
      "step": 6565
    },
    {
      "entropy": 5.828999805450439,
      "epoch": 5.644606789858186,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002581257342273657,
      "loss": 5.2191,
      "mean_token_accuracy": 0.21002251505851746,
      "num_tokens": 11939840.0,
      "step": 6570
    },
    {
      "entropy": 5.790510845184326,
      "epoch": 5.648904168457241,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002577942063565504,
      "loss": 5.1473,
      "mean_token_accuracy": 0.21715394109487535,
      "num_tokens": 11948260.0,
      "step": 6575
    },
    {
      "entropy": 5.780890369415284,
      "epoch": 5.6532015470562955,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002574627160563114,
      "loss": 5.2054,
      "mean_token_accuracy": 0.21656007170677186,
      "num_tokens": 11956776.0,
      "step": 6580
    },
    {
      "entropy": 5.833474206924438,
      "epoch": 5.65749892565535,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002571312640504909,
      "loss": 5.2272,
      "mean_token_accuracy": 0.20785245895385743,
      "num_tokens": 11966375.0,
      "step": 6585
    },
    {
      "entropy": 5.903450536727905,
      "epoch": 5.661796304254405,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002567998510628476,
      "loss": 5.2413,
      "mean_token_accuracy": 0.20541951507329942,
      "num_tokens": 11975835.0,
      "step": 6590
    },
    {
      "entropy": 5.7732603549957275,
      "epoch": 5.666093682853459,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025646847781705506,
      "loss": 5.1226,
      "mean_token_accuracy": 0.21898027807474135,
      "num_tokens": 11984672.0,
      "step": 6595
    },
    {
      "entropy": 5.7607639789581295,
      "epoch": 5.670391061452514,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002561371450367,
      "loss": 5.1289,
      "mean_token_accuracy": 0.2148086816072464,
      "num_tokens": 11993954.0,
      "step": 6600
    },
    {
      "entropy": 5.799904012680054,
      "epoch": 5.674688440051568,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025580585344528076,
      "loss": 5.1941,
      "mean_token_accuracy": 0.2090480461716652,
      "num_tokens": 12002523.0,
      "step": 6605
    },
    {
      "entropy": 5.802334547042847,
      "epoch": 5.678985818650623,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002554746037662058,
      "loss": 5.2153,
      "mean_token_accuracy": 0.20566553473472596,
      "num_tokens": 12011638.0,
      "step": 6610
    },
    {
      "entropy": 5.868585634231567,
      "epoch": 5.683283197249677,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002551433967227919,
      "loss": 5.1641,
      "mean_token_accuracy": 0.21109885722398758,
      "num_tokens": 12021319.0,
      "step": 6615
    },
    {
      "entropy": 5.80119833946228,
      "epoch": 5.687580575848732,
      "grad_norm": 1.1875,
      "learning_rate": 0.000254812233038263,
      "loss": 5.106,
      "mean_token_accuracy": 0.22026268839836122,
      "num_tokens": 12030255.0,
      "step": 6620
    },
    {
      "entropy": 5.785006141662597,
      "epoch": 5.691877954447786,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025448111343574813,
      "loss": 5.1206,
      "mean_token_accuracy": 0.2177802160382271,
      "num_tokens": 12038884.0,
      "step": 6625
    },
    {
      "entropy": 5.755448532104492,
      "epoch": 5.696175333046842,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002541500386382802,
      "loss": 5.1032,
      "mean_token_accuracy": 0.2135314717888832,
      "num_tokens": 12047477.0,
      "step": 6630
    },
    {
      "entropy": 5.795413208007813,
      "epoch": 5.7004727116458955,
      "grad_norm": 1.203125,
      "learning_rate": 0.00025381900936879433,
      "loss": 5.1846,
      "mean_token_accuracy": 0.21453225016593933,
      "num_tokens": 12056902.0,
      "step": 6635
    },
    {
      "entropy": 5.7829125881195065,
      "epoch": 5.704770090244951,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002534880263501259,
      "loss": 5.1542,
      "mean_token_accuracy": 0.21434427946805953,
      "num_tokens": 12065721.0,
      "step": 6640
    },
    {
      "entropy": 5.759543371200562,
      "epoch": 5.7090674688440055,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002531570903050097,
      "loss": 5.1091,
      "mean_token_accuracy": 0.21880163848400117,
      "num_tokens": 12074870.0,
      "step": 6645
    },
    {
      "entropy": 5.818104219436646,
      "epoch": 5.71336484744306,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002528262019560776,
      "loss": 5.158,
      "mean_token_accuracy": 0.21322663277387618,
      "num_tokens": 12084557.0,
      "step": 6650
    },
    {
      "entropy": 5.762011098861694,
      "epoch": 5.717662226042115,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002524953620258579,
      "loss": 5.0262,
      "mean_token_accuracy": 0.22854391783475875,
      "num_tokens": 12093074.0,
      "step": 6655
    },
    {
      "entropy": 5.73213381767273,
      "epoch": 5.721959604641169,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002521645712367724,
      "loss": 5.0496,
      "mean_token_accuracy": 0.22822125405073165,
      "num_tokens": 12102785.0,
      "step": 6660
    },
    {
      "entropy": 5.755954074859619,
      "epoch": 5.726256983240224,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025183383031113606,
      "loss": 5.0837,
      "mean_token_accuracy": 0.2249210149049759,
      "num_tokens": 12112535.0,
      "step": 6665
    },
    {
      "entropy": 5.667020416259765,
      "epoch": 5.730554361839278,
      "grad_norm": 1.15625,
      "learning_rate": 0.00025150313997115476,
      "loss": 5.0416,
      "mean_token_accuracy": 0.22201696187257766,
      "num_tokens": 12121604.0,
      "step": 6670
    },
    {
      "entropy": 5.77487530708313,
      "epoch": 5.734851740438333,
      "grad_norm": 1.125,
      "learning_rate": 0.0002511725009389244,
      "loss": 5.1933,
      "mean_token_accuracy": 0.21116092801094055,
      "num_tokens": 12131276.0,
      "step": 6675
    },
    {
      "entropy": 5.765606832504273,
      "epoch": 5.739149119037387,
      "grad_norm": 1.3125,
      "learning_rate": 0.000250841913936428,
      "loss": 5.1503,
      "mean_token_accuracy": 0.21786546260118483,
      "num_tokens": 12140180.0,
      "step": 6680
    },
    {
      "entropy": 5.75055570602417,
      "epoch": 5.743446497636442,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002505113796855357,
      "loss": 5.143,
      "mean_token_accuracy": 0.21634120494127274,
      "num_tokens": 12149635.0,
      "step": 6685
    },
    {
      "entropy": 5.770916938781738,
      "epoch": 5.747743876235496,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025018089890800225,
      "loss": 5.0908,
      "mean_token_accuracy": 0.22405355870723725,
      "num_tokens": 12157565.0,
      "step": 6690
    },
    {
      "entropy": 5.780913352966309,
      "epoch": 5.752041254834551,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024985047232546544,
      "loss": 5.1884,
      "mean_token_accuracy": 0.21726172119379045,
      "num_tokens": 12166647.0,
      "step": 6695
    },
    {
      "entropy": 5.8112811088562015,
      "epoch": 5.7563386334336055,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00024952010065944485,
      "loss": 5.1644,
      "mean_token_accuracy": 0.21350513994693757,
      "num_tokens": 12175554.0,
      "step": 6700
    },
    {
      "entropy": 5.794380140304566,
      "epoch": 5.76063601203266,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002491897846313402,
      "loss": 5.1415,
      "mean_token_accuracy": 0.21891088634729386,
      "num_tokens": 12184756.0,
      "step": 6705
    },
    {
      "entropy": 5.844700193405151,
      "epoch": 5.764933390631715,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002488595249624297,
      "loss": 5.2308,
      "mean_token_accuracy": 0.20482745319604873,
      "num_tokens": 12194724.0,
      "step": 6710
    },
    {
      "entropy": 5.777750396728516,
      "epoch": 5.769230769230769,
      "grad_norm": 1.09375,
      "learning_rate": 0.00024852932237386837,
      "loss": 5.1602,
      "mean_token_accuracy": 0.21278387904167176,
      "num_tokens": 12203804.0,
      "step": 6715
    },
    {
      "entropy": 5.781616306304931,
      "epoch": 5.773528147829824,
      "grad_norm": 1.125,
      "learning_rate": 0.00024819917758668673,
      "loss": 5.1177,
      "mean_token_accuracy": 0.21435283571481706,
      "num_tokens": 12212868.0,
      "step": 6720
    },
    {
      "entropy": 5.795768356323242,
      "epoch": 5.777825526428878,
      "grad_norm": 1.171875,
      "learning_rate": 0.00024786909132178906,
      "loss": 5.1967,
      "mean_token_accuracy": 0.21346242129802703,
      "num_tokens": 12221650.0,
      "step": 6725
    },
    {
      "entropy": 5.770245027542114,
      "epoch": 5.782122905027933,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024753906429995194,
      "loss": 5.0708,
      "mean_token_accuracy": 0.2283392906188965,
      "num_tokens": 12231541.0,
      "step": 6730
    },
    {
      "entropy": 5.815408086776733,
      "epoch": 5.786420283626987,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002472090972418222,
      "loss": 5.1723,
      "mean_token_accuracy": 0.21145310401916503,
      "num_tokens": 12240899.0,
      "step": 6735
    },
    {
      "entropy": 5.7969895362854,
      "epoch": 5.790717662226042,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002468791908679163,
      "loss": 5.16,
      "mean_token_accuracy": 0.20907359272241594,
      "num_tokens": 12250352.0,
      "step": 6740
    },
    {
      "entropy": 5.8266209125518795,
      "epoch": 5.795015040825096,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002465493458986175,
      "loss": 5.224,
      "mean_token_accuracy": 0.20271035730838777,
      "num_tokens": 12259975.0,
      "step": 6745
    },
    {
      "entropy": 5.690389919281006,
      "epoch": 5.799312419424151,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024621956305417587,
      "loss": 5.0584,
      "mean_token_accuracy": 0.22248426228761672,
      "num_tokens": 12269203.0,
      "step": 6750
    },
    {
      "entropy": 5.801789426803589,
      "epoch": 5.8036097980232055,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000245889843054705,
      "loss": 5.1674,
      "mean_token_accuracy": 0.2146278813481331,
      "num_tokens": 12279481.0,
      "step": 6755
    },
    {
      "entropy": 5.766292572021484,
      "epoch": 5.80790717662226,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00024556018662018163,
      "loss": 5.1355,
      "mean_token_accuracy": 0.21804405450820924,
      "num_tokens": 12288848.0,
      "step": 6760
    },
    {
      "entropy": 5.825388383865357,
      "epoch": 5.812204555221315,
      "grad_norm": 1.140625,
      "learning_rate": 0.00024523059447044377,
      "loss": 5.1359,
      "mean_token_accuracy": 0.2106882631778717,
      "num_tokens": 12297346.0,
      "step": 6765
    },
    {
      "entropy": 5.811703062057495,
      "epoch": 5.816501933820369,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002449010673251887,
      "loss": 5.1457,
      "mean_token_accuracy": 0.2106386587023735,
      "num_tokens": 12306233.0,
      "step": 6770
    },
    {
      "entropy": 5.837384557723999,
      "epoch": 5.820799312419425,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002445716059039723,
      "loss": 5.2415,
      "mean_token_accuracy": 0.20431807935237883,
      "num_tokens": 12315609.0,
      "step": 6775
    },
    {
      "entropy": 5.803850889205933,
      "epoch": 5.825096691018478,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024424221092620644,
      "loss": 5.1938,
      "mean_token_accuracy": 0.21159621477127075,
      "num_tokens": 12323915.0,
      "step": 6780
    },
    {
      "entropy": 5.721775150299072,
      "epoch": 5.829394069617534,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00024391288311115822,
      "loss": 5.1136,
      "mean_token_accuracy": 0.2114120438694954,
      "num_tokens": 12334077.0,
      "step": 6785
    },
    {
      "entropy": 5.771698474884033,
      "epoch": 5.833691448216588,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002435836231779478,
      "loss": 5.1534,
      "mean_token_accuracy": 0.21626046299934387,
      "num_tokens": 12342411.0,
      "step": 6790
    },
    {
      "entropy": 5.740696430206299,
      "epoch": 5.837988826815643,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00024325443184554724,
      "loss": 5.0498,
      "mean_token_accuracy": 0.2268505409359932,
      "num_tokens": 12351308.0,
      "step": 6795
    },
    {
      "entropy": 5.778064870834351,
      "epoch": 5.842286205414697,
      "grad_norm": 1.1875,
      "learning_rate": 0.00024292530983277904,
      "loss": 5.2086,
      "mean_token_accuracy": 0.211841943860054,
      "num_tokens": 12359673.0,
      "step": 6800
    },
    {
      "entropy": 5.745600652694702,
      "epoch": 5.846583584013752,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024259625785831408,
      "loss": 5.0643,
      "mean_token_accuracy": 0.2172771066427231,
      "num_tokens": 12367876.0,
      "step": 6805
    },
    {
      "entropy": 5.812221240997315,
      "epoch": 5.850880962612806,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024226727664067023,
      "loss": 5.2167,
      "mean_token_accuracy": 0.2113609328866005,
      "num_tokens": 12377040.0,
      "step": 6810
    },
    {
      "entropy": 5.85981330871582,
      "epoch": 5.855178341211861,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024193836689821109,
      "loss": 5.2745,
      "mean_token_accuracy": 0.20549044460058213,
      "num_tokens": 12387622.0,
      "step": 6815
    },
    {
      "entropy": 5.698194265365601,
      "epoch": 5.8594757198109155,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002416095293491439,
      "loss": 5.0575,
      "mean_token_accuracy": 0.22093253284692765,
      "num_tokens": 12396447.0,
      "step": 6820
    },
    {
      "entropy": 5.789422225952149,
      "epoch": 5.86377309840997,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002412807647115186,
      "loss": 5.0657,
      "mean_token_accuracy": 0.2230805352330208,
      "num_tokens": 12405887.0,
      "step": 6825
    },
    {
      "entropy": 5.756113576889038,
      "epoch": 5.868070477009025,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024095207370322574,
      "loss": 5.0966,
      "mean_token_accuracy": 0.21677415072917938,
      "num_tokens": 12414543.0,
      "step": 6830
    },
    {
      "entropy": 5.7737664699554445,
      "epoch": 5.872367855608079,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024062345704199507,
      "loss": 5.1109,
      "mean_token_accuracy": 0.21734392493963242,
      "num_tokens": 12423370.0,
      "step": 6835
    },
    {
      "entropy": 5.804680013656617,
      "epoch": 5.876665234207134,
      "grad_norm": 1.0,
      "learning_rate": 0.00024029491544539405,
      "loss": 5.2056,
      "mean_token_accuracy": 0.20914409458637237,
      "num_tokens": 12433980.0,
      "step": 6840
    },
    {
      "entropy": 5.737584018707276,
      "epoch": 5.880962612806188,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023996644963082616,
      "loss": 5.1546,
      "mean_token_accuracy": 0.22128169238567352,
      "num_tokens": 12443300.0,
      "step": 6845
    },
    {
      "entropy": 5.804375219345093,
      "epoch": 5.885259991405243,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00023963806031552948,
      "loss": 5.1943,
      "mean_token_accuracy": 0.2085866376757622,
      "num_tokens": 12452462.0,
      "step": 6850
    },
    {
      "entropy": 5.748406171798706,
      "epoch": 5.889557370004297,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00023930974821657504,
      "loss": 5.1142,
      "mean_token_accuracy": 0.21527279913425446,
      "num_tokens": 12461605.0,
      "step": 6855
    },
    {
      "entropy": 5.866064405441284,
      "epoch": 5.893854748603352,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00023898151405086533,
      "loss": 5.185,
      "mean_token_accuracy": 0.21090082675218583,
      "num_tokens": 12470905.0,
      "step": 6860
    },
    {
      "entropy": 5.825461673736572,
      "epoch": 5.8981521272024064,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023865335853513232,
      "loss": 5.1773,
      "mean_token_accuracy": 0.21529610455036163,
      "num_tokens": 12478913.0,
      "step": 6865
    },
    {
      "entropy": 5.788094234466553,
      "epoch": 5.902449505801461,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00023832528238593677,
      "loss": 5.2454,
      "mean_token_accuracy": 0.20996989905834199,
      "num_tokens": 12487561.0,
      "step": 6870
    },
    {
      "entropy": 5.747205543518066,
      "epoch": 5.9067468844005155,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023799728631966556,
      "loss": 5.1313,
      "mean_token_accuracy": 0.21550016105175018,
      "num_tokens": 12496781.0,
      "step": 6875
    },
    {
      "entropy": 5.8067419052124025,
      "epoch": 5.91104426299957,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002376693710525313,
      "loss": 5.2261,
      "mean_token_accuracy": 0.21084323525428772,
      "num_tokens": 12505716.0,
      "step": 6880
    },
    {
      "entropy": 5.8443159580230715,
      "epoch": 5.915341641598625,
      "grad_norm": 1.109375,
      "learning_rate": 0.00023734153730056967,
      "loss": 5.1854,
      "mean_token_accuracy": 0.2098903939127922,
      "num_tokens": 12515594.0,
      "step": 6885
    },
    {
      "entropy": 5.75254225730896,
      "epoch": 5.919639020197679,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023701378577963873,
      "loss": 5.0097,
      "mean_token_accuracy": 0.22866766154766083,
      "num_tokens": 12523439.0,
      "step": 6890
    },
    {
      "entropy": 5.754075956344605,
      "epoch": 5.923936398796734,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002366861172054166,
      "loss": 5.1066,
      "mean_token_accuracy": 0.21780942380428314,
      "num_tokens": 12532100.0,
      "step": 6895
    },
    {
      "entropy": 5.77516279220581,
      "epoch": 5.928233777395788,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023635853229340054,
      "loss": 5.1563,
      "mean_token_accuracy": 0.21333520114421844,
      "num_tokens": 12539689.0,
      "step": 6900
    },
    {
      "entropy": 5.788961362838745,
      "epoch": 5.932531155994843,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023603103175890512,
      "loss": 5.1674,
      "mean_token_accuracy": 0.20800318717956542,
      "num_tokens": 12548486.0,
      "step": 6905
    },
    {
      "entropy": 5.780201768875122,
      "epoch": 5.936828534593897,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023570361631706062,
      "loss": 5.0482,
      "mean_token_accuracy": 0.22682550996541978,
      "num_tokens": 12557423.0,
      "step": 6910
    },
    {
      "entropy": 5.752371120452881,
      "epoch": 5.941125913192952,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023537628668281142,
      "loss": 5.1793,
      "mean_token_accuracy": 0.21411100775003433,
      "num_tokens": 12566086.0,
      "step": 6915
    },
    {
      "entropy": 5.739467906951904,
      "epoch": 5.945423291792007,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00023504904357091468,
      "loss": 5.0906,
      "mean_token_accuracy": 0.21916348338127137,
      "num_tokens": 12575827.0,
      "step": 6920
    },
    {
      "entropy": 5.817741394042969,
      "epoch": 5.949720670391061,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002347218876959384,
      "loss": 5.0988,
      "mean_token_accuracy": 0.22555966377258302,
      "num_tokens": 12585044.0,
      "step": 6925
    },
    {
      "entropy": 5.774907779693604,
      "epoch": 5.954018048990116,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002343948197722604,
      "loss": 5.1268,
      "mean_token_accuracy": 0.2213137552142143,
      "num_tokens": 12594677.0,
      "step": 6930
    },
    {
      "entropy": 5.723889780044556,
      "epoch": 5.958315427589171,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023406784051406638,
      "loss": 5.1446,
      "mean_token_accuracy": 0.21569790840148925,
      "num_tokens": 12604829.0,
      "step": 6935
    },
    {
      "entropy": 5.844821977615356,
      "epoch": 5.9626128061882255,
      "grad_norm": 1.1875,
      "learning_rate": 0.00023374095063534816,
      "loss": 5.1469,
      "mean_token_accuracy": 0.22094390839338302,
      "num_tokens": 12613869.0,
      "step": 6940
    },
    {
      "entropy": 5.773779249191284,
      "epoch": 5.96691018478728,
      "grad_norm": 1.1875,
      "learning_rate": 0.00023341415084990276,
      "loss": 5.1121,
      "mean_token_accuracy": 0.21411748081445695,
      "num_tokens": 12623248.0,
      "step": 6945
    },
    {
      "entropy": 5.678793716430664,
      "epoch": 5.971207563386335,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023308744187132996,
      "loss": 5.0518,
      "mean_token_accuracy": 0.2295151397585869,
      "num_tokens": 12631973.0,
      "step": 6950
    },
    {
      "entropy": 5.788469696044922,
      "epoch": 5.975504941985389,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023276082441303197,
      "loss": 5.1718,
      "mean_token_accuracy": 0.21482646316289902,
      "num_tokens": 12641435.0,
      "step": 6955
    },
    {
      "entropy": 5.832605695724487,
      "epoch": 5.979802320584444,
      "grad_norm": 1.1875,
      "learning_rate": 0.00023243429918821056,
      "loss": 5.1501,
      "mean_token_accuracy": 0.2146319031715393,
      "num_tokens": 12651077.0,
      "step": 6960
    },
    {
      "entropy": 5.868136739730835,
      "epoch": 5.984099699183498,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023210786690986646,
      "loss": 5.2385,
      "mean_token_accuracy": 0.2106163904070854,
      "num_tokens": 12659929.0,
      "step": 6965
    },
    {
      "entropy": 5.773500680923462,
      "epoch": 5.988397077782553,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023178152829079712,
      "loss": 5.0916,
      "mean_token_accuracy": 0.22088077813386917,
      "num_tokens": 12670725.0,
      "step": 6970
    },
    {
      "entropy": 5.763846731185913,
      "epoch": 5.992694456381607,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023145528404359562,
      "loss": 5.1313,
      "mean_token_accuracy": 0.22139803171157837,
      "num_tokens": 12680820.0,
      "step": 6975
    },
    {
      "entropy": 5.784390497207641,
      "epoch": 5.996991834980662,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002311291348806492,
      "loss": 5.2048,
      "mean_token_accuracy": 0.20281474888324738,
      "num_tokens": 12689785.0,
      "step": 6980
    },
    {
      "entropy": 5.7638511127895775,
      "epoch": 6.000859475719811,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002308030815141372,
      "loss": 5.1312,
      "mean_token_accuracy": 0.21178502175543043,
      "num_tokens": 12697221.0,
      "step": 6985
    },
    {
      "entropy": 5.737544345855713,
      "epoch": 6.005156854318866,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023047712465602976,
      "loss": 4.9411,
      "mean_token_accuracy": 0.23268725723028183,
      "num_tokens": 12707127.0,
      "step": 6990
    },
    {
      "entropy": 5.735475063323975,
      "epoch": 6.00945423291792,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023015126501808641,
      "loss": 4.966,
      "mean_token_accuracy": 0.2215101584792137,
      "num_tokens": 12715364.0,
      "step": 6995
    },
    {
      "entropy": 5.776022481918335,
      "epoch": 6.013751611516975,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00022982550331185437,
      "loss": 4.9637,
      "mean_token_accuracy": 0.2314266100525856,
      "num_tokens": 12724914.0,
      "step": 7000
    },
    {
      "epoch": 6.013751611516975,
      "eval_entropy": 5.5485868609703335,
      "eval_loss": 5.897665977478027,
      "eval_mean_token_accuracy": 0.18092229928787765,
      "eval_num_tokens": 12724914.0,
      "eval_runtime": 2.0511,
      "eval_samples_per_second": 1730.293,
      "eval_steps_per_second": 216.469,
      "step": 7000
    },
    {
      "entropy": 5.691806173324585,
      "epoch": 6.0180489901160295,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00022949984024866704,
      "loss": 4.9672,
      "mean_token_accuracy": 0.23159780353307724,
      "num_tokens": 12735193.0,
      "step": 7005
    },
    {
      "entropy": 5.781547927856446,
      "epoch": 6.022346368715084,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002291742765396424,
      "loss": 5.0222,
      "mean_token_accuracy": 0.22906866669654846,
      "num_tokens": 12743945.0,
      "step": 7010
    },
    {
      "entropy": 5.777514505386352,
      "epoch": 6.0266437473141385,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00022884881289568133,
      "loss": 5.0386,
      "mean_token_accuracy": 0.22094691544771194,
      "num_tokens": 12753130.0,
      "step": 7015
    },
    {
      "entropy": 5.6990760326385494,
      "epoch": 6.030941125913193,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002285234500274665,
      "loss": 5.0016,
      "mean_token_accuracy": 0.2268824100494385,
      "num_tokens": 12762108.0,
      "step": 7020
    },
    {
      "entropy": 5.840245294570923,
      "epoch": 6.035238504512248,
      "grad_norm": 1.25,
      "learning_rate": 0.00022819818864546016,
      "loss": 5.0613,
      "mean_token_accuracy": 0.22050540000200272,
      "num_tokens": 12772102.0,
      "step": 7025
    },
    {
      "entropy": 5.788558721542358,
      "epoch": 6.039535883111302,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022787302945990345,
      "loss": 5.0144,
      "mean_token_accuracy": 0.23190406560897828,
      "num_tokens": 12781225.0,
      "step": 7030
    },
    {
      "entropy": 5.784692239761353,
      "epoch": 6.043833261710357,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022754797318081383,
      "loss": 4.9778,
      "mean_token_accuracy": 0.23033921122550965,
      "num_tokens": 12789896.0,
      "step": 7035
    },
    {
      "entropy": 5.75786280632019,
      "epoch": 6.048130640309411,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022722302051798442,
      "loss": 4.9264,
      "mean_token_accuracy": 0.22915157228708266,
      "num_tokens": 12798596.0,
      "step": 7040
    },
    {
      "entropy": 5.666304588317871,
      "epoch": 6.052428018908466,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002268981721809819,
      "loss": 4.9324,
      "mean_token_accuracy": 0.23248923122882842,
      "num_tokens": 12807285.0,
      "step": 7045
    },
    {
      "entropy": 5.76234278678894,
      "epoch": 6.05672539750752,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002265734288791451,
      "loss": 5.0015,
      "mean_token_accuracy": 0.22686269730329514,
      "num_tokens": 12816668.0,
      "step": 7050
    },
    {
      "entropy": 5.819028425216675,
      "epoch": 6.061022776106575,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022624879132158377,
      "loss": 5.0876,
      "mean_token_accuracy": 0.22044486552476883,
      "num_tokens": 12825943.0,
      "step": 7055
    },
    {
      "entropy": 5.733125495910644,
      "epoch": 6.0653201547056295,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00022592426021717654,
      "loss": 4.9027,
      "mean_token_accuracy": 0.24135779589414597,
      "num_tokens": 12835693.0,
      "step": 7060
    },
    {
      "entropy": 5.726211166381836,
      "epoch": 6.069617533304684,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002255998362745696,
      "loss": 4.8067,
      "mean_token_accuracy": 0.2527772426605225,
      "num_tokens": 12844201.0,
      "step": 7065
    },
    {
      "entropy": 5.697748184204102,
      "epoch": 6.073914911903739,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022527552020217513,
      "loss": 4.9578,
      "mean_token_accuracy": 0.23135053366422653,
      "num_tokens": 12853220.0,
      "step": 7070
    },
    {
      "entropy": 5.7508221626281735,
      "epoch": 6.078212290502793,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002249513127081697,
      "loss": 5.0379,
      "mean_token_accuracy": 0.22537065148353577,
      "num_tokens": 12862486.0,
      "step": 7075
    },
    {
      "entropy": 5.821695613861084,
      "epoch": 6.082509669101848,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022462721450049316,
      "loss": 5.0331,
      "mean_token_accuracy": 0.22243736535310746,
      "num_tokens": 12871717.0,
      "step": 7080
    },
    {
      "entropy": 5.768175172805786,
      "epoch": 6.086807047700902,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002243032262868464,
      "loss": 4.98,
      "mean_token_accuracy": 0.22336842119693756,
      "num_tokens": 12881278.0,
      "step": 7085
    },
    {
      "entropy": 5.767650985717774,
      "epoch": 6.091104426299957,
      "grad_norm": 1.140625,
      "learning_rate": 0.00022397934877469,
      "loss": 5.0105,
      "mean_token_accuracy": 0.2182280093431473,
      "num_tokens": 12890720.0,
      "step": 7090
    },
    {
      "entropy": 5.7904541969299315,
      "epoch": 6.095401804899011,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002236555826712432,
      "loss": 5.0389,
      "mean_token_accuracy": 0.2230403557419777,
      "num_tokens": 12900428.0,
      "step": 7095
    },
    {
      "entropy": 5.838481426239014,
      "epoch": 6.099699183498066,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022333192868348152,
      "loss": 5.0203,
      "mean_token_accuracy": 0.22564003616571426,
      "num_tokens": 12910177.0,
      "step": 7100
    },
    {
      "entropy": 5.762051153182983,
      "epoch": 6.10399656209712,
      "grad_norm": 1.140625,
      "learning_rate": 0.00022300838751813606,
      "loss": 5.049,
      "mean_token_accuracy": 0.2290334463119507,
      "num_tokens": 12920734.0,
      "step": 7105
    },
    {
      "entropy": 5.759262037277222,
      "epoch": 6.108293940696175,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022268495988169145,
      "loss": 4.9445,
      "mean_token_accuracy": 0.23307121843099593,
      "num_tokens": 12929585.0,
      "step": 7110
    },
    {
      "entropy": 5.68530306816101,
      "epoch": 6.1125913192952295,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022236164648038433,
      "loss": 4.9332,
      "mean_token_accuracy": 0.2352724328637123,
      "num_tokens": 12938933.0,
      "step": 7115
    },
    {
      "entropy": 5.717230653762817,
      "epoch": 6.116888697894285,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002220384480202019,
      "loss": 4.9107,
      "mean_token_accuracy": 0.23734596371650696,
      "num_tokens": 12947461.0,
      "step": 7120
    },
    {
      "entropy": 5.7552255153656,
      "epoch": 6.1211860764933395,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022171536520688046,
      "loss": 4.9756,
      "mean_token_accuracy": 0.23315537869930267,
      "num_tokens": 12956507.0,
      "step": 7125
    },
    {
      "entropy": 5.797720336914063,
      "epoch": 6.125483455092394,
      "grad_norm": 1.25,
      "learning_rate": 0.00022139239874590362,
      "loss": 5.0782,
      "mean_token_accuracy": 0.22219871133565902,
      "num_tokens": 12965740.0,
      "step": 7130
    },
    {
      "entropy": 5.767501020431519,
      "epoch": 6.1297808336914485,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002210695493425013,
      "loss": 4.9994,
      "mean_token_accuracy": 0.22486449778079987,
      "num_tokens": 12975057.0,
      "step": 7135
    },
    {
      "entropy": 5.700614976882934,
      "epoch": 6.134078212290503,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022074681770164735,
      "loss": 4.9421,
      "mean_token_accuracy": 0.22969190180301666,
      "num_tokens": 12984087.0,
      "step": 7140
    },
    {
      "entropy": 5.704544973373413,
      "epoch": 6.138375590889558,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00022042420452805868,
      "loss": 4.9477,
      "mean_token_accuracy": 0.23626155406236649,
      "num_tokens": 12992793.0,
      "step": 7145
    },
    {
      "entropy": 5.783449697494507,
      "epoch": 6.142672969488612,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022010171052619365,
      "loss": 5.0133,
      "mean_token_accuracy": 0.2346527874469757,
      "num_tokens": 13000769.0,
      "step": 7150
    },
    {
      "entropy": 5.748849725723266,
      "epoch": 6.146970348087667,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021977933640025,
      "loss": 4.9984,
      "mean_token_accuracy": 0.22886652797460555,
      "num_tokens": 13010677.0,
      "step": 7155
    },
    {
      "entropy": 5.6607763290405275,
      "epoch": 6.151267726686721,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021945708285416434,
      "loss": 4.8752,
      "mean_token_accuracy": 0.24084128588438034,
      "num_tokens": 13019791.0,
      "step": 7160
    },
    {
      "entropy": 5.758824348449707,
      "epoch": 6.155565105285776,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002191349505916093,
      "loss": 5.0239,
      "mean_token_accuracy": 0.22872246205806732,
      "num_tokens": 13029223.0,
      "step": 7165
    },
    {
      "entropy": 5.775124549865723,
      "epoch": 6.15986248388483,
      "grad_norm": 1.171875,
      "learning_rate": 0.00021881294031599318,
      "loss": 4.9713,
      "mean_token_accuracy": 0.23292697370052337,
      "num_tokens": 13038716.0,
      "step": 7170
    },
    {
      "entropy": 5.781008958816528,
      "epoch": 6.164159862483885,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002184910527304576,
      "loss": 5.0122,
      "mean_token_accuracy": 0.2295176088809967,
      "num_tokens": 13047915.0,
      "step": 7175
    },
    {
      "entropy": 5.699311828613281,
      "epoch": 6.1684572410829395,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021816928853787636,
      "loss": 4.9639,
      "mean_token_accuracy": 0.23525033593177797,
      "num_tokens": 13056613.0,
      "step": 7180
    },
    {
      "entropy": 5.756593084335327,
      "epoch": 6.172754619681994,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021784764844085398,
      "loss": 5.0108,
      "mean_token_accuracy": 0.2276516392827034,
      "num_tokens": 13066658.0,
      "step": 7185
    },
    {
      "entropy": 5.863659620285034,
      "epoch": 6.177051998281049,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002175261331417238,
      "loss": 5.0337,
      "mean_token_accuracy": 0.2199842929840088,
      "num_tokens": 13074798.0,
      "step": 7190
    },
    {
      "entropy": 5.73055624961853,
      "epoch": 6.181349376880103,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021720474334254675,
      "loss": 4.9447,
      "mean_token_accuracy": 0.22651222050189973,
      "num_tokens": 13084173.0,
      "step": 7195
    },
    {
      "entropy": 5.759624099731445,
      "epoch": 6.185646755479158,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021688347974510962,
      "loss": 4.9736,
      "mean_token_accuracy": 0.22720245718955995,
      "num_tokens": 13093096.0,
      "step": 7200
    },
    {
      "entropy": 5.735679626464844,
      "epoch": 6.189944134078212,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021656234305092377,
      "loss": 4.975,
      "mean_token_accuracy": 0.2323437213897705,
      "num_tokens": 13101191.0,
      "step": 7205
    },
    {
      "entropy": 5.745137310028076,
      "epoch": 6.194241512677267,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002162413339612234,
      "loss": 5.0005,
      "mean_token_accuracy": 0.22740160077810287,
      "num_tokens": 13109829.0,
      "step": 7210
    },
    {
      "entropy": 5.718288230895996,
      "epoch": 6.198538891276321,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021592045317696406,
      "loss": 4.9514,
      "mean_token_accuracy": 0.23580033779144288,
      "num_tokens": 13119314.0,
      "step": 7215
    },
    {
      "entropy": 5.710982465744019,
      "epoch": 6.202836269875376,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021559970139882102,
      "loss": 4.9237,
      "mean_token_accuracy": 0.2312058076262474,
      "num_tokens": 13128113.0,
      "step": 7220
    },
    {
      "entropy": 5.825408935546875,
      "epoch": 6.20713364847443,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002152790793271881,
      "loss": 5.0725,
      "mean_token_accuracy": 0.21972693353891373,
      "num_tokens": 13136892.0,
      "step": 7225
    },
    {
      "entropy": 5.790244007110596,
      "epoch": 6.211431027073485,
      "grad_norm": 1.09375,
      "learning_rate": 0.00021495858766217558,
      "loss": 5.0404,
      "mean_token_accuracy": 0.22273045629262925,
      "num_tokens": 13146960.0,
      "step": 7230
    },
    {
      "entropy": 5.7194140434265135,
      "epoch": 6.2157284056725395,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021463822710360932,
      "loss": 4.9014,
      "mean_token_accuracy": 0.24048465341329575,
      "num_tokens": 13156147.0,
      "step": 7235
    },
    {
      "entropy": 5.779071998596192,
      "epoch": 6.220025784271594,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021431799835102867,
      "loss": 4.9939,
      "mean_token_accuracy": 0.21844286173582078,
      "num_tokens": 13164588.0,
      "step": 7240
    },
    {
      "entropy": 5.685627555847168,
      "epoch": 6.224323162870649,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021399790210368524,
      "loss": 4.9544,
      "mean_token_accuracy": 0.23447279185056685,
      "num_tokens": 13174361.0,
      "step": 7245
    },
    {
      "entropy": 5.796992540359497,
      "epoch": 6.228620541469703,
      "grad_norm": 1.09375,
      "learning_rate": 0.00021367793906054133,
      "loss": 5.1382,
      "mean_token_accuracy": 0.21883737295866013,
      "num_tokens": 13185266.0,
      "step": 7250
    },
    {
      "entropy": 5.745621395111084,
      "epoch": 6.232917920068758,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021335810992026823,
      "loss": 4.9842,
      "mean_token_accuracy": 0.23670983910560608,
      "num_tokens": 13194227.0,
      "step": 7255
    },
    {
      "entropy": 5.774166917800903,
      "epoch": 6.237215298667812,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021303841538124497,
      "loss": 5.0301,
      "mean_token_accuracy": 0.22270644307136536,
      "num_tokens": 13202569.0,
      "step": 7260
    },
    {
      "entropy": 5.739467716217041,
      "epoch": 6.241512677266867,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021271885614155685,
      "loss": 4.9759,
      "mean_token_accuracy": 0.2304987356066704,
      "num_tokens": 13212201.0,
      "step": 7265
    },
    {
      "entropy": 5.677531290054321,
      "epoch": 6.245810055865922,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002123994328989932,
      "loss": 4.9028,
      "mean_token_accuracy": 0.23297007828950883,
      "num_tokens": 13220802.0,
      "step": 7270
    },
    {
      "entropy": 5.756556558609009,
      "epoch": 6.250107434464977,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021208014635104688,
      "loss": 5.0535,
      "mean_token_accuracy": 0.22601591944694518,
      "num_tokens": 13229519.0,
      "step": 7275
    },
    {
      "entropy": 5.749277067184448,
      "epoch": 6.254404813064031,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021176099719491209,
      "loss": 5.0007,
      "mean_token_accuracy": 0.2263491541147232,
      "num_tokens": 13238865.0,
      "step": 7280
    },
    {
      "entropy": 5.685416126251221,
      "epoch": 6.258702191663086,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021144198612748312,
      "loss": 4.9414,
      "mean_token_accuracy": 0.23242014199495314,
      "num_tokens": 13247259.0,
      "step": 7285
    },
    {
      "entropy": 5.764989757537842,
      "epoch": 6.26299957026214,
      "grad_norm": 1.1875,
      "learning_rate": 0.00021112311384535243,
      "loss": 5.0379,
      "mean_token_accuracy": 0.2274124652147293,
      "num_tokens": 13256692.0,
      "step": 7290
    },
    {
      "entropy": 5.769240140914917,
      "epoch": 6.267296948861195,
      "grad_norm": 1.140625,
      "learning_rate": 0.00021080438104480976,
      "loss": 4.9756,
      "mean_token_accuracy": 0.22864962965250016,
      "num_tokens": 13266109.0,
      "step": 7295
    },
    {
      "entropy": 5.7997815132141115,
      "epoch": 6.2715943274602495,
      "grad_norm": 1.125,
      "learning_rate": 0.00021048578842184019,
      "loss": 5.0686,
      "mean_token_accuracy": 0.2189205065369606,
      "num_tokens": 13275484.0,
      "step": 7300
    },
    {
      "entropy": 5.721390676498413,
      "epoch": 6.275891706059304,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021016733667212245,
      "loss": 4.9673,
      "mean_token_accuracy": 0.23466121554374694,
      "num_tokens": 13284755.0,
      "step": 7305
    },
    {
      "entropy": 5.712722826004028,
      "epoch": 6.280189084658359,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00020984902649102806,
      "loss": 4.9483,
      "mean_token_accuracy": 0.23197327852249144,
      "num_tokens": 13294386.0,
      "step": 7310
    },
    {
      "entropy": 5.741777181625366,
      "epoch": 6.284486463257413,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020953085857361924,
      "loss": 4.9594,
      "mean_token_accuracy": 0.23087407499551774,
      "num_tokens": 13303926.0,
      "step": 7315
    },
    {
      "entropy": 5.748653316497803,
      "epoch": 6.288783841856468,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020921283361464754,
      "loss": 5.0503,
      "mean_token_accuracy": 0.2205579772591591,
      "num_tokens": 13312727.0,
      "step": 7320
    },
    {
      "entropy": 5.742294883728027,
      "epoch": 6.293081220455522,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020889495230855232,
      "loss": 5.0249,
      "mean_token_accuracy": 0.22402707636356353,
      "num_tokens": 13321706.0,
      "step": 7325
    },
    {
      "entropy": 5.708799171447754,
      "epoch": 6.297378599054577,
      "grad_norm": 1.203125,
      "learning_rate": 0.00020857721534945923,
      "loss": 4.9556,
      "mean_token_accuracy": 0.24121423661708832,
      "num_tokens": 13330436.0,
      "step": 7330
    },
    {
      "entropy": 5.784104108810425,
      "epoch": 6.301675977653631,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002082596234311789,
      "loss": 5.0456,
      "mean_token_accuracy": 0.219855660200119,
      "num_tokens": 13339334.0,
      "step": 7335
    },
    {
      "entropy": 5.7998706817626955,
      "epoch": 6.305973356252686,
      "grad_norm": 1.25,
      "learning_rate": 0.0002079421772472051,
      "loss": 5.0971,
      "mean_token_accuracy": 0.21683503985404967,
      "num_tokens": 13348969.0,
      "step": 7340
    },
    {
      "entropy": 5.7793018341064455,
      "epoch": 6.31027073485174,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002076248774907134,
      "loss": 5.0597,
      "mean_token_accuracy": 0.22228462994098663,
      "num_tokens": 13358467.0,
      "step": 7345
    },
    {
      "entropy": 5.72367901802063,
      "epoch": 6.314568113450795,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020730772485455962,
      "loss": 4.8861,
      "mean_token_accuracy": 0.23987846970558166,
      "num_tokens": 13366413.0,
      "step": 7350
    },
    {
      "entropy": 5.775143098831177,
      "epoch": 6.3188654920498495,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002069907200312785,
      "loss": 5.0529,
      "mean_token_accuracy": 0.22475336343050004,
      "num_tokens": 13376620.0,
      "step": 7355
    },
    {
      "entropy": 5.733593416213989,
      "epoch": 6.323162870648904,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020667386371308162,
      "loss": 5.0338,
      "mean_token_accuracy": 0.22324397265911103,
      "num_tokens": 13385492.0,
      "step": 7360
    },
    {
      "entropy": 5.727029609680176,
      "epoch": 6.327460249247959,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00020635715659185673,
      "loss": 4.9713,
      "mean_token_accuracy": 0.22889943718910216,
      "num_tokens": 13395562.0,
      "step": 7365
    },
    {
      "entropy": 5.694593620300293,
      "epoch": 6.331757627847013,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020604059935916551,
      "loss": 4.9342,
      "mean_token_accuracy": 0.23273587673902513,
      "num_tokens": 13403357.0,
      "step": 7370
    },
    {
      "entropy": 5.74878888130188,
      "epoch": 6.336055006446068,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020572419270624255,
      "loss": 5.0243,
      "mean_token_accuracy": 0.22675324082374573,
      "num_tokens": 13412527.0,
      "step": 7375
    },
    {
      "entropy": 5.752114963531494,
      "epoch": 6.340352385045122,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020540793732399339,
      "loss": 5.089,
      "mean_token_accuracy": 0.21984241157770157,
      "num_tokens": 13422455.0,
      "step": 7380
    },
    {
      "entropy": 5.836174726486206,
      "epoch": 6.344649763644177,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020509183390299325,
      "loss": 5.1788,
      "mean_token_accuracy": 0.20899358093738557,
      "num_tokens": 13431677.0,
      "step": 7385
    },
    {
      "entropy": 5.69023060798645,
      "epoch": 6.348947142243231,
      "grad_norm": 1.21875,
      "learning_rate": 0.00020477588313348594,
      "loss": 4.889,
      "mean_token_accuracy": 0.24538520127534866,
      "num_tokens": 13440522.0,
      "step": 7390
    },
    {
      "entropy": 5.802966070175171,
      "epoch": 6.353244520842286,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020446008570538154,
      "loss": 5.0105,
      "mean_token_accuracy": 0.21989744454622268,
      "num_tokens": 13450021.0,
      "step": 7395
    },
    {
      "entropy": 5.722886657714843,
      "epoch": 6.35754189944134,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002041444423082554,
      "loss": 4.951,
      "mean_token_accuracy": 0.2331815704703331,
      "num_tokens": 13458115.0,
      "step": 7400
    },
    {
      "entropy": 5.771057939529419,
      "epoch": 6.361839278040395,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020382895363134652,
      "loss": 5.0203,
      "mean_token_accuracy": 0.21479823887348176,
      "num_tokens": 13466798.0,
      "step": 7405
    },
    {
      "entropy": 5.738609075546265,
      "epoch": 6.3661366566394495,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020351362036355602,
      "loss": 5.0325,
      "mean_token_accuracy": 0.2201032057404518,
      "num_tokens": 13476096.0,
      "step": 7410
    },
    {
      "entropy": 5.718414974212647,
      "epoch": 6.370434035238505,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002031984431934459,
      "loss": 4.9876,
      "mean_token_accuracy": 0.23637728095054628,
      "num_tokens": 13484601.0,
      "step": 7415
    },
    {
      "entropy": 5.77166337966919,
      "epoch": 6.3747314138375595,
      "grad_norm": 1.21875,
      "learning_rate": 0.00020288342280923695,
      "loss": 4.9674,
      "mean_token_accuracy": 0.2354439154267311,
      "num_tokens": 13493994.0,
      "step": 7420
    },
    {
      "entropy": 5.739843845367432,
      "epoch": 6.379028792436614,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020256855989880785,
      "loss": 5.0041,
      "mean_token_accuracy": 0.22339748293161393,
      "num_tokens": 13502890.0,
      "step": 7425
    },
    {
      "entropy": 5.659921026229858,
      "epoch": 6.383326171035669,
      "grad_norm": 1.25,
      "learning_rate": 0.00020225385514969336,
      "loss": 4.9686,
      "mean_token_accuracy": 0.23669744431972503,
      "num_tokens": 13512980.0,
      "step": 7430
    },
    {
      "entropy": 5.755312919616699,
      "epoch": 6.387623549634723,
      "grad_norm": 1.21875,
      "learning_rate": 0.00020193930924908277,
      "loss": 5.0504,
      "mean_token_accuracy": 0.22853563725948334,
      "num_tokens": 13521558.0,
      "step": 7435
    },
    {
      "entropy": 5.736734771728516,
      "epoch": 6.391920928233778,
      "grad_norm": 1.125,
      "learning_rate": 0.00020162492288381867,
      "loss": 4.9714,
      "mean_token_accuracy": 0.228364035487175,
      "num_tokens": 13531506.0,
      "step": 7440
    },
    {
      "entropy": 5.719764518737793,
      "epoch": 6.396218306832832,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002013106967403953,
      "loss": 4.9222,
      "mean_token_accuracy": 0.23592597395181655,
      "num_tokens": 13540559.0,
      "step": 7445
    },
    {
      "entropy": 5.787017345428467,
      "epoch": 6.400515685431887,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002009966315049569,
      "loss": 5.0318,
      "mean_token_accuracy": 0.22964833080768585,
      "num_tokens": 13550654.0,
      "step": 7450
    },
    {
      "entropy": 5.761588859558105,
      "epoch": 6.404813064030941,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002006827278632964,
      "loss": 5.0439,
      "mean_token_accuracy": 0.2232115238904953,
      "num_tokens": 13560708.0,
      "step": 7455
    },
    {
      "entropy": 5.755409479141235,
      "epoch": 6.409110442629996,
      "grad_norm": 1.203125,
      "learning_rate": 0.00020036898650085377,
      "loss": 4.9525,
      "mean_token_accuracy": 0.2230154424905777,
      "num_tokens": 13569330.0,
      "step": 7460
    },
    {
      "entropy": 5.803361701965332,
      "epoch": 6.41340782122905,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020005540810271493,
      "loss": 5.1119,
      "mean_token_accuracy": 0.2120390772819519,
      "num_tokens": 13577500.0,
      "step": 7465
    },
    {
      "entropy": 5.738029098510742,
      "epoch": 6.417705199828105,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019974199335360976,
      "loss": 5.0176,
      "mean_token_accuracy": 0.22733247876167298,
      "num_tokens": 13586087.0,
      "step": 7470
    },
    {
      "entropy": 5.715477800369262,
      "epoch": 6.4220025784271595,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019942874293791068,
      "loss": 4.9729,
      "mean_token_accuracy": 0.22815978080034255,
      "num_tokens": 13595346.0,
      "step": 7475
    },
    {
      "entropy": 5.7811461925506595,
      "epoch": 6.426299957026214,
      "grad_norm": 1.265625,
      "learning_rate": 0.00019911565753963145,
      "loss": 5.1105,
      "mean_token_accuracy": 0.21590404063463212,
      "num_tokens": 13604755.0,
      "step": 7480
    },
    {
      "entropy": 5.77233510017395,
      "epoch": 6.430597335625269,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001988027378424254,
      "loss": 5.0182,
      "mean_token_accuracy": 0.23077472299337387,
      "num_tokens": 13613860.0,
      "step": 7485
    },
    {
      "entropy": 5.686518621444702,
      "epoch": 6.434894714224323,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019848998452958429,
      "loss": 4.9077,
      "mean_token_accuracy": 0.23786175549030303,
      "num_tokens": 13622574.0,
      "step": 7490
    },
    {
      "entropy": 5.765685796737671,
      "epoch": 6.439192092823378,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019817739828403602,
      "loss": 5.0519,
      "mean_token_accuracy": 0.21631892621517182,
      "num_tokens": 13632366.0,
      "step": 7495
    },
    {
      "entropy": 5.826940727233887,
      "epoch": 6.443489471422432,
      "grad_norm": 1.15625,
      "learning_rate": 0.00019786497978834422,
      "loss": 4.9489,
      "mean_token_accuracy": 0.23657962381839753,
      "num_tokens": 13640682.0,
      "step": 7500
    },
    {
      "epoch": 6.443489471422432,
      "eval_entropy": 5.557100028605075,
      "eval_loss": 5.904560089111328,
      "eval_mean_token_accuracy": 0.18100960485570067,
      "eval_num_tokens": 13640682.0,
      "eval_runtime": 2.0529,
      "eval_samples_per_second": 1728.761,
      "eval_steps_per_second": 216.278,
      "step": 7500
    },
    {
      "entropy": 5.727175378799439,
      "epoch": 6.447786850021487,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019755272972470602,
      "loss": 5.0162,
      "mean_token_accuracy": 0.2280375748872757,
      "num_tokens": 13649675.0,
      "step": 7505
    },
    {
      "entropy": 5.728924798965454,
      "epoch": 6.452084228620541,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019724064877495057,
      "loss": 5.0136,
      "mean_token_accuracy": 0.22045981138944626,
      "num_tokens": 13658260.0,
      "step": 7510
    },
    {
      "entropy": 5.708987665176392,
      "epoch": 6.456381607219596,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019692873762053808,
      "loss": 4.9363,
      "mean_token_accuracy": 0.23248431831598282,
      "num_tokens": 13666571.0,
      "step": 7515
    },
    {
      "entropy": 5.81019959449768,
      "epoch": 6.46067898581865,
      "grad_norm": 1.109375,
      "learning_rate": 0.00019661699694255785,
      "loss": 5.0342,
      "mean_token_accuracy": 0.22640468925237656,
      "num_tokens": 13675707.0,
      "step": 7520
    },
    {
      "entropy": 5.823298740386963,
      "epoch": 6.464976364417705,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019630542742172692,
      "loss": 4.9734,
      "mean_token_accuracy": 0.23476444482803344,
      "num_tokens": 13684796.0,
      "step": 7525
    },
    {
      "entropy": 5.7843626022338865,
      "epoch": 6.4692737430167595,
      "grad_norm": 1.234375,
      "learning_rate": 0.00019599402973838854,
      "loss": 5.0717,
      "mean_token_accuracy": 0.22082927227020263,
      "num_tokens": 13693158.0,
      "step": 7530
    },
    {
      "entropy": 5.755879306793213,
      "epoch": 6.473571121615814,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001956828045725107,
      "loss": 5.0679,
      "mean_token_accuracy": 0.2204177588224411,
      "num_tokens": 13703521.0,
      "step": 7535
    },
    {
      "entropy": 5.725975036621094,
      "epoch": 6.477868500214869,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001953717526036849,
      "loss": 4.9523,
      "mean_token_accuracy": 0.23214234560728073,
      "num_tokens": 13712337.0,
      "step": 7540
    },
    {
      "entropy": 5.751292133331299,
      "epoch": 6.482165878813923,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019506087451112437,
      "loss": 4.9708,
      "mean_token_accuracy": 0.2247476503252983,
      "num_tokens": 13721605.0,
      "step": 7545
    },
    {
      "entropy": 5.700674152374267,
      "epoch": 6.486463257412978,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019475017097366244,
      "loss": 4.8844,
      "mean_token_accuracy": 0.23299427032470704,
      "num_tokens": 13730827.0,
      "step": 7550
    },
    {
      "entropy": 5.7830650806427,
      "epoch": 6.490760636012032,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019443964266975156,
      "loss": 5.0499,
      "mean_token_accuracy": 0.21963311433792115,
      "num_tokens": 13740128.0,
      "step": 7555
    },
    {
      "entropy": 5.730917930603027,
      "epoch": 6.495058014611088,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001941292902774614,
      "loss": 5.0172,
      "mean_token_accuracy": 0.2203517660498619,
      "num_tokens": 13748428.0,
      "step": 7560
    },
    {
      "entropy": 5.710373449325561,
      "epoch": 6.499355393210142,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019381911447447742,
      "loss": 5.0008,
      "mean_token_accuracy": 0.22881685942411423,
      "num_tokens": 13757109.0,
      "step": 7565
    },
    {
      "entropy": 5.804109764099121,
      "epoch": 6.503652771809197,
      "grad_norm": 1.375,
      "learning_rate": 0.00019350911593809977,
      "loss": 5.0596,
      "mean_token_accuracy": 0.2185147598385811,
      "num_tokens": 13766281.0,
      "step": 7570
    },
    {
      "entropy": 5.796328210830689,
      "epoch": 6.507950150408251,
      "grad_norm": 1.15625,
      "learning_rate": 0.00019319929534524128,
      "loss": 5.0527,
      "mean_token_accuracy": 0.21946843713521957,
      "num_tokens": 13775535.0,
      "step": 7575
    },
    {
      "entropy": 5.753546905517578,
      "epoch": 6.512247529007306,
      "grad_norm": 1.25,
      "learning_rate": 0.00019288965337242636,
      "loss": 5.0015,
      "mean_token_accuracy": 0.22280348986387252,
      "num_tokens": 13784099.0,
      "step": 7580
    },
    {
      "entropy": 5.834137105941773,
      "epoch": 6.51654490760636,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019258019069578924,
      "loss": 5.1504,
      "mean_token_accuracy": 0.21845682412385942,
      "num_tokens": 13793098.0,
      "step": 7585
    },
    {
      "entropy": 5.684184217453003,
      "epoch": 6.520842286205415,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019227090799107266,
      "loss": 4.9946,
      "mean_token_accuracy": 0.2286797657608986,
      "num_tokens": 13801847.0,
      "step": 7590
    },
    {
      "entropy": 5.70582799911499,
      "epoch": 6.5251396648044695,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001919618059336265,
      "loss": 4.8744,
      "mean_token_accuracy": 0.2416084885597229,
      "num_tokens": 13810599.0,
      "step": 7595
    },
    {
      "entropy": 5.78764591217041,
      "epoch": 6.529437043403524,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019165288519840617,
      "loss": 5.0296,
      "mean_token_accuracy": 0.2230227395892143,
      "num_tokens": 13819602.0,
      "step": 7600
    },
    {
      "entropy": 5.698198080062866,
      "epoch": 6.533734422002579,
      "grad_norm": 1.328125,
      "learning_rate": 0.000191344146459971,
      "loss": 4.961,
      "mean_token_accuracy": 0.22928732633590698,
      "num_tokens": 13828254.0,
      "step": 7605
    },
    {
      "entropy": 5.721844005584717,
      "epoch": 6.538031800601633,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00019103559039248302,
      "loss": 5.0155,
      "mean_token_accuracy": 0.2294679179787636,
      "num_tokens": 13837163.0,
      "step": 7610
    },
    {
      "entropy": 5.793222379684448,
      "epoch": 6.542329179200688,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001907272176697052,
      "loss": 5.0293,
      "mean_token_accuracy": 0.22645544707775117,
      "num_tokens": 13846373.0,
      "step": 7615
    },
    {
      "entropy": 5.677641916275024,
      "epoch": 6.546626557799742,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019041902896500059,
      "loss": 4.9707,
      "mean_token_accuracy": 0.23405378609895705,
      "num_tokens": 13855846.0,
      "step": 7620
    },
    {
      "entropy": 5.752673006057739,
      "epoch": 6.550923936398797,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019011102495132993,
      "loss": 4.9868,
      "mean_token_accuracy": 0.22904991656541823,
      "num_tokens": 13864723.0,
      "step": 7625
    },
    {
      "entropy": 5.830902004241944,
      "epoch": 6.555221314997851,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00018980320630125104,
      "loss": 5.1423,
      "mean_token_accuracy": 0.2158060148358345,
      "num_tokens": 13873418.0,
      "step": 7630
    },
    {
      "entropy": 5.79814453125,
      "epoch": 6.559518693596906,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018949557368691666,
      "loss": 5.0261,
      "mean_token_accuracy": 0.22538177222013472,
      "num_tokens": 13881890.0,
      "step": 7635
    },
    {
      "entropy": 5.767509126663208,
      "epoch": 6.56381607219596,
      "grad_norm": 1.171875,
      "learning_rate": 0.00018918812778007343,
      "loss": 5.0624,
      "mean_token_accuracy": 0.22678313702344893,
      "num_tokens": 13891289.0,
      "step": 7640
    },
    {
      "entropy": 5.741885089874268,
      "epoch": 6.568113450795015,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018888086925206054,
      "loss": 5.0,
      "mean_token_accuracy": 0.22480978667736054,
      "num_tokens": 13900344.0,
      "step": 7645
    },
    {
      "entropy": 5.779563140869141,
      "epoch": 6.5724108293940695,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018857379877380763,
      "loss": 5.0356,
      "mean_token_accuracy": 0.228300940990448,
      "num_tokens": 13909108.0,
      "step": 7650
    },
    {
      "entropy": 5.724088335037232,
      "epoch": 6.576708207993124,
      "grad_norm": 1.1875,
      "learning_rate": 0.00018826691701583404,
      "loss": 5.0442,
      "mean_token_accuracy": 0.22599031925201415,
      "num_tokens": 13918458.0,
      "step": 7655
    },
    {
      "entropy": 5.766824007034302,
      "epoch": 6.581005586592179,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018796022464824663,
      "loss": 4.9931,
      "mean_token_accuracy": 0.2276449218392372,
      "num_tokens": 13927186.0,
      "step": 7660
    },
    {
      "entropy": 5.812344980239868,
      "epoch": 6.585302965191233,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018765372234073912,
      "loss": 5.0941,
      "mean_token_accuracy": 0.21494560092687606,
      "num_tokens": 13936701.0,
      "step": 7665
    },
    {
      "entropy": 5.75049843788147,
      "epoch": 6.589600343790288,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018734741076259005,
      "loss": 5.1121,
      "mean_token_accuracy": 0.2118251845240593,
      "num_tokens": 13945842.0,
      "step": 7670
    },
    {
      "entropy": 5.690277910232544,
      "epoch": 6.593897722389342,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00018704129058266152,
      "loss": 4.9681,
      "mean_token_accuracy": 0.23356643468141555,
      "num_tokens": 13955675.0,
      "step": 7675
    },
    {
      "entropy": 5.818869304656983,
      "epoch": 6.598195100988397,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018673536246939743,
      "loss": 5.0863,
      "mean_token_accuracy": 0.2218109503388405,
      "num_tokens": 13964153.0,
      "step": 7680
    },
    {
      "entropy": 5.748479413986206,
      "epoch": 6.602492479587451,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018642962709082274,
      "loss": 5.0471,
      "mean_token_accuracy": 0.22789418399333955,
      "num_tokens": 13972141.0,
      "step": 7685
    },
    {
      "entropy": 5.7032506465911865,
      "epoch": 6.606789858186506,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018612408511454103,
      "loss": 4.9048,
      "mean_token_accuracy": 0.23088316321372987,
      "num_tokens": 13981369.0,
      "step": 7690
    },
    {
      "entropy": 5.7426300048828125,
      "epoch": 6.61108723678556,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018581873720773423,
      "loss": 4.9813,
      "mean_token_accuracy": 0.23071062564849854,
      "num_tokens": 13990626.0,
      "step": 7695
    },
    {
      "entropy": 5.743307399749756,
      "epoch": 6.615384615384615,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018551358403715989,
      "loss": 5.0658,
      "mean_token_accuracy": 0.21717096418142318,
      "num_tokens": 13999399.0,
      "step": 7700
    },
    {
      "entropy": 5.780385732650757,
      "epoch": 6.61968199398367,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018520862626915052,
      "loss": 4.9593,
      "mean_token_accuracy": 0.2303201287984848,
      "num_tokens": 14007487.0,
      "step": 7705
    },
    {
      "entropy": 5.758582687377929,
      "epoch": 6.623979372582724,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00018490386456961223,
      "loss": 4.9535,
      "mean_token_accuracy": 0.229754213988781,
      "num_tokens": 14016779.0,
      "step": 7710
    },
    {
      "entropy": 5.761456108093261,
      "epoch": 6.6282767511817795,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001845992996040224,
      "loss": 5.0385,
      "mean_token_accuracy": 0.22799862772226334,
      "num_tokens": 14025770.0,
      "step": 7715
    },
    {
      "entropy": 5.693759107589722,
      "epoch": 6.632574129780834,
      "grad_norm": 1.25,
      "learning_rate": 0.00018429493203742946,
      "loss": 4.9434,
      "mean_token_accuracy": 0.23615066856145858,
      "num_tokens": 14035304.0,
      "step": 7720
    },
    {
      "entropy": 5.743652820587158,
      "epoch": 6.636871508379889,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018399076253445052,
      "loss": 4.903,
      "mean_token_accuracy": 0.23021777719259262,
      "num_tokens": 14044058.0,
      "step": 7725
    },
    {
      "entropy": 5.745870208740234,
      "epoch": 6.641168886978943,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018368679175927012,
      "loss": 4.9386,
      "mean_token_accuracy": 0.238407401740551,
      "num_tokens": 14052709.0,
      "step": 7730
    },
    {
      "entropy": 5.683894968032837,
      "epoch": 6.645466265577998,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018338302037563885,
      "loss": 4.898,
      "mean_token_accuracy": 0.24091770648956298,
      "num_tokens": 14061346.0,
      "step": 7735
    },
    {
      "entropy": 5.7170616626739506,
      "epoch": 6.649763644177052,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018307944904687211,
      "loss": 4.9888,
      "mean_token_accuracy": 0.22879648357629775,
      "num_tokens": 14070409.0,
      "step": 7740
    },
    {
      "entropy": 5.774936103820801,
      "epoch": 6.654061022776107,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001827760784358483,
      "loss": 5.0723,
      "mean_token_accuracy": 0.21478112488985063,
      "num_tokens": 14079448.0,
      "step": 7745
    },
    {
      "entropy": 5.800827121734619,
      "epoch": 6.658358401375161,
      "grad_norm": 1.25,
      "learning_rate": 0.00018247290920500776,
      "loss": 5.0417,
      "mean_token_accuracy": 0.22413471043109895,
      "num_tokens": 14088452.0,
      "step": 7750
    },
    {
      "entropy": 5.783909034729004,
      "epoch": 6.662655779974216,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018216994201635062,
      "loss": 5.0219,
      "mean_token_accuracy": 0.2250862166285515,
      "num_tokens": 14097956.0,
      "step": 7755
    },
    {
      "entropy": 5.768687295913696,
      "epoch": 6.66695315857327,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018186717753143633,
      "loss": 4.9803,
      "mean_token_accuracy": 0.22687984108924866,
      "num_tokens": 14106771.0,
      "step": 7760
    },
    {
      "entropy": 5.759476661682129,
      "epoch": 6.671250537172325,
      "grad_norm": 1.171875,
      "learning_rate": 0.00018156461641138133,
      "loss": 5.047,
      "mean_token_accuracy": 0.2235451176762581,
      "num_tokens": 14115871.0,
      "step": 7765
    },
    {
      "entropy": 5.760025405883789,
      "epoch": 6.6755479157713795,
      "grad_norm": 1.203125,
      "learning_rate": 0.00018126225931685836,
      "loss": 4.9904,
      "mean_token_accuracy": 0.2259608507156372,
      "num_tokens": 14125316.0,
      "step": 7770
    },
    {
      "entropy": 5.67072057723999,
      "epoch": 6.679845294370434,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018096010690809444,
      "loss": 4.9147,
      "mean_token_accuracy": 0.23884862661361694,
      "num_tokens": 14134244.0,
      "step": 7775
    },
    {
      "entropy": 5.767531013488769,
      "epoch": 6.684142672969489,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018065815984486962,
      "loss": 5.0719,
      "mean_token_accuracy": 0.2180321991443634,
      "num_tokens": 14143600.0,
      "step": 7780
    },
    {
      "entropy": 5.758095407485962,
      "epoch": 6.688440051568543,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018035641878651548,
      "loss": 4.9592,
      "mean_token_accuracy": 0.22789613008499146,
      "num_tokens": 14152112.0,
      "step": 7785
    },
    {
      "entropy": 5.728698348999023,
      "epoch": 6.692737430167598,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018005488439191408,
      "loss": 4.9382,
      "mean_token_accuracy": 0.23120807409286498,
      "num_tokens": 14161044.0,
      "step": 7790
    },
    {
      "entropy": 5.755050468444824,
      "epoch": 6.697034808766652,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001797535573194959,
      "loss": 5.069,
      "mean_token_accuracy": 0.22951823174953462,
      "num_tokens": 14169629.0,
      "step": 7795
    },
    {
      "entropy": 5.732662105560303,
      "epoch": 6.701332187365707,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001794524382272389,
      "loss": 4.9929,
      "mean_token_accuracy": 0.22708707451820373,
      "num_tokens": 14179234.0,
      "step": 7800
    },
    {
      "entropy": 5.719165086746216,
      "epoch": 6.705629565964761,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001791515277726667,
      "loss": 5.0178,
      "mean_token_accuracy": 0.23272869884967803,
      "num_tokens": 14188887.0,
      "step": 7805
    },
    {
      "entropy": 5.685248327255249,
      "epoch": 6.709926944563816,
      "grad_norm": 1.234375,
      "learning_rate": 0.00017885082661284763,
      "loss": 4.9347,
      "mean_token_accuracy": 0.23434123396873474,
      "num_tokens": 14197731.0,
      "step": 7810
    },
    {
      "entropy": 5.723914051055909,
      "epoch": 6.71422432316287,
      "grad_norm": 1.171875,
      "learning_rate": 0.00017855033540439274,
      "loss": 5.0669,
      "mean_token_accuracy": 0.22158386409282685,
      "num_tokens": 14206851.0,
      "step": 7815
    },
    {
      "entropy": 5.7335141658782955,
      "epoch": 6.718521701761925,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017825005480345463,
      "loss": 5.0577,
      "mean_token_accuracy": 0.22551996409893035,
      "num_tokens": 14215879.0,
      "step": 7820
    },
    {
      "entropy": 5.736531066894531,
      "epoch": 6.7228190803609795,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017794998546572627,
      "loss": 4.9062,
      "mean_token_accuracy": 0.2371249318122864,
      "num_tokens": 14223798.0,
      "step": 7825
    },
    {
      "entropy": 5.791633462905883,
      "epoch": 6.727116458960034,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001776501280464391,
      "loss": 5.0048,
      "mean_token_accuracy": 0.22887726873159409,
      "num_tokens": 14233234.0,
      "step": 7830
    },
    {
      "entropy": 5.764397811889649,
      "epoch": 6.731413837559089,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017735048320036197,
      "loss": 4.9561,
      "mean_token_accuracy": 0.22443322390317916,
      "num_tokens": 14241851.0,
      "step": 7835
    },
    {
      "entropy": 5.752367305755615,
      "epoch": 6.735711216158143,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017705105158179917,
      "loss": 5.0901,
      "mean_token_accuracy": 0.2143208220601082,
      "num_tokens": 14251578.0,
      "step": 7840
    },
    {
      "entropy": 5.8044846534729,
      "epoch": 6.740008594757198,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00017675183384458987,
      "loss": 5.087,
      "mean_token_accuracy": 0.2191782310605049,
      "num_tokens": 14261122.0,
      "step": 7845
    },
    {
      "entropy": 5.775960302352905,
      "epoch": 6.744305973356253,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017645283064210616,
      "loss": 4.9647,
      "mean_token_accuracy": 0.2325132444500923,
      "num_tokens": 14270594.0,
      "step": 7850
    },
    {
      "entropy": 5.749928951263428,
      "epoch": 6.748603351955307,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017615404262725132,
      "loss": 5.0314,
      "mean_token_accuracy": 0.22753766030073166,
      "num_tokens": 14279646.0,
      "step": 7855
    },
    {
      "entropy": 5.734545373916626,
      "epoch": 6.752900730554362,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017585547045245885,
      "loss": 4.9699,
      "mean_token_accuracy": 0.23684488236904144,
      "num_tokens": 14288555.0,
      "step": 7860
    },
    {
      "entropy": 5.7363245487213135,
      "epoch": 6.757198109153417,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017555711476969138,
      "loss": 4.9907,
      "mean_token_accuracy": 0.22701586335897445,
      "num_tokens": 14297813.0,
      "step": 7865
    },
    {
      "entropy": 5.775456809997559,
      "epoch": 6.761495487752471,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017525897623043806,
      "loss": 4.9727,
      "mean_token_accuracy": 0.23112321645021439,
      "num_tokens": 14305779.0,
      "step": 7870
    },
    {
      "entropy": 5.800004673004151,
      "epoch": 6.765792866351526,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017496105548571472,
      "loss": 5.0746,
      "mean_token_accuracy": 0.21700311601161956,
      "num_tokens": 14314419.0,
      "step": 7875
    },
    {
      "entropy": 5.758483266830444,
      "epoch": 6.77009024495058,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017466335318606086,
      "loss": 5.0075,
      "mean_token_accuracy": 0.22559448331594467,
      "num_tokens": 14322959.0,
      "step": 7880
    },
    {
      "entropy": 5.755596733093261,
      "epoch": 6.774387623549635,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00017436586998153947,
      "loss": 4.9798,
      "mean_token_accuracy": 0.2367397665977478,
      "num_tokens": 14332270.0,
      "step": 7885
    },
    {
      "entropy": 5.720788049697876,
      "epoch": 6.7786850021486895,
      "grad_norm": 1.140625,
      "learning_rate": 0.00017406860652173495,
      "loss": 4.9967,
      "mean_token_accuracy": 0.2273873805999756,
      "num_tokens": 14341701.0,
      "step": 7890
    },
    {
      "entropy": 5.808590507507324,
      "epoch": 6.782982380747744,
      "grad_norm": 1.375,
      "learning_rate": 0.00017377156345575176,
      "loss": 5.0183,
      "mean_token_accuracy": 0.22628037482500077,
      "num_tokens": 14349551.0,
      "step": 7895
    },
    {
      "entropy": 5.682976245880127,
      "epoch": 6.787279759346799,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017347474143221338,
      "loss": 4.9242,
      "mean_token_accuracy": 0.23774727880954744,
      "num_tokens": 14358577.0,
      "step": 7900
    },
    {
      "entropy": 5.8283007621765135,
      "epoch": 6.791577137945853,
      "grad_norm": 1.234375,
      "learning_rate": 0.00017317814109926044,
      "loss": 5.0366,
      "mean_token_accuracy": 0.21837185621261596,
      "num_tokens": 14367862.0,
      "step": 7905
    },
    {
      "entropy": 5.667750120162964,
      "epoch": 6.795874516544908,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001728817631045495,
      "loss": 5.0011,
      "mean_token_accuracy": 0.2289063662290573,
      "num_tokens": 14377414.0,
      "step": 7910
    },
    {
      "entropy": 5.640456008911133,
      "epoch": 6.800171895143962,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001725856080952516,
      "loss": 4.9272,
      "mean_token_accuracy": 0.24358929991722106,
      "num_tokens": 14387239.0,
      "step": 7915
    },
    {
      "entropy": 5.670576524734497,
      "epoch": 6.804469273743017,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001722896767180509,
      "loss": 4.9289,
      "mean_token_accuracy": 0.23665995746850968,
      "num_tokens": 14396076.0,
      "step": 7920
    },
    {
      "entropy": 5.732269239425659,
      "epoch": 6.808766652342071,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017199396961914334,
      "loss": 5.0311,
      "mean_token_accuracy": 0.22220169007778168,
      "num_tokens": 14404982.0,
      "step": 7925
    },
    {
      "entropy": 5.715225648880005,
      "epoch": 6.813064030941126,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017169848744423506,
      "loss": 4.9637,
      "mean_token_accuracy": 0.2225647196173668,
      "num_tokens": 14413364.0,
      "step": 7930
    },
    {
      "entropy": 5.790431547164917,
      "epoch": 6.81736140954018,
      "grad_norm": 1.125,
      "learning_rate": 0.00017140323083854076,
      "loss": 5.0946,
      "mean_token_accuracy": 0.2200735867023468,
      "num_tokens": 14424279.0,
      "step": 7935
    },
    {
      "entropy": 5.755757904052734,
      "epoch": 6.821658788139235,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00017110820044678317,
      "loss": 5.0775,
      "mean_token_accuracy": 0.21869605481624604,
      "num_tokens": 14432931.0,
      "step": 7940
    },
    {
      "entropy": 5.736237001419068,
      "epoch": 6.8259561667382895,
      "grad_norm": 1.234375,
      "learning_rate": 0.00017081339691319054,
      "loss": 4.9942,
      "mean_token_accuracy": 0.22483924329280852,
      "num_tokens": 14442652.0,
      "step": 7945
    },
    {
      "entropy": 5.748126602172851,
      "epoch": 6.830253545337344,
      "grad_norm": 1.234375,
      "learning_rate": 0.00017051882088149612,
      "loss": 4.9811,
      "mean_token_accuracy": 0.22612235695123672,
      "num_tokens": 14452061.0,
      "step": 7950
    },
    {
      "entropy": 5.7769842624664305,
      "epoch": 6.834550923936399,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017022447299493599,
      "loss": 4.9711,
      "mean_token_accuracy": 0.22626492232084275,
      "num_tokens": 14460771.0,
      "step": 7955
    },
    {
      "entropy": 5.7788252353668215,
      "epoch": 6.838848302535453,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016993035389624854,
      "loss": 5.0858,
      "mean_token_accuracy": 0.21990473121404647,
      "num_tokens": 14469983.0,
      "step": 7960
    },
    {
      "entropy": 5.741082191467285,
      "epoch": 6.843145681134508,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001696364642276722,
      "loss": 4.9851,
      "mean_token_accuracy": 0.22639018148183823,
      "num_tokens": 14478641.0,
      "step": 7965
    },
    {
      "entropy": 5.7685614109039305,
      "epoch": 6.847443059733562,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00016934280463094448,
      "loss": 5.0548,
      "mean_token_accuracy": 0.22486061602830887,
      "num_tokens": 14487922.0,
      "step": 7970
    },
    {
      "entropy": 5.685476779937744,
      "epoch": 6.851740438332617,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016904937574730062,
      "loss": 4.8962,
      "mean_token_accuracy": 0.2398315191268921,
      "num_tokens": 14496259.0,
      "step": 7975
    },
    {
      "entropy": 5.743455696105957,
      "epoch": 6.856037816931671,
      "grad_norm": 1.5,
      "learning_rate": 0.00016875617821747208,
      "loss": 5.0165,
      "mean_token_accuracy": 0.22508815228939055,
      "num_tokens": 14504366.0,
      "step": 7980
    },
    {
      "entropy": 5.707944107055664,
      "epoch": 6.860335195530726,
      "grad_norm": 1.109375,
      "learning_rate": 0.00016846321268168508,
      "loss": 5.0198,
      "mean_token_accuracy": 0.22508743554353713,
      "num_tokens": 14513561.0,
      "step": 7985
    },
    {
      "entropy": 5.765905380249023,
      "epoch": 6.8646325741297805,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016817047977965905,
      "loss": 4.9763,
      "mean_token_accuracy": 0.23366623669862746,
      "num_tokens": 14522531.0,
      "step": 7990
    },
    {
      "entropy": 5.791642904281616,
      "epoch": 6.868929952728836,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001678779801506058,
      "loss": 5.012,
      "mean_token_accuracy": 0.2229594960808754,
      "num_tokens": 14531375.0,
      "step": 7995
    },
    {
      "entropy": 5.769171142578125,
      "epoch": 6.8732273313278895,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016758571443322774,
      "loss": 5.032,
      "mean_token_accuracy": 0.2259478285908699,
      "num_tokens": 14541081.0,
      "step": 8000
    },
    {
      "epoch": 6.8732273313278895,
      "eval_entropy": 5.577934592156796,
      "eval_loss": 5.886641502380371,
      "eval_mean_token_accuracy": 0.1824896966887487,
      "eval_num_tokens": 14541081.0,
      "eval_runtime": 2.0519,
      "eval_samples_per_second": 1729.627,
      "eval_steps_per_second": 216.386,
      "step": 8000
    },
    {
      "entropy": 5.687022161483765,
      "epoch": 6.877524709926945,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001672936832657162,
      "loss": 4.9468,
      "mean_token_accuracy": 0.23986830711364746,
      "num_tokens": 14550129.0,
      "step": 8005
    },
    {
      "entropy": 5.782396030426026,
      "epoch": 6.8818220885259995,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016700188728575047,
      "loss": 5.1143,
      "mean_token_accuracy": 0.21532667577266693,
      "num_tokens": 14559051.0,
      "step": 8010
    },
    {
      "entropy": 5.728854513168335,
      "epoch": 6.886119467125054,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016671032713049655,
      "loss": 5.032,
      "mean_token_accuracy": 0.22852370887994766,
      "num_tokens": 14567719.0,
      "step": 8015
    },
    {
      "entropy": 5.794173049926758,
      "epoch": 6.890416845724109,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016641900343660515,
      "loss": 5.0466,
      "mean_token_accuracy": 0.2290108770132065,
      "num_tokens": 14576256.0,
      "step": 8020
    },
    {
      "entropy": 5.784524774551391,
      "epoch": 6.894714224323163,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001661279168402107,
      "loss": 5.057,
      "mean_token_accuracy": 0.2221679225564003,
      "num_tokens": 14586392.0,
      "step": 8025
    },
    {
      "entropy": 5.757458209991455,
      "epoch": 6.899011602922218,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016583706797693008,
      "loss": 5.018,
      "mean_token_accuracy": 0.22664075791835786,
      "num_tokens": 14595448.0,
      "step": 8030
    },
    {
      "entropy": 5.768005990982056,
      "epoch": 6.903308981521272,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00016554645748186105,
      "loss": 5.0602,
      "mean_token_accuracy": 0.2278745874762535,
      "num_tokens": 14604242.0,
      "step": 8035
    },
    {
      "entropy": 5.827671146392822,
      "epoch": 6.907606360120327,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016525608598958063,
      "loss": 5.1147,
      "mean_token_accuracy": 0.21708249151706696,
      "num_tokens": 14614983.0,
      "step": 8040
    },
    {
      "entropy": 5.760879182815552,
      "epoch": 6.911903738719381,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016496595413414421,
      "loss": 5.0022,
      "mean_token_accuracy": 0.2230290189385414,
      "num_tokens": 14624748.0,
      "step": 8045
    },
    {
      "entropy": 5.73979549407959,
      "epoch": 6.916201117318436,
      "grad_norm": 1.140625,
      "learning_rate": 0.00016467606254908355,
      "loss": 4.9485,
      "mean_token_accuracy": 0.23578315526247023,
      "num_tokens": 14633642.0,
      "step": 8050
    },
    {
      "entropy": 5.775144433975219,
      "epoch": 6.9204984959174904,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016438641186740632,
      "loss": 5.0437,
      "mean_token_accuracy": 0.22425224334001542,
      "num_tokens": 14642549.0,
      "step": 8055
    },
    {
      "entropy": 5.746579980850219,
      "epoch": 6.924795874516545,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016409700272159371,
      "loss": 5.0538,
      "mean_token_accuracy": 0.2254525676369667,
      "num_tokens": 14651642.0,
      "step": 8060
    },
    {
      "entropy": 5.767870950698852,
      "epoch": 6.9290932531155995,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016380783574359957,
      "loss": 5.0124,
      "mean_token_accuracy": 0.22681328803300857,
      "num_tokens": 14661052.0,
      "step": 8065
    },
    {
      "entropy": 5.733950090408325,
      "epoch": 6.933390631714654,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001635189115648491,
      "loss": 5.0125,
      "mean_token_accuracy": 0.22621434628963472,
      "num_tokens": 14670292.0,
      "step": 8070
    },
    {
      "entropy": 5.69383544921875,
      "epoch": 6.937688010313709,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016323023081623705,
      "loss": 4.9481,
      "mean_token_accuracy": 0.23518468737602233,
      "num_tokens": 14679735.0,
      "step": 8075
    },
    {
      "entropy": 5.801571559906006,
      "epoch": 6.941985388912763,
      "grad_norm": 1.21875,
      "learning_rate": 0.00016294179412812702,
      "loss": 5.0666,
      "mean_token_accuracy": 0.21386227756738663,
      "num_tokens": 14688710.0,
      "step": 8080
    },
    {
      "entropy": 5.769373989105224,
      "epoch": 6.946282767511818,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016265360213034923,
      "loss": 5.1267,
      "mean_token_accuracy": 0.20773181766271592,
      "num_tokens": 14698523.0,
      "step": 8085
    },
    {
      "entropy": 5.71826205253601,
      "epoch": 6.950580146110872,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00016236565545220007,
      "loss": 5.0173,
      "mean_token_accuracy": 0.2249780148267746,
      "num_tokens": 14707674.0,
      "step": 8090
    },
    {
      "entropy": 5.745194053649902,
      "epoch": 6.954877524709927,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00016207795472243975,
      "loss": 4.9274,
      "mean_token_accuracy": 0.23496925532817842,
      "num_tokens": 14716600.0,
      "step": 8095
    },
    {
      "entropy": 5.830715799331665,
      "epoch": 6.959174903308981,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016179050056929173,
      "loss": 5.115,
      "mean_token_accuracy": 0.21429842114448547,
      "num_tokens": 14726112.0,
      "step": 8100
    },
    {
      "entropy": 5.752058267593384,
      "epoch": 6.963472281908036,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016150329362044102,
      "loss": 4.9985,
      "mean_token_accuracy": 0.2300025776028633,
      "num_tokens": 14735126.0,
      "step": 8105
    },
    {
      "entropy": 5.76392502784729,
      "epoch": 6.9677696605070905,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016121633450303285,
      "loss": 5.0026,
      "mean_token_accuracy": 0.22468472868204117,
      "num_tokens": 14744346.0,
      "step": 8110
    },
    {
      "entropy": 5.854573678970337,
      "epoch": 6.972067039106145,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00016092962384367122,
      "loss": 5.1148,
      "mean_token_accuracy": 0.22219625413417815,
      "num_tokens": 14753322.0,
      "step": 8115
    },
    {
      "entropy": 5.706526899337769,
      "epoch": 6.9763644177051995,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001606431622684176,
      "loss": 5.0008,
      "mean_token_accuracy": 0.232183800637722,
      "num_tokens": 14762384.0,
      "step": 8120
    },
    {
      "entropy": 5.743291187286377,
      "epoch": 6.980661796304254,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00016035695040278935,
      "loss": 5.0402,
      "mean_token_accuracy": 0.22681437730789183,
      "num_tokens": 14771451.0,
      "step": 8125
    },
    {
      "entropy": 5.808922386169433,
      "epoch": 6.984959174903309,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016007098887175914,
      "loss": 5.0685,
      "mean_token_accuracy": 0.2192844942212105,
      "num_tokens": 14780662.0,
      "step": 8130
    },
    {
      "entropy": 5.790175151824951,
      "epoch": 6.989256553502363,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015978527829975254,
      "loss": 4.983,
      "mean_token_accuracy": 0.22973400801420213,
      "num_tokens": 14789201.0,
      "step": 8135
    },
    {
      "entropy": 5.751619052886963,
      "epoch": 6.993553932101419,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015949981931064714,
      "loss": 5.0224,
      "mean_token_accuracy": 0.2253705859184265,
      "num_tokens": 14797857.0,
      "step": 8140
    },
    {
      "entropy": 5.768443250656128,
      "epoch": 6.997851310700472,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001592146125277714,
      "loss": 5.0127,
      "mean_token_accuracy": 0.22637296169996263,
      "num_tokens": 14806271.0,
      "step": 8145
    },
    {
      "entropy": 5.774147987365723,
      "epoch": 7.001718951439622,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00015892965857390278,
      "loss": 5.0911,
      "mean_token_accuracy": 0.21749180224206713,
      "num_tokens": 14815568.0,
      "step": 8150
    },
    {
      "entropy": 5.817372131347656,
      "epoch": 7.006016330038676,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015864495807126704,
      "loss": 4.9783,
      "mean_token_accuracy": 0.23229083567857742,
      "num_tokens": 14825140.0,
      "step": 8155
    },
    {
      "entropy": 5.7854362487792965,
      "epoch": 7.010313708637731,
      "grad_norm": 1.1875,
      "learning_rate": 0.00015836051164153602,
      "loss": 4.9856,
      "mean_token_accuracy": 0.23200100064277648,
      "num_tokens": 14834459.0,
      "step": 8160
    },
    {
      "entropy": 5.78501672744751,
      "epoch": 7.014611087236785,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015807631990582733,
      "loss": 4.9071,
      "mean_token_accuracy": 0.23723485916852952,
      "num_tokens": 14843632.0,
      "step": 8165
    },
    {
      "entropy": 5.789565229415894,
      "epoch": 7.01890846583584,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00015779238348470192,
      "loss": 4.8983,
      "mean_token_accuracy": 0.24426205605268478,
      "num_tokens": 14852626.0,
      "step": 8170
    },
    {
      "entropy": 5.664606523513794,
      "epoch": 7.023205844434894,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015750870299816345,
      "loss": 4.8615,
      "mean_token_accuracy": 0.23574336618185043,
      "num_tokens": 14861571.0,
      "step": 8175
    },
    {
      "entropy": 5.739161968231201,
      "epoch": 7.027503223033949,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015722527906565672,
      "loss": 4.8483,
      "mean_token_accuracy": 0.24561480730772017,
      "num_tokens": 14870383.0,
      "step": 8180
    },
    {
      "entropy": 5.741978025436401,
      "epoch": 7.0318006016330035,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015694211230606647,
      "loss": 4.9331,
      "mean_token_accuracy": 0.22312296479940413,
      "num_tokens": 14880212.0,
      "step": 8185
    },
    {
      "entropy": 5.733472061157227,
      "epoch": 7.036097980232058,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00015665920333771564,
      "loss": 4.9149,
      "mean_token_accuracy": 0.23446937501430512,
      "num_tokens": 14889347.0,
      "step": 8190
    },
    {
      "entropy": 5.761220645904541,
      "epoch": 7.0403953588311134,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00015637655277836427,
      "loss": 4.9316,
      "mean_token_accuracy": 0.22786078751087188,
      "num_tokens": 14898553.0,
      "step": 8195
    },
    {
      "entropy": 5.763401174545288,
      "epoch": 7.044692737430168,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001560941612452081,
      "loss": 4.9279,
      "mean_token_accuracy": 0.23226336091756822,
      "num_tokens": 14907275.0,
      "step": 8200
    },
    {
      "entropy": 5.747837781906128,
      "epoch": 7.0489901160292225,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001558120293548777,
      "loss": 4.9118,
      "mean_token_accuracy": 0.24116831719875337,
      "num_tokens": 14916409.0,
      "step": 8205
    },
    {
      "entropy": 5.794670104980469,
      "epoch": 7.053287494628277,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015553015772343614,
      "loss": 5.0002,
      "mean_token_accuracy": 0.22521151453256608,
      "num_tokens": 14927144.0,
      "step": 8210
    },
    {
      "entropy": 5.691853094100952,
      "epoch": 7.057584873227332,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015524854696637847,
      "loss": 4.835,
      "mean_token_accuracy": 0.2416848361492157,
      "num_tokens": 14936310.0,
      "step": 8215
    },
    {
      "entropy": 5.698910331726074,
      "epoch": 7.061882251826386,
      "grad_norm": 1.15625,
      "learning_rate": 0.00015496719769862981,
      "loss": 4.8624,
      "mean_token_accuracy": 0.2372149869799614,
      "num_tokens": 14945571.0,
      "step": 8220
    },
    {
      "entropy": 5.711551094055176,
      "epoch": 7.066179630425441,
      "grad_norm": 1.375,
      "learning_rate": 0.00015468611053454478,
      "loss": 4.8733,
      "mean_token_accuracy": 0.2346884161233902,
      "num_tokens": 14954586.0,
      "step": 8225
    },
    {
      "entropy": 5.747192287445069,
      "epoch": 7.070477009024495,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015440528608790533,
      "loss": 4.9477,
      "mean_token_accuracy": 0.2333863615989685,
      "num_tokens": 14963048.0,
      "step": 8230
    },
    {
      "entropy": 5.771088981628418,
      "epoch": 7.07477438762355,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001541247249719197,
      "loss": 4.9389,
      "mean_token_accuracy": 0.2317440077662468,
      "num_tokens": 14972158.0,
      "step": 8235
    },
    {
      "entropy": 5.768611907958984,
      "epoch": 7.079071766222604,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00015384442779922135,
      "loss": 4.8873,
      "mean_token_accuracy": 0.2359219416975975,
      "num_tokens": 14980869.0,
      "step": 8240
    },
    {
      "entropy": 5.763613080978393,
      "epoch": 7.083369144821659,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015356439518186726,
      "loss": 4.8693,
      "mean_token_accuracy": 0.234014393389225,
      "num_tokens": 14989103.0,
      "step": 8245
    },
    {
      "entropy": 5.719690227508545,
      "epoch": 7.0876665234207135,
      "grad_norm": 1.09375,
      "learning_rate": 0.00015328462773133672,
      "loss": 4.9012,
      "mean_token_accuracy": 0.24253213852643968,
      "num_tokens": 14999378.0,
      "step": 8250
    },
    {
      "entropy": 5.740370655059815,
      "epoch": 7.091963902019768,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015300512605852977,
      "loss": 4.982,
      "mean_token_accuracy": 0.2293758898973465,
      "num_tokens": 15007971.0,
      "step": 8255
    },
    {
      "entropy": 5.799237632751465,
      "epoch": 7.0962612806188226,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001527258907737668,
      "loss": 4.9721,
      "mean_token_accuracy": 0.22907347977161407,
      "num_tokens": 15017537.0,
      "step": 8260
    },
    {
      "entropy": 5.73674783706665,
      "epoch": 7.100558659217877,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015244692248678586,
      "loss": 4.8556,
      "mean_token_accuracy": 0.24554019421339035,
      "num_tokens": 15025684.0,
      "step": 8265
    },
    {
      "entropy": 5.751060104370117,
      "epoch": 7.104856037816932,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001521682218067421,
      "loss": 4.8882,
      "mean_token_accuracy": 0.2414063036441803,
      "num_tokens": 15034753.0,
      "step": 8270
    },
    {
      "entropy": 5.755550146102905,
      "epoch": 7.109153416415986,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015188978934220642,
      "loss": 4.9135,
      "mean_token_accuracy": 0.23815136551856994,
      "num_tokens": 15044685.0,
      "step": 8275
    },
    {
      "entropy": 5.79474835395813,
      "epoch": 7.113450795015041,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001516116257011641,
      "loss": 4.9627,
      "mean_token_accuracy": 0.23860295563936235,
      "num_tokens": 15054853.0,
      "step": 8280
    },
    {
      "entropy": 5.733561706542969,
      "epoch": 7.117748173614095,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001513337314910134,
      "loss": 4.9315,
      "mean_token_accuracy": 0.22673808336257933,
      "num_tokens": 15065244.0,
      "step": 8285
    },
    {
      "entropy": 5.687442588806152,
      "epoch": 7.12204555221315,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015105610731856416,
      "loss": 4.8645,
      "mean_token_accuracy": 0.24764200150966645,
      "num_tokens": 15074046.0,
      "step": 8290
    },
    {
      "entropy": 5.72289457321167,
      "epoch": 7.126342930812204,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015077875379003653,
      "loss": 4.9086,
      "mean_token_accuracy": 0.23524617105722428,
      "num_tokens": 15083518.0,
      "step": 8295
    },
    {
      "entropy": 5.751125144958496,
      "epoch": 7.130640309411259,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00015050167151105988,
      "loss": 4.9784,
      "mean_token_accuracy": 0.2345997780561447,
      "num_tokens": 15092512.0,
      "step": 8300
    },
    {
      "entropy": 5.805935955047607,
      "epoch": 7.1349376880103135,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000150224861086671,
      "loss": 4.9293,
      "mean_token_accuracy": 0.23530755788087845,
      "num_tokens": 15101722.0,
      "step": 8305
    },
    {
      "entropy": 5.757468891143799,
      "epoch": 7.139235066609368,
      "grad_norm": 1.234375,
      "learning_rate": 0.00014994832312131332,
      "loss": 4.8837,
      "mean_token_accuracy": 0.23100620359182358,
      "num_tokens": 15110114.0,
      "step": 8310
    },
    {
      "entropy": 5.739134931564331,
      "epoch": 7.143532445208423,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014967205821883532,
      "loss": 4.9683,
      "mean_token_accuracy": 0.2341601148247719,
      "num_tokens": 15119461.0,
      "step": 8315
    },
    {
      "entropy": 5.77350263595581,
      "epoch": 7.147829823807477,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000149396066982489,
      "loss": 4.9054,
      "mean_token_accuracy": 0.23744661062955857,
      "num_tokens": 15127518.0,
      "step": 8320
    },
    {
      "entropy": 5.745607709884643,
      "epoch": 7.152127202406532,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014912035001492897,
      "loss": 4.985,
      "mean_token_accuracy": 0.2232433095574379,
      "num_tokens": 15136741.0,
      "step": 8325
    },
    {
      "entropy": 5.759474182128907,
      "epoch": 7.156424581005586,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014884490791821058,
      "loss": 4.9243,
      "mean_token_accuracy": 0.22860487401485444,
      "num_tokens": 15145193.0,
      "step": 8330
    },
    {
      "entropy": 5.7424744129180905,
      "epoch": 7.160721959604641,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014856974129378981,
      "loss": 4.9284,
      "mean_token_accuracy": 0.23071371614933014,
      "num_tokens": 15154117.0,
      "step": 8335
    },
    {
      "entropy": 5.77603006362915,
      "epoch": 7.165019338203695,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001482948507425203,
      "loss": 4.9383,
      "mean_token_accuracy": 0.23303324580192566,
      "num_tokens": 15163221.0,
      "step": 8340
    },
    {
      "entropy": 5.763977718353272,
      "epoch": 7.169316716802751,
      "grad_norm": 1.1875,
      "learning_rate": 0.00014802023686465314,
      "loss": 4.9967,
      "mean_token_accuracy": 0.22445127367973328,
      "num_tokens": 15173234.0,
      "step": 8345
    },
    {
      "entropy": 5.731019926071167,
      "epoch": 7.173614095401805,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014774590025983523,
      "loss": 4.8845,
      "mean_token_accuracy": 0.2370949223637581,
      "num_tokens": 15181436.0,
      "step": 8350
    },
    {
      "entropy": 5.717350101470947,
      "epoch": 7.17791147400086,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014747184152710807,
      "loss": 4.9213,
      "mean_token_accuracy": 0.24155859798192977,
      "num_tokens": 15191697.0,
      "step": 8355
    },
    {
      "entropy": 5.7321617126464846,
      "epoch": 7.182208852599914,
      "grad_norm": 1.25,
      "learning_rate": 0.00014719806126490658,
      "loss": 4.9061,
      "mean_token_accuracy": 0.23953365683555602,
      "num_tokens": 15201563.0,
      "step": 8360
    },
    {
      "entropy": 5.721116971969605,
      "epoch": 7.186506231198969,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001469245600710573,
      "loss": 4.9142,
      "mean_token_accuracy": 0.23049890249967575,
      "num_tokens": 15210886.0,
      "step": 8365
    },
    {
      "entropy": 5.765688562393189,
      "epoch": 7.1908036097980235,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014665133854277742,
      "loss": 4.9404,
      "mean_token_accuracy": 0.23336558789014816,
      "num_tokens": 15219254.0,
      "step": 8370
    },
    {
      "entropy": 5.735818195343017,
      "epoch": 7.195100988397078,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001463783972766737,
      "loss": 4.9266,
      "mean_token_accuracy": 0.2335692301392555,
      "num_tokens": 15228117.0,
      "step": 8375
    },
    {
      "entropy": 5.662823820114136,
      "epoch": 7.199398366996133,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001461057368687407,
      "loss": 4.8986,
      "mean_token_accuracy": 0.23986562192440034,
      "num_tokens": 15236621.0,
      "step": 8380
    },
    {
      "entropy": 5.674626255035401,
      "epoch": 7.203695745595187,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014583335791435971,
      "loss": 4.8528,
      "mean_token_accuracy": 0.23885561674833297,
      "num_tokens": 15245487.0,
      "step": 8385
    },
    {
      "entropy": 5.70451512336731,
      "epoch": 7.207993124194242,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014556126100829774,
      "loss": 4.892,
      "mean_token_accuracy": 0.23383982479572296,
      "num_tokens": 15255321.0,
      "step": 8390
    },
    {
      "entropy": 5.693902921676636,
      "epoch": 7.212290502793296,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014528944674470546,
      "loss": 4.8329,
      "mean_token_accuracy": 0.24579360634088515,
      "num_tokens": 15264788.0,
      "step": 8395
    },
    {
      "entropy": 5.755379295349121,
      "epoch": 7.216587881392351,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001450179157171166,
      "loss": 4.9272,
      "mean_token_accuracy": 0.22800333052873611,
      "num_tokens": 15273448.0,
      "step": 8400
    },
    {
      "entropy": 5.779269552230835,
      "epoch": 7.220885259991405,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014474666851844632,
      "loss": 5.0026,
      "mean_token_accuracy": 0.22174182534217834,
      "num_tokens": 15283071.0,
      "step": 8405
    },
    {
      "entropy": 5.700813627243042,
      "epoch": 7.22518263859046,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014447570574099028,
      "loss": 4.8268,
      "mean_token_accuracy": 0.24628832787275315,
      "num_tokens": 15291537.0,
      "step": 8410
    },
    {
      "entropy": 5.798207378387451,
      "epoch": 7.229480017189514,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014420502797642283,
      "loss": 4.9036,
      "mean_token_accuracy": 0.2389177218079567,
      "num_tokens": 15300531.0,
      "step": 8415
    },
    {
      "entropy": 5.763958597183228,
      "epoch": 7.233777395788569,
      "grad_norm": 1.328125,
      "learning_rate": 0.000143934635815796,
      "loss": 4.9717,
      "mean_token_accuracy": 0.23086747974157334,
      "num_tokens": 15309820.0,
      "step": 8420
    },
    {
      "entropy": 5.7297776222229,
      "epoch": 7.2380747743876235,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001436645298495381,
      "loss": 4.94,
      "mean_token_accuracy": 0.23809971809387206,
      "num_tokens": 15318604.0,
      "step": 8425
    },
    {
      "entropy": 5.714357185363769,
      "epoch": 7.242372152986678,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014339471066745262,
      "loss": 4.9314,
      "mean_token_accuracy": 0.2272237628698349,
      "num_tokens": 15327737.0,
      "step": 8430
    },
    {
      "entropy": 5.724672031402588,
      "epoch": 7.246669531585733,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000143125178858717,
      "loss": 4.9833,
      "mean_token_accuracy": 0.23164504915475845,
      "num_tokens": 15336663.0,
      "step": 8435
    },
    {
      "entropy": 5.796184444427491,
      "epoch": 7.250966910184787,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014285593501188083,
      "loss": 4.9863,
      "mean_token_accuracy": 0.23062566369771959,
      "num_tokens": 15345278.0,
      "step": 8440
    },
    {
      "entropy": 5.771311235427857,
      "epoch": 7.255264288783842,
      "grad_norm": 1.21875,
      "learning_rate": 0.00014258697971486492,
      "loss": 4.9502,
      "mean_token_accuracy": 0.2327966347336769,
      "num_tokens": 15354230.0,
      "step": 8445
    },
    {
      "entropy": 5.756986951828003,
      "epoch": 7.259561667382896,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014231831355496045,
      "loss": 4.9571,
      "mean_token_accuracy": 0.23196402341127395,
      "num_tokens": 15362838.0,
      "step": 8450
    },
    {
      "entropy": 5.693323469161987,
      "epoch": 7.263859045981951,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00014204993711882662,
      "loss": 4.9305,
      "mean_token_accuracy": 0.23500606417655945,
      "num_tokens": 15372593.0,
      "step": 8455
    },
    {
      "entropy": 5.764274072647095,
      "epoch": 7.268156424581005,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001417818509924906,
      "loss": 4.9943,
      "mean_token_accuracy": 0.23025406748056412,
      "num_tokens": 15381945.0,
      "step": 8460
    },
    {
      "entropy": 5.704707050323487,
      "epoch": 7.27245380318006,
      "grad_norm": 1.1875,
      "learning_rate": 0.000141514055761345,
      "loss": 4.9009,
      "mean_token_accuracy": 0.23704942613840102,
      "num_tokens": 15391487.0,
      "step": 8465
    },
    {
      "entropy": 5.759627199172973,
      "epoch": 7.276751181779114,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014124655201014786,
      "loss": 4.8612,
      "mean_token_accuracy": 0.2423098489642143,
      "num_tokens": 15399891.0,
      "step": 8470
    },
    {
      "entropy": 5.754227495193481,
      "epoch": 7.281048560378169,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014097934032302037,
      "loss": 4.8634,
      "mean_token_accuracy": 0.237603360414505,
      "num_tokens": 15408693.0,
      "step": 8475
    },
    {
      "entropy": 5.696844053268433,
      "epoch": 7.2853459389772235,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014071242128344593,
      "loss": 4.9555,
      "mean_token_accuracy": 0.23301202952861785,
      "num_tokens": 15417779.0,
      "step": 8480
    },
    {
      "entropy": 5.71226954460144,
      "epoch": 7.289643317576278,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001404457954742691,
      "loss": 4.8754,
      "mean_token_accuracy": 0.24334413707256317,
      "num_tokens": 15425826.0,
      "step": 8485
    },
    {
      "entropy": 5.732396030426026,
      "epoch": 7.2939406961753335,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014017946347769423,
      "loss": 4.9541,
      "mean_token_accuracy": 0.23008962273597716,
      "num_tokens": 15435811.0,
      "step": 8490
    },
    {
      "entropy": 5.693571281433106,
      "epoch": 7.298238074774388,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013991342587528377,
      "loss": 4.8234,
      "mean_token_accuracy": 0.24324662387371063,
      "num_tokens": 15444949.0,
      "step": 8495
    },
    {
      "entropy": 5.704442405700684,
      "epoch": 7.302535453373443,
      "grad_norm": 1.28125,
      "learning_rate": 0.00013964768324795752,
      "loss": 4.8677,
      "mean_token_accuracy": 0.24231224954128266,
      "num_tokens": 15453398.0,
      "step": 8500
    },
    {
      "epoch": 7.302535453373443,
      "eval_entropy": 5.546972805852288,
      "eval_loss": 5.898355960845947,
      "eval_mean_token_accuracy": 0.1819473335032796,
      "eval_num_tokens": 15453398.0,
      "eval_runtime": 2.0519,
      "eval_samples_per_second": 1729.602,
      "eval_steps_per_second": 216.383,
      "step": 8500
    },
    {
      "entropy": 5.66387529373169,
      "epoch": 7.306832831972497,
      "grad_norm": 1.203125,
      "learning_rate": 0.00013938223617599124,
      "loss": 4.9421,
      "mean_token_accuracy": 0.23342001885175706,
      "num_tokens": 15462785.0,
      "step": 8505
    },
    {
      "entropy": 5.74355845451355,
      "epoch": 7.311130210571552,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013911708523901514,
      "loss": 4.9755,
      "mean_token_accuracy": 0.23636211156845094,
      "num_tokens": 15471718.0,
      "step": 8510
    },
    {
      "entropy": 5.763104438781738,
      "epoch": 7.315427589170606,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00013885223101601303,
      "loss": 4.8955,
      "mean_token_accuracy": 0.23637697100639343,
      "num_tokens": 15480204.0,
      "step": 8515
    },
    {
      "entropy": 5.71501669883728,
      "epoch": 7.319724967769661,
      "grad_norm": 1.265625,
      "learning_rate": 0.00013858767408532051,
      "loss": 4.8816,
      "mean_token_accuracy": 0.24099365025758743,
      "num_tokens": 15489388.0,
      "step": 8520
    },
    {
      "entropy": 5.723988056182861,
      "epoch": 7.324022346368715,
      "grad_norm": 1.375,
      "learning_rate": 0.00013832341502462432,
      "loss": 4.8774,
      "mean_token_accuracy": 0.23590993583202363,
      "num_tokens": 15498028.0,
      "step": 8525
    },
    {
      "entropy": 5.737170457839966,
      "epoch": 7.32831972496777,
      "grad_norm": 1.46875,
      "learning_rate": 0.00013805945441096057,
      "loss": 4.9277,
      "mean_token_accuracy": 0.23668332844972612,
      "num_tokens": 15506382.0,
      "step": 8530
    },
    {
      "entropy": 5.7540687084197994,
      "epoch": 7.332617103566824,
      "grad_norm": 1.375,
      "learning_rate": 0.00013779579282071364,
      "loss": 4.9306,
      "mean_token_accuracy": 0.2349459186196327,
      "num_tokens": 15515271.0,
      "step": 8535
    },
    {
      "entropy": 5.676801681518555,
      "epoch": 7.336914482165879,
      "grad_norm": 1.28125,
      "learning_rate": 0.00013753243082961512,
      "loss": 4.8768,
      "mean_token_accuracy": 0.24011008739471434,
      "num_tokens": 15524396.0,
      "step": 8540
    },
    {
      "entropy": 5.68070273399353,
      "epoch": 7.3412118607649335,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013726936901274246,
      "loss": 4.8393,
      "mean_token_accuracy": 0.24872156083583832,
      "num_tokens": 15532829.0,
      "step": 8545
    },
    {
      "entropy": 5.800928020477295,
      "epoch": 7.345509239363988,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001370066079445174,
      "loss": 4.9704,
      "mean_token_accuracy": 0.23076788932085038,
      "num_tokens": 15541726.0,
      "step": 8550
    },
    {
      "entropy": 5.79030556678772,
      "epoch": 7.349806617963043,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00013674414819870502,
      "loss": 5.0745,
      "mean_token_accuracy": 0.22369800060987471,
      "num_tokens": 15551539.0,
      "step": 8555
    },
    {
      "entropy": 5.730133438110352,
      "epoch": 7.354103996562097,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013648199034841264,
      "loss": 4.92,
      "mean_token_accuracy": 0.2387056604027748,
      "num_tokens": 15560147.0,
      "step": 8560
    },
    {
      "entropy": 5.69276180267334,
      "epoch": 7.358401375161152,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001362201349660882,
      "loss": 4.8912,
      "mean_token_accuracy": 0.24466140568256378,
      "num_tokens": 15568983.0,
      "step": 8565
    },
    {
      "entropy": 5.666532707214356,
      "epoch": 7.362698753760206,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001359585826235192,
      "loss": 4.8956,
      "mean_token_accuracy": 0.24125270992517472,
      "num_tokens": 15578065.0,
      "step": 8570
    },
    {
      "entropy": 5.799620962142944,
      "epoch": 7.366996132359261,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013569733389183126,
      "loss": 4.9952,
      "mean_token_accuracy": 0.22822352200746537,
      "num_tokens": 15587181.0,
      "step": 8575
    },
    {
      "entropy": 5.758121490478516,
      "epoch": 7.371293510958315,
      "grad_norm": 1.234375,
      "learning_rate": 0.00013543638934148736,
      "loss": 4.9296,
      "mean_token_accuracy": 0.23435905277729036,
      "num_tokens": 15596602.0,
      "step": 8580
    },
    {
      "entropy": 5.814731168746948,
      "epoch": 7.37559088955737,
      "grad_norm": 1.359375,
      "learning_rate": 0.000135175749542286,
      "loss": 4.9821,
      "mean_token_accuracy": 0.22124236524105073,
      "num_tokens": 15605857.0,
      "step": 8585
    },
    {
      "entropy": 5.67050518989563,
      "epoch": 7.379888268156424,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001349154150633604,
      "loss": 4.8697,
      "mean_token_accuracy": 0.2459047943353653,
      "num_tokens": 15615320.0,
      "step": 8590
    },
    {
      "entropy": 5.741513633728028,
      "epoch": 7.384185646755479,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000134655386473177,
      "loss": 5.0171,
      "mean_token_accuracy": 0.22208292037248611,
      "num_tokens": 15624193.0,
      "step": 8595
    },
    {
      "entropy": 5.637994432449341,
      "epoch": 7.3884830253545335,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013439566433953427,
      "loss": 4.892,
      "mean_token_accuracy": 0.2412777841091156,
      "num_tokens": 15632924.0,
      "step": 8600
    },
    {
      "entropy": 5.720129156112671,
      "epoch": 7.392780403953588,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001341362492295616,
      "loss": 4.89,
      "mean_token_accuracy": 0.2424454867839813,
      "num_tokens": 15642201.0,
      "step": 8605
    },
    {
      "entropy": 5.777584362030029,
      "epoch": 7.397077782552643,
      "grad_norm": 1.46875,
      "learning_rate": 0.00013387714170971776,
      "loss": 4.9287,
      "mean_token_accuracy": 0.22944550216197968,
      "num_tokens": 15651608.0,
      "step": 8610
    },
    {
      "entropy": 5.785937261581421,
      "epoch": 7.401375161151697,
      "grad_norm": 1.265625,
      "learning_rate": 0.00013361834234579012,
      "loss": 5.0178,
      "mean_token_accuracy": 0.22616576850414277,
      "num_tokens": 15661768.0,
      "step": 8615
    },
    {
      "entropy": 5.6717626571655275,
      "epoch": 7.405672539750752,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001333598517028931,
      "loss": 4.9006,
      "mean_token_accuracy": 0.23272989243268966,
      "num_tokens": 15670270.0,
      "step": 8620
    },
    {
      "entropy": 5.688118124008179,
      "epoch": 7.409969918349806,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013310167034546688,
      "loss": 4.887,
      "mean_token_accuracy": 0.2374843627214432,
      "num_tokens": 15679587.0,
      "step": 8625
    },
    {
      "entropy": 5.749850463867188,
      "epoch": 7.414267296948861,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001328437988372763,
      "loss": 4.9595,
      "mean_token_accuracy": 0.22750054448843002,
      "num_tokens": 15688838.0,
      "step": 8630
    },
    {
      "entropy": 5.756968069076538,
      "epoch": 7.418564675547916,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013258623774140967,
      "loss": 4.9342,
      "mean_token_accuracy": 0.2360052227973938,
      "num_tokens": 15697744.0,
      "step": 8635
    },
    {
      "entropy": 5.706060266494751,
      "epoch": 7.422862054146971,
      "grad_norm": 1.375,
      "learning_rate": 0.00013232898762027766,
      "loss": 4.8852,
      "mean_token_accuracy": 0.2418963521718979,
      "num_tokens": 15707643.0,
      "step": 8640
    },
    {
      "entropy": 5.772034740447998,
      "epoch": 7.427159432746025,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00013207204903561154,
      "loss": 4.978,
      "mean_token_accuracy": 0.23209577053785324,
      "num_tokens": 15717568.0,
      "step": 8645
    },
    {
      "entropy": 5.702054405212403,
      "epoch": 7.43145681134508,
      "grad_norm": 1.234375,
      "learning_rate": 0.00013181542254846247,
      "loss": 4.8473,
      "mean_token_accuracy": 0.2422897547483444,
      "num_tokens": 15726467.0,
      "step": 8650
    },
    {
      "entropy": 5.75566291809082,
      "epoch": 7.435754189944134,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001315591087192002,
      "loss": 4.9831,
      "mean_token_accuracy": 0.2323268622159958,
      "num_tokens": 15736533.0,
      "step": 8655
    },
    {
      "entropy": 5.655916357040406,
      "epoch": 7.440051568543189,
      "grad_norm": 1.234375,
      "learning_rate": 0.00013130310810751162,
      "loss": 4.8862,
      "mean_token_accuracy": 0.23978102058172227,
      "num_tokens": 15745853.0,
      "step": 8660
    },
    {
      "entropy": 5.784155988693238,
      "epoch": 7.4443489471422435,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013104742127239983,
      "loss": 5.016,
      "mean_token_accuracy": 0.22613199800252914,
      "num_tokens": 15755534.0,
      "step": 8665
    },
    {
      "entropy": 5.7693627834320065,
      "epoch": 7.448646325741298,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001307920487721826,
      "loss": 4.9986,
      "mean_token_accuracy": 0.22481437474489213,
      "num_tokens": 15766182.0,
      "step": 8670
    },
    {
      "entropy": 5.737584114074707,
      "epoch": 7.452943704340353,
      "grad_norm": 1.375,
      "learning_rate": 0.00013053699116449144,
      "loss": 4.9084,
      "mean_token_accuracy": 0.22960878312587737,
      "num_tokens": 15775454.0,
      "step": 8675
    },
    {
      "entropy": 5.89331545829773,
      "epoch": 7.457241082939407,
      "grad_norm": 1.203125,
      "learning_rate": 0.00013028224900627026,
      "loss": 5.0301,
      "mean_token_accuracy": 0.21912138164043427,
      "num_tokens": 15784768.0,
      "step": 8680
    },
    {
      "entropy": 5.803331756591797,
      "epoch": 7.461538461538462,
      "grad_norm": 1.0625,
      "learning_rate": 0.00013002782285377395,
      "loss": 4.9622,
      "mean_token_accuracy": 0.2376553386449814,
      "num_tokens": 15794255.0,
      "step": 8685
    },
    {
      "entropy": 5.750441551208496,
      "epoch": 7.465835840137516,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001297737132625677,
      "loss": 4.9946,
      "mean_token_accuracy": 0.2235792651772499,
      "num_tokens": 15803722.0,
      "step": 8690
    },
    {
      "entropy": 5.789198493957519,
      "epoch": 7.470133218736571,
      "grad_norm": 1.5,
      "learning_rate": 0.00012951992078752528,
      "loss": 4.9474,
      "mean_token_accuracy": 0.2321646735072136,
      "num_tokens": 15811819.0,
      "step": 8695
    },
    {
      "entropy": 5.744178915023804,
      "epoch": 7.474430597335625,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00012926644598282798,
      "loss": 4.9459,
      "mean_token_accuracy": 0.22402314394712447,
      "num_tokens": 15821446.0,
      "step": 8700
    },
    {
      "entropy": 5.74596061706543,
      "epoch": 7.47872797593468,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001290132894019634,
      "loss": 4.9661,
      "mean_token_accuracy": 0.23607563078403473,
      "num_tokens": 15830585.0,
      "step": 8705
    },
    {
      "entropy": 5.778988885879516,
      "epoch": 7.483025354533734,
      "grad_norm": 1.28125,
      "learning_rate": 0.00012876045159772442,
      "loss": 4.9489,
      "mean_token_accuracy": 0.23937921822071076,
      "num_tokens": 15838872.0,
      "step": 8710
    },
    {
      "entropy": 5.773708295822144,
      "epoch": 7.487322733132789,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012850793312220766,
      "loss": 4.9283,
      "mean_token_accuracy": 0.23426500260829924,
      "num_tokens": 15847561.0,
      "step": 8715
    },
    {
      "entropy": 5.674078750610351,
      "epoch": 7.4916201117318435,
      "grad_norm": 1.25,
      "learning_rate": 0.00012825573452681266,
      "loss": 4.8773,
      "mean_token_accuracy": 0.24968714416027069,
      "num_tokens": 15856405.0,
      "step": 8720
    },
    {
      "entropy": 5.735969209671021,
      "epoch": 7.495917490330898,
      "grad_norm": 1.375,
      "learning_rate": 0.00012800385636224017,
      "loss": 4.9358,
      "mean_token_accuracy": 0.23501814007759095,
      "num_tokens": 15865856.0,
      "step": 8725
    },
    {
      "entropy": 5.747404956817627,
      "epoch": 7.500214868929953,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012775229917849162,
      "loss": 4.9054,
      "mean_token_accuracy": 0.23629093021154404,
      "num_tokens": 15873605.0,
      "step": 8730
    },
    {
      "entropy": 5.761697673797608,
      "epoch": 7.504512247529007,
      "grad_norm": 1.234375,
      "learning_rate": 0.00012750106352486728,
      "loss": 4.8979,
      "mean_token_accuracy": 0.2428216427564621,
      "num_tokens": 15883123.0,
      "step": 8735
    },
    {
      "entropy": 5.700611209869384,
      "epoch": 7.508809626128062,
      "grad_norm": 1.171875,
      "learning_rate": 0.00012725014994996534,
      "loss": 4.9339,
      "mean_token_accuracy": 0.24504542350769043,
      "num_tokens": 15892713.0,
      "step": 8740
    },
    {
      "entropy": 5.807457494735718,
      "epoch": 7.513107004727116,
      "grad_norm": 1.25,
      "learning_rate": 0.00012699955900168075,
      "loss": 4.9282,
      "mean_token_accuracy": 0.229953795671463,
      "num_tokens": 15902913.0,
      "step": 8745
    },
    {
      "entropy": 5.729871797561645,
      "epoch": 7.517404383326171,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012674929122720414,
      "loss": 4.9777,
      "mean_token_accuracy": 0.2232517972588539,
      "num_tokens": 15912721.0,
      "step": 8750
    },
    {
      "entropy": 5.635332775115967,
      "epoch": 7.521701761925225,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001264993471730202,
      "loss": 4.8575,
      "mean_token_accuracy": 0.24274646639823913,
      "num_tokens": 15921520.0,
      "step": 8755
    },
    {
      "entropy": 5.737993383407593,
      "epoch": 7.52599914052428,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012624972738490675,
      "loss": 5.0013,
      "mean_token_accuracy": 0.22250534743070602,
      "num_tokens": 15930753.0,
      "step": 8760
    },
    {
      "entropy": 5.810367679595947,
      "epoch": 7.530296519123334,
      "grad_norm": 1.21875,
      "learning_rate": 0.00012600043240793368,
      "loss": 4.9765,
      "mean_token_accuracy": 0.2321833461523056,
      "num_tokens": 15939957.0,
      "step": 8765
    },
    {
      "entropy": 5.74908938407898,
      "epoch": 7.534593897722389,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012575146278646175,
      "loss": 4.8922,
      "mean_token_accuracy": 0.23532313406467437,
      "num_tokens": 15949555.0,
      "step": 8770
    },
    {
      "entropy": 5.7044130802154545,
      "epoch": 7.5388912763214435,
      "grad_norm": 1.21875,
      "learning_rate": 0.00012550281906414097,
      "loss": 4.9005,
      "mean_token_accuracy": 0.23530834317207336,
      "num_tokens": 15958395.0,
      "step": 8775
    },
    {
      "entropy": 5.754625558853149,
      "epoch": 7.543188654920499,
      "grad_norm": 1.234375,
      "learning_rate": 0.00012525450178390972,
      "loss": 4.9416,
      "mean_token_accuracy": 0.2437831625342369,
      "num_tokens": 15967522.0,
      "step": 8780
    },
    {
      "entropy": 5.779452610015869,
      "epoch": 7.547486033519553,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001250065114879939,
      "loss": 4.9461,
      "mean_token_accuracy": 0.23520458936691285,
      "num_tokens": 15976311.0,
      "step": 8785
    },
    {
      "entropy": 5.705014038085937,
      "epoch": 7.551783412118608,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00012475884871790505,
      "loss": 4.8667,
      "mean_token_accuracy": 0.2412562444806099,
      "num_tokens": 15985202.0,
      "step": 8790
    },
    {
      "entropy": 5.8083795547485355,
      "epoch": 7.556080790717663,
      "grad_norm": 1.296875,
      "learning_rate": 0.00012451151401443982,
      "loss": 5.0036,
      "mean_token_accuracy": 0.22445941120386123,
      "num_tokens": 15995043.0,
      "step": 8795
    },
    {
      "entropy": 5.658902978897094,
      "epoch": 7.560378169316717,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012426450791767815,
      "loss": 4.8297,
      "mean_token_accuracy": 0.24630579203367234,
      "num_tokens": 16004355.0,
      "step": 8800
    },
    {
      "entropy": 5.727620363235474,
      "epoch": 7.564675547915772,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012401783096698283,
      "loss": 4.7861,
      "mean_token_accuracy": 0.24524826556444168,
      "num_tokens": 16013069.0,
      "step": 8805
    },
    {
      "entropy": 5.739884567260742,
      "epoch": 7.568972926514826,
      "grad_norm": 1.203125,
      "learning_rate": 0.00012377148370099764,
      "loss": 4.9565,
      "mean_token_accuracy": 0.22828071862459182,
      "num_tokens": 16023757.0,
      "step": 8810
    },
    {
      "entropy": 5.752638149261474,
      "epoch": 7.573270305113881,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012352546665764642,
      "loss": 4.9625,
      "mean_token_accuracy": 0.22861632257699965,
      "num_tokens": 16032550.0,
      "step": 8815
    },
    {
      "entropy": 5.631821250915527,
      "epoch": 7.577567683712935,
      "grad_norm": 1.25,
      "learning_rate": 0.00012327978037413219,
      "loss": 4.8393,
      "mean_token_accuracy": 0.2455634668469429,
      "num_tokens": 16041580.0,
      "step": 8820
    },
    {
      "entropy": 5.749710655212402,
      "epoch": 7.58186506231199,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012303442538693564,
      "loss": 4.9425,
      "mean_token_accuracy": 0.23156363964080812,
      "num_tokens": 16049845.0,
      "step": 8825
    },
    {
      "entropy": 5.689564323425293,
      "epoch": 7.586162440911044,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012278940223181393,
      "loss": 4.8338,
      "mean_token_accuracy": 0.23270008265972136,
      "num_tokens": 16059703.0,
      "step": 8830
    },
    {
      "entropy": 5.725271081924438,
      "epoch": 7.590459819510099,
      "grad_norm": 1.25,
      "learning_rate": 0.00012254471144379964,
      "loss": 4.8128,
      "mean_token_accuracy": 0.25256053656339644,
      "num_tokens": 16068416.0,
      "step": 8835
    },
    {
      "entropy": 5.727273273468017,
      "epoch": 7.5947571981091535,
      "grad_norm": 1.21875,
      "learning_rate": 0.00012230035355719968,
      "loss": 4.9651,
      "mean_token_accuracy": 0.23480646908283234,
      "num_tokens": 16078067.0,
      "step": 8840
    },
    {
      "entropy": 5.772902631759644,
      "epoch": 7.599054576708208,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001220563291055941,
      "loss": 4.9437,
      "mean_token_accuracy": 0.23261293321847915,
      "num_tokens": 16086591.0,
      "step": 8845
    },
    {
      "entropy": 5.765447664260864,
      "epoch": 7.603351955307263,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001218126386218347,
      "loss": 4.9353,
      "mean_token_accuracy": 0.23733194619417192,
      "num_tokens": 16096138.0,
      "step": 8850
    },
    {
      "entropy": 5.6730225563049315,
      "epoch": 7.607649333906317,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012156928263804403,
      "loss": 4.8637,
      "mean_token_accuracy": 0.24171005487442015,
      "num_tokens": 16105182.0,
      "step": 8855
    },
    {
      "entropy": 5.743314933776856,
      "epoch": 7.611946712505372,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001213262616856144,
      "loss": 5.0012,
      "mean_token_accuracy": 0.2306993141770363,
      "num_tokens": 16113940.0,
      "step": 8860
    },
    {
      "entropy": 5.747490549087525,
      "epoch": 7.616244091104426,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012108357629520635,
      "loss": 4.9057,
      "mean_token_accuracy": 0.2387942299246788,
      "num_tokens": 16123036.0,
      "step": 8865
    },
    {
      "entropy": 5.7210612297058105,
      "epoch": 7.620541469703481,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012084122699674785,
      "loss": 4.8936,
      "mean_token_accuracy": 0.23947659581899644,
      "num_tokens": 16131057.0,
      "step": 8870
    },
    {
      "entropy": 5.73658299446106,
      "epoch": 7.624838848302535,
      "grad_norm": 1.375,
      "learning_rate": 0.00012059921431943278,
      "loss": 4.943,
      "mean_token_accuracy": 0.22805553525686265,
      "num_tokens": 16140259.0,
      "step": 8875
    },
    {
      "entropy": 5.778007841110229,
      "epoch": 7.62913622690159,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012035753879172026,
      "loss": 5.1314,
      "mean_token_accuracy": 0.21736743450164794,
      "num_tokens": 16149585.0,
      "step": 8880
    },
    {
      "entropy": 5.67980465888977,
      "epoch": 7.633433605500644,
      "grad_norm": 1.375,
      "learning_rate": 0.00012011620094133296,
      "loss": 4.7809,
      "mean_token_accuracy": 0.25052584558725355,
      "num_tokens": 16157656.0,
      "step": 8885
    },
    {
      "entropy": 5.683636951446533,
      "epoch": 7.637730984099699,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00011987520129525622,
      "loss": 4.923,
      "mean_token_accuracy": 0.2359049841761589,
      "num_tokens": 16166900.0,
      "step": 8890
    },
    {
      "entropy": 5.739316749572754,
      "epoch": 7.6420283626987535,
      "grad_norm": 1.46875,
      "learning_rate": 0.000119634540379737,
      "loss": 4.9459,
      "mean_token_accuracy": 0.23375426828861237,
      "num_tokens": 16174859.0,
      "step": 8895
    },
    {
      "entropy": 5.769543027877807,
      "epoch": 7.646325741297808,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011939421872028262,
      "loss": 4.9518,
      "mean_token_accuracy": 0.2293008342385292,
      "num_tokens": 16183660.0,
      "step": 8900
    },
    {
      "entropy": 5.688256788253784,
      "epoch": 7.650623119896863,
      "grad_norm": 1.296875,
      "learning_rate": 0.00011915423684165948,
      "loss": 4.8695,
      "mean_token_accuracy": 0.2363400101661682,
      "num_tokens": 16192344.0,
      "step": 8905
    },
    {
      "entropy": 5.784082841873169,
      "epoch": 7.654920498495917,
      "grad_norm": 1.21875,
      "learning_rate": 0.00011891459526789198,
      "loss": 4.9513,
      "mean_token_accuracy": 0.23087790459394456,
      "num_tokens": 16202060.0,
      "step": 8910
    },
    {
      "entropy": 5.750850200653076,
      "epoch": 7.659217877094972,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001186752945222616,
      "loss": 4.9408,
      "mean_token_accuracy": 0.230193392932415,
      "num_tokens": 16211297.0,
      "step": 8915
    },
    {
      "entropy": 5.741831827163696,
      "epoch": 7.663515255694026,
      "grad_norm": 1.171875,
      "learning_rate": 0.00011843633512730562,
      "loss": 4.891,
      "mean_token_accuracy": 0.23511849343776703,
      "num_tokens": 16219812.0,
      "step": 8920
    },
    {
      "entropy": 5.710381603240966,
      "epoch": 7.667812634293082,
      "grad_norm": 1.28125,
      "learning_rate": 0.00011819771760481576,
      "loss": 4.9056,
      "mean_token_accuracy": 0.24004074782133103,
      "num_tokens": 16229197.0,
      "step": 8925
    },
    {
      "entropy": 5.659539985656738,
      "epoch": 7.672110012892135,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011795944247583725,
      "loss": 4.827,
      "mean_token_accuracy": 0.2396245777606964,
      "num_tokens": 16238154.0,
      "step": 8930
    },
    {
      "entropy": 5.710673809051514,
      "epoch": 7.676407391491191,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00011772151026066789,
      "loss": 4.8812,
      "mean_token_accuracy": 0.23135425001382828,
      "num_tokens": 16247206.0,
      "step": 8935
    },
    {
      "entropy": 5.770887327194214,
      "epoch": 7.680704770090245,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00011748392147885642,
      "loss": 5.0019,
      "mean_token_accuracy": 0.22777645140886307,
      "num_tokens": 16256571.0,
      "step": 8940
    },
    {
      "entropy": 5.739024114608765,
      "epoch": 7.6850021486893,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011724667664920177,
      "loss": 4.9316,
      "mean_token_accuracy": 0.23300305306911467,
      "num_tokens": 16265429.0,
      "step": 8945
    },
    {
      "entropy": 5.772635078430175,
      "epoch": 7.689299527288354,
      "grad_norm": 1.375,
      "learning_rate": 0.00011700977628975183,
      "loss": 5.0259,
      "mean_token_accuracy": 0.22466233223676682,
      "num_tokens": 16273804.0,
      "step": 8950
    },
    {
      "entropy": 5.75628547668457,
      "epoch": 7.693596905887409,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011677322091780243,
      "loss": 4.9441,
      "mean_token_accuracy": 0.230313977599144,
      "num_tokens": 16282894.0,
      "step": 8955
    },
    {
      "entropy": 5.782389211654663,
      "epoch": 7.6978942844864635,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001165370110498958,
      "loss": 4.9823,
      "mean_token_accuracy": 0.22912768721580506,
      "num_tokens": 16291568.0,
      "step": 8960
    },
    {
      "entropy": 5.8072168827056885,
      "epoch": 7.702191663085518,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011630114720181989,
      "loss": 4.9706,
      "mean_token_accuracy": 0.2286185145378113,
      "num_tokens": 16300650.0,
      "step": 8965
    },
    {
      "entropy": 5.725326013565064,
      "epoch": 7.706489041684573,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011606562988860711,
      "loss": 4.9052,
      "mean_token_accuracy": 0.23271812498569489,
      "num_tokens": 16309712.0,
      "step": 8970
    },
    {
      "entropy": 5.704691696166992,
      "epoch": 7.710786420283627,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001158304596245332,
      "loss": 4.8801,
      "mean_token_accuracy": 0.23317846804857253,
      "num_tokens": 16319440.0,
      "step": 8975
    },
    {
      "entropy": 5.673254013061523,
      "epoch": 7.715083798882682,
      "grad_norm": 1.265625,
      "learning_rate": 0.00011559563692311595,
      "loss": 4.8871,
      "mean_token_accuracy": 0.2425983503460884,
      "num_tokens": 16328752.0,
      "step": 8980
    },
    {
      "entropy": 5.715083312988281,
      "epoch": 7.719381177481736,
      "grad_norm": 1.25,
      "learning_rate": 0.00011536116229711422,
      "loss": 4.87,
      "mean_token_accuracy": 0.23960913121700286,
      "num_tokens": 16338045.0,
      "step": 8985
    },
    {
      "entropy": 5.771728086471557,
      "epoch": 7.723678556080791,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000115127036258527,
      "loss": 4.9468,
      "mean_token_accuracy": 0.22923657298088074,
      "num_tokens": 16347174.0,
      "step": 8990
    },
    {
      "entropy": 5.707568883895874,
      "epoch": 7.727975934679845,
      "grad_norm": 1.375,
      "learning_rate": 0.00011489325931859185,
      "loss": 4.8257,
      "mean_token_accuracy": 0.2456026166677475,
      "num_tokens": 16355371.0,
      "step": 8995
    },
    {
      "entropy": 5.703813886642456,
      "epoch": 7.7322733132789,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001146598319877843,
      "loss": 4.884,
      "mean_token_accuracy": 0.2391938716173172,
      "num_tokens": 16363938.0,
      "step": 9000
    },
    {
      "epoch": 7.7322733132789,
      "eval_entropy": 5.542284931685473,
      "eval_loss": 5.893875598907471,
      "eval_mean_token_accuracy": 0.18242550859445925,
      "eval_num_tokens": 16363938.0,
      "eval_runtime": 2.0469,
      "eval_samples_per_second": 1733.83,
      "eval_steps_per_second": 216.912,
      "step": 9000
    },
    {
      "entropy": 5.730257511138916,
      "epoch": 7.736570691877954,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011442675477581621,
      "loss": 4.946,
      "mean_token_accuracy": 0.23982110470533372,
      "num_tokens": 16373110.0,
      "step": 9005
    },
    {
      "entropy": 5.769806432723999,
      "epoch": 7.740868070477009,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001141940281916352,
      "loss": 4.8663,
      "mean_token_accuracy": 0.23052393943071364,
      "num_tokens": 16381521.0,
      "step": 9010
    },
    {
      "entropy": 5.7794060707092285,
      "epoch": 7.7451654490760635,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00011396165274342304,
      "loss": 4.9934,
      "mean_token_accuracy": 0.2219093069434166,
      "num_tokens": 16391322.0,
      "step": 9015
    },
    {
      "entropy": 5.702830696105957,
      "epoch": 7.749462827675118,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00011372962893859471,
      "loss": 4.9375,
      "mean_token_accuracy": 0.22916385978460313,
      "num_tokens": 16400653.0,
      "step": 9020
    },
    {
      "entropy": 5.685720300674438,
      "epoch": 7.753760206274173,
      "grad_norm": 1.28125,
      "learning_rate": 0.00011349795728379759,
      "loss": 4.9439,
      "mean_token_accuracy": 0.2307956948876381,
      "num_tokens": 16410133.0,
      "step": 9025
    },
    {
      "entropy": 5.757158088684082,
      "epoch": 7.758057584873227,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011326663828491,
      "loss": 5.0021,
      "mean_token_accuracy": 0.22755355685949324,
      "num_tokens": 16419302.0,
      "step": 9030
    },
    {
      "entropy": 5.731455373764038,
      "epoch": 7.762354963472282,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011303567244704015,
      "loss": 4.9722,
      "mean_token_accuracy": 0.23510894626379014,
      "num_tokens": 16428020.0,
      "step": 9035
    },
    {
      "entropy": 5.787262058258056,
      "epoch": 7.766652342071336,
      "grad_norm": 1.296875,
      "learning_rate": 0.00011280506027452502,
      "loss": 5.0529,
      "mean_token_accuracy": 0.21928364634513856,
      "num_tokens": 16438033.0,
      "step": 9040
    },
    {
      "entropy": 5.7590248584747314,
      "epoch": 7.770949720670391,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001125748022709295,
      "loss": 4.9702,
      "mean_token_accuracy": 0.23275834918022156,
      "num_tokens": 16447067.0,
      "step": 9045
    },
    {
      "entropy": 5.787170457839966,
      "epoch": 7.775247099269445,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00011234489893904509,
      "loss": 4.9795,
      "mean_token_accuracy": 0.2291556641459465,
      "num_tokens": 16457146.0,
      "step": 9050
    },
    {
      "entropy": 5.732330751419068,
      "epoch": 7.7795444778685,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011211535078088869,
      "loss": 4.8429,
      "mean_token_accuracy": 0.24395294189453126,
      "num_tokens": 16466428.0,
      "step": 9055
    },
    {
      "entropy": 5.7382737159729,
      "epoch": 7.783841856467554,
      "grad_norm": 1.375,
      "learning_rate": 0.00011188615829770171,
      "loss": 4.9092,
      "mean_token_accuracy": 0.23593527376651763,
      "num_tokens": 16474198.0,
      "step": 9060
    },
    {
      "entropy": 5.710181808471679,
      "epoch": 7.788139235066609,
      "grad_norm": 1.375,
      "learning_rate": 0.00011165732198994905,
      "loss": 4.9434,
      "mean_token_accuracy": 0.23322095423936845,
      "num_tokens": 16483464.0,
      "step": 9065
    },
    {
      "entropy": 5.713530969619751,
      "epoch": 7.792436613665664,
      "grad_norm": 1.375,
      "learning_rate": 0.00011142884235731756,
      "loss": 4.9257,
      "mean_token_accuracy": 0.23757666498422622,
      "num_tokens": 16492619.0,
      "step": 9070
    },
    {
      "entropy": 5.7721363544464115,
      "epoch": 7.796733992264718,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011120071989871564,
      "loss": 5.0154,
      "mean_token_accuracy": 0.2297023668885231,
      "num_tokens": 16501690.0,
      "step": 9075
    },
    {
      "entropy": 5.661825942993164,
      "epoch": 7.8010313708637735,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00011097295511227134,
      "loss": 4.8963,
      "mean_token_accuracy": 0.2394334778189659,
      "num_tokens": 16510158.0,
      "step": 9080
    },
    {
      "entropy": 5.722220945358276,
      "epoch": 7.805328749462827,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001107455484953321,
      "loss": 4.9024,
      "mean_token_accuracy": 0.23700267672538758,
      "num_tokens": 16518722.0,
      "step": 9085
    },
    {
      "entropy": 5.6848217964172365,
      "epoch": 7.809626128061883,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011051850054446306,
      "loss": 4.9151,
      "mean_token_accuracy": 0.23075505644083022,
      "num_tokens": 16527404.0,
      "step": 9090
    },
    {
      "entropy": 5.663630819320678,
      "epoch": 7.813923506660937,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011029181175544603,
      "loss": 4.8506,
      "mean_token_accuracy": 0.2431934058666229,
      "num_tokens": 16536210.0,
      "step": 9095
    },
    {
      "entropy": 5.792266893386841,
      "epoch": 7.818220885259992,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011006548262327884,
      "loss": 5.0166,
      "mean_token_accuracy": 0.23040284514427184,
      "num_tokens": 16544707.0,
      "step": 9100
    },
    {
      "entropy": 5.806525135040284,
      "epoch": 7.822518263859046,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001098395136421739,
      "loss": 4.9812,
      "mean_token_accuracy": 0.22473449409008026,
      "num_tokens": 16553883.0,
      "step": 9105
    },
    {
      "entropy": 5.760424566268921,
      "epoch": 7.826815642458101,
      "grad_norm": 1.1875,
      "learning_rate": 0.00010961390530555712,
      "loss": 4.9704,
      "mean_token_accuracy": 0.22575420886278152,
      "num_tokens": 16562537.0,
      "step": 9110
    },
    {
      "entropy": 5.756677484512329,
      "epoch": 7.831113021057155,
      "grad_norm": 1.265625,
      "learning_rate": 0.00010938865810606682,
      "loss": 4.9295,
      "mean_token_accuracy": 0.23352717459201813,
      "num_tokens": 16571665.0,
      "step": 9115
    },
    {
      "entropy": 5.738250303268432,
      "epoch": 7.83541039965621,
      "grad_norm": 1.1875,
      "learning_rate": 0.00010916377253555293,
      "loss": 4.9336,
      "mean_token_accuracy": 0.22934698015451432,
      "num_tokens": 16581102.0,
      "step": 9120
    },
    {
      "entropy": 5.708257818222046,
      "epoch": 7.839707778255264,
      "grad_norm": 1.328125,
      "learning_rate": 0.00010893924908507573,
      "loss": 4.8916,
      "mean_token_accuracy": 0.2412150412797928,
      "num_tokens": 16589958.0,
      "step": 9125
    },
    {
      "entropy": 5.7717221736907955,
      "epoch": 7.844005156854319,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001087150882449046,
      "loss": 4.9687,
      "mean_token_accuracy": 0.23071876615285875,
      "num_tokens": 16598800.0,
      "step": 9130
    },
    {
      "entropy": 5.744738292694092,
      "epoch": 7.8483025354533735,
      "grad_norm": 1.265625,
      "learning_rate": 0.00010849129050451717,
      "loss": 4.9183,
      "mean_token_accuracy": 0.23448994606733323,
      "num_tokens": 16607751.0,
      "step": 9135
    },
    {
      "entropy": 5.703471088409424,
      "epoch": 7.852599914052428,
      "grad_norm": 1.234375,
      "learning_rate": 0.00010826785635259842,
      "loss": 4.8756,
      "mean_token_accuracy": 0.23654707819223403,
      "num_tokens": 16616041.0,
      "step": 9140
    },
    {
      "entropy": 5.713053512573242,
      "epoch": 7.856897292651483,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010804478627703903,
      "loss": 4.8403,
      "mean_token_accuracy": 0.2534535199403763,
      "num_tokens": 16624800.0,
      "step": 9145
    },
    {
      "entropy": 5.770458030700683,
      "epoch": 7.861194671250537,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00010782208076493508,
      "loss": 4.9587,
      "mean_token_accuracy": 0.23223767578601837,
      "num_tokens": 16632808.0,
      "step": 9150
    },
    {
      "entropy": 5.747208118438721,
      "epoch": 7.865492049849592,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00010759974030258621,
      "loss": 4.9428,
      "mean_token_accuracy": 0.2311663582921028,
      "num_tokens": 16641179.0,
      "step": 9155
    },
    {
      "entropy": 5.7243444442749025,
      "epoch": 7.869789428448646,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00010737776537549531,
      "loss": 4.9326,
      "mean_token_accuracy": 0.24079572409391403,
      "num_tokens": 16650402.0,
      "step": 9160
    },
    {
      "entropy": 5.710455989837646,
      "epoch": 7.874086807047701,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010715615646836679,
      "loss": 4.8984,
      "mean_token_accuracy": 0.23725621849298478,
      "num_tokens": 16659661.0,
      "step": 9165
    },
    {
      "entropy": 5.641438770294189,
      "epoch": 7.878384185646755,
      "grad_norm": 1.21875,
      "learning_rate": 0.00010693491406510585,
      "loss": 4.8848,
      "mean_token_accuracy": 0.2445200726389885,
      "num_tokens": 16668630.0,
      "step": 9170
    },
    {
      "entropy": 5.747666835784912,
      "epoch": 7.88268156424581,
      "grad_norm": 1.328125,
      "learning_rate": 0.00010671403864881757,
      "loss": 4.9711,
      "mean_token_accuracy": 0.2317958801984787,
      "num_tokens": 16678023.0,
      "step": 9175
    },
    {
      "entropy": 5.759106779098511,
      "epoch": 7.8869789428448644,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00010649353070180562,
      "loss": 4.9623,
      "mean_token_accuracy": 0.2380196452140808,
      "num_tokens": 16686751.0,
      "step": 9180
    },
    {
      "entropy": 5.734574794769287,
      "epoch": 7.891276321443919,
      "grad_norm": 1.296875,
      "learning_rate": 0.00010627339070557118,
      "loss": 4.9401,
      "mean_token_accuracy": 0.24096089154481887,
      "num_tokens": 16696672.0,
      "step": 9185
    },
    {
      "entropy": 5.685746860504151,
      "epoch": 7.8955737000429735,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00010605361914081194,
      "loss": 4.854,
      "mean_token_accuracy": 0.24479493349790574,
      "num_tokens": 16706018.0,
      "step": 9190
    },
    {
      "entropy": 5.758388519287109,
      "epoch": 7.899871078642028,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00010583421648742125,
      "loss": 4.8878,
      "mean_token_accuracy": 0.24218843132257462,
      "num_tokens": 16715206.0,
      "step": 9195
    },
    {
      "entropy": 5.73368878364563,
      "epoch": 7.904168457241083,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010561518322448673,
      "loss": 4.993,
      "mean_token_accuracy": 0.22797715663909912,
      "num_tokens": 16724479.0,
      "step": 9200
    },
    {
      "entropy": 5.702595281600952,
      "epoch": 7.908465835840137,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010539651983028955,
      "loss": 4.868,
      "mean_token_accuracy": 0.2413019374012947,
      "num_tokens": 16733304.0,
      "step": 9205
    },
    {
      "entropy": 5.685127592086792,
      "epoch": 7.912763214439192,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001051782267823031,
      "loss": 4.8558,
      "mean_token_accuracy": 0.24108621031045913,
      "num_tokens": 16741447.0,
      "step": 9210
    },
    {
      "entropy": 5.721054267883301,
      "epoch": 7.917060593038247,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010496030455719225,
      "loss": 4.9251,
      "mean_token_accuracy": 0.23336865305900573,
      "num_tokens": 16751487.0,
      "step": 9215
    },
    {
      "entropy": 5.724647521972656,
      "epoch": 7.921357971637301,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010474275363081193,
      "loss": 4.9328,
      "mean_token_accuracy": 0.23269153237342835,
      "num_tokens": 16760795.0,
      "step": 9220
    },
    {
      "entropy": 5.702741193771362,
      "epoch": 7.925655350236356,
      "grad_norm": 1.375,
      "learning_rate": 0.0001045255744782064,
      "loss": 4.9025,
      "mean_token_accuracy": 0.24020729959011078,
      "num_tokens": 16769639.0,
      "step": 9225
    },
    {
      "entropy": 5.816455984115601,
      "epoch": 7.92995272883541,
      "grad_norm": 1.28125,
      "learning_rate": 0.00010430876757360817,
      "loss": 4.9837,
      "mean_token_accuracy": 0.22814977616071702,
      "num_tokens": 16779195.0,
      "step": 9230
    },
    {
      "entropy": 5.668118476867676,
      "epoch": 7.934250107434465,
      "grad_norm": 1.359375,
      "learning_rate": 0.00010409233339043694,
      "loss": 4.8047,
      "mean_token_accuracy": 0.24403645545244218,
      "num_tokens": 16787531.0,
      "step": 9235
    },
    {
      "entropy": 5.670865440368653,
      "epoch": 7.93854748603352,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00010387627240129838,
      "loss": 4.8691,
      "mean_token_accuracy": 0.23790572285652162,
      "num_tokens": 16796392.0,
      "step": 9240
    },
    {
      "entropy": 5.739948797225952,
      "epoch": 7.942844864632574,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00010366058507798326,
      "loss": 4.9214,
      "mean_token_accuracy": 0.2339615508913994,
      "num_tokens": 16804942.0,
      "step": 9245
    },
    {
      "entropy": 5.705602979660034,
      "epoch": 7.947142243231629,
      "grad_norm": 1.296875,
      "learning_rate": 0.00010344527189146655,
      "loss": 4.9528,
      "mean_token_accuracy": 0.2274705156683922,
      "num_tokens": 16813754.0,
      "step": 9250
    },
    {
      "entropy": 5.739188241958618,
      "epoch": 7.9514396218306835,
      "grad_norm": 1.1875,
      "learning_rate": 0.00010323033331190626,
      "loss": 4.9274,
      "mean_token_accuracy": 0.23373161852359772,
      "num_tokens": 16823010.0,
      "step": 9255
    },
    {
      "entropy": 5.76418023109436,
      "epoch": 7.955737000429738,
      "grad_norm": 1.3125,
      "learning_rate": 0.00010301576980864228,
      "loss": 4.8603,
      "mean_token_accuracy": 0.2446628138422966,
      "num_tokens": 16831909.0,
      "step": 9260
    },
    {
      "entropy": 5.694305419921875,
      "epoch": 7.960034379028793,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010280158185019547,
      "loss": 4.9073,
      "mean_token_accuracy": 0.24446062594652176,
      "num_tokens": 16841460.0,
      "step": 9265
    },
    {
      "entropy": 5.697159719467163,
      "epoch": 7.964331757627847,
      "grad_norm": 1.15625,
      "learning_rate": 0.00010258776990426686,
      "loss": 4.927,
      "mean_token_accuracy": 0.24162409752607344,
      "num_tokens": 16850592.0,
      "step": 9270
    },
    {
      "entropy": 5.729753589630127,
      "epoch": 7.968629136226902,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010237433443773612,
      "loss": 4.9118,
      "mean_token_accuracy": 0.23850051164627076,
      "num_tokens": 16859736.0,
      "step": 9275
    },
    {
      "entropy": 5.655977869033814,
      "epoch": 7.972926514825956,
      "grad_norm": 1.203125,
      "learning_rate": 0.00010216127591666115,
      "loss": 4.9269,
      "mean_token_accuracy": 0.24168841391801835,
      "num_tokens": 16870084.0,
      "step": 9280
    },
    {
      "entropy": 5.739932727813721,
      "epoch": 7.977223893425011,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00010194859480627648,
      "loss": 4.9567,
      "mean_token_accuracy": 0.2386748567223549,
      "num_tokens": 16877771.0,
      "step": 9285
    },
    {
      "entropy": 5.675134468078613,
      "epoch": 7.981521272024065,
      "grad_norm": 1.203125,
      "learning_rate": 0.00010173629157099279,
      "loss": 4.9575,
      "mean_token_accuracy": 0.2281126633286476,
      "num_tokens": 16887487.0,
      "step": 9290
    },
    {
      "entropy": 5.714840936660766,
      "epoch": 7.98581865062312,
      "grad_norm": 1.15625,
      "learning_rate": 0.00010152436667439537,
      "loss": 4.9519,
      "mean_token_accuracy": 0.23581907600164415,
      "num_tokens": 16897286.0,
      "step": 9295
    },
    {
      "entropy": 5.750794839859009,
      "epoch": 7.9901160292221745,
      "grad_norm": 1.3125,
      "learning_rate": 0.00010131282057924345,
      "loss": 4.888,
      "mean_token_accuracy": 0.23815936595201492,
      "num_tokens": 16905968.0,
      "step": 9300
    },
    {
      "entropy": 5.780832004547119,
      "epoch": 7.994413407821229,
      "grad_norm": 1.25,
      "learning_rate": 0.00010110165374746924,
      "loss": 4.8951,
      "mean_token_accuracy": 0.24371608346700668,
      "num_tokens": 16914604.0,
      "step": 9305
    },
    {
      "entropy": 5.744164609909058,
      "epoch": 7.9987107864202835,
      "grad_norm": 1.28125,
      "learning_rate": 0.00010089086664017674,
      "loss": 5.0412,
      "mean_token_accuracy": 0.22926397770643234,
      "num_tokens": 16925085.0,
      "step": 9310
    },
    {
      "entropy": 5.742586241828071,
      "epoch": 8.002578427159433,
      "grad_norm": 1.296875,
      "learning_rate": 0.00010068045971764067,
      "loss": 4.8669,
      "mean_token_accuracy": 0.240800850921207,
      "num_tokens": 16932717.0,
      "step": 9315
    },
    {
      "entropy": 5.752298831939697,
      "epoch": 8.006875805758487,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00010047043343930561,
      "loss": 4.9556,
      "mean_token_accuracy": 0.239510677754879,
      "num_tokens": 16941332.0,
      "step": 9320
    },
    {
      "entropy": 5.703397274017334,
      "epoch": 8.011173184357542,
      "grad_norm": 1.359375,
      "learning_rate": 0.00010026078826378502,
      "loss": 4.7841,
      "mean_token_accuracy": 0.2428373247385025,
      "num_tokens": 16949732.0,
      "step": 9325
    },
    {
      "entropy": 5.756609344482422,
      "epoch": 8.015470562956596,
      "grad_norm": 1.265625,
      "learning_rate": 0.00010005152464886031,
      "loss": 4.8276,
      "mean_token_accuracy": 0.23863828033208848,
      "num_tokens": 16958013.0,
      "step": 9330
    },
    {
      "entropy": 5.721332359313965,
      "epoch": 8.019767941555651,
      "grad_norm": 1.21875,
      "learning_rate": 9.984264305147941e-05,
      "loss": 4.7953,
      "mean_token_accuracy": 0.2455083206295967,
      "num_tokens": 16966050.0,
      "step": 9335
    },
    {
      "entropy": 5.687017393112183,
      "epoch": 8.024065320154705,
      "grad_norm": 1.21875,
      "learning_rate": 9.963414392775627e-05,
      "loss": 4.7865,
      "mean_token_accuracy": 0.2540111944079399,
      "num_tokens": 16975178.0,
      "step": 9340
    },
    {
      "entropy": 5.612176132202149,
      "epoch": 8.02836269875376,
      "grad_norm": 1.234375,
      "learning_rate": 9.942602773296971e-05,
      "loss": 4.7719,
      "mean_token_accuracy": 0.2532846137881279,
      "num_tokens": 16984247.0,
      "step": 9345
    },
    {
      "entropy": 5.7741395950317385,
      "epoch": 8.032660077352816,
      "grad_norm": 1.2578125,
      "learning_rate": 9.921829492156223e-05,
      "loss": 4.9684,
      "mean_token_accuracy": 0.23141839802265168,
      "num_tokens": 16995048.0,
      "step": 9350
    },
    {
      "entropy": 5.75913782119751,
      "epoch": 8.03695745595187,
      "grad_norm": 1.5,
      "learning_rate": 9.901094594713933e-05,
      "loss": 4.9403,
      "mean_token_accuracy": 0.23183894604444505,
      "num_tokens": 17003748.0,
      "step": 9355
    },
    {
      "entropy": 5.784180068969727,
      "epoch": 8.041254834550925,
      "grad_norm": 1.28125,
      "learning_rate": 9.88039812624682e-05,
      "loss": 4.9053,
      "mean_token_accuracy": 0.23992706537246705,
      "num_tokens": 17011639.0,
      "step": 9360
    },
    {
      "entropy": 5.799221658706665,
      "epoch": 8.045552213149978,
      "grad_norm": 1.1328125,
      "learning_rate": 9.859740131947715e-05,
      "loss": 4.8935,
      "mean_token_accuracy": 0.2353417918086052,
      "num_tokens": 17021056.0,
      "step": 9365
    },
    {
      "entropy": 5.754740142822266,
      "epoch": 8.049849591749034,
      "grad_norm": 1.265625,
      "learning_rate": 9.839120656925407e-05,
      "loss": 4.9271,
      "mean_token_accuracy": 0.2309979572892189,
      "num_tokens": 17030944.0,
      "step": 9370
    },
    {
      "entropy": 5.699716949462891,
      "epoch": 8.054146970348087,
      "grad_norm": 1.3046875,
      "learning_rate": 9.818539746204588e-05,
      "loss": 4.8744,
      "mean_token_accuracy": 0.23718750476837158,
      "num_tokens": 17040127.0,
      "step": 9375
    },
    {
      "entropy": 5.7823670387268065,
      "epoch": 8.058444348947143,
      "grad_norm": 1.5,
      "learning_rate": 9.797997444725745e-05,
      "loss": 4.8596,
      "mean_token_accuracy": 0.24111505299806596,
      "num_tokens": 17049418.0,
      "step": 9380
    },
    {
      "entropy": 5.69295825958252,
      "epoch": 8.062741727546197,
      "grad_norm": 1.28125,
      "learning_rate": 9.77749379734506e-05,
      "loss": 4.8774,
      "mean_token_accuracy": 0.24546006321907043,
      "num_tokens": 17058270.0,
      "step": 9385
    },
    {
      "entropy": 5.734280729293824,
      "epoch": 8.067039106145252,
      "grad_norm": 1.3359375,
      "learning_rate": 9.757028848834293e-05,
      "loss": 4.8781,
      "mean_token_accuracy": 0.23956651538610457,
      "num_tokens": 17068011.0,
      "step": 9390
    },
    {
      "entropy": 5.702370309829712,
      "epoch": 8.071336484744306,
      "grad_norm": 1.2734375,
      "learning_rate": 9.736602643880712e-05,
      "loss": 4.8929,
      "mean_token_accuracy": 0.23692873120307922,
      "num_tokens": 17077356.0,
      "step": 9395
    },
    {
      "entropy": 5.742789077758789,
      "epoch": 8.075633863343361,
      "grad_norm": 1.265625,
      "learning_rate": 9.716215227086997e-05,
      "loss": 4.832,
      "mean_token_accuracy": 0.24452268034219743,
      "num_tokens": 17085679.0,
      "step": 9400
    },
    {
      "entropy": 5.737786293029785,
      "epoch": 8.079931241942415,
      "grad_norm": 1.2421875,
      "learning_rate": 9.695866642971098e-05,
      "loss": 4.8761,
      "mean_token_accuracy": 0.23521246314048766,
      "num_tokens": 17094925.0,
      "step": 9405
    },
    {
      "entropy": 5.735822296142578,
      "epoch": 8.08422862054147,
      "grad_norm": 1.3046875,
      "learning_rate": 9.67555693596621e-05,
      "loss": 4.922,
      "mean_token_accuracy": 0.23624190241098403,
      "num_tokens": 17105278.0,
      "step": 9410
    },
    {
      "entropy": 5.697363328933716,
      "epoch": 8.088525999140524,
      "grad_norm": 1.2890625,
      "learning_rate": 9.655286150420595e-05,
      "loss": 4.843,
      "mean_token_accuracy": 0.24786596596240998,
      "num_tokens": 17114070.0,
      "step": 9415
    },
    {
      "entropy": 5.631302261352539,
      "epoch": 8.09282337773958,
      "grad_norm": 1.234375,
      "learning_rate": 9.635054330597565e-05,
      "loss": 4.7635,
      "mean_token_accuracy": 0.2543473169207573,
      "num_tokens": 17122862.0,
      "step": 9420
    },
    {
      "entropy": 5.75023717880249,
      "epoch": 8.097120756338633,
      "grad_norm": 1.3515625,
      "learning_rate": 9.614861520675322e-05,
      "loss": 4.8539,
      "mean_token_accuracy": 0.24282941818237305,
      "num_tokens": 17131555.0,
      "step": 9425
    },
    {
      "entropy": 5.703547573089599,
      "epoch": 8.101418134937688,
      "grad_norm": 1.3359375,
      "learning_rate": 9.594707764746881e-05,
      "loss": 4.8151,
      "mean_token_accuracy": 0.24557766914367676,
      "num_tokens": 17140841.0,
      "step": 9430
    },
    {
      "entropy": 5.64467306137085,
      "epoch": 8.105715513536742,
      "grad_norm": 1.2265625,
      "learning_rate": 9.57459310682e-05,
      "loss": 4.8587,
      "mean_token_accuracy": 0.23592619597911835,
      "num_tokens": 17150027.0,
      "step": 9435
    },
    {
      "entropy": 5.759531688690186,
      "epoch": 8.110012892135797,
      "grad_norm": 1.2265625,
      "learning_rate": 9.554517590817055e-05,
      "loss": 4.9168,
      "mean_token_accuracy": 0.22910912930965424,
      "num_tokens": 17159589.0,
      "step": 9440
    },
    {
      "entropy": 5.783081340789795,
      "epoch": 8.114310270734851,
      "grad_norm": 1.1953125,
      "learning_rate": 9.534481260574944e-05,
      "loss": 4.9094,
      "mean_token_accuracy": 0.2321816936135292,
      "num_tokens": 17168219.0,
      "step": 9445
    },
    {
      "entropy": 5.742900657653808,
      "epoch": 8.118607649333907,
      "grad_norm": 1.25,
      "learning_rate": 9.514484159844997e-05,
      "loss": 4.8874,
      "mean_token_accuracy": 0.2349926143884659,
      "num_tokens": 17177364.0,
      "step": 9450
    },
    {
      "entropy": 5.771203088760376,
      "epoch": 8.12290502793296,
      "grad_norm": 1.3359375,
      "learning_rate": 9.494526332292899e-05,
      "loss": 4.9252,
      "mean_token_accuracy": 0.23223545551300048,
      "num_tokens": 17186572.0,
      "step": 9455
    },
    {
      "entropy": 5.734961462020874,
      "epoch": 8.127202406532016,
      "grad_norm": 1.265625,
      "learning_rate": 9.47460782149857e-05,
      "loss": 4.9111,
      "mean_token_accuracy": 0.2386125445365906,
      "num_tokens": 17195645.0,
      "step": 9460
    },
    {
      "entropy": 5.759470748901367,
      "epoch": 8.13149978513107,
      "grad_norm": 1.234375,
      "learning_rate": 9.454728670956073e-05,
      "loss": 4.9004,
      "mean_token_accuracy": 0.22731278389692305,
      "num_tokens": 17205279.0,
      "step": 9465
    },
    {
      "entropy": 5.70623517036438,
      "epoch": 8.135797163730125,
      "grad_norm": 1.2890625,
      "learning_rate": 9.43488892407352e-05,
      "loss": 4.8315,
      "mean_token_accuracy": 0.24042012691497802,
      "num_tokens": 17214536.0,
      "step": 9470
    },
    {
      "entropy": 5.785218620300293,
      "epoch": 8.140094542329178,
      "grad_norm": 1.2421875,
      "learning_rate": 9.415088624172997e-05,
      "loss": 4.9279,
      "mean_token_accuracy": 0.23537793904542922,
      "num_tokens": 17223336.0,
      "step": 9475
    },
    {
      "entropy": 5.756191349029541,
      "epoch": 8.144391920928234,
      "grad_norm": 1.296875,
      "learning_rate": 9.395327814490439e-05,
      "loss": 4.931,
      "mean_token_accuracy": 0.23400909453630447,
      "num_tokens": 17232991.0,
      "step": 9480
    },
    {
      "entropy": 5.731492424011231,
      "epoch": 8.148689299527287,
      "grad_norm": 1.3515625,
      "learning_rate": 9.375606538175566e-05,
      "loss": 4.8581,
      "mean_token_accuracy": 0.24122172892093657,
      "num_tokens": 17241760.0,
      "step": 9485
    },
    {
      "entropy": 5.7561249256134035,
      "epoch": 8.152986678126343,
      "grad_norm": 1.2265625,
      "learning_rate": 9.35592483829175e-05,
      "loss": 4.9404,
      "mean_token_accuracy": 0.23250404745340347,
      "num_tokens": 17251514.0,
      "step": 9490
    },
    {
      "entropy": 5.7169372081756595,
      "epoch": 8.157284056725398,
      "grad_norm": 1.3046875,
      "learning_rate": 9.336282757815964e-05,
      "loss": 4.9114,
      "mean_token_accuracy": 0.23848976343870162,
      "num_tokens": 17260876.0,
      "step": 9495
    },
    {
      "entropy": 5.70046215057373,
      "epoch": 8.161581435324452,
      "grad_norm": 1.3125,
      "learning_rate": 9.316680339638664e-05,
      "loss": 4.9403,
      "mean_token_accuracy": 0.23490447849035262,
      "num_tokens": 17270051.0,
      "step": 9500
    },
    {
      "epoch": 8.161581435324452,
      "eval_entropy": 5.563689900948121,
      "eval_loss": 5.894437789916992,
      "eval_mean_token_accuracy": 0.18221602075406024,
      "eval_num_tokens": 17270051.0,
      "eval_runtime": 2.0525,
      "eval_samples_per_second": 1729.101,
      "eval_steps_per_second": 216.32,
      "step": 9500
    },
    {
      "entropy": 5.79104528427124,
      "epoch": 8.165878813923507,
      "grad_norm": 1.4140625,
      "learning_rate": 9.297117626563687e-05,
      "loss": 4.9393,
      "mean_token_accuracy": 0.2395332857966423,
      "num_tokens": 17279744.0,
      "step": 9505
    },
    {
      "entropy": 5.732816505432129,
      "epoch": 8.170176192522561,
      "grad_norm": 1.2890625,
      "learning_rate": 9.27759466130818e-05,
      "loss": 4.8743,
      "mean_token_accuracy": 0.24095906615257262,
      "num_tokens": 17289107.0,
      "step": 9510
    },
    {
      "entropy": 5.6882086277008055,
      "epoch": 8.174473571121617,
      "grad_norm": 1.1796875,
      "learning_rate": 9.25811148650251e-05,
      "loss": 4.7732,
      "mean_token_accuracy": 0.25059460252523424,
      "num_tokens": 17298385.0,
      "step": 9515
    },
    {
      "entropy": 5.7441253662109375,
      "epoch": 8.17877094972067,
      "grad_norm": 1.2421875,
      "learning_rate": 9.238668144690133e-05,
      "loss": 4.8633,
      "mean_token_accuracy": 0.23739856630563735,
      "num_tokens": 17307759.0,
      "step": 9520
    },
    {
      "entropy": 5.712777185440063,
      "epoch": 8.183068328319726,
      "grad_norm": 1.21875,
      "learning_rate": 9.219264678327527e-05,
      "loss": 4.8583,
      "mean_token_accuracy": 0.24061242938041688,
      "num_tokens": 17317135.0,
      "step": 9525
    },
    {
      "entropy": 5.691803932189941,
      "epoch": 8.18736570691878,
      "grad_norm": 1.2890625,
      "learning_rate": 9.199901129784121e-05,
      "loss": 4.8485,
      "mean_token_accuracy": 0.24647330194711686,
      "num_tokens": 17326950.0,
      "step": 9530
    },
    {
      "entropy": 5.782800197601318,
      "epoch": 8.191663085517835,
      "grad_norm": 1.21875,
      "learning_rate": 9.180577541342164e-05,
      "loss": 4.964,
      "mean_token_accuracy": 0.23249762654304504,
      "num_tokens": 17335873.0,
      "step": 9535
    },
    {
      "entropy": 5.714890718460083,
      "epoch": 8.195960464116888,
      "grad_norm": 1.328125,
      "learning_rate": 9.161293955196648e-05,
      "loss": 4.8682,
      "mean_token_accuracy": 0.24433981478214264,
      "num_tokens": 17344659.0,
      "step": 9540
    },
    {
      "entropy": 5.783798694610596,
      "epoch": 8.200257842715944,
      "grad_norm": 1.328125,
      "learning_rate": 9.142050413455214e-05,
      "loss": 4.9286,
      "mean_token_accuracy": 0.23019878268241883,
      "num_tokens": 17353274.0,
      "step": 9545
    },
    {
      "entropy": 5.750073575973511,
      "epoch": 8.204555221314997,
      "grad_norm": 1.2109375,
      "learning_rate": 9.12284695813807e-05,
      "loss": 4.9038,
      "mean_token_accuracy": 0.24520579278469085,
      "num_tokens": 17362212.0,
      "step": 9550
    },
    {
      "entropy": 5.727174139022827,
      "epoch": 8.208852599914053,
      "grad_norm": 1.3359375,
      "learning_rate": 9.103683631177878e-05,
      "loss": 4.8673,
      "mean_token_accuracy": 0.24208441823720933,
      "num_tokens": 17370567.0,
      "step": 9555
    },
    {
      "entropy": 5.729988098144531,
      "epoch": 8.213149978513107,
      "grad_norm": 1.53125,
      "learning_rate": 9.084560474419701e-05,
      "loss": 4.8634,
      "mean_token_accuracy": 0.2391197621822357,
      "num_tokens": 17378521.0,
      "step": 9560
    },
    {
      "entropy": 5.759287405014038,
      "epoch": 8.217447357112162,
      "grad_norm": 1.3671875,
      "learning_rate": 9.065477529620852e-05,
      "loss": 4.8879,
      "mean_token_accuracy": 0.23471090346574783,
      "num_tokens": 17386988.0,
      "step": 9565
    },
    {
      "entropy": 5.7283953666687015,
      "epoch": 8.221744735711216,
      "grad_norm": 1.3828125,
      "learning_rate": 9.046434838450868e-05,
      "loss": 4.8983,
      "mean_token_accuracy": 0.23679854422807695,
      "num_tokens": 17396317.0,
      "step": 9570
    },
    {
      "entropy": 5.797110462188721,
      "epoch": 8.226042114310271,
      "grad_norm": 1.375,
      "learning_rate": 9.027432442491369e-05,
      "loss": 4.9725,
      "mean_token_accuracy": 0.22504952549934387,
      "num_tokens": 17405768.0,
      "step": 9575
    },
    {
      "entropy": 5.758984994888306,
      "epoch": 8.230339492909325,
      "grad_norm": 1.359375,
      "learning_rate": 9.008470383235991e-05,
      "loss": 4.9832,
      "mean_token_accuracy": 0.23178735971450806,
      "num_tokens": 17416098.0,
      "step": 9580
    },
    {
      "entropy": 5.73343677520752,
      "epoch": 8.23463687150838,
      "grad_norm": 1.3671875,
      "learning_rate": 8.989548702090295e-05,
      "loss": 4.8935,
      "mean_token_accuracy": 0.24406977444887162,
      "num_tokens": 17426091.0,
      "step": 9585
    },
    {
      "entropy": 5.700839996337891,
      "epoch": 8.238934250107434,
      "grad_norm": 1.3359375,
      "learning_rate": 8.970667440371676e-05,
      "loss": 4.8358,
      "mean_token_accuracy": 0.24239999651908875,
      "num_tokens": 17434956.0,
      "step": 9590
    },
    {
      "entropy": 5.75614857673645,
      "epoch": 8.24323162870649,
      "grad_norm": 1.25,
      "learning_rate": 8.951826639309257e-05,
      "loss": 4.9163,
      "mean_token_accuracy": 0.23965495973825454,
      "num_tokens": 17443980.0,
      "step": 9595
    },
    {
      "entropy": 5.700466012954712,
      "epoch": 8.247529007305543,
      "grad_norm": 1.375,
      "learning_rate": 8.933026340043811e-05,
      "loss": 4.8575,
      "mean_token_accuracy": 0.24223389178514482,
      "num_tokens": 17452982.0,
      "step": 9600
    },
    {
      "entropy": 5.698475551605225,
      "epoch": 8.251826385904598,
      "grad_norm": 1.359375,
      "learning_rate": 8.914266583627684e-05,
      "loss": 4.8291,
      "mean_token_accuracy": 0.24232960790395736,
      "num_tokens": 17461946.0,
      "step": 9605
    },
    {
      "entropy": 5.786036348342895,
      "epoch": 8.256123764503652,
      "grad_norm": 1.4375,
      "learning_rate": 8.89554741102469e-05,
      "loss": 4.905,
      "mean_token_accuracy": 0.23367423713207244,
      "num_tokens": 17470125.0,
      "step": 9610
    },
    {
      "entropy": 5.71298189163208,
      "epoch": 8.260421143102707,
      "grad_norm": 1.296875,
      "learning_rate": 8.876868863110013e-05,
      "loss": 4.9215,
      "mean_token_accuracy": 0.23382412493228913,
      "num_tokens": 17479536.0,
      "step": 9615
    },
    {
      "entropy": 5.784276628494263,
      "epoch": 8.264718521701761,
      "grad_norm": 1.234375,
      "learning_rate": 8.858230980670134e-05,
      "loss": 4.8705,
      "mean_token_accuracy": 0.23859835118055345,
      "num_tokens": 17488709.0,
      "step": 9620
    },
    {
      "entropy": 5.768792533874512,
      "epoch": 8.269015900300817,
      "grad_norm": 1.25,
      "learning_rate": 8.839633804402747e-05,
      "loss": 4.9358,
      "mean_token_accuracy": 0.23131754696369172,
      "num_tokens": 17498196.0,
      "step": 9625
    },
    {
      "entropy": 5.767839860916138,
      "epoch": 8.27331327889987,
      "grad_norm": 1.4453125,
      "learning_rate": 8.821077374916647e-05,
      "loss": 4.8827,
      "mean_token_accuracy": 0.24630177915096282,
      "num_tokens": 17507374.0,
      "step": 9630
    },
    {
      "entropy": 5.680627489089966,
      "epoch": 8.277610657498926,
      "grad_norm": 1.375,
      "learning_rate": 8.802561732731654e-05,
      "loss": 4.8574,
      "mean_token_accuracy": 0.2398720934987068,
      "num_tokens": 17516163.0,
      "step": 9635
    },
    {
      "entropy": 5.7917811393737795,
      "epoch": 8.281908036097981,
      "grad_norm": 1.2265625,
      "learning_rate": 8.784086918278534e-05,
      "loss": 4.9316,
      "mean_token_accuracy": 0.22891969829797745,
      "num_tokens": 17525567.0,
      "step": 9640
    },
    {
      "entropy": 5.692333650588989,
      "epoch": 8.286205414697035,
      "grad_norm": 1.2890625,
      "learning_rate": 8.765652971898908e-05,
      "loss": 4.8077,
      "mean_token_accuracy": 0.24528621435165404,
      "num_tokens": 17534433.0,
      "step": 9645
    },
    {
      "entropy": 5.738155460357666,
      "epoch": 8.29050279329609,
      "grad_norm": 1.4140625,
      "learning_rate": 8.747259933845134e-05,
      "loss": 4.9463,
      "mean_token_accuracy": 0.24599668383598328,
      "num_tokens": 17542981.0,
      "step": 9650
    },
    {
      "entropy": 5.774512910842896,
      "epoch": 8.294800171895144,
      "grad_norm": 1.2265625,
      "learning_rate": 8.728907844280254e-05,
      "loss": 4.9443,
      "mean_token_accuracy": 0.23179508298635482,
      "num_tokens": 17552425.0,
      "step": 9655
    },
    {
      "entropy": 5.730475330352784,
      "epoch": 8.2990975504942,
      "grad_norm": 1.421875,
      "learning_rate": 8.710596743277901e-05,
      "loss": 4.8814,
      "mean_token_accuracy": 0.2452605828642845,
      "num_tokens": 17560954.0,
      "step": 9660
    },
    {
      "entropy": 5.730258941650391,
      "epoch": 8.303394929093253,
      "grad_norm": 1.3671875,
      "learning_rate": 8.692326670822207e-05,
      "loss": 4.8157,
      "mean_token_accuracy": 0.2448845013976097,
      "num_tokens": 17569769.0,
      "step": 9665
    },
    {
      "entropy": 5.740361738204956,
      "epoch": 8.307692307692308,
      "grad_norm": 1.2890625,
      "learning_rate": 8.6740976668077e-05,
      "loss": 4.8495,
      "mean_token_accuracy": 0.2443626791238785,
      "num_tokens": 17578875.0,
      "step": 9670
    },
    {
      "entropy": 5.683072328567505,
      "epoch": 8.311989686291362,
      "grad_norm": 1.4296875,
      "learning_rate": 8.655909771039239e-05,
      "loss": 4.8408,
      "mean_token_accuracy": 0.24255885183811188,
      "num_tokens": 17587556.0,
      "step": 9675
    },
    {
      "entropy": 5.764631128311157,
      "epoch": 8.316287064890417,
      "grad_norm": 1.5546875,
      "learning_rate": 8.637763023231922e-05,
      "loss": 4.9092,
      "mean_token_accuracy": 0.23544830977916717,
      "num_tokens": 17595077.0,
      "step": 9680
    },
    {
      "entropy": 5.629233980178833,
      "epoch": 8.320584443489471,
      "grad_norm": 1.3828125,
      "learning_rate": 8.619657463010987e-05,
      "loss": 4.7745,
      "mean_token_accuracy": 0.24312002211809158,
      "num_tokens": 17603864.0,
      "step": 9685
    },
    {
      "entropy": 5.785550355911255,
      "epoch": 8.324881822088527,
      "grad_norm": 1.4296875,
      "learning_rate": 8.60159312991175e-05,
      "loss": 4.9774,
      "mean_token_accuracy": 0.23317618519067765,
      "num_tokens": 17613570.0,
      "step": 9690
    },
    {
      "entropy": 5.765936279296875,
      "epoch": 8.32917920068758,
      "grad_norm": 1.25,
      "learning_rate": 8.583570063379487e-05,
      "loss": 4.908,
      "mean_token_accuracy": 0.24077682793140412,
      "num_tokens": 17622352.0,
      "step": 9695
    },
    {
      "entropy": 5.728990364074707,
      "epoch": 8.333476579286636,
      "grad_norm": 1.390625,
      "learning_rate": 8.565588302769374e-05,
      "loss": 4.848,
      "mean_token_accuracy": 0.23626988530158996,
      "num_tokens": 17631556.0,
      "step": 9700
    },
    {
      "entropy": 5.673337030410766,
      "epoch": 8.33777395788569,
      "grad_norm": 1.3046875,
      "learning_rate": 8.54764788734639e-05,
      "loss": 4.8609,
      "mean_token_accuracy": 0.24892795085906982,
      "num_tokens": 17641182.0,
      "step": 9705
    },
    {
      "entropy": 5.765714406967163,
      "epoch": 8.342071336484745,
      "grad_norm": 1.2734375,
      "learning_rate": 8.529748856285228e-05,
      "loss": 4.8606,
      "mean_token_accuracy": 0.2417272850871086,
      "num_tokens": 17650842.0,
      "step": 9710
    },
    {
      "entropy": 5.737577295303344,
      "epoch": 8.346368715083798,
      "grad_norm": 1.359375,
      "learning_rate": 8.511891248670217e-05,
      "loss": 4.8778,
      "mean_token_accuracy": 0.2430541396141052,
      "num_tokens": 17659963.0,
      "step": 9715
    },
    {
      "entropy": 5.78171010017395,
      "epoch": 8.350666093682854,
      "grad_norm": 1.4375,
      "learning_rate": 8.494075103495245e-05,
      "loss": 4.9191,
      "mean_token_accuracy": 0.23714551329612732,
      "num_tokens": 17668637.0,
      "step": 9720
    },
    {
      "entropy": 5.782537889480591,
      "epoch": 8.354963472281907,
      "grad_norm": 1.3125,
      "learning_rate": 8.476300459663643e-05,
      "loss": 4.8997,
      "mean_token_accuracy": 0.22940298169851303,
      "num_tokens": 17678212.0,
      "step": 9725
    },
    {
      "entropy": 5.75561261177063,
      "epoch": 8.359260850880963,
      "grad_norm": 1.2890625,
      "learning_rate": 8.458567355988123e-05,
      "loss": 4.9546,
      "mean_token_accuracy": 0.22492529600858688,
      "num_tokens": 17686766.0,
      "step": 9730
    },
    {
      "entropy": 5.705477094650268,
      "epoch": 8.363558229480017,
      "grad_norm": 1.1171875,
      "learning_rate": 8.440875831190704e-05,
      "loss": 4.8802,
      "mean_token_accuracy": 0.2351899713277817,
      "num_tokens": 17696846.0,
      "step": 9735
    },
    {
      "entropy": 5.730382871627808,
      "epoch": 8.367855608079072,
      "grad_norm": 1.2578125,
      "learning_rate": 8.423225923902608e-05,
      "loss": 4.8545,
      "mean_token_accuracy": 0.2428334668278694,
      "num_tokens": 17705846.0,
      "step": 9740
    },
    {
      "entropy": 5.678792095184326,
      "epoch": 8.372152986678126,
      "grad_norm": 1.3515625,
      "learning_rate": 8.405617672664175e-05,
      "loss": 4.8015,
      "mean_token_accuracy": 0.2536479622125626,
      "num_tokens": 17714255.0,
      "step": 9745
    },
    {
      "entropy": 5.7832560539245605,
      "epoch": 8.376450365277181,
      "grad_norm": 1.2265625,
      "learning_rate": 8.388051115924781e-05,
      "loss": 4.9787,
      "mean_token_accuracy": 0.22999690473079681,
      "num_tokens": 17724246.0,
      "step": 9750
    },
    {
      "entropy": 5.733207416534424,
      "epoch": 8.380747743876235,
      "grad_norm": 1.328125,
      "learning_rate": 8.370526292042774e-05,
      "loss": 4.8385,
      "mean_token_accuracy": 0.24650791734457017,
      "num_tokens": 17733081.0,
      "step": 9755
    },
    {
      "entropy": 5.665503215789795,
      "epoch": 8.38504512247529,
      "grad_norm": 1.5234375,
      "learning_rate": 8.353043239285357e-05,
      "loss": 4.8383,
      "mean_token_accuracy": 0.2454715371131897,
      "num_tokens": 17741220.0,
      "step": 9760
    },
    {
      "entropy": 5.736371994018555,
      "epoch": 8.389342501074344,
      "grad_norm": 1.4140625,
      "learning_rate": 8.335601995828534e-05,
      "loss": 4.8971,
      "mean_token_accuracy": 0.2377567008137703,
      "num_tokens": 17750057.0,
      "step": 9765
    },
    {
      "entropy": 5.739968490600586,
      "epoch": 8.3936398796734,
      "grad_norm": 1.2890625,
      "learning_rate": 8.318202599757008e-05,
      "loss": 4.9148,
      "mean_token_accuracy": 0.23634277582168578,
      "num_tokens": 17759016.0,
      "step": 9770
    },
    {
      "entropy": 5.708938121795654,
      "epoch": 8.397937258272453,
      "grad_norm": 1.3671875,
      "learning_rate": 8.30084508906411e-05,
      "loss": 4.8183,
      "mean_token_accuracy": 0.24032998383045195,
      "num_tokens": 17767170.0,
      "step": 9775
    },
    {
      "entropy": 5.693941688537597,
      "epoch": 8.402234636871508,
      "grad_norm": 1.3828125,
      "learning_rate": 8.283529501651698e-05,
      "loss": 4.8933,
      "mean_token_accuracy": 0.2311558097600937,
      "num_tokens": 17776508.0,
      "step": 9780
    },
    {
      "entropy": 5.720448160171509,
      "epoch": 8.406532015470564,
      "grad_norm": 1.2734375,
      "learning_rate": 8.266255875330095e-05,
      "loss": 4.8376,
      "mean_token_accuracy": 0.24367564618587495,
      "num_tokens": 17785494.0,
      "step": 9785
    },
    {
      "entropy": 5.707212591171265,
      "epoch": 8.410829394069617,
      "grad_norm": 1.359375,
      "learning_rate": 8.249024247817998e-05,
      "loss": 4.8865,
      "mean_token_accuracy": 0.23894913047552108,
      "num_tokens": 17795158.0,
      "step": 9790
    },
    {
      "entropy": 5.717279958724975,
      "epoch": 8.415126772668673,
      "grad_norm": 1.21875,
      "learning_rate": 8.231834656742402e-05,
      "loss": 4.9086,
      "mean_token_accuracy": 0.2380632221698761,
      "num_tokens": 17805130.0,
      "step": 9795
    },
    {
      "entropy": 5.745273447036743,
      "epoch": 8.419424151267727,
      "grad_norm": 1.359375,
      "learning_rate": 8.214687139638494e-05,
      "loss": 4.9124,
      "mean_token_accuracy": 0.24023134857416154,
      "num_tokens": 17813775.0,
      "step": 9800
    },
    {
      "entropy": 5.7758644104003904,
      "epoch": 8.423721529866782,
      "grad_norm": 1.453125,
      "learning_rate": 8.197581733949597e-05,
      "loss": 4.8666,
      "mean_token_accuracy": 0.25315709263086317,
      "num_tokens": 17822599.0,
      "step": 9805
    },
    {
      "entropy": 5.774426078796386,
      "epoch": 8.428018908465836,
      "grad_norm": 1.4609375,
      "learning_rate": 8.180518477027083e-05,
      "loss": 4.8961,
      "mean_token_accuracy": 0.23910606354475022,
      "num_tokens": 17830742.0,
      "step": 9810
    },
    {
      "entropy": 5.620084571838379,
      "epoch": 8.432316287064891,
      "grad_norm": 1.28125,
      "learning_rate": 8.163497406130287e-05,
      "loss": 4.7803,
      "mean_token_accuracy": 0.24886348992586135,
      "num_tokens": 17840101.0,
      "step": 9815
    },
    {
      "entropy": 5.8178997993469235,
      "epoch": 8.436613665663945,
      "grad_norm": 1.265625,
      "learning_rate": 8.146518558426426e-05,
      "loss": 4.9902,
      "mean_token_accuracy": 0.2252660185098648,
      "num_tokens": 17850464.0,
      "step": 9820
    },
    {
      "entropy": 5.741954326629639,
      "epoch": 8.440911044263,
      "grad_norm": 1.4140625,
      "learning_rate": 8.129581970990507e-05,
      "loss": 4.8553,
      "mean_token_accuracy": 0.24823236614465713,
      "num_tokens": 17858792.0,
      "step": 9825
    },
    {
      "entropy": 5.732809400558471,
      "epoch": 8.445208422862054,
      "grad_norm": 1.4140625,
      "learning_rate": 8.11268768080528e-05,
      "loss": 4.8564,
      "mean_token_accuracy": 0.23900128602981568,
      "num_tokens": 17867225.0,
      "step": 9830
    },
    {
      "entropy": 5.7832074642181395,
      "epoch": 8.44950580146111,
      "grad_norm": 1.4609375,
      "learning_rate": 8.09583572476111e-05,
      "loss": 4.9673,
      "mean_token_accuracy": 0.22997304052114487,
      "num_tokens": 17876423.0,
      "step": 9835
    },
    {
      "entropy": 5.718025875091553,
      "epoch": 8.453803180060163,
      "grad_norm": 1.46875,
      "learning_rate": 8.079026139655946e-05,
      "loss": 4.9255,
      "mean_token_accuracy": 0.23200517892837524,
      "num_tokens": 17886847.0,
      "step": 9840
    },
    {
      "entropy": 5.685693693161011,
      "epoch": 8.458100558659218,
      "grad_norm": 1.296875,
      "learning_rate": 8.062258962195192e-05,
      "loss": 4.8057,
      "mean_token_accuracy": 0.2456534430384636,
      "num_tokens": 17897395.0,
      "step": 9845
    },
    {
      "entropy": 5.726854228973389,
      "epoch": 8.462397937258272,
      "grad_norm": 1.46875,
      "learning_rate": 8.04553422899167e-05,
      "loss": 4.8925,
      "mean_token_accuracy": 0.23469093143939973,
      "num_tokens": 17906190.0,
      "step": 9850
    },
    {
      "entropy": 5.708572816848755,
      "epoch": 8.466695315857327,
      "grad_norm": 1.2890625,
      "learning_rate": 8.028851976565508e-05,
      "loss": 4.946,
      "mean_token_accuracy": 0.23586695939302443,
      "num_tokens": 17916056.0,
      "step": 9855
    },
    {
      "entropy": 5.7173912525177,
      "epoch": 8.470992694456381,
      "grad_norm": 1.3125,
      "learning_rate": 8.01221224134408e-05,
      "loss": 4.8411,
      "mean_token_accuracy": 0.24219889491796492,
      "num_tokens": 17925790.0,
      "step": 9860
    },
    {
      "entropy": 5.743465614318848,
      "epoch": 8.475290073055437,
      "grad_norm": 1.3671875,
      "learning_rate": 7.995615059661907e-05,
      "loss": 4.9206,
      "mean_token_accuracy": 0.2368515193462372,
      "num_tokens": 17935007.0,
      "step": 9865
    },
    {
      "entropy": 5.759069204330444,
      "epoch": 8.47958745165449,
      "grad_norm": 1.3359375,
      "learning_rate": 7.979060467760616e-05,
      "loss": 4.904,
      "mean_token_accuracy": 0.24082816541194915,
      "num_tokens": 17943363.0,
      "step": 9870
    },
    {
      "entropy": 5.689574384689331,
      "epoch": 8.483884830253546,
      "grad_norm": 1.4375,
      "learning_rate": 7.962548501788811e-05,
      "loss": 4.8149,
      "mean_token_accuracy": 0.24400141090154648,
      "num_tokens": 17951667.0,
      "step": 9875
    },
    {
      "entropy": 5.72466893196106,
      "epoch": 8.4881822088526,
      "grad_norm": 1.390625,
      "learning_rate": 7.946079197802028e-05,
      "loss": 4.9225,
      "mean_token_accuracy": 0.23353277146816254,
      "num_tokens": 17960920.0,
      "step": 9880
    },
    {
      "entropy": 5.713424777984619,
      "epoch": 8.492479587451655,
      "grad_norm": 1.5,
      "learning_rate": 7.929652591762636e-05,
      "loss": 4.911,
      "mean_token_accuracy": 0.2364127442240715,
      "num_tokens": 17969462.0,
      "step": 9885
    },
    {
      "entropy": 5.730654716491699,
      "epoch": 8.496776966050708,
      "grad_norm": 1.4296875,
      "learning_rate": 7.913268719539785e-05,
      "loss": 4.9171,
      "mean_token_accuracy": 0.2429963693022728,
      "num_tokens": 17978608.0,
      "step": 9890
    },
    {
      "entropy": 5.689770078659057,
      "epoch": 8.501074344649764,
      "grad_norm": 1.359375,
      "learning_rate": 7.896927616909304e-05,
      "loss": 4.7824,
      "mean_token_accuracy": 0.2481196016073227,
      "num_tokens": 17987563.0,
      "step": 9895
    },
    {
      "entropy": 5.751080274581909,
      "epoch": 8.505371723248818,
      "grad_norm": 1.25,
      "learning_rate": 7.880629319553623e-05,
      "loss": 4.8956,
      "mean_token_accuracy": 0.24801230430603027,
      "num_tokens": 17996493.0,
      "step": 9900
    },
    {
      "entropy": 5.73831958770752,
      "epoch": 8.509669101847873,
      "grad_norm": 1.296875,
      "learning_rate": 7.864373863061717e-05,
      "loss": 4.8902,
      "mean_token_accuracy": 0.24250307977199553,
      "num_tokens": 18004578.0,
      "step": 9905
    },
    {
      "entropy": 5.7382752895355225,
      "epoch": 8.513966480446927,
      "grad_norm": 1.4453125,
      "learning_rate": 7.848161282929006e-05,
      "loss": 4.8881,
      "mean_token_accuracy": 0.2407060906291008,
      "num_tokens": 18013366.0,
      "step": 9910
    },
    {
      "entropy": 5.776788902282715,
      "epoch": 8.518263859045982,
      "grad_norm": 1.3671875,
      "learning_rate": 7.831991614557274e-05,
      "loss": 4.903,
      "mean_token_accuracy": 0.23019613772630693,
      "num_tokens": 18021718.0,
      "step": 9915
    },
    {
      "entropy": 5.7528174877166744,
      "epoch": 8.522561237645036,
      "grad_norm": 1.359375,
      "learning_rate": 7.815864893254619e-05,
      "loss": 4.891,
      "mean_token_accuracy": 0.2360119715332985,
      "num_tokens": 18030770.0,
      "step": 9920
    },
    {
      "entropy": 5.7316086769104,
      "epoch": 8.526858616244091,
      "grad_norm": 1.3203125,
      "learning_rate": 7.799781154235361e-05,
      "loss": 4.9092,
      "mean_token_accuracy": 0.24154060184955597,
      "num_tokens": 18039009.0,
      "step": 9925
    },
    {
      "entropy": 5.808560800552368,
      "epoch": 8.531155994843147,
      "grad_norm": 1.453125,
      "learning_rate": 7.783740432619954e-05,
      "loss": 4.9671,
      "mean_token_accuracy": 0.22287265360355377,
      "num_tokens": 18046677.0,
      "step": 9930
    },
    {
      "entropy": 5.714200210571289,
      "epoch": 8.5354533734422,
      "grad_norm": 1.3828125,
      "learning_rate": 7.767742763434922e-05,
      "loss": 4.9079,
      "mean_token_accuracy": 0.23346690237522125,
      "num_tokens": 18056772.0,
      "step": 9935
    },
    {
      "entropy": 5.666274166107177,
      "epoch": 8.539750752041256,
      "grad_norm": 1.28125,
      "learning_rate": 7.75178818161277e-05,
      "loss": 4.7845,
      "mean_token_accuracy": 0.24585407674312593,
      "num_tokens": 18065387.0,
      "step": 9940
    },
    {
      "entropy": 5.755437517166138,
      "epoch": 8.54404813064031,
      "grad_norm": 1.390625,
      "learning_rate": 7.735876721991945e-05,
      "loss": 4.9507,
      "mean_token_accuracy": 0.23148660957813263,
      "num_tokens": 18074495.0,
      "step": 9945
    },
    {
      "entropy": 5.743089199066162,
      "epoch": 8.548345509239365,
      "grad_norm": 1.2265625,
      "learning_rate": 7.720008419316708e-05,
      "loss": 4.9242,
      "mean_token_accuracy": 0.2357686460018158,
      "num_tokens": 18083351.0,
      "step": 9950
    },
    {
      "entropy": 5.700260353088379,
      "epoch": 8.552642887838418,
      "grad_norm": 1.2734375,
      "learning_rate": 7.704183308237089e-05,
      "loss": 4.8213,
      "mean_token_accuracy": 0.24597719609737395,
      "num_tokens": 18092632.0,
      "step": 9955
    },
    {
      "entropy": 5.662213611602783,
      "epoch": 8.556940266437474,
      "grad_norm": 1.2421875,
      "learning_rate": 7.688401423308799e-05,
      "loss": 4.8233,
      "mean_token_accuracy": 0.2466511309146881,
      "num_tokens": 18102380.0,
      "step": 9960
    },
    {
      "entropy": 5.671461486816407,
      "epoch": 8.561237645036528,
      "grad_norm": 1.4296875,
      "learning_rate": 7.672662798993174e-05,
      "loss": 4.8247,
      "mean_token_accuracy": 0.2511079117655754,
      "num_tokens": 18111254.0,
      "step": 9965
    },
    {
      "entropy": 5.7380656719207765,
      "epoch": 8.565535023635583,
      "grad_norm": 1.328125,
      "learning_rate": 7.656967469657083e-05,
      "loss": 4.8936,
      "mean_token_accuracy": 0.242263326048851,
      "num_tokens": 18119998.0,
      "step": 9970
    },
    {
      "entropy": 5.707980823516846,
      "epoch": 8.569832402234637,
      "grad_norm": 1.328125,
      "learning_rate": 7.641315469572841e-05,
      "loss": 4.8431,
      "mean_token_accuracy": 0.24148163497447966,
      "num_tokens": 18129237.0,
      "step": 9975
    },
    {
      "entropy": 5.769561052322388,
      "epoch": 8.574129780833692,
      "grad_norm": 1.1953125,
      "learning_rate": 7.625706832918172e-05,
      "loss": 4.8962,
      "mean_token_accuracy": 0.24106623828411103,
      "num_tokens": 18138440.0,
      "step": 9980
    },
    {
      "entropy": 5.753009605407715,
      "epoch": 8.578427159432746,
      "grad_norm": 1.3359375,
      "learning_rate": 7.610141593776091e-05,
      "loss": 4.996,
      "mean_token_accuracy": 0.2327938124537468,
      "num_tokens": 18147712.0,
      "step": 9985
    },
    {
      "entropy": 5.748133611679077,
      "epoch": 8.582724538031801,
      "grad_norm": 1.3984375,
      "learning_rate": 7.59461978613486e-05,
      "loss": 4.9078,
      "mean_token_accuracy": 0.23580171167850494,
      "num_tokens": 18156710.0,
      "step": 9990
    },
    {
      "entropy": 5.725798463821411,
      "epoch": 8.587021916630855,
      "grad_norm": 1.4140625,
      "learning_rate": 7.579141443887901e-05,
      "loss": 4.8523,
      "mean_token_accuracy": 0.2376137688755989,
      "num_tokens": 18165412.0,
      "step": 9995
    },
    {
      "entropy": 5.7189679622650145,
      "epoch": 8.59131929522991,
      "grad_norm": 1.4765625,
      "learning_rate": 7.563706600833737e-05,
      "loss": 4.9063,
      "mean_token_accuracy": 0.2349215939640999,
      "num_tokens": 18174415.0,
      "step": 10000
    },
    {
      "epoch": 8.59131929522991,
      "eval_entropy": 5.534239016137682,
      "eval_loss": 5.892942428588867,
      "eval_mean_token_accuracy": 0.18259063309377377,
      "eval_num_tokens": 18174415.0,
      "eval_runtime": 2.0512,
      "eval_samples_per_second": 1730.231,
      "eval_steps_per_second": 216.462,
      "step": 10000
    },
    {
      "entropy": 5.641706037521362,
      "epoch": 8.595616673828964,
      "grad_norm": 1.4921875,
      "learning_rate": 7.548315290675886e-05,
      "loss": 4.8364,
      "mean_token_accuracy": 0.24370339065790175,
      "num_tokens": 18183468.0,
      "step": 10005
    },
    {
      "entropy": 5.677710723876953,
      "epoch": 8.59991405242802,
      "grad_norm": 1.265625,
      "learning_rate": 7.532967547022825e-05,
      "loss": 4.8421,
      "mean_token_accuracy": 0.24187945276498796,
      "num_tokens": 18192552.0,
      "step": 10010
    },
    {
      "entropy": 5.703868246078491,
      "epoch": 8.604211431027073,
      "grad_norm": 1.4296875,
      "learning_rate": 7.517663403387874e-05,
      "loss": 4.9387,
      "mean_token_accuracy": 0.24121222794055938,
      "num_tokens": 18202056.0,
      "step": 10015
    },
    {
      "entropy": 5.684072017669678,
      "epoch": 8.608508809626128,
      "grad_norm": 1.390625,
      "learning_rate": 7.502402893189191e-05,
      "loss": 4.8142,
      "mean_token_accuracy": 0.24965679943561553,
      "num_tokens": 18210329.0,
      "step": 10020
    },
    {
      "entropy": 5.795586156845093,
      "epoch": 8.612806188225182,
      "grad_norm": 1.2421875,
      "learning_rate": 7.487186049749618e-05,
      "loss": 4.94,
      "mean_token_accuracy": 0.2302323043346405,
      "num_tokens": 18219143.0,
      "step": 10025
    },
    {
      "entropy": 5.720360326766968,
      "epoch": 8.617103566824238,
      "grad_norm": 1.296875,
      "learning_rate": 7.472012906296658e-05,
      "loss": 4.8775,
      "mean_token_accuracy": 0.2409013941884041,
      "num_tokens": 18227679.0,
      "step": 10030
    },
    {
      "entropy": 5.712895488739013,
      "epoch": 8.621400945423291,
      "grad_norm": 1.140625,
      "learning_rate": 7.45688349596239e-05,
      "loss": 4.8676,
      "mean_token_accuracy": 0.24006159454584122,
      "num_tokens": 18237522.0,
      "step": 10035
    },
    {
      "entropy": 5.727213191986084,
      "epoch": 8.625698324022347,
      "grad_norm": 1.3515625,
      "learning_rate": 7.441797851783402e-05,
      "loss": 4.91,
      "mean_token_accuracy": 0.23364521861076354,
      "num_tokens": 18245969.0,
      "step": 10040
    },
    {
      "entropy": 5.717015218734741,
      "epoch": 8.6299957026214,
      "grad_norm": 1.3984375,
      "learning_rate": 7.426756006700716e-05,
      "loss": 4.8539,
      "mean_token_accuracy": 0.2373197704553604,
      "num_tokens": 18254886.0,
      "step": 10045
    },
    {
      "entropy": 5.744679689407349,
      "epoch": 8.634293081220456,
      "grad_norm": 1.5078125,
      "learning_rate": 7.411757993559702e-05,
      "loss": 4.9657,
      "mean_token_accuracy": 0.233807535469532,
      "num_tokens": 18264222.0,
      "step": 10050
    },
    {
      "entropy": 5.769802045822144,
      "epoch": 8.63859045981951,
      "grad_norm": 1.375,
      "learning_rate": 7.396803845110032e-05,
      "loss": 4.9244,
      "mean_token_accuracy": 0.24051254242658615,
      "num_tokens": 18273548.0,
      "step": 10055
    },
    {
      "entropy": 5.7355225563049315,
      "epoch": 8.642887838418565,
      "grad_norm": 1.34375,
      "learning_rate": 7.381893594005585e-05,
      "loss": 4.888,
      "mean_token_accuracy": 0.23913937211036682,
      "num_tokens": 18282842.0,
      "step": 10060
    },
    {
      "entropy": 5.714138078689575,
      "epoch": 8.647185217017618,
      "grad_norm": 1.359375,
      "learning_rate": 7.367027272804387e-05,
      "loss": 4.8535,
      "mean_token_accuracy": 0.23634449243545533,
      "num_tokens": 18291133.0,
      "step": 10065
    },
    {
      "entropy": 5.708312606811523,
      "epoch": 8.651482595616674,
      "grad_norm": 1.3203125,
      "learning_rate": 7.352204913968546e-05,
      "loss": 4.8902,
      "mean_token_accuracy": 0.2456353023648262,
      "num_tokens": 18300936.0,
      "step": 10070
    },
    {
      "entropy": 5.745303821563721,
      "epoch": 8.65577997421573,
      "grad_norm": 1.171875,
      "learning_rate": 7.337426549864175e-05,
      "loss": 4.9334,
      "mean_token_accuracy": 0.2344992771744728,
      "num_tokens": 18311372.0,
      "step": 10075
    },
    {
      "entropy": 5.734985256195069,
      "epoch": 8.660077352814783,
      "grad_norm": 1.234375,
      "learning_rate": 7.322692212761306e-05,
      "loss": 4.9354,
      "mean_token_accuracy": 0.2361741617321968,
      "num_tokens": 18321166.0,
      "step": 10080
    },
    {
      "entropy": 5.7485943794250485,
      "epoch": 8.664374731413837,
      "grad_norm": 1.40625,
      "learning_rate": 7.308001934833844e-05,
      "loss": 4.8799,
      "mean_token_accuracy": 0.23693183660507203,
      "num_tokens": 18330399.0,
      "step": 10085
    },
    {
      "entropy": 5.751676893234253,
      "epoch": 8.668672110012892,
      "grad_norm": 1.375,
      "learning_rate": 7.29335574815948e-05,
      "loss": 4.8746,
      "mean_token_accuracy": 0.2360835403203964,
      "num_tokens": 18338920.0,
      "step": 10090
    },
    {
      "entropy": 5.717150449752808,
      "epoch": 8.672969488611947,
      "grad_norm": 1.2421875,
      "learning_rate": 7.278753684719636e-05,
      "loss": 4.8798,
      "mean_token_accuracy": 0.23987387418746947,
      "num_tokens": 18348515.0,
      "step": 10095
    },
    {
      "entropy": 5.727327728271485,
      "epoch": 8.677266867211001,
      "grad_norm": 1.3828125,
      "learning_rate": 7.264195776399386e-05,
      "loss": 4.9107,
      "mean_token_accuracy": 0.2364715039730072,
      "num_tokens": 18357353.0,
      "step": 10100
    },
    {
      "entropy": 5.72551212310791,
      "epoch": 8.681564245810057,
      "grad_norm": 1.359375,
      "learning_rate": 7.249682054987381e-05,
      "loss": 4.8387,
      "mean_token_accuracy": 0.24308911412954332,
      "num_tokens": 18366434.0,
      "step": 10105
    },
    {
      "entropy": 5.7901074409484865,
      "epoch": 8.68586162440911,
      "grad_norm": 1.2734375,
      "learning_rate": 7.23521255217578e-05,
      "loss": 4.9607,
      "mean_token_accuracy": 0.22461001873016356,
      "num_tokens": 18375589.0,
      "step": 10110
    },
    {
      "entropy": 5.70100040435791,
      "epoch": 8.690159003008166,
      "grad_norm": 1.28125,
      "learning_rate": 7.220787299560205e-05,
      "loss": 4.8989,
      "mean_token_accuracy": 0.24264244586229325,
      "num_tokens": 18384904.0,
      "step": 10115
    },
    {
      "entropy": 5.7278729438781735,
      "epoch": 8.69445638160722,
      "grad_norm": 1.375,
      "learning_rate": 7.20640632863963e-05,
      "loss": 4.7711,
      "mean_token_accuracy": 0.24757508635520936,
      "num_tokens": 18393833.0,
      "step": 10120
    },
    {
      "entropy": 5.711657571792602,
      "epoch": 8.698753760206275,
      "grad_norm": 1.328125,
      "learning_rate": 7.192069670816359e-05,
      "loss": 4.9244,
      "mean_token_accuracy": 0.23145184218883513,
      "num_tokens": 18403444.0,
      "step": 10125
    },
    {
      "entropy": 5.706142997741699,
      "epoch": 8.703051138805328,
      "grad_norm": 1.4921875,
      "learning_rate": 7.177777357395912e-05,
      "loss": 4.869,
      "mean_token_accuracy": 0.24197168052196502,
      "num_tokens": 18412091.0,
      "step": 10130
    },
    {
      "entropy": 5.698826265335083,
      "epoch": 8.707348517404384,
      "grad_norm": 1.375,
      "learning_rate": 7.163529419587002e-05,
      "loss": 4.8753,
      "mean_token_accuracy": 0.23961719274520873,
      "num_tokens": 18421655.0,
      "step": 10135
    },
    {
      "entropy": 5.773096561431885,
      "epoch": 8.711645896003438,
      "grad_norm": 1.328125,
      "learning_rate": 7.149325888501418e-05,
      "loss": 4.9558,
      "mean_token_accuracy": 0.23473044633865356,
      "num_tokens": 18431239.0,
      "step": 10140
    },
    {
      "entropy": 5.736195468902588,
      "epoch": 8.715943274602493,
      "grad_norm": 1.234375,
      "learning_rate": 7.135166795153992e-05,
      "loss": 4.8505,
      "mean_token_accuracy": 0.24684498012065886,
      "num_tokens": 18440572.0,
      "step": 10145
    },
    {
      "entropy": 5.727258586883545,
      "epoch": 8.720240653201547,
      "grad_norm": 1.359375,
      "learning_rate": 7.121052170462541e-05,
      "loss": 4.9381,
      "mean_token_accuracy": 0.23492728620767594,
      "num_tokens": 18449245.0,
      "step": 10150
    },
    {
      "entropy": 5.653642177581787,
      "epoch": 8.724538031800602,
      "grad_norm": 1.4453125,
      "learning_rate": 7.106982045247754e-05,
      "loss": 4.7844,
      "mean_token_accuracy": 0.2551065072417259,
      "num_tokens": 18457961.0,
      "step": 10155
    },
    {
      "entropy": 5.728663206100464,
      "epoch": 8.728835410399656,
      "grad_norm": 1.265625,
      "learning_rate": 7.092956450233162e-05,
      "loss": 4.9391,
      "mean_token_accuracy": 0.24017365425825118,
      "num_tokens": 18467956.0,
      "step": 10160
    },
    {
      "entropy": 5.741170835494995,
      "epoch": 8.733132788998711,
      "grad_norm": 1.265625,
      "learning_rate": 7.078975416045055e-05,
      "loss": 4.8778,
      "mean_token_accuracy": 0.23747024983167647,
      "num_tokens": 18477313.0,
      "step": 10165
    },
    {
      "entropy": 5.669637441635132,
      "epoch": 8.737430167597765,
      "grad_norm": 1.53125,
      "learning_rate": 7.065038973212424e-05,
      "loss": 4.7584,
      "mean_token_accuracy": 0.24721809923648835,
      "num_tokens": 18485690.0,
      "step": 10170
    },
    {
      "entropy": 5.723529863357544,
      "epoch": 8.74172754619682,
      "grad_norm": 1.4375,
      "learning_rate": 7.051147152166896e-05,
      "loss": 4.879,
      "mean_token_accuracy": 0.24353028535842897,
      "num_tokens": 18494150.0,
      "step": 10175
    },
    {
      "entropy": 5.661523246765137,
      "epoch": 8.746024924795874,
      "grad_norm": 1.140625,
      "learning_rate": 7.037299983242652e-05,
      "loss": 4.8357,
      "mean_token_accuracy": 0.246757473051548,
      "num_tokens": 18504091.0,
      "step": 10180
    },
    {
      "entropy": 5.713498783111572,
      "epoch": 8.75032230339493,
      "grad_norm": 1.3203125,
      "learning_rate": 7.023497496676371e-05,
      "loss": 4.8883,
      "mean_token_accuracy": 0.23697651624679567,
      "num_tokens": 18513695.0,
      "step": 10185
    },
    {
      "entropy": 5.72203483581543,
      "epoch": 8.754619681993983,
      "grad_norm": 1.21875,
      "learning_rate": 7.009739722607173e-05,
      "loss": 4.8597,
      "mean_token_accuracy": 0.24536042660474777,
      "num_tokens": 18522551.0,
      "step": 10190
    },
    {
      "entropy": 5.74442343711853,
      "epoch": 8.758917060593038,
      "grad_norm": 1.265625,
      "learning_rate": 6.996026691076531e-05,
      "loss": 5.0015,
      "mean_token_accuracy": 0.22911013662815094,
      "num_tokens": 18533290.0,
      "step": 10195
    },
    {
      "entropy": 5.722520256042481,
      "epoch": 8.763214439192092,
      "grad_norm": 1.2890625,
      "learning_rate": 6.982358432028234e-05,
      "loss": 4.8089,
      "mean_token_accuracy": 0.24856774657964706,
      "num_tokens": 18542990.0,
      "step": 10200
    },
    {
      "entropy": 5.815678882598877,
      "epoch": 8.767511817791148,
      "grad_norm": 1.3515625,
      "learning_rate": 6.968734975308283e-05,
      "loss": 4.9813,
      "mean_token_accuracy": 0.22747452706098556,
      "num_tokens": 18552525.0,
      "step": 10205
    },
    {
      "entropy": 5.694711017608642,
      "epoch": 8.771809196390201,
      "grad_norm": 1.34375,
      "learning_rate": 6.955156350664876e-05,
      "loss": 4.8089,
      "mean_token_accuracy": 0.24590101689100266,
      "num_tokens": 18561741.0,
      "step": 10210
    },
    {
      "entropy": 5.781192684173584,
      "epoch": 8.776106574989257,
      "grad_norm": 1.359375,
      "learning_rate": 6.941622587748298e-05,
      "loss": 4.951,
      "mean_token_accuracy": 0.22873226106166838,
      "num_tokens": 18570130.0,
      "step": 10215
    },
    {
      "entropy": 5.702016830444336,
      "epoch": 8.780403953588312,
      "grad_norm": 1.3359375,
      "learning_rate": 6.928133716110863e-05,
      "loss": 4.8938,
      "mean_token_accuracy": 0.23844954818487168,
      "num_tokens": 18579149.0,
      "step": 10220
    },
    {
      "entropy": 5.728512763977051,
      "epoch": 8.784701332187366,
      "grad_norm": 1.359375,
      "learning_rate": 6.914689765206899e-05,
      "loss": 4.7987,
      "mean_token_accuracy": 0.24630094319581985,
      "num_tokens": 18587604.0,
      "step": 10225
    },
    {
      "entropy": 5.751327037811279,
      "epoch": 8.78899871078642,
      "grad_norm": 1.421875,
      "learning_rate": 6.901290764392609e-05,
      "loss": 4.8531,
      "mean_token_accuracy": 0.24052831828594207,
      "num_tokens": 18596056.0,
      "step": 10230
    },
    {
      "entropy": 5.749263572692871,
      "epoch": 8.793296089385475,
      "grad_norm": 1.1640625,
      "learning_rate": 6.887936742926058e-05,
      "loss": 4.9199,
      "mean_token_accuracy": 0.23125884532928467,
      "num_tokens": 18605575.0,
      "step": 10235
    },
    {
      "entropy": 5.678548955917359,
      "epoch": 8.79759346798453,
      "grad_norm": 1.3046875,
      "learning_rate": 6.874627729967086e-05,
      "loss": 4.8344,
      "mean_token_accuracy": 0.24363392889499663,
      "num_tokens": 18614206.0,
      "step": 10240
    },
    {
      "entropy": 5.715464544296265,
      "epoch": 8.801890846583584,
      "grad_norm": 1.203125,
      "learning_rate": 6.861363754577262e-05,
      "loss": 4.8535,
      "mean_token_accuracy": 0.23847707211971284,
      "num_tokens": 18623067.0,
      "step": 10245
    },
    {
      "entropy": 5.643958950042725,
      "epoch": 8.80618822518264,
      "grad_norm": 1.453125,
      "learning_rate": 6.848144845719808e-05,
      "loss": 4.8387,
      "mean_token_accuracy": 0.24321892559528352,
      "num_tokens": 18632266.0,
      "step": 10250
    },
    {
      "entropy": 5.664030075073242,
      "epoch": 8.810485603781693,
      "grad_norm": 1.3984375,
      "learning_rate": 6.834971032259537e-05,
      "loss": 4.8397,
      "mean_token_accuracy": 0.24072007983922958,
      "num_tokens": 18640516.0,
      "step": 10255
    },
    {
      "entropy": 5.690808057785034,
      "epoch": 8.814782982380748,
      "grad_norm": 1.390625,
      "learning_rate": 6.821842342962786e-05,
      "loss": 4.8475,
      "mean_token_accuracy": 0.24447838515043258,
      "num_tokens": 18649368.0,
      "step": 10260
    },
    {
      "entropy": 5.677191877365113,
      "epoch": 8.819080360979802,
      "grad_norm": 1.359375,
      "learning_rate": 6.808758806497375e-05,
      "loss": 4.8447,
      "mean_token_accuracy": 0.2524683356285095,
      "num_tokens": 18658306.0,
      "step": 10265
    },
    {
      "entropy": 5.745322799682617,
      "epoch": 8.823377739578858,
      "grad_norm": 1.375,
      "learning_rate": 6.795720451432509e-05,
      "loss": 4.8406,
      "mean_token_accuracy": 0.24409451335668564,
      "num_tokens": 18668676.0,
      "step": 10270
    },
    {
      "entropy": 5.740674781799316,
      "epoch": 8.827675118177911,
      "grad_norm": 1.28125,
      "learning_rate": 6.782727306238749e-05,
      "loss": 4.9852,
      "mean_token_accuracy": 0.22357712537050248,
      "num_tokens": 18677840.0,
      "step": 10275
    },
    {
      "entropy": 5.644893312454224,
      "epoch": 8.831972496776967,
      "grad_norm": 1.421875,
      "learning_rate": 6.769779399287928e-05,
      "loss": 4.7822,
      "mean_token_accuracy": 0.24160815626382828,
      "num_tokens": 18685733.0,
      "step": 10280
    },
    {
      "entropy": 5.71365647315979,
      "epoch": 8.83626987537602,
      "grad_norm": 1.3046875,
      "learning_rate": 6.756876758853104e-05,
      "loss": 4.8656,
      "mean_token_accuracy": 0.2417924225330353,
      "num_tokens": 18694648.0,
      "step": 10285
    },
    {
      "entropy": 5.710398960113525,
      "epoch": 8.840567253975076,
      "grad_norm": 1.5546875,
      "learning_rate": 6.744019413108486e-05,
      "loss": 4.8521,
      "mean_token_accuracy": 0.2444690704345703,
      "num_tokens": 18702758.0,
      "step": 10290
    },
    {
      "entropy": 5.7455525398254395,
      "epoch": 8.84486463257413,
      "grad_norm": 1.3671875,
      "learning_rate": 6.731207390129366e-05,
      "loss": 4.9434,
      "mean_token_accuracy": 0.23295524716377258,
      "num_tokens": 18712257.0,
      "step": 10295
    },
    {
      "entropy": 5.734651517868042,
      "epoch": 8.849162011173185,
      "grad_norm": 1.1875,
      "learning_rate": 6.7184407178921e-05,
      "loss": 4.888,
      "mean_token_accuracy": 0.23881399929523467,
      "num_tokens": 18722337.0,
      "step": 10300
    },
    {
      "entropy": 5.749492120742798,
      "epoch": 8.853459389772238,
      "grad_norm": 1.4453125,
      "learning_rate": 6.70571942427399e-05,
      "loss": 4.9559,
      "mean_token_accuracy": 0.23205118924379348,
      "num_tokens": 18731260.0,
      "step": 10305
    },
    {
      "entropy": 5.75962724685669,
      "epoch": 8.857756768371294,
      "grad_norm": 1.3515625,
      "learning_rate": 6.693043537053254e-05,
      "loss": 4.9813,
      "mean_token_accuracy": 0.23050196915864946,
      "num_tokens": 18741357.0,
      "step": 10310
    },
    {
      "entropy": 5.70353946685791,
      "epoch": 8.862054146970348,
      "grad_norm": 1.328125,
      "learning_rate": 6.68041308390896e-05,
      "loss": 4.929,
      "mean_token_accuracy": 0.23658778071403502,
      "num_tokens": 18751694.0,
      "step": 10315
    },
    {
      "entropy": 5.786187076568604,
      "epoch": 8.866351525569403,
      "grad_norm": 1.2421875,
      "learning_rate": 6.667828092420977e-05,
      "loss": 4.9266,
      "mean_token_accuracy": 0.2431049346923828,
      "num_tokens": 18760679.0,
      "step": 10320
    },
    {
      "entropy": 5.698891353607178,
      "epoch": 8.870648904168457,
      "grad_norm": 1.2734375,
      "learning_rate": 6.655288590069889e-05,
      "loss": 4.8974,
      "mean_token_accuracy": 0.24344006329774856,
      "num_tokens": 18769650.0,
      "step": 10325
    },
    {
      "entropy": 5.65026216506958,
      "epoch": 8.874946282767512,
      "grad_norm": 1.2890625,
      "learning_rate": 6.642794604236965e-05,
      "loss": 4.8883,
      "mean_token_accuracy": 0.2407523512840271,
      "num_tokens": 18780076.0,
      "step": 10330
    },
    {
      "entropy": 5.705488920211792,
      "epoch": 8.879243661366566,
      "grad_norm": 1.3046875,
      "learning_rate": 6.630346162204069e-05,
      "loss": 4.8524,
      "mean_token_accuracy": 0.2402828097343445,
      "num_tokens": 18789034.0,
      "step": 10335
    },
    {
      "entropy": 5.685500717163086,
      "epoch": 8.883541039965621,
      "grad_norm": 1.484375,
      "learning_rate": 6.617943291153631e-05,
      "loss": 4.7906,
      "mean_token_accuracy": 0.2516307681798935,
      "num_tokens": 18797916.0,
      "step": 10340
    },
    {
      "entropy": 5.78669924736023,
      "epoch": 8.887838418564675,
      "grad_norm": 1.515625,
      "learning_rate": 6.605586018168558e-05,
      "loss": 4.9008,
      "mean_token_accuracy": 0.24112697392702104,
      "num_tokens": 18806895.0,
      "step": 10345
    },
    {
      "entropy": 5.7594961643219,
      "epoch": 8.89213579716373,
      "grad_norm": 1.25,
      "learning_rate": 6.593274370232191e-05,
      "loss": 4.888,
      "mean_token_accuracy": 0.23973492234945298,
      "num_tokens": 18816049.0,
      "step": 10350
    },
    {
      "entropy": 5.6986674785614015,
      "epoch": 8.896433175762784,
      "grad_norm": 1.265625,
      "learning_rate": 6.581008374228255e-05,
      "loss": 4.8829,
      "mean_token_accuracy": 0.2380861073732376,
      "num_tokens": 18825491.0,
      "step": 10355
    },
    {
      "entropy": 5.688055086135864,
      "epoch": 8.90073055436184,
      "grad_norm": 1.3125,
      "learning_rate": 6.568788056940785e-05,
      "loss": 4.8894,
      "mean_token_accuracy": 0.239205239713192,
      "num_tokens": 18834677.0,
      "step": 10360
    },
    {
      "entropy": 5.6841371059417725,
      "epoch": 8.905027932960895,
      "grad_norm": 1.28125,
      "learning_rate": 6.556613445054063e-05,
      "loss": 4.7971,
      "mean_token_accuracy": 0.24701832830905915,
      "num_tokens": 18844283.0,
      "step": 10365
    },
    {
      "entropy": 5.7468496799469,
      "epoch": 8.909325311559948,
      "grad_norm": 1.3046875,
      "learning_rate": 6.544484565152577e-05,
      "loss": 4.8709,
      "mean_token_accuracy": 0.24514500945806503,
      "num_tokens": 18853403.0,
      "step": 10370
    },
    {
      "entropy": 5.707927703857422,
      "epoch": 8.913622690159002,
      "grad_norm": 1.484375,
      "learning_rate": 6.532401443720951e-05,
      "loss": 4.8844,
      "mean_token_accuracy": 0.24281774312257767,
      "num_tokens": 18861530.0,
      "step": 10375
    },
    {
      "entropy": 5.734713888168335,
      "epoch": 8.917920068758058,
      "grad_norm": 1.171875,
      "learning_rate": 6.520364107143898e-05,
      "loss": 4.9245,
      "mean_token_accuracy": 0.24035619348287582,
      "num_tokens": 18871495.0,
      "step": 10380
    },
    {
      "entropy": 5.757951641082764,
      "epoch": 8.922217447357113,
      "grad_norm": 1.3046875,
      "learning_rate": 6.50837258170615e-05,
      "loss": 4.914,
      "mean_token_accuracy": 0.2439088225364685,
      "num_tokens": 18881099.0,
      "step": 10385
    },
    {
      "entropy": 5.776555347442627,
      "epoch": 8.926514825956167,
      "grad_norm": 1.2890625,
      "learning_rate": 6.496426893592397e-05,
      "loss": 4.9279,
      "mean_token_accuracy": 0.22435057163238525,
      "num_tokens": 18890368.0,
      "step": 10390
    },
    {
      "entropy": 5.771154832839966,
      "epoch": 8.930812204555222,
      "grad_norm": 1.421875,
      "learning_rate": 6.484527068887258e-05,
      "loss": 4.9245,
      "mean_token_accuracy": 0.22901748716831208,
      "num_tokens": 18898576.0,
      "step": 10395
    },
    {
      "entropy": 5.7181800365447994,
      "epoch": 8.935109583154276,
      "grad_norm": 1.4453125,
      "learning_rate": 6.472673133575181e-05,
      "loss": 4.8585,
      "mean_token_accuracy": 0.23982157111167907,
      "num_tokens": 18907276.0,
      "step": 10400
    },
    {
      "entropy": 5.724564266204834,
      "epoch": 8.939406961753331,
      "grad_norm": 1.4140625,
      "learning_rate": 6.460865113540437e-05,
      "loss": 4.8424,
      "mean_token_accuracy": 0.24264838844537734,
      "num_tokens": 18915920.0,
      "step": 10405
    },
    {
      "entropy": 5.750702953338623,
      "epoch": 8.943704340352385,
      "grad_norm": 1.34375,
      "learning_rate": 6.449103034567011e-05,
      "loss": 4.8413,
      "mean_token_accuracy": 0.24670601040124893,
      "num_tokens": 18925529.0,
      "step": 10410
    },
    {
      "entropy": 5.744922637939453,
      "epoch": 8.94800171895144,
      "grad_norm": 1.375,
      "learning_rate": 6.437386922338591e-05,
      "loss": 4.9222,
      "mean_token_accuracy": 0.23244228661060334,
      "num_tokens": 18934859.0,
      "step": 10415
    },
    {
      "entropy": 5.760837602615356,
      "epoch": 8.952299097550494,
      "grad_norm": 1.359375,
      "learning_rate": 6.425716802438479e-05,
      "loss": 4.9191,
      "mean_token_accuracy": 0.23593176901340485,
      "num_tokens": 18943804.0,
      "step": 10420
    },
    {
      "entropy": 5.6986171245574955,
      "epoch": 8.95659647614955,
      "grad_norm": 1.3984375,
      "learning_rate": 6.414092700349548e-05,
      "loss": 4.8862,
      "mean_token_accuracy": 0.2431177631020546,
      "num_tokens": 18954290.0,
      "step": 10425
    },
    {
      "entropy": 5.7580320835113525,
      "epoch": 8.960893854748603,
      "grad_norm": 1.390625,
      "learning_rate": 6.402514641454192e-05,
      "loss": 4.9962,
      "mean_token_accuracy": 0.22172346264123916,
      "num_tokens": 18963448.0,
      "step": 10430
    },
    {
      "entropy": 5.723747587203979,
      "epoch": 8.965191233347658,
      "grad_norm": 1.265625,
      "learning_rate": 6.390982651034274e-05,
      "loss": 4.8953,
      "mean_token_accuracy": 0.23904290795326233,
      "num_tokens": 18972314.0,
      "step": 10435
    },
    {
      "entropy": 5.672523498535156,
      "epoch": 8.969488611946712,
      "grad_norm": 1.46875,
      "learning_rate": 6.379496754271044e-05,
      "loss": 4.7194,
      "mean_token_accuracy": 0.26136820316314696,
      "num_tokens": 18980720.0,
      "step": 10440
    },
    {
      "entropy": 5.65635724067688,
      "epoch": 8.973785990545768,
      "grad_norm": 1.4375,
      "learning_rate": 6.368056976245107e-05,
      "loss": 4.7667,
      "mean_token_accuracy": 0.25870931446552276,
      "num_tokens": 18988267.0,
      "step": 10445
    },
    {
      "entropy": 5.698014831542968,
      "epoch": 8.978083369144821,
      "grad_norm": 1.421875,
      "learning_rate": 6.356663341936368e-05,
      "loss": 4.8663,
      "mean_token_accuracy": 0.24248220026493073,
      "num_tokens": 18997036.0,
      "step": 10450
    },
    {
      "entropy": 5.697885036468506,
      "epoch": 8.982380747743877,
      "grad_norm": 1.3828125,
      "learning_rate": 6.345315876223977e-05,
      "loss": 4.8258,
      "mean_token_accuracy": 0.24143291264772415,
      "num_tokens": 19005440.0,
      "step": 10455
    },
    {
      "entropy": 5.692233467102051,
      "epoch": 8.98667812634293,
      "grad_norm": 1.3359375,
      "learning_rate": 6.334014603886256e-05,
      "loss": 4.881,
      "mean_token_accuracy": 0.24324591606855392,
      "num_tokens": 19014449.0,
      "step": 10460
    },
    {
      "entropy": 5.684511327743531,
      "epoch": 8.990975504941986,
      "grad_norm": 1.4296875,
      "learning_rate": 6.322759549600665e-05,
      "loss": 4.8448,
      "mean_token_accuracy": 0.24234582334756852,
      "num_tokens": 19024061.0,
      "step": 10465
    },
    {
      "entropy": 5.649232959747314,
      "epoch": 8.99527288354104,
      "grad_norm": 1.3671875,
      "learning_rate": 6.311550737943753e-05,
      "loss": 4.8331,
      "mean_token_accuracy": 0.24488990157842636,
      "num_tokens": 19033400.0,
      "step": 10470
    },
    {
      "entropy": 5.7437886714935305,
      "epoch": 8.999570262140095,
      "grad_norm": 1.34375,
      "learning_rate": 6.300388193391075e-05,
      "loss": 4.9086,
      "mean_token_accuracy": 0.2353186011314392,
      "num_tokens": 19042633.0,
      "step": 10475
    },
    {
      "entropy": 5.738433043162028,
      "epoch": 9.003437902879243,
      "grad_norm": 1.2734375,
      "learning_rate": 6.289271940317174e-05,
      "loss": 4.9605,
      "mean_token_accuracy": 0.22919231487645042,
      "num_tokens": 19051120.0,
      "step": 10480
    },
    {
      "entropy": 5.701286554336548,
      "epoch": 9.007735281478299,
      "grad_norm": 1.328125,
      "learning_rate": 6.278202002995497e-05,
      "loss": 4.8759,
      "mean_token_accuracy": 0.238884337246418,
      "num_tokens": 19060755.0,
      "step": 10485
    },
    {
      "entropy": 5.717434740066528,
      "epoch": 9.012032660077352,
      "grad_norm": 1.40625,
      "learning_rate": 6.267178405598371e-05,
      "loss": 4.7925,
      "mean_token_accuracy": 0.2569643914699554,
      "num_tokens": 19069197.0,
      "step": 10490
    },
    {
      "entropy": 5.755046939849853,
      "epoch": 9.016330038676408,
      "grad_norm": 1.3515625,
      "learning_rate": 6.256201172196921e-05,
      "loss": 4.8559,
      "mean_token_accuracy": 0.2463230550289154,
      "num_tokens": 19078452.0,
      "step": 10495
    },
    {
      "entropy": 5.735121393203736,
      "epoch": 9.020627417275461,
      "grad_norm": 1.28125,
      "learning_rate": 6.245270326761034e-05,
      "loss": 4.8732,
      "mean_token_accuracy": 0.23314425498247146,
      "num_tokens": 19087127.0,
      "step": 10500
    },
    {
      "epoch": 9.020627417275461,
      "eval_entropy": 5.559686202723701,
      "eval_loss": 5.891933917999268,
      "eval_mean_token_accuracy": 0.18310657089894955,
      "eval_num_tokens": 19087127.0,
      "eval_runtime": 2.0537,
      "eval_samples_per_second": 1728.062,
      "eval_steps_per_second": 216.19,
      "step": 10500
    },
    {
      "entropy": 5.768205690383911,
      "epoch": 9.024924795874517,
      "grad_norm": 1.4375,
      "learning_rate": 6.234385893159311e-05,
      "loss": 4.8745,
      "mean_token_accuracy": 0.24339525550603866,
      "num_tokens": 19095610.0,
      "step": 10505
    },
    {
      "entropy": 5.7007143020629885,
      "epoch": 9.02922217447357,
      "grad_norm": 1.390625,
      "learning_rate": 6.223547895159009e-05,
      "loss": 4.8149,
      "mean_token_accuracy": 0.25003656595945356,
      "num_tokens": 19103649.0,
      "step": 10510
    },
    {
      "entropy": 5.641485261917114,
      "epoch": 9.033519553072626,
      "grad_norm": 1.3046875,
      "learning_rate": 6.212756356425978e-05,
      "loss": 4.7692,
      "mean_token_accuracy": 0.25189306139945983,
      "num_tokens": 19112426.0,
      "step": 10515
    },
    {
      "entropy": 5.722223091125488,
      "epoch": 9.03781693167168,
      "grad_norm": 1.3359375,
      "learning_rate": 6.202011300524623e-05,
      "loss": 4.8962,
      "mean_token_accuracy": 0.2431654766201973,
      "num_tokens": 19121307.0,
      "step": 10520
    },
    {
      "entropy": 5.768562936782837,
      "epoch": 9.042114310270735,
      "grad_norm": 1.2578125,
      "learning_rate": 6.191312750917855e-05,
      "loss": 4.9119,
      "mean_token_accuracy": 0.24403591454029083,
      "num_tokens": 19129728.0,
      "step": 10525
    },
    {
      "entropy": 5.696633958816529,
      "epoch": 9.046411688869789,
      "grad_norm": 1.375,
      "learning_rate": 6.180660730967036e-05,
      "loss": 4.8368,
      "mean_token_accuracy": 0.24488697648048402,
      "num_tokens": 19139931.0,
      "step": 10530
    },
    {
      "entropy": 5.760554838180542,
      "epoch": 9.050709067468844,
      "grad_norm": 1.34375,
      "learning_rate": 6.170055263931912e-05,
      "loss": 4.8218,
      "mean_token_accuracy": 0.24506218880414962,
      "num_tokens": 19148711.0,
      "step": 10535
    },
    {
      "entropy": 5.7619397163391115,
      "epoch": 9.055006446067898,
      "grad_norm": 1.375,
      "learning_rate": 6.159496372970584e-05,
      "loss": 4.8071,
      "mean_token_accuracy": 0.24310505241155625,
      "num_tokens": 19156937.0,
      "step": 10540
    },
    {
      "entropy": 5.677298069000244,
      "epoch": 9.059303824666953,
      "grad_norm": 1.2265625,
      "learning_rate": 6.148984081139454e-05,
      "loss": 4.8373,
      "mean_token_accuracy": 0.2453565239906311,
      "num_tokens": 19166449.0,
      "step": 10545
    },
    {
      "entropy": 5.751143884658814,
      "epoch": 9.063601203266007,
      "grad_norm": 1.1328125,
      "learning_rate": 6.138518411393163e-05,
      "loss": 4.9664,
      "mean_token_accuracy": 0.23417426645755768,
      "num_tokens": 19175757.0,
      "step": 10550
    },
    {
      "entropy": 5.7382289409637455,
      "epoch": 9.067898581865062,
      "grad_norm": 1.390625,
      "learning_rate": 6.128099386584548e-05,
      "loss": 4.8604,
      "mean_token_accuracy": 0.2502607494592667,
      "num_tokens": 19184638.0,
      "step": 10555
    },
    {
      "entropy": 5.714648246765137,
      "epoch": 9.072195960464116,
      "grad_norm": 1.390625,
      "learning_rate": 6.1177270294646e-05,
      "loss": 4.9032,
      "mean_token_accuracy": 0.24113665670156478,
      "num_tokens": 19193791.0,
      "step": 10560
    },
    {
      "entropy": 5.68681697845459,
      "epoch": 9.076493339063171,
      "grad_norm": 1.34375,
      "learning_rate": 6.107401362682401e-05,
      "loss": 4.9186,
      "mean_token_accuracy": 0.23942002207040786,
      "num_tokens": 19202595.0,
      "step": 10565
    },
    {
      "entropy": 5.715084075927734,
      "epoch": 9.080790717662227,
      "grad_norm": 1.2265625,
      "learning_rate": 6.097122408785076e-05,
      "loss": 4.881,
      "mean_token_accuracy": 0.24075437635183333,
      "num_tokens": 19211994.0,
      "step": 10570
    },
    {
      "entropy": 5.72618145942688,
      "epoch": 9.08508809626128,
      "grad_norm": 1.3046875,
      "learning_rate": 6.086890190217752e-05,
      "loss": 4.8462,
      "mean_token_accuracy": 0.2431514397263527,
      "num_tokens": 19220588.0,
      "step": 10575
    },
    {
      "entropy": 5.6821825981140135,
      "epoch": 9.089385474860336,
      "grad_norm": 1.3984375,
      "learning_rate": 6.076704729323507e-05,
      "loss": 4.7938,
      "mean_token_accuracy": 0.24978432357311248,
      "num_tokens": 19229679.0,
      "step": 10580
    },
    {
      "entropy": 5.72755184173584,
      "epoch": 9.09368285345939,
      "grad_norm": 1.3359375,
      "learning_rate": 6.0665660483433173e-05,
      "loss": 4.9212,
      "mean_token_accuracy": 0.23904457241296767,
      "num_tokens": 19239269.0,
      "step": 10585
    },
    {
      "entropy": 5.739861583709716,
      "epoch": 9.097980232058445,
      "grad_norm": 1.375,
      "learning_rate": 6.05647416941601e-05,
      "loss": 4.9112,
      "mean_token_accuracy": 0.24730296581983566,
      "num_tokens": 19249046.0,
      "step": 10590
    },
    {
      "entropy": 5.682226276397705,
      "epoch": 9.102277610657499,
      "grad_norm": 1.34375,
      "learning_rate": 6.046429114578212e-05,
      "loss": 4.784,
      "mean_token_accuracy": 0.2475551262497902,
      "num_tokens": 19257398.0,
      "step": 10595
    },
    {
      "entropy": 5.744178533554077,
      "epoch": 9.106574989256554,
      "grad_norm": 1.34375,
      "learning_rate": 6.0364309057643084e-05,
      "loss": 4.8856,
      "mean_token_accuracy": 0.2361735597252846,
      "num_tokens": 19266619.0,
      "step": 10600
    },
    {
      "entropy": 5.779590082168579,
      "epoch": 9.110872367855608,
      "grad_norm": 1.3515625,
      "learning_rate": 6.0264795648063904e-05,
      "loss": 4.8832,
      "mean_token_accuracy": 0.24283566623926162,
      "num_tokens": 19275837.0,
      "step": 10605
    },
    {
      "entropy": 5.7637005805969235,
      "epoch": 9.115169746454663,
      "grad_norm": 1.203125,
      "learning_rate": 6.0165751134342155e-05,
      "loss": 4.8548,
      "mean_token_accuracy": 0.2424402043223381,
      "num_tokens": 19284745.0,
      "step": 10610
    },
    {
      "entropy": 5.722152042388916,
      "epoch": 9.119467125053717,
      "grad_norm": 1.203125,
      "learning_rate": 6.006717573275138e-05,
      "loss": 4.8022,
      "mean_token_accuracy": 0.24550892412662506,
      "num_tokens": 19293956.0,
      "step": 10615
    },
    {
      "entropy": 5.700108480453491,
      "epoch": 9.123764503652772,
      "grad_norm": 1.09375,
      "learning_rate": 5.996906965854093e-05,
      "loss": 4.9131,
      "mean_token_accuracy": 0.23563133031129838,
      "num_tokens": 19303629.0,
      "step": 10620
    },
    {
      "entropy": 5.715306329727173,
      "epoch": 9.128061882251826,
      "grad_norm": 1.421875,
      "learning_rate": 5.987143312593522e-05,
      "loss": 4.9243,
      "mean_token_accuracy": 0.23819719702005387,
      "num_tokens": 19312458.0,
      "step": 10625
    },
    {
      "entropy": 5.679522514343262,
      "epoch": 9.132359260850881,
      "grad_norm": 1.3671875,
      "learning_rate": 5.977426634813338e-05,
      "loss": 4.8295,
      "mean_token_accuracy": 0.2426922008395195,
      "num_tokens": 19321428.0,
      "step": 10630
    },
    {
      "entropy": 5.7862955093383786,
      "epoch": 9.136656639449935,
      "grad_norm": 1.3828125,
      "learning_rate": 5.9677569537308866e-05,
      "loss": 4.9686,
      "mean_token_accuracy": 0.22704976350069045,
      "num_tokens": 19331176.0,
      "step": 10635
    },
    {
      "entropy": 5.7446370124816895,
      "epoch": 9.14095401804899,
      "grad_norm": 1.2890625,
      "learning_rate": 5.958134290460888e-05,
      "loss": 4.9083,
      "mean_token_accuracy": 0.2379206582903862,
      "num_tokens": 19340401.0,
      "step": 10640
    },
    {
      "entropy": 5.678402233123779,
      "epoch": 9.145251396648044,
      "grad_norm": 1.140625,
      "learning_rate": 5.94855866601539e-05,
      "loss": 4.8058,
      "mean_token_accuracy": 0.24439001679420472,
      "num_tokens": 19350700.0,
      "step": 10645
    },
    {
      "entropy": 5.744732236862182,
      "epoch": 9.1495487752471,
      "grad_norm": 1.3828125,
      "learning_rate": 5.939030101303724e-05,
      "loss": 4.8695,
      "mean_token_accuracy": 0.2402612566947937,
      "num_tokens": 19360040.0,
      "step": 10650
    },
    {
      "entropy": 5.7309928894042965,
      "epoch": 9.153846153846153,
      "grad_norm": 1.4140625,
      "learning_rate": 5.929548617132472e-05,
      "loss": 4.8179,
      "mean_token_accuracy": 0.2454057216644287,
      "num_tokens": 19368546.0,
      "step": 10655
    },
    {
      "entropy": 5.6870287418365475,
      "epoch": 9.158143532445209,
      "grad_norm": 1.4296875,
      "learning_rate": 5.920114234205407e-05,
      "loss": 4.7862,
      "mean_token_accuracy": 0.24816205948591233,
      "num_tokens": 19377137.0,
      "step": 10660
    },
    {
      "entropy": 5.759473419189453,
      "epoch": 9.162440911044262,
      "grad_norm": 1.3515625,
      "learning_rate": 5.910726973123451e-05,
      "loss": 4.9586,
      "mean_token_accuracy": 0.2330634355545044,
      "num_tokens": 19385781.0,
      "step": 10665
    },
    {
      "entropy": 5.72282657623291,
      "epoch": 9.166738289643318,
      "grad_norm": 1.3515625,
      "learning_rate": 5.901386854384622e-05,
      "loss": 4.8898,
      "mean_token_accuracy": 0.24274778068065644,
      "num_tokens": 19395186.0,
      "step": 10670
    },
    {
      "entropy": 5.77225079536438,
      "epoch": 9.171035668242371,
      "grad_norm": 1.3828125,
      "learning_rate": 5.892093898384017e-05,
      "loss": 4.8566,
      "mean_token_accuracy": 0.24656753838062287,
      "num_tokens": 19404140.0,
      "step": 10675
    },
    {
      "entropy": 5.730916357040405,
      "epoch": 9.175333046841427,
      "grad_norm": 1.34375,
      "learning_rate": 5.8828481254137276e-05,
      "loss": 4.8786,
      "mean_token_accuracy": 0.24277020394802093,
      "num_tokens": 19414481.0,
      "step": 10680
    },
    {
      "entropy": 5.60438551902771,
      "epoch": 9.17963042544048,
      "grad_norm": 1.5234375,
      "learning_rate": 5.873649555662836e-05,
      "loss": 4.7244,
      "mean_token_accuracy": 0.24888451993465424,
      "num_tokens": 19423259.0,
      "step": 10685
    },
    {
      "entropy": 5.724711990356445,
      "epoch": 9.183927804039536,
      "grad_norm": 1.3359375,
      "learning_rate": 5.8644982092173335e-05,
      "loss": 4.865,
      "mean_token_accuracy": 0.24325377345085145,
      "num_tokens": 19432011.0,
      "step": 10690
    },
    {
      "entropy": 5.696267986297608,
      "epoch": 9.18822518263859,
      "grad_norm": 1.328125,
      "learning_rate": 5.85539410606011e-05,
      "loss": 4.8452,
      "mean_token_accuracy": 0.24142812043428422,
      "num_tokens": 19441221.0,
      "step": 10695
    },
    {
      "entropy": 5.712868976593017,
      "epoch": 9.192522561237645,
      "grad_norm": 1.3984375,
      "learning_rate": 5.8463372660708836e-05,
      "loss": 4.8102,
      "mean_token_accuracy": 0.2449094161391258,
      "num_tokens": 19450014.0,
      "step": 10700
    },
    {
      "entropy": 5.724195671081543,
      "epoch": 9.196819939836699,
      "grad_norm": 1.3359375,
      "learning_rate": 5.837327709026171e-05,
      "loss": 4.9036,
      "mean_token_accuracy": 0.2391239047050476,
      "num_tokens": 19459480.0,
      "step": 10705
    },
    {
      "entropy": 5.737326002120971,
      "epoch": 9.201117318435754,
      "grad_norm": 1.265625,
      "learning_rate": 5.8283654545992464e-05,
      "loss": 4.9029,
      "mean_token_accuracy": 0.23848976492881774,
      "num_tokens": 19469111.0,
      "step": 10710
    },
    {
      "entropy": 5.751066207885742,
      "epoch": 9.20541469703481,
      "grad_norm": 1.40625,
      "learning_rate": 5.819450522360096e-05,
      "loss": 4.8419,
      "mean_token_accuracy": 0.23905779421329498,
      "num_tokens": 19477531.0,
      "step": 10715
    },
    {
      "entropy": 5.737590551376343,
      "epoch": 9.209712075633863,
      "grad_norm": 1.4140625,
      "learning_rate": 5.810582931775362e-05,
      "loss": 4.8554,
      "mean_token_accuracy": 0.23738077729940416,
      "num_tokens": 19486041.0,
      "step": 10720
    },
    {
      "entropy": 5.7077555656433105,
      "epoch": 9.214009454232919,
      "grad_norm": 1.3515625,
      "learning_rate": 5.801762702208317e-05,
      "loss": 4.8924,
      "mean_token_accuracy": 0.24125652760267258,
      "num_tokens": 19495514.0,
      "step": 10725
    },
    {
      "entropy": 5.6811192512512205,
      "epoch": 9.218306832831972,
      "grad_norm": 1.390625,
      "learning_rate": 5.7929898529188215e-05,
      "loss": 4.8245,
      "mean_token_accuracy": 0.2412465736269951,
      "num_tokens": 19503879.0,
      "step": 10730
    },
    {
      "entropy": 5.684185552597046,
      "epoch": 9.222604211431028,
      "grad_norm": 1.1640625,
      "learning_rate": 5.784264403063272e-05,
      "loss": 4.7873,
      "mean_token_accuracy": 0.24483999311923982,
      "num_tokens": 19513922.0,
      "step": 10735
    },
    {
      "entropy": 5.6932319641113285,
      "epoch": 9.226901590030081,
      "grad_norm": 1.4765625,
      "learning_rate": 5.775586371694561e-05,
      "loss": 4.8255,
      "mean_token_accuracy": 0.2445918560028076,
      "num_tokens": 19522555.0,
      "step": 10740
    },
    {
      "entropy": 5.701971054077148,
      "epoch": 9.231198968629137,
      "grad_norm": 1.1953125,
      "learning_rate": 5.7669557777620376e-05,
      "loss": 4.8253,
      "mean_token_accuracy": 0.2440289631485939,
      "num_tokens": 19531245.0,
      "step": 10745
    },
    {
      "entropy": 5.6902079582214355,
      "epoch": 9.23549634722819,
      "grad_norm": 1.40625,
      "learning_rate": 5.7583726401114756e-05,
      "loss": 4.8014,
      "mean_token_accuracy": 0.25117970257997513,
      "num_tokens": 19540258.0,
      "step": 10750
    },
    {
      "entropy": 5.7464635372161865,
      "epoch": 9.239793725827246,
      "grad_norm": 1.390625,
      "learning_rate": 5.749836977485013e-05,
      "loss": 4.8656,
      "mean_token_accuracy": 0.2349914312362671,
      "num_tokens": 19550071.0,
      "step": 10755
    },
    {
      "entropy": 5.686515665054321,
      "epoch": 9.2440911044263,
      "grad_norm": 1.25,
      "learning_rate": 5.741348808521128e-05,
      "loss": 4.7854,
      "mean_token_accuracy": 0.25020948648452757,
      "num_tokens": 19558972.0,
      "step": 10760
    },
    {
      "entropy": 5.709124660491943,
      "epoch": 9.248388483025355,
      "grad_norm": 1.28125,
      "learning_rate": 5.7329081517545846e-05,
      "loss": 4.8781,
      "mean_token_accuracy": 0.23870760351419448,
      "num_tokens": 19568270.0,
      "step": 10765
    },
    {
      "entropy": 5.750066089630127,
      "epoch": 9.252685861624409,
      "grad_norm": 1.5703125,
      "learning_rate": 5.724515025616409e-05,
      "loss": 4.8639,
      "mean_token_accuracy": 0.2456628665328026,
      "num_tokens": 19576726.0,
      "step": 10770
    },
    {
      "entropy": 5.721097707748413,
      "epoch": 9.256983240223464,
      "grad_norm": 1.296875,
      "learning_rate": 5.716169448433832e-05,
      "loss": 4.8602,
      "mean_token_accuracy": 0.23966725170612335,
      "num_tokens": 19586706.0,
      "step": 10775
    },
    {
      "entropy": 5.713471555709839,
      "epoch": 9.261280618822518,
      "grad_norm": 1.3671875,
      "learning_rate": 5.707871438430255e-05,
      "loss": 4.8842,
      "mean_token_accuracy": 0.23704735487699508,
      "num_tokens": 19595370.0,
      "step": 10780
    },
    {
      "entropy": 5.720728015899658,
      "epoch": 9.265577997421573,
      "grad_norm": 1.4453125,
      "learning_rate": 5.699621013725218e-05,
      "loss": 4.9015,
      "mean_token_accuracy": 0.24206158965826036,
      "num_tokens": 19604134.0,
      "step": 10785
    },
    {
      "entropy": 5.6510015487670895,
      "epoch": 9.269875376020627,
      "grad_norm": 1.359375,
      "learning_rate": 5.691418192334352e-05,
      "loss": 4.8221,
      "mean_token_accuracy": 0.25330779552459715,
      "num_tokens": 19612716.0,
      "step": 10790
    },
    {
      "entropy": 5.697645568847657,
      "epoch": 9.274172754619682,
      "grad_norm": 1.296875,
      "learning_rate": 5.683262992169341e-05,
      "loss": 4.909,
      "mean_token_accuracy": 0.2355555221438408,
      "num_tokens": 19621850.0,
      "step": 10795
    },
    {
      "entropy": 5.688250255584717,
      "epoch": 9.278470133218736,
      "grad_norm": 1.296875,
      "learning_rate": 5.675155431037876e-05,
      "loss": 4.8704,
      "mean_token_accuracy": 0.24086801260709761,
      "num_tokens": 19630706.0,
      "step": 10800
    },
    {
      "entropy": 5.662147855758667,
      "epoch": 9.282767511817791,
      "grad_norm": 1.359375,
      "learning_rate": 5.6670955266436365e-05,
      "loss": 4.8389,
      "mean_token_accuracy": 0.2558609426021576,
      "num_tokens": 19639364.0,
      "step": 10805
    },
    {
      "entropy": 5.741171836853027,
      "epoch": 9.287064890416845,
      "grad_norm": 1.328125,
      "learning_rate": 5.659083296586229e-05,
      "loss": 4.7906,
      "mean_token_accuracy": 0.24558178037405015,
      "num_tokens": 19647816.0,
      "step": 10810
    },
    {
      "entropy": 5.730060243606568,
      "epoch": 9.2913622690159,
      "grad_norm": 1.2109375,
      "learning_rate": 5.6511187583611663e-05,
      "loss": 4.8929,
      "mean_token_accuracy": 0.24423112869262695,
      "num_tokens": 19657136.0,
      "step": 10815
    },
    {
      "entropy": 5.771777868270874,
      "epoch": 9.295659647614954,
      "grad_norm": 1.4609375,
      "learning_rate": 5.643201929359809e-05,
      "loss": 4.8806,
      "mean_token_accuracy": 0.2378945991396904,
      "num_tokens": 19665676.0,
      "step": 10820
    },
    {
      "entropy": 5.734367990493775,
      "epoch": 9.29995702621401,
      "grad_norm": 1.4765625,
      "learning_rate": 5.635332826869353e-05,
      "loss": 4.8337,
      "mean_token_accuracy": 0.24641830772161483,
      "num_tokens": 19674415.0,
      "step": 10825
    },
    {
      "entropy": 5.673384141921997,
      "epoch": 9.304254404813063,
      "grad_norm": 1.265625,
      "learning_rate": 5.6275114680727716e-05,
      "loss": 4.8245,
      "mean_token_accuracy": 0.24797581434249877,
      "num_tokens": 19684411.0,
      "step": 10830
    },
    {
      "entropy": 5.698321294784546,
      "epoch": 9.308551783412119,
      "grad_norm": 1.4453125,
      "learning_rate": 5.619737870048783e-05,
      "loss": 4.8224,
      "mean_token_accuracy": 0.24377230405807496,
      "num_tokens": 19693192.0,
      "step": 10835
    },
    {
      "entropy": 5.6213613986969,
      "epoch": 9.312849162011172,
      "grad_norm": 1.34375,
      "learning_rate": 5.612012049771823e-05,
      "loss": 4.7103,
      "mean_token_accuracy": 0.25083507001399996,
      "num_tokens": 19701710.0,
      "step": 10840
    },
    {
      "entropy": 5.7836895942687985,
      "epoch": 9.317146540610228,
      "grad_norm": 1.265625,
      "learning_rate": 5.6043340241119924e-05,
      "loss": 4.9229,
      "mean_token_accuracy": 0.23689377903938294,
      "num_tokens": 19710909.0,
      "step": 10845
    },
    {
      "entropy": 5.72804274559021,
      "epoch": 9.321443919209282,
      "grad_norm": 1.296875,
      "learning_rate": 5.596703809835033e-05,
      "loss": 4.874,
      "mean_token_accuracy": 0.23930975049734116,
      "num_tokens": 19720634.0,
      "step": 10850
    },
    {
      "entropy": 5.708077764511108,
      "epoch": 9.325741297808337,
      "grad_norm": 1.5390625,
      "learning_rate": 5.589121423602277e-05,
      "loss": 4.8003,
      "mean_token_accuracy": 0.24899453520774842,
      "num_tokens": 19729763.0,
      "step": 10855
    },
    {
      "entropy": 5.772850513458252,
      "epoch": 9.33003867640739,
      "grad_norm": 1.328125,
      "learning_rate": 5.581586881970631e-05,
      "loss": 4.9386,
      "mean_token_accuracy": 0.23242200613021852,
      "num_tokens": 19739065.0,
      "step": 10860
    },
    {
      "entropy": 5.6640668392181395,
      "epoch": 9.334336055006446,
      "grad_norm": 1.2734375,
      "learning_rate": 5.574100201392522e-05,
      "loss": 4.7625,
      "mean_token_accuracy": 0.25813745260238646,
      "num_tokens": 19748519.0,
      "step": 10865
    },
    {
      "entropy": 5.7350257396697994,
      "epoch": 9.338633433605501,
      "grad_norm": 1.3515625,
      "learning_rate": 5.5666613982158665e-05,
      "loss": 4.839,
      "mean_token_accuracy": 0.2506720334291458,
      "num_tokens": 19757573.0,
      "step": 10870
    },
    {
      "entropy": 5.762840938568115,
      "epoch": 9.342930812204555,
      "grad_norm": 1.34375,
      "learning_rate": 5.559270488684036e-05,
      "loss": 4.9695,
      "mean_token_accuracy": 0.22792510986328124,
      "num_tokens": 19767534.0,
      "step": 10875
    },
    {
      "entropy": 5.7635585308074955,
      "epoch": 9.34722819080361,
      "grad_norm": 1.2265625,
      "learning_rate": 5.551927488935826e-05,
      "loss": 4.915,
      "mean_token_accuracy": 0.2328255593776703,
      "num_tokens": 19776929.0,
      "step": 10880
    },
    {
      "entropy": 5.7432098388671875,
      "epoch": 9.351525569402664,
      "grad_norm": 1.3515625,
      "learning_rate": 5.5446324150054086e-05,
      "loss": 4.8966,
      "mean_token_accuracy": 0.24054146707057952,
      "num_tokens": 19785772.0,
      "step": 10885
    },
    {
      "entropy": 5.744037818908692,
      "epoch": 9.35582294800172,
      "grad_norm": 1.15625,
      "learning_rate": 5.537385282822315e-05,
      "loss": 4.8791,
      "mean_token_accuracy": 0.2423680067062378,
      "num_tokens": 19796073.0,
      "step": 10890
    },
    {
      "entropy": 5.719657611846924,
      "epoch": 9.360120326600773,
      "grad_norm": 1.5234375,
      "learning_rate": 5.53018610821138e-05,
      "loss": 4.8875,
      "mean_token_accuracy": 0.24139334410429,
      "num_tokens": 19804511.0,
      "step": 10895
    },
    {
      "entropy": 5.771108245849609,
      "epoch": 9.364417705199829,
      "grad_norm": 1.3671875,
      "learning_rate": 5.523034906892728e-05,
      "loss": 4.9206,
      "mean_token_accuracy": 0.24052533209323884,
      "num_tokens": 19813348.0,
      "step": 10900
    },
    {
      "entropy": 5.759429550170898,
      "epoch": 9.368715083798882,
      "grad_norm": 1.3125,
      "learning_rate": 5.515931694481722e-05,
      "loss": 4.9221,
      "mean_token_accuracy": 0.23646650612354278,
      "num_tokens": 19822390.0,
      "step": 10905
    },
    {
      "entropy": 5.789000034332275,
      "epoch": 9.373012462397938,
      "grad_norm": 1.34375,
      "learning_rate": 5.508876486488936e-05,
      "loss": 4.9348,
      "mean_token_accuracy": 0.23186779618263245,
      "num_tokens": 19832423.0,
      "step": 10910
    },
    {
      "entropy": 5.708720397949219,
      "epoch": 9.377309840996991,
      "grad_norm": 1.421875,
      "learning_rate": 5.501869298320128e-05,
      "loss": 4.8786,
      "mean_token_accuracy": 0.23571010082960128,
      "num_tokens": 19841271.0,
      "step": 10915
    },
    {
      "entropy": 5.74044771194458,
      "epoch": 9.381607219596047,
      "grad_norm": 1.4296875,
      "learning_rate": 5.4949101452761995e-05,
      "loss": 4.8809,
      "mean_token_accuracy": 0.24290914088487625,
      "num_tokens": 19850648.0,
      "step": 10920
    },
    {
      "entropy": 5.731217288970948,
      "epoch": 9.3859045981951,
      "grad_norm": 1.2109375,
      "learning_rate": 5.4879990425531534e-05,
      "loss": 4.8185,
      "mean_token_accuracy": 0.25113645792007444,
      "num_tokens": 19860102.0,
      "step": 10925
    },
    {
      "entropy": 5.745176935195923,
      "epoch": 9.390201976794156,
      "grad_norm": 1.1875,
      "learning_rate": 5.4811360052420754e-05,
      "loss": 4.9093,
      "mean_token_accuracy": 0.23800594508647918,
      "num_tokens": 19870008.0,
      "step": 10930
    },
    {
      "entropy": 5.709982967376709,
      "epoch": 9.39449935539321,
      "grad_norm": 1.15625,
      "learning_rate": 5.4743210483290974e-05,
      "loss": 4.8154,
      "mean_token_accuracy": 0.24858268350362778,
      "num_tokens": 19879452.0,
      "step": 10935
    },
    {
      "entropy": 5.79788875579834,
      "epoch": 9.398796733992265,
      "grad_norm": 1.25,
      "learning_rate": 5.467554186695364e-05,
      "loss": 4.9251,
      "mean_token_accuracy": 0.23087782859802247,
      "num_tokens": 19888956.0,
      "step": 10940
    },
    {
      "entropy": 5.713366889953614,
      "epoch": 9.403094112591319,
      "grad_norm": 1.2734375,
      "learning_rate": 5.4608354351169944e-05,
      "loss": 4.8697,
      "mean_token_accuracy": 0.24130331128835678,
      "num_tokens": 19897760.0,
      "step": 10945
    },
    {
      "entropy": 5.6680543422698975,
      "epoch": 9.407391491190374,
      "grad_norm": 1.3984375,
      "learning_rate": 5.454164808265057e-05,
      "loss": 4.8502,
      "mean_token_accuracy": 0.23947279751300812,
      "num_tokens": 19905692.0,
      "step": 10950
    },
    {
      "entropy": 5.671962881088257,
      "epoch": 9.411688869789428,
      "grad_norm": 1.359375,
      "learning_rate": 5.447542320705532e-05,
      "loss": 4.7892,
      "mean_token_accuracy": 0.2486485093832016,
      "num_tokens": 19914633.0,
      "step": 10955
    },
    {
      "entropy": 5.754352760314942,
      "epoch": 9.415986248388483,
      "grad_norm": 1.3984375,
      "learning_rate": 5.440967986899289e-05,
      "loss": 4.885,
      "mean_token_accuracy": 0.2311162158846855,
      "num_tokens": 19922749.0,
      "step": 10960
    },
    {
      "entropy": 5.722763872146606,
      "epoch": 9.420283626987537,
      "grad_norm": 1.34375,
      "learning_rate": 5.434441821202042e-05,
      "loss": 4.8282,
      "mean_token_accuracy": 0.24647813886404038,
      "num_tokens": 19931274.0,
      "step": 10965
    },
    {
      "entropy": 5.728282833099366,
      "epoch": 9.424581005586592,
      "grad_norm": 1.359375,
      "learning_rate": 5.42796383786433e-05,
      "loss": 4.8605,
      "mean_token_accuracy": 0.2373078092932701,
      "num_tokens": 19940052.0,
      "step": 10970
    },
    {
      "entropy": 5.72069993019104,
      "epoch": 9.428878384185646,
      "grad_norm": 1.234375,
      "learning_rate": 5.4215340510314805e-05,
      "loss": 4.8635,
      "mean_token_accuracy": 0.24039194881916046,
      "num_tokens": 19948890.0,
      "step": 10975
    },
    {
      "entropy": 5.700727844238282,
      "epoch": 9.433175762784701,
      "grad_norm": 1.234375,
      "learning_rate": 5.41515247474358e-05,
      "loss": 4.9025,
      "mean_token_accuracy": 0.24478237628936766,
      "num_tokens": 19958158.0,
      "step": 10980
    },
    {
      "entropy": 5.762474441528321,
      "epoch": 9.437473141383755,
      "grad_norm": 1.2578125,
      "learning_rate": 5.4088191229354306e-05,
      "loss": 4.9555,
      "mean_token_accuracy": 0.2400050938129425,
      "num_tokens": 19967432.0,
      "step": 10985
    },
    {
      "entropy": 5.700712060928344,
      "epoch": 9.44177051998281,
      "grad_norm": 1.296875,
      "learning_rate": 5.402534009436552e-05,
      "loss": 4.8405,
      "mean_token_accuracy": 0.24093716442584992,
      "num_tokens": 19977661.0,
      "step": 10990
    },
    {
      "entropy": 5.7578575134277346,
      "epoch": 9.446067898581864,
      "grad_norm": 1.359375,
      "learning_rate": 5.396297147971116e-05,
      "loss": 4.8967,
      "mean_token_accuracy": 0.23771874606609344,
      "num_tokens": 19986359.0,
      "step": 10995
    },
    {
      "entropy": 5.716905927658081,
      "epoch": 9.45036527718092,
      "grad_norm": 1.28125,
      "learning_rate": 5.390108552157935e-05,
      "loss": 4.8097,
      "mean_token_accuracy": 0.24744485914707184,
      "num_tokens": 19994618.0,
      "step": 11000
    },
    {
      "epoch": 9.45036527718092,
      "eval_entropy": 5.561021716208072,
      "eval_loss": 5.893560409545898,
      "eval_mean_token_accuracy": 0.18303356171996743,
      "eval_num_tokens": 19994618.0,
      "eval_runtime": 2.0478,
      "eval_samples_per_second": 1733.059,
      "eval_steps_per_second": 216.816,
      "step": 11000
    },
    {
      "entropy": 5.751536750793457,
      "epoch": 9.454662655779973,
      "grad_norm": 1.3046875,
      "learning_rate": 5.383968235510427e-05,
      "loss": 4.9125,
      "mean_token_accuracy": 0.24575117975473404,
      "num_tokens": 20004050.0,
      "step": 11005
    },
    {
      "entropy": 5.719616365432739,
      "epoch": 9.458960034379029,
      "grad_norm": 1.2890625,
      "learning_rate": 5.377876211436592e-05,
      "loss": 4.8358,
      "mean_token_accuracy": 0.24225348532199859,
      "num_tokens": 20013480.0,
      "step": 11010
    },
    {
      "entropy": 5.794860315322876,
      "epoch": 9.463257412978084,
      "grad_norm": 1.4453125,
      "learning_rate": 5.371832493238973e-05,
      "loss": 4.9287,
      "mean_token_accuracy": 0.2346111848950386,
      "num_tokens": 20022133.0,
      "step": 11015
    },
    {
      "entropy": 5.674678897857666,
      "epoch": 9.467554791577138,
      "grad_norm": 1.375,
      "learning_rate": 5.365837094114639e-05,
      "loss": 4.808,
      "mean_token_accuracy": 0.25014188438653945,
      "num_tokens": 20031462.0,
      "step": 11020
    },
    {
      "entropy": 5.768085956573486,
      "epoch": 9.471852170176193,
      "grad_norm": 1.390625,
      "learning_rate": 5.3598900271551396e-05,
      "loss": 4.9489,
      "mean_token_accuracy": 0.2319101795554161,
      "num_tokens": 20042055.0,
      "step": 11025
    },
    {
      "entropy": 5.718179845809937,
      "epoch": 9.476149548775247,
      "grad_norm": 1.15625,
      "learning_rate": 5.353991305346499e-05,
      "loss": 4.8555,
      "mean_token_accuracy": 0.2395426645874977,
      "num_tokens": 20051379.0,
      "step": 11030
    },
    {
      "entropy": 5.6791009426116945,
      "epoch": 9.480446927374302,
      "grad_norm": 1.3359375,
      "learning_rate": 5.348140941569165e-05,
      "loss": 4.7847,
      "mean_token_accuracy": 0.2500495076179504,
      "num_tokens": 20060289.0,
      "step": 11035
    },
    {
      "entropy": 5.699596500396728,
      "epoch": 9.484744305973356,
      "grad_norm": 1.3203125,
      "learning_rate": 5.342338948597989e-05,
      "loss": 4.8269,
      "mean_token_accuracy": 0.24260480403900148,
      "num_tokens": 20070133.0,
      "step": 11040
    },
    {
      "entropy": 5.75848708152771,
      "epoch": 9.489041684572411,
      "grad_norm": 1.375,
      "learning_rate": 5.336585339102209e-05,
      "loss": 4.8636,
      "mean_token_accuracy": 0.24093058556318284,
      "num_tokens": 20078384.0,
      "step": 11045
    },
    {
      "entropy": 5.649259424209594,
      "epoch": 9.493339063171465,
      "grad_norm": 1.34375,
      "learning_rate": 5.33088012564541e-05,
      "loss": 4.7802,
      "mean_token_accuracy": 0.2498975083231926,
      "num_tokens": 20087116.0,
      "step": 11050
    },
    {
      "entropy": 5.608646535873413,
      "epoch": 9.49763644177052,
      "grad_norm": 1.34375,
      "learning_rate": 5.3252233206854955e-05,
      "loss": 4.7779,
      "mean_token_accuracy": 0.24721285551786423,
      "num_tokens": 20096918.0,
      "step": 11055
    },
    {
      "entropy": 5.720380878448486,
      "epoch": 9.501933820369574,
      "grad_norm": 1.3203125,
      "learning_rate": 5.3196149365746656e-05,
      "loss": 4.9427,
      "mean_token_accuracy": 0.23233859241008759,
      "num_tokens": 20107602.0,
      "step": 11060
    },
    {
      "entropy": 5.647018575668335,
      "epoch": 9.50623119896863,
      "grad_norm": 1.171875,
      "learning_rate": 5.31405498555939e-05,
      "loss": 4.8205,
      "mean_token_accuracy": 0.24482105374336244,
      "num_tokens": 20117732.0,
      "step": 11065
    },
    {
      "entropy": 5.787069416046142,
      "epoch": 9.510528577567683,
      "grad_norm": 1.21875,
      "learning_rate": 5.308543479780384e-05,
      "loss": 4.9499,
      "mean_token_accuracy": 0.22889099568128585,
      "num_tokens": 20127765.0,
      "step": 11070
    },
    {
      "entropy": 5.748648309707642,
      "epoch": 9.514825956166739,
      "grad_norm": 1.375,
      "learning_rate": 5.303080431272567e-05,
      "loss": 4.9253,
      "mean_token_accuracy": 0.23980981707572938,
      "num_tokens": 20137135.0,
      "step": 11075
    },
    {
      "entropy": 5.704532814025879,
      "epoch": 9.519123334765792,
      "grad_norm": 1.3984375,
      "learning_rate": 5.297665851965055e-05,
      "loss": 4.8654,
      "mean_token_accuracy": 0.24282841831445695,
      "num_tokens": 20146156.0,
      "step": 11080
    },
    {
      "entropy": 5.735879802703858,
      "epoch": 9.523420713364848,
      "grad_norm": 1.4921875,
      "learning_rate": 5.292299753681129e-05,
      "loss": 4.8687,
      "mean_token_accuracy": 0.2410810708999634,
      "num_tokens": 20154433.0,
      "step": 11085
    },
    {
      "entropy": 5.735851383209228,
      "epoch": 9.527718091963902,
      "grad_norm": 1.28125,
      "learning_rate": 5.286982148138196e-05,
      "loss": 4.852,
      "mean_token_accuracy": 0.24170392155647277,
      "num_tokens": 20164209.0,
      "step": 11090
    },
    {
      "entropy": 5.724815750122071,
      "epoch": 9.532015470562957,
      "grad_norm": 1.3515625,
      "learning_rate": 5.281713046947787e-05,
      "loss": 4.8645,
      "mean_token_accuracy": 0.2355036363005638,
      "num_tokens": 20173116.0,
      "step": 11095
    },
    {
      "entropy": 5.773639297485351,
      "epoch": 9.53631284916201,
      "grad_norm": 1.40625,
      "learning_rate": 5.2764924616155116e-05,
      "loss": 4.9453,
      "mean_token_accuracy": 0.23105342984199523,
      "num_tokens": 20182933.0,
      "step": 11100
    },
    {
      "entropy": 5.726321268081665,
      "epoch": 9.540610227761066,
      "grad_norm": 1.5078125,
      "learning_rate": 5.271320403541038e-05,
      "loss": 4.8288,
      "mean_token_accuracy": 0.24428213983774186,
      "num_tokens": 20191899.0,
      "step": 11105
    },
    {
      "entropy": 5.751812839508057,
      "epoch": 9.54490760636012,
      "grad_norm": 1.3671875,
      "learning_rate": 5.266196884018081e-05,
      "loss": 4.8788,
      "mean_token_accuracy": 0.24230959862470627,
      "num_tokens": 20201108.0,
      "step": 11110
    },
    {
      "entropy": 5.69306116104126,
      "epoch": 9.549204984959175,
      "grad_norm": 1.3671875,
      "learning_rate": 5.2611219142343494e-05,
      "loss": 4.854,
      "mean_token_accuracy": 0.24209704995155334,
      "num_tokens": 20209137.0,
      "step": 11115
    },
    {
      "entropy": 5.660806035995483,
      "epoch": 9.553502363558229,
      "grad_norm": 1.25,
      "learning_rate": 5.2560955052715574e-05,
      "loss": 4.7671,
      "mean_token_accuracy": 0.24614596515893936,
      "num_tokens": 20217879.0,
      "step": 11120
    },
    {
      "entropy": 5.724523305892944,
      "epoch": 9.557799742157284,
      "grad_norm": 1.3515625,
      "learning_rate": 5.2511176681053704e-05,
      "loss": 4.8475,
      "mean_token_accuracy": 0.24516242742538452,
      "num_tokens": 20226783.0,
      "step": 11125
    },
    {
      "entropy": 5.725910329818726,
      "epoch": 9.562097120756338,
      "grad_norm": 1.375,
      "learning_rate": 5.246188413605393e-05,
      "loss": 4.8643,
      "mean_token_accuracy": 0.24164182394742967,
      "num_tokens": 20235008.0,
      "step": 11130
    },
    {
      "entropy": 5.716705322265625,
      "epoch": 9.566394499355393,
      "grad_norm": 1.359375,
      "learning_rate": 5.241307752535149e-05,
      "loss": 4.7719,
      "mean_token_accuracy": 0.25252772122621536,
      "num_tokens": 20243157.0,
      "step": 11135
    },
    {
      "entropy": 5.709731864929199,
      "epoch": 9.570691877954447,
      "grad_norm": 1.2890625,
      "learning_rate": 5.236475695552052e-05,
      "loss": 4.8463,
      "mean_token_accuracy": 0.24593894481658934,
      "num_tokens": 20251548.0,
      "step": 11140
    },
    {
      "entropy": 5.611426973342896,
      "epoch": 9.574989256553502,
      "grad_norm": 1.4375,
      "learning_rate": 5.2316922532073796e-05,
      "loss": 4.7391,
      "mean_token_accuracy": 0.25582781732082366,
      "num_tokens": 20260954.0,
      "step": 11145
    },
    {
      "entropy": 5.783295440673828,
      "epoch": 9.579286635152556,
      "grad_norm": 1.3359375,
      "learning_rate": 5.226957435946265e-05,
      "loss": 4.9657,
      "mean_token_accuracy": 0.2272557646036148,
      "num_tokens": 20269976.0,
      "step": 11150
    },
    {
      "entropy": 5.731646537780762,
      "epoch": 9.583584013751612,
      "grad_norm": 1.3203125,
      "learning_rate": 5.2222712541076464e-05,
      "loss": 4.9138,
      "mean_token_accuracy": 0.2371578335762024,
      "num_tokens": 20279824.0,
      "step": 11155
    },
    {
      "entropy": 5.7411322593688965,
      "epoch": 9.587881392350667,
      "grad_norm": 1.34375,
      "learning_rate": 5.217633717924282e-05,
      "loss": 4.8572,
      "mean_token_accuracy": 0.2577525168657303,
      "num_tokens": 20289284.0,
      "step": 11160
    },
    {
      "entropy": 5.7514575004577635,
      "epoch": 9.59217877094972,
      "grad_norm": 1.1796875,
      "learning_rate": 5.213044837522689e-05,
      "loss": 4.8106,
      "mean_token_accuracy": 0.24952750355005265,
      "num_tokens": 20297905.0,
      "step": 11165
    },
    {
      "entropy": 5.672228050231934,
      "epoch": 9.596476149548776,
      "grad_norm": 1.3828125,
      "learning_rate": 5.208504622923154e-05,
      "loss": 4.8154,
      "mean_token_accuracy": 0.24929569363594056,
      "num_tokens": 20307219.0,
      "step": 11170
    },
    {
      "entropy": 5.722142553329467,
      "epoch": 9.60077352814783,
      "grad_norm": 1.3984375,
      "learning_rate": 5.204013084039687e-05,
      "loss": 4.8724,
      "mean_token_accuracy": 0.2437526673078537,
      "num_tokens": 20316369.0,
      "step": 11175
    },
    {
      "entropy": 5.6969505786895756,
      "epoch": 9.605070906746885,
      "grad_norm": 1.4375,
      "learning_rate": 5.199570230680017e-05,
      "loss": 4.8473,
      "mean_token_accuracy": 0.24211715310811996,
      "num_tokens": 20325764.0,
      "step": 11180
    },
    {
      "entropy": 5.803487777709961,
      "epoch": 9.609368285345939,
      "grad_norm": 1.328125,
      "learning_rate": 5.19517607254556e-05,
      "loss": 4.9431,
      "mean_token_accuracy": 0.23417452871799468,
      "num_tokens": 20334671.0,
      "step": 11185
    },
    {
      "entropy": 5.7173380851745605,
      "epoch": 9.613665663944994,
      "grad_norm": 1.34375,
      "learning_rate": 5.190830619231397e-05,
      "loss": 4.8684,
      "mean_token_accuracy": 0.24233201891183853,
      "num_tokens": 20343585.0,
      "step": 11190
    },
    {
      "entropy": 5.701148939132691,
      "epoch": 9.617963042544048,
      "grad_norm": 1.328125,
      "learning_rate": 5.186533880226263e-05,
      "loss": 4.8113,
      "mean_token_accuracy": 0.2474406123161316,
      "num_tokens": 20352569.0,
      "step": 11195
    },
    {
      "entropy": 5.730457019805908,
      "epoch": 9.622260421143103,
      "grad_norm": 1.4453125,
      "learning_rate": 5.1822858649125197e-05,
      "loss": 4.8833,
      "mean_token_accuracy": 0.23313068896532058,
      "num_tokens": 20361479.0,
      "step": 11200
    },
    {
      "entropy": 5.774404764175415,
      "epoch": 9.626557799742157,
      "grad_norm": 1.453125,
      "learning_rate": 5.178086582566134e-05,
      "loss": 4.9157,
      "mean_token_accuracy": 0.2357160970568657,
      "num_tokens": 20370286.0,
      "step": 11205
    },
    {
      "entropy": 5.705867576599121,
      "epoch": 9.630855178341212,
      "grad_norm": 1.3125,
      "learning_rate": 5.1739360423566596e-05,
      "loss": 4.8895,
      "mean_token_accuracy": 0.23882053792476654,
      "num_tokens": 20379253.0,
      "step": 11210
    },
    {
      "entropy": 5.724793434143066,
      "epoch": 9.635152556940266,
      "grad_norm": 1.390625,
      "learning_rate": 5.16983425334722e-05,
      "loss": 4.8406,
      "mean_token_accuracy": 0.24639717787504195,
      "num_tokens": 20387796.0,
      "step": 11215
    },
    {
      "entropy": 5.711037969589233,
      "epoch": 9.639449935539322,
      "grad_norm": 1.3515625,
      "learning_rate": 5.1657812244944796e-05,
      "loss": 4.7793,
      "mean_token_accuracy": 0.24511823803186417,
      "num_tokens": 20396209.0,
      "step": 11220
    },
    {
      "entropy": 5.801721382141113,
      "epoch": 9.643747314138375,
      "grad_norm": 1.40625,
      "learning_rate": 5.1617769646486344e-05,
      "loss": 4.9309,
      "mean_token_accuracy": 0.24493779987096786,
      "num_tokens": 20405102.0,
      "step": 11225
    },
    {
      "entropy": 5.7793690204620365,
      "epoch": 9.64804469273743,
      "grad_norm": 1.234375,
      "learning_rate": 5.157821482553389e-05,
      "loss": 4.984,
      "mean_token_accuracy": 0.22937669306993486,
      "num_tokens": 20415126.0,
      "step": 11230
    },
    {
      "entropy": 5.72899432182312,
      "epoch": 9.652342071336484,
      "grad_norm": 1.375,
      "learning_rate": 5.153914786845932e-05,
      "loss": 4.9059,
      "mean_token_accuracy": 0.235084168612957,
      "num_tokens": 20424166.0,
      "step": 11235
    },
    {
      "entropy": 5.665798377990723,
      "epoch": 9.65663944993554,
      "grad_norm": 1.53125,
      "learning_rate": 5.1500568860569285e-05,
      "loss": 4.7348,
      "mean_token_accuracy": 0.26206835061311723,
      "num_tokens": 20431984.0,
      "step": 11240
    },
    {
      "entropy": 5.659931087493897,
      "epoch": 9.660936828534593,
      "grad_norm": 1.4296875,
      "learning_rate": 5.1462477886104904e-05,
      "loss": 4.7933,
      "mean_token_accuracy": 0.24927198141813278,
      "num_tokens": 20440923.0,
      "step": 11245
    },
    {
      "entropy": 5.728438901901245,
      "epoch": 9.665234207133649,
      "grad_norm": 1.4453125,
      "learning_rate": 5.1424875028241625e-05,
      "loss": 4.8127,
      "mean_token_accuracy": 0.2419541597366333,
      "num_tokens": 20449525.0,
      "step": 11250
    },
    {
      "entropy": 5.731784915924072,
      "epoch": 9.669531585732702,
      "grad_norm": 1.203125,
      "learning_rate": 5.138776036908911e-05,
      "loss": 4.883,
      "mean_token_accuracy": 0.24157910495996476,
      "num_tokens": 20459193.0,
      "step": 11255
    },
    {
      "entropy": 5.709342765808105,
      "epoch": 9.673828964331758,
      "grad_norm": 1.5234375,
      "learning_rate": 5.135113398969091e-05,
      "loss": 4.8233,
      "mean_token_accuracy": 0.2402026414871216,
      "num_tokens": 20467564.0,
      "step": 11260
    },
    {
      "entropy": 5.78696117401123,
      "epoch": 9.678126342930812,
      "grad_norm": 1.328125,
      "learning_rate": 5.131499597002437e-05,
      "loss": 4.9665,
      "mean_token_accuracy": 0.234141606092453,
      "num_tokens": 20476573.0,
      "step": 11265
    },
    {
      "entropy": 5.719672203063965,
      "epoch": 9.682423721529867,
      "grad_norm": 1.3515625,
      "learning_rate": 5.12793463890005e-05,
      "loss": 4.9004,
      "mean_token_accuracy": 0.23852619677782058,
      "num_tokens": 20486788.0,
      "step": 11270
    },
    {
      "entropy": 5.757355690002441,
      "epoch": 9.68672110012892,
      "grad_norm": 1.21875,
      "learning_rate": 5.124418532446376e-05,
      "loss": 4.9113,
      "mean_token_accuracy": 0.23218149691820145,
      "num_tokens": 20495982.0,
      "step": 11275
    },
    {
      "entropy": 5.786857461929321,
      "epoch": 9.691018478727976,
      "grad_norm": 1.2890625,
      "learning_rate": 5.120951285319187e-05,
      "loss": 4.9401,
      "mean_token_accuracy": 0.23337600380182266,
      "num_tokens": 20504678.0,
      "step": 11280
    },
    {
      "entropy": 5.71163010597229,
      "epoch": 9.69531585732703,
      "grad_norm": 1.4140625,
      "learning_rate": 5.1175329050895584e-05,
      "loss": 4.7812,
      "mean_token_accuracy": 0.24595889896154405,
      "num_tokens": 20513362.0,
      "step": 11285
    },
    {
      "entropy": 5.695388889312744,
      "epoch": 9.699613235926085,
      "grad_norm": 1.3046875,
      "learning_rate": 5.114163399221871e-05,
      "loss": 4.9018,
      "mean_token_accuracy": 0.24140434116125106,
      "num_tokens": 20522280.0,
      "step": 11290
    },
    {
      "entropy": 5.683009386062622,
      "epoch": 9.703910614525139,
      "grad_norm": 1.46875,
      "learning_rate": 5.110842775073778e-05,
      "loss": 4.8258,
      "mean_token_accuracy": 0.24388017654418945,
      "num_tokens": 20531175.0,
      "step": 11295
    },
    {
      "entropy": 5.700381183624268,
      "epoch": 9.708207993124194,
      "grad_norm": 1.2734375,
      "learning_rate": 5.107571039896196e-05,
      "loss": 4.8818,
      "mean_token_accuracy": 0.24136573374271392,
      "num_tokens": 20540592.0,
      "step": 11300
    },
    {
      "entropy": 5.795329952239991,
      "epoch": 9.71250537172325,
      "grad_norm": 1.46875,
      "learning_rate": 5.1043482008332864e-05,
      "loss": 4.9774,
      "mean_token_accuracy": 0.22769286781549453,
      "num_tokens": 20549963.0,
      "step": 11305
    },
    {
      "entropy": 5.708670043945313,
      "epoch": 9.716802750322303,
      "grad_norm": 1.2421875,
      "learning_rate": 5.1011742649224394e-05,
      "loss": 4.8863,
      "mean_token_accuracy": 0.24391805678606032,
      "num_tokens": 20558928.0,
      "step": 11310
    },
    {
      "entropy": 5.684968519210815,
      "epoch": 9.721100128921359,
      "grad_norm": 1.5,
      "learning_rate": 5.098049239094267e-05,
      "loss": 4.7638,
      "mean_token_accuracy": 0.25189766734838487,
      "num_tokens": 20566483.0,
      "step": 11315
    },
    {
      "entropy": 5.751645565032959,
      "epoch": 9.725397507520412,
      "grad_norm": 1.390625,
      "learning_rate": 5.094973130172573e-05,
      "loss": 4.8502,
      "mean_token_accuracy": 0.24031373858451843,
      "num_tokens": 20575349.0,
      "step": 11320
    },
    {
      "entropy": 5.764202070236206,
      "epoch": 9.729694886119468,
      "grad_norm": 1.375,
      "learning_rate": 5.09194594487435e-05,
      "loss": 4.888,
      "mean_token_accuracy": 0.23896111100912093,
      "num_tokens": 20584235.0,
      "step": 11325
    },
    {
      "entropy": 5.65966477394104,
      "epoch": 9.733992264718522,
      "grad_norm": 1.40625,
      "learning_rate": 5.088967689809763e-05,
      "loss": 4.7997,
      "mean_token_accuracy": 0.2483944997191429,
      "num_tokens": 20593705.0,
      "step": 11330
    },
    {
      "entropy": 5.726091384887695,
      "epoch": 9.738289643317577,
      "grad_norm": 1.140625,
      "learning_rate": 5.086038371482128e-05,
      "loss": 4.9113,
      "mean_token_accuracy": 0.24341873675584794,
      "num_tokens": 20603886.0,
      "step": 11335
    },
    {
      "entropy": 5.5950987339019775,
      "epoch": 9.74258702191663,
      "grad_norm": 1.3046875,
      "learning_rate": 5.0831579962879074e-05,
      "loss": 4.7436,
      "mean_token_accuracy": 0.26151440143585203,
      "num_tokens": 20613181.0,
      "step": 11340
    },
    {
      "entropy": 5.748124647140503,
      "epoch": 9.746884400515686,
      "grad_norm": 1.34375,
      "learning_rate": 5.080326570516686e-05,
      "loss": 4.8578,
      "mean_token_accuracy": 0.24287873059511184,
      "num_tokens": 20620938.0,
      "step": 11345
    },
    {
      "entropy": 5.775603008270264,
      "epoch": 9.75118177911474,
      "grad_norm": 1.2734375,
      "learning_rate": 5.077544100351172e-05,
      "loss": 4.955,
      "mean_token_accuracy": 0.23114031553268433,
      "num_tokens": 20630992.0,
      "step": 11350
    },
    {
      "entropy": 5.784262037277221,
      "epoch": 9.755479157713795,
      "grad_norm": 1.40625,
      "learning_rate": 5.0748105918671616e-05,
      "loss": 4.9767,
      "mean_token_accuracy": 0.23523692786693573,
      "num_tokens": 20640543.0,
      "step": 11355
    },
    {
      "entropy": 5.749115753173828,
      "epoch": 9.759776536312849,
      "grad_norm": 1.2890625,
      "learning_rate": 5.072126051033551e-05,
      "loss": 4.88,
      "mean_token_accuracy": 0.23079000860452653,
      "num_tokens": 20649814.0,
      "step": 11360
    },
    {
      "entropy": 5.744604825973511,
      "epoch": 9.764073914911904,
      "grad_norm": 1.3046875,
      "learning_rate": 5.069490483712298e-05,
      "loss": 4.8877,
      "mean_token_accuracy": 0.24896368533372878,
      "num_tokens": 20658215.0,
      "step": 11365
    },
    {
      "entropy": 5.736106920242309,
      "epoch": 9.768371293510958,
      "grad_norm": 1.265625,
      "learning_rate": 5.066903895658433e-05,
      "loss": 4.8536,
      "mean_token_accuracy": 0.24528583735227585,
      "num_tokens": 20667412.0,
      "step": 11370
    },
    {
      "entropy": 5.701273393630982,
      "epoch": 9.772668672110013,
      "grad_norm": 1.40625,
      "learning_rate": 5.064366292520028e-05,
      "loss": 4.8855,
      "mean_token_accuracy": 0.24092838913202286,
      "num_tokens": 20676366.0,
      "step": 11375
    },
    {
      "entropy": 5.788467454910278,
      "epoch": 9.776966050709067,
      "grad_norm": 1.1953125,
      "learning_rate": 5.061877679838192e-05,
      "loss": 4.9789,
      "mean_token_accuracy": 0.22519542872905732,
      "num_tokens": 20686872.0,
      "step": 11380
    },
    {
      "entropy": 5.737813997268677,
      "epoch": 9.781263429308122,
      "grad_norm": 1.3984375,
      "learning_rate": 5.059438063047066e-05,
      "loss": 4.8193,
      "mean_token_accuracy": 0.24957498162984848,
      "num_tokens": 20695162.0,
      "step": 11385
    },
    {
      "entropy": 5.724996185302734,
      "epoch": 9.785560807907176,
      "grad_norm": 1.2890625,
      "learning_rate": 5.057047447473796e-05,
      "loss": 4.8617,
      "mean_token_accuracy": 0.2426105782389641,
      "num_tokens": 20703897.0,
      "step": 11390
    },
    {
      "entropy": 5.687663936614991,
      "epoch": 9.789858186506232,
      "grad_norm": 1.25,
      "learning_rate": 5.054705838338529e-05,
      "loss": 4.8257,
      "mean_token_accuracy": 0.2476524829864502,
      "num_tokens": 20712115.0,
      "step": 11395
    },
    {
      "entropy": 5.6803947448730465,
      "epoch": 9.794155565105285,
      "grad_norm": 1.28125,
      "learning_rate": 5.052413240754404e-05,
      "loss": 4.7912,
      "mean_token_accuracy": 0.24645674079656602,
      "num_tokens": 20721073.0,
      "step": 11400
    },
    {
      "entropy": 5.723237371444702,
      "epoch": 9.79845294370434,
      "grad_norm": 1.1796875,
      "learning_rate": 5.0501696597275376e-05,
      "loss": 4.8563,
      "mean_token_accuracy": 0.24140281826257706,
      "num_tokens": 20731370.0,
      "step": 11405
    },
    {
      "entropy": 5.745257997512818,
      "epoch": 9.802750322303394,
      "grad_norm": 1.3671875,
      "learning_rate": 5.047975100157018e-05,
      "loss": 4.9027,
      "mean_token_accuracy": 0.2401305392384529,
      "num_tokens": 20740744.0,
      "step": 11410
    },
    {
      "entropy": 5.727187061309815,
      "epoch": 9.80704770090245,
      "grad_norm": 1.2734375,
      "learning_rate": 5.045829566834879e-05,
      "loss": 4.8634,
      "mean_token_accuracy": 0.2440107509493828,
      "num_tokens": 20749838.0,
      "step": 11415
    },
    {
      "entropy": 5.711826944351197,
      "epoch": 9.811345079501503,
      "grad_norm": 1.21875,
      "learning_rate": 5.043733064446113e-05,
      "loss": 4.8006,
      "mean_token_accuracy": 0.24765003770589827,
      "num_tokens": 20759689.0,
      "step": 11420
    },
    {
      "entropy": 5.709393310546875,
      "epoch": 9.815642458100559,
      "grad_norm": 1.421875,
      "learning_rate": 5.041685597568641e-05,
      "loss": 4.8315,
      "mean_token_accuracy": 0.24392541050910949,
      "num_tokens": 20769234.0,
      "step": 11425
    },
    {
      "entropy": 5.69757490158081,
      "epoch": 9.819939836699612,
      "grad_norm": 1.3203125,
      "learning_rate": 5.039687170673315e-05,
      "loss": 4.9075,
      "mean_token_accuracy": 0.2447763815522194,
      "num_tokens": 20778794.0,
      "step": 11430
    },
    {
      "entropy": 5.717040491104126,
      "epoch": 9.824237215298668,
      "grad_norm": 1.2734375,
      "learning_rate": 5.037737788123895e-05,
      "loss": 4.8499,
      "mean_token_accuracy": 0.24264205247163773,
      "num_tokens": 20788327.0,
      "step": 11435
    },
    {
      "entropy": 5.753022718429565,
      "epoch": 9.828534593897722,
      "grad_norm": 1.265625,
      "learning_rate": 5.03583745417706e-05,
      "loss": 4.9045,
      "mean_token_accuracy": 0.23663776665925979,
      "num_tokens": 20797161.0,
      "step": 11440
    },
    {
      "entropy": 5.795938730239868,
      "epoch": 9.832831972496777,
      "grad_norm": 1.2734375,
      "learning_rate": 5.033986172982375e-05,
      "loss": 4.9604,
      "mean_token_accuracy": 0.236555577814579,
      "num_tokens": 20806264.0,
      "step": 11445
    },
    {
      "entropy": 5.737242841720581,
      "epoch": 9.837129351095832,
      "grad_norm": 1.3203125,
      "learning_rate": 5.0321839485823014e-05,
      "loss": 4.906,
      "mean_token_accuracy": 0.22974789142608643,
      "num_tokens": 20815569.0,
      "step": 11450
    },
    {
      "entropy": 5.778546380996704,
      "epoch": 9.841426729694886,
      "grad_norm": 1.375,
      "learning_rate": 5.030430784912177e-05,
      "loss": 4.9837,
      "mean_token_accuracy": 0.23175105601549148,
      "num_tokens": 20825234.0,
      "step": 11455
    },
    {
      "entropy": 5.661735200881958,
      "epoch": 9.845724108293942,
      "grad_norm": 1.375,
      "learning_rate": 5.0287266858002054e-05,
      "loss": 4.8381,
      "mean_token_accuracy": 0.2472504422068596,
      "num_tokens": 20834601.0,
      "step": 11460
    },
    {
      "entropy": 5.75831446647644,
      "epoch": 9.850021486892995,
      "grad_norm": 1.1640625,
      "learning_rate": 5.027071654967465e-05,
      "loss": 4.8487,
      "mean_token_accuracy": 0.24770977795124055,
      "num_tokens": 20843686.0,
      "step": 11465
    },
    {
      "entropy": 5.7528746128082275,
      "epoch": 9.85431886549205,
      "grad_norm": 1.3203125,
      "learning_rate": 5.025465696027875e-05,
      "loss": 4.8806,
      "mean_token_accuracy": 0.2449487879872322,
      "num_tokens": 20852009.0,
      "step": 11470
    },
    {
      "entropy": 5.733751058578491,
      "epoch": 9.858616244091104,
      "grad_norm": 1.375,
      "learning_rate": 5.023908812488211e-05,
      "loss": 4.843,
      "mean_token_accuracy": 0.24153771251440048,
      "num_tokens": 20861241.0,
      "step": 11475
    },
    {
      "entropy": 5.742742109298706,
      "epoch": 9.86291362269016,
      "grad_norm": 1.2109375,
      "learning_rate": 5.022401007748087e-05,
      "loss": 4.8754,
      "mean_token_accuracy": 0.23860252499580384,
      "num_tokens": 20869774.0,
      "step": 11480
    },
    {
      "entropy": 5.717617797851562,
      "epoch": 9.867211001289213,
      "grad_norm": 1.3984375,
      "learning_rate": 5.0209422850999414e-05,
      "loss": 4.9157,
      "mean_token_accuracy": 0.23705103248357773,
      "num_tokens": 20878844.0,
      "step": 11485
    },
    {
      "entropy": 5.723273706436157,
      "epoch": 9.871508379888269,
      "grad_norm": 1.296875,
      "learning_rate": 5.019532647729046e-05,
      "loss": 4.8968,
      "mean_token_accuracy": 0.23562048077583314,
      "num_tokens": 20888054.0,
      "step": 11490
    },
    {
      "entropy": 5.744698572158813,
      "epoch": 9.875805758487322,
      "grad_norm": 1.1640625,
      "learning_rate": 5.0181720987134815e-05,
      "loss": 4.8726,
      "mean_token_accuracy": 0.24331187903881074,
      "num_tokens": 20897752.0,
      "step": 11495
    },
    {
      "entropy": 5.623719358444214,
      "epoch": 9.880103137086378,
      "grad_norm": 1.453125,
      "learning_rate": 5.016860641024143e-05,
      "loss": 4.7521,
      "mean_token_accuracy": 0.25455262660980227,
      "num_tokens": 20906186.0,
      "step": 11500
    },
    {
      "epoch": 9.880103137086378,
      "eval_entropy": 5.549299109089482,
      "eval_loss": 5.895005702972412,
      "eval_mean_token_accuracy": 0.18285801942955265,
      "eval_num_tokens": 20906186.0,
      "eval_runtime": 2.0641,
      "eval_samples_per_second": 1719.433,
      "eval_steps_per_second": 215.111,
      "step": 11500
    }
  ],
  "logging_steps": 5,
  "max_steps": 11630,
  "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": 4703050292981760.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}