{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.741167129813418,
  "eval_steps": 3000,
  "global_step": 78000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 4.898969125747681,
      "epoch": 0.0004962286621675268,
      "grad_norm": 13.5625,
      "learning_rate": 2e-06,
      "loss": 14.3924,
      "mean_token_accuracy": 0.0,
      "num_tokens": 11448.0,
      "step": 5
    },
    {
      "entropy": 4.908018112182617,
      "epoch": 0.0009924573243350536,
      "grad_norm": 13.875,
      "learning_rate": 4.5e-06,
      "loss": 14.4058,
      "mean_token_accuracy": 0.0,
      "num_tokens": 22646.0,
      "step": 10
    },
    {
      "entropy": 4.919006061553955,
      "epoch": 0.0014886859865025804,
      "grad_norm": 13.875,
      "learning_rate": 7e-06,
      "loss": 14.3319,
      "mean_token_accuracy": 6.66666659526527e-05,
      "num_tokens": 34011.0,
      "step": 15
    },
    {
      "entropy": 4.953374242782592,
      "epoch": 0.001984914648670107,
      "grad_norm": 17.0,
      "learning_rate": 9.5e-06,
      "loss": 14.1459,
      "mean_token_accuracy": 0.0,
      "num_tokens": 44566.0,
      "step": 20
    },
    {
      "entropy": 5.0529521942138675,
      "epoch": 0.002481143310837634,
      "grad_norm": 20.375,
      "learning_rate": 1.2e-05,
      "loss": 13.7995,
      "mean_token_accuracy": 8.56164377182722e-05,
      "num_tokens": 56274.0,
      "step": 25
    },
    {
      "entropy": 5.29259762763977,
      "epoch": 0.0029773719730051607,
      "grad_norm": 25.75,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 13.2456,
      "mean_token_accuracy": 0.0,
      "num_tokens": 66584.0,
      "step": 30
    },
    {
      "entropy": 6.583654308319092,
      "epoch": 0.0034736006351726877,
      "grad_norm": 25.625,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 12.143,
      "mean_token_accuracy": 0.0,
      "num_tokens": 76948.0,
      "step": 35
    },
    {
      "entropy": 9.861779689788818,
      "epoch": 0.003969829297340214,
      "grad_norm": 3.375,
      "learning_rate": 1.95e-05,
      "loss": 10.9562,
      "mean_token_accuracy": 0.00021723387762904168,
      "num_tokens": 87253.0,
      "step": 40
    },
    {
      "entropy": 10.708572006225586,
      "epoch": 0.004466057959507741,
      "grad_norm": 3.4375,
      "learning_rate": 2.2e-05,
      "loss": 10.7237,
      "mean_token_accuracy": 0.010100901639088989,
      "num_tokens": 98273.0,
      "step": 45
    },
    {
      "entropy": 10.741462326049804,
      "epoch": 0.004962286621675268,
      "grad_norm": 3.015625,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 10.5536,
      "mean_token_accuracy": 0.017115887999534608,
      "num_tokens": 107788.0,
      "step": 50
    },
    {
      "entropy": 10.71127405166626,
      "epoch": 0.0054585152838427945,
      "grad_norm": 2.125,
      "learning_rate": 2.7e-05,
      "loss": 10.3864,
      "mean_token_accuracy": 0.015602491609752179,
      "num_tokens": 118632.0,
      "step": 55
    },
    {
      "entropy": 10.660900974273682,
      "epoch": 0.0059547439460103215,
      "grad_norm": 2.3125,
      "learning_rate": 2.95e-05,
      "loss": 10.1743,
      "mean_token_accuracy": 0.016560734901577234,
      "num_tokens": 128649.0,
      "step": 60
    },
    {
      "entropy": 10.705959510803222,
      "epoch": 0.0064509726081778485,
      "grad_norm": 1.703125,
      "learning_rate": 3.2e-05,
      "loss": 10.0457,
      "mean_token_accuracy": 0.01719390321522951,
      "num_tokens": 138958.0,
      "step": 65
    },
    {
      "entropy": 10.725889587402344,
      "epoch": 0.0069472012703453755,
      "grad_norm": 1.7734375,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.9768,
      "mean_token_accuracy": 0.04213028624653816,
      "num_tokens": 151445.0,
      "step": 70
    },
    {
      "entropy": 10.721480083465575,
      "epoch": 0.007443429932512902,
      "grad_norm": 1.6328125,
      "learning_rate": 3.7e-05,
      "loss": 9.8683,
      "mean_token_accuracy": 0.04831866100430489,
      "num_tokens": 162239.0,
      "step": 75
    },
    {
      "entropy": 10.70771484375,
      "epoch": 0.007939658594680429,
      "grad_norm": 1.7265625,
      "learning_rate": 3.95e-05,
      "loss": 9.8133,
      "mean_token_accuracy": 0.042871393263339996,
      "num_tokens": 173261.0,
      "step": 80
    },
    {
      "entropy": 10.706668090820312,
      "epoch": 0.008435887256847956,
      "grad_norm": 1.71875,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.6933,
      "mean_token_accuracy": 0.042860785871744154,
      "num_tokens": 183857.0,
      "step": 85
    },
    {
      "entropy": 10.696372127532959,
      "epoch": 0.008932115919015483,
      "grad_norm": 1.6015625,
      "learning_rate": 4.45e-05,
      "loss": 9.6957,
      "mean_token_accuracy": 0.04004955124109984,
      "num_tokens": 196635.0,
      "step": 90
    },
    {
      "entropy": 10.685241222381592,
      "epoch": 0.00942834458118301,
      "grad_norm": 1.6796875,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 9.5188,
      "mean_token_accuracy": 0.04844313133507967,
      "num_tokens": 208700.0,
      "step": 95
    },
    {
      "entropy": 10.666670799255371,
      "epoch": 0.009924573243350537,
      "grad_norm": 1.7421875,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 9.3761,
      "mean_token_accuracy": 0.046723760291934015,
      "num_tokens": 218915.0,
      "step": 100
    },
    {
      "entropy": 10.638776302337646,
      "epoch": 0.010420801905518062,
      "grad_norm": 1.640625,
      "learning_rate": 5.2e-05,
      "loss": 9.2697,
      "mean_token_accuracy": 0.052684008702635766,
      "num_tokens": 230211.0,
      "step": 105
    },
    {
      "entropy": 10.60411787033081,
      "epoch": 0.010917030567685589,
      "grad_norm": 1.6875,
      "learning_rate": 5.45e-05,
      "loss": 9.1633,
      "mean_token_accuracy": 0.05152133964002133,
      "num_tokens": 241174.0,
      "step": 110
    },
    {
      "entropy": 10.56174831390381,
      "epoch": 0.011413259229853116,
      "grad_norm": 1.546875,
      "learning_rate": 5.7e-05,
      "loss": 9.0895,
      "mean_token_accuracy": 0.047252026945352556,
      "num_tokens": 251708.0,
      "step": 115
    },
    {
      "entropy": 10.521473789215088,
      "epoch": 0.011909487892020643,
      "grad_norm": 1.4140625,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 9.0937,
      "mean_token_accuracy": 0.045639040693640706,
      "num_tokens": 263287.0,
      "step": 120
    },
    {
      "entropy": 10.479307842254638,
      "epoch": 0.01240571655418817,
      "grad_norm": 1.3515625,
      "learning_rate": 6.2e-05,
      "loss": 8.924,
      "mean_token_accuracy": 0.043540829047560695,
      "num_tokens": 275179.0,
      "step": 125
    },
    {
      "entropy": 10.418986606597901,
      "epoch": 0.012901945216355697,
      "grad_norm": 1.578125,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.7387,
      "mean_token_accuracy": 0.04855534341186285,
      "num_tokens": 287223.0,
      "step": 130
    },
    {
      "entropy": 10.346403312683105,
      "epoch": 0.013398173878523224,
      "grad_norm": 1.3984375,
      "learning_rate": 6.7e-05,
      "loss": 8.6159,
      "mean_token_accuracy": 0.04886538349092007,
      "num_tokens": 297868.0,
      "step": 135
    },
    {
      "entropy": 10.254747295379639,
      "epoch": 0.013894402540690751,
      "grad_norm": 1.328125,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.5351,
      "mean_token_accuracy": 0.050269151106476784,
      "num_tokens": 309702.0,
      "step": 140
    },
    {
      "entropy": 10.153191566467285,
      "epoch": 0.014390631202858276,
      "grad_norm": 1.2109375,
      "learning_rate": 7.2e-05,
      "loss": 8.5012,
      "mean_token_accuracy": 0.045820210874080655,
      "num_tokens": 321613.0,
      "step": 145
    },
    {
      "entropy": 10.02096643447876,
      "epoch": 0.014886859865025803,
      "grad_norm": 1.171875,
      "learning_rate": 7.45e-05,
      "loss": 8.3155,
      "mean_token_accuracy": 0.050194774940609935,
      "num_tokens": 332668.0,
      "step": 150
    },
    {
      "entropy": 9.884018039703369,
      "epoch": 0.01538308852719333,
      "grad_norm": 1.0703125,
      "learning_rate": 7.7e-05,
      "loss": 8.2768,
      "mean_token_accuracy": 0.050016530230641364,
      "num_tokens": 342579.0,
      "step": 155
    },
    {
      "entropy": 9.711176681518555,
      "epoch": 0.015879317189360857,
      "grad_norm": 1.1171875,
      "learning_rate": 7.950000000000001e-05,
      "loss": 8.126,
      "mean_token_accuracy": 0.04717333130538463,
      "num_tokens": 352271.0,
      "step": 160
    },
    {
      "entropy": 9.53788137435913,
      "epoch": 0.016375545851528384,
      "grad_norm": 1.0078125,
      "learning_rate": 8.2e-05,
      "loss": 8.0862,
      "mean_token_accuracy": 0.05482652448117733,
      "num_tokens": 364033.0,
      "step": 165
    },
    {
      "entropy": 9.262330722808837,
      "epoch": 0.01687177451369591,
      "grad_norm": 0.84375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.9892,
      "mean_token_accuracy": 0.04694691449403763,
      "num_tokens": 373788.0,
      "step": 170
    },
    {
      "entropy": 9.129754638671875,
      "epoch": 0.01736800317586344,
      "grad_norm": 0.8125,
      "learning_rate": 8.7e-05,
      "loss": 7.9617,
      "mean_token_accuracy": 0.044896982982754706,
      "num_tokens": 384883.0,
      "step": 175
    },
    {
      "entropy": 8.975637817382813,
      "epoch": 0.017864231838030965,
      "grad_norm": 0.93359375,
      "learning_rate": 8.95e-05,
      "loss": 7.8555,
      "mean_token_accuracy": 0.04881042614579201,
      "num_tokens": 396044.0,
      "step": 180
    },
    {
      "entropy": 8.740224266052246,
      "epoch": 0.018360460500198492,
      "grad_norm": 0.76953125,
      "learning_rate": 9.2e-05,
      "loss": 7.9152,
      "mean_token_accuracy": 0.047916683554649356,
      "num_tokens": 406995.0,
      "step": 185
    },
    {
      "entropy": 8.553805732727051,
      "epoch": 0.01885668916236602,
      "grad_norm": 0.78515625,
      "learning_rate": 9.45e-05,
      "loss": 7.7613,
      "mean_token_accuracy": 0.05165142118930817,
      "num_tokens": 417238.0,
      "step": 190
    },
    {
      "entropy": 8.464924621582032,
      "epoch": 0.019352917824533546,
      "grad_norm": 0.734375,
      "learning_rate": 9.7e-05,
      "loss": 7.8352,
      "mean_token_accuracy": 0.044884103909134866,
      "num_tokens": 427263.0,
      "step": 195
    },
    {
      "entropy": 8.39473524093628,
      "epoch": 0.019849146486701073,
      "grad_norm": 0.83984375,
      "learning_rate": 9.95e-05,
      "loss": 7.7506,
      "mean_token_accuracy": 0.04875692799687385,
      "num_tokens": 438357.0,
      "step": 200
    },
    {
      "entropy": 8.318683052062989,
      "epoch": 0.0203453751488686,
      "grad_norm": 0.67578125,
      "learning_rate": 0.000102,
      "loss": 7.7541,
      "mean_token_accuracy": 0.04936891198158264,
      "num_tokens": 448738.0,
      "step": 205
    },
    {
      "entropy": 8.268324375152588,
      "epoch": 0.020841603811036124,
      "grad_norm": 0.71875,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.76,
      "mean_token_accuracy": 0.056964365392923356,
      "num_tokens": 457873.0,
      "step": 210
    },
    {
      "entropy": 8.257455635070801,
      "epoch": 0.02133783247320365,
      "grad_norm": 0.71875,
      "learning_rate": 0.000107,
      "loss": 7.7046,
      "mean_token_accuracy": 0.05989706926047802,
      "num_tokens": 468640.0,
      "step": 215
    },
    {
      "entropy": 8.317792892456055,
      "epoch": 0.021834061135371178,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001095,
      "loss": 7.7039,
      "mean_token_accuracy": 0.06188451088964939,
      "num_tokens": 479922.0,
      "step": 220
    },
    {
      "entropy": 8.245875453948974,
      "epoch": 0.022330289797538705,
      "grad_norm": 0.75390625,
      "learning_rate": 0.000112,
      "loss": 7.663,
      "mean_token_accuracy": 0.053246379271149634,
      "num_tokens": 491056.0,
      "step": 225
    },
    {
      "entropy": 8.234527969360352,
      "epoch": 0.022826518459706232,
      "grad_norm": 0.875,
      "learning_rate": 0.0001145,
      "loss": 7.7427,
      "mean_token_accuracy": 0.05878411866724491,
      "num_tokens": 504365.0,
      "step": 230
    },
    {
      "entropy": 8.225464916229248,
      "epoch": 0.02332274712187376,
      "grad_norm": 1.0,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.638,
      "mean_token_accuracy": 0.057496064156293866,
      "num_tokens": 514347.0,
      "step": 235
    },
    {
      "entropy": 8.184865188598632,
      "epoch": 0.023818975784041286,
      "grad_norm": 1.15625,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.6521,
      "mean_token_accuracy": 0.062484626099467276,
      "num_tokens": 525912.0,
      "step": 240
    },
    {
      "entropy": 8.202389526367188,
      "epoch": 0.024315204446208813,
      "grad_norm": 1.0,
      "learning_rate": 0.000122,
      "loss": 7.5455,
      "mean_token_accuracy": 0.06251657865941525,
      "num_tokens": 535566.0,
      "step": 245
    },
    {
      "entropy": 8.209773063659668,
      "epoch": 0.02481143310837634,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0001245,
      "loss": 7.593,
      "mean_token_accuracy": 0.0647435687482357,
      "num_tokens": 546474.0,
      "step": 250
    },
    {
      "entropy": 8.14191026687622,
      "epoch": 0.025307661770543867,
      "grad_norm": 0.84375,
      "learning_rate": 0.000127,
      "loss": 7.6509,
      "mean_token_accuracy": 0.061424101889133456,
      "num_tokens": 558132.0,
      "step": 255
    },
    {
      "entropy": 8.203132820129394,
      "epoch": 0.025803890432711394,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001295,
      "loss": 7.6428,
      "mean_token_accuracy": 0.06428770869970321,
      "num_tokens": 568309.0,
      "step": 260
    },
    {
      "entropy": 8.081572723388671,
      "epoch": 0.02630011909487892,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000132,
      "loss": 7.5249,
      "mean_token_accuracy": 0.06533985063433648,
      "num_tokens": 579701.0,
      "step": 265
    },
    {
      "entropy": 8.113927555084228,
      "epoch": 0.026796347757046448,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.4971,
      "mean_token_accuracy": 0.06705491952598094,
      "num_tokens": 590832.0,
      "step": 270
    },
    {
      "entropy": 8.06861686706543,
      "epoch": 0.027292576419213975,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.4523,
      "mean_token_accuracy": 0.06724155806005001,
      "num_tokens": 601094.0,
      "step": 275
    },
    {
      "entropy": 8.04738645553589,
      "epoch": 0.027788805081381502,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0001395,
      "loss": 7.4503,
      "mean_token_accuracy": 0.0721239697188139,
      "num_tokens": 612027.0,
      "step": 280
    },
    {
      "entropy": 7.953780841827393,
      "epoch": 0.02828503374354903,
      "grad_norm": 0.84375,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.4716,
      "mean_token_accuracy": 0.0679593924432993,
      "num_tokens": 623698.0,
      "step": 285
    },
    {
      "entropy": 7.966112470626831,
      "epoch": 0.028781262405716553,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0001445,
      "loss": 7.4753,
      "mean_token_accuracy": 0.06937897726893424,
      "num_tokens": 635631.0,
      "step": 290
    },
    {
      "entropy": 7.981750106811523,
      "epoch": 0.02927749106788408,
      "grad_norm": 0.828125,
      "learning_rate": 0.000147,
      "loss": 7.4567,
      "mean_token_accuracy": 0.06901448890566826,
      "num_tokens": 647663.0,
      "step": 295
    },
    {
      "entropy": 7.9344415187835695,
      "epoch": 0.029773719730051607,
      "grad_norm": 0.96875,
      "learning_rate": 0.0001495,
      "loss": 7.4202,
      "mean_token_accuracy": 0.06708400063216687,
      "num_tokens": 659352.0,
      "step": 300
    },
    {
      "entropy": 7.962864351272583,
      "epoch": 0.030269948392219134,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000152,
      "loss": 7.3883,
      "mean_token_accuracy": 0.0732420600950718,
      "num_tokens": 669379.0,
      "step": 305
    },
    {
      "entropy": 7.942562055587769,
      "epoch": 0.03076617705438666,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.3844,
      "mean_token_accuracy": 0.07024326287209988,
      "num_tokens": 680144.0,
      "step": 310
    },
    {
      "entropy": 7.996751976013184,
      "epoch": 0.03126240571655419,
      "grad_norm": 1.15625,
      "learning_rate": 0.000157,
      "loss": 7.3426,
      "mean_token_accuracy": 0.06623242497444153,
      "num_tokens": 691165.0,
      "step": 315
    },
    {
      "entropy": 7.823547601699829,
      "epoch": 0.031758634378721715,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001595,
      "loss": 7.3996,
      "mean_token_accuracy": 0.06626546904444694,
      "num_tokens": 703767.0,
      "step": 320
    },
    {
      "entropy": 7.798118686676025,
      "epoch": 0.032254863040889245,
      "grad_norm": 1.03125,
      "learning_rate": 0.000162,
      "loss": 7.389,
      "mean_token_accuracy": 0.07417060770094394,
      "num_tokens": 714298.0,
      "step": 325
    },
    {
      "entropy": 7.857796287536621,
      "epoch": 0.03275109170305677,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.3782,
      "mean_token_accuracy": 0.07758076079189777,
      "num_tokens": 724135.0,
      "step": 330
    },
    {
      "entropy": 7.84428334236145,
      "epoch": 0.03324732036522429,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.4209,
      "mean_token_accuracy": 0.06709365807473659,
      "num_tokens": 735050.0,
      "step": 335
    },
    {
      "entropy": 7.832436752319336,
      "epoch": 0.03374354902739182,
      "grad_norm": 0.7265625,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.3356,
      "mean_token_accuracy": 0.07104014568030834,
      "num_tokens": 746073.0,
      "step": 340
    },
    {
      "entropy": 7.795631265640258,
      "epoch": 0.034239777689559346,
      "grad_norm": 0.828125,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.3294,
      "mean_token_accuracy": 0.06888764202594758,
      "num_tokens": 758998.0,
      "step": 345
    },
    {
      "entropy": 7.798632431030273,
      "epoch": 0.03473600635172688,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.2774,
      "mean_token_accuracy": 0.06999869644641876,
      "num_tokens": 771608.0,
      "step": 350
    },
    {
      "entropy": 7.839082527160644,
      "epoch": 0.0352322350138944,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000177,
      "loss": 7.3336,
      "mean_token_accuracy": 0.07213714122772216,
      "num_tokens": 782394.0,
      "step": 355
    },
    {
      "entropy": 7.719696950912476,
      "epoch": 0.03572846367606193,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001795,
      "loss": 7.2421,
      "mean_token_accuracy": 0.07258376255631446,
      "num_tokens": 792389.0,
      "step": 360
    },
    {
      "entropy": 7.709195804595947,
      "epoch": 0.036224692338229454,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000182,
      "loss": 7.2874,
      "mean_token_accuracy": 0.07547833137214184,
      "num_tokens": 803219.0,
      "step": 365
    },
    {
      "entropy": 7.799539947509766,
      "epoch": 0.036720921000396985,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001845,
      "loss": 7.3472,
      "mean_token_accuracy": 0.07126162201166153,
      "num_tokens": 814249.0,
      "step": 370
    },
    {
      "entropy": 7.732659769058228,
      "epoch": 0.03721714966256451,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000187,
      "loss": 7.2464,
      "mean_token_accuracy": 0.07738562412559986,
      "num_tokens": 825138.0,
      "step": 375
    },
    {
      "entropy": 7.711156368255615,
      "epoch": 0.03771337832473204,
      "grad_norm": 1.0625,
      "learning_rate": 0.0001895,
      "loss": 7.2878,
      "mean_token_accuracy": 0.07714832685887814,
      "num_tokens": 836024.0,
      "step": 380
    },
    {
      "entropy": 7.6839886665344235,
      "epoch": 0.03820960698689956,
      "grad_norm": 0.890625,
      "learning_rate": 0.000192,
      "loss": 7.28,
      "mean_token_accuracy": 0.0712902493774891,
      "num_tokens": 847765.0,
      "step": 385
    },
    {
      "entropy": 7.674788284301758,
      "epoch": 0.03870583564906709,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0001945,
      "loss": 7.274,
      "mean_token_accuracy": 0.07328340820968152,
      "num_tokens": 859980.0,
      "step": 390
    },
    {
      "entropy": 7.707912158966065,
      "epoch": 0.039202064311234616,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.2697,
      "mean_token_accuracy": 0.07509268596768379,
      "num_tokens": 871514.0,
      "step": 395
    },
    {
      "entropy": 7.807592153549194,
      "epoch": 0.03969829297340215,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.2594,
      "mean_token_accuracy": 0.06952392719686032,
      "num_tokens": 881984.0,
      "step": 400
    },
    {
      "entropy": 7.621029710769653,
      "epoch": 0.04019452163556967,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000202,
      "loss": 7.2334,
      "mean_token_accuracy": 0.07512535229325294,
      "num_tokens": 894198.0,
      "step": 405
    },
    {
      "entropy": 7.6571179866790775,
      "epoch": 0.0406907502977372,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00020449999999999998,
      "loss": 7.2387,
      "mean_token_accuracy": 0.07779023945331573,
      "num_tokens": 905394.0,
      "step": 410
    },
    {
      "entropy": 7.729351425170899,
      "epoch": 0.041186978959904724,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000207,
      "loss": 7.2892,
      "mean_token_accuracy": 0.0726883053779602,
      "num_tokens": 917010.0,
      "step": 415
    },
    {
      "entropy": 7.617552757263184,
      "epoch": 0.04168320762207225,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002095,
      "loss": 7.1201,
      "mean_token_accuracy": 0.07954838424921036,
      "num_tokens": 926761.0,
      "step": 420
    },
    {
      "entropy": 7.515660905838013,
      "epoch": 0.04217943628423978,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000212,
      "loss": 7.1273,
      "mean_token_accuracy": 0.07839518934488296,
      "num_tokens": 938385.0,
      "step": 425
    },
    {
      "entropy": 7.5848071575164795,
      "epoch": 0.0426756649464073,
      "grad_norm": 0.90625,
      "learning_rate": 0.0002145,
      "loss": 7.1617,
      "mean_token_accuracy": 0.07512866109609603,
      "num_tokens": 949709.0,
      "step": 430
    },
    {
      "entropy": 7.638444280624389,
      "epoch": 0.04317189360857483,
      "grad_norm": 1.0,
      "learning_rate": 0.00021700000000000002,
      "loss": 7.2004,
      "mean_token_accuracy": 0.07354060970246792,
      "num_tokens": 960837.0,
      "step": 435
    },
    {
      "entropy": 7.593792295455932,
      "epoch": 0.043668122270742356,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002195,
      "loss": 7.0385,
      "mean_token_accuracy": 0.08071191161870957,
      "num_tokens": 971036.0,
      "step": 440
    },
    {
      "entropy": 7.5954118251800535,
      "epoch": 0.044164350932909886,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000222,
      "loss": 7.1804,
      "mean_token_accuracy": 0.0766368854790926,
      "num_tokens": 983057.0,
      "step": 445
    },
    {
      "entropy": 7.580991935729981,
      "epoch": 0.04466057959507741,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0002245,
      "loss": 7.1365,
      "mean_token_accuracy": 0.08031548969447613,
      "num_tokens": 994303.0,
      "step": 450
    },
    {
      "entropy": 7.499250650405884,
      "epoch": 0.04515680825724494,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00022700000000000002,
      "loss": 7.135,
      "mean_token_accuracy": 0.07660448849201203,
      "num_tokens": 1005033.0,
      "step": 455
    },
    {
      "entropy": 7.613832330703735,
      "epoch": 0.045653036919412464,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00022950000000000002,
      "loss": 7.193,
      "mean_token_accuracy": 0.08052344620227814,
      "num_tokens": 1015940.0,
      "step": 460
    },
    {
      "entropy": 7.549146223068237,
      "epoch": 0.046149265581579994,
      "grad_norm": 1.109375,
      "learning_rate": 0.00023200000000000003,
      "loss": 7.1365,
      "mean_token_accuracy": 0.08448440171778202,
      "num_tokens": 1027582.0,
      "step": 465
    },
    {
      "entropy": 7.550874137878418,
      "epoch": 0.04664549424374752,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023449999999999998,
      "loss": 7.1147,
      "mean_token_accuracy": 0.07626789584755897,
      "num_tokens": 1040046.0,
      "step": 470
    },
    {
      "entropy": 7.5545329570770265,
      "epoch": 0.04714172290591505,
      "grad_norm": 1.40625,
      "learning_rate": 0.000237,
      "loss": 7.1922,
      "mean_token_accuracy": 0.0756704431027174,
      "num_tokens": 1050636.0,
      "step": 475
    },
    {
      "entropy": 7.618306350708008,
      "epoch": 0.04763795156808257,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002395,
      "loss": 7.0724,
      "mean_token_accuracy": 0.08306895196437836,
      "num_tokens": 1060748.0,
      "step": 480
    },
    {
      "entropy": 7.600764751434326,
      "epoch": 0.0481341802302501,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000242,
      "loss": 7.1277,
      "mean_token_accuracy": 0.07725501246750355,
      "num_tokens": 1071940.0,
      "step": 485
    },
    {
      "entropy": 7.485190486907959,
      "epoch": 0.048630408892417626,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002445,
      "loss": 7.1243,
      "mean_token_accuracy": 0.08531232550740242,
      "num_tokens": 1082962.0,
      "step": 490
    },
    {
      "entropy": 7.577714586257935,
      "epoch": 0.04912663755458515,
      "grad_norm": 1.078125,
      "learning_rate": 0.000247,
      "loss": 7.0988,
      "mean_token_accuracy": 0.08504232913255691,
      "num_tokens": 1093082.0,
      "step": 495
    },
    {
      "entropy": 7.54087610244751,
      "epoch": 0.04962286621675268,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0002495,
      "loss": 7.161,
      "mean_token_accuracy": 0.07727786526083946,
      "num_tokens": 1103528.0,
      "step": 500
    },
    {
      "entropy": 7.444737100601197,
      "epoch": 0.0501190948789202,
      "grad_norm": 1.359375,
      "learning_rate": 0.000252,
      "loss": 7.0602,
      "mean_token_accuracy": 0.07846606895327568,
      "num_tokens": 1114813.0,
      "step": 505
    },
    {
      "entropy": 7.511781406402588,
      "epoch": 0.050615323541087734,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002545,
      "loss": 7.0536,
      "mean_token_accuracy": 0.07985799759626389,
      "num_tokens": 1125345.0,
      "step": 510
    },
    {
      "entropy": 7.452861642837524,
      "epoch": 0.05111155220325526,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000257,
      "loss": 7.0618,
      "mean_token_accuracy": 0.08063343465328217,
      "num_tokens": 1134985.0,
      "step": 515
    },
    {
      "entropy": 7.497338771820068,
      "epoch": 0.05160778086542279,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0002595,
      "loss": 7.0742,
      "mean_token_accuracy": 0.08803893476724625,
      "num_tokens": 1145796.0,
      "step": 520
    },
    {
      "entropy": 7.637927579879761,
      "epoch": 0.05210400952759031,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000262,
      "loss": 7.207,
      "mean_token_accuracy": 0.07291420958936215,
      "num_tokens": 1156936.0,
      "step": 525
    },
    {
      "entropy": 7.462926721572876,
      "epoch": 0.05260023818975784,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00026450000000000003,
      "loss": 7.057,
      "mean_token_accuracy": 0.08704347461462021,
      "num_tokens": 1167526.0,
      "step": 530
    },
    {
      "entropy": 7.486857652664185,
      "epoch": 0.053096466851925365,
      "grad_norm": 1.109375,
      "learning_rate": 0.00026700000000000004,
      "loss": 7.1034,
      "mean_token_accuracy": 0.0767381489276886,
      "num_tokens": 1177647.0,
      "step": 535
    },
    {
      "entropy": 7.492286443710327,
      "epoch": 0.053592695514092896,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026950000000000005,
      "loss": 7.0725,
      "mean_token_accuracy": 0.08671703785657883,
      "num_tokens": 1188255.0,
      "step": 540
    },
    {
      "entropy": 7.378701829910279,
      "epoch": 0.05408892417626042,
      "grad_norm": 0.796875,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.9983,
      "mean_token_accuracy": 0.09135269373655319,
      "num_tokens": 1199433.0,
      "step": 545
    },
    {
      "entropy": 7.424235725402832,
      "epoch": 0.05458515283842795,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002745,
      "loss": 7.0643,
      "mean_token_accuracy": 0.08066366724669934,
      "num_tokens": 1210067.0,
      "step": 550
    },
    {
      "entropy": 7.368731451034546,
      "epoch": 0.055081381500595473,
      "grad_norm": 1.171875,
      "learning_rate": 0.000277,
      "loss": 6.956,
      "mean_token_accuracy": 0.0897083193063736,
      "num_tokens": 1220639.0,
      "step": 555
    },
    {
      "entropy": 7.4522312641143795,
      "epoch": 0.055577610162763004,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002795,
      "loss": 7.005,
      "mean_token_accuracy": 0.08417323902249337,
      "num_tokens": 1233158.0,
      "step": 560
    },
    {
      "entropy": 7.401830005645752,
      "epoch": 0.05607383882493053,
      "grad_norm": 1.125,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.9805,
      "mean_token_accuracy": 0.08172584101557731,
      "num_tokens": 1244552.0,
      "step": 565
    },
    {
      "entropy": 7.434595537185669,
      "epoch": 0.05657006748709806,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002845,
      "loss": 7.0868,
      "mean_token_accuracy": 0.07824886590242386,
      "num_tokens": 1255478.0,
      "step": 570
    },
    {
      "entropy": 7.333680963516235,
      "epoch": 0.05706629614926558,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000287,
      "loss": 6.9104,
      "mean_token_accuracy": 0.09092385321855545,
      "num_tokens": 1267507.0,
      "step": 575
    },
    {
      "entropy": 7.288129615783691,
      "epoch": 0.057562524811433105,
      "grad_norm": 0.953125,
      "learning_rate": 0.0002895,
      "loss": 6.9139,
      "mean_token_accuracy": 0.08322850465774537,
      "num_tokens": 1279377.0,
      "step": 580
    },
    {
      "entropy": 7.501294803619385,
      "epoch": 0.058058753473600636,
      "grad_norm": 1.28125,
      "learning_rate": 0.000292,
      "loss": 7.0026,
      "mean_token_accuracy": 0.09129699394106865,
      "num_tokens": 1290671.0,
      "step": 585
    },
    {
      "entropy": 7.294687795639038,
      "epoch": 0.05855498213576816,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0002945,
      "loss": 6.9489,
      "mean_token_accuracy": 0.09131397828459739,
      "num_tokens": 1302792.0,
      "step": 590
    },
    {
      "entropy": 7.357182216644287,
      "epoch": 0.05905121079793569,
      "grad_norm": 1.015625,
      "learning_rate": 0.000297,
      "loss": 6.9027,
      "mean_token_accuracy": 0.0812704510986805,
      "num_tokens": 1313315.0,
      "step": 595
    },
    {
      "entropy": 7.276323318481445,
      "epoch": 0.05954743946010321,
      "grad_norm": 0.9375,
      "learning_rate": 0.0002995,
      "loss": 6.9766,
      "mean_token_accuracy": 0.0856074646115303,
      "num_tokens": 1326358.0,
      "step": 600
    },
    {
      "entropy": 7.370176696777344,
      "epoch": 0.060043668122270744,
      "grad_norm": 0.921875,
      "learning_rate": 0.000302,
      "loss": 6.859,
      "mean_token_accuracy": 0.08850097134709359,
      "num_tokens": 1336266.0,
      "step": 605
    },
    {
      "entropy": 7.235168743133545,
      "epoch": 0.06053989678443827,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003045,
      "loss": 6.8108,
      "mean_token_accuracy": 0.0915584422647953,
      "num_tokens": 1347610.0,
      "step": 610
    },
    {
      "entropy": 7.234266328811645,
      "epoch": 0.0610361254466058,
      "grad_norm": 1.078125,
      "learning_rate": 0.000307,
      "loss": 6.7687,
      "mean_token_accuracy": 0.09413886740803719,
      "num_tokens": 1358317.0,
      "step": 615
    },
    {
      "entropy": 7.267807722091675,
      "epoch": 0.06153235410877332,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003095,
      "loss": 6.8764,
      "mean_token_accuracy": 0.08720959722995758,
      "num_tokens": 1370126.0,
      "step": 620
    },
    {
      "entropy": 7.298430395126343,
      "epoch": 0.06202858277094085,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000312,
      "loss": 6.837,
      "mean_token_accuracy": 0.0917629949748516,
      "num_tokens": 1381571.0,
      "step": 625
    },
    {
      "entropy": 7.124814748764038,
      "epoch": 0.06252481143310838,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003145,
      "loss": 6.8635,
      "mean_token_accuracy": 0.09271070286631584,
      "num_tokens": 1392486.0,
      "step": 630
    },
    {
      "entropy": 7.334286689758301,
      "epoch": 0.0630210400952759,
      "grad_norm": 0.984375,
      "learning_rate": 0.000317,
      "loss": 6.8681,
      "mean_token_accuracy": 0.08915356397628785,
      "num_tokens": 1403086.0,
      "step": 635
    },
    {
      "entropy": 7.2069190502166744,
      "epoch": 0.06351726875744343,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003195,
      "loss": 6.8319,
      "mean_token_accuracy": 0.09226809442043304,
      "num_tokens": 1414537.0,
      "step": 640
    },
    {
      "entropy": 7.251292371749878,
      "epoch": 0.06401349741961096,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000322,
      "loss": 6.8459,
      "mean_token_accuracy": 0.09251495376229286,
      "num_tokens": 1426998.0,
      "step": 645
    },
    {
      "entropy": 7.164758586883545,
      "epoch": 0.06450972608177849,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.874,
      "mean_token_accuracy": 0.09265056326985359,
      "num_tokens": 1437667.0,
      "step": 650
    },
    {
      "entropy": 7.226160907745362,
      "epoch": 0.065005954743946,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.7432,
      "mean_token_accuracy": 0.0969434417784214,
      "num_tokens": 1448280.0,
      "step": 655
    },
    {
      "entropy": 7.15698504447937,
      "epoch": 0.06550218340611354,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.8019,
      "mean_token_accuracy": 0.0953099973499775,
      "num_tokens": 1458608.0,
      "step": 660
    },
    {
      "entropy": 7.315108156204223,
      "epoch": 0.06599841206828107,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.9802,
      "mean_token_accuracy": 0.08709674105048179,
      "num_tokens": 1470435.0,
      "step": 665
    },
    {
      "entropy": 7.140678405761719,
      "epoch": 0.06649464073044858,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.8258,
      "mean_token_accuracy": 0.09219390302896499,
      "num_tokens": 1481959.0,
      "step": 670
    },
    {
      "entropy": 7.185245370864868,
      "epoch": 0.06699086939261611,
      "grad_norm": 1.09375,
      "learning_rate": 0.000337,
      "loss": 6.8687,
      "mean_token_accuracy": 0.08819933906197548,
      "num_tokens": 1493798.0,
      "step": 675
    },
    {
      "entropy": 7.220840215682983,
      "epoch": 0.06748709805478365,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003395,
      "loss": 6.6833,
      "mean_token_accuracy": 0.0931725412607193,
      "num_tokens": 1502969.0,
      "step": 680
    },
    {
      "entropy": 7.103082990646362,
      "epoch": 0.06798332671695118,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000342,
      "loss": 6.8289,
      "mean_token_accuracy": 0.09088226556777954,
      "num_tokens": 1513825.0,
      "step": 685
    },
    {
      "entropy": 7.194331789016724,
      "epoch": 0.06847955537911869,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.7387,
      "mean_token_accuracy": 0.09087632670998573,
      "num_tokens": 1525328.0,
      "step": 690
    },
    {
      "entropy": 7.107731008529663,
      "epoch": 0.06897578404128622,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000347,
      "loss": 6.7526,
      "mean_token_accuracy": 0.09229976609349251,
      "num_tokens": 1537383.0,
      "step": 695
    },
    {
      "entropy": 7.252395153045654,
      "epoch": 0.06947201270345375,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003495,
      "loss": 6.8497,
      "mean_token_accuracy": 0.09393966495990753,
      "num_tokens": 1549671.0,
      "step": 700
    },
    {
      "entropy": 7.091624927520752,
      "epoch": 0.06996824136562128,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000352,
      "loss": 6.728,
      "mean_token_accuracy": 0.08959824368357658,
      "num_tokens": 1559974.0,
      "step": 705
    },
    {
      "entropy": 7.1363232135772705,
      "epoch": 0.0704644700277888,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0003545,
      "loss": 6.8239,
      "mean_token_accuracy": 0.08678433373570442,
      "num_tokens": 1570890.0,
      "step": 710
    },
    {
      "entropy": 7.107376194000244,
      "epoch": 0.07096069868995633,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000357,
      "loss": 6.7323,
      "mean_token_accuracy": 0.0978621043264866,
      "num_tokens": 1581434.0,
      "step": 715
    },
    {
      "entropy": 7.089070653915405,
      "epoch": 0.07145692735212386,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003595,
      "loss": 6.7863,
      "mean_token_accuracy": 0.09178474843502045,
      "num_tokens": 1593710.0,
      "step": 720
    },
    {
      "entropy": 7.1520946502685545,
      "epoch": 0.07195315601429139,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000362,
      "loss": 6.7536,
      "mean_token_accuracy": 0.09313018918037415,
      "num_tokens": 1604635.0,
      "step": 725
    },
    {
      "entropy": 7.222806167602539,
      "epoch": 0.07244938467645891,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0003645,
      "loss": 6.8202,
      "mean_token_accuracy": 0.09525534659624099,
      "num_tokens": 1615258.0,
      "step": 730
    },
    {
      "entropy": 7.074941730499267,
      "epoch": 0.07294561333862644,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000367,
      "loss": 6.7125,
      "mean_token_accuracy": 0.09651728793978691,
      "num_tokens": 1625942.0,
      "step": 735
    },
    {
      "entropy": 7.0538249015808105,
      "epoch": 0.07344184200079397,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003695,
      "loss": 6.7046,
      "mean_token_accuracy": 0.1009585440158844,
      "num_tokens": 1636270.0,
      "step": 740
    },
    {
      "entropy": 7.074357271194458,
      "epoch": 0.07393807066296149,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000372,
      "loss": 6.6799,
      "mean_token_accuracy": 0.0913244754076004,
      "num_tokens": 1647029.0,
      "step": 745
    },
    {
      "entropy": 7.095760726928711,
      "epoch": 0.07443429932512902,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003745,
      "loss": 6.8498,
      "mean_token_accuracy": 0.09200523346662522,
      "num_tokens": 1659483.0,
      "step": 750
    },
    {
      "entropy": 6.94871416091919,
      "epoch": 0.07493052798729655,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000377,
      "loss": 6.7055,
      "mean_token_accuracy": 0.09416541829705238,
      "num_tokens": 1669425.0,
      "step": 755
    },
    {
      "entropy": 7.080055141448975,
      "epoch": 0.07542675664946408,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003795,
      "loss": 6.6949,
      "mean_token_accuracy": 0.09541306495666504,
      "num_tokens": 1680426.0,
      "step": 760
    },
    {
      "entropy": 7.007839250564575,
      "epoch": 0.0759229853116316,
      "grad_norm": 0.83203125,
      "learning_rate": 0.000382,
      "loss": 6.7124,
      "mean_token_accuracy": 0.09531945809721946,
      "num_tokens": 1692803.0,
      "step": 765
    },
    {
      "entropy": 7.0682658672332765,
      "epoch": 0.07641921397379912,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003845,
      "loss": 6.7343,
      "mean_token_accuracy": 0.09483553916215896,
      "num_tokens": 1703919.0,
      "step": 770
    },
    {
      "entropy": 7.002940464019775,
      "epoch": 0.07691544263596665,
      "grad_norm": 0.96875,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.7095,
      "mean_token_accuracy": 0.10153669938445091,
      "num_tokens": 1714487.0,
      "step": 775
    },
    {
      "entropy": 7.094982624053955,
      "epoch": 0.07741167129813419,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.7585,
      "mean_token_accuracy": 0.09340820983052253,
      "num_tokens": 1726601.0,
      "step": 780
    },
    {
      "entropy": 7.01391978263855,
      "epoch": 0.0779078999603017,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.6475,
      "mean_token_accuracy": 0.09487889260053635,
      "num_tokens": 1738743.0,
      "step": 785
    },
    {
      "entropy": 6.975992584228516,
      "epoch": 0.07840412862246923,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.7308,
      "mean_token_accuracy": 0.0951120376586914,
      "num_tokens": 1748286.0,
      "step": 790
    },
    {
      "entropy": 6.987929821014404,
      "epoch": 0.07890035728463676,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.7515,
      "mean_token_accuracy": 0.0957480825483799,
      "num_tokens": 1758917.0,
      "step": 795
    },
    {
      "entropy": 6.972993183135986,
      "epoch": 0.0793965859468043,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0003995,
      "loss": 6.5889,
      "mean_token_accuracy": 0.09930885285139084,
      "num_tokens": 1769179.0,
      "step": 800
    },
    {
      "entropy": 6.96280460357666,
      "epoch": 0.07989281460897181,
      "grad_norm": 0.9375,
      "learning_rate": 0.000402,
      "loss": 6.7138,
      "mean_token_accuracy": 0.08964183181524277,
      "num_tokens": 1782369.0,
      "step": 805
    },
    {
      "entropy": 6.994895315170288,
      "epoch": 0.08038904327113934,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004045,
      "loss": 6.7162,
      "mean_token_accuracy": 0.10041663944721221,
      "num_tokens": 1795992.0,
      "step": 810
    },
    {
      "entropy": 6.964673948287964,
      "epoch": 0.08088527193330687,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.6023,
      "mean_token_accuracy": 0.10524952039122581,
      "num_tokens": 1806938.0,
      "step": 815
    },
    {
      "entropy": 6.863758850097656,
      "epoch": 0.0813815005954744,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004095,
      "loss": 6.5447,
      "mean_token_accuracy": 0.0992793746292591,
      "num_tokens": 1817663.0,
      "step": 820
    },
    {
      "entropy": 6.998508739471435,
      "epoch": 0.08187772925764192,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000412,
      "loss": 6.6978,
      "mean_token_accuracy": 0.09649931266903877,
      "num_tokens": 1828254.0,
      "step": 825
    },
    {
      "entropy": 6.933160877227783,
      "epoch": 0.08237395791980945,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004145,
      "loss": 6.6293,
      "mean_token_accuracy": 0.09699171110987663,
      "num_tokens": 1839001.0,
      "step": 830
    },
    {
      "entropy": 6.939033889770508,
      "epoch": 0.08287018658197698,
      "grad_norm": 1.015625,
      "learning_rate": 0.000417,
      "loss": 6.7288,
      "mean_token_accuracy": 0.09662937819957733,
      "num_tokens": 1849906.0,
      "step": 835
    },
    {
      "entropy": 6.815816164016724,
      "epoch": 0.0833664152441445,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004195,
      "loss": 6.6076,
      "mean_token_accuracy": 0.09685141816735268,
      "num_tokens": 1860982.0,
      "step": 840
    },
    {
      "entropy": 7.031003141403199,
      "epoch": 0.08386264390631203,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000422,
      "loss": 6.5616,
      "mean_token_accuracy": 0.10065253376960755,
      "num_tokens": 1872177.0,
      "step": 845
    },
    {
      "entropy": 6.826659727096557,
      "epoch": 0.08435887256847956,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004245,
      "loss": 6.5362,
      "mean_token_accuracy": 0.0993971161544323,
      "num_tokens": 1882783.0,
      "step": 850
    },
    {
      "entropy": 7.006473255157471,
      "epoch": 0.08485510123064709,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000427,
      "loss": 6.7184,
      "mean_token_accuracy": 0.08898815661668777,
      "num_tokens": 1893706.0,
      "step": 855
    },
    {
      "entropy": 6.823424482345581,
      "epoch": 0.0853513298928146,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004295,
      "loss": 6.5757,
      "mean_token_accuracy": 0.09971656277775764,
      "num_tokens": 1903852.0,
      "step": 860
    },
    {
      "entropy": 6.9098859310150145,
      "epoch": 0.08584755855498213,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000432,
      "loss": 6.6087,
      "mean_token_accuracy": 0.09987007081508636,
      "num_tokens": 1914157.0,
      "step": 865
    },
    {
      "entropy": 6.880821752548218,
      "epoch": 0.08634378721714966,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004345,
      "loss": 6.5541,
      "mean_token_accuracy": 0.09982468262314796,
      "num_tokens": 1924684.0,
      "step": 870
    },
    {
      "entropy": 6.945840787887573,
      "epoch": 0.0868400158793172,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000437,
      "loss": 6.7066,
      "mean_token_accuracy": 0.0949609711766243,
      "num_tokens": 1935657.0,
      "step": 875
    },
    {
      "entropy": 6.870462036132812,
      "epoch": 0.08733624454148471,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004395,
      "loss": 6.5897,
      "mean_token_accuracy": 0.09813723415136337,
      "num_tokens": 1946735.0,
      "step": 880
    },
    {
      "entropy": 6.932937145233154,
      "epoch": 0.08783247320365224,
      "grad_norm": 1.046875,
      "learning_rate": 0.000442,
      "loss": 6.7502,
      "mean_token_accuracy": 0.09312896803021431,
      "num_tokens": 1957871.0,
      "step": 885
    },
    {
      "entropy": 6.838114881515503,
      "epoch": 0.08832870186581977,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004445,
      "loss": 6.5663,
      "mean_token_accuracy": 0.10240828990936279,
      "num_tokens": 1969032.0,
      "step": 890
    },
    {
      "entropy": 6.92586464881897,
      "epoch": 0.0888249305279873,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000447,
      "loss": 6.6063,
      "mean_token_accuracy": 0.09725166410207749,
      "num_tokens": 1980212.0,
      "step": 895
    },
    {
      "entropy": 6.780085325241089,
      "epoch": 0.08932115919015482,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.5267,
      "mean_token_accuracy": 0.10072881132364273,
      "num_tokens": 1990333.0,
      "step": 900
    },
    {
      "entropy": 6.879197597503662,
      "epoch": 0.08981738785232235,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.6025,
      "mean_token_accuracy": 0.10396269112825393,
      "num_tokens": 2001486.0,
      "step": 905
    },
    {
      "entropy": 6.857023143768311,
      "epoch": 0.09031361651448988,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.5679,
      "mean_token_accuracy": 0.1022559143602848,
      "num_tokens": 2012535.0,
      "step": 910
    },
    {
      "entropy": 6.766489458084107,
      "epoch": 0.0908098451766574,
      "grad_norm": 0.984375,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.5318,
      "mean_token_accuracy": 0.10583447441458702,
      "num_tokens": 2024116.0,
      "step": 915
    },
    {
      "entropy": 6.746376848220825,
      "epoch": 0.09130607383882493,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.5749,
      "mean_token_accuracy": 0.09981027767062187,
      "num_tokens": 2036341.0,
      "step": 920
    },
    {
      "entropy": 6.938368415832519,
      "epoch": 0.09180230250099246,
      "grad_norm": 0.9375,
      "learning_rate": 0.000462,
      "loss": 6.5504,
      "mean_token_accuracy": 0.10081811845302582,
      "num_tokens": 2046768.0,
      "step": 925
    },
    {
      "entropy": 6.7602132797241214,
      "epoch": 0.09229853116315999,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004645,
      "loss": 6.5178,
      "mean_token_accuracy": 0.09979332089424134,
      "num_tokens": 2057315.0,
      "step": 930
    },
    {
      "entropy": 6.767438125610352,
      "epoch": 0.0927947598253275,
      "grad_norm": 1.125,
      "learning_rate": 0.000467,
      "loss": 6.4968,
      "mean_token_accuracy": 0.10354413539171219,
      "num_tokens": 2068267.0,
      "step": 935
    },
    {
      "entropy": 6.767202854156494,
      "epoch": 0.09329098848749504,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004695,
      "loss": 6.5673,
      "mean_token_accuracy": 0.10121043622493744,
      "num_tokens": 2078271.0,
      "step": 940
    },
    {
      "entropy": 6.74484601020813,
      "epoch": 0.09378721714966257,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000472,
      "loss": 6.4897,
      "mean_token_accuracy": 0.09888025969266892,
      "num_tokens": 2088381.0,
      "step": 945
    },
    {
      "entropy": 6.789056253433228,
      "epoch": 0.0942834458118301,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004745,
      "loss": 6.451,
      "mean_token_accuracy": 0.09835516735911369,
      "num_tokens": 2098899.0,
      "step": 950
    },
    {
      "entropy": 6.749150371551513,
      "epoch": 0.09477967447399761,
      "grad_norm": 1.046875,
      "learning_rate": 0.000477,
      "loss": 6.5972,
      "mean_token_accuracy": 0.09927409738302231,
      "num_tokens": 2110057.0,
      "step": 955
    },
    {
      "entropy": 6.713812303543091,
      "epoch": 0.09527590313616514,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004795,
      "loss": 6.5065,
      "mean_token_accuracy": 0.09916102886199951,
      "num_tokens": 2121758.0,
      "step": 960
    },
    {
      "entropy": 6.771513843536377,
      "epoch": 0.09577213179833267,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000482,
      "loss": 6.5131,
      "mean_token_accuracy": 0.10310358032584191,
      "num_tokens": 2131840.0,
      "step": 965
    },
    {
      "entropy": 6.744855642318726,
      "epoch": 0.0962683604605002,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004845,
      "loss": 6.5007,
      "mean_token_accuracy": 0.1013672724366188,
      "num_tokens": 2142084.0,
      "step": 970
    },
    {
      "entropy": 6.749075984954834,
      "epoch": 0.09676458912266772,
      "grad_norm": 1.015625,
      "learning_rate": 0.000487,
      "loss": 6.5483,
      "mean_token_accuracy": 0.10051179081201553,
      "num_tokens": 2152785.0,
      "step": 975
    },
    {
      "entropy": 6.751393842697143,
      "epoch": 0.09726081778483525,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004895,
      "loss": 6.5224,
      "mean_token_accuracy": 0.1026117317378521,
      "num_tokens": 2163774.0,
      "step": 980
    },
    {
      "entropy": 6.7088648796081545,
      "epoch": 0.09775704644700278,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000492,
      "loss": 6.4589,
      "mean_token_accuracy": 0.10463614240288735,
      "num_tokens": 2175378.0,
      "step": 985
    },
    {
      "entropy": 6.910195636749267,
      "epoch": 0.0982532751091703,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004945,
      "loss": 6.6604,
      "mean_token_accuracy": 0.09883119091391564,
      "num_tokens": 2186510.0,
      "step": 990
    },
    {
      "entropy": 6.734104061126709,
      "epoch": 0.09874950377133783,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000497,
      "loss": 6.4553,
      "mean_token_accuracy": 0.10956107974052429,
      "num_tokens": 2197482.0,
      "step": 995
    },
    {
      "entropy": 6.622636842727661,
      "epoch": 0.09924573243350536,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995,
      "loss": 6.4604,
      "mean_token_accuracy": 0.10386316254734992,
      "num_tokens": 2208329.0,
      "step": 1000
    },
    {
      "entropy": 6.609873104095459,
      "epoch": 0.09974196109567289,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999999982145551,
      "loss": 6.362,
      "mean_token_accuracy": 0.10516428351402282,
      "num_tokens": 2218632.0,
      "step": 1005
    },
    {
      "entropy": 6.709473085403443,
      "epoch": 0.1002381897578404,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004999999909611855,
      "loss": 6.4814,
      "mean_token_accuracy": 0.09935311302542686,
      "num_tokens": 2231335.0,
      "step": 1010
    },
    {
      "entropy": 6.6939784526824955,
      "epoch": 0.10073441842000794,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999999781283008,
      "loss": 6.4733,
      "mean_token_accuracy": 0.10079516991972923,
      "num_tokens": 2241704.0,
      "step": 1015
    },
    {
      "entropy": 6.665507650375366,
      "epoch": 0.10123064708217547,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999999597159017,
      "loss": 6.505,
      "mean_token_accuracy": 0.1012033224105835,
      "num_tokens": 2252310.0,
      "step": 1020
    },
    {
      "entropy": 6.7658100605010985,
      "epoch": 0.101726875744343,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999999357239882,
      "loss": 6.4323,
      "mean_token_accuracy": 0.10344030633568764,
      "num_tokens": 2262270.0,
      "step": 1025
    },
    {
      "entropy": 6.723705816268921,
      "epoch": 0.10222310440651051,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999999061525612,
      "loss": 6.5677,
      "mean_token_accuracy": 0.10005603730678558,
      "num_tokens": 2273986.0,
      "step": 1030
    },
    {
      "entropy": 6.7837742328643795,
      "epoch": 0.10271933306867805,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999998710016214,
      "loss": 6.4629,
      "mean_token_accuracy": 0.10525803044438362,
      "num_tokens": 2283430.0,
      "step": 1035
    },
    {
      "entropy": 6.677142333984375,
      "epoch": 0.10321556173084558,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999998302711696,
      "loss": 6.4669,
      "mean_token_accuracy": 0.10035635456442833,
      "num_tokens": 2294201.0,
      "step": 1040
    },
    {
      "entropy": 6.609557294845581,
      "epoch": 0.1037117903930131,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004999997839612069,
      "loss": 6.3612,
      "mean_token_accuracy": 0.1090113341808319,
      "num_tokens": 2303876.0,
      "step": 1045
    },
    {
      "entropy": 6.626906967163086,
      "epoch": 0.10420801905518062,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999997320717342,
      "loss": 6.4015,
      "mean_token_accuracy": 0.10890848338603973,
      "num_tokens": 2313713.0,
      "step": 1050
    },
    {
      "entropy": 6.605928659439087,
      "epoch": 0.10470424771734815,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999996746027531,
      "loss": 6.4148,
      "mean_token_accuracy": 0.10125040039420127,
      "num_tokens": 2324862.0,
      "step": 1055
    },
    {
      "entropy": 6.717818832397461,
      "epoch": 0.10520047637951568,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499999611554265,
      "loss": 6.4953,
      "mean_token_accuracy": 0.10330296978354454,
      "num_tokens": 2334974.0,
      "step": 1060
    },
    {
      "entropy": 6.62673773765564,
      "epoch": 0.10569670504168321,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999995429262712,
      "loss": 6.5196,
      "mean_token_accuracy": 0.10395341366529465,
      "num_tokens": 2345945.0,
      "step": 1065
    },
    {
      "entropy": 6.722126197814942,
      "epoch": 0.10619293370385073,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999994687187737,
      "loss": 6.4608,
      "mean_token_accuracy": 0.10263997614383698,
      "num_tokens": 2356749.0,
      "step": 1070
    },
    {
      "entropy": 6.70174617767334,
      "epoch": 0.10668916236601826,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999993889317742,
      "loss": 6.5289,
      "mean_token_accuracy": 0.10041191279888154,
      "num_tokens": 2368181.0,
      "step": 1075
    },
    {
      "entropy": 6.65612006187439,
      "epoch": 0.10718539102818579,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999993035652746,
      "loss": 6.432,
      "mean_token_accuracy": 0.10550632029771805,
      "num_tokens": 2377737.0,
      "step": 1080
    },
    {
      "entropy": 6.570956897735596,
      "epoch": 0.10768161969035331,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004999992126192771,
      "loss": 6.3551,
      "mean_token_accuracy": 0.10739153623580933,
      "num_tokens": 2388944.0,
      "step": 1085
    },
    {
      "entropy": 6.695246934890747,
      "epoch": 0.10817784835252084,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499999116093784,
      "loss": 6.4357,
      "mean_token_accuracy": 0.10068753212690354,
      "num_tokens": 2400758.0,
      "step": 1090
    },
    {
      "entropy": 6.534470844268799,
      "epoch": 0.10867407701468837,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999990139887976,
      "loss": 6.4253,
      "mean_token_accuracy": 0.10705526322126388,
      "num_tokens": 2411916.0,
      "step": 1095
    },
    {
      "entropy": 6.811460256576538,
      "epoch": 0.1091703056768559,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999989063043206,
      "loss": 6.463,
      "mean_token_accuracy": 0.10148089230060578,
      "num_tokens": 2423310.0,
      "step": 1100
    },
    {
      "entropy": 6.5493433475494385,
      "epoch": 0.10966653433902342,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999987930403555,
      "loss": 6.3854,
      "mean_token_accuracy": 0.10383407324552536,
      "num_tokens": 2432100.0,
      "step": 1105
    },
    {
      "entropy": 6.6149815082550045,
      "epoch": 0.11016276300119095,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999986741969051,
      "loss": 6.4639,
      "mean_token_accuracy": 0.10388477817177773,
      "num_tokens": 2443563.0,
      "step": 1110
    },
    {
      "entropy": 6.726413249969482,
      "epoch": 0.11065899166335848,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999985497739724,
      "loss": 6.4566,
      "mean_token_accuracy": 0.10313622802495956,
      "num_tokens": 2455435.0,
      "step": 1115
    },
    {
      "entropy": 6.534851551055908,
      "epoch": 0.11115522032552601,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999984197715606,
      "loss": 6.3981,
      "mean_token_accuracy": 0.10903462842106819,
      "num_tokens": 2466173.0,
      "step": 1120
    },
    {
      "entropy": 6.549497175216675,
      "epoch": 0.11165144898769352,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999982841896728,
      "loss": 6.3521,
      "mean_token_accuracy": 0.10385128781199456,
      "num_tokens": 2476349.0,
      "step": 1125
    },
    {
      "entropy": 6.66382737159729,
      "epoch": 0.11214767764986106,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999981430283123,
      "loss": 6.4763,
      "mean_token_accuracy": 0.10946066528558732,
      "num_tokens": 2487864.0,
      "step": 1130
    },
    {
      "entropy": 6.5958610534667965,
      "epoch": 0.11264390631202859,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999979962874827,
      "loss": 6.4164,
      "mean_token_accuracy": 0.10289325043559075,
      "num_tokens": 2499188.0,
      "step": 1135
    },
    {
      "entropy": 6.616852045059204,
      "epoch": 0.11314013497419612,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999978439671877,
      "loss": 6.3896,
      "mean_token_accuracy": 0.10375408679246903,
      "num_tokens": 2509813.0,
      "step": 1140
    },
    {
      "entropy": 6.66924524307251,
      "epoch": 0.11363636363636363,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004999976860674309,
      "loss": 6.4185,
      "mean_token_accuracy": 0.10202216356992722,
      "num_tokens": 2520813.0,
      "step": 1145
    },
    {
      "entropy": 6.483026504516602,
      "epoch": 0.11413259229853116,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999975225882164,
      "loss": 6.4173,
      "mean_token_accuracy": 0.1033550575375557,
      "num_tokens": 2531802.0,
      "step": 1150
    },
    {
      "entropy": 6.586180353164673,
      "epoch": 0.1146288209606987,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499997353529548,
      "loss": 6.2121,
      "mean_token_accuracy": 0.1153721533715725,
      "num_tokens": 2541935.0,
      "step": 1155
    },
    {
      "entropy": 6.49102201461792,
      "epoch": 0.11512504962286621,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999971788914303,
      "loss": 6.3252,
      "mean_token_accuracy": 0.10732494816184043,
      "num_tokens": 2553224.0,
      "step": 1160
    },
    {
      "entropy": 6.557978200912475,
      "epoch": 0.11562127828503374,
      "grad_norm": 0.7421875,
      "learning_rate": 0.0004999969986738672,
      "loss": 6.3578,
      "mean_token_accuracy": 0.10425093993544579,
      "num_tokens": 2565355.0,
      "step": 1165
    },
    {
      "entropy": 6.631026411056519,
      "epoch": 0.11611750694720127,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999968128768634,
      "loss": 6.4405,
      "mean_token_accuracy": 0.10526153668761254,
      "num_tokens": 2575962.0,
      "step": 1170
    },
    {
      "entropy": 6.586783170700073,
      "epoch": 0.1166137356093688,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999966215004236,
      "loss": 6.4478,
      "mean_token_accuracy": 0.11039511635899543,
      "num_tokens": 2588136.0,
      "step": 1175
    },
    {
      "entropy": 6.563306188583374,
      "epoch": 0.11710996427153632,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999964245445522,
      "loss": 6.3313,
      "mean_token_accuracy": 0.10910526439547538,
      "num_tokens": 2597785.0,
      "step": 1180
    },
    {
      "entropy": 6.518818330764771,
      "epoch": 0.11760619293370385,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004999962220092545,
      "loss": 6.3785,
      "mean_token_accuracy": 0.11630177646875381,
      "num_tokens": 2609738.0,
      "step": 1185
    },
    {
      "entropy": 6.455830097198486,
      "epoch": 0.11810242159587138,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004999960138945351,
      "loss": 6.2671,
      "mean_token_accuracy": 0.11771797314286232,
      "num_tokens": 2620179.0,
      "step": 1190
    },
    {
      "entropy": 6.5756360530853275,
      "epoch": 0.11859865025803891,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999958002003996,
      "loss": 6.3035,
      "mean_token_accuracy": 0.11005059778690338,
      "num_tokens": 2631550.0,
      "step": 1195
    },
    {
      "entropy": 6.5420623302459715,
      "epoch": 0.11909487892020643,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004999955809268529,
      "loss": 6.3474,
      "mean_token_accuracy": 0.1102082036435604,
      "num_tokens": 2642467.0,
      "step": 1200
    },
    {
      "entropy": 6.430347681045532,
      "epoch": 0.11959110758237396,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999953560739008,
      "loss": 6.3182,
      "mean_token_accuracy": 0.1130327932536602,
      "num_tokens": 2652132.0,
      "step": 1205
    },
    {
      "entropy": 6.609172010421753,
      "epoch": 0.12008733624454149,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004999951256415485,
      "loss": 6.3786,
      "mean_token_accuracy": 0.1051353357732296,
      "num_tokens": 2663336.0,
      "step": 1210
    },
    {
      "entropy": 6.602077627182007,
      "epoch": 0.12058356490670902,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000499994889629802,
      "loss": 6.4557,
      "mean_token_accuracy": 0.09855944886803628,
      "num_tokens": 2674201.0,
      "step": 1215
    },
    {
      "entropy": 6.591161394119263,
      "epoch": 0.12107979356887653,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000499994648038667,
      "loss": 6.3755,
      "mean_token_accuracy": 0.11215761303901672,
      "num_tokens": 2685073.0,
      "step": 1220
    },
    {
      "entropy": 6.588635969161987,
      "epoch": 0.12157602223104406,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999944008681496,
      "loss": 6.3189,
      "mean_token_accuracy": 0.11205925047397614,
      "num_tokens": 2695024.0,
      "step": 1225
    },
    {
      "entropy": 6.502772092819214,
      "epoch": 0.1220722508932116,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999941481182559,
      "loss": 6.4082,
      "mean_token_accuracy": 0.10132400244474411,
      "num_tokens": 2706232.0,
      "step": 1230
    },
    {
      "entropy": 6.634514093399048,
      "epoch": 0.12256847955537911,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999938897889921,
      "loss": 6.3776,
      "mean_token_accuracy": 0.11380477249622345,
      "num_tokens": 2715874.0,
      "step": 1235
    },
    {
      "entropy": 6.467554330825806,
      "epoch": 0.12306470821754664,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004999936258803647,
      "loss": 6.3171,
      "mean_token_accuracy": 0.11320996284484863,
      "num_tokens": 2727409.0,
      "step": 1240
    },
    {
      "entropy": 6.4485221862792965,
      "epoch": 0.12356093687971417,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999933563923802,
      "loss": 6.2336,
      "mean_token_accuracy": 0.11560766994953156,
      "num_tokens": 2737259.0,
      "step": 1245
    },
    {
      "entropy": 6.5671069622039795,
      "epoch": 0.1240571655418817,
      "grad_norm": 0.7734375,
      "learning_rate": 0.0004999930813250451,
      "loss": 6.3636,
      "mean_token_accuracy": 0.10681080147624016,
      "num_tokens": 2748416.0,
      "step": 1250
    },
    {
      "entropy": 6.504324531555175,
      "epoch": 0.12455339420404922,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999928006783667,
      "loss": 6.3299,
      "mean_token_accuracy": 0.10958396941423416,
      "num_tokens": 2760470.0,
      "step": 1255
    },
    {
      "entropy": 6.538706254959107,
      "epoch": 0.12504962286621676,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999925144523515,
      "loss": 6.3608,
      "mean_token_accuracy": 0.10441428348422051,
      "num_tokens": 2770222.0,
      "step": 1260
    },
    {
      "entropy": 6.533433151245117,
      "epoch": 0.12554585152838427,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999922226470066,
      "loss": 6.2804,
      "mean_token_accuracy": 0.10863575115799903,
      "num_tokens": 2780212.0,
      "step": 1265
    },
    {
      "entropy": 6.4365581512451175,
      "epoch": 0.1260420801905518,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999919252623397,
      "loss": 6.2603,
      "mean_token_accuracy": 0.11084286272525787,
      "num_tokens": 2790239.0,
      "step": 1270
    },
    {
      "entropy": 6.580913019180298,
      "epoch": 0.12653830885271933,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999916222983577,
      "loss": 6.2971,
      "mean_token_accuracy": 0.1081952191889286,
      "num_tokens": 2801902.0,
      "step": 1275
    },
    {
      "entropy": 6.360956907272339,
      "epoch": 0.12703453751488686,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999913137550682,
      "loss": 6.2987,
      "mean_token_accuracy": 0.10948901474475861,
      "num_tokens": 2813268.0,
      "step": 1280
    },
    {
      "entropy": 6.5253222465515135,
      "epoch": 0.1275307661770544,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004999909996324791,
      "loss": 6.3583,
      "mean_token_accuracy": 0.1083506740629673,
      "num_tokens": 2824506.0,
      "step": 1285
    },
    {
      "entropy": 6.4849005222320555,
      "epoch": 0.12802699483922192,
      "grad_norm": 0.84375,
      "learning_rate": 0.000499990679930598,
      "loss": 6.2729,
      "mean_token_accuracy": 0.11370046511292457,
      "num_tokens": 2835144.0,
      "step": 1290
    },
    {
      "entropy": 6.4290557384490965,
      "epoch": 0.12852322350138945,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999903546494327,
      "loss": 6.2832,
      "mean_token_accuracy": 0.1095927357673645,
      "num_tokens": 2845382.0,
      "step": 1295
    },
    {
      "entropy": 6.500859642028809,
      "epoch": 0.12901945216355698,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999900237889916,
      "loss": 6.2942,
      "mean_token_accuracy": 0.11329072564840317,
      "num_tokens": 2854697.0,
      "step": 1300
    },
    {
      "entropy": 6.496008634567261,
      "epoch": 0.12951568082572448,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999896873492827,
      "loss": 6.313,
      "mean_token_accuracy": 0.10269764885306358,
      "num_tokens": 2866216.0,
      "step": 1305
    },
    {
      "entropy": 6.478988075256348,
      "epoch": 0.130011909487892,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999893453303143,
      "loss": 6.2662,
      "mean_token_accuracy": 0.11140501946210861,
      "num_tokens": 2877734.0,
      "step": 1310
    },
    {
      "entropy": 6.496612596511841,
      "epoch": 0.13050813815005954,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999889977320949,
      "loss": 6.3097,
      "mean_token_accuracy": 0.11451938673853874,
      "num_tokens": 2889445.0,
      "step": 1315
    },
    {
      "entropy": 6.386645030975342,
      "epoch": 0.13100436681222707,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004999886445546332,
      "loss": 6.2526,
      "mean_token_accuracy": 0.117067901045084,
      "num_tokens": 2900485.0,
      "step": 1320
    },
    {
      "entropy": 6.536042022705078,
      "epoch": 0.1315005954743946,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000499988285797938,
      "loss": 6.2462,
      "mean_token_accuracy": 0.11725969165563584,
      "num_tokens": 2909795.0,
      "step": 1325
    },
    {
      "entropy": 6.310559272766113,
      "epoch": 0.13199682413656214,
      "grad_norm": 0.859375,
      "learning_rate": 0.000499987921462018,
      "loss": 6.2739,
      "mean_token_accuracy": 0.120875982940197,
      "num_tokens": 2920153.0,
      "step": 1330
    },
    {
      "entropy": 6.545910024642945,
      "epoch": 0.13249305279872967,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999875515468825,
      "loss": 6.2668,
      "mean_token_accuracy": 0.11391363963484764,
      "num_tokens": 2930457.0,
      "step": 1335
    },
    {
      "entropy": 6.387372589111328,
      "epoch": 0.13298928146089717,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999871760525403,
      "loss": 6.3293,
      "mean_token_accuracy": 0.11570871099829674,
      "num_tokens": 2941305.0,
      "step": 1340
    },
    {
      "entropy": 6.462648534774781,
      "epoch": 0.1334855101230647,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999867949790011,
      "loss": 6.2594,
      "mean_token_accuracy": 0.11346142515540122,
      "num_tokens": 2951798.0,
      "step": 1345
    },
    {
      "entropy": 6.408753728866577,
      "epoch": 0.13398173878523223,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999864083262743,
      "loss": 6.1704,
      "mean_token_accuracy": 0.11709224805235863,
      "num_tokens": 2962486.0,
      "step": 1350
    },
    {
      "entropy": 6.384300184249878,
      "epoch": 0.13447796744739976,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999860160943691,
      "loss": 6.2681,
      "mean_token_accuracy": 0.11269479319453239,
      "num_tokens": 2974317.0,
      "step": 1355
    },
    {
      "entropy": 6.450789546966552,
      "epoch": 0.1349741961095673,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999856182832958,
      "loss": 6.1904,
      "mean_token_accuracy": 0.11520234048366547,
      "num_tokens": 2985440.0,
      "step": 1360
    },
    {
      "entropy": 6.400120830535888,
      "epoch": 0.13547042477173482,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999852148930637,
      "loss": 6.2605,
      "mean_token_accuracy": 0.10958047136664391,
      "num_tokens": 2997044.0,
      "step": 1365
    },
    {
      "entropy": 6.447021579742431,
      "epoch": 0.13596665343390235,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999848059236832,
      "loss": 6.2687,
      "mean_token_accuracy": 0.1083021655678749,
      "num_tokens": 3005965.0,
      "step": 1370
    },
    {
      "entropy": 6.3724065780639645,
      "epoch": 0.13646288209606988,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999843913751643,
      "loss": 6.1383,
      "mean_token_accuracy": 0.11827749311923981,
      "num_tokens": 3015031.0,
      "step": 1375
    },
    {
      "entropy": 6.378967905044556,
      "epoch": 0.13695911075823738,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999839712475174,
      "loss": 6.2074,
      "mean_token_accuracy": 0.1124014399945736,
      "num_tokens": 3026275.0,
      "step": 1380
    },
    {
      "entropy": 6.385042381286621,
      "epoch": 0.13745533942040492,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999835455407526,
      "loss": 6.2493,
      "mean_token_accuracy": 0.11228244602680207,
      "num_tokens": 3036778.0,
      "step": 1385
    },
    {
      "entropy": 6.488673353195191,
      "epoch": 0.13795156808257245,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999831142548808,
      "loss": 6.3516,
      "mean_token_accuracy": 0.11366798952221871,
      "num_tokens": 3048149.0,
      "step": 1390
    },
    {
      "entropy": 6.426563882827759,
      "epoch": 0.13844779674473998,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999826773899124,
      "loss": 6.2154,
      "mean_token_accuracy": 0.11358957141637802,
      "num_tokens": 3059149.0,
      "step": 1395
    },
    {
      "entropy": 6.404417276382446,
      "epoch": 0.1389440254069075,
      "grad_norm": 0.74609375,
      "learning_rate": 0.0004999822349458586,
      "loss": 6.2257,
      "mean_token_accuracy": 0.11496646702289581,
      "num_tokens": 3071613.0,
      "step": 1400
    },
    {
      "entropy": 6.5001030445098875,
      "epoch": 0.13944025406907504,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999817869227301,
      "loss": 6.369,
      "mean_token_accuracy": 0.1097478836774826,
      "num_tokens": 3084742.0,
      "step": 1405
    },
    {
      "entropy": 6.461193561553955,
      "epoch": 0.13993648273124257,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999813333205381,
      "loss": 6.2857,
      "mean_token_accuracy": 0.11472547352313996,
      "num_tokens": 3095356.0,
      "step": 1410
    },
    {
      "entropy": 6.409702301025391,
      "epoch": 0.14043271139341007,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999808741392937,
      "loss": 6.2669,
      "mean_token_accuracy": 0.10798696279525757,
      "num_tokens": 3107717.0,
      "step": 1415
    },
    {
      "entropy": 6.36297869682312,
      "epoch": 0.1409289400555776,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004999804093790086,
      "loss": 6.2518,
      "mean_token_accuracy": 0.1127444364130497,
      "num_tokens": 3119205.0,
      "step": 1420
    },
    {
      "entropy": 6.448381614685059,
      "epoch": 0.14142516871774513,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000499979939039694,
      "loss": 6.3038,
      "mean_token_accuracy": 0.11262011229991913,
      "num_tokens": 3130564.0,
      "step": 1425
    },
    {
      "entropy": 6.450186538696289,
      "epoch": 0.14192139737991266,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004999794631213618,
      "loss": 6.2395,
      "mean_token_accuracy": 0.11082697659730911,
      "num_tokens": 3141883.0,
      "step": 1430
    },
    {
      "entropy": 6.420169115066528,
      "epoch": 0.1424176260420802,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004999789816240236,
      "loss": 6.2394,
      "mean_token_accuracy": 0.10963290259242058,
      "num_tokens": 3152817.0,
      "step": 1435
    },
    {
      "entropy": 6.358748054504394,
      "epoch": 0.14291385470424772,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999784945476915,
      "loss": 6.1567,
      "mean_token_accuracy": 0.11966277733445167,
      "num_tokens": 3163546.0,
      "step": 1440
    },
    {
      "entropy": 6.412687873840332,
      "epoch": 0.14341008336641525,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999780018923775,
      "loss": 6.207,
      "mean_token_accuracy": 0.10822459608316422,
      "num_tokens": 3173899.0,
      "step": 1445
    },
    {
      "entropy": 6.287933111190796,
      "epoch": 0.14390631202858278,
      "grad_norm": 0.875,
      "learning_rate": 0.0004999775036580939,
      "loss": 6.11,
      "mean_token_accuracy": 0.11830191910266877,
      "num_tokens": 3184433.0,
      "step": 1450
    },
    {
      "entropy": 6.4349367141723635,
      "epoch": 0.1444025406907503,
      "grad_norm": 0.875,
      "learning_rate": 0.0004999769998448529,
      "loss": 6.22,
      "mean_token_accuracy": 0.11617036014795304,
      "num_tokens": 3196077.0,
      "step": 1455
    },
    {
      "entropy": 6.256659460067749,
      "epoch": 0.14489876935291782,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999764904526672,
      "loss": 6.1422,
      "mean_token_accuracy": 0.11813490390777588,
      "num_tokens": 3206645.0,
      "step": 1460
    },
    {
      "entropy": 6.442942762374878,
      "epoch": 0.14539499801508535,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999759754815492,
      "loss": 6.2781,
      "mean_token_accuracy": 0.11664444580674171,
      "num_tokens": 3216786.0,
      "step": 1465
    },
    {
      "entropy": 6.370675897598266,
      "epoch": 0.14589122667725288,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004999754549315118,
      "loss": 6.1909,
      "mean_token_accuracy": 0.11156717166304589,
      "num_tokens": 3227384.0,
      "step": 1470
    },
    {
      "entropy": 6.271988248825073,
      "epoch": 0.1463874553394204,
      "grad_norm": 0.80859375,
      "learning_rate": 0.000499974928802568,
      "loss": 6.0496,
      "mean_token_accuracy": 0.1194495476782322,
      "num_tokens": 3238800.0,
      "step": 1475
    },
    {
      "entropy": 6.343875980377197,
      "epoch": 0.14688368400158794,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999743970947306,
      "loss": 6.1983,
      "mean_token_accuracy": 0.11200206577777863,
      "num_tokens": 3248627.0,
      "step": 1480
    },
    {
      "entropy": 6.364979076385498,
      "epoch": 0.14737991266375547,
      "grad_norm": 0.890625,
      "learning_rate": 0.000499973859808013,
      "loss": 6.2264,
      "mean_token_accuracy": 0.11476045995950698,
      "num_tokens": 3258989.0,
      "step": 1485
    },
    {
      "entropy": 6.351506185531616,
      "epoch": 0.14787614132592297,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999733169424284,
      "loss": 6.1945,
      "mean_token_accuracy": 0.12000712603330613,
      "num_tokens": 3269474.0,
      "step": 1490
    },
    {
      "entropy": 6.345127201080322,
      "epoch": 0.1483723699880905,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999727684979904,
      "loss": 6.1851,
      "mean_token_accuracy": 0.11789674162864686,
      "num_tokens": 3280453.0,
      "step": 1495
    },
    {
      "entropy": 6.31245641708374,
      "epoch": 0.14886859865025803,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999722144747125,
      "loss": 6.1203,
      "mean_token_accuracy": 0.11437854543328285,
      "num_tokens": 3290234.0,
      "step": 1500
    },
    {
      "entropy": 6.290363025665283,
      "epoch": 0.14936482731242556,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999716548726085,
      "loss": 6.1151,
      "mean_token_accuracy": 0.12060677483677865,
      "num_tokens": 3300943.0,
      "step": 1505
    },
    {
      "entropy": 6.312642621994018,
      "epoch": 0.1498610559745931,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999710896916921,
      "loss": 6.0684,
      "mean_token_accuracy": 0.11790815740823746,
      "num_tokens": 3311296.0,
      "step": 1510
    },
    {
      "entropy": 6.33518762588501,
      "epoch": 0.15035728463676062,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999705189319774,
      "loss": 6.2183,
      "mean_token_accuracy": 0.11259444132447242,
      "num_tokens": 3322310.0,
      "step": 1515
    },
    {
      "entropy": 6.35755672454834,
      "epoch": 0.15085351329892815,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999699425934788,
      "loss": 6.2299,
      "mean_token_accuracy": 0.11464733257889748,
      "num_tokens": 3334176.0,
      "step": 1520
    },
    {
      "entropy": 6.353241586685181,
      "epoch": 0.15134974196109569,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004999693606762103,
      "loss": 6.2134,
      "mean_token_accuracy": 0.11368680894374847,
      "num_tokens": 3344839.0,
      "step": 1525
    },
    {
      "entropy": 6.369286584854126,
      "epoch": 0.1518459706232632,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999687731801865,
      "loss": 6.1841,
      "mean_token_accuracy": 0.11411794796586036,
      "num_tokens": 3357046.0,
      "step": 1530
    },
    {
      "entropy": 6.3504150867462155,
      "epoch": 0.15234219928543072,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999681801054217,
      "loss": 6.2434,
      "mean_token_accuracy": 0.11438931301236152,
      "num_tokens": 3367783.0,
      "step": 1535
    },
    {
      "entropy": 6.324097442626953,
      "epoch": 0.15283842794759825,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000499967581451931,
      "loss": 6.1145,
      "mean_token_accuracy": 0.11726456582546234,
      "num_tokens": 3378529.0,
      "step": 1540
    },
    {
      "entropy": 6.375859260559082,
      "epoch": 0.15333465660976578,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499966977219729,
      "loss": 6.3061,
      "mean_token_accuracy": 0.11245756894350052,
      "num_tokens": 3389318.0,
      "step": 1545
    },
    {
      "entropy": 6.410177373886109,
      "epoch": 0.1538308852719333,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999663674088306,
      "loss": 6.3082,
      "mean_token_accuracy": 0.1074009008705616,
      "num_tokens": 3400603.0,
      "step": 1550
    },
    {
      "entropy": 6.406285333633423,
      "epoch": 0.15432711393410084,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999657520192511,
      "loss": 6.2016,
      "mean_token_accuracy": 0.11285926327109337,
      "num_tokens": 3410316.0,
      "step": 1555
    },
    {
      "entropy": 6.356779527664185,
      "epoch": 0.15482334259626837,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004999651310510058,
      "loss": 6.1416,
      "mean_token_accuracy": 0.1178070843219757,
      "num_tokens": 3422155.0,
      "step": 1560
    },
    {
      "entropy": 6.345295333862305,
      "epoch": 0.15531957125843587,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999645045041099,
      "loss": 6.0819,
      "mean_token_accuracy": 0.11718083545565605,
      "num_tokens": 3432892.0,
      "step": 1565
    },
    {
      "entropy": 6.321739959716797,
      "epoch": 0.1558157999206034,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004999638723785791,
      "loss": 6.2004,
      "mean_token_accuracy": 0.11840709745883941,
      "num_tokens": 3443967.0,
      "step": 1570
    },
    {
      "entropy": 6.270855236053467,
      "epoch": 0.15631202858277093,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000499963234674429,
      "loss": 6.0916,
      "mean_token_accuracy": 0.12166468054056168,
      "num_tokens": 3454354.0,
      "step": 1575
    },
    {
      "entropy": 6.233002376556397,
      "epoch": 0.15680825724493846,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999625913916754,
      "loss": 6.1319,
      "mean_token_accuracy": 0.11802673935890198,
      "num_tokens": 3465232.0,
      "step": 1580
    },
    {
      "entropy": 6.242019701004028,
      "epoch": 0.157304485907106,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999619425303343,
      "loss": 6.0925,
      "mean_token_accuracy": 0.11856844201683998,
      "num_tokens": 3475443.0,
      "step": 1585
    },
    {
      "entropy": 6.3711628913879395,
      "epoch": 0.15780071456927353,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004999612880904219,
      "loss": 6.1274,
      "mean_token_accuracy": 0.11918400153517723,
      "num_tokens": 3486643.0,
      "step": 1590
    },
    {
      "entropy": 6.28782172203064,
      "epoch": 0.15829694323144106,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999606280719542,
      "loss": 6.185,
      "mean_token_accuracy": 0.11784008368849755,
      "num_tokens": 3497557.0,
      "step": 1595
    },
    {
      "entropy": 6.421589040756226,
      "epoch": 0.1587931718936086,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999599624749475,
      "loss": 6.1637,
      "mean_token_accuracy": 0.11806859821081161,
      "num_tokens": 3509790.0,
      "step": 1600
    },
    {
      "entropy": 6.215350532531739,
      "epoch": 0.1592894005557761,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004999592912994187,
      "loss": 6.0908,
      "mean_token_accuracy": 0.11676915585994721,
      "num_tokens": 3521577.0,
      "step": 1605
    },
    {
      "entropy": 6.3262897491455075,
      "epoch": 0.15978562921794362,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999586145453842,
      "loss": 6.1432,
      "mean_token_accuracy": 0.11143859475851059,
      "num_tokens": 3532126.0,
      "step": 1610
    },
    {
      "entropy": 6.2068733215332035,
      "epoch": 0.16028185788011115,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999579322128608,
      "loss": 5.972,
      "mean_token_accuracy": 0.11642664968967438,
      "num_tokens": 3542560.0,
      "step": 1615
    },
    {
      "entropy": 6.257067346572876,
      "epoch": 0.16077808654227868,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999572443018654,
      "loss": 6.1404,
      "mean_token_accuracy": 0.11960313320159913,
      "num_tokens": 3553863.0,
      "step": 1620
    },
    {
      "entropy": 6.282857131958008,
      "epoch": 0.1612743152044462,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999565508124151,
      "loss": 6.1603,
      "mean_token_accuracy": 0.12152265384793282,
      "num_tokens": 3564953.0,
      "step": 1625
    },
    {
      "entropy": 6.252542781829834,
      "epoch": 0.16177054386661374,
      "grad_norm": 0.78125,
      "learning_rate": 0.000499955851744527,
      "loss": 6.1588,
      "mean_token_accuracy": 0.1182580165565014,
      "num_tokens": 3576247.0,
      "step": 1630
    },
    {
      "entropy": 6.4327959537506105,
      "epoch": 0.16226677252878127,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004999551470982185,
      "loss": 6.1834,
      "mean_token_accuracy": 0.11378399357199669,
      "num_tokens": 3587600.0,
      "step": 1635
    },
    {
      "entropy": 6.352491140365601,
      "epoch": 0.1627630011909488,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999544368735073,
      "loss": 6.1676,
      "mean_token_accuracy": 0.1208123430609703,
      "num_tokens": 3597727.0,
      "step": 1640
    },
    {
      "entropy": 6.288443946838379,
      "epoch": 0.1632592298531163,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999537210704106,
      "loss": 6.1383,
      "mean_token_accuracy": 0.1199887067079544,
      "num_tokens": 3608066.0,
      "step": 1645
    },
    {
      "entropy": 6.309280347824097,
      "epoch": 0.16375545851528384,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004999529996889463,
      "loss": 6.2006,
      "mean_token_accuracy": 0.11445491313934326,
      "num_tokens": 3618263.0,
      "step": 1650
    },
    {
      "entropy": 6.344433927536011,
      "epoch": 0.16425168717745137,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999522727291325,
      "loss": 6.1369,
      "mean_token_accuracy": 0.12186064049601555,
      "num_tokens": 3629465.0,
      "step": 1655
    },
    {
      "entropy": 6.342315721511841,
      "epoch": 0.1647479158396189,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000499951540190987,
      "loss": 6.1583,
      "mean_token_accuracy": 0.11373883932828903,
      "num_tokens": 3639657.0,
      "step": 1660
    },
    {
      "entropy": 6.328229236602783,
      "epoch": 0.16524414450178643,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004999508020745281,
      "loss": 6.1261,
      "mean_token_accuracy": 0.11836064383387565,
      "num_tokens": 3652374.0,
      "step": 1665
    },
    {
      "entropy": 6.265553331375122,
      "epoch": 0.16574037316395396,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004999500583797739,
      "loss": 6.1072,
      "mean_token_accuracy": 0.12078006193041801,
      "num_tokens": 3664083.0,
      "step": 1670
    },
    {
      "entropy": 6.2652623653411865,
      "epoch": 0.1662366018261215,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004999493091067431,
      "loss": 6.1452,
      "mean_token_accuracy": 0.12134512290358543,
      "num_tokens": 3675314.0,
      "step": 1675
    },
    {
      "entropy": 6.273673057556152,
      "epoch": 0.166732830488289,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004999485542554541,
      "loss": 6.0998,
      "mean_token_accuracy": 0.11119712963700294,
      "num_tokens": 3686019.0,
      "step": 1680
    },
    {
      "entropy": 6.370255327224731,
      "epoch": 0.16722905915045652,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004999477938259256,
      "loss": 6.2015,
      "mean_token_accuracy": 0.1111437164247036,
      "num_tokens": 3698170.0,
      "step": 1685
    },
    {
      "entropy": 6.2584208965301515,
      "epoch": 0.16772528781262405,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004999470278181766,
      "loss": 6.181,
      "mean_token_accuracy": 0.11148267313838005,
      "num_tokens": 3709243.0,
      "step": 1690
    },
    {
      "entropy": 6.344109964370728,
      "epoch": 0.16822151647479158,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999462562322261,
      "loss": 6.1492,
      "mean_token_accuracy": 0.11816644594073296,
      "num_tokens": 3720196.0,
      "step": 1695
    },
    {
      "entropy": 6.374466037750244,
      "epoch": 0.1687177451369591,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000499945479068093,
      "loss": 6.1937,
      "mean_token_accuracy": 0.1157242514193058,
      "num_tokens": 3732714.0,
      "step": 1700
    },
    {
      "entropy": 6.284525728225708,
      "epoch": 0.16921397379912664,
      "grad_norm": 0.72265625,
      "learning_rate": 0.0004999446963257969,
      "loss": 6.1734,
      "mean_token_accuracy": 0.11130392551422119,
      "num_tokens": 3746474.0,
      "step": 1705
    },
    {
      "entropy": 6.26992883682251,
      "epoch": 0.16971020246129417,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999439080053571,
      "loss": 6.0446,
      "mean_token_accuracy": 0.12372053563594818,
      "num_tokens": 3757528.0,
      "step": 1710
    },
    {
      "entropy": 6.382873582839966,
      "epoch": 0.1702064311234617,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999431141067928,
      "loss": 6.18,
      "mean_token_accuracy": 0.11336961910128593,
      "num_tokens": 3768360.0,
      "step": 1715
    },
    {
      "entropy": 6.226411628723144,
      "epoch": 0.1707026597856292,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999423146301242,
      "loss": 6.053,
      "mean_token_accuracy": 0.11719192564487457,
      "num_tokens": 3779809.0,
      "step": 1720
    },
    {
      "entropy": 6.197938585281372,
      "epoch": 0.17119888844779674,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999415095753709,
      "loss": 6.0517,
      "mean_token_accuracy": 0.11497935056686401,
      "num_tokens": 3791455.0,
      "step": 1725
    },
    {
      "entropy": 6.194182014465332,
      "epoch": 0.17169511710996427,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999406989425527,
      "loss": 6.0768,
      "mean_token_accuracy": 0.11933901458978653,
      "num_tokens": 3803209.0,
      "step": 1730
    },
    {
      "entropy": 6.272842121124268,
      "epoch": 0.1721913457721318,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999398827316901,
      "loss": 6.118,
      "mean_token_accuracy": 0.11888522580265999,
      "num_tokens": 3813553.0,
      "step": 1735
    },
    {
      "entropy": 6.242817163467407,
      "epoch": 0.17268757443429933,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000499939060942803,
      "loss": 6.1491,
      "mean_token_accuracy": 0.1162618339061737,
      "num_tokens": 3824856.0,
      "step": 1740
    },
    {
      "entropy": 6.3598274230957035,
      "epoch": 0.17318380309646686,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999382335759119,
      "loss": 6.226,
      "mean_token_accuracy": 0.1098976731300354,
      "num_tokens": 3835489.0,
      "step": 1745
    },
    {
      "entropy": 6.3387504577636715,
      "epoch": 0.1736800317586344,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999374006310372,
      "loss": 6.0514,
      "mean_token_accuracy": 0.11634560897946358,
      "num_tokens": 3844838.0,
      "step": 1750
    },
    {
      "entropy": 6.168754911422729,
      "epoch": 0.1741762604208019,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004999365621081999,
      "loss": 6.0342,
      "mean_token_accuracy": 0.11432853490114211,
      "num_tokens": 3855288.0,
      "step": 1755
    },
    {
      "entropy": 6.248133897781372,
      "epoch": 0.17467248908296942,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999357180074204,
      "loss": 6.0702,
      "mean_token_accuracy": 0.12020740732550621,
      "num_tokens": 3865585.0,
      "step": 1760
    },
    {
      "entropy": 6.307549571990966,
      "epoch": 0.17516871774513695,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004999348683287199,
      "loss": 6.0612,
      "mean_token_accuracy": 0.11682183891534806,
      "num_tokens": 3877694.0,
      "step": 1765
    },
    {
      "entropy": 6.190867567062378,
      "epoch": 0.17566494640730448,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004999340130721192,
      "loss": 6.0438,
      "mean_token_accuracy": 0.12537719830870628,
      "num_tokens": 3888802.0,
      "step": 1770
    },
    {
      "entropy": 6.209321069717407,
      "epoch": 0.17616117506947201,
      "grad_norm": 0.875,
      "learning_rate": 0.0004999331522376396,
      "loss": 6.0208,
      "mean_token_accuracy": 0.11791236400604248,
      "num_tokens": 3900294.0,
      "step": 1775
    },
    {
      "entropy": 6.202359390258789,
      "epoch": 0.17665740373163955,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999322858253027,
      "loss": 6.0417,
      "mean_token_accuracy": 0.12317918911576271,
      "num_tokens": 3910865.0,
      "step": 1780
    },
    {
      "entropy": 6.220766210556031,
      "epoch": 0.17715363239380708,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999314138351297,
      "loss": 6.1082,
      "mean_token_accuracy": 0.1180627316236496,
      "num_tokens": 3921960.0,
      "step": 1785
    },
    {
      "entropy": 6.266336345672608,
      "epoch": 0.1776498610559746,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999305362671424,
      "loss": 6.1053,
      "mean_token_accuracy": 0.11726907566189766,
      "num_tokens": 3934085.0,
      "step": 1790
    },
    {
      "entropy": 6.166120910644532,
      "epoch": 0.1781460897181421,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004999296531213624,
      "loss": 5.9587,
      "mean_token_accuracy": 0.1321515955030918,
      "num_tokens": 3944532.0,
      "step": 1795
    },
    {
      "entropy": 6.186281824111939,
      "epoch": 0.17864231838030964,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999287643978117,
      "loss": 5.9965,
      "mean_token_accuracy": 0.12600135430693626,
      "num_tokens": 3954995.0,
      "step": 1800
    },
    {
      "entropy": 6.13634295463562,
      "epoch": 0.17913854704247717,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004999278700965124,
      "loss": 6.045,
      "mean_token_accuracy": 0.12481234520673752,
      "num_tokens": 3966621.0,
      "step": 1805
    },
    {
      "entropy": 6.31223611831665,
      "epoch": 0.1796347757046447,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004999269702174866,
      "loss": 6.1086,
      "mean_token_accuracy": 0.11732375845313073,
      "num_tokens": 3977568.0,
      "step": 1810
    },
    {
      "entropy": 6.178112936019898,
      "epoch": 0.18013100436681223,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999260647607565,
      "loss": 6.1296,
      "mean_token_accuracy": 0.11571917533874512,
      "num_tokens": 3988571.0,
      "step": 1815
    },
    {
      "entropy": 6.251215600967408,
      "epoch": 0.18062723302897976,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999251537263448,
      "loss": 5.9563,
      "mean_token_accuracy": 0.12610897570848464,
      "num_tokens": 3998380.0,
      "step": 1820
    },
    {
      "entropy": 6.153609037399292,
      "epoch": 0.1811234616911473,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004999242371142739,
      "loss": 6.1257,
      "mean_token_accuracy": 0.12231617867946624,
      "num_tokens": 4010017.0,
      "step": 1825
    },
    {
      "entropy": 6.336256694793701,
      "epoch": 0.1816196903533148,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004999233149245665,
      "loss": 6.0924,
      "mean_token_accuracy": 0.11703047156333923,
      "num_tokens": 4020922.0,
      "step": 1830
    },
    {
      "entropy": 6.280692195892334,
      "epoch": 0.18211591901548232,
      "grad_norm": 0.875,
      "learning_rate": 0.0004999223871572456,
      "loss": 6.0664,
      "mean_token_accuracy": 0.1190974049270153,
      "num_tokens": 4030927.0,
      "step": 1835
    },
    {
      "entropy": 6.214899921417237,
      "epoch": 0.18261214767764986,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999214538123342,
      "loss": 6.1446,
      "mean_token_accuracy": 0.11380955725908279,
      "num_tokens": 4041375.0,
      "step": 1840
    },
    {
      "entropy": 6.2237224102020265,
      "epoch": 0.18310837633981739,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999205148898554,
      "loss": 6.1269,
      "mean_token_accuracy": 0.11845389157533645,
      "num_tokens": 4053213.0,
      "step": 1845
    },
    {
      "entropy": 6.275475740432739,
      "epoch": 0.18360460500198492,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999195703898325,
      "loss": 5.9784,
      "mean_token_accuracy": 0.12606104388833045,
      "num_tokens": 4064536.0,
      "step": 1850
    },
    {
      "entropy": 6.190648603439331,
      "epoch": 0.18410083366415245,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999186203122889,
      "loss": 6.06,
      "mean_token_accuracy": 0.12409329041838646,
      "num_tokens": 4075282.0,
      "step": 1855
    },
    {
      "entropy": 6.171055507659912,
      "epoch": 0.18459706232631998,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999176646572482,
      "loss": 6.057,
      "mean_token_accuracy": 0.12450164556503296,
      "num_tokens": 4085501.0,
      "step": 1860
    },
    {
      "entropy": 6.222785091400146,
      "epoch": 0.1850932909884875,
      "grad_norm": 0.890625,
      "learning_rate": 0.000499916703424734,
      "loss": 6.0077,
      "mean_token_accuracy": 0.11859095394611359,
      "num_tokens": 4095697.0,
      "step": 1865
    },
    {
      "entropy": 6.199232578277588,
      "epoch": 0.185589519650655,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004999157366147702,
      "loss": 6.0605,
      "mean_token_accuracy": 0.1195147417485714,
      "num_tokens": 4106459.0,
      "step": 1870
    },
    {
      "entropy": 6.221849536895752,
      "epoch": 0.18608574831282254,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004999147642273809,
      "loss": 5.9945,
      "mean_token_accuracy": 0.12537158504128457,
      "num_tokens": 4118320.0,
      "step": 1875
    },
    {
      "entropy": 6.132491636276245,
      "epoch": 0.18658197697499007,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00049991378626259,
      "loss": 6.018,
      "mean_token_accuracy": 0.12773770093917847,
      "num_tokens": 4131249.0,
      "step": 1880
    },
    {
      "entropy": 6.259493064880371,
      "epoch": 0.1870782056371576,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999128027204219,
      "loss": 6.0923,
      "mean_token_accuracy": 0.12224588692188262,
      "num_tokens": 4141480.0,
      "step": 1885
    },
    {
      "entropy": 6.227820491790771,
      "epoch": 0.18757443429932513,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000499911813600901,
      "loss": 6.1457,
      "mean_token_accuracy": 0.11545902863144875,
      "num_tokens": 4152344.0,
      "step": 1890
    },
    {
      "entropy": 6.2940919399261475,
      "epoch": 0.18807066296149266,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999108189040516,
      "loss": 6.1088,
      "mean_token_accuracy": 0.12004400044679642,
      "num_tokens": 4164772.0,
      "step": 1895
    },
    {
      "entropy": 6.163094329833984,
      "epoch": 0.1885668916236602,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999098186298986,
      "loss": 6.0398,
      "mean_token_accuracy": 0.11989247053861618,
      "num_tokens": 4175511.0,
      "step": 1900
    },
    {
      "entropy": 6.184041833877563,
      "epoch": 0.1890631202858277,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999088127784667,
      "loss": 6.0138,
      "mean_token_accuracy": 0.1198397345840931,
      "num_tokens": 4187445.0,
      "step": 1905
    },
    {
      "entropy": 6.224808263778686,
      "epoch": 0.18955934894799523,
      "grad_norm": 0.828125,
      "learning_rate": 0.000499907801349781,
      "loss": 5.9992,
      "mean_token_accuracy": 0.1206267774105072,
      "num_tokens": 4198124.0,
      "step": 1910
    },
    {
      "entropy": 6.183636951446533,
      "epoch": 0.19005557761016276,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999067843438664,
      "loss": 6.0454,
      "mean_token_accuracy": 0.11894752085208893,
      "num_tokens": 4208954.0,
      "step": 1915
    },
    {
      "entropy": 6.204626941680909,
      "epoch": 0.1905518062723303,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999057617607482,
      "loss": 6.1093,
      "mean_token_accuracy": 0.1197871819138527,
      "num_tokens": 4218896.0,
      "step": 1920
    },
    {
      "entropy": 6.1618143081665036,
      "epoch": 0.19104803493449782,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999047336004517,
      "loss": 6.0051,
      "mean_token_accuracy": 0.11776421442627907,
      "num_tokens": 4228851.0,
      "step": 1925
    },
    {
      "entropy": 6.110290908813477,
      "epoch": 0.19154426359666535,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999036998630024,
      "loss": 5.9683,
      "mean_token_accuracy": 0.12174378857016563,
      "num_tokens": 4240377.0,
      "step": 1930
    },
    {
      "entropy": 6.116988277435302,
      "epoch": 0.19204049225883288,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000499902660548426,
      "loss": 5.9615,
      "mean_token_accuracy": 0.12193656638264656,
      "num_tokens": 4251341.0,
      "step": 1935
    },
    {
      "entropy": 6.247050094604492,
      "epoch": 0.1925367209210004,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999016156567482,
      "loss": 6.0695,
      "mean_token_accuracy": 0.12280801013112068,
      "num_tokens": 4261206.0,
      "step": 1940
    },
    {
      "entropy": 6.215968275070191,
      "epoch": 0.1930329495831679,
      "grad_norm": 0.8203125,
      "learning_rate": 0.000499900565187995,
      "loss": 6.0677,
      "mean_token_accuracy": 0.1141348622739315,
      "num_tokens": 4272662.0,
      "step": 1945
    },
    {
      "entropy": 6.240064001083374,
      "epoch": 0.19352917824533544,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004998995091421924,
      "loss": 6.0459,
      "mean_token_accuracy": 0.11888503208756447,
      "num_tokens": 4283302.0,
      "step": 1950
    },
    {
      "entropy": 6.114093589782715,
      "epoch": 0.19402540690750297,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998984475193665,
      "loss": 5.9975,
      "mean_token_accuracy": 0.1221638023853302,
      "num_tokens": 4296087.0,
      "step": 1955
    },
    {
      "entropy": 6.181862306594849,
      "epoch": 0.1945216355696705,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004998973803195439,
      "loss": 6.0053,
      "mean_token_accuracy": 0.12597885355353355,
      "num_tokens": 4305918.0,
      "step": 1960
    },
    {
      "entropy": 6.123769617080688,
      "epoch": 0.19501786423183803,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004998963075427508,
      "loss": 6.0646,
      "mean_token_accuracy": 0.12296420186758042,
      "num_tokens": 4316507.0,
      "step": 1965
    },
    {
      "entropy": 6.2180873394012455,
      "epoch": 0.19551409289400556,
      "grad_norm": 0.76953125,
      "learning_rate": 0.0004998952291890138,
      "loss": 6.0,
      "mean_token_accuracy": 0.12596616744995118,
      "num_tokens": 4329091.0,
      "step": 1970
    },
    {
      "entropy": 6.206251239776611,
      "epoch": 0.1960103215561731,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004998941452583599,
      "loss": 6.0603,
      "mean_token_accuracy": 0.12475561425089836,
      "num_tokens": 4339266.0,
      "step": 1975
    },
    {
      "entropy": 6.0974133014678955,
      "epoch": 0.1965065502183406,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998930557508156,
      "loss": 5.984,
      "mean_token_accuracy": 0.1275751493871212,
      "num_tokens": 4349300.0,
      "step": 1980
    },
    {
      "entropy": 6.0830080032348635,
      "epoch": 0.19700277888050813,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004998919606664082,
      "loss": 5.9019,
      "mean_token_accuracy": 0.13283863738179208,
      "num_tokens": 4360314.0,
      "step": 1985
    },
    {
      "entropy": 6.183152389526367,
      "epoch": 0.19749900754267566,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004998908600051648,
      "loss": 5.9678,
      "mean_token_accuracy": 0.1221761591732502,
      "num_tokens": 4370077.0,
      "step": 1990
    },
    {
      "entropy": 6.084688329696656,
      "epoch": 0.1979952362048432,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004998897537671126,
      "loss": 5.9713,
      "mean_token_accuracy": 0.12218144908547401,
      "num_tokens": 4381000.0,
      "step": 1995
    },
    {
      "entropy": 6.211720275878906,
      "epoch": 0.19849146486701072,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998886419522792,
      "loss": 6.1942,
      "mean_token_accuracy": 0.11456988155841827,
      "num_tokens": 4393485.0,
      "step": 2000
    },
    {
      "entropy": 6.298164701461792,
      "epoch": 0.19898769352917825,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004998875245606919,
      "loss": 6.0114,
      "mean_token_accuracy": 0.12225927039980888,
      "num_tokens": 4404354.0,
      "step": 2005
    },
    {
      "entropy": 6.105899810791016,
      "epoch": 0.19948392219134578,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998864015923788,
      "loss": 6.1597,
      "mean_token_accuracy": 0.12076655179262161,
      "num_tokens": 4414818.0,
      "step": 2010
    },
    {
      "entropy": 6.200961780548096,
      "epoch": 0.1999801508535133,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004998852730473674,
      "loss": 6.0344,
      "mean_token_accuracy": 0.11524304151535034,
      "num_tokens": 4426271.0,
      "step": 2015
    },
    {
      "entropy": 6.12263708114624,
      "epoch": 0.2004763795156808,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004998841389256859,
      "loss": 6.0903,
      "mean_token_accuracy": 0.12121652588248252,
      "num_tokens": 4436981.0,
      "step": 2020
    },
    {
      "entropy": 6.139917802810669,
      "epoch": 0.20097260817784834,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998829992273622,
      "loss": 5.9254,
      "mean_token_accuracy": 0.12804987505078316,
      "num_tokens": 4446494.0,
      "step": 2025
    },
    {
      "entropy": 6.094701433181763,
      "epoch": 0.20146883684001587,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004998818539524248,
      "loss": 6.007,
      "mean_token_accuracy": 0.12410036325454712,
      "num_tokens": 4458535.0,
      "step": 2030
    },
    {
      "entropy": 6.231687307357788,
      "epoch": 0.2019650655021834,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000499880703100902,
      "loss": 5.9882,
      "mean_token_accuracy": 0.1257662445306778,
      "num_tokens": 4469025.0,
      "step": 2035
    },
    {
      "entropy": 6.119232892990112,
      "epoch": 0.20246129416435094,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004998795466728223,
      "loss": 5.9832,
      "mean_token_accuracy": 0.1266017124056816,
      "num_tokens": 4478431.0,
      "step": 2040
    },
    {
      "entropy": 6.141779518127441,
      "epoch": 0.20295752282651847,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998783846682144,
      "loss": 6.1012,
      "mean_token_accuracy": 0.12176022753119468,
      "num_tokens": 4490215.0,
      "step": 2045
    },
    {
      "entropy": 6.160814476013184,
      "epoch": 0.203453751488686,
      "grad_norm": 0.76171875,
      "learning_rate": 0.0004998772170871072,
      "loss": 5.9883,
      "mean_token_accuracy": 0.12457352355122567,
      "num_tokens": 4502895.0,
      "step": 2050
    },
    {
      "entropy": 6.102251768112183,
      "epoch": 0.20394998015085353,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004998760439295295,
      "loss": 6.0066,
      "mean_token_accuracy": 0.12112522423267365,
      "num_tokens": 4514468.0,
      "step": 2055
    },
    {
      "entropy": 6.218370962142944,
      "epoch": 0.20444620881302103,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004998748651955105,
      "loss": 6.0276,
      "mean_token_accuracy": 0.11916945278644561,
      "num_tokens": 4525787.0,
      "step": 2060
    },
    {
      "entropy": 6.161030578613281,
      "epoch": 0.20494243747518856,
      "grad_norm": 0.765625,
      "learning_rate": 0.0004998736808850795,
      "loss": 6.1047,
      "mean_token_accuracy": 0.11898912787437439,
      "num_tokens": 4537536.0,
      "step": 2065
    },
    {
      "entropy": 6.2428888320922855,
      "epoch": 0.2054386661373561,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004998724909982656,
      "loss": 6.0472,
      "mean_token_accuracy": 0.12252518236637115,
      "num_tokens": 4549619.0,
      "step": 2070
    },
    {
      "entropy": 6.097667264938354,
      "epoch": 0.20593489479952362,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004998712955350986,
      "loss": 5.9866,
      "mean_token_accuracy": 0.12310111299157142,
      "num_tokens": 4560248.0,
      "step": 2075
    },
    {
      "entropy": 6.195798301696778,
      "epoch": 0.20643112346169115,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004998700944956081,
      "loss": 6.0975,
      "mean_token_accuracy": 0.12141188234090805,
      "num_tokens": 4571888.0,
      "step": 2080
    },
    {
      "entropy": 6.179491949081421,
      "epoch": 0.20692735212385868,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004998688878798237,
      "loss": 5.9365,
      "mean_token_accuracy": 0.1305774599313736,
      "num_tokens": 4583234.0,
      "step": 2085
    },
    {
      "entropy": 6.161084604263306,
      "epoch": 0.2074235807860262,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004998676756877753,
      "loss": 6.1601,
      "mean_token_accuracy": 0.11846253573894501,
      "num_tokens": 4595252.0,
      "step": 2090
    },
    {
      "entropy": 6.131542110443116,
      "epoch": 0.20791980944819372,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004998664579194934,
      "loss": 5.9236,
      "mean_token_accuracy": 0.1237091027200222,
      "num_tokens": 4607091.0,
      "step": 2095
    },
    {
      "entropy": 6.140899753570556,
      "epoch": 0.20841603811036125,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004998652345750077,
      "loss": 5.8958,
      "mean_token_accuracy": 0.1278851717710495,
      "num_tokens": 4618003.0,
      "step": 2100
    },
    {
      "entropy": 6.041016483306885,
      "epoch": 0.20891226677252878,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004998640056543487,
      "loss": 5.8996,
      "mean_token_accuracy": 0.12397338300943375,
      "num_tokens": 4628429.0,
      "step": 2105
    },
    {
      "entropy": 6.15621018409729,
      "epoch": 0.2094084954346963,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998627711575469,
      "loss": 5.9676,
      "mean_token_accuracy": 0.12933185324072838,
      "num_tokens": 4639414.0,
      "step": 2110
    },
    {
      "entropy": 6.167100429534912,
      "epoch": 0.20990472409686384,
      "grad_norm": 0.8046875,
      "learning_rate": 0.000499861531084633,
      "loss": 6.0398,
      "mean_token_accuracy": 0.12240175604820251,
      "num_tokens": 4652594.0,
      "step": 2115
    },
    {
      "entropy": 6.079341125488281,
      "epoch": 0.21040095275903137,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004998602854356376,
      "loss": 5.8855,
      "mean_token_accuracy": 0.1307040624320507,
      "num_tokens": 4662981.0,
      "step": 2120
    },
    {
      "entropy": 6.053981876373291,
      "epoch": 0.2108971814211989,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998590342105917,
      "loss": 5.9951,
      "mean_token_accuracy": 0.12460904493927956,
      "num_tokens": 4672469.0,
      "step": 2125
    },
    {
      "entropy": 6.146594381332397,
      "epoch": 0.21139341008336643,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004998577774095262,
      "loss": 5.9799,
      "mean_token_accuracy": 0.12103921249508857,
      "num_tokens": 4683640.0,
      "step": 2130
    },
    {
      "entropy": 6.100749874114991,
      "epoch": 0.21188963874553393,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004998565150324724,
      "loss": 5.9284,
      "mean_token_accuracy": 0.12689334973692895,
      "num_tokens": 4694032.0,
      "step": 2135
    },
    {
      "entropy": 6.005304336547852,
      "epoch": 0.21238586740770146,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004998552470794616,
      "loss": 5.8162,
      "mean_token_accuracy": 0.12930263206362724,
      "num_tokens": 4704253.0,
      "step": 2140
    },
    {
      "entropy": 6.0581738471984865,
      "epoch": 0.212882096069869,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004998539735505251,
      "loss": 5.9544,
      "mean_token_accuracy": 0.12062570005655289,
      "num_tokens": 4715327.0,
      "step": 2145
    },
    {
      "entropy": 6.2206943988800045,
      "epoch": 0.21337832473203652,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004998526944456946,
      "loss": 5.9851,
      "mean_token_accuracy": 0.12153983116149902,
      "num_tokens": 4726450.0,
      "step": 2150
    },
    {
      "entropy": 6.0526422500610355,
      "epoch": 0.21387455339420405,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004998514097650018,
      "loss": 6.0067,
      "mean_token_accuracy": 0.12227371484041213,
      "num_tokens": 4736630.0,
      "step": 2155
    },
    {
      "entropy": 6.161049175262451,
      "epoch": 0.21437078205637158,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004998501195084786,
      "loss": 5.8963,
      "mean_token_accuracy": 0.12918635457754135,
      "num_tokens": 4746592.0,
      "step": 2160
    },
    {
      "entropy": 6.086433601379395,
      "epoch": 0.21486701071853911,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004998488236761569,
      "loss": 6.072,
      "mean_token_accuracy": 0.12488910630345344,
      "num_tokens": 4757824.0,
      "step": 2165
    },
    {
      "entropy": 6.144498586654663,
      "epoch": 0.21536323938070662,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004998475222680688,
      "loss": 5.9959,
      "mean_token_accuracy": 0.1257177084684372,
      "num_tokens": 4768453.0,
      "step": 2170
    },
    {
      "entropy": 6.051328659057617,
      "epoch": 0.21585946804287415,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004998462152842468,
      "loss": 5.9771,
      "mean_token_accuracy": 0.12266893535852433,
      "num_tokens": 4779664.0,
      "step": 2175
    },
    {
      "entropy": 6.203923177719116,
      "epoch": 0.21635569670504168,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000499844902724723,
      "loss": 5.997,
      "mean_token_accuracy": 0.12400704175233841,
      "num_tokens": 4791012.0,
      "step": 2180
    },
    {
      "entropy": 6.080739498138428,
      "epoch": 0.2168519253672092,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004998435845895302,
      "loss": 5.8727,
      "mean_token_accuracy": 0.1246095396578312,
      "num_tokens": 4800578.0,
      "step": 2185
    },
    {
      "entropy": 6.023191165924072,
      "epoch": 0.21734815402937674,
      "grad_norm": 0.90625,
      "learning_rate": 0.000499842260878701,
      "loss": 5.8847,
      "mean_token_accuracy": 0.12296021655201912,
      "num_tokens": 4810940.0,
      "step": 2190
    },
    {
      "entropy": 6.089705514907837,
      "epoch": 0.21784438269154427,
      "grad_norm": 0.7734375,
      "learning_rate": 0.0004998409315922683,
      "loss": 6.0182,
      "mean_token_accuracy": 0.12229126617312432,
      "num_tokens": 4822696.0,
      "step": 2195
    },
    {
      "entropy": 6.09975118637085,
      "epoch": 0.2183406113537118,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004998395967302649,
      "loss": 5.8893,
      "mean_token_accuracy": 0.12959264144301413,
      "num_tokens": 4833669.0,
      "step": 2200
    },
    {
      "entropy": 6.1851764678955075,
      "epoch": 0.21883684001587933,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000499838256292724,
      "loss": 6.0744,
      "mean_token_accuracy": 0.12145713344216347,
      "num_tokens": 4844021.0,
      "step": 2205
    },
    {
      "entropy": 6.052931642532348,
      "epoch": 0.21933306867804683,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004998369102796787,
      "loss": 5.9505,
      "mean_token_accuracy": 0.12860263735055924,
      "num_tokens": 4855336.0,
      "step": 2210
    },
    {
      "entropy": 6.104745626449585,
      "epoch": 0.21982929734021436,
      "grad_norm": 0.7734375,
      "learning_rate": 0.0004998355586911627,
      "loss": 5.9443,
      "mean_token_accuracy": 0.12385106608271598,
      "num_tokens": 4867713.0,
      "step": 2215
    },
    {
      "entropy": 6.0858691215515135,
      "epoch": 0.2203255260023819,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0004998342015272093,
      "loss": 5.9244,
      "mean_token_accuracy": 0.1223891668021679,
      "num_tokens": 4879134.0,
      "step": 2220
    },
    {
      "entropy": 6.108216285705566,
      "epoch": 0.22082175466454942,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004998328387878521,
      "loss": 6.0808,
      "mean_token_accuracy": 0.12277590483427048,
      "num_tokens": 4891755.0,
      "step": 2225
    },
    {
      "entropy": 6.2470619678497314,
      "epoch": 0.22131798332671695,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004998314704731249,
      "loss": 6.0916,
      "mean_token_accuracy": 0.12046385034918786,
      "num_tokens": 4903264.0,
      "step": 2230
    },
    {
      "entropy": 6.077593660354614,
      "epoch": 0.22181421198888449,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998300965830618,
      "loss": 5.9,
      "mean_token_accuracy": 0.12378231808543205,
      "num_tokens": 4912533.0,
      "step": 2235
    },
    {
      "entropy": 5.9839921474456785,
      "epoch": 0.22231044065105202,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004998287171176968,
      "loss": 5.8996,
      "mean_token_accuracy": 0.1311133660376072,
      "num_tokens": 4923288.0,
      "step": 2240
    },
    {
      "entropy": 6.105050659179687,
      "epoch": 0.22280666931321952,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004998273320770641,
      "loss": 5.9154,
      "mean_token_accuracy": 0.12592516243457794,
      "num_tokens": 4934759.0,
      "step": 2245
    },
    {
      "entropy": 6.072722053527832,
      "epoch": 0.22330289797538705,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499825941461198,
      "loss": 5.9621,
      "mean_token_accuracy": 0.12398277670145035,
      "num_tokens": 4947024.0,
      "step": 2250
    },
    {
      "entropy": 6.092696714401245,
      "epoch": 0.22379912663755458,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004998245452701329,
      "loss": 5.9152,
      "mean_token_accuracy": 0.12754956483840943,
      "num_tokens": 4959281.0,
      "step": 2255
    },
    {
      "entropy": 6.13616452217102,
      "epoch": 0.2242953552997221,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998231435039036,
      "loss": 5.9725,
      "mean_token_accuracy": 0.12141189724206924,
      "num_tokens": 4970466.0,
      "step": 2260
    },
    {
      "entropy": 6.069933271408081,
      "epoch": 0.22479158396188964,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004998217361625449,
      "loss": 5.9236,
      "mean_token_accuracy": 0.12711238041520118,
      "num_tokens": 4981135.0,
      "step": 2265
    },
    {
      "entropy": 6.105356216430664,
      "epoch": 0.22528781262405717,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004998203232460915,
      "loss": 5.9797,
      "mean_token_accuracy": 0.12980691865086555,
      "num_tokens": 4991292.0,
      "step": 2270
    },
    {
      "entropy": 6.059658765792847,
      "epoch": 0.2257840412862247,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004998189047545785,
      "loss": 5.9345,
      "mean_token_accuracy": 0.13011907190084457,
      "num_tokens": 5002806.0,
      "step": 2275
    },
    {
      "entropy": 6.100206136703491,
      "epoch": 0.22628026994839223,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004998174806880411,
      "loss": 5.8843,
      "mean_token_accuracy": 0.1296177588403225,
      "num_tokens": 5013474.0,
      "step": 2280
    },
    {
      "entropy": 6.06908540725708,
      "epoch": 0.22677649861055973,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004998160510465146,
      "loss": 5.9249,
      "mean_token_accuracy": 0.1318567730486393,
      "num_tokens": 5025151.0,
      "step": 2285
    },
    {
      "entropy": 5.967230415344238,
      "epoch": 0.22727272727272727,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004998146158300345,
      "loss": 5.9196,
      "mean_token_accuracy": 0.1306504562497139,
      "num_tokens": 5037166.0,
      "step": 2290
    },
    {
      "entropy": 6.142732429504394,
      "epoch": 0.2277689559348948,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004998131750386364,
      "loss": 5.9706,
      "mean_token_accuracy": 0.12619220688939095,
      "num_tokens": 5047452.0,
      "step": 2295
    },
    {
      "entropy": 6.048137044906616,
      "epoch": 0.22826518459706233,
      "grad_norm": 0.875,
      "learning_rate": 0.0004998117286723559,
      "loss": 5.9756,
      "mean_token_accuracy": 0.12324338406324387,
      "num_tokens": 5058501.0,
      "step": 2300
    },
    {
      "entropy": 6.132171249389648,
      "epoch": 0.22876141325922986,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000499810276731229,
      "loss": 5.974,
      "mean_token_accuracy": 0.12393502593040466,
      "num_tokens": 5069716.0,
      "step": 2305
    },
    {
      "entropy": 6.158532667160034,
      "epoch": 0.2292576419213974,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998088192152918,
      "loss": 5.9587,
      "mean_token_accuracy": 0.12171528786420822,
      "num_tokens": 5078509.0,
      "step": 2310
    },
    {
      "entropy": 6.056684732437134,
      "epoch": 0.22975387058356492,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004998073561245801,
      "loss": 5.9908,
      "mean_token_accuracy": 0.12283852398395538,
      "num_tokens": 5090355.0,
      "step": 2315
    },
    {
      "entropy": 6.118961763381958,
      "epoch": 0.23025009924573242,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004998058874591305,
      "loss": 5.8839,
      "mean_token_accuracy": 0.12737965807318688,
      "num_tokens": 5100582.0,
      "step": 2320
    },
    {
      "entropy": 6.098464822769165,
      "epoch": 0.23074632790789995,
      "grad_norm": 0.76953125,
      "learning_rate": 0.0004998044132189793,
      "loss": 5.9514,
      "mean_token_accuracy": 0.12292122989892959,
      "num_tokens": 5112176.0,
      "step": 2325
    },
    {
      "entropy": 6.076971912384034,
      "epoch": 0.23124255657006748,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499802933404163,
      "loss": 5.8787,
      "mean_token_accuracy": 0.1285627380013466,
      "num_tokens": 5122840.0,
      "step": 2330
    },
    {
      "entropy": 6.050347518920899,
      "epoch": 0.231738785232235,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004998014480147183,
      "loss": 5.8587,
      "mean_token_accuracy": 0.13070411905646323,
      "num_tokens": 5132926.0,
      "step": 2335
    },
    {
      "entropy": 6.0548614978790285,
      "epoch": 0.23223501389440254,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004997999570506823,
      "loss": 5.8476,
      "mean_token_accuracy": 0.12983267456293107,
      "num_tokens": 5143394.0,
      "step": 2340
    },
    {
      "entropy": 5.978992509841919,
      "epoch": 0.23273124255657007,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004997984605120917,
      "loss": 5.8757,
      "mean_token_accuracy": 0.12346786707639694,
      "num_tokens": 5155710.0,
      "step": 2345
    },
    {
      "entropy": 5.942594051361084,
      "epoch": 0.2332274712187376,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004997969583989835,
      "loss": 5.7929,
      "mean_token_accuracy": 0.12445841357111931,
      "num_tokens": 5167065.0,
      "step": 2350
    },
    {
      "entropy": 6.06046576499939,
      "epoch": 0.23372369988090513,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004997954507113952,
      "loss": 5.9018,
      "mean_token_accuracy": 0.12778859511017798,
      "num_tokens": 5177535.0,
      "step": 2355
    },
    {
      "entropy": 6.021834182739258,
      "epoch": 0.23421992854307264,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004997939374493642,
      "loss": 5.8305,
      "mean_token_accuracy": 0.12911173850297927,
      "num_tokens": 5188324.0,
      "step": 2360
    },
    {
      "entropy": 6.018092298507691,
      "epoch": 0.23471615720524017,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000499792418612928,
      "loss": 5.9024,
      "mean_token_accuracy": 0.13002333790063858,
      "num_tokens": 5199223.0,
      "step": 2365
    },
    {
      "entropy": 5.9740711688995365,
      "epoch": 0.2352123858674077,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004997908942021241,
      "loss": 5.89,
      "mean_token_accuracy": 0.13267461881041526,
      "num_tokens": 5210080.0,
      "step": 2370
    },
    {
      "entropy": 6.135750961303711,
      "epoch": 0.23570861452957523,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004997893642169904,
      "loss": 6.0108,
      "mean_token_accuracy": 0.12664430662989618,
      "num_tokens": 5220867.0,
      "step": 2375
    },
    {
      "entropy": 6.101852035522461,
      "epoch": 0.23620484319174276,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997878286575648,
      "loss": 5.8597,
      "mean_token_accuracy": 0.12812694609165193,
      "num_tokens": 5231451.0,
      "step": 2380
    },
    {
      "entropy": 6.028659343719482,
      "epoch": 0.2367010718539103,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004997862875238855,
      "loss": 5.9642,
      "mean_token_accuracy": 0.12428494766354561,
      "num_tokens": 5244643.0,
      "step": 2385
    },
    {
      "entropy": 6.113288068771363,
      "epoch": 0.23719730051607782,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004997847408159907,
      "loss": 5.979,
      "mean_token_accuracy": 0.12018007338047028,
      "num_tokens": 5256279.0,
      "step": 2390
    },
    {
      "entropy": 6.0652384757995605,
      "epoch": 0.23769352917824532,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004997831885339186,
      "loss": 5.9475,
      "mean_token_accuracy": 0.1281052775681019,
      "num_tokens": 5268498.0,
      "step": 2395
    },
    {
      "entropy": 6.014727401733398,
      "epoch": 0.23818975784041285,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004997816306777077,
      "loss": 5.908,
      "mean_token_accuracy": 0.1289649024605751,
      "num_tokens": 5280869.0,
      "step": 2400
    },
    {
      "entropy": 6.235253620147705,
      "epoch": 0.23868598650258038,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004997800672473969,
      "loss": 6.0058,
      "mean_token_accuracy": 0.12169916182756424,
      "num_tokens": 5293082.0,
      "step": 2405
    },
    {
      "entropy": 5.997746467590332,
      "epoch": 0.2391822151647479,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997784982430247,
      "loss": 5.8106,
      "mean_token_accuracy": 0.13437482044100763,
      "num_tokens": 5304039.0,
      "step": 2410
    },
    {
      "entropy": 5.944843721389771,
      "epoch": 0.23967844382691544,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004997769236646301,
      "loss": 5.8851,
      "mean_token_accuracy": 0.1293253131210804,
      "num_tokens": 5315016.0,
      "step": 2415
    },
    {
      "entropy": 6.173792791366577,
      "epoch": 0.24017467248908297,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004997753435122522,
      "loss": 6.0082,
      "mean_token_accuracy": 0.12303022742271423,
      "num_tokens": 5326090.0,
      "step": 2420
    },
    {
      "entropy": 6.016424989700317,
      "epoch": 0.2406709011512505,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00049977375778593,
      "loss": 5.7936,
      "mean_token_accuracy": 0.1281319923698902,
      "num_tokens": 5336423.0,
      "step": 2425
    },
    {
      "entropy": 6.052116632461548,
      "epoch": 0.24116712981341804,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004997721664857032,
      "loss": 5.9728,
      "mean_token_accuracy": 0.12626754492521286,
      "num_tokens": 5346982.0,
      "step": 2430
    },
    {
      "entropy": 5.9658690929412845,
      "epoch": 0.24166335847558554,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004997705696116108,
      "loss": 5.8618,
      "mean_token_accuracy": 0.1352499283850193,
      "num_tokens": 5359210.0,
      "step": 2435
    },
    {
      "entropy": 6.060595321655273,
      "epoch": 0.24215958713775307,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004997689671636926,
      "loss": 5.8833,
      "mean_token_accuracy": 0.1232607200741768,
      "num_tokens": 5369891.0,
      "step": 2440
    },
    {
      "entropy": 6.024409484863281,
      "epoch": 0.2426558157999206,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004997673591419884,
      "loss": 5.8619,
      "mean_token_accuracy": 0.1305900402367115,
      "num_tokens": 5380850.0,
      "step": 2445
    },
    {
      "entropy": 6.061957502365113,
      "epoch": 0.24315204446208813,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0004997657455465381,
      "loss": 5.928,
      "mean_token_accuracy": 0.13309514969587327,
      "num_tokens": 5391903.0,
      "step": 2450
    },
    {
      "entropy": 6.0940954208374025,
      "epoch": 0.24364827312425566,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004997641263773815,
      "loss": 5.978,
      "mean_token_accuracy": 0.12671018466353418,
      "num_tokens": 5402517.0,
      "step": 2455
    },
    {
      "entropy": 5.941332578659058,
      "epoch": 0.2441445017864232,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000499762501634559,
      "loss": 5.8878,
      "mean_token_accuracy": 0.12774158492684365,
      "num_tokens": 5414110.0,
      "step": 2460
    },
    {
      "entropy": 6.125403642654419,
      "epoch": 0.24464073044859072,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004997608713181108,
      "loss": 5.9006,
      "mean_token_accuracy": 0.12905448973178862,
      "num_tokens": 5424466.0,
      "step": 2465
    },
    {
      "entropy": 6.040532350540161,
      "epoch": 0.24513695911075822,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004997592354280772,
      "loss": 5.9095,
      "mean_token_accuracy": 0.134385783970356,
      "num_tokens": 5434461.0,
      "step": 2470
    },
    {
      "entropy": 5.967836189270019,
      "epoch": 0.24563318777292575,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000499757593964499,
      "loss": 5.8913,
      "mean_token_accuracy": 0.11896291747689247,
      "num_tokens": 5445921.0,
      "step": 2475
    },
    {
      "entropy": 6.038508701324463,
      "epoch": 0.24612941643509328,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997559469274166,
      "loss": 5.8771,
      "mean_token_accuracy": 0.13031587675213813,
      "num_tokens": 5456283.0,
      "step": 2480
    },
    {
      "entropy": 6.024613332748413,
      "epoch": 0.24662564509726081,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004997542943168712,
      "loss": 5.9114,
      "mean_token_accuracy": 0.1300675705075264,
      "num_tokens": 5467616.0,
      "step": 2485
    },
    {
      "entropy": 6.012180805206299,
      "epoch": 0.24712187375942835,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997526361329035,
      "loss": 5.8744,
      "mean_token_accuracy": 0.13522633239626886,
      "num_tokens": 5478050.0,
      "step": 2490
    },
    {
      "entropy": 6.102998495101929,
      "epoch": 0.24761810242159588,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997509723755547,
      "loss": 5.8714,
      "mean_token_accuracy": 0.12875011935830116,
      "num_tokens": 5487185.0,
      "step": 2495
    },
    {
      "entropy": 5.998979711532593,
      "epoch": 0.2481143310837634,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004997493030448662,
      "loss": 5.8924,
      "mean_token_accuracy": 0.12383673936128617,
      "num_tokens": 5498802.0,
      "step": 2500
    },
    {
      "entropy": 6.0036334037780765,
      "epoch": 0.24861055974593094,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004997476281408793,
      "loss": 5.9009,
      "mean_token_accuracy": 0.12360760122537613,
      "num_tokens": 5510285.0,
      "step": 2505
    },
    {
      "entropy": 6.045953416824341,
      "epoch": 0.24910678840809844,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004997459476636354,
      "loss": 5.7868,
      "mean_token_accuracy": 0.13784014880657197,
      "num_tokens": 5520028.0,
      "step": 2510
    },
    {
      "entropy": 5.9735174655914305,
      "epoch": 0.24960301707026597,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997442616131763,
      "loss": 5.912,
      "mean_token_accuracy": 0.12859713956713675,
      "num_tokens": 5529271.0,
      "step": 2515
    },
    {
      "entropy": 6.054106569290161,
      "epoch": 0.25009924573243353,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004997425699895438,
      "loss": 5.826,
      "mean_token_accuracy": 0.12208940386772156,
      "num_tokens": 5539396.0,
      "step": 2520
    },
    {
      "entropy": 6.049839782714844,
      "epoch": 0.25059547439460106,
      "grad_norm": 0.76953125,
      "learning_rate": 0.0004997408727927798,
      "loss": 5.9723,
      "mean_token_accuracy": 0.1294700786471367,
      "num_tokens": 5551674.0,
      "step": 2525
    },
    {
      "entropy": 6.055530405044555,
      "epoch": 0.25109170305676853,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004997391700229264,
      "loss": 5.8948,
      "mean_token_accuracy": 0.1298077531158924,
      "num_tokens": 5562749.0,
      "step": 2530
    },
    {
      "entropy": 6.030540132522583,
      "epoch": 0.25158793171893606,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000499737461680026,
      "loss": 5.9,
      "mean_token_accuracy": 0.12933559864759445,
      "num_tokens": 5573478.0,
      "step": 2535
    },
    {
      "entropy": 5.963146209716797,
      "epoch": 0.2520841603811036,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004997357477641207,
      "loss": 5.8908,
      "mean_token_accuracy": 0.13493381813168526,
      "num_tokens": 5584546.0,
      "step": 2540
    },
    {
      "entropy": 5.964659547805786,
      "epoch": 0.2525803890432711,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004997340282752532,
      "loss": 5.7847,
      "mean_token_accuracy": 0.1342097096145153,
      "num_tokens": 5594004.0,
      "step": 2545
    },
    {
      "entropy": 6.085031366348266,
      "epoch": 0.25307661770543866,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000499732303213466,
      "loss": 5.8353,
      "mean_token_accuracy": 0.12420367822051048,
      "num_tokens": 5606095.0,
      "step": 2550
    },
    {
      "entropy": 6.014924669265747,
      "epoch": 0.2535728463676062,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997305725788019,
      "loss": 5.8899,
      "mean_token_accuracy": 0.12461917325854302,
      "num_tokens": 5615596.0,
      "step": 2555
    },
    {
      "entropy": 5.9634756565094,
      "epoch": 0.2540690750297737,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000499728836371304,
      "loss": 5.8139,
      "mean_token_accuracy": 0.12745841071009636,
      "num_tokens": 5626753.0,
      "step": 2560
    },
    {
      "entropy": 5.963172578811646,
      "epoch": 0.25456530369194125,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004997270945910151,
      "loss": 5.8388,
      "mean_token_accuracy": 0.12756719589233398,
      "num_tokens": 5636135.0,
      "step": 2565
    },
    {
      "entropy": 6.079827070236206,
      "epoch": 0.2550615323541088,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004997253472379785,
      "loss": 5.9086,
      "mean_token_accuracy": 0.12276982814073563,
      "num_tokens": 5647074.0,
      "step": 2570
    },
    {
      "entropy": 6.04095311164856,
      "epoch": 0.2555577610162763,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0004997235943122375,
      "loss": 5.9298,
      "mean_token_accuracy": 0.12193780019879341,
      "num_tokens": 5658600.0,
      "step": 2575
    },
    {
      "entropy": 5.954120779037476,
      "epoch": 0.25605398967844384,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004997218358138356,
      "loss": 5.8184,
      "mean_token_accuracy": 0.12850069105625153,
      "num_tokens": 5668453.0,
      "step": 2580
    },
    {
      "entropy": 6.052977800369263,
      "epoch": 0.25655021834061137,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997200717428165,
      "loss": 5.8802,
      "mean_token_accuracy": 0.12774283811450005,
      "num_tokens": 5678855.0,
      "step": 2585
    },
    {
      "entropy": 6.0446836948394775,
      "epoch": 0.2570464470027789,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004997183020992237,
      "loss": 5.9438,
      "mean_token_accuracy": 0.12405224218964576,
      "num_tokens": 5690063.0,
      "step": 2590
    },
    {
      "entropy": 5.979913902282715,
      "epoch": 0.25754267566494643,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004997165268831014,
      "loss": 5.843,
      "mean_token_accuracy": 0.1319989301264286,
      "num_tokens": 5701487.0,
      "step": 2595
    },
    {
      "entropy": 5.977937984466553,
      "epoch": 0.25803890432711396,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004997147460944933,
      "loss": 5.8313,
      "mean_token_accuracy": 0.131146914511919,
      "num_tokens": 5713034.0,
      "step": 2600
    },
    {
      "entropy": 6.07123966217041,
      "epoch": 0.25853513298928144,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004997129597334438,
      "loss": 5.8811,
      "mean_token_accuracy": 0.13053280711174012,
      "num_tokens": 5725469.0,
      "step": 2605
    },
    {
      "entropy": 5.960973072052002,
      "epoch": 0.25903136165144897,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0004997111677999971,
      "loss": 5.87,
      "mean_token_accuracy": 0.12475038319826126,
      "num_tokens": 5737099.0,
      "step": 2610
    },
    {
      "entropy": 6.032576656341552,
      "epoch": 0.2595275903136165,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004997093702941976,
      "loss": 5.872,
      "mean_token_accuracy": 0.1248766042292118,
      "num_tokens": 5748756.0,
      "step": 2615
    },
    {
      "entropy": 6.0417296409606935,
      "epoch": 0.260023818975784,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00049970756721609,
      "loss": 5.8727,
      "mean_token_accuracy": 0.12974709644913673,
      "num_tokens": 5759598.0,
      "step": 2620
    },
    {
      "entropy": 6.012347316741943,
      "epoch": 0.26052004763795156,
      "grad_norm": 0.796875,
      "learning_rate": 0.000499705758565719,
      "loss": 5.9157,
      "mean_token_accuracy": 0.12809571251273155,
      "num_tokens": 5770397.0,
      "step": 2625
    },
    {
      "entropy": 6.098965167999268,
      "epoch": 0.2610162763001191,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004997039443431292,
      "loss": 5.983,
      "mean_token_accuracy": 0.11926335170865059,
      "num_tokens": 5780935.0,
      "step": 2630
    },
    {
      "entropy": 6.08192253112793,
      "epoch": 0.2615125049622866,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997021245483659,
      "loss": 5.8366,
      "mean_token_accuracy": 0.12855549082159995,
      "num_tokens": 5791557.0,
      "step": 2635
    },
    {
      "entropy": 5.944977807998657,
      "epoch": 0.26200873362445415,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000499700299181474,
      "loss": 5.874,
      "mean_token_accuracy": 0.12412384524941444,
      "num_tokens": 5802233.0,
      "step": 2640
    },
    {
      "entropy": 5.984538125991821,
      "epoch": 0.2625049622866217,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004996984682424989,
      "loss": 5.7929,
      "mean_token_accuracy": 0.13563756346702577,
      "num_tokens": 5813424.0,
      "step": 2645
    },
    {
      "entropy": 5.963291454315185,
      "epoch": 0.2630011909487892,
      "grad_norm": 0.74609375,
      "learning_rate": 0.000499696631731486,
      "loss": 5.7802,
      "mean_token_accuracy": 0.13106578662991525,
      "num_tokens": 5825652.0,
      "step": 2650
    },
    {
      "entropy": 6.034928655624389,
      "epoch": 0.26349741961095674,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004996947896484809,
      "loss": 5.865,
      "mean_token_accuracy": 0.13201481997966766,
      "num_tokens": 5836603.0,
      "step": 2655
    },
    {
      "entropy": 5.982569742202759,
      "epoch": 0.26399364827312427,
      "grad_norm": 0.890625,
      "learning_rate": 0.000499692941993529,
      "loss": 5.9128,
      "mean_token_accuracy": 0.1318881943821907,
      "num_tokens": 5849365.0,
      "step": 2660
    },
    {
      "entropy": 6.044679498672485,
      "epoch": 0.2644898769352918,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004996910887666765,
      "loss": 5.8354,
      "mean_token_accuracy": 0.1336039572954178,
      "num_tokens": 5860305.0,
      "step": 2665
    },
    {
      "entropy": 5.926130056381226,
      "epoch": 0.26498610559745933,
      "grad_norm": 0.828125,
      "learning_rate": 0.000499689229967969,
      "loss": 5.7666,
      "mean_token_accuracy": 0.13250631913542749,
      "num_tokens": 5872272.0,
      "step": 2670
    },
    {
      "entropy": 5.9937135696411135,
      "epoch": 0.26548233425962686,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996873655974529,
      "loss": 5.8152,
      "mean_token_accuracy": 0.12750425711274146,
      "num_tokens": 5884107.0,
      "step": 2675
    },
    {
      "entropy": 5.906749153137207,
      "epoch": 0.26597856292179434,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004996854956551743,
      "loss": 5.8092,
      "mean_token_accuracy": 0.13456383347511292,
      "num_tokens": 5894060.0,
      "step": 2680
    },
    {
      "entropy": 5.9789345264434814,
      "epoch": 0.26647479158396187,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004996836201411794,
      "loss": 5.8238,
      "mean_token_accuracy": 0.1331551544368267,
      "num_tokens": 5904804.0,
      "step": 2685
    },
    {
      "entropy": 6.056610536575318,
      "epoch": 0.2669710202461294,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499681739055515,
      "loss": 5.8861,
      "mean_token_accuracy": 0.1274963565170765,
      "num_tokens": 5916496.0,
      "step": 2690
    },
    {
      "entropy": 5.9887939453125,
      "epoch": 0.26746724890829693,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004996798523982276,
      "loss": 5.831,
      "mean_token_accuracy": 0.1256636656820774,
      "num_tokens": 5926920.0,
      "step": 2695
    },
    {
      "entropy": 6.095494651794434,
      "epoch": 0.26796347757046446,
      "grad_norm": 0.875,
      "learning_rate": 0.000499677960169364,
      "loss": 5.935,
      "mean_token_accuracy": 0.12686496153473853,
      "num_tokens": 5940098.0,
      "step": 2700
    },
    {
      "entropy": 5.972264385223388,
      "epoch": 0.268459706232632,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004996760623689711,
      "loss": 5.7841,
      "mean_token_accuracy": 0.13265340775251389,
      "num_tokens": 5949687.0,
      "step": 2705
    },
    {
      "entropy": 5.992936277389527,
      "epoch": 0.2689559348947995,
      "grad_norm": 0.8125,
      "learning_rate": 0.000499674158997096,
      "loss": 5.9045,
      "mean_token_accuracy": 0.1267220564186573,
      "num_tokens": 5960777.0,
      "step": 2710
    },
    {
      "entropy": 6.038538122177124,
      "epoch": 0.26945216355696705,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004996722500537859,
      "loss": 5.8095,
      "mean_token_accuracy": 0.12924240455031394,
      "num_tokens": 5972298.0,
      "step": 2715
    },
    {
      "entropy": 5.961693143844604,
      "epoch": 0.2699483922191346,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004996703355390881,
      "loss": 5.9214,
      "mean_token_accuracy": 0.12794365584850312,
      "num_tokens": 5985613.0,
      "step": 2720
    },
    {
      "entropy": 6.043231201171875,
      "epoch": 0.2704446208813021,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00049966841545305,
      "loss": 5.9656,
      "mean_token_accuracy": 0.1257025733590126,
      "num_tokens": 5997530.0,
      "step": 2725
    },
    {
      "entropy": 6.031536388397217,
      "epoch": 0.27094084954346964,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004996664897957195,
      "loss": 5.8365,
      "mean_token_accuracy": 0.12756068184971808,
      "num_tokens": 6008123.0,
      "step": 2730
    },
    {
      "entropy": 6.018695735931397,
      "epoch": 0.2714370782056372,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004996645585671441,
      "loss": 5.852,
      "mean_token_accuracy": 0.13391014859080314,
      "num_tokens": 6020578.0,
      "step": 2735
    },
    {
      "entropy": 5.907762765884399,
      "epoch": 0.2719333068678047,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004996626217673717,
      "loss": 5.7793,
      "mean_token_accuracy": 0.1337601587176323,
      "num_tokens": 6031625.0,
      "step": 2740
    },
    {
      "entropy": 5.958955860137939,
      "epoch": 0.27242953552997223,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004996606793964504,
      "loss": 5.8286,
      "mean_token_accuracy": 0.13649252206087112,
      "num_tokens": 6041320.0,
      "step": 2745
    },
    {
      "entropy": 6.01719069480896,
      "epoch": 0.27292576419213976,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004996587314544284,
      "loss": 5.8369,
      "mean_token_accuracy": 0.12664347812533377,
      "num_tokens": 6052627.0,
      "step": 2750
    },
    {
      "entropy": 5.954709434509278,
      "epoch": 0.27342199285430724,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004996567779413539,
      "loss": 5.8036,
      "mean_token_accuracy": 0.13628075420856475,
      "num_tokens": 6063468.0,
      "step": 2755
    },
    {
      "entropy": 5.991281652450562,
      "epoch": 0.27391822151647477,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004996548188572755,
      "loss": 5.7814,
      "mean_token_accuracy": 0.1363678827881813,
      "num_tokens": 6074208.0,
      "step": 2760
    },
    {
      "entropy": 5.994483947753906,
      "epoch": 0.2744144501786423,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004996528542022416,
      "loss": 5.8366,
      "mean_token_accuracy": 0.12587395310401917,
      "num_tokens": 6085068.0,
      "step": 2765
    },
    {
      "entropy": 5.959670734405518,
      "epoch": 0.27491067884080983,
      "grad_norm": 0.8046875,
      "learning_rate": 0.000499650883976301,
      "loss": 5.844,
      "mean_token_accuracy": 0.12782294154167176,
      "num_tokens": 6096445.0,
      "step": 2770
    },
    {
      "entropy": 5.91315975189209,
      "epoch": 0.27540690750297736,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004996489081795026,
      "loss": 5.7401,
      "mean_token_accuracy": 0.13289162740111352,
      "num_tokens": 6106527.0,
      "step": 2775
    },
    {
      "entropy": 5.91305251121521,
      "epoch": 0.2759031361651449,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996469268118953,
      "loss": 5.7589,
      "mean_token_accuracy": 0.13679778575897217,
      "num_tokens": 6115923.0,
      "step": 2780
    },
    {
      "entropy": 5.868895196914673,
      "epoch": 0.2763993648273124,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004996449398735283,
      "loss": 5.6526,
      "mean_token_accuracy": 0.13616614267230034,
      "num_tokens": 6125861.0,
      "step": 2785
    },
    {
      "entropy": 6.010179805755615,
      "epoch": 0.27689559348947995,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004996429473644509,
      "loss": 5.8537,
      "mean_token_accuracy": 0.13069555088877677,
      "num_tokens": 6137189.0,
      "step": 2790
    },
    {
      "entropy": 5.97644534111023,
      "epoch": 0.2773918221516475,
      "grad_norm": 0.77734375,
      "learning_rate": 0.0004996409492847124,
      "loss": 5.7804,
      "mean_token_accuracy": 0.13484650626778602,
      "num_tokens": 6147920.0,
      "step": 2795
    },
    {
      "entropy": 5.939912557601929,
      "epoch": 0.277888050813815,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004996389456343624,
      "loss": 5.7712,
      "mean_token_accuracy": 0.13595055788755417,
      "num_tokens": 6159371.0,
      "step": 2800
    },
    {
      "entropy": 6.066695976257324,
      "epoch": 0.27838427947598254,
      "grad_norm": 0.75,
      "learning_rate": 0.0004996369364134507,
      "loss": 5.9699,
      "mean_token_accuracy": 0.12258584052324295,
      "num_tokens": 6172679.0,
      "step": 2805
    },
    {
      "entropy": 5.927506589889527,
      "epoch": 0.2788805081381501,
      "grad_norm": 0.8359375,
      "learning_rate": 0.000499634921622027,
      "loss": 5.8298,
      "mean_token_accuracy": 0.12628915533423424,
      "num_tokens": 6184244.0,
      "step": 2810
    },
    {
      "entropy": 5.933115148544312,
      "epoch": 0.2793767368003176,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004996329012601413,
      "loss": 5.706,
      "mean_token_accuracy": 0.1285858377814293,
      "num_tokens": 6194826.0,
      "step": 2815
    },
    {
      "entropy": 5.979615592956543,
      "epoch": 0.27987296546248513,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004996308753278437,
      "loss": 5.8456,
      "mean_token_accuracy": 0.13074849992990495,
      "num_tokens": 6206808.0,
      "step": 2820
    },
    {
      "entropy": 6.013004970550537,
      "epoch": 0.28036919412465267,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004996288438251843,
      "loss": 5.8605,
      "mean_token_accuracy": 0.12552075758576392,
      "num_tokens": 6217303.0,
      "step": 2825
    },
    {
      "entropy": 5.992705678939819,
      "epoch": 0.28086542278682014,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004996268067522137,
      "loss": 5.8603,
      "mean_token_accuracy": 0.12485205754637718,
      "num_tokens": 6227965.0,
      "step": 2830
    },
    {
      "entropy": 6.082558679580688,
      "epoch": 0.28136165144898767,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996247641089823,
      "loss": 5.919,
      "mean_token_accuracy": 0.12674327939748764,
      "num_tokens": 6239011.0,
      "step": 2835
    },
    {
      "entropy": 5.880522966384888,
      "epoch": 0.2818578801111552,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004996227158955409,
      "loss": 5.7612,
      "mean_token_accuracy": 0.13352836593985556,
      "num_tokens": 6249084.0,
      "step": 2840
    },
    {
      "entropy": 5.958843946456909,
      "epoch": 0.28235410877332273,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00049962066211194,
      "loss": 5.8334,
      "mean_token_accuracy": 0.12839528918266296,
      "num_tokens": 6260706.0,
      "step": 2845
    },
    {
      "entropy": 5.931930923461914,
      "epoch": 0.28285033743549026,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004996186027582308,
      "loss": 5.7522,
      "mean_token_accuracy": 0.13635355457663537,
      "num_tokens": 6272848.0,
      "step": 2850
    },
    {
      "entropy": 5.946898508071899,
      "epoch": 0.2833465660976578,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004996165378344642,
      "loss": 5.7663,
      "mean_token_accuracy": 0.13342537730932236,
      "num_tokens": 6283820.0,
      "step": 2855
    },
    {
      "entropy": 5.972999238967896,
      "epoch": 0.2838427947598253,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004996144673406914,
      "loss": 5.9339,
      "mean_token_accuracy": 0.12071966528892517,
      "num_tokens": 6294193.0,
      "step": 2860
    },
    {
      "entropy": 5.960216379165649,
      "epoch": 0.28433902342199285,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004996123912769639,
      "loss": 5.7676,
      "mean_token_accuracy": 0.1383022852241993,
      "num_tokens": 6304407.0,
      "step": 2865
    },
    {
      "entropy": 5.893539810180664,
      "epoch": 0.2848352520841604,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004996103096433332,
      "loss": 5.8421,
      "mean_token_accuracy": 0.12886647060513495,
      "num_tokens": 6316311.0,
      "step": 2870
    },
    {
      "entropy": 6.00091586112976,
      "epoch": 0.2853314807463279,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004996082224398506,
      "loss": 5.8144,
      "mean_token_accuracy": 0.13291785418987273,
      "num_tokens": 6327248.0,
      "step": 2875
    },
    {
      "entropy": 5.986534309387207,
      "epoch": 0.28582770940849545,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004996061296665682,
      "loss": 5.8325,
      "mean_token_accuracy": 0.12824170365929605,
      "num_tokens": 6338494.0,
      "step": 2880
    },
    {
      "entropy": 5.9684656143188475,
      "epoch": 0.286323938070663,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004996040313235377,
      "loss": 5.7233,
      "mean_token_accuracy": 0.13372908607125283,
      "num_tokens": 6349038.0,
      "step": 2885
    },
    {
      "entropy": 5.902253913879394,
      "epoch": 0.2868201667328305,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004996019274108112,
      "loss": 5.8233,
      "mean_token_accuracy": 0.13003271594643592,
      "num_tokens": 6360180.0,
      "step": 2890
    },
    {
      "entropy": 5.906910562515259,
      "epoch": 0.28731639539499804,
      "grad_norm": 0.875,
      "learning_rate": 0.000499599817928441,
      "loss": 5.7596,
      "mean_token_accuracy": 0.13310305699706076,
      "num_tokens": 6370931.0,
      "step": 2895
    },
    {
      "entropy": 5.877241802215576,
      "epoch": 0.28781262405716557,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004995977028764792,
      "loss": 5.7375,
      "mean_token_accuracy": 0.13817125037312508,
      "num_tokens": 6381810.0,
      "step": 2900
    },
    {
      "entropy": 5.958535528182983,
      "epoch": 0.28830885271933304,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004995955822549783,
      "loss": 5.7447,
      "mean_token_accuracy": 0.1337203674018383,
      "num_tokens": 6392268.0,
      "step": 2905
    },
    {
      "entropy": 5.885456800460815,
      "epoch": 0.2888050813815006,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004995934560639909,
      "loss": 5.8172,
      "mean_token_accuracy": 0.13473332896828652,
      "num_tokens": 6403650.0,
      "step": 2910
    },
    {
      "entropy": 5.995740032196045,
      "epoch": 0.2893013100436681,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004995913243035699,
      "loss": 5.7959,
      "mean_token_accuracy": 0.12757119312882423,
      "num_tokens": 6414672.0,
      "step": 2915
    },
    {
      "entropy": 5.970022487640381,
      "epoch": 0.28979753870583563,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004995891869737678,
      "loss": 5.8593,
      "mean_token_accuracy": 0.12999678775668144,
      "num_tokens": 6424916.0,
      "step": 2920
    },
    {
      "entropy": 5.960818958282471,
      "epoch": 0.29029376736800316,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004995870440746379,
      "loss": 5.8059,
      "mean_token_accuracy": 0.13466615676879884,
      "num_tokens": 6435950.0,
      "step": 2925
    },
    {
      "entropy": 5.905315399169922,
      "epoch": 0.2907899960301707,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004995848956062333,
      "loss": 5.7742,
      "mean_token_accuracy": 0.1341630332171917,
      "num_tokens": 6446336.0,
      "step": 2930
    },
    {
      "entropy": 5.917783546447754,
      "epoch": 0.2912862246923382,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004995827415686071,
      "loss": 5.8452,
      "mean_token_accuracy": 0.13145026564598083,
      "num_tokens": 6457655.0,
      "step": 2935
    },
    {
      "entropy": 5.985751056671143,
      "epoch": 0.29178245335450576,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000499580581961813,
      "loss": 5.7235,
      "mean_token_accuracy": 0.13669102862477303,
      "num_tokens": 6469627.0,
      "step": 2940
    },
    {
      "entropy": 5.918372678756714,
      "epoch": 0.2922786820166733,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004995784167859043,
      "loss": 5.825,
      "mean_token_accuracy": 0.129074639827013,
      "num_tokens": 6480351.0,
      "step": 2945
    },
    {
      "entropy": 6.008989477157593,
      "epoch": 0.2927749106788408,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004995762460409349,
      "loss": 5.8548,
      "mean_token_accuracy": 0.1337804116308689,
      "num_tokens": 6491651.0,
      "step": 2950
    },
    {
      "entropy": 5.920342969894409,
      "epoch": 0.29327113934100835,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004995740697269584,
      "loss": 5.6923,
      "mean_token_accuracy": 0.13204924538731574,
      "num_tokens": 6502531.0,
      "step": 2955
    },
    {
      "entropy": 5.878091621398926,
      "epoch": 0.2937673680031759,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004995718878440289,
      "loss": 5.7131,
      "mean_token_accuracy": 0.14013013020157813,
      "num_tokens": 6511938.0,
      "step": 2960
    },
    {
      "entropy": 6.0145423412323,
      "epoch": 0.2942635966653434,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995697003922006,
      "loss": 5.8628,
      "mean_token_accuracy": 0.1239828608930111,
      "num_tokens": 6522775.0,
      "step": 2965
    },
    {
      "entropy": 5.953707361221314,
      "epoch": 0.29475982532751094,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004995675073715274,
      "loss": 5.681,
      "mean_token_accuracy": 0.13056408539414405,
      "num_tokens": 6534660.0,
      "step": 2970
    },
    {
      "entropy": 5.901784515380859,
      "epoch": 0.29525605398967847,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004995653087820641,
      "loss": 5.8299,
      "mean_token_accuracy": 0.13245153054594994,
      "num_tokens": 6545673.0,
      "step": 2975
    },
    {
      "entropy": 5.9352250576019285,
      "epoch": 0.29575228265184594,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499563104623865,
      "loss": 5.7862,
      "mean_token_accuracy": 0.1360513038933277,
      "num_tokens": 6556912.0,
      "step": 2980
    },
    {
      "entropy": 5.981804084777832,
      "epoch": 0.2962485113140135,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004995608948969849,
      "loss": 5.8789,
      "mean_token_accuracy": 0.12770932540297508,
      "num_tokens": 6567532.0,
      "step": 2985
    },
    {
      "entropy": 5.955900192260742,
      "epoch": 0.296744739976181,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004995586796014784,
      "loss": 5.8553,
      "mean_token_accuracy": 0.13003772795200347,
      "num_tokens": 6577830.0,
      "step": 2990
    },
    {
      "entropy": 5.927827978134156,
      "epoch": 0.29724096863834853,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004995564587374005,
      "loss": 5.7476,
      "mean_token_accuracy": 0.13181802555918692,
      "num_tokens": 6587254.0,
      "step": 2995
    },
    {
      "entropy": 5.962104320526123,
      "epoch": 0.29773719730051607,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004995542323048063,
      "loss": 5.934,
      "mean_token_accuracy": 0.12504435703158379,
      "num_tokens": 6598294.0,
      "step": 3000
    },
    {
      "epoch": 0.29773719730051607,
      "eval_entropy": 5.7928955505681055,
      "eval_loss": 5.8702263832092285,
      "eval_mean_token_accuracy": 0.13254506512458533,
      "eval_num_tokens": 6598294.0,
      "eval_runtime": 20.7058,
      "eval_samples_per_second": 1626.066,
      "eval_steps_per_second": 203.276,
      "step": 3000
    },
    {
      "entropy": 6.02299394607544,
      "epoch": 0.2982334259626836,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000499552000303751,
      "loss": 5.8132,
      "mean_token_accuracy": 0.13359044566750528,
      "num_tokens": 6608772.0,
      "step": 3005
    },
    {
      "entropy": 5.907866477966309,
      "epoch": 0.2987296546248511,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00049954976273429,
      "loss": 5.8093,
      "mean_token_accuracy": 0.1337615132331848,
      "num_tokens": 6619973.0,
      "step": 3010
    },
    {
      "entropy": 5.9264623641967775,
      "epoch": 0.29922588328701866,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004995475195964787,
      "loss": 5.8292,
      "mean_token_accuracy": 0.13057641312479973,
      "num_tokens": 6631172.0,
      "step": 3015
    },
    {
      "entropy": 5.9544103145599365,
      "epoch": 0.2997221119491862,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004995452708903727,
      "loss": 5.8188,
      "mean_token_accuracy": 0.12836946919560432,
      "num_tokens": 6641343.0,
      "step": 3020
    },
    {
      "entropy": 5.912361097335816,
      "epoch": 0.3002183406113537,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000499543016616028,
      "loss": 5.7628,
      "mean_token_accuracy": 0.13423253819346428,
      "num_tokens": 6652892.0,
      "step": 3025
    },
    {
      "entropy": 5.965789175033569,
      "epoch": 0.30071456927352125,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004995407567735002,
      "loss": 5.7887,
      "mean_token_accuracy": 0.13260234743356705,
      "num_tokens": 6664177.0,
      "step": 3030
    },
    {
      "entropy": 5.903903007507324,
      "epoch": 0.3012107979356888,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004995384913628456,
      "loss": 5.7612,
      "mean_token_accuracy": 0.13183784186840058,
      "num_tokens": 6674402.0,
      "step": 3035
    },
    {
      "entropy": 5.972148370742798,
      "epoch": 0.3017070265978563,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004995362203841202,
      "loss": 5.8938,
      "mean_token_accuracy": 0.12839276492595672,
      "num_tokens": 6684886.0,
      "step": 3040
    },
    {
      "entropy": 5.954754209518432,
      "epoch": 0.30220325526002384,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004995339438373804,
      "loss": 5.7202,
      "mean_token_accuracy": 0.135014508664608,
      "num_tokens": 6694774.0,
      "step": 3045
    },
    {
      "entropy": 5.888301420211792,
      "epoch": 0.30269948392219137,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004995316617226825,
      "loss": 5.7818,
      "mean_token_accuracy": 0.13432007953524588,
      "num_tokens": 6705346.0,
      "step": 3050
    },
    {
      "entropy": 5.871857500076294,
      "epoch": 0.30319571258435885,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004995293740400833,
      "loss": 5.7312,
      "mean_token_accuracy": 0.13942001089453698,
      "num_tokens": 6715579.0,
      "step": 3055
    },
    {
      "entropy": 5.8535606384277346,
      "epoch": 0.3036919412465264,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004995270807896395,
      "loss": 5.6946,
      "mean_token_accuracy": 0.13676538467407226,
      "num_tokens": 6726382.0,
      "step": 3060
    },
    {
      "entropy": 5.935056352615357,
      "epoch": 0.3041881699086939,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004995247819714079,
      "loss": 5.7698,
      "mean_token_accuracy": 0.1337558187544346,
      "num_tokens": 6737245.0,
      "step": 3065
    },
    {
      "entropy": 5.979394245147705,
      "epoch": 0.30468439857086144,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004995224775854454,
      "loss": 5.9624,
      "mean_token_accuracy": 0.12612314149737358,
      "num_tokens": 6749113.0,
      "step": 3070
    },
    {
      "entropy": 6.009639263153076,
      "epoch": 0.30518062723302897,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004995201676318093,
      "loss": 5.8219,
      "mean_token_accuracy": 0.1318033404648304,
      "num_tokens": 6761987.0,
      "step": 3075
    },
    {
      "entropy": 5.9234983921051025,
      "epoch": 0.3056768558951965,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004995178521105568,
      "loss": 5.766,
      "mean_token_accuracy": 0.12556311190128328,
      "num_tokens": 6771279.0,
      "step": 3080
    },
    {
      "entropy": 5.889677429199219,
      "epoch": 0.30617308455736403,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004995155310217454,
      "loss": 5.6911,
      "mean_token_accuracy": 0.1361276350915432,
      "num_tokens": 6781038.0,
      "step": 3085
    },
    {
      "entropy": 5.8956520557403564,
      "epoch": 0.30666931321953156,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004995132043654327,
      "loss": 5.8346,
      "mean_token_accuracy": 0.13275470435619355,
      "num_tokens": 6792686.0,
      "step": 3090
    },
    {
      "entropy": 5.951267910003662,
      "epoch": 0.3071655418816991,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004995108721416763,
      "loss": 5.8074,
      "mean_token_accuracy": 0.139174947142601,
      "num_tokens": 6804414.0,
      "step": 3095
    },
    {
      "entropy": 5.929559230804443,
      "epoch": 0.3076617705438666,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004995085343505339,
      "loss": 5.7652,
      "mean_token_accuracy": 0.13180396631360053,
      "num_tokens": 6816180.0,
      "step": 3100
    },
    {
      "entropy": 5.945132160186768,
      "epoch": 0.30815799920603415,
      "grad_norm": 0.875,
      "learning_rate": 0.0004995061909920638,
      "loss": 5.778,
      "mean_token_accuracy": 0.1298554077744484,
      "num_tokens": 6827114.0,
      "step": 3105
    },
    {
      "entropy": 5.969542980194092,
      "epoch": 0.3086542278682017,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004995038420663238,
      "loss": 5.8373,
      "mean_token_accuracy": 0.13185039311647415,
      "num_tokens": 6836953.0,
      "step": 3110
    },
    {
      "entropy": 5.944265985488892,
      "epoch": 0.3091504565303692,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004995014875733724,
      "loss": 5.8746,
      "mean_token_accuracy": 0.12786564826965333,
      "num_tokens": 6847963.0,
      "step": 3115
    },
    {
      "entropy": 5.981411457061768,
      "epoch": 0.30964668519253674,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004994991275132678,
      "loss": 5.8869,
      "mean_token_accuracy": 0.12628084421157837,
      "num_tokens": 6858264.0,
      "step": 3120
    },
    {
      "entropy": 5.948660182952881,
      "epoch": 0.31014291385470427,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004994967618860685,
      "loss": 5.7736,
      "mean_token_accuracy": 0.13805429711937905,
      "num_tokens": 6868861.0,
      "step": 3125
    },
    {
      "entropy": 5.981192207336425,
      "epoch": 0.31063914251687175,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004994943906918333,
      "loss": 5.7972,
      "mean_token_accuracy": 0.13693608418107034,
      "num_tokens": 6879797.0,
      "step": 3130
    },
    {
      "entropy": 5.884922504425049,
      "epoch": 0.3111353711790393,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004994920139306211,
      "loss": 5.7463,
      "mean_token_accuracy": 0.13194068968296052,
      "num_tokens": 6891323.0,
      "step": 3135
    },
    {
      "entropy": 5.909702634811401,
      "epoch": 0.3116315998412068,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004994896316024906,
      "loss": 5.8044,
      "mean_token_accuracy": 0.13013609647750854,
      "num_tokens": 6903345.0,
      "step": 3140
    },
    {
      "entropy": 5.932398700714112,
      "epoch": 0.31212782850337434,
      "grad_norm": 0.9375,
      "learning_rate": 0.000499487243707501,
      "loss": 5.7547,
      "mean_token_accuracy": 0.13403790220618247,
      "num_tokens": 6914342.0,
      "step": 3145
    },
    {
      "entropy": 5.918168926239014,
      "epoch": 0.31262405716554187,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004994848502457114,
      "loss": 5.8166,
      "mean_token_accuracy": 0.13116133585572243,
      "num_tokens": 6926691.0,
      "step": 3150
    },
    {
      "entropy": 5.885048723220825,
      "epoch": 0.3131202858277094,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004994824512171812,
      "loss": 5.7052,
      "mean_token_accuracy": 0.13244526460766792,
      "num_tokens": 6937841.0,
      "step": 3155
    },
    {
      "entropy": 5.895932054519653,
      "epoch": 0.31361651448987693,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004994800466219701,
      "loss": 5.781,
      "mean_token_accuracy": 0.12934989854693413,
      "num_tokens": 6948601.0,
      "step": 3160
    },
    {
      "entropy": 5.95720477104187,
      "epoch": 0.31411274315204446,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004994776364601374,
      "loss": 5.7171,
      "mean_token_accuracy": 0.13609042316675185,
      "num_tokens": 6959775.0,
      "step": 3165
    },
    {
      "entropy": 5.85569396018982,
      "epoch": 0.314608971814212,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004994752207317431,
      "loss": 5.7471,
      "mean_token_accuracy": 0.1365107588469982,
      "num_tokens": 6969202.0,
      "step": 3170
    },
    {
      "entropy": 5.846676874160766,
      "epoch": 0.3151052004763795,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000499472799436847,
      "loss": 5.7296,
      "mean_token_accuracy": 0.136111731082201,
      "num_tokens": 6978633.0,
      "step": 3175
    },
    {
      "entropy": 6.046626424789428,
      "epoch": 0.31560142913854705,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004994703725755092,
      "loss": 5.7934,
      "mean_token_accuracy": 0.13145529180765153,
      "num_tokens": 6989638.0,
      "step": 3180
    },
    {
      "entropy": 5.903738021850586,
      "epoch": 0.3160976578007146,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004994679401477899,
      "loss": 5.7616,
      "mean_token_accuracy": 0.1309955298900604,
      "num_tokens": 7001606.0,
      "step": 3185
    },
    {
      "entropy": 5.860150480270386,
      "epoch": 0.3165938864628821,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004994655021537492,
      "loss": 5.7334,
      "mean_token_accuracy": 0.13141413554549217,
      "num_tokens": 7012915.0,
      "step": 3190
    },
    {
      "entropy": 5.89152774810791,
      "epoch": 0.31709011512504964,
      "grad_norm": 0.875,
      "learning_rate": 0.0004994630585934478,
      "loss": 5.7898,
      "mean_token_accuracy": 0.12930853962898253,
      "num_tokens": 7023191.0,
      "step": 3195
    },
    {
      "entropy": 5.914811658859253,
      "epoch": 0.3175863437872172,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004994606094669463,
      "loss": 5.681,
      "mean_token_accuracy": 0.13937111124396323,
      "num_tokens": 7032506.0,
      "step": 3200
    },
    {
      "entropy": 5.89855432510376,
      "epoch": 0.3180825724493847,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004994581547743053,
      "loss": 5.814,
      "mean_token_accuracy": 0.12698504626750945,
      "num_tokens": 7046098.0,
      "step": 3205
    },
    {
      "entropy": 6.016089820861817,
      "epoch": 0.3185788011115522,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004994556945155857,
      "loss": 5.8264,
      "mean_token_accuracy": 0.13558279499411582,
      "num_tokens": 7057434.0,
      "step": 3210
    },
    {
      "entropy": 5.82761664390564,
      "epoch": 0.3190750297737197,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004994532286908486,
      "loss": 5.7577,
      "mean_token_accuracy": 0.12784940525889396,
      "num_tokens": 7067674.0,
      "step": 3215
    },
    {
      "entropy": 5.943560075759888,
      "epoch": 0.31957125843588724,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499450757300155,
      "loss": 5.8249,
      "mean_token_accuracy": 0.13068325966596603,
      "num_tokens": 7077898.0,
      "step": 3220
    },
    {
      "entropy": 5.973318004608155,
      "epoch": 0.32006748709805477,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004994482803435664,
      "loss": 5.7882,
      "mean_token_accuracy": 0.12848865017294883,
      "num_tokens": 7088566.0,
      "step": 3225
    },
    {
      "entropy": 5.90263147354126,
      "epoch": 0.3205637157602223,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000499445797821144,
      "loss": 5.7808,
      "mean_token_accuracy": 0.13472721502184867,
      "num_tokens": 7099533.0,
      "step": 3230
    },
    {
      "entropy": 5.94389238357544,
      "epoch": 0.32105994442238983,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004994433097329494,
      "loss": 5.7632,
      "mean_token_accuracy": 0.13370973244309425,
      "num_tokens": 7112254.0,
      "step": 3235
    },
    {
      "entropy": 5.9783838272094725,
      "epoch": 0.32155617308455736,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004994408160790444,
      "loss": 5.7846,
      "mean_token_accuracy": 0.13278804942965508,
      "num_tokens": 7124704.0,
      "step": 3240
    },
    {
      "entropy": 6.045426225662231,
      "epoch": 0.3220524017467249,
      "grad_norm": 0.77734375,
      "learning_rate": 0.0004994383168594907,
      "loss": 5.9735,
      "mean_token_accuracy": 0.11922514736652375,
      "num_tokens": 7137741.0,
      "step": 3245
    },
    {
      "entropy": 5.94087586402893,
      "epoch": 0.3225486304088924,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004994358120743506,
      "loss": 5.7579,
      "mean_token_accuracy": 0.12984591871500015,
      "num_tokens": 7147888.0,
      "step": 3250
    },
    {
      "entropy": 5.7819092750549315,
      "epoch": 0.32304485907105995,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004994333017236858,
      "loss": 5.7076,
      "mean_token_accuracy": 0.14143425077199936,
      "num_tokens": 7158950.0,
      "step": 3255
    },
    {
      "entropy": 5.94332447052002,
      "epoch": 0.3235410877332275,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004994307858075587,
      "loss": 5.8421,
      "mean_token_accuracy": 0.12570593357086182,
      "num_tokens": 7170156.0,
      "step": 3260
    },
    {
      "entropy": 5.932934379577636,
      "epoch": 0.324037316395395,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004994282643260319,
      "loss": 5.7048,
      "mean_token_accuracy": 0.13710035309195517,
      "num_tokens": 7180576.0,
      "step": 3265
    },
    {
      "entropy": 5.811103487014771,
      "epoch": 0.32453354505756254,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004994257372791676,
      "loss": 5.7565,
      "mean_token_accuracy": 0.13245908543467522,
      "num_tokens": 7192380.0,
      "step": 3270
    },
    {
      "entropy": 5.963694858551025,
      "epoch": 0.3250297737197301,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0004994232046670286,
      "loss": 5.8183,
      "mean_token_accuracy": 0.13459155336022377,
      "num_tokens": 7203898.0,
      "step": 3275
    },
    {
      "entropy": 5.989812850952148,
      "epoch": 0.3255260023818976,
      "grad_norm": 0.875,
      "learning_rate": 0.0004994206664896778,
      "loss": 5.7841,
      "mean_token_accuracy": 0.13584727346897124,
      "num_tokens": 7215241.0,
      "step": 3280
    },
    {
      "entropy": 5.931274271011352,
      "epoch": 0.3260222310440651,
      "grad_norm": 0.875,
      "learning_rate": 0.0004994181227471779,
      "loss": 5.7383,
      "mean_token_accuracy": 0.13464230597019194,
      "num_tokens": 7226039.0,
      "step": 3285
    },
    {
      "entropy": 5.8311902523040775,
      "epoch": 0.3265184597062326,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004994155734395923,
      "loss": 5.7687,
      "mean_token_accuracy": 0.13406949862837791,
      "num_tokens": 7237148.0,
      "step": 3290
    },
    {
      "entropy": 5.923242807388306,
      "epoch": 0.32701468836840014,
      "grad_norm": 0.90625,
      "learning_rate": 0.000499413018566984,
      "loss": 5.7621,
      "mean_token_accuracy": 0.1315522961318493,
      "num_tokens": 7249284.0,
      "step": 3295
    },
    {
      "entropy": 5.875470447540283,
      "epoch": 0.32751091703056767,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004994104581294163,
      "loss": 5.7631,
      "mean_token_accuracy": 0.12935446053743363,
      "num_tokens": 7259967.0,
      "step": 3300
    },
    {
      "entropy": 5.827996301651001,
      "epoch": 0.3280071456927352,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004994078921269529,
      "loss": 5.7235,
      "mean_token_accuracy": 0.13562380596995355,
      "num_tokens": 7271145.0,
      "step": 3305
    },
    {
      "entropy": 6.002976369857788,
      "epoch": 0.32850337435490273,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004994053205596573,
      "loss": 5.7351,
      "mean_token_accuracy": 0.13544280007481574,
      "num_tokens": 7281872.0,
      "step": 3310
    },
    {
      "entropy": 5.815413761138916,
      "epoch": 0.32899960301707026,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004994027434275934,
      "loss": 5.7639,
      "mean_token_accuracy": 0.1359103187918663,
      "num_tokens": 7293330.0,
      "step": 3315
    },
    {
      "entropy": 5.9990698337554935,
      "epoch": 0.3294958316792378,
      "grad_norm": 0.765625,
      "learning_rate": 0.0004994001607308249,
      "loss": 5.8214,
      "mean_token_accuracy": 0.12711838632822037,
      "num_tokens": 7304187.0,
      "step": 3320
    },
    {
      "entropy": 5.941404533386231,
      "epoch": 0.3299920603414053,
      "grad_norm": 0.8046875,
      "learning_rate": 0.000499397572469416,
      "loss": 5.7195,
      "mean_token_accuracy": 0.13139874935150148,
      "num_tokens": 7314756.0,
      "step": 3325
    },
    {
      "entropy": 5.826371097564698,
      "epoch": 0.33048828900357285,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0004993949786434308,
      "loss": 5.7561,
      "mean_token_accuracy": 0.13403744399547576,
      "num_tokens": 7325711.0,
      "step": 3330
    },
    {
      "entropy": 5.872751712799072,
      "epoch": 0.3309845176657404,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004993923792529337,
      "loss": 5.6831,
      "mean_token_accuracy": 0.1406965471804142,
      "num_tokens": 7335639.0,
      "step": 3335
    },
    {
      "entropy": 5.956606531143189,
      "epoch": 0.3314807463279079,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004993897742979892,
      "loss": 5.8027,
      "mean_token_accuracy": 0.13377716019749641,
      "num_tokens": 7346621.0,
      "step": 3340
    },
    {
      "entropy": 5.9286905288696286,
      "epoch": 0.33197697499007545,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004993871637786616,
      "loss": 5.7705,
      "mean_token_accuracy": 0.13041678220033645,
      "num_tokens": 7358950.0,
      "step": 3345
    },
    {
      "entropy": 5.821332550048828,
      "epoch": 0.332473203652243,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000499384547695016,
      "loss": 5.626,
      "mean_token_accuracy": 0.13846936598420143,
      "num_tokens": 7369629.0,
      "step": 3350
    },
    {
      "entropy": 5.872072839736939,
      "epoch": 0.3329694323144105,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000499381926047117,
      "loss": 5.7262,
      "mean_token_accuracy": 0.13408969193696976,
      "num_tokens": 7379853.0,
      "step": 3355
    },
    {
      "entropy": 5.865009212493897,
      "epoch": 0.333465660976578,
      "grad_norm": 0.90625,
      "learning_rate": 0.00049937929883503,
      "loss": 5.7563,
      "mean_token_accuracy": 0.13046114146709442,
      "num_tokens": 7391323.0,
      "step": 3360
    },
    {
      "entropy": 5.887697219848633,
      "epoch": 0.3339618896387455,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993766660588196,
      "loss": 5.7304,
      "mean_token_accuracy": 0.13350997790694236,
      "num_tokens": 7402201.0,
      "step": 3365
    },
    {
      "entropy": 6.01474552154541,
      "epoch": 0.33445811830091304,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004993740277185515,
      "loss": 5.9623,
      "mean_token_accuracy": 0.12484050020575524,
      "num_tokens": 7413783.0,
      "step": 3370
    },
    {
      "entropy": 5.840613555908203,
      "epoch": 0.3349543469630806,
      "grad_norm": 0.859375,
      "learning_rate": 0.000499371383814291,
      "loss": 5.7223,
      "mean_token_accuracy": 0.13400618061423303,
      "num_tokens": 7425909.0,
      "step": 3375
    },
    {
      "entropy": 5.847539567947388,
      "epoch": 0.3354505756252481,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004993687343461036,
      "loss": 5.6318,
      "mean_token_accuracy": 0.13809972107410431,
      "num_tokens": 7436346.0,
      "step": 3380
    },
    {
      "entropy": 5.904675722122192,
      "epoch": 0.33594680428741563,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004993660793140551,
      "loss": 5.6827,
      "mean_token_accuracy": 0.14161865711212157,
      "num_tokens": 7446596.0,
      "step": 3385
    },
    {
      "entropy": 5.782830667495728,
      "epoch": 0.33644303294958317,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004993634187182113,
      "loss": 5.7883,
      "mean_token_accuracy": 0.12615693137049674,
      "num_tokens": 7457697.0,
      "step": 3390
    },
    {
      "entropy": 6.065376043319702,
      "epoch": 0.3369392616117507,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004993607525586381,
      "loss": 5.8802,
      "mean_token_accuracy": 0.131008081138134,
      "num_tokens": 7468541.0,
      "step": 3395
    },
    {
      "entropy": 5.9343438148498535,
      "epoch": 0.3374354902739182,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004993580808354018,
      "loss": 5.7764,
      "mean_token_accuracy": 0.13551992177963257,
      "num_tokens": 7479781.0,
      "step": 3400
    },
    {
      "entropy": 5.837701988220215,
      "epoch": 0.33793171893608576,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004993554035485685,
      "loss": 5.6816,
      "mean_token_accuracy": 0.1342918574810028,
      "num_tokens": 7490722.0,
      "step": 3405
    },
    {
      "entropy": 5.948045158386231,
      "epoch": 0.3384279475982533,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004993527206982047,
      "loss": 5.7867,
      "mean_token_accuracy": 0.13522917330265044,
      "num_tokens": 7502293.0,
      "step": 3410
    },
    {
      "entropy": 5.918159008026123,
      "epoch": 0.3389241762604208,
      "grad_norm": 0.75390625,
      "learning_rate": 0.0004993500322843767,
      "loss": 5.7208,
      "mean_token_accuracy": 0.13539437279105188,
      "num_tokens": 7513846.0,
      "step": 3415
    },
    {
      "entropy": 5.830818223953247,
      "epoch": 0.33942040492258835,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004993473383071514,
      "loss": 5.7466,
      "mean_token_accuracy": 0.13467329666018485,
      "num_tokens": 7524062.0,
      "step": 3420
    },
    {
      "entropy": 5.901687240600586,
      "epoch": 0.3399166335847559,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004993446387665956,
      "loss": 5.7989,
      "mean_token_accuracy": 0.13709105178713799,
      "num_tokens": 7535533.0,
      "step": 3425
    },
    {
      "entropy": 5.914133930206299,
      "epoch": 0.3404128622469234,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004993419336627761,
      "loss": 5.6902,
      "mean_token_accuracy": 0.13548238351941108,
      "num_tokens": 7545951.0,
      "step": 3430
    },
    {
      "entropy": 5.817704439163208,
      "epoch": 0.3409090909090909,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004993392229957602,
      "loss": 5.6334,
      "mean_token_accuracy": 0.1382601298391819,
      "num_tokens": 7557320.0,
      "step": 3435
    },
    {
      "entropy": 5.805458402633667,
      "epoch": 0.3414053195712584,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004993365067656148,
      "loss": 5.7513,
      "mean_token_accuracy": 0.1329974986612797,
      "num_tokens": 7568078.0,
      "step": 3440
    },
    {
      "entropy": 5.883556461334228,
      "epoch": 0.34190154823342594,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004993337849724076,
      "loss": 5.7412,
      "mean_token_accuracy": 0.13214808478951454,
      "num_tokens": 7578940.0,
      "step": 3445
    },
    {
      "entropy": 5.893185567855835,
      "epoch": 0.3423977768955935,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004993310576162059,
      "loss": 5.7479,
      "mean_token_accuracy": 0.13639405518770217,
      "num_tokens": 7589864.0,
      "step": 3450
    },
    {
      "entropy": 5.922738981246948,
      "epoch": 0.342894005557761,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004993283246970772,
      "loss": 5.7057,
      "mean_token_accuracy": 0.13013398125767708,
      "num_tokens": 7599795.0,
      "step": 3455
    },
    {
      "entropy": 5.854125308990478,
      "epoch": 0.34339023421992854,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004993255862150896,
      "loss": 5.6876,
      "mean_token_accuracy": 0.13669478967785836,
      "num_tokens": 7610056.0,
      "step": 3460
    },
    {
      "entropy": 5.802547788619995,
      "epoch": 0.34388646288209607,
      "grad_norm": 0.875,
      "learning_rate": 0.0004993228421703108,
      "loss": 5.6874,
      "mean_token_accuracy": 0.13286467269062996,
      "num_tokens": 7620218.0,
      "step": 3465
    },
    {
      "entropy": 5.884554147720337,
      "epoch": 0.3443826915442636,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004993200925628088,
      "loss": 5.7779,
      "mean_token_accuracy": 0.1346588559448719,
      "num_tokens": 7632781.0,
      "step": 3470
    },
    {
      "entropy": 5.8774271488189695,
      "epoch": 0.3448789202064311,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000499317337392652,
      "loss": 5.7513,
      "mean_token_accuracy": 0.13203491717576982,
      "num_tokens": 7643845.0,
      "step": 3475
    },
    {
      "entropy": 5.910117483139038,
      "epoch": 0.34537514886859866,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004993145766599085,
      "loss": 5.7793,
      "mean_token_accuracy": 0.13024887666106225,
      "num_tokens": 7654583.0,
      "step": 3480
    },
    {
      "entropy": 5.8562994480133055,
      "epoch": 0.3458713775307662,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004993118103646469,
      "loss": 5.6299,
      "mean_token_accuracy": 0.14034538567066193,
      "num_tokens": 7665714.0,
      "step": 3485
    },
    {
      "entropy": 5.8359884262084964,
      "epoch": 0.3463676061929337,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004993090385069357,
      "loss": 5.6924,
      "mean_token_accuracy": 0.1398855820298195,
      "num_tokens": 7676011.0,
      "step": 3490
    },
    {
      "entropy": 5.825574779510498,
      "epoch": 0.34686383485510125,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004993062610868437,
      "loss": 5.7376,
      "mean_token_accuracy": 0.13499814942479132,
      "num_tokens": 7686927.0,
      "step": 3495
    },
    {
      "entropy": 5.846639966964721,
      "epoch": 0.3473600635172688,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004993034781044398,
      "loss": 5.6989,
      "mean_token_accuracy": 0.135499270260334,
      "num_tokens": 7698419.0,
      "step": 3500
    },
    {
      "entropy": 5.902319765090942,
      "epoch": 0.3478562921794363,
      "grad_norm": 0.859375,
      "learning_rate": 0.000499300689559793,
      "loss": 5.6731,
      "mean_token_accuracy": 0.13935840278863906,
      "num_tokens": 7709695.0,
      "step": 3505
    },
    {
      "entropy": 5.833289051055909,
      "epoch": 0.3483525208416038,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004992978954529724,
      "loss": 5.6456,
      "mean_token_accuracy": 0.1389060892164707,
      "num_tokens": 7721309.0,
      "step": 3510
    },
    {
      "entropy": 5.848011827468872,
      "epoch": 0.3488487495037713,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004992950957840473,
      "loss": 5.7182,
      "mean_token_accuracy": 0.13931151628494262,
      "num_tokens": 7732074.0,
      "step": 3515
    },
    {
      "entropy": 5.895964670181274,
      "epoch": 0.34934497816593885,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000499292290553087,
      "loss": 5.7953,
      "mean_token_accuracy": 0.13158057406544685,
      "num_tokens": 7742869.0,
      "step": 3520
    },
    {
      "entropy": 5.8405561447143555,
      "epoch": 0.3498412068281064,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004992894797601614,
      "loss": 5.6901,
      "mean_token_accuracy": 0.13555181324481963,
      "num_tokens": 7754895.0,
      "step": 3525
    },
    {
      "entropy": 5.8825901508331295,
      "epoch": 0.3503374354902739,
      "grad_norm": 0.90625,
      "learning_rate": 0.00049928666340534,
      "loss": 5.6677,
      "mean_token_accuracy": 0.13643246665596961,
      "num_tokens": 7765262.0,
      "step": 3530
    },
    {
      "entropy": 5.8095235347747805,
      "epoch": 0.35083366415244144,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004992838414886925,
      "loss": 5.739,
      "mean_token_accuracy": 0.1327873632311821,
      "num_tokens": 7777348.0,
      "step": 3535
    },
    {
      "entropy": 5.916355848312378,
      "epoch": 0.35132989281460897,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004992810140102891,
      "loss": 5.7297,
      "mean_token_accuracy": 0.13413830399513244,
      "num_tokens": 7788161.0,
      "step": 3540
    },
    {
      "entropy": 5.840317058563232,
      "epoch": 0.3518261214767765,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004992781809701999,
      "loss": 5.7161,
      "mean_token_accuracy": 0.13731137067079544,
      "num_tokens": 7798629.0,
      "step": 3545
    },
    {
      "entropy": 5.829579782485962,
      "epoch": 0.35232235013894403,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004992753423684951,
      "loss": 5.7198,
      "mean_token_accuracy": 0.13580114766955376,
      "num_tokens": 7810148.0,
      "step": 3550
    },
    {
      "entropy": 5.863365554809571,
      "epoch": 0.35281857880111156,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004992724982052449,
      "loss": 5.71,
      "mean_token_accuracy": 0.13254974409937859,
      "num_tokens": 7821667.0,
      "step": 3555
    },
    {
      "entropy": 5.798212766647339,
      "epoch": 0.3533148074632791,
      "grad_norm": 0.7890625,
      "learning_rate": 0.00049926964848052,
      "loss": 5.6957,
      "mean_token_accuracy": 0.13729101940989494,
      "num_tokens": 7832184.0,
      "step": 3560
    },
    {
      "entropy": 5.927416610717773,
      "epoch": 0.3538110361254466,
      "grad_norm": 0.875,
      "learning_rate": 0.0004992667931943913,
      "loss": 5.7183,
      "mean_token_accuracy": 0.13577507585287094,
      "num_tokens": 7842983.0,
      "step": 3565
    },
    {
      "entropy": 5.856418895721435,
      "epoch": 0.35430726478761415,
      "grad_norm": 0.7734375,
      "learning_rate": 0.0004992639323469293,
      "loss": 5.6354,
      "mean_token_accuracy": 0.14334375709295272,
      "num_tokens": 7854783.0,
      "step": 3570
    },
    {
      "entropy": 5.773731470108032,
      "epoch": 0.3548034934497817,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000499261065938205,
      "loss": 5.6583,
      "mean_token_accuracy": 0.1432967111468315,
      "num_tokens": 7864684.0,
      "step": 3575
    },
    {
      "entropy": 5.895873069763184,
      "epoch": 0.3552997221119492,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004992581939682895,
      "loss": 5.8056,
      "mean_token_accuracy": 0.1261812210083008,
      "num_tokens": 7876846.0,
      "step": 3580
    },
    {
      "entropy": 5.935712575912476,
      "epoch": 0.3557959507741167,
      "grad_norm": 0.890625,
      "learning_rate": 0.000499255316437254,
      "loss": 5.6779,
      "mean_token_accuracy": 0.1337505742907524,
      "num_tokens": 7886826.0,
      "step": 3585
    },
    {
      "entropy": 5.801439237594605,
      "epoch": 0.3562921794362842,
      "grad_norm": 0.875,
      "learning_rate": 0.0004992524333451699,
      "loss": 5.7537,
      "mean_token_accuracy": 0.13139742091298104,
      "num_tokens": 7898941.0,
      "step": 3590
    },
    {
      "entropy": 5.849826717376709,
      "epoch": 0.35678840809845175,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992495446921087,
      "loss": 5.6674,
      "mean_token_accuracy": 0.13365738093852997,
      "num_tokens": 7909916.0,
      "step": 3595
    },
    {
      "entropy": 5.808553171157837,
      "epoch": 0.3572846367606193,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004992466504781419,
      "loss": 5.6792,
      "mean_token_accuracy": 0.13627909868955612,
      "num_tokens": 7922137.0,
      "step": 3600
    },
    {
      "entropy": 5.79820442199707,
      "epoch": 0.3577808654227868,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004992437507033415,
      "loss": 5.616,
      "mean_token_accuracy": 0.13959808573126792,
      "num_tokens": 7932940.0,
      "step": 3605
    },
    {
      "entropy": 5.881356430053711,
      "epoch": 0.35827709408495434,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004992408453677792,
      "loss": 5.7851,
      "mean_token_accuracy": 0.12732164859771727,
      "num_tokens": 7943735.0,
      "step": 3610
    },
    {
      "entropy": 5.798199939727783,
      "epoch": 0.35877332274712187,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004992379344715273,
      "loss": 5.672,
      "mean_token_accuracy": 0.13556718379259108,
      "num_tokens": 7955323.0,
      "step": 3615
    },
    {
      "entropy": 5.827555418014526,
      "epoch": 0.3592695514092894,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004992350180146577,
      "loss": 5.6371,
      "mean_token_accuracy": 0.13796332776546477,
      "num_tokens": 7966073.0,
      "step": 3620
    },
    {
      "entropy": 5.883342456817627,
      "epoch": 0.35976578007145693,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004992320959972428,
      "loss": 5.7951,
      "mean_token_accuracy": 0.13291794434189796,
      "num_tokens": 7977027.0,
      "step": 3625
    },
    {
      "entropy": 5.816143798828125,
      "epoch": 0.36026200873362446,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004992291684193551,
      "loss": 5.6621,
      "mean_token_accuracy": 0.13837531432509423,
      "num_tokens": 7987368.0,
      "step": 3630
    },
    {
      "entropy": 5.862054634094238,
      "epoch": 0.360758237395792,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004992262352810672,
      "loss": 5.7323,
      "mean_token_accuracy": 0.13465685471892358,
      "num_tokens": 8000060.0,
      "step": 3635
    },
    {
      "entropy": 5.904985523223877,
      "epoch": 0.3612544660579595,
      "grad_norm": 0.875,
      "learning_rate": 0.0004992232965824518,
      "loss": 5.779,
      "mean_token_accuracy": 0.13045103549957277,
      "num_tokens": 8011388.0,
      "step": 3640
    },
    {
      "entropy": 5.78095817565918,
      "epoch": 0.36175069472012705,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000499220352323582,
      "loss": 5.6446,
      "mean_token_accuracy": 0.13995237201452254,
      "num_tokens": 8021682.0,
      "step": 3645
    },
    {
      "entropy": 5.828157138824463,
      "epoch": 0.3622469233822946,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004992174025045304,
      "loss": 5.6637,
      "mean_token_accuracy": 0.1378515474498272,
      "num_tokens": 8032799.0,
      "step": 3650
    },
    {
      "entropy": 5.791543912887573,
      "epoch": 0.3627431520444621,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004992144471253704,
      "loss": 5.5542,
      "mean_token_accuracy": 0.14514420479536055,
      "num_tokens": 8042742.0,
      "step": 3655
    },
    {
      "entropy": 5.836719179153443,
      "epoch": 0.3632393807066296,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004992114861861753,
      "loss": 5.726,
      "mean_token_accuracy": 0.13775497898459435,
      "num_tokens": 8054068.0,
      "step": 3660
    },
    {
      "entropy": 5.895388793945313,
      "epoch": 0.3637356093687971,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004992085196870184,
      "loss": 5.7041,
      "mean_token_accuracy": 0.13520997688174247,
      "num_tokens": 8065286.0,
      "step": 3665
    },
    {
      "entropy": 5.8350739002227785,
      "epoch": 0.36423183803096465,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004992055476279734,
      "loss": 5.7644,
      "mean_token_accuracy": 0.12864286303520203,
      "num_tokens": 8077122.0,
      "step": 3670
    },
    {
      "entropy": 5.914077806472778,
      "epoch": 0.3647280666931322,
      "grad_norm": 0.875,
      "learning_rate": 0.0004992025700091139,
      "loss": 5.7053,
      "mean_token_accuracy": 0.1394345611333847,
      "num_tokens": 8087036.0,
      "step": 3675
    },
    {
      "entropy": 5.887608432769776,
      "epoch": 0.3652242953552997,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004991995868305138,
      "loss": 5.7439,
      "mean_token_accuracy": 0.13455166965723037,
      "num_tokens": 8098517.0,
      "step": 3680
    },
    {
      "entropy": 5.778636121749878,
      "epoch": 0.36572052401746724,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004991965980922471,
      "loss": 5.6182,
      "mean_token_accuracy": 0.13923544213175773,
      "num_tokens": 8107475.0,
      "step": 3685
    },
    {
      "entropy": 5.82491021156311,
      "epoch": 0.36621675267963477,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004991936037943878,
      "loss": 5.7266,
      "mean_token_accuracy": 0.1303709253668785,
      "num_tokens": 8118529.0,
      "step": 3690
    },
    {
      "entropy": 5.9018360614776615,
      "epoch": 0.3667129813418023,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004991906039370104,
      "loss": 5.6292,
      "mean_token_accuracy": 0.14039468988776208,
      "num_tokens": 8128951.0,
      "step": 3695
    },
    {
      "entropy": 5.746214437484741,
      "epoch": 0.36720921000396983,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004991875985201889,
      "loss": 5.6492,
      "mean_token_accuracy": 0.1410698860883713,
      "num_tokens": 8139385.0,
      "step": 3700
    },
    {
      "entropy": 5.918550968170166,
      "epoch": 0.36770543866613736,
      "grad_norm": 0.859375,
      "learning_rate": 0.000499184587543998,
      "loss": 5.6843,
      "mean_token_accuracy": 0.1389869749546051,
      "num_tokens": 8149376.0,
      "step": 3705
    },
    {
      "entropy": 5.8605245590209964,
      "epoch": 0.3682016673283049,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004991815710085126,
      "loss": 5.7092,
      "mean_token_accuracy": 0.13653566539287568,
      "num_tokens": 8160687.0,
      "step": 3710
    },
    {
      "entropy": 5.872888708114624,
      "epoch": 0.3686978959904724,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004991785489138072,
      "loss": 5.7141,
      "mean_token_accuracy": 0.1377060107886791,
      "num_tokens": 8171762.0,
      "step": 3715
    },
    {
      "entropy": 5.724028301239014,
      "epoch": 0.36919412465263995,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004991755212599569,
      "loss": 5.4791,
      "mean_token_accuracy": 0.14507829397916794,
      "num_tokens": 8181643.0,
      "step": 3720
    },
    {
      "entropy": 5.697611093521118,
      "epoch": 0.3696903533148075,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004991724880470366,
      "loss": 5.7306,
      "mean_token_accuracy": 0.13918372690677644,
      "num_tokens": 8192692.0,
      "step": 3725
    },
    {
      "entropy": 5.91280164718628,
      "epoch": 0.370186581976975,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004991694492751217,
      "loss": 5.678,
      "mean_token_accuracy": 0.13002298176288604,
      "num_tokens": 8202941.0,
      "step": 3730
    },
    {
      "entropy": 5.799401473999024,
      "epoch": 0.3706828106391425,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004991664049442876,
      "loss": 5.6749,
      "mean_token_accuracy": 0.13614187836647035,
      "num_tokens": 8214206.0,
      "step": 3735
    },
    {
      "entropy": 5.792055654525757,
      "epoch": 0.37117903930131,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004991633550546097,
      "loss": 5.5088,
      "mean_token_accuracy": 0.15157946944236755,
      "num_tokens": 8223672.0,
      "step": 3740
    },
    {
      "entropy": 5.7840086936950685,
      "epoch": 0.37167526796347755,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004991602996061635,
      "loss": 5.6511,
      "mean_token_accuracy": 0.14170872122049333,
      "num_tokens": 8234189.0,
      "step": 3745
    },
    {
      "entropy": 5.839711046218872,
      "epoch": 0.3721714966256451,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004991572385990249,
      "loss": 5.6868,
      "mean_token_accuracy": 0.139129176735878,
      "num_tokens": 8245382.0,
      "step": 3750
    },
    {
      "entropy": 5.895852756500244,
      "epoch": 0.3726677252878126,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004991541720332698,
      "loss": 5.6406,
      "mean_token_accuracy": 0.13922024741768838,
      "num_tokens": 8254530.0,
      "step": 3755
    },
    {
      "entropy": 5.741476011276245,
      "epoch": 0.37316395394998014,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004991510999089743,
      "loss": 5.6856,
      "mean_token_accuracy": 0.13898830264806747,
      "num_tokens": 8264030.0,
      "step": 3760
    },
    {
      "entropy": 5.792221021652222,
      "epoch": 0.3736601826121477,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004991480222262145,
      "loss": 5.6503,
      "mean_token_accuracy": 0.14368472024798393,
      "num_tokens": 8273537.0,
      "step": 3765
    },
    {
      "entropy": 5.88636155128479,
      "epoch": 0.3741564112743152,
      "grad_norm": 0.70703125,
      "learning_rate": 0.0004991449389850668,
      "loss": 5.6834,
      "mean_token_accuracy": 0.13585606068372727,
      "num_tokens": 8286123.0,
      "step": 3770
    },
    {
      "entropy": 5.796380996704102,
      "epoch": 0.37465263993648273,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004991418501856076,
      "loss": 5.7091,
      "mean_token_accuracy": 0.13685489520430566,
      "num_tokens": 8298115.0,
      "step": 3775
    },
    {
      "entropy": 5.87733302116394,
      "epoch": 0.37514886859865026,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004991387558279134,
      "loss": 5.7145,
      "mean_token_accuracy": 0.1364159256219864,
      "num_tokens": 8309376.0,
      "step": 3780
    },
    {
      "entropy": 5.805009841918945,
      "epoch": 0.3756450972608178,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004991356559120612,
      "loss": 5.6778,
      "mean_token_accuracy": 0.13979559317231177,
      "num_tokens": 8319272.0,
      "step": 3785
    },
    {
      "entropy": 5.818359422683716,
      "epoch": 0.3761413259229853,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004991325504381275,
      "loss": 5.6691,
      "mean_token_accuracy": 0.13450096398591996,
      "num_tokens": 8330042.0,
      "step": 3790
    },
    {
      "entropy": 5.800448036193847,
      "epoch": 0.37663755458515286,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004991294394061896,
      "loss": 5.622,
      "mean_token_accuracy": 0.1370707221329212,
      "num_tokens": 8340394.0,
      "step": 3795
    },
    {
      "entropy": 5.775977277755738,
      "epoch": 0.3771337832473204,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004991263228163246,
      "loss": 5.6841,
      "mean_token_accuracy": 0.1365865007042885,
      "num_tokens": 8351103.0,
      "step": 3800
    },
    {
      "entropy": 5.865830612182617,
      "epoch": 0.3776300119094879,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0004991232006686096,
      "loss": 5.6929,
      "mean_token_accuracy": 0.13774350062012672,
      "num_tokens": 8363215.0,
      "step": 3805
    },
    {
      "entropy": 5.919228649139404,
      "epoch": 0.3781262405716554,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004991200729631223,
      "loss": 5.7373,
      "mean_token_accuracy": 0.12794058844447137,
      "num_tokens": 8374602.0,
      "step": 3810
    },
    {
      "entropy": 5.787239599227905,
      "epoch": 0.3786224692338229,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004991169396999401,
      "loss": 5.6555,
      "mean_token_accuracy": 0.13108878657221795,
      "num_tokens": 8386470.0,
      "step": 3815
    },
    {
      "entropy": 5.842096328735352,
      "epoch": 0.37911869789599045,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004991138008791408,
      "loss": 5.6799,
      "mean_token_accuracy": 0.13508746549487113,
      "num_tokens": 8396284.0,
      "step": 3820
    },
    {
      "entropy": 5.8682530403137205,
      "epoch": 0.379614926558158,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004991106565008021,
      "loss": 5.7021,
      "mean_token_accuracy": 0.13416668847203256,
      "num_tokens": 8407856.0,
      "step": 3825
    },
    {
      "entropy": 5.822223329544068,
      "epoch": 0.3801111552203255,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004991075065650019,
      "loss": 5.6105,
      "mean_token_accuracy": 0.1422433525323868,
      "num_tokens": 8419560.0,
      "step": 3830
    },
    {
      "entropy": 5.871718502044677,
      "epoch": 0.38060738388249304,
      "grad_norm": 0.875,
      "learning_rate": 0.0004991043510718186,
      "loss": 5.6177,
      "mean_token_accuracy": 0.1404113858938217,
      "num_tokens": 8430594.0,
      "step": 3835
    },
    {
      "entropy": 5.7367877006530765,
      "epoch": 0.3811036125446606,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004991011900213303,
      "loss": 5.588,
      "mean_token_accuracy": 0.1445833697915077,
      "num_tokens": 8440549.0,
      "step": 3840
    },
    {
      "entropy": 5.7166365623474125,
      "epoch": 0.3815998412068281,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004990980234136152,
      "loss": 5.6232,
      "mean_token_accuracy": 0.13667350262403488,
      "num_tokens": 8451553.0,
      "step": 3845
    },
    {
      "entropy": 5.814205455780029,
      "epoch": 0.38209606986899564,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004990948512487522,
      "loss": 5.6252,
      "mean_token_accuracy": 0.1421022191643715,
      "num_tokens": 8462899.0,
      "step": 3850
    },
    {
      "entropy": 5.8078104019165036,
      "epoch": 0.38259229853116317,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004990916735268198,
      "loss": 5.6796,
      "mean_token_accuracy": 0.13645819798111916,
      "num_tokens": 8474566.0,
      "step": 3855
    },
    {
      "entropy": 5.869540357589722,
      "epoch": 0.3830885271933307,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004990884902478966,
      "loss": 5.6865,
      "mean_token_accuracy": 0.14038145467638968,
      "num_tokens": 8484786.0,
      "step": 3860
    },
    {
      "entropy": 5.750705623626709,
      "epoch": 0.3835847558554982,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004990853014120619,
      "loss": 5.625,
      "mean_token_accuracy": 0.14226635843515395,
      "num_tokens": 8496416.0,
      "step": 3865
    },
    {
      "entropy": 5.798192834854126,
      "epoch": 0.38408098451766576,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004990821070193944,
      "loss": 5.6561,
      "mean_token_accuracy": 0.136373158544302,
      "num_tokens": 8507287.0,
      "step": 3870
    },
    {
      "entropy": 5.865294790267944,
      "epoch": 0.3845772131798333,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004990789070699736,
      "loss": 5.6948,
      "mean_token_accuracy": 0.13455733358860017,
      "num_tokens": 8517922.0,
      "step": 3875
    },
    {
      "entropy": 5.765826034545898,
      "epoch": 0.3850734418420008,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004990757015638787,
      "loss": 5.631,
      "mean_token_accuracy": 0.14302095845341684,
      "num_tokens": 8529532.0,
      "step": 3880
    },
    {
      "entropy": 5.816843843460083,
      "epoch": 0.3855696705041683,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004990724905011894,
      "loss": 5.6033,
      "mean_token_accuracy": 0.14065930545330046,
      "num_tokens": 8540207.0,
      "step": 3885
    },
    {
      "entropy": 5.801877880096436,
      "epoch": 0.3860658991663358,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000499069273881985,
      "loss": 5.6383,
      "mean_token_accuracy": 0.13733408302068711,
      "num_tokens": 8551634.0,
      "step": 3890
    },
    {
      "entropy": 5.8024838924407955,
      "epoch": 0.38656212782850335,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004990660517063455,
      "loss": 5.7033,
      "mean_token_accuracy": 0.1306481033563614,
      "num_tokens": 8563256.0,
      "step": 3895
    },
    {
      "entropy": 5.81333646774292,
      "epoch": 0.3870583564906709,
      "grad_norm": 0.875,
      "learning_rate": 0.0004990628239743508,
      "loss": 5.7163,
      "mean_token_accuracy": 0.13748448565602303,
      "num_tokens": 8573833.0,
      "step": 3900
    },
    {
      "entropy": 5.79175386428833,
      "epoch": 0.3875545851528384,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004990595906860807,
      "loss": 5.6222,
      "mean_token_accuracy": 0.1423626087605953,
      "num_tokens": 8584283.0,
      "step": 3905
    },
    {
      "entropy": 5.789517545700074,
      "epoch": 0.38805081381500595,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004990563518416157,
      "loss": 5.6789,
      "mean_token_accuracy": 0.1358125925064087,
      "num_tokens": 8595344.0,
      "step": 3910
    },
    {
      "entropy": 5.814813470840454,
      "epoch": 0.3885470424771735,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004990531074410358,
      "loss": 5.6036,
      "mean_token_accuracy": 0.137834145873785,
      "num_tokens": 8605625.0,
      "step": 3915
    },
    {
      "entropy": 5.7738450050354,
      "epoch": 0.389043271139341,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004990498574844218,
      "loss": 5.5894,
      "mean_token_accuracy": 0.14018410369753836,
      "num_tokens": 8616259.0,
      "step": 3920
    },
    {
      "entropy": 5.82464485168457,
      "epoch": 0.38953949980150854,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004990466019718542,
      "loss": 5.6034,
      "mean_token_accuracy": 0.1412269026041031,
      "num_tokens": 8627179.0,
      "step": 3925
    },
    {
      "entropy": 5.831127595901489,
      "epoch": 0.39003572846367607,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004990433409034134,
      "loss": 5.71,
      "mean_token_accuracy": 0.14471403062343596,
      "num_tokens": 8637448.0,
      "step": 3930
    },
    {
      "entropy": 5.725909614562989,
      "epoch": 0.3905319571258436,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004990400742791806,
      "loss": 5.6293,
      "mean_token_accuracy": 0.1356988586485386,
      "num_tokens": 8649168.0,
      "step": 3935
    },
    {
      "entropy": 5.88536319732666,
      "epoch": 0.39102818578801113,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004990368020992368,
      "loss": 5.7237,
      "mean_token_accuracy": 0.13518148735165597,
      "num_tokens": 8660653.0,
      "step": 3940
    },
    {
      "entropy": 5.793624401092529,
      "epoch": 0.39152441445017866,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000499033524363663,
      "loss": 5.6971,
      "mean_token_accuracy": 0.1437949612736702,
      "num_tokens": 8672000.0,
      "step": 3945
    },
    {
      "entropy": 5.733256244659424,
      "epoch": 0.3920206431123462,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004990302410725405,
      "loss": 5.6268,
      "mean_token_accuracy": 0.14762071818113326,
      "num_tokens": 8682455.0,
      "step": 3950
    },
    {
      "entropy": 5.769017505645752,
      "epoch": 0.3925168717745137,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004990269522259509,
      "loss": 5.6153,
      "mean_token_accuracy": 0.1415529504418373,
      "num_tokens": 8692975.0,
      "step": 3955
    },
    {
      "entropy": 5.729673624038696,
      "epoch": 0.3930131004366812,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004990236578239755,
      "loss": 5.5754,
      "mean_token_accuracy": 0.14106195271015168,
      "num_tokens": 8703569.0,
      "step": 3960
    },
    {
      "entropy": 5.820702600479126,
      "epoch": 0.3935093290988487,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004990203578666962,
      "loss": 5.6878,
      "mean_token_accuracy": 0.13368871957063674,
      "num_tokens": 8713734.0,
      "step": 3965
    },
    {
      "entropy": 5.783185625076294,
      "epoch": 0.39400555776101626,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004990170523541948,
      "loss": 5.5848,
      "mean_token_accuracy": 0.14641537964344026,
      "num_tokens": 8723852.0,
      "step": 3970
    },
    {
      "entropy": 5.793915796279907,
      "epoch": 0.3945017864231838,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004990137412865531,
      "loss": 5.747,
      "mean_token_accuracy": 0.13510857298970222,
      "num_tokens": 8734300.0,
      "step": 3975
    },
    {
      "entropy": 5.898438024520874,
      "epoch": 0.3949980150853513,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004990104246638533,
      "loss": 5.7356,
      "mean_token_accuracy": 0.1344385825097561,
      "num_tokens": 8745022.0,
      "step": 3980
    },
    {
      "entropy": 5.785170221328736,
      "epoch": 0.39549424374751885,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004990071024861778,
      "loss": 5.6298,
      "mean_token_accuracy": 0.13691987618803977,
      "num_tokens": 8754870.0,
      "step": 3985
    },
    {
      "entropy": 5.872950315475464,
      "epoch": 0.3959904724096864,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004990037747536087,
      "loss": 5.6865,
      "mean_token_accuracy": 0.1410525433719158,
      "num_tokens": 8765536.0,
      "step": 3990
    },
    {
      "entropy": 5.788661527633667,
      "epoch": 0.3964867010718539,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990004414662288,
      "loss": 5.6444,
      "mean_token_accuracy": 0.13316506817936896,
      "num_tokens": 8776898.0,
      "step": 3995
    },
    {
      "entropy": 5.795246744155884,
      "epoch": 0.39698292973402144,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004989971026241205,
      "loss": 5.7092,
      "mean_token_accuracy": 0.1341576285660267,
      "num_tokens": 8788176.0,
      "step": 4000
    },
    {
      "entropy": 5.745181846618652,
      "epoch": 0.39747915839618897,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004989937582273668,
      "loss": 5.6423,
      "mean_token_accuracy": 0.13551107943058013,
      "num_tokens": 8798939.0,
      "step": 4005
    },
    {
      "entropy": 5.790009880065918,
      "epoch": 0.3979753870583565,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004989904082760506,
      "loss": 5.6695,
      "mean_token_accuracy": 0.14157403707504274,
      "num_tokens": 8809092.0,
      "step": 4010
    },
    {
      "entropy": 5.820794916152954,
      "epoch": 0.39847161572052403,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000498987052770255,
      "loss": 5.5906,
      "mean_token_accuracy": 0.14081012904644014,
      "num_tokens": 8819352.0,
      "step": 4015
    },
    {
      "entropy": 5.824670314788818,
      "epoch": 0.39896784438269156,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000498983691710063,
      "loss": 5.7098,
      "mean_token_accuracy": 0.1329125739634037,
      "num_tokens": 8829878.0,
      "step": 4020
    },
    {
      "entropy": 5.778005743026734,
      "epoch": 0.3994640730448591,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000498980325095558,
      "loss": 5.5798,
      "mean_token_accuracy": 0.1393500655889511,
      "num_tokens": 8840364.0,
      "step": 4025
    },
    {
      "entropy": 5.7912201404571535,
      "epoch": 0.3999603017070266,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004989769529268238,
      "loss": 5.6914,
      "mean_token_accuracy": 0.1342613399028778,
      "num_tokens": 8851340.0,
      "step": 4030
    },
    {
      "entropy": 5.750426435470581,
      "epoch": 0.4004565303691941,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004989735752039436,
      "loss": 5.6583,
      "mean_token_accuracy": 0.13800134211778642,
      "num_tokens": 8862571.0,
      "step": 4035
    },
    {
      "entropy": 5.796105480194091,
      "epoch": 0.4009527590313616,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004989701919270015,
      "loss": 5.6605,
      "mean_token_accuracy": 0.1415472649037838,
      "num_tokens": 8873927.0,
      "step": 4040
    },
    {
      "entropy": 5.866248321533203,
      "epoch": 0.40144898769352916,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004989668030960812,
      "loss": 5.8541,
      "mean_token_accuracy": 0.12870026156306266,
      "num_tokens": 8886542.0,
      "step": 4045
    },
    {
      "entropy": 5.886906385421753,
      "epoch": 0.4019452163556967,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004989634087112667,
      "loss": 5.6018,
      "mean_token_accuracy": 0.1396520234644413,
      "num_tokens": 8896114.0,
      "step": 4050
    },
    {
      "entropy": 5.802579689025879,
      "epoch": 0.4024414450178642,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004989600087726425,
      "loss": 5.6846,
      "mean_token_accuracy": 0.1316719926893711,
      "num_tokens": 8906303.0,
      "step": 4055
    },
    {
      "entropy": 5.829805946350097,
      "epoch": 0.40293767368003175,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004989566032802924,
      "loss": 5.7132,
      "mean_token_accuracy": 0.13243434056639672,
      "num_tokens": 8916961.0,
      "step": 4060
    },
    {
      "entropy": 5.879288005828857,
      "epoch": 0.4034339023421993,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004989531922343013,
      "loss": 5.805,
      "mean_token_accuracy": 0.13305630013346673,
      "num_tokens": 8927700.0,
      "step": 4065
    },
    {
      "entropy": 5.819786024093628,
      "epoch": 0.4039301310043668,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004989497756347534,
      "loss": 5.6316,
      "mean_token_accuracy": 0.13920575231313706,
      "num_tokens": 8939655.0,
      "step": 4070
    },
    {
      "entropy": 5.780047941207886,
      "epoch": 0.40442635966653434,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004989463534817339,
      "loss": 5.5705,
      "mean_token_accuracy": 0.14523619860410691,
      "num_tokens": 8951238.0,
      "step": 4075
    },
    {
      "entropy": 5.802511358261109,
      "epoch": 0.40492258832870187,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004989429257753273,
      "loss": 5.6305,
      "mean_token_accuracy": 0.14453877732157708,
      "num_tokens": 8962019.0,
      "step": 4080
    },
    {
      "entropy": 5.7828209400177,
      "epoch": 0.4054188169908694,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004989394925156187,
      "loss": 5.6351,
      "mean_token_accuracy": 0.13892287462949754,
      "num_tokens": 8973861.0,
      "step": 4085
    },
    {
      "entropy": 5.741808176040649,
      "epoch": 0.40591504565303693,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004989360537026932,
      "loss": 5.6414,
      "mean_token_accuracy": 0.14035039395093918,
      "num_tokens": 8985981.0,
      "step": 4090
    },
    {
      "entropy": 5.7918037414550785,
      "epoch": 0.40641127431520446,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000498932609336636,
      "loss": 5.5418,
      "mean_token_accuracy": 0.14218456000089646,
      "num_tokens": 8995318.0,
      "step": 4095
    },
    {
      "entropy": 5.7594521045684814,
      "epoch": 0.406907502977372,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004989291594175327,
      "loss": 5.5765,
      "mean_token_accuracy": 0.1396835669875145,
      "num_tokens": 9004679.0,
      "step": 4100
    },
    {
      "entropy": 5.91703233718872,
      "epoch": 0.4074037316395395,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004989257039454689,
      "loss": 5.7059,
      "mean_token_accuracy": 0.1379046157002449,
      "num_tokens": 9016422.0,
      "step": 4105
    },
    {
      "entropy": 5.724780511856079,
      "epoch": 0.40789996030170705,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00049892224292053,
      "loss": 5.6229,
      "mean_token_accuracy": 0.14078758656978607,
      "num_tokens": 9027914.0,
      "step": 4110
    },
    {
      "entropy": 5.761678218841553,
      "epoch": 0.40839618896387453,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004989187763428021,
      "loss": 5.5918,
      "mean_token_accuracy": 0.144223216176033,
      "num_tokens": 9038701.0,
      "step": 4115
    },
    {
      "entropy": 5.768312168121338,
      "epoch": 0.40889241762604206,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000498915304212371,
      "loss": 5.6561,
      "mean_token_accuracy": 0.13807908743619918,
      "num_tokens": 9050620.0,
      "step": 4120
    },
    {
      "entropy": 5.846501874923706,
      "epoch": 0.4093886462882096,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004989118265293228,
      "loss": 5.6829,
      "mean_token_accuracy": 0.14339982569217682,
      "num_tokens": 9060900.0,
      "step": 4125
    },
    {
      "entropy": 5.792847585678101,
      "epoch": 0.4098848749503771,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004989083432937439,
      "loss": 5.5866,
      "mean_token_accuracy": 0.13959128633141518,
      "num_tokens": 9072122.0,
      "step": 4130
    },
    {
      "entropy": 5.749606561660767,
      "epoch": 0.41038110361254465,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004989048545057204,
      "loss": 5.617,
      "mean_token_accuracy": 0.13930945098400116,
      "num_tokens": 9082780.0,
      "step": 4135
    },
    {
      "entropy": 5.782565975189209,
      "epoch": 0.4108773322747122,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004989013601653391,
      "loss": 5.6658,
      "mean_token_accuracy": 0.1379122644662857,
      "num_tokens": 9093518.0,
      "step": 4140
    },
    {
      "entropy": 5.767605924606324,
      "epoch": 0.4113735609368797,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004988978602726866,
      "loss": 5.6779,
      "mean_token_accuracy": 0.1437633991241455,
      "num_tokens": 9105035.0,
      "step": 4145
    },
    {
      "entropy": 5.812577152252198,
      "epoch": 0.41186978959904724,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004988943548278494,
      "loss": 5.587,
      "mean_token_accuracy": 0.14110235944390298,
      "num_tokens": 9117154.0,
      "step": 4150
    },
    {
      "entropy": 5.77940182685852,
      "epoch": 0.4123660182612148,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004988908438309149,
      "loss": 5.6667,
      "mean_token_accuracy": 0.13823975771665573,
      "num_tokens": 9128818.0,
      "step": 4155
    },
    {
      "entropy": 5.8112109184265135,
      "epoch": 0.4128622469233823,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004988873272819697,
      "loss": 5.67,
      "mean_token_accuracy": 0.1413659855723381,
      "num_tokens": 9140188.0,
      "step": 4160
    },
    {
      "entropy": 5.94907431602478,
      "epoch": 0.41335847558554983,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004988838051811014,
      "loss": 5.7523,
      "mean_token_accuracy": 0.134553412348032,
      "num_tokens": 9150565.0,
      "step": 4165
    },
    {
      "entropy": 5.801510667800903,
      "epoch": 0.41385470424771736,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000498880277528397,
      "loss": 5.6054,
      "mean_token_accuracy": 0.13975557759404184,
      "num_tokens": 9160130.0,
      "step": 4170
    },
    {
      "entropy": 5.799229383468628,
      "epoch": 0.4143509329098849,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004988767443239443,
      "loss": 5.6117,
      "mean_token_accuracy": 0.1398836188018322,
      "num_tokens": 9171740.0,
      "step": 4175
    },
    {
      "entropy": 5.76237120628357,
      "epoch": 0.4148471615720524,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004988732055678307,
      "loss": 5.5544,
      "mean_token_accuracy": 0.15382059216499328,
      "num_tokens": 9182459.0,
      "step": 4180
    },
    {
      "entropy": 5.805677366256714,
      "epoch": 0.41534339023421996,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000498869661260144,
      "loss": 5.6322,
      "mean_token_accuracy": 0.14206244200468063,
      "num_tokens": 9192284.0,
      "step": 4185
    },
    {
      "entropy": 5.780238819122315,
      "epoch": 0.41583961889638743,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000498866111400972,
      "loss": 5.6115,
      "mean_token_accuracy": 0.1399123787879944,
      "num_tokens": 9202700.0,
      "step": 4190
    },
    {
      "entropy": 5.7953057289123535,
      "epoch": 0.41633584755855496,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004988625559904029,
      "loss": 5.6243,
      "mean_token_accuracy": 0.13381586745381355,
      "num_tokens": 9213013.0,
      "step": 4195
    },
    {
      "entropy": 5.83180456161499,
      "epoch": 0.4168320762207225,
      "grad_norm": 0.7421875,
      "learning_rate": 0.0004988589950285248,
      "loss": 5.6966,
      "mean_token_accuracy": 0.13726072534918785,
      "num_tokens": 9224568.0,
      "step": 4200
    },
    {
      "entropy": 5.76538610458374,
      "epoch": 0.41732830488289,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004988554285154259,
      "loss": 5.5748,
      "mean_token_accuracy": 0.13923351615667343,
      "num_tokens": 9233791.0,
      "step": 4205
    },
    {
      "entropy": 5.792706155776978,
      "epoch": 0.41782453354505755,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004988518564511948,
      "loss": 5.6178,
      "mean_token_accuracy": 0.13635936975479127,
      "num_tokens": 9245048.0,
      "step": 4210
    },
    {
      "entropy": 5.860803031921387,
      "epoch": 0.4183207622072251,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004988482788359202,
      "loss": 5.6396,
      "mean_token_accuracy": 0.14628660529851914,
      "num_tokens": 9256508.0,
      "step": 4215
    },
    {
      "entropy": 5.803260707855225,
      "epoch": 0.4188169908693926,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004988446956696904,
      "loss": 5.668,
      "mean_token_accuracy": 0.14071763530373574,
      "num_tokens": 9267871.0,
      "step": 4220
    },
    {
      "entropy": 5.85851092338562,
      "epoch": 0.41931321953156014,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004988411069525947,
      "loss": 5.7168,
      "mean_token_accuracy": 0.13775873184204102,
      "num_tokens": 9279287.0,
      "step": 4225
    },
    {
      "entropy": 5.862564992904663,
      "epoch": 0.4198094481937277,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004988375126847216,
      "loss": 5.7249,
      "mean_token_accuracy": 0.13378578275442124,
      "num_tokens": 9291257.0,
      "step": 4230
    },
    {
      "entropy": 5.840371370315552,
      "epoch": 0.4203056768558952,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004988339128661607,
      "loss": 5.7021,
      "mean_token_accuracy": 0.1337619088590145,
      "num_tokens": 9302590.0,
      "step": 4235
    },
    {
      "entropy": 5.699577569961548,
      "epoch": 0.42080190551806274,
      "grad_norm": 0.80859375,
      "learning_rate": 0.000498830307497001,
      "loss": 5.5512,
      "mean_token_accuracy": 0.14313922077417374,
      "num_tokens": 9313174.0,
      "step": 4240
    },
    {
      "entropy": 5.794048738479614,
      "epoch": 0.42129813418023027,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000498826696577332,
      "loss": 5.6396,
      "mean_token_accuracy": 0.1417992301285267,
      "num_tokens": 9325034.0,
      "step": 4245
    },
    {
      "entropy": 5.806614542007447,
      "epoch": 0.4217943628423978,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004988230801072432,
      "loss": 5.6206,
      "mean_token_accuracy": 0.13834859877824784,
      "num_tokens": 9336834.0,
      "step": 4250
    },
    {
      "entropy": 5.737400197982788,
      "epoch": 0.4222905915045653,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004988194580868244,
      "loss": 5.6237,
      "mean_token_accuracy": 0.13811139315366744,
      "num_tokens": 9347747.0,
      "step": 4255
    },
    {
      "entropy": 5.771370553970337,
      "epoch": 0.42278682016673286,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004988158305161654,
      "loss": 5.6391,
      "mean_token_accuracy": 0.1432290256023407,
      "num_tokens": 9359186.0,
      "step": 4260
    },
    {
      "entropy": 5.782405090332031,
      "epoch": 0.42328304882890033,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000498812197395356,
      "loss": 5.5621,
      "mean_token_accuracy": 0.14198907613754272,
      "num_tokens": 9369485.0,
      "step": 4265
    },
    {
      "entropy": 5.782733201980591,
      "epoch": 0.42377927749106786,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004988085587244863,
      "loss": 5.6577,
      "mean_token_accuracy": 0.13524223864078522,
      "num_tokens": 9382658.0,
      "step": 4270
    },
    {
      "entropy": 5.74179515838623,
      "epoch": 0.4242755061532354,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004988049145036466,
      "loss": 5.6175,
      "mean_token_accuracy": 0.1354674480855465,
      "num_tokens": 9394151.0,
      "step": 4275
    },
    {
      "entropy": 5.861121129989624,
      "epoch": 0.4247717348154029,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004988012647329273,
      "loss": 5.7112,
      "mean_token_accuracy": 0.13729124367237092,
      "num_tokens": 9405956.0,
      "step": 4280
    },
    {
      "entropy": 5.821110439300537,
      "epoch": 0.42526796347757045,
      "grad_norm": 0.90625,
      "learning_rate": 0.000498797609412419,
      "loss": 5.6038,
      "mean_token_accuracy": 0.13974196910858155,
      "num_tokens": 9416889.0,
      "step": 4285
    },
    {
      "entropy": 5.7748674869537355,
      "epoch": 0.425764192139738,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004987939485422122,
      "loss": 5.7055,
      "mean_token_accuracy": 0.136608724296093,
      "num_tokens": 9427602.0,
      "step": 4290
    },
    {
      "entropy": 5.823736667633057,
      "epoch": 0.4262604208019055,
      "grad_norm": 0.7578125,
      "learning_rate": 0.0004987902821223976,
      "loss": 5.629,
      "mean_token_accuracy": 0.14252188354730605,
      "num_tokens": 9438557.0,
      "step": 4295
    },
    {
      "entropy": 5.777904653549195,
      "epoch": 0.42675664946407305,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004987866101530663,
      "loss": 5.667,
      "mean_token_accuracy": 0.13800618946552276,
      "num_tokens": 9449598.0,
      "step": 4300
    },
    {
      "entropy": 5.8158505916595455,
      "epoch": 0.4272528781262406,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004987829326343092,
      "loss": 5.6076,
      "mean_token_accuracy": 0.14416556805372238,
      "num_tokens": 9459740.0,
      "step": 4305
    },
    {
      "entropy": 5.753842973709107,
      "epoch": 0.4277491067884081,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004987792495662177,
      "loss": 5.6073,
      "mean_token_accuracy": 0.14074130430817605,
      "num_tokens": 9469835.0,
      "step": 4310
    },
    {
      "entropy": 5.818010139465332,
      "epoch": 0.42824533545057564,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000498775560948883,
      "loss": 5.6917,
      "mean_token_accuracy": 0.13898994252085686,
      "num_tokens": 9480735.0,
      "step": 4315
    },
    {
      "entropy": 5.785709524154663,
      "epoch": 0.42874156411274317,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004987718667823965,
      "loss": 5.5887,
      "mean_token_accuracy": 0.14314745515584945,
      "num_tokens": 9491321.0,
      "step": 4320
    },
    {
      "entropy": 5.817282676696777,
      "epoch": 0.4292377927749107,
      "grad_norm": 0.78125,
      "learning_rate": 0.00049876816706685,
      "loss": 5.721,
      "mean_token_accuracy": 0.13546350300312043,
      "num_tokens": 9503507.0,
      "step": 4325
    },
    {
      "entropy": 5.835805463790893,
      "epoch": 0.42973402143707823,
      "grad_norm": 1.6953125,
      "learning_rate": 0.000498764461802335,
      "loss": 5.7267,
      "mean_token_accuracy": 0.14347420930862426,
      "num_tokens": 9514404.0,
      "step": 4330
    },
    {
      "entropy": 5.736133670806884,
      "epoch": 0.43023025009924576,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004987607509889437,
      "loss": 5.5945,
      "mean_token_accuracy": 0.13721355572342872,
      "num_tokens": 9525555.0,
      "step": 4335
    },
    {
      "entropy": 5.844026803970337,
      "epoch": 0.43072647876141323,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004987570346267677,
      "loss": 5.593,
      "mean_token_accuracy": 0.14471945762634278,
      "num_tokens": 9535084.0,
      "step": 4340
    },
    {
      "entropy": 5.716549587249756,
      "epoch": 0.43122270742358076,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004987533127158996,
      "loss": 5.5996,
      "mean_token_accuracy": 0.1444016769528389,
      "num_tokens": 9546170.0,
      "step": 4345
    },
    {
      "entropy": 5.811110973358154,
      "epoch": 0.4317189360857483,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004987495852564314,
      "loss": 5.6225,
      "mean_token_accuracy": 0.14201515763998032,
      "num_tokens": 9556793.0,
      "step": 4350
    },
    {
      "entropy": 5.784374189376831,
      "epoch": 0.4322151647479158,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004987458522484556,
      "loss": 5.566,
      "mean_token_accuracy": 0.14575942158699035,
      "num_tokens": 9568114.0,
      "step": 4355
    },
    {
      "entropy": 5.740352964401245,
      "epoch": 0.43271139341008336,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004987421136920648,
      "loss": 5.5771,
      "mean_token_accuracy": 0.13610847890377045,
      "num_tokens": 9579119.0,
      "step": 4360
    },
    {
      "entropy": 5.722053813934326,
      "epoch": 0.4332076220722509,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004987383695873517,
      "loss": 5.6147,
      "mean_token_accuracy": 0.14022180810570717,
      "num_tokens": 9590259.0,
      "step": 4365
    },
    {
      "entropy": 5.802224493026733,
      "epoch": 0.4337038507344184,
      "grad_norm": 0.875,
      "learning_rate": 0.0004987346199344092,
      "loss": 5.6512,
      "mean_token_accuracy": 0.13241124674677848,
      "num_tokens": 9600431.0,
      "step": 4370
    },
    {
      "entropy": 5.747055196762085,
      "epoch": 0.43420007939658595,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004987308647333303,
      "loss": 5.6298,
      "mean_token_accuracy": 0.14093587100505828,
      "num_tokens": 9610286.0,
      "step": 4375
    },
    {
      "entropy": 5.844609880447388,
      "epoch": 0.4346963080587535,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004987271039842079,
      "loss": 5.6392,
      "mean_token_accuracy": 0.14039682298898698,
      "num_tokens": 9621148.0,
      "step": 4380
    },
    {
      "entropy": 5.768031358718872,
      "epoch": 0.435192536720921,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004987233376871355,
      "loss": 5.5234,
      "mean_token_accuracy": 0.1461140587925911,
      "num_tokens": 9631300.0,
      "step": 4385
    },
    {
      "entropy": 5.78350191116333,
      "epoch": 0.43568876538308854,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004987195658422063,
      "loss": 5.6269,
      "mean_token_accuracy": 0.13842125460505486,
      "num_tokens": 9641986.0,
      "step": 4390
    },
    {
      "entropy": 5.781039094924926,
      "epoch": 0.43618499404525607,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004987157884495141,
      "loss": 5.6395,
      "mean_token_accuracy": 0.1402071826159954,
      "num_tokens": 9652962.0,
      "step": 4395
    },
    {
      "entropy": 5.757225465774536,
      "epoch": 0.4366812227074236,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004987120055091523,
      "loss": 5.6788,
      "mean_token_accuracy": 0.13589339554309846,
      "num_tokens": 9664084.0,
      "step": 4400
    },
    {
      "entropy": 5.754967546463012,
      "epoch": 0.43717745136959113,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004987082170212149,
      "loss": 5.5724,
      "mean_token_accuracy": 0.1397831603884697,
      "num_tokens": 9675821.0,
      "step": 4405
    },
    {
      "entropy": 5.774995422363281,
      "epoch": 0.43767368003175866,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004987044229857957,
      "loss": 5.6874,
      "mean_token_accuracy": 0.1395527347922325,
      "num_tokens": 9687964.0,
      "step": 4410
    },
    {
      "entropy": 5.788907766342163,
      "epoch": 0.43816990869392614,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004987006234029889,
      "loss": 5.6718,
      "mean_token_accuracy": 0.139390766620636,
      "num_tokens": 9699653.0,
      "step": 4415
    },
    {
      "entropy": 5.857262229919433,
      "epoch": 0.43866613735609367,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004986968182728885,
      "loss": 5.6781,
      "mean_token_accuracy": 0.13918035179376603,
      "num_tokens": 9709872.0,
      "step": 4420
    },
    {
      "entropy": 5.817832374572754,
      "epoch": 0.4391623660182612,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004986930075955892,
      "loss": 5.601,
      "mean_token_accuracy": 0.13685947135090828,
      "num_tokens": 9720589.0,
      "step": 4425
    },
    {
      "entropy": 5.700802898406982,
      "epoch": 0.4396585946804287,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004986891913711853,
      "loss": 5.5351,
      "mean_token_accuracy": 0.14478714317083358,
      "num_tokens": 9731862.0,
      "step": 4430
    },
    {
      "entropy": 5.73977108001709,
      "epoch": 0.44015482334259626,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004986853695997715,
      "loss": 5.6872,
      "mean_token_accuracy": 0.14369284883141517,
      "num_tokens": 9743710.0,
      "step": 4435
    },
    {
      "entropy": 5.799801588058472,
      "epoch": 0.4406510520047638,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004986815422814425,
      "loss": 5.6418,
      "mean_token_accuracy": 0.13150426223874093,
      "num_tokens": 9753855.0,
      "step": 4440
    },
    {
      "entropy": 5.875456953048706,
      "epoch": 0.4411472806669313,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004986777094162932,
      "loss": 5.6379,
      "mean_token_accuracy": 0.14705183878540992,
      "num_tokens": 9763883.0,
      "step": 4445
    },
    {
      "entropy": 5.748485708236695,
      "epoch": 0.44164350932909885,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004986738710044187,
      "loss": 5.5989,
      "mean_token_accuracy": 0.14280258789658545,
      "num_tokens": 9774890.0,
      "step": 4450
    },
    {
      "entropy": 5.732300472259522,
      "epoch": 0.4421397379912664,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004986700270459142,
      "loss": 5.5783,
      "mean_token_accuracy": 0.1434408903121948,
      "num_tokens": 9785249.0,
      "step": 4455
    },
    {
      "entropy": 5.791567468643189,
      "epoch": 0.4426359666534339,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004986661775408749,
      "loss": 5.5913,
      "mean_token_accuracy": 0.14472425431013108,
      "num_tokens": 9795723.0,
      "step": 4460
    },
    {
      "entropy": 5.7759202003479,
      "epoch": 0.44313219531560144,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004986623224893965,
      "loss": 5.5394,
      "mean_token_accuracy": 0.1415581427514553,
      "num_tokens": 9807637.0,
      "step": 4465
    },
    {
      "entropy": 5.784197425842285,
      "epoch": 0.44362842397776897,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004986584618915744,
      "loss": 5.648,
      "mean_token_accuracy": 0.1421603761613369,
      "num_tokens": 9820298.0,
      "step": 4470
    },
    {
      "entropy": 5.709079837799072,
      "epoch": 0.4441246526399365,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004986545957475045,
      "loss": 5.5383,
      "mean_token_accuracy": 0.14711108356714248,
      "num_tokens": 9831251.0,
      "step": 4475
    },
    {
      "entropy": 5.734958457946777,
      "epoch": 0.44462088130210403,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004986507240572825,
      "loss": 5.596,
      "mean_token_accuracy": 0.14287913516163825,
      "num_tokens": 9842884.0,
      "step": 4480
    },
    {
      "entropy": 5.72746171951294,
      "epoch": 0.44511710996427156,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004986468468210042,
      "loss": 5.4933,
      "mean_token_accuracy": 0.14691198095679284,
      "num_tokens": 9854413.0,
      "step": 4485
    },
    {
      "entropy": 5.793558883666992,
      "epoch": 0.44561333862643904,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004986429640387663,
      "loss": 5.6508,
      "mean_token_accuracy": 0.1404290720820427,
      "num_tokens": 9865771.0,
      "step": 4490
    },
    {
      "entropy": 5.766204071044922,
      "epoch": 0.44610956728860657,
      "grad_norm": 0.7734375,
      "learning_rate": 0.0004986390757106648,
      "loss": 5.5707,
      "mean_token_accuracy": 0.1469174288213253,
      "num_tokens": 9876717.0,
      "step": 4495
    },
    {
      "entropy": 5.730377721786499,
      "epoch": 0.4466057959507741,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004986351818367959,
      "loss": 5.6486,
      "mean_token_accuracy": 0.140429724752903,
      "num_tokens": 9887616.0,
      "step": 4500
    },
    {
      "entropy": 5.818638896942138,
      "epoch": 0.44710202461294163,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004986312824172566,
      "loss": 5.595,
      "mean_token_accuracy": 0.14335185587406157,
      "num_tokens": 9898979.0,
      "step": 4505
    },
    {
      "entropy": 5.797679376602173,
      "epoch": 0.44759825327510916,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0004986273774521433,
      "loss": 5.6306,
      "mean_token_accuracy": 0.13533177375793456,
      "num_tokens": 9912158.0,
      "step": 4510
    },
    {
      "entropy": 5.7367393493652346,
      "epoch": 0.4480944819372767,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004986234669415529,
      "loss": 5.6312,
      "mean_token_accuracy": 0.13736921548843384,
      "num_tokens": 9923886.0,
      "step": 4515
    },
    {
      "entropy": 5.776280212402344,
      "epoch": 0.4485907105994442,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004986195508855824,
      "loss": 5.6023,
      "mean_token_accuracy": 0.14297038465738296,
      "num_tokens": 9934979.0,
      "step": 4520
    },
    {
      "entropy": 5.79187970161438,
      "epoch": 0.44908693926161175,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000498615629284329,
      "loss": 5.7071,
      "mean_token_accuracy": 0.13817890509963035,
      "num_tokens": 9944715.0,
      "step": 4525
    },
    {
      "entropy": 5.775436878204346,
      "epoch": 0.4495831679237793,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004986117021378896,
      "loss": 5.5741,
      "mean_token_accuracy": 0.14012717232108116,
      "num_tokens": 9954736.0,
      "step": 4530
    },
    {
      "entropy": 5.79448766708374,
      "epoch": 0.4500793965859468,
      "grad_norm": 0.75,
      "learning_rate": 0.000498607769446362,
      "loss": 5.6594,
      "mean_token_accuracy": 0.1388283774256706,
      "num_tokens": 9965952.0,
      "step": 4535
    },
    {
      "entropy": 5.7624434471130375,
      "epoch": 0.45057562524811434,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004986038312098435,
      "loss": 5.604,
      "mean_token_accuracy": 0.14314566776156426,
      "num_tokens": 9976428.0,
      "step": 4540
    },
    {
      "entropy": 5.75066294670105,
      "epoch": 0.4510718539102819,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004985998874284319,
      "loss": 5.546,
      "mean_token_accuracy": 0.1484223261475563,
      "num_tokens": 9986100.0,
      "step": 4545
    },
    {
      "entropy": 5.655505847930908,
      "epoch": 0.4515680825724494,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004985959381022248,
      "loss": 5.5458,
      "mean_token_accuracy": 0.14303347319364548,
      "num_tokens": 9996584.0,
      "step": 4550
    },
    {
      "entropy": 5.739107084274292,
      "epoch": 0.45206431123461693,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004985919832313203,
      "loss": 5.557,
      "mean_token_accuracy": 0.14798323214054107,
      "num_tokens": 10007792.0,
      "step": 4555
    },
    {
      "entropy": 5.851394081115723,
      "epoch": 0.45256053989678446,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004985880228158164,
      "loss": 5.756,
      "mean_token_accuracy": 0.13254465758800507,
      "num_tokens": 10019615.0,
      "step": 4560
    },
    {
      "entropy": 5.697079610824585,
      "epoch": 0.45305676855895194,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004985840568558112,
      "loss": 5.5814,
      "mean_token_accuracy": 0.14153416082262993,
      "num_tokens": 10030223.0,
      "step": 4565
    },
    {
      "entropy": 5.775788831710815,
      "epoch": 0.45355299722111947,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004985800853514033,
      "loss": 5.5967,
      "mean_token_accuracy": 0.137832148373127,
      "num_tokens": 10041522.0,
      "step": 4570
    },
    {
      "entropy": 5.785793733596802,
      "epoch": 0.454049225883287,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004985761083026911,
      "loss": 5.5212,
      "mean_token_accuracy": 0.15029784739017488,
      "num_tokens": 10052847.0,
      "step": 4575
    },
    {
      "entropy": 5.655645513534546,
      "epoch": 0.45454545454545453,
      "grad_norm": 0.875,
      "learning_rate": 0.0004985721257097731,
      "loss": 5.5434,
      "mean_token_accuracy": 0.14386311173439026,
      "num_tokens": 10062259.0,
      "step": 4580
    },
    {
      "entropy": 5.820441770553589,
      "epoch": 0.45504168320762206,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004985681375727482,
      "loss": 5.7118,
      "mean_token_accuracy": 0.13377906680107116,
      "num_tokens": 10073341.0,
      "step": 4585
    },
    {
      "entropy": 5.822879838943481,
      "epoch": 0.4555379118697896,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0004985641438917151,
      "loss": 5.5629,
      "mean_token_accuracy": 0.14327463507652283,
      "num_tokens": 10083992.0,
      "step": 4590
    },
    {
      "entropy": 5.765067672729492,
      "epoch": 0.4560341405319571,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004985601446667731,
      "loss": 5.5608,
      "mean_token_accuracy": 0.14663187712430953,
      "num_tokens": 10094245.0,
      "step": 4595
    },
    {
      "entropy": 5.7631641864776615,
      "epoch": 0.45653036919412465,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004985561398980213,
      "loss": 5.5366,
      "mean_token_accuracy": 0.14789687395095824,
      "num_tokens": 10105976.0,
      "step": 4600
    },
    {
      "entropy": 5.672163963317871,
      "epoch": 0.4570265978562922,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004985521295855588,
      "loss": 5.6168,
      "mean_token_accuracy": 0.14210953339934349,
      "num_tokens": 10117373.0,
      "step": 4605
    },
    {
      "entropy": 5.754079294204712,
      "epoch": 0.4575228265184597,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004985481137294853,
      "loss": 5.5685,
      "mean_token_accuracy": 0.13775459825992584,
      "num_tokens": 10128470.0,
      "step": 4610
    },
    {
      "entropy": 5.714517211914062,
      "epoch": 0.45801905518062724,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004985440923299002,
      "loss": 5.5662,
      "mean_token_accuracy": 0.1423081412911415,
      "num_tokens": 10140217.0,
      "step": 4615
    },
    {
      "entropy": 5.710232543945312,
      "epoch": 0.4585152838427948,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004985400653869034,
      "loss": 5.633,
      "mean_token_accuracy": 0.14135775342583656,
      "num_tokens": 10152308.0,
      "step": 4620
    },
    {
      "entropy": 5.8700785636901855,
      "epoch": 0.4590115125049623,
      "grad_norm": 0.875,
      "learning_rate": 0.0004985360329005945,
      "loss": 5.6582,
      "mean_token_accuracy": 0.14096900820732117,
      "num_tokens": 10164234.0,
      "step": 4625
    },
    {
      "entropy": 5.874130821228027,
      "epoch": 0.45950774116712984,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004985319948710738,
      "loss": 5.7163,
      "mean_token_accuracy": 0.13750708624720573,
      "num_tokens": 10175287.0,
      "step": 4630
    },
    {
      "entropy": 5.801472568511963,
      "epoch": 0.46000396982929737,
      "grad_norm": 1.0,
      "learning_rate": 0.0004985279512984413,
      "loss": 5.5875,
      "mean_token_accuracy": 0.14970208927989007,
      "num_tokens": 10184658.0,
      "step": 4635
    },
    {
      "entropy": 5.737123918533325,
      "epoch": 0.46050019849146484,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004985239021827972,
      "loss": 5.5849,
      "mean_token_accuracy": 0.1390370771288872,
      "num_tokens": 10194593.0,
      "step": 4640
    },
    {
      "entropy": 5.701331233978271,
      "epoch": 0.46099642715363237,
      "grad_norm": 0.79296875,
      "learning_rate": 0.000498519847524242,
      "loss": 5.5092,
      "mean_token_accuracy": 0.1458827719092369,
      "num_tokens": 10205200.0,
      "step": 4645
    },
    {
      "entropy": 5.724951648712159,
      "epoch": 0.4614926558157999,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004985157873228762,
      "loss": 5.5194,
      "mean_token_accuracy": 0.14664195030927657,
      "num_tokens": 10216848.0,
      "step": 4650
    },
    {
      "entropy": 5.757262468338013,
      "epoch": 0.46198888447796743,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004985117215788005,
      "loss": 5.5703,
      "mean_token_accuracy": 0.14250332042574881,
      "num_tokens": 10228188.0,
      "step": 4655
    },
    {
      "entropy": 5.751273918151855,
      "epoch": 0.46248511314013496,
      "grad_norm": 0.875,
      "learning_rate": 0.0004985076502921158,
      "loss": 5.659,
      "mean_token_accuracy": 0.1386226199567318,
      "num_tokens": 10238253.0,
      "step": 4660
    },
    {
      "entropy": 5.799253034591675,
      "epoch": 0.4629813418023025,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000498503573462923,
      "loss": 5.604,
      "mean_token_accuracy": 0.14217669069766997,
      "num_tokens": 10248169.0,
      "step": 4665
    },
    {
      "entropy": 5.719962024688721,
      "epoch": 0.46347757046447,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004984994910913232,
      "loss": 5.4916,
      "mean_token_accuracy": 0.14536735266447068,
      "num_tokens": 10258962.0,
      "step": 4670
    },
    {
      "entropy": 5.754688787460327,
      "epoch": 0.46397379912663755,
      "grad_norm": 0.765625,
      "learning_rate": 0.0004984954031774175,
      "loss": 5.5963,
      "mean_token_accuracy": 0.14113237708806992,
      "num_tokens": 10270303.0,
      "step": 4675
    },
    {
      "entropy": 5.649554014205933,
      "epoch": 0.4644700277888051,
      "grad_norm": 0.7421875,
      "learning_rate": 0.0004984913097213075,
      "loss": 5.5111,
      "mean_token_accuracy": 0.1448366239666939,
      "num_tokens": 10281782.0,
      "step": 4680
    },
    {
      "entropy": 5.7514485836029055,
      "epoch": 0.4649662564509726,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004984872107230945,
      "loss": 5.5542,
      "mean_token_accuracy": 0.14501627534627914,
      "num_tokens": 10292937.0,
      "step": 4685
    },
    {
      "entropy": 5.758346080780029,
      "epoch": 0.46546248511314015,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004984831061828803,
      "loss": 5.606,
      "mean_token_accuracy": 0.13853585198521615,
      "num_tokens": 10305443.0,
      "step": 4690
    },
    {
      "entropy": 5.717559385299682,
      "epoch": 0.4659587137753077,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004984789961007667,
      "loss": 5.5427,
      "mean_token_accuracy": 0.14828034415841101,
      "num_tokens": 10316489.0,
      "step": 4695
    },
    {
      "entropy": 5.701294040679931,
      "epoch": 0.4664549424374752,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004984748804768554,
      "loss": 5.5884,
      "mean_token_accuracy": 0.1406767785549164,
      "num_tokens": 10326690.0,
      "step": 4700
    },
    {
      "entropy": 5.7987219333648685,
      "epoch": 0.46695117109964274,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004984707593112487,
      "loss": 5.5973,
      "mean_token_accuracy": 0.13755468428134918,
      "num_tokens": 10337059.0,
      "step": 4705
    },
    {
      "entropy": 5.678113794326782,
      "epoch": 0.46744739976181027,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004984666326040487,
      "loss": 5.4636,
      "mean_token_accuracy": 0.15040105506777762,
      "num_tokens": 10347035.0,
      "step": 4710
    },
    {
      "entropy": 5.685402202606201,
      "epoch": 0.46794362842397774,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004984625003553578,
      "loss": 5.5931,
      "mean_token_accuracy": 0.14260579720139505,
      "num_tokens": 10359492.0,
      "step": 4715
    },
    {
      "entropy": 5.814010429382324,
      "epoch": 0.4684398570861453,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004984583625652783,
      "loss": 5.573,
      "mean_token_accuracy": 0.1426217183470726,
      "num_tokens": 10370128.0,
      "step": 4720
    },
    {
      "entropy": 5.729959917068482,
      "epoch": 0.4689360857483128,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000498454219233913,
      "loss": 5.5762,
      "mean_token_accuracy": 0.13988015651702881,
      "num_tokens": 10380276.0,
      "step": 4725
    },
    {
      "entropy": 5.6514264106750485,
      "epoch": 0.46943231441048033,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004984500703613645,
      "loss": 5.5602,
      "mean_token_accuracy": 0.1456937611103058,
      "num_tokens": 10392489.0,
      "step": 4730
    },
    {
      "entropy": 5.83931975364685,
      "epoch": 0.46992854307264786,
      "grad_norm": 0.7421875,
      "learning_rate": 0.0004984459159477358,
      "loss": 5.6432,
      "mean_token_accuracy": 0.13943550512194633,
      "num_tokens": 10404259.0,
      "step": 4735
    },
    {
      "entropy": 5.690709781646729,
      "epoch": 0.4704247717348154,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004984417559931298,
      "loss": 5.5047,
      "mean_token_accuracy": 0.14229437559843064,
      "num_tokens": 10415000.0,
      "step": 4740
    },
    {
      "entropy": 5.74327278137207,
      "epoch": 0.4709210003969829,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004984375904976498,
      "loss": 5.6375,
      "mean_token_accuracy": 0.13753386437892914,
      "num_tokens": 10426078.0,
      "step": 4745
    },
    {
      "entropy": 5.792974853515625,
      "epoch": 0.47141722905915046,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004984334194613989,
      "loss": 5.5616,
      "mean_token_accuracy": 0.14064555689692498,
      "num_tokens": 10436838.0,
      "step": 4750
    },
    {
      "entropy": 5.712934160232544,
      "epoch": 0.471913457721318,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004984292428844809,
      "loss": 5.5534,
      "mean_token_accuracy": 0.1431037314236164,
      "num_tokens": 10448544.0,
      "step": 4755
    },
    {
      "entropy": 5.819026613235474,
      "epoch": 0.4724096863834855,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004984250607669989,
      "loss": 5.6213,
      "mean_token_accuracy": 0.1387665741145611,
      "num_tokens": 10459730.0,
      "step": 4760
    },
    {
      "entropy": 5.8150371551513675,
      "epoch": 0.47290591504565305,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004984208731090569,
      "loss": 5.6797,
      "mean_token_accuracy": 0.14077720865607263,
      "num_tokens": 10470257.0,
      "step": 4765
    },
    {
      "entropy": 5.776945304870606,
      "epoch": 0.4734021437078206,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004984166799107587,
      "loss": 5.6371,
      "mean_token_accuracy": 0.14257957339286803,
      "num_tokens": 10480835.0,
      "step": 4770
    },
    {
      "entropy": 5.759775972366333,
      "epoch": 0.4738983723699881,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004984124811722081,
      "loss": 5.6488,
      "mean_token_accuracy": 0.14650474786758422,
      "num_tokens": 10491720.0,
      "step": 4775
    },
    {
      "entropy": 5.753606271743775,
      "epoch": 0.47439460103215564,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004984082768935096,
      "loss": 5.6651,
      "mean_token_accuracy": 0.14373267889022828,
      "num_tokens": 10502313.0,
      "step": 4780
    },
    {
      "entropy": 5.78562560081482,
      "epoch": 0.47489082969432317,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004984040670747672,
      "loss": 5.54,
      "mean_token_accuracy": 0.14707302302122116,
      "num_tokens": 10513333.0,
      "step": 4785
    },
    {
      "entropy": 5.711219787597656,
      "epoch": 0.47538705835649064,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004983998517160852,
      "loss": 5.5278,
      "mean_token_accuracy": 0.15109342858195304,
      "num_tokens": 10522766.0,
      "step": 4790
    },
    {
      "entropy": 5.755279207229615,
      "epoch": 0.4758832870186582,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004983956308175684,
      "loss": 5.5675,
      "mean_token_accuracy": 0.14042461812496185,
      "num_tokens": 10534362.0,
      "step": 4795
    },
    {
      "entropy": 5.7386070728302006,
      "epoch": 0.4763795156808257,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004983914043793212,
      "loss": 5.5811,
      "mean_token_accuracy": 0.14227042123675346,
      "num_tokens": 10545072.0,
      "step": 4800
    },
    {
      "entropy": 5.673999500274658,
      "epoch": 0.47687574434299324,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004983871724014487,
      "loss": 5.5169,
      "mean_token_accuracy": 0.14101085662841797,
      "num_tokens": 10555953.0,
      "step": 4805
    },
    {
      "entropy": 5.778817129135132,
      "epoch": 0.47737197300516077,
      "grad_norm": 0.77734375,
      "learning_rate": 0.0004983829348840556,
      "loss": 5.5675,
      "mean_token_accuracy": 0.14799523502588272,
      "num_tokens": 10567757.0,
      "step": 4810
    },
    {
      "entropy": 5.668074464797973,
      "epoch": 0.4778682016673283,
      "grad_norm": 0.82421875,
      "learning_rate": 0.000498378691827247,
      "loss": 5.4701,
      "mean_token_accuracy": 0.1479513630270958,
      "num_tokens": 10579390.0,
      "step": 4815
    },
    {
      "entropy": 5.742642593383789,
      "epoch": 0.4783644303294958,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004983744432311282,
      "loss": 5.6669,
      "mean_token_accuracy": 0.13707347735762596,
      "num_tokens": 10589624.0,
      "step": 4820
    },
    {
      "entropy": 5.829803085327148,
      "epoch": 0.47886065899166336,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004983701890958046,
      "loss": 5.6361,
      "mean_token_accuracy": 0.13275278136134147,
      "num_tokens": 10601927.0,
      "step": 4825
    },
    {
      "entropy": 5.667952632904052,
      "epoch": 0.4793568876538309,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004983659294213815,
      "loss": 5.5147,
      "mean_token_accuracy": 0.1479553312063217,
      "num_tokens": 10614073.0,
      "step": 4830
    },
    {
      "entropy": 5.720580911636352,
      "epoch": 0.4798531163159984,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004983616642079649,
      "loss": 5.5511,
      "mean_token_accuracy": 0.1447890058159828,
      "num_tokens": 10625535.0,
      "step": 4835
    },
    {
      "entropy": 5.812899208068847,
      "epoch": 0.48034934497816595,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004983573934556601,
      "loss": 5.5241,
      "mean_token_accuracy": 0.1474653646349907,
      "num_tokens": 10637160.0,
      "step": 4840
    },
    {
      "entropy": 5.780987215042114,
      "epoch": 0.4808455736403335,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004983531171645734,
      "loss": 5.6279,
      "mean_token_accuracy": 0.14129644334316255,
      "num_tokens": 10649061.0,
      "step": 4845
    },
    {
      "entropy": 5.638265609741211,
      "epoch": 0.481341802302501,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004983488353348106,
      "loss": 5.4829,
      "mean_token_accuracy": 0.14804023057222365,
      "num_tokens": 10661130.0,
      "step": 4850
    },
    {
      "entropy": 5.721083116531372,
      "epoch": 0.48183803096466854,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004983445479664779,
      "loss": 5.5476,
      "mean_token_accuracy": 0.1464819297194481,
      "num_tokens": 10673017.0,
      "step": 4855
    },
    {
      "entropy": 5.7977289199829105,
      "epoch": 0.48233425962683607,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004983402550596817,
      "loss": 5.6084,
      "mean_token_accuracy": 0.14504522681236268,
      "num_tokens": 10683268.0,
      "step": 4860
    },
    {
      "entropy": 5.833132791519165,
      "epoch": 0.48283048828900355,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004983359566145285,
      "loss": 5.6222,
      "mean_token_accuracy": 0.14250127971172333,
      "num_tokens": 10694220.0,
      "step": 4865
    },
    {
      "entropy": 5.747547292709351,
      "epoch": 0.4833267169511711,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004983316526311248,
      "loss": 5.5401,
      "mean_token_accuracy": 0.14009479209780692,
      "num_tokens": 10704829.0,
      "step": 4870
    },
    {
      "entropy": 5.74469518661499,
      "epoch": 0.4838229456133386,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004983273431095773,
      "loss": 5.7219,
      "mean_token_accuracy": 0.13392212018370628,
      "num_tokens": 10717160.0,
      "step": 4875
    },
    {
      "entropy": 5.78354058265686,
      "epoch": 0.48431917427550614,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004983230280499929,
      "loss": 5.5425,
      "mean_token_accuracy": 0.14195659756660461,
      "num_tokens": 10728638.0,
      "step": 4880
    },
    {
      "entropy": 5.737115001678466,
      "epoch": 0.48481540293767367,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004983187074524787,
      "loss": 5.5888,
      "mean_token_accuracy": 0.13973118513822555,
      "num_tokens": 10740122.0,
      "step": 4885
    },
    {
      "entropy": 5.766622018814087,
      "epoch": 0.4853116315998412,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004983143813171418,
      "loss": 5.5563,
      "mean_token_accuracy": 0.1492931678891182,
      "num_tokens": 10751103.0,
      "step": 4890
    },
    {
      "entropy": 5.706625509262085,
      "epoch": 0.48580786026200873,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004983100496440895,
      "loss": 5.5063,
      "mean_token_accuracy": 0.1490258678793907,
      "num_tokens": 10761937.0,
      "step": 4895
    },
    {
      "entropy": 5.796269845962525,
      "epoch": 0.48630408892417626,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000498305712433429,
      "loss": 5.7346,
      "mean_token_accuracy": 0.14223510399460793,
      "num_tokens": 10773527.0,
      "step": 4900
    },
    {
      "entropy": 5.765747499465943,
      "epoch": 0.4868003175863438,
      "grad_norm": 0.828125,
      "learning_rate": 0.000498301369685268,
      "loss": 5.5754,
      "mean_token_accuracy": 0.14397116079926492,
      "num_tokens": 10783873.0,
      "step": 4905
    },
    {
      "entropy": 5.6565389156341555,
      "epoch": 0.4872965462485113,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004982970213997142,
      "loss": 5.4238,
      "mean_token_accuracy": 0.15091024041175843,
      "num_tokens": 10796124.0,
      "step": 4910
    },
    {
      "entropy": 5.614575958251953,
      "epoch": 0.48779277491067885,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004982926675768756,
      "loss": 5.5215,
      "mean_token_accuracy": 0.14468540474772454,
      "num_tokens": 10806468.0,
      "step": 4915
    },
    {
      "entropy": 5.737712860107422,
      "epoch": 0.4882890035728464,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004982883082168599,
      "loss": 5.5778,
      "mean_token_accuracy": 0.14341849982738494,
      "num_tokens": 10817126.0,
      "step": 4920
    },
    {
      "entropy": 5.780298519134521,
      "epoch": 0.4887852322350139,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004982839433197752,
      "loss": 5.597,
      "mean_token_accuracy": 0.13673522770404817,
      "num_tokens": 10828985.0,
      "step": 4925
    },
    {
      "entropy": 5.686007690429688,
      "epoch": 0.48928146089718144,
      "grad_norm": 0.8125,
      "learning_rate": 0.00049827957288573,
      "loss": 5.5045,
      "mean_token_accuracy": 0.14484182596206666,
      "num_tokens": 10840528.0,
      "step": 4930
    },
    {
      "entropy": 5.708341026306153,
      "epoch": 0.48977768955934897,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004982751969148324,
      "loss": 5.5301,
      "mean_token_accuracy": 0.14575338512659072,
      "num_tokens": 10851312.0,
      "step": 4935
    },
    {
      "entropy": 5.67660961151123,
      "epoch": 0.49027391822151645,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004982708154071912,
      "loss": 5.4893,
      "mean_token_accuracy": 0.1472529575228691,
      "num_tokens": 10861882.0,
      "step": 4940
    },
    {
      "entropy": 5.73410964012146,
      "epoch": 0.490770146883684,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004982664283629147,
      "loss": 5.6418,
      "mean_token_accuracy": 0.135673226416111,
      "num_tokens": 10873187.0,
      "step": 4945
    },
    {
      "entropy": 5.750768995285034,
      "epoch": 0.4912663755458515,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004982620357821118,
      "loss": 5.529,
      "mean_token_accuracy": 0.1487978346645832,
      "num_tokens": 10884254.0,
      "step": 4950
    },
    {
      "entropy": 5.62707486152649,
      "epoch": 0.49176260420801904,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004982576376648918,
      "loss": 5.5165,
      "mean_token_accuracy": 0.14257091730833055,
      "num_tokens": 10895770.0,
      "step": 4955
    },
    {
      "entropy": 5.730557155609131,
      "epoch": 0.49225883287018657,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004982532340113633,
      "loss": 5.5434,
      "mean_token_accuracy": 0.14722897857427597,
      "num_tokens": 10906436.0,
      "step": 4960
    },
    {
      "entropy": 5.758408737182617,
      "epoch": 0.4927550615323541,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004982488248216357,
      "loss": 5.4945,
      "mean_token_accuracy": 0.14825854152441026,
      "num_tokens": 10917655.0,
      "step": 4965
    },
    {
      "entropy": 5.69063892364502,
      "epoch": 0.49325129019452163,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004982444100958182,
      "loss": 5.515,
      "mean_token_accuracy": 0.1498719871044159,
      "num_tokens": 10928359.0,
      "step": 4970
    },
    {
      "entropy": 5.748080778121948,
      "epoch": 0.49374751885668916,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004982399898340205,
      "loss": 5.6798,
      "mean_token_accuracy": 0.1363233283162117,
      "num_tokens": 10940494.0,
      "step": 4975
    },
    {
      "entropy": 5.799513626098633,
      "epoch": 0.4942437475188567,
      "grad_norm": 0.828125,
      "learning_rate": 0.000498235564036352,
      "loss": 5.6694,
      "mean_token_accuracy": 0.13602429032325744,
      "num_tokens": 10951298.0,
      "step": 4980
    },
    {
      "entropy": 5.7167809963226315,
      "epoch": 0.4947399761810242,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004982311327029226,
      "loss": 5.565,
      "mean_token_accuracy": 0.14564179703593255,
      "num_tokens": 10961680.0,
      "step": 4985
    },
    {
      "entropy": 5.820360136032105,
      "epoch": 0.49523620484319175,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004982266958338421,
      "loss": 5.6719,
      "mean_token_accuracy": 0.14173018112778663,
      "num_tokens": 10974968.0,
      "step": 4990
    },
    {
      "entropy": 5.712174558639527,
      "epoch": 0.4957324335053593,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004982222534292205,
      "loss": 5.6018,
      "mean_token_accuracy": 0.13866912946105003,
      "num_tokens": 10986383.0,
      "step": 4995
    },
    {
      "entropy": 5.779634046554565,
      "epoch": 0.4962286621675268,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004982178054891681,
      "loss": 5.6601,
      "mean_token_accuracy": 0.14250716120004653,
      "num_tokens": 10998384.0,
      "step": 5000
    },
    {
      "entropy": 5.722892570495605,
      "epoch": 0.49672489082969434,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004982133520137951,
      "loss": 5.4276,
      "mean_token_accuracy": 0.15285419523715973,
      "num_tokens": 11009420.0,
      "step": 5005
    },
    {
      "entropy": 5.763294076919555,
      "epoch": 0.4972211194918619,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004982088930032119,
      "loss": 5.5968,
      "mean_token_accuracy": 0.1417885646224022,
      "num_tokens": 11019601.0,
      "step": 5010
    },
    {
      "entropy": 5.721580982208252,
      "epoch": 0.4977173481540294,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004982044284575293,
      "loss": 5.5333,
      "mean_token_accuracy": 0.1445724494755268,
      "num_tokens": 11031138.0,
      "step": 5015
    },
    {
      "entropy": 5.723393726348877,
      "epoch": 0.4982135768161969,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004981999583768576,
      "loss": 5.5418,
      "mean_token_accuracy": 0.14509216398000718,
      "num_tokens": 11041643.0,
      "step": 5020
    },
    {
      "entropy": 5.676524686813354,
      "epoch": 0.4987098054783644,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000498195482761308,
      "loss": 5.5537,
      "mean_token_accuracy": 0.13784084841609,
      "num_tokens": 11052116.0,
      "step": 5025
    },
    {
      "entropy": 5.821243333816528,
      "epoch": 0.49920603414053194,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004981910016109914,
      "loss": 5.734,
      "mean_token_accuracy": 0.136544182151556,
      "num_tokens": 11064796.0,
      "step": 5030
    },
    {
      "entropy": 5.8484121799469,
      "epoch": 0.49970226280269947,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004981865149260189,
      "loss": 5.5833,
      "mean_token_accuracy": 0.14325116872787474,
      "num_tokens": 11075029.0,
      "step": 5035
    },
    {
      "entropy": 5.731780004501343,
      "epoch": 0.5001984914648671,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004981820227065016,
      "loss": 5.6085,
      "mean_token_accuracy": 0.14171847850084304,
      "num_tokens": 11086702.0,
      "step": 5040
    },
    {
      "entropy": 5.700930786132813,
      "epoch": 0.5006947201270345,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004981775249525513,
      "loss": 5.5816,
      "mean_token_accuracy": 0.1423800431191921,
      "num_tokens": 11097417.0,
      "step": 5045
    },
    {
      "entropy": 5.67112922668457,
      "epoch": 0.5011909487892021,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004981730216642791,
      "loss": 5.4835,
      "mean_token_accuracy": 0.14785366728901864,
      "num_tokens": 11108458.0,
      "step": 5050
    },
    {
      "entropy": 5.775296688079834,
      "epoch": 0.5016871774513696,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004981685128417969,
      "loss": 5.6521,
      "mean_token_accuracy": 0.1389369361102581,
      "num_tokens": 11121466.0,
      "step": 5055
    },
    {
      "entropy": 5.7190674304962155,
      "epoch": 0.5021834061135371,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004981639984852165,
      "loss": 5.5956,
      "mean_token_accuracy": 0.13504938408732414,
      "num_tokens": 11132489.0,
      "step": 5060
    },
    {
      "entropy": 5.695435190200806,
      "epoch": 0.5026796347757047,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004981594785946498,
      "loss": 5.5532,
      "mean_token_accuracy": 0.14500434398651124,
      "num_tokens": 11143087.0,
      "step": 5065
    },
    {
      "entropy": 5.7991992950439455,
      "epoch": 0.5031758634378721,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004981549531702089,
      "loss": 5.5835,
      "mean_token_accuracy": 0.1371206223964691,
      "num_tokens": 11154229.0,
      "step": 5070
    },
    {
      "entropy": 5.748044872283936,
      "epoch": 0.5036720921000397,
      "grad_norm": 0.8046875,
      "learning_rate": 0.000498150422212006,
      "loss": 5.5454,
      "mean_token_accuracy": 0.143742111325264,
      "num_tokens": 11165698.0,
      "step": 5075
    },
    {
      "entropy": 5.725100755691528,
      "epoch": 0.5041683207622072,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004981458857201534,
      "loss": 5.5462,
      "mean_token_accuracy": 0.14240316525101662,
      "num_tokens": 11177025.0,
      "step": 5080
    },
    {
      "entropy": 5.78506989479065,
      "epoch": 0.5046645494243748,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004981413436947639,
      "loss": 5.6209,
      "mean_token_accuracy": 0.14323396608233452,
      "num_tokens": 11188766.0,
      "step": 5085
    },
    {
      "entropy": 5.720196533203125,
      "epoch": 0.5051607780865423,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004981367961359496,
      "loss": 5.6546,
      "mean_token_accuracy": 0.13432737216353416,
      "num_tokens": 11201626.0,
      "step": 5090
    },
    {
      "entropy": 5.732619857788086,
      "epoch": 0.5056570067487098,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004981322430438239,
      "loss": 5.564,
      "mean_token_accuracy": 0.13936000987887381,
      "num_tokens": 11213975.0,
      "step": 5095
    },
    {
      "entropy": 5.688735437393189,
      "epoch": 0.5061532354108773,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004981276844184991,
      "loss": 5.5301,
      "mean_token_accuracy": 0.14741144627332686,
      "num_tokens": 11223636.0,
      "step": 5100
    },
    {
      "entropy": 5.6913628578186035,
      "epoch": 0.5066494640730449,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004981231202600886,
      "loss": 5.497,
      "mean_token_accuracy": 0.14770336896181108,
      "num_tokens": 11233549.0,
      "step": 5105
    },
    {
      "entropy": 5.639012718200684,
      "epoch": 0.5071456927352124,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004981185505687056,
      "loss": 5.5018,
      "mean_token_accuracy": 0.14809941947460176,
      "num_tokens": 11242915.0,
      "step": 5110
    },
    {
      "entropy": 5.788972234725952,
      "epoch": 0.50764192139738,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004981139753444632,
      "loss": 5.6213,
      "mean_token_accuracy": 0.13717583566904068,
      "num_tokens": 11254370.0,
      "step": 5115
    },
    {
      "entropy": 5.745554065704345,
      "epoch": 0.5081381500595474,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004981093945874752,
      "loss": 5.5478,
      "mean_token_accuracy": 0.15008534342050553,
      "num_tokens": 11264226.0,
      "step": 5120
    },
    {
      "entropy": 5.689509296417237,
      "epoch": 0.508634378721715,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004981048082978547,
      "loss": 5.4536,
      "mean_token_accuracy": 0.1530928671360016,
      "num_tokens": 11275304.0,
      "step": 5125
    },
    {
      "entropy": 5.705717992782593,
      "epoch": 0.5091306073838825,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004981002164757158,
      "loss": 5.5157,
      "mean_token_accuracy": 0.14436992928385733,
      "num_tokens": 11286179.0,
      "step": 5130
    },
    {
      "entropy": 5.761962938308716,
      "epoch": 0.50962683604605,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004980956191211723,
      "loss": 5.5448,
      "mean_token_accuracy": 0.14527795910835267,
      "num_tokens": 11297244.0,
      "step": 5135
    },
    {
      "entropy": 5.739287281036377,
      "epoch": 0.5101230647082176,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004980910162343382,
      "loss": 5.5531,
      "mean_token_accuracy": 0.14658074975013732,
      "num_tokens": 11307251.0,
      "step": 5140
    },
    {
      "entropy": 5.726897239685059,
      "epoch": 0.510619293370385,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004980864078153275,
      "loss": 5.5294,
      "mean_token_accuracy": 0.1411370761692524,
      "num_tokens": 11318441.0,
      "step": 5145
    },
    {
      "entropy": 5.605728054046631,
      "epoch": 0.5111155220325526,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004980817938642548,
      "loss": 5.497,
      "mean_token_accuracy": 0.1480163112282753,
      "num_tokens": 11328743.0,
      "step": 5150
    },
    {
      "entropy": 5.608948802947998,
      "epoch": 0.5116117506947201,
      "grad_norm": 0.890625,
      "learning_rate": 0.000498077174381234,
      "loss": 5.4985,
      "mean_token_accuracy": 0.148304795473814,
      "num_tokens": 11340745.0,
      "step": 5155
    },
    {
      "entropy": 5.767784595489502,
      "epoch": 0.5121079793568877,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004980725493663802,
      "loss": 5.5726,
      "mean_token_accuracy": 0.14685158357024192,
      "num_tokens": 11352060.0,
      "step": 5160
    },
    {
      "entropy": 5.702456951141357,
      "epoch": 0.5126042080190552,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004980679188198078,
      "loss": 5.5702,
      "mean_token_accuracy": 0.14360618591308594,
      "num_tokens": 11363097.0,
      "step": 5165
    },
    {
      "entropy": 5.721812438964844,
      "epoch": 0.5131004366812227,
      "grad_norm": 0.875,
      "learning_rate": 0.0004980632827416317,
      "loss": 5.602,
      "mean_token_accuracy": 0.15401604920625686,
      "num_tokens": 11373410.0,
      "step": 5170
    },
    {
      "entropy": 5.758921146392822,
      "epoch": 0.5135966653433902,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004980586411319669,
      "loss": 5.6816,
      "mean_token_accuracy": 0.13971229791641235,
      "num_tokens": 11385615.0,
      "step": 5175
    },
    {
      "entropy": 5.666999673843383,
      "epoch": 0.5140928940055578,
      "grad_norm": 0.76953125,
      "learning_rate": 0.0004980539939909283,
      "loss": 5.4783,
      "mean_token_accuracy": 0.14054736942052842,
      "num_tokens": 11396492.0,
      "step": 5180
    },
    {
      "entropy": 5.781865406036377,
      "epoch": 0.5145891226677253,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004980493413186314,
      "loss": 5.5828,
      "mean_token_accuracy": 0.13886439129710199,
      "num_tokens": 11406377.0,
      "step": 5185
    },
    {
      "entropy": 5.740067386627198,
      "epoch": 0.5150853513298929,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004980446831151914,
      "loss": 5.5575,
      "mean_token_accuracy": 0.1424478217959404,
      "num_tokens": 11416862.0,
      "step": 5190
    },
    {
      "entropy": 5.631788444519043,
      "epoch": 0.5155815799920603,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004980400193807239,
      "loss": 5.5077,
      "mean_token_accuracy": 0.14980673640966416,
      "num_tokens": 11428246.0,
      "step": 5195
    },
    {
      "entropy": 5.748811864852906,
      "epoch": 0.5160778086542279,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004980353501153446,
      "loss": 5.5369,
      "mean_token_accuracy": 0.14516637101769447,
      "num_tokens": 11438147.0,
      "step": 5200
    },
    {
      "entropy": 5.762879467010498,
      "epoch": 0.5165740373163954,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004980306753191691,
      "loss": 5.6576,
      "mean_token_accuracy": 0.13611230552196502,
      "num_tokens": 11449154.0,
      "step": 5205
    },
    {
      "entropy": 5.747657585144043,
      "epoch": 0.5170702659785629,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004980259949923135,
      "loss": 5.5571,
      "mean_token_accuracy": 0.14405836015939713,
      "num_tokens": 11459262.0,
      "step": 5210
    },
    {
      "entropy": 5.719294309616089,
      "epoch": 0.5175664946407305,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004980213091348938,
      "loss": 5.5391,
      "mean_token_accuracy": 0.14385731667280197,
      "num_tokens": 11471991.0,
      "step": 5215
    },
    {
      "entropy": 5.700556945800781,
      "epoch": 0.5180627233028979,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004980166177470262,
      "loss": 5.4846,
      "mean_token_accuracy": 0.14262256026268005,
      "num_tokens": 11483176.0,
      "step": 5220
    },
    {
      "entropy": 5.722855520248413,
      "epoch": 0.5185589519650655,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004980119208288271,
      "loss": 5.5731,
      "mean_token_accuracy": 0.14170767962932587,
      "num_tokens": 11494015.0,
      "step": 5225
    },
    {
      "entropy": 5.6596784591674805,
      "epoch": 0.519055180627233,
      "grad_norm": 0.7421875,
      "learning_rate": 0.0004980072183804128,
      "loss": 5.4895,
      "mean_token_accuracy": 0.1490239754319191,
      "num_tokens": 11506365.0,
      "step": 5230
    },
    {
      "entropy": 5.706877708435059,
      "epoch": 0.5195514092894006,
      "grad_norm": 0.875,
      "learning_rate": 0.0004980025104019,
      "loss": 5.5856,
      "mean_token_accuracy": 0.15004663541913033,
      "num_tokens": 11517199.0,
      "step": 5235
    },
    {
      "entropy": 5.682269048690796,
      "epoch": 0.520047637951568,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004979977968934056,
      "loss": 5.4337,
      "mean_token_accuracy": 0.15386080890893936,
      "num_tokens": 11527591.0,
      "step": 5240
    },
    {
      "entropy": 5.74504714012146,
      "epoch": 0.5205438666137356,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004979930778550463,
      "loss": 5.6126,
      "mean_token_accuracy": 0.14512009173631668,
      "num_tokens": 11538657.0,
      "step": 5245
    },
    {
      "entropy": 5.662412548065186,
      "epoch": 0.5210400952759031,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004979883532869392,
      "loss": 5.4288,
      "mean_token_accuracy": 0.16117970496416092,
      "num_tokens": 11549513.0,
      "step": 5250
    },
    {
      "entropy": 5.721430921554566,
      "epoch": 0.5215363239380707,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004979836231892013,
      "loss": 5.5206,
      "mean_token_accuracy": 0.1425759568810463,
      "num_tokens": 11560376.0,
      "step": 5255
    },
    {
      "entropy": 5.672984027862549,
      "epoch": 0.5220325526002382,
      "grad_norm": 0.875,
      "learning_rate": 0.0004979788875619501,
      "loss": 5.4948,
      "mean_token_accuracy": 0.14300124421715737,
      "num_tokens": 11571218.0,
      "step": 5260
    },
    {
      "entropy": 5.7140038967132565,
      "epoch": 0.5225287812624058,
      "grad_norm": 0.828125,
      "learning_rate": 0.000497974146405303,
      "loss": 5.5693,
      "mean_token_accuracy": 0.14280975982546806,
      "num_tokens": 11581981.0,
      "step": 5265
    },
    {
      "entropy": 5.675239849090576,
      "epoch": 0.5230250099245732,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004979693997193776,
      "loss": 5.5226,
      "mean_token_accuracy": 0.14427190124988556,
      "num_tokens": 11592561.0,
      "step": 5270
    },
    {
      "entropy": 5.732921123504639,
      "epoch": 0.5235212385867408,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004979646475042914,
      "loss": 5.4962,
      "mean_token_accuracy": 0.14883433431386947,
      "num_tokens": 11601933.0,
      "step": 5275
    },
    {
      "entropy": 5.660780525207519,
      "epoch": 0.5240174672489083,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004979598897601625,
      "loss": 5.4274,
      "mean_token_accuracy": 0.15457454025745393,
      "num_tokens": 11613938.0,
      "step": 5280
    },
    {
      "entropy": 5.766283369064331,
      "epoch": 0.5245136959110758,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004979551264871088,
      "loss": 5.6035,
      "mean_token_accuracy": 0.13649525493383408,
      "num_tokens": 11624890.0,
      "step": 5285
    },
    {
      "entropy": 5.738007640838623,
      "epoch": 0.5250099245732434,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004979503576852482,
      "loss": 5.6069,
      "mean_token_accuracy": 0.14732139855623244,
      "num_tokens": 11635080.0,
      "step": 5290
    },
    {
      "entropy": 5.656884241104126,
      "epoch": 0.5255061532354108,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004979455833546994,
      "loss": 5.5011,
      "mean_token_accuracy": 0.14415451809763907,
      "num_tokens": 11645180.0,
      "step": 5295
    },
    {
      "entropy": 5.749808835983276,
      "epoch": 0.5260023818975784,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004979408034955804,
      "loss": 5.5431,
      "mean_token_accuracy": 0.14449771195650102,
      "num_tokens": 11657599.0,
      "step": 5300
    },
    {
      "entropy": 5.738276576995849,
      "epoch": 0.5264986105597459,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004979360181080099,
      "loss": 5.5817,
      "mean_token_accuracy": 0.13297581300139427,
      "num_tokens": 11669177.0,
      "step": 5305
    },
    {
      "entropy": 5.658540487289429,
      "epoch": 0.5269948392219135,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004979312271921065,
      "loss": 5.4448,
      "mean_token_accuracy": 0.14841168075799943,
      "num_tokens": 11679756.0,
      "step": 5310
    },
    {
      "entropy": 5.681262445449829,
      "epoch": 0.527491067884081,
      "grad_norm": 0.8359375,
      "learning_rate": 0.000497926430747989,
      "loss": 5.5043,
      "mean_token_accuracy": 0.1484253525733948,
      "num_tokens": 11689805.0,
      "step": 5315
    },
    {
      "entropy": 5.6945432186126705,
      "epoch": 0.5279872965462485,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004979216287757764,
      "loss": 5.5277,
      "mean_token_accuracy": 0.14734025299549103,
      "num_tokens": 11701150.0,
      "step": 5320
    },
    {
      "entropy": 5.623287963867187,
      "epoch": 0.528483525208416,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000497916821275588,
      "loss": 5.4639,
      "mean_token_accuracy": 0.1452282652258873,
      "num_tokens": 11712256.0,
      "step": 5325
    },
    {
      "entropy": 5.703384447097778,
      "epoch": 0.5289797538705836,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004979120082475424,
      "loss": 5.5123,
      "mean_token_accuracy": 0.14741147682070732,
      "num_tokens": 11722547.0,
      "step": 5330
    },
    {
      "entropy": 5.716305112838745,
      "epoch": 0.5294759825327511,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004979071896917595,
      "loss": 5.522,
      "mean_token_accuracy": 0.1516739696264267,
      "num_tokens": 11732309.0,
      "step": 5335
    },
    {
      "entropy": 5.720453691482544,
      "epoch": 0.5299722111949187,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004979023656083586,
      "loss": 5.5843,
      "mean_token_accuracy": 0.14252596721053123,
      "num_tokens": 11743982.0,
      "step": 5340
    },
    {
      "entropy": 5.66952691078186,
      "epoch": 0.5304684398570861,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004978975359974594,
      "loss": 5.5189,
      "mean_token_accuracy": 0.14486275911331176,
      "num_tokens": 11755506.0,
      "step": 5345
    },
    {
      "entropy": 5.751617050170898,
      "epoch": 0.5309646685192537,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004978927008591816,
      "loss": 5.5216,
      "mean_token_accuracy": 0.14368319660425186,
      "num_tokens": 11767024.0,
      "step": 5350
    },
    {
      "entropy": 5.774104166030884,
      "epoch": 0.5314608971814212,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004978878601936451,
      "loss": 5.5214,
      "mean_token_accuracy": 0.14008311927318573,
      "num_tokens": 11777625.0,
      "step": 5355
    },
    {
      "entropy": 5.676749420166016,
      "epoch": 0.5319571258435887,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004978830140009698,
      "loss": 5.4998,
      "mean_token_accuracy": 0.15135693401098252,
      "num_tokens": 11788610.0,
      "step": 5360
    },
    {
      "entropy": 5.6927237033844,
      "epoch": 0.5324533545057563,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004978781622812762,
      "loss": 5.5998,
      "mean_token_accuracy": 0.14594103246927262,
      "num_tokens": 11800828.0,
      "step": 5365
    },
    {
      "entropy": 5.745589065551758,
      "epoch": 0.5329495831679237,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0004978733050346844,
      "loss": 5.598,
      "mean_token_accuracy": 0.14758748263120652,
      "num_tokens": 11811729.0,
      "step": 5370
    },
    {
      "entropy": 5.737847137451172,
      "epoch": 0.5334458118300913,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004978684422613149,
      "loss": 5.6203,
      "mean_token_accuracy": 0.14279971048235893,
      "num_tokens": 11822081.0,
      "step": 5375
    },
    {
      "entropy": 5.704253196716309,
      "epoch": 0.5339420404922588,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004978635739612882,
      "loss": 5.4959,
      "mean_token_accuracy": 0.1478683464229107,
      "num_tokens": 11832984.0,
      "step": 5380
    },
    {
      "entropy": 5.69680380821228,
      "epoch": 0.5344382691544264,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004978587001347252,
      "loss": 5.4676,
      "mean_token_accuracy": 0.1478103816509247,
      "num_tokens": 11842918.0,
      "step": 5385
    },
    {
      "entropy": 5.63281888961792,
      "epoch": 0.5349344978165939,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004978538207817465,
      "loss": 5.501,
      "mean_token_accuracy": 0.1504437059164047,
      "num_tokens": 11854364.0,
      "step": 5390
    },
    {
      "entropy": 5.684811115264893,
      "epoch": 0.5354307264787614,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004978489359024734,
      "loss": 5.5732,
      "mean_token_accuracy": 0.1437056079506874,
      "num_tokens": 11866400.0,
      "step": 5395
    },
    {
      "entropy": 5.774307823181152,
      "epoch": 0.5359269551409289,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004978440454970268,
      "loss": 5.5738,
      "mean_token_accuracy": 0.14117259830236434,
      "num_tokens": 11877094.0,
      "step": 5400
    },
    {
      "entropy": 5.687881755828857,
      "epoch": 0.5364231838030965,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004978391495655281,
      "loss": 5.5245,
      "mean_token_accuracy": 0.14514778330922126,
      "num_tokens": 11887289.0,
      "step": 5405
    },
    {
      "entropy": 5.739621353149414,
      "epoch": 0.536919412465264,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004978342481080987,
      "loss": 5.5079,
      "mean_token_accuracy": 0.14592916518449783,
      "num_tokens": 11897700.0,
      "step": 5410
    },
    {
      "entropy": 5.5940704345703125,
      "epoch": 0.5374156411274316,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004978293411248601,
      "loss": 5.4367,
      "mean_token_accuracy": 0.15131850391626359,
      "num_tokens": 11908623.0,
      "step": 5415
    },
    {
      "entropy": 5.679359340667725,
      "epoch": 0.537911869789599,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004978244286159339,
      "loss": 5.4525,
      "mean_token_accuracy": 0.15489538311958312,
      "num_tokens": 11917883.0,
      "step": 5420
    },
    {
      "entropy": 5.658584928512573,
      "epoch": 0.5384080984517666,
      "grad_norm": 0.71875,
      "learning_rate": 0.0004978195105814421,
      "loss": 5.4932,
      "mean_token_accuracy": 0.1469782829284668,
      "num_tokens": 11929145.0,
      "step": 5425
    },
    {
      "entropy": 5.692590570449829,
      "epoch": 0.5389043271139341,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004978145870215065,
      "loss": 5.4835,
      "mean_token_accuracy": 0.1445659376680851,
      "num_tokens": 11940073.0,
      "step": 5430
    },
    {
      "entropy": 5.716174697875976,
      "epoch": 0.5394005557761016,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004978096579362493,
      "loss": 5.6122,
      "mean_token_accuracy": 0.14428975731134414,
      "num_tokens": 11953140.0,
      "step": 5435
    },
    {
      "entropy": 5.725712585449219,
      "epoch": 0.5398967844382692,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004978047233257927,
      "loss": 5.5355,
      "mean_token_accuracy": 0.1476729243993759,
      "num_tokens": 11963853.0,
      "step": 5440
    },
    {
      "entropy": 5.696603202819825,
      "epoch": 0.5403930131004366,
      "grad_norm": 0.7734375,
      "learning_rate": 0.0004977997831902591,
      "loss": 5.4953,
      "mean_token_accuracy": 0.14950083196163177,
      "num_tokens": 11975469.0,
      "step": 5445
    },
    {
      "entropy": 5.667461633682251,
      "epoch": 0.5408892417626042,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004977948375297709,
      "loss": 5.4805,
      "mean_token_accuracy": 0.1535550147294998,
      "num_tokens": 11985227.0,
      "step": 5450
    },
    {
      "entropy": 5.554087400436401,
      "epoch": 0.5413854704247717,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004977898863444509,
      "loss": 5.4905,
      "mean_token_accuracy": 0.1473379760980606,
      "num_tokens": 11995459.0,
      "step": 5455
    },
    {
      "entropy": 5.694110012054443,
      "epoch": 0.5418816990869393,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004977849296344217,
      "loss": 5.5509,
      "mean_token_accuracy": 0.14240046888589858,
      "num_tokens": 12007705.0,
      "step": 5460
    },
    {
      "entropy": 5.6833288192749025,
      "epoch": 0.5423779277491068,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004977799673998063,
      "loss": 5.4428,
      "mean_token_accuracy": 0.14829470664262773,
      "num_tokens": 12019729.0,
      "step": 5465
    },
    {
      "entropy": 5.694163846969604,
      "epoch": 0.5428741564112743,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004977749996407278,
      "loss": 5.5204,
      "mean_token_accuracy": 0.14065904766321183,
      "num_tokens": 12031009.0,
      "step": 5470
    },
    {
      "entropy": 5.71015944480896,
      "epoch": 0.5433703850734418,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004977700263573094,
      "loss": 5.4757,
      "mean_token_accuracy": 0.149942646920681,
      "num_tokens": 12041396.0,
      "step": 5475
    },
    {
      "entropy": 5.714011335372925,
      "epoch": 0.5438666137356094,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004977650475496743,
      "loss": 5.6327,
      "mean_token_accuracy": 0.13937563449144363,
      "num_tokens": 12053112.0,
      "step": 5480
    },
    {
      "entropy": 5.701058769226075,
      "epoch": 0.5443628423977769,
      "grad_norm": 0.76953125,
      "learning_rate": 0.0004977600632179459,
      "loss": 5.4853,
      "mean_token_accuracy": 0.1432749643921852,
      "num_tokens": 12064823.0,
      "step": 5485
    },
    {
      "entropy": 5.741170930862427,
      "epoch": 0.5448590710599445,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004977550733622482,
      "loss": 5.611,
      "mean_token_accuracy": 0.1431275188922882,
      "num_tokens": 12076656.0,
      "step": 5490
    },
    {
      "entropy": 5.702113914489746,
      "epoch": 0.5453552997221119,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004977500779827045,
      "loss": 5.417,
      "mean_token_accuracy": 0.14641434401273729,
      "num_tokens": 12086324.0,
      "step": 5495
    },
    {
      "entropy": 5.584799289703369,
      "epoch": 0.5458515283842795,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004977450770794389,
      "loss": 5.4573,
      "mean_token_accuracy": 0.14669612646102906,
      "num_tokens": 12098076.0,
      "step": 5500
    },
    {
      "entropy": 5.658282661437989,
      "epoch": 0.546347757046447,
      "grad_norm": 0.734375,
      "learning_rate": 0.0004977400706525755,
      "loss": 5.5176,
      "mean_token_accuracy": 0.14338865503668785,
      "num_tokens": 12111282.0,
      "step": 5505
    },
    {
      "entropy": 5.67042498588562,
      "epoch": 0.5468439857086145,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004977350587022381,
      "loss": 5.4646,
      "mean_token_accuracy": 0.15267725735902787,
      "num_tokens": 12121932.0,
      "step": 5510
    },
    {
      "entropy": 5.663971471786499,
      "epoch": 0.5473402143707821,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004977300412285513,
      "loss": 5.4506,
      "mean_token_accuracy": 0.1547014519572258,
      "num_tokens": 12133040.0,
      "step": 5515
    },
    {
      "entropy": 5.7008575916290285,
      "epoch": 0.5478364430329495,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004977250182316394,
      "loss": 5.5286,
      "mean_token_accuracy": 0.14612772613763808,
      "num_tokens": 12142577.0,
      "step": 5520
    },
    {
      "entropy": 5.695830965042115,
      "epoch": 0.5483326716951171,
      "grad_norm": 0.875,
      "learning_rate": 0.000497719989711627,
      "loss": 5.4969,
      "mean_token_accuracy": 0.1528421387076378,
      "num_tokens": 12153941.0,
      "step": 5525
    },
    {
      "entropy": 5.6863785743713375,
      "epoch": 0.5488289003572846,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004977149556686387,
      "loss": 5.5501,
      "mean_token_accuracy": 0.14650570154190062,
      "num_tokens": 12165347.0,
      "step": 5530
    },
    {
      "entropy": 5.599493026733398,
      "epoch": 0.5493251290194522,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004977099161027996,
      "loss": 5.4374,
      "mean_token_accuracy": 0.14999375939369203,
      "num_tokens": 12175705.0,
      "step": 5535
    },
    {
      "entropy": 5.706617259979248,
      "epoch": 0.5498213576816197,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004977048710142342,
      "loss": 5.4914,
      "mean_token_accuracy": 0.1500910520553589,
      "num_tokens": 12186781.0,
      "step": 5540
    },
    {
      "entropy": 5.7512706279754635,
      "epoch": 0.5503175863437872,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004976998204030681,
      "loss": 5.5249,
      "mean_token_accuracy": 0.14498836621642114,
      "num_tokens": 12197145.0,
      "step": 5545
    },
    {
      "entropy": 5.624058055877685,
      "epoch": 0.5508138150059547,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004976947642694262,
      "loss": 5.5192,
      "mean_token_accuracy": 0.15011530891060829,
      "num_tokens": 12207637.0,
      "step": 5550
    },
    {
      "entropy": 5.618358469009399,
      "epoch": 0.5513100436681223,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004976897026134341,
      "loss": 5.4405,
      "mean_token_accuracy": 0.15141630321741104,
      "num_tokens": 12218156.0,
      "step": 5555
    },
    {
      "entropy": 5.689433479309082,
      "epoch": 0.5518062723302898,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004976846354352172,
      "loss": 5.528,
      "mean_token_accuracy": 0.1447851613163948,
      "num_tokens": 12229987.0,
      "step": 5560
    },
    {
      "entropy": 5.69376802444458,
      "epoch": 0.5523025009924574,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004976795627349012,
      "loss": 5.5116,
      "mean_token_accuracy": 0.1453092396259308,
      "num_tokens": 12240508.0,
      "step": 5565
    },
    {
      "entropy": 5.652896738052368,
      "epoch": 0.5527987296546248,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004976744845126118,
      "loss": 5.4503,
      "mean_token_accuracy": 0.1521064132452011,
      "num_tokens": 12250317.0,
      "step": 5570
    },
    {
      "entropy": 5.620663976669311,
      "epoch": 0.5532949583167924,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004976694007684751,
      "loss": 5.4953,
      "mean_token_accuracy": 0.1509055018424988,
      "num_tokens": 12262104.0,
      "step": 5575
    },
    {
      "entropy": 5.74803204536438,
      "epoch": 0.5537911869789599,
      "grad_norm": 0.8359375,
      "learning_rate": 0.000497664311502617,
      "loss": 5.5393,
      "mean_token_accuracy": 0.1466676525771618,
      "num_tokens": 12273259.0,
      "step": 5580
    },
    {
      "entropy": 5.736172199249268,
      "epoch": 0.5542874156411274,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004976592167151638,
      "loss": 5.5704,
      "mean_token_accuracy": 0.13990923017263412,
      "num_tokens": 12285504.0,
      "step": 5585
    },
    {
      "entropy": 5.692285537719727,
      "epoch": 0.554783644303295,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004976541164062418,
      "loss": 5.4988,
      "mean_token_accuracy": 0.14217816665768623,
      "num_tokens": 12296512.0,
      "step": 5590
    },
    {
      "entropy": 5.577828121185303,
      "epoch": 0.5552798729654624,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004976490105759774,
      "loss": 5.3666,
      "mean_token_accuracy": 0.15274341851472856,
      "num_tokens": 12307948.0,
      "step": 5595
    },
    {
      "entropy": 5.6738968849182125,
      "epoch": 0.55577610162763,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004976438992244976,
      "loss": 5.5376,
      "mean_token_accuracy": 0.14299441426992415,
      "num_tokens": 12320020.0,
      "step": 5600
    },
    {
      "entropy": 5.689694452285766,
      "epoch": 0.5562723302897975,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004976387823519286,
      "loss": 5.4035,
      "mean_token_accuracy": 0.1494772896170616,
      "num_tokens": 12331091.0,
      "step": 5605
    },
    {
      "entropy": 5.6960502624511715,
      "epoch": 0.5567685589519651,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004976336599583975,
      "loss": 5.5089,
      "mean_token_accuracy": 0.14410399943590163,
      "num_tokens": 12342621.0,
      "step": 5610
    },
    {
      "entropy": 5.788615989685058,
      "epoch": 0.5572647876141326,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004976285320440315,
      "loss": 5.7636,
      "mean_token_accuracy": 0.13273412361741066,
      "num_tokens": 12353330.0,
      "step": 5615
    },
    {
      "entropy": 5.751302766799927,
      "epoch": 0.5577610162763001,
      "grad_norm": 0.77734375,
      "learning_rate": 0.0004976233986089577,
      "loss": 5.4999,
      "mean_token_accuracy": 0.14577182084321977,
      "num_tokens": 12364418.0,
      "step": 5620
    },
    {
      "entropy": 5.610360383987427,
      "epoch": 0.5582572449384676,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004976182596533031,
      "loss": 5.4916,
      "mean_token_accuracy": 0.1467949479818344,
      "num_tokens": 12377022.0,
      "step": 5625
    },
    {
      "entropy": 5.739885330200195,
      "epoch": 0.5587534736006352,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004976131151771955,
      "loss": 5.548,
      "mean_token_accuracy": 0.14481229856610298,
      "num_tokens": 12387799.0,
      "step": 5630
    },
    {
      "entropy": 5.670582485198975,
      "epoch": 0.5592497022628027,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004976079651807623,
      "loss": 5.5007,
      "mean_token_accuracy": 0.14010098576545715,
      "num_tokens": 12397387.0,
      "step": 5635
    },
    {
      "entropy": 5.609793567657471,
      "epoch": 0.5597459309249703,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004976028096641313,
      "loss": 5.4277,
      "mean_token_accuracy": 0.15353151261806489,
      "num_tokens": 12407994.0,
      "step": 5640
    },
    {
      "entropy": 5.786797094345093,
      "epoch": 0.5602421595871377,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004975976486274302,
      "loss": 5.6054,
      "mean_token_accuracy": 0.14164167046546935,
      "num_tokens": 12419124.0,
      "step": 5645
    },
    {
      "entropy": 5.701025772094726,
      "epoch": 0.5607383882493053,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004975924820707872,
      "loss": 5.4893,
      "mean_token_accuracy": 0.14726104438304902,
      "num_tokens": 12428699.0,
      "step": 5650
    },
    {
      "entropy": 5.689584064483642,
      "epoch": 0.5612346169114728,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004975873099943302,
      "loss": 5.57,
      "mean_token_accuracy": 0.14373983442783356,
      "num_tokens": 12439025.0,
      "step": 5655
    },
    {
      "entropy": 5.737751007080078,
      "epoch": 0.5617308455736403,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004975821323981875,
      "loss": 5.5221,
      "mean_token_accuracy": 0.14871042519807814,
      "num_tokens": 12449439.0,
      "step": 5660
    },
    {
      "entropy": 5.723925542831421,
      "epoch": 0.5622270742358079,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004975769492824877,
      "loss": 5.4628,
      "mean_token_accuracy": 0.14492267221212388,
      "num_tokens": 12460066.0,
      "step": 5665
    },
    {
      "entropy": 5.664363098144531,
      "epoch": 0.5627233028979753,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004975717606473589,
      "loss": 5.4883,
      "mean_token_accuracy": 0.15085864216089248,
      "num_tokens": 12471093.0,
      "step": 5670
    },
    {
      "entropy": 5.739382266998291,
      "epoch": 0.5632195315601429,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004975665664929302,
      "loss": 5.513,
      "mean_token_accuracy": 0.14348138272762298,
      "num_tokens": 12481181.0,
      "step": 5675
    },
    {
      "entropy": 5.70921950340271,
      "epoch": 0.5637157602223104,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004975613668193303,
      "loss": 5.528,
      "mean_token_accuracy": 0.14321721345186234,
      "num_tokens": 12493068.0,
      "step": 5680
    },
    {
      "entropy": 5.631594800949097,
      "epoch": 0.564211988884478,
      "grad_norm": 0.921875,
      "learning_rate": 0.000497556161626688,
      "loss": 5.4839,
      "mean_token_accuracy": 0.1489529438316822,
      "num_tokens": 12504025.0,
      "step": 5685
    },
    {
      "entropy": 5.679368209838867,
      "epoch": 0.5647082175466455,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004975509509151325,
      "loss": 5.4858,
      "mean_token_accuracy": 0.1454532712697983,
      "num_tokens": 12514567.0,
      "step": 5690
    },
    {
      "entropy": 5.785870027542114,
      "epoch": 0.565204446208813,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004975457346847929,
      "loss": 5.5249,
      "mean_token_accuracy": 0.1441642612218857,
      "num_tokens": 12525088.0,
      "step": 5695
    },
    {
      "entropy": 5.677102327346802,
      "epoch": 0.5657006748709805,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004975405129357986,
      "loss": 5.4866,
      "mean_token_accuracy": 0.14762923568487168,
      "num_tokens": 12536002.0,
      "step": 5700
    },
    {
      "entropy": 5.676094245910645,
      "epoch": 0.5661969035331481,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004975352856682792,
      "loss": 5.4974,
      "mean_token_accuracy": 0.15148521959781647,
      "num_tokens": 12547468.0,
      "step": 5705
    },
    {
      "entropy": 5.715325212478637,
      "epoch": 0.5666931321953156,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004975300528823642,
      "loss": 5.4656,
      "mean_token_accuracy": 0.1467609941959381,
      "num_tokens": 12557670.0,
      "step": 5710
    },
    {
      "entropy": 5.710341072082519,
      "epoch": 0.5671893608574832,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004975248145781834,
      "loss": 5.5278,
      "mean_token_accuracy": 0.1446359783411026,
      "num_tokens": 12568425.0,
      "step": 5715
    },
    {
      "entropy": 5.608817863464355,
      "epoch": 0.5676855895196506,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004975195707558667,
      "loss": 5.4998,
      "mean_token_accuracy": 0.14632719457149507,
      "num_tokens": 12579138.0,
      "step": 5720
    },
    {
      "entropy": 5.64209623336792,
      "epoch": 0.5681818181818182,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004975143214155443,
      "loss": 5.4958,
      "mean_token_accuracy": 0.14967800229787825,
      "num_tokens": 12589743.0,
      "step": 5725
    },
    {
      "entropy": 5.688083267211914,
      "epoch": 0.5686780468439857,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004975090665573459,
      "loss": 5.4851,
      "mean_token_accuracy": 0.152118718624115,
      "num_tokens": 12600648.0,
      "step": 5730
    },
    {
      "entropy": 5.645006799697876,
      "epoch": 0.5691742755061532,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004975038061814023,
      "loss": 5.5031,
      "mean_token_accuracy": 0.14906034767627716,
      "num_tokens": 12611369.0,
      "step": 5735
    },
    {
      "entropy": 5.7086090564727785,
      "epoch": 0.5696705041683208,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0004974985402878437,
      "loss": 5.4672,
      "mean_token_accuracy": 0.1463977113366127,
      "num_tokens": 12621992.0,
      "step": 5740
    },
    {
      "entropy": 5.697903108596802,
      "epoch": 0.5701667328304882,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004974932688768007,
      "loss": 5.5454,
      "mean_token_accuracy": 0.14656417667865754,
      "num_tokens": 12632463.0,
      "step": 5745
    },
    {
      "entropy": 5.6192210674285885,
      "epoch": 0.5706629614926558,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004974879919484042,
      "loss": 5.4058,
      "mean_token_accuracy": 0.15206122696399688,
      "num_tokens": 12642282.0,
      "step": 5750
    },
    {
      "entropy": 5.636960077285766,
      "epoch": 0.5711591901548233,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004974827095027848,
      "loss": 5.4346,
      "mean_token_accuracy": 0.15423172563314438,
      "num_tokens": 12653495.0,
      "step": 5755
    },
    {
      "entropy": 5.664530849456787,
      "epoch": 0.5716554188169909,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004974774215400736,
      "loss": 5.3785,
      "mean_token_accuracy": 0.15366745442152024,
      "num_tokens": 12664322.0,
      "step": 5760
    },
    {
      "entropy": 5.542517852783203,
      "epoch": 0.5721516474791584,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004974721280604019,
      "loss": 5.4768,
      "mean_token_accuracy": 0.14642566740512847,
      "num_tokens": 12673981.0,
      "step": 5765
    },
    {
      "entropy": 5.599040508270264,
      "epoch": 0.572647876141326,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004974668290639007,
      "loss": 5.4781,
      "mean_token_accuracy": 0.14832795709371566,
      "num_tokens": 12684778.0,
      "step": 5770
    },
    {
      "entropy": 5.705513000488281,
      "epoch": 0.5731441048034934,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004974615245507014,
      "loss": 5.4421,
      "mean_token_accuracy": 0.14995291382074355,
      "num_tokens": 12695287.0,
      "step": 5775
    },
    {
      "entropy": 5.598648929595948,
      "epoch": 0.573640333465661,
      "grad_norm": 0.875,
      "learning_rate": 0.0004974562145209359,
      "loss": 5.5041,
      "mean_token_accuracy": 0.14448004961013794,
      "num_tokens": 12705773.0,
      "step": 5780
    },
    {
      "entropy": 5.6014838218688965,
      "epoch": 0.5741365621278285,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004974508989747354,
      "loss": 5.4276,
      "mean_token_accuracy": 0.15292005538940429,
      "num_tokens": 12716783.0,
      "step": 5785
    },
    {
      "entropy": 5.6707727909088135,
      "epoch": 0.5746327907899961,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004974455779122321,
      "loss": 5.4976,
      "mean_token_accuracy": 0.14310954585671426,
      "num_tokens": 12728146.0,
      "step": 5790
    },
    {
      "entropy": 5.705024862289429,
      "epoch": 0.5751290194521635,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004974402513335577,
      "loss": 5.5146,
      "mean_token_accuracy": 0.14299591705203057,
      "num_tokens": 12739683.0,
      "step": 5795
    },
    {
      "entropy": 5.713213539123535,
      "epoch": 0.5756252481143311,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004974349192388444,
      "loss": 5.4891,
      "mean_token_accuracy": 0.14356330558657646,
      "num_tokens": 12750257.0,
      "step": 5800
    },
    {
      "entropy": 5.72646746635437,
      "epoch": 0.5761214767764986,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004974295816282245,
      "loss": 5.5675,
      "mean_token_accuracy": 0.14630476087331773,
      "num_tokens": 12761055.0,
      "step": 5805
    },
    {
      "entropy": 5.67992148399353,
      "epoch": 0.5766177054386661,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004974242385018303,
      "loss": 5.5052,
      "mean_token_accuracy": 0.14898856729269028,
      "num_tokens": 12771014.0,
      "step": 5810
    },
    {
      "entropy": 5.635032606124878,
      "epoch": 0.5771139341008337,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000497418889859794,
      "loss": 5.424,
      "mean_token_accuracy": 0.15025612711906433,
      "num_tokens": 12782623.0,
      "step": 5815
    },
    {
      "entropy": 5.713941335678101,
      "epoch": 0.5776101627630011,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004974135357022488,
      "loss": 5.543,
      "mean_token_accuracy": 0.14384849816560746,
      "num_tokens": 12793533.0,
      "step": 5820
    },
    {
      "entropy": 5.593503761291504,
      "epoch": 0.5781063914251687,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004974081760293269,
      "loss": 5.4105,
      "mean_token_accuracy": 0.14829638749361038,
      "num_tokens": 12804389.0,
      "step": 5825
    },
    {
      "entropy": 5.666346645355224,
      "epoch": 0.5786026200873362,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004974028108411617,
      "loss": 5.512,
      "mean_token_accuracy": 0.15230901390314103,
      "num_tokens": 12815672.0,
      "step": 5830
    },
    {
      "entropy": 5.732553863525391,
      "epoch": 0.5790988487495038,
      "grad_norm": 0.875,
      "learning_rate": 0.0004973974401378859,
      "loss": 5.5116,
      "mean_token_accuracy": 0.15112632364034653,
      "num_tokens": 12826113.0,
      "step": 5835
    },
    {
      "entropy": 5.612425994873047,
      "epoch": 0.5795950774116713,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004973920639196329,
      "loss": 5.4408,
      "mean_token_accuracy": 0.1494893580675125,
      "num_tokens": 12835941.0,
      "step": 5840
    },
    {
      "entropy": 5.639367532730103,
      "epoch": 0.5800913060738389,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004973866821865358,
      "loss": 5.5282,
      "mean_token_accuracy": 0.14228470176458358,
      "num_tokens": 12846186.0,
      "step": 5845
    },
    {
      "entropy": 5.718220901489258,
      "epoch": 0.5805875347360063,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004973812949387283,
      "loss": 5.4929,
      "mean_token_accuracy": 0.14542728066444396,
      "num_tokens": 12856380.0,
      "step": 5850
    },
    {
      "entropy": 5.649071073532104,
      "epoch": 0.5810837633981739,
      "grad_norm": 0.875,
      "learning_rate": 0.0004973759021763439,
      "loss": 5.5286,
      "mean_token_accuracy": 0.14764680862426757,
      "num_tokens": 12865999.0,
      "step": 5855
    },
    {
      "entropy": 5.68231954574585,
      "epoch": 0.5815799920603414,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004973705038995162,
      "loss": 5.4126,
      "mean_token_accuracy": 0.14954375773668288,
      "num_tokens": 12876388.0,
      "step": 5860
    },
    {
      "entropy": 5.625105524063111,
      "epoch": 0.582076220722509,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004973651001083792,
      "loss": 5.4145,
      "mean_token_accuracy": 0.14912877306342126,
      "num_tokens": 12885862.0,
      "step": 5865
    },
    {
      "entropy": 5.605915117263794,
      "epoch": 0.5825724493846764,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004973596908030669,
      "loss": 5.5022,
      "mean_token_accuracy": 0.1445961207151413,
      "num_tokens": 12897489.0,
      "step": 5870
    },
    {
      "entropy": 5.703808736801148,
      "epoch": 0.583068678046844,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004973542759837135,
      "loss": 5.5464,
      "mean_token_accuracy": 0.14082904383540154,
      "num_tokens": 12907395.0,
      "step": 5875
    },
    {
      "entropy": 5.62304458618164,
      "epoch": 0.5835649067090115,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000497348855650453,
      "loss": 5.4717,
      "mean_token_accuracy": 0.1517767146229744,
      "num_tokens": 12919439.0,
      "step": 5880
    },
    {
      "entropy": 5.69926609992981,
      "epoch": 0.584061135371179,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004973434298034201,
      "loss": 5.5616,
      "mean_token_accuracy": 0.14480644688010216,
      "num_tokens": 12931892.0,
      "step": 5885
    },
    {
      "entropy": 5.770834016799927,
      "epoch": 0.5845573640333466,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004973379984427492,
      "loss": 5.603,
      "mean_token_accuracy": 0.14584836214780808,
      "num_tokens": 12943189.0,
      "step": 5890
    },
    {
      "entropy": 5.670014476776123,
      "epoch": 0.585053592695514,
      "grad_norm": 0.875,
      "learning_rate": 0.000497332561568575,
      "loss": 5.5406,
      "mean_token_accuracy": 0.14588731974363328,
      "num_tokens": 12954662.0,
      "step": 5895
    },
    {
      "entropy": 5.63857250213623,
      "epoch": 0.5855498213576816,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004973271191810325,
      "loss": 5.4833,
      "mean_token_accuracy": 0.14281948506832123,
      "num_tokens": 12964924.0,
      "step": 5900
    },
    {
      "entropy": 5.689583873748779,
      "epoch": 0.5860460500198491,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004973216712802564,
      "loss": 5.5549,
      "mean_token_accuracy": 0.14132398664951323,
      "num_tokens": 12976483.0,
      "step": 5905
    },
    {
      "entropy": 5.654120254516601,
      "epoch": 0.5865422786820167,
      "grad_norm": 0.83203125,
      "learning_rate": 0.000497316217866382,
      "loss": 5.4601,
      "mean_token_accuracy": 0.15196489095687865,
      "num_tokens": 12988054.0,
      "step": 5910
    },
    {
      "entropy": 5.69523344039917,
      "epoch": 0.5870385073441842,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004973107589395444,
      "loss": 5.5062,
      "mean_token_accuracy": 0.1495279923081398,
      "num_tokens": 13000596.0,
      "step": 5915
    },
    {
      "entropy": 5.558714103698731,
      "epoch": 0.5875347360063518,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004973052944998791,
      "loss": 5.3601,
      "mean_token_accuracy": 0.15159481316804885,
      "num_tokens": 13012360.0,
      "step": 5920
    },
    {
      "entropy": 5.637037706375122,
      "epoch": 0.5880309646685192,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004972998245475214,
      "loss": 5.5061,
      "mean_token_accuracy": 0.14521099850535393,
      "num_tokens": 13023806.0,
      "step": 5925
    },
    {
      "entropy": 5.674616146087646,
      "epoch": 0.5885271933306868,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000497294349082607,
      "loss": 5.4444,
      "mean_token_accuracy": 0.14743458330631257,
      "num_tokens": 13034049.0,
      "step": 5930
    },
    {
      "entropy": 5.698005294799804,
      "epoch": 0.5890234219928543,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004972888681052719,
      "loss": 5.5418,
      "mean_token_accuracy": 0.14588503241539003,
      "num_tokens": 13045006.0,
      "step": 5935
    },
    {
      "entropy": 5.713470411300659,
      "epoch": 0.5895196506550219,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004972833816156518,
      "loss": 5.454,
      "mean_token_accuracy": 0.151131209731102,
      "num_tokens": 13055337.0,
      "step": 5940
    },
    {
      "entropy": 5.613440036773682,
      "epoch": 0.5900158793171894,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004972778896138828,
      "loss": 5.4414,
      "mean_token_accuracy": 0.151805180311203,
      "num_tokens": 13065267.0,
      "step": 5945
    },
    {
      "entropy": 5.648981380462646,
      "epoch": 0.5905121079793569,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004972723921001011,
      "loss": 5.5115,
      "mean_token_accuracy": 0.14798768609762192,
      "num_tokens": 13075173.0,
      "step": 5950
    },
    {
      "entropy": 5.707030391693115,
      "epoch": 0.5910083366415244,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004972668890744431,
      "loss": 5.4911,
      "mean_token_accuracy": 0.14562270641326905,
      "num_tokens": 13086848.0,
      "step": 5955
    },
    {
      "entropy": 5.640896415710449,
      "epoch": 0.5915045653036919,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004972613805370452,
      "loss": 5.4768,
      "mean_token_accuracy": 0.1498607374727726,
      "num_tokens": 13096865.0,
      "step": 5960
    },
    {
      "entropy": 5.612109804153443,
      "epoch": 0.5920007939658595,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004972558664880439,
      "loss": 5.4543,
      "mean_token_accuracy": 0.14806280583143233,
      "num_tokens": 13107907.0,
      "step": 5965
    },
    {
      "entropy": 5.591845846176147,
      "epoch": 0.592497022628027,
      "grad_norm": 1.015625,
      "learning_rate": 0.000497250346927576,
      "loss": 5.373,
      "mean_token_accuracy": 0.15522033870220184,
      "num_tokens": 13118171.0,
      "step": 5970
    },
    {
      "entropy": 5.639474821090698,
      "epoch": 0.5929932512901945,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004972448218557785,
      "loss": 5.3608,
      "mean_token_accuracy": 0.15381580740213394,
      "num_tokens": 13127042.0,
      "step": 5975
    },
    {
      "entropy": 5.682769632339477,
      "epoch": 0.593489479952362,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004972392912727882,
      "loss": 5.6267,
      "mean_token_accuracy": 0.1483294628560543,
      "num_tokens": 13137026.0,
      "step": 5980
    },
    {
      "entropy": 5.6737512111663815,
      "epoch": 0.5939857086145296,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004972337551787425,
      "loss": 5.4745,
      "mean_token_accuracy": 0.15264537185430527,
      "num_tokens": 13146947.0,
      "step": 5985
    },
    {
      "entropy": 5.762002801895141,
      "epoch": 0.5944819372766971,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004972282135737785,
      "loss": 5.528,
      "mean_token_accuracy": 0.14844981729984283,
      "num_tokens": 13158714.0,
      "step": 5990
    },
    {
      "entropy": 5.709932327270508,
      "epoch": 0.5949781659388647,
      "grad_norm": 0.875,
      "learning_rate": 0.0004972226664580337,
      "loss": 5.5318,
      "mean_token_accuracy": 0.14496293887495995,
      "num_tokens": 13170156.0,
      "step": 5995
    },
    {
      "entropy": 5.593976449966431,
      "epoch": 0.5954743946010321,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004972171138316456,
      "loss": 5.4315,
      "mean_token_accuracy": 0.14985745549201965,
      "num_tokens": 13179919.0,
      "step": 6000
    },
    {
      "epoch": 0.5954743946010321,
      "eval_entropy": 5.474441283727434,
      "eval_loss": 5.552264213562012,
      "eval_mean_token_accuracy": 0.14955780961736417,
      "eval_num_tokens": 13179919.0,
      "eval_runtime": 20.3484,
      "eval_samples_per_second": 1654.624,
      "eval_steps_per_second": 206.846,
      "step": 6000
    },
    {
      "entropy": 5.6688230514526365,
      "epoch": 0.5959706232631997,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004972115556947518,
      "loss": 5.5127,
      "mean_token_accuracy": 0.14665141627192496,
      "num_tokens": 13190777.0,
      "step": 6005
    },
    {
      "entropy": 5.711255836486816,
      "epoch": 0.5964668519253672,
      "grad_norm": 0.875,
      "learning_rate": 0.0004972059920474904,
      "loss": 5.5007,
      "mean_token_accuracy": 0.14610673263669013,
      "num_tokens": 13202179.0,
      "step": 6010
    },
    {
      "entropy": 5.662798738479614,
      "epoch": 0.5969630805875348,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004972004228899991,
      "loss": 5.4586,
      "mean_token_accuracy": 0.15040792152285576,
      "num_tokens": 13212764.0,
      "step": 6015
    },
    {
      "entropy": 5.642608690261841,
      "epoch": 0.5974593092497023,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004971948482224162,
      "loss": 5.5022,
      "mean_token_accuracy": 0.1451937548816204,
      "num_tokens": 13223343.0,
      "step": 6020
    },
    {
      "entropy": 5.653862953186035,
      "epoch": 0.5979555379118698,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004971892680448798,
      "loss": 5.4709,
      "mean_token_accuracy": 0.1503230482339859,
      "num_tokens": 13233839.0,
      "step": 6025
    },
    {
      "entropy": 5.720564508438111,
      "epoch": 0.5984517665740373,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004971836823575283,
      "loss": 5.5738,
      "mean_token_accuracy": 0.1428346298635006,
      "num_tokens": 13243854.0,
      "step": 6030
    },
    {
      "entropy": 5.697343015670777,
      "epoch": 0.5989479952362048,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004971780911605003,
      "loss": 5.4835,
      "mean_token_accuracy": 0.1490764081478119,
      "num_tokens": 13256073.0,
      "step": 6035
    },
    {
      "entropy": 5.665929317474365,
      "epoch": 0.5994442238983724,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004971724944539343,
      "loss": 5.515,
      "mean_token_accuracy": 0.14822529405355453,
      "num_tokens": 13267568.0,
      "step": 6040
    },
    {
      "entropy": 5.720554208755493,
      "epoch": 0.5999404525605398,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004971668922379692,
      "loss": 5.4985,
      "mean_token_accuracy": 0.1423833392560482,
      "num_tokens": 13278829.0,
      "step": 6045
    },
    {
      "entropy": 5.669347763061523,
      "epoch": 0.6004366812227074,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000497161284512744,
      "loss": 5.458,
      "mean_token_accuracy": 0.15174900591373444,
      "num_tokens": 13289156.0,
      "step": 6050
    },
    {
      "entropy": 5.547655439376831,
      "epoch": 0.6009329098848749,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004971556712783976,
      "loss": 5.3382,
      "mean_token_accuracy": 0.15496094673871993,
      "num_tokens": 13299817.0,
      "step": 6055
    },
    {
      "entropy": 5.56444673538208,
      "epoch": 0.6014291385470425,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004971500525350692,
      "loss": 5.4054,
      "mean_token_accuracy": 0.15332763493061066,
      "num_tokens": 13310290.0,
      "step": 6060
    },
    {
      "entropy": 5.607007026672363,
      "epoch": 0.60192536720921,
      "grad_norm": 0.76953125,
      "learning_rate": 0.0004971444282828983,
      "loss": 5.5244,
      "mean_token_accuracy": 0.1488925039768219,
      "num_tokens": 13321747.0,
      "step": 6065
    },
    {
      "entropy": 5.725985717773438,
      "epoch": 0.6024215958713776,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004971387985220242,
      "loss": 5.4554,
      "mean_token_accuracy": 0.15337199568748475,
      "num_tokens": 13333249.0,
      "step": 6070
    },
    {
      "entropy": 5.558460092544555,
      "epoch": 0.602917824533545,
      "grad_norm": 0.875,
      "learning_rate": 0.0004971331632525866,
      "loss": 5.4383,
      "mean_token_accuracy": 0.1514162078499794,
      "num_tokens": 13345336.0,
      "step": 6075
    },
    {
      "entropy": 5.7256739139556885,
      "epoch": 0.6034140531957126,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000497127522474725,
      "loss": 5.5893,
      "mean_token_accuracy": 0.14125749319791794,
      "num_tokens": 13355797.0,
      "step": 6080
    },
    {
      "entropy": 5.720564508438111,
      "epoch": 0.6039102818578801,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004971218761885797,
      "loss": 5.5372,
      "mean_token_accuracy": 0.14828825294971465,
      "num_tokens": 13367058.0,
      "step": 6085
    },
    {
      "entropy": 5.668391180038452,
      "epoch": 0.6044065105200477,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004971162243942904,
      "loss": 5.4761,
      "mean_token_accuracy": 0.14879498779773712,
      "num_tokens": 13378579.0,
      "step": 6090
    },
    {
      "entropy": 5.670612144470215,
      "epoch": 0.6049027391822152,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004971105670919975,
      "loss": 5.4647,
      "mean_token_accuracy": 0.14927419573068618,
      "num_tokens": 13389240.0,
      "step": 6095
    },
    {
      "entropy": 5.708739519119263,
      "epoch": 0.6053989678443827,
      "grad_norm": 0.8359375,
      "learning_rate": 0.000497104904281841,
      "loss": 5.5996,
      "mean_token_accuracy": 0.14185432642698287,
      "num_tokens": 13401895.0,
      "step": 6100
    },
    {
      "entropy": 5.748668956756592,
      "epoch": 0.6058951965065502,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004970992359639616,
      "loss": 5.5673,
      "mean_token_accuracy": 0.1404925614595413,
      "num_tokens": 13413334.0,
      "step": 6105
    },
    {
      "entropy": 5.686299133300781,
      "epoch": 0.6063914251687177,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004970935621384996,
      "loss": 5.4763,
      "mean_token_accuracy": 0.15073925480246544,
      "num_tokens": 13423371.0,
      "step": 6110
    },
    {
      "entropy": 5.716171646118164,
      "epoch": 0.6068876538308853,
      "grad_norm": 1.875,
      "learning_rate": 0.0004970878828055959,
      "loss": 5.4628,
      "mean_token_accuracy": 0.15296884328126908,
      "num_tokens": 13434782.0,
      "step": 6115
    },
    {
      "entropy": 5.639015197753906,
      "epoch": 0.6073838824930528,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004970821979653912,
      "loss": 5.4516,
      "mean_token_accuracy": 0.1492540270090103,
      "num_tokens": 13444493.0,
      "step": 6120
    },
    {
      "entropy": 5.634776020050049,
      "epoch": 0.6078801111552203,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004970765076180265,
      "loss": 5.4336,
      "mean_token_accuracy": 0.14924686998128892,
      "num_tokens": 13455518.0,
      "step": 6125
    },
    {
      "entropy": 5.729962301254273,
      "epoch": 0.6083763398173878,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004970708117636431,
      "loss": 5.5256,
      "mean_token_accuracy": 0.15012605488300323,
      "num_tokens": 13467631.0,
      "step": 6130
    },
    {
      "entropy": 5.6114765167236325,
      "epoch": 0.6088725684795554,
      "grad_norm": 0.80078125,
      "learning_rate": 0.000497065110402382,
      "loss": 5.4627,
      "mean_token_accuracy": 0.14560690820217131,
      "num_tokens": 13478217.0,
      "step": 6135
    },
    {
      "entropy": 5.63499870300293,
      "epoch": 0.6093687971417229,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004970594035343846,
      "loss": 5.4374,
      "mean_token_accuracy": 0.15331496596336364,
      "num_tokens": 13489169.0,
      "step": 6140
    },
    {
      "entropy": 5.675707912445068,
      "epoch": 0.6098650258038905,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004970536911597926,
      "loss": 5.4679,
      "mean_token_accuracy": 0.14365995824337005,
      "num_tokens": 13499273.0,
      "step": 6145
    },
    {
      "entropy": 5.737535095214843,
      "epoch": 0.6103612544660579,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004970479732787476,
      "loss": 5.4907,
      "mean_token_accuracy": 0.15293558239936828,
      "num_tokens": 13509664.0,
      "step": 6150
    },
    {
      "entropy": 5.632332801818848,
      "epoch": 0.6108574831282255,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004970422498913911,
      "loss": 5.383,
      "mean_token_accuracy": 0.14625990018248558,
      "num_tokens": 13520288.0,
      "step": 6155
    },
    {
      "entropy": 5.617789173126221,
      "epoch": 0.611353711790393,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004970365209978655,
      "loss": 5.482,
      "mean_token_accuracy": 0.14435981065034867,
      "num_tokens": 13531167.0,
      "step": 6160
    },
    {
      "entropy": 5.633021974563599,
      "epoch": 0.6118499404525606,
      "grad_norm": 0.77734375,
      "learning_rate": 0.0004970307865983124,
      "loss": 5.5136,
      "mean_token_accuracy": 0.15051327794790267,
      "num_tokens": 13544109.0,
      "step": 6165
    },
    {
      "entropy": 5.711878728866577,
      "epoch": 0.6123461691147281,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004970250466928744,
      "loss": 5.4445,
      "mean_token_accuracy": 0.15177036300301552,
      "num_tokens": 13554935.0,
      "step": 6170
    },
    {
      "entropy": 5.650032234191895,
      "epoch": 0.6128423977768956,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004970193012816936,
      "loss": 5.4592,
      "mean_token_accuracy": 0.1498909927904606,
      "num_tokens": 13565959.0,
      "step": 6175
    },
    {
      "entropy": 5.586412048339843,
      "epoch": 0.6133386264390631,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004970135503649125,
      "loss": 5.4659,
      "mean_token_accuracy": 0.14620325416326524,
      "num_tokens": 13577883.0,
      "step": 6180
    },
    {
      "entropy": 5.672992372512818,
      "epoch": 0.6138348551012306,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004970077939426738,
      "loss": 5.4442,
      "mean_token_accuracy": 0.1484336093068123,
      "num_tokens": 13588597.0,
      "step": 6185
    },
    {
      "entropy": 5.665392017364502,
      "epoch": 0.6143310837633982,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004970020320151202,
      "loss": 5.3882,
      "mean_token_accuracy": 0.15516335219144822,
      "num_tokens": 13598772.0,
      "step": 6190
    },
    {
      "entropy": 5.672236442565918,
      "epoch": 0.6148273124255657,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004969962645823946,
      "loss": 5.4717,
      "mean_token_accuracy": 0.1497521921992302,
      "num_tokens": 13609696.0,
      "step": 6195
    },
    {
      "entropy": 5.592589807510376,
      "epoch": 0.6153235410877332,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00049699049164464,
      "loss": 5.4598,
      "mean_token_accuracy": 0.15013887584209443,
      "num_tokens": 13619389.0,
      "step": 6200
    },
    {
      "entropy": 5.624180746078491,
      "epoch": 0.6158197697499007,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004969847132019995,
      "loss": 5.4294,
      "mean_token_accuracy": 0.15189450085163117,
      "num_tokens": 13632133.0,
      "step": 6205
    },
    {
      "entropy": 5.660528612136841,
      "epoch": 0.6163159984120683,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004969789292546164,
      "loss": 5.3865,
      "mean_token_accuracy": 0.1556358143687248,
      "num_tokens": 13642715.0,
      "step": 6210
    },
    {
      "entropy": 5.621987342834473,
      "epoch": 0.6168122270742358,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004969731398026343,
      "loss": 5.4931,
      "mean_token_accuracy": 0.14751377552747727,
      "num_tokens": 13653223.0,
      "step": 6215
    },
    {
      "entropy": 5.559430122375488,
      "epoch": 0.6173084557364034,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004969673448461966,
      "loss": 5.4629,
      "mean_token_accuracy": 0.1489397704601288,
      "num_tokens": 13664206.0,
      "step": 6220
    },
    {
      "entropy": 5.66539363861084,
      "epoch": 0.6178046843985708,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000496961544385447,
      "loss": 5.486,
      "mean_token_accuracy": 0.14996751099824907,
      "num_tokens": 13673974.0,
      "step": 6225
    },
    {
      "entropy": 5.628175449371338,
      "epoch": 0.6183009130607384,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004969557384205295,
      "loss": 5.4018,
      "mean_token_accuracy": 0.15043520480394362,
      "num_tokens": 13685189.0,
      "step": 6230
    },
    {
      "entropy": 5.642521572113037,
      "epoch": 0.6187971417229059,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004969499269515878,
      "loss": 5.4195,
      "mean_token_accuracy": 0.15626146644353867,
      "num_tokens": 13696181.0,
      "step": 6235
    },
    {
      "entropy": 5.738152933120728,
      "epoch": 0.6192933703850735,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004969441099787663,
      "loss": 5.5273,
      "mean_token_accuracy": 0.14354099780321122,
      "num_tokens": 13706719.0,
      "step": 6240
    },
    {
      "entropy": 5.740992403030395,
      "epoch": 0.619789599047241,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000496938287502209,
      "loss": 5.4874,
      "mean_token_accuracy": 0.1491748109459877,
      "num_tokens": 13716653.0,
      "step": 6245
    },
    {
      "entropy": 5.581117296218872,
      "epoch": 0.6202858277094085,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004969324595220605,
      "loss": 5.3952,
      "mean_token_accuracy": 0.15741943269968034,
      "num_tokens": 13728220.0,
      "step": 6250
    },
    {
      "entropy": 5.562930679321289,
      "epoch": 0.620782056371576,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004969266260384652,
      "loss": 5.3781,
      "mean_token_accuracy": 0.15155331492424012,
      "num_tokens": 13738103.0,
      "step": 6255
    },
    {
      "entropy": 5.641094112396241,
      "epoch": 0.6212782850337435,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004969207870515678,
      "loss": 5.4541,
      "mean_token_accuracy": 0.1509600356221199,
      "num_tokens": 13749845.0,
      "step": 6260
    },
    {
      "entropy": 5.6544554233551025,
      "epoch": 0.6217745136959111,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000496914942561513,
      "loss": 5.4965,
      "mean_token_accuracy": 0.15008585378527642,
      "num_tokens": 13759983.0,
      "step": 6265
    },
    {
      "entropy": 5.687742471694946,
      "epoch": 0.6222707423580786,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004969090925684459,
      "loss": 5.5695,
      "mean_token_accuracy": 0.14714422523975373,
      "num_tokens": 13770045.0,
      "step": 6270
    },
    {
      "entropy": 5.5897331714630125,
      "epoch": 0.6227669710202461,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004969032370725114,
      "loss": 5.428,
      "mean_token_accuracy": 0.15328796654939653,
      "num_tokens": 13779823.0,
      "step": 6275
    },
    {
      "entropy": 5.629849863052368,
      "epoch": 0.6232631996824136,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0004968973760738548,
      "loss": 5.4199,
      "mean_token_accuracy": 0.1493604764342308,
      "num_tokens": 13791165.0,
      "step": 6280
    },
    {
      "entropy": 5.711244440078735,
      "epoch": 0.6237594283445812,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004968915095726216,
      "loss": 5.5162,
      "mean_token_accuracy": 0.1434049963951111,
      "num_tokens": 13802872.0,
      "step": 6285
    },
    {
      "entropy": 5.658536672592163,
      "epoch": 0.6242556570067487,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004968856375689568,
      "loss": 5.4538,
      "mean_token_accuracy": 0.15151210278272628,
      "num_tokens": 13813893.0,
      "step": 6290
    },
    {
      "entropy": 5.6452268123626705,
      "epoch": 0.6247518856689163,
      "grad_norm": 0.875,
      "learning_rate": 0.0004968797600630065,
      "loss": 5.5219,
      "mean_token_accuracy": 0.15299668312072753,
      "num_tokens": 13823613.0,
      "step": 6295
    },
    {
      "entropy": 5.7115713596344,
      "epoch": 0.6252481143310837,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004968738770549162,
      "loss": 5.525,
      "mean_token_accuracy": 0.14985218942165374,
      "num_tokens": 13835413.0,
      "step": 6300
    },
    {
      "entropy": 5.634415054321289,
      "epoch": 0.6257443429932513,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004968679885448318,
      "loss": 5.4584,
      "mean_token_accuracy": 0.14723524004220961,
      "num_tokens": 13845806.0,
      "step": 6305
    },
    {
      "entropy": 5.693110227584839,
      "epoch": 0.6262405716554188,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004968620945328995,
      "loss": 5.5625,
      "mean_token_accuracy": 0.14774496108293533,
      "num_tokens": 13856540.0,
      "step": 6310
    },
    {
      "entropy": 5.641265916824341,
      "epoch": 0.6267368003175864,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004968561950192652,
      "loss": 5.4035,
      "mean_token_accuracy": 0.15361193418502808,
      "num_tokens": 13867609.0,
      "step": 6315
    },
    {
      "entropy": 5.698054599761963,
      "epoch": 0.6272330289797539,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004968502900040753,
      "loss": 5.4735,
      "mean_token_accuracy": 0.14571210592985154,
      "num_tokens": 13878299.0,
      "step": 6320
    },
    {
      "entropy": 5.694347095489502,
      "epoch": 0.6277292576419214,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004968443794874764,
      "loss": 5.4807,
      "mean_token_accuracy": 0.14888293594121932,
      "num_tokens": 13889625.0,
      "step": 6325
    },
    {
      "entropy": 5.62461142539978,
      "epoch": 0.6282254863040889,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004968384634696149,
      "loss": 5.4585,
      "mean_token_accuracy": 0.1475509375333786,
      "num_tokens": 13899680.0,
      "step": 6330
    },
    {
      "entropy": 5.574176692962647,
      "epoch": 0.6287217149662565,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004968325419506375,
      "loss": 5.4308,
      "mean_token_accuracy": 0.15305441319942475,
      "num_tokens": 13910883.0,
      "step": 6335
    },
    {
      "entropy": 5.608674001693726,
      "epoch": 0.629217943628424,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004968266149306909,
      "loss": 5.4367,
      "mean_token_accuracy": 0.15015380084514618,
      "num_tokens": 13921442.0,
      "step": 6340
    },
    {
      "entropy": 5.6577832221984865,
      "epoch": 0.6297141722905915,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004968206824099224,
      "loss": 5.4799,
      "mean_token_accuracy": 0.15244491249322892,
      "num_tokens": 13930309.0,
      "step": 6345
    },
    {
      "entropy": 5.672576761245727,
      "epoch": 0.630210400952759,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004968147443884789,
      "loss": 5.4624,
      "mean_token_accuracy": 0.1559087797999382,
      "num_tokens": 13941346.0,
      "step": 6350
    },
    {
      "entropy": 5.636896467208862,
      "epoch": 0.6307066296149265,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004968088008665078,
      "loss": 5.4418,
      "mean_token_accuracy": 0.1529480427503586,
      "num_tokens": 13951933.0,
      "step": 6355
    },
    {
      "entropy": 5.587031555175781,
      "epoch": 0.6312028582770941,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004968028518441563,
      "loss": 5.4884,
      "mean_token_accuracy": 0.1458326257765293,
      "num_tokens": 13963112.0,
      "step": 6360
    },
    {
      "entropy": 5.713586330413818,
      "epoch": 0.6316990869392616,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004967968973215721,
      "loss": 5.5219,
      "mean_token_accuracy": 0.15321382731199265,
      "num_tokens": 13974218.0,
      "step": 6365
    },
    {
      "entropy": 5.71185941696167,
      "epoch": 0.6321953156014292,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004967909372989027,
      "loss": 5.4483,
      "mean_token_accuracy": 0.15306579768657685,
      "num_tokens": 13984841.0,
      "step": 6370
    },
    {
      "entropy": 5.609977102279663,
      "epoch": 0.6326915442635966,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0004967849717762959,
      "loss": 5.4589,
      "mean_token_accuracy": 0.1499641492962837,
      "num_tokens": 13997504.0,
      "step": 6375
    },
    {
      "entropy": 5.571201181411743,
      "epoch": 0.6331877729257642,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004967790007538999,
      "loss": 5.4043,
      "mean_token_accuracy": 0.15468272864818572,
      "num_tokens": 14009244.0,
      "step": 6380
    },
    {
      "entropy": 5.617036819458008,
      "epoch": 0.6336840015879317,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004967730242318624,
      "loss": 5.439,
      "mean_token_accuracy": 0.1511642351746559,
      "num_tokens": 14021190.0,
      "step": 6385
    },
    {
      "entropy": 5.607939720153809,
      "epoch": 0.6341802302500993,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000496767042210332,
      "loss": 5.3981,
      "mean_token_accuracy": 0.1521475076675415,
      "num_tokens": 14032191.0,
      "step": 6390
    },
    {
      "entropy": 5.543286514282227,
      "epoch": 0.6346764589122668,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004967610546894566,
      "loss": 5.3764,
      "mean_token_accuracy": 0.15400539189577103,
      "num_tokens": 14042817.0,
      "step": 6395
    },
    {
      "entropy": 5.626496887207031,
      "epoch": 0.6351726875744343,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000496755061669385,
      "loss": 5.3917,
      "mean_token_accuracy": 0.15254785865545273,
      "num_tokens": 14052725.0,
      "step": 6400
    },
    {
      "entropy": 5.65068039894104,
      "epoch": 0.6356689162366018,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004967490631502656,
      "loss": 5.4728,
      "mean_token_accuracy": 0.14567496478557587,
      "num_tokens": 14063621.0,
      "step": 6405
    },
    {
      "entropy": 5.735702085494995,
      "epoch": 0.6361651448987694,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004967430591322474,
      "loss": 5.5721,
      "mean_token_accuracy": 0.147471821308136,
      "num_tokens": 14074350.0,
      "step": 6410
    },
    {
      "entropy": 5.694315576553345,
      "epoch": 0.6366613735609369,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004967370496154791,
      "loss": 5.5215,
      "mean_token_accuracy": 0.1483861267566681,
      "num_tokens": 14084600.0,
      "step": 6415
    },
    {
      "entropy": 5.700325345993042,
      "epoch": 0.6371576022231044,
      "grad_norm": 0.875,
      "learning_rate": 0.0004967310346001098,
      "loss": 5.4329,
      "mean_token_accuracy": 0.1556906893849373,
      "num_tokens": 14094739.0,
      "step": 6420
    },
    {
      "entropy": 5.769415760040284,
      "epoch": 0.6376538308852719,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004967250140862885,
      "loss": 5.594,
      "mean_token_accuracy": 0.1452789142727852,
      "num_tokens": 14106020.0,
      "step": 6425
    },
    {
      "entropy": 5.731551599502564,
      "epoch": 0.6381500595474394,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004967189880741647,
      "loss": 5.5722,
      "mean_token_accuracy": 0.14545285180211068,
      "num_tokens": 14117699.0,
      "step": 6430
    },
    {
      "entropy": 5.623116683959961,
      "epoch": 0.638646288209607,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004967129565638878,
      "loss": 5.4402,
      "mean_token_accuracy": 0.1498525395989418,
      "num_tokens": 14128901.0,
      "step": 6435
    },
    {
      "entropy": 5.62310471534729,
      "epoch": 0.6391425168717745,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004967069195556073,
      "loss": 5.4796,
      "mean_token_accuracy": 0.14773380234837533,
      "num_tokens": 14140832.0,
      "step": 6440
    },
    {
      "entropy": 5.678237342834473,
      "epoch": 0.6396387455339421,
      "grad_norm": 0.953125,
      "learning_rate": 0.000496700877049473,
      "loss": 5.5207,
      "mean_token_accuracy": 0.15002197176218032,
      "num_tokens": 14151936.0,
      "step": 6445
    },
    {
      "entropy": 5.700641059875489,
      "epoch": 0.6401349741961095,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004966948290456345,
      "loss": 5.518,
      "mean_token_accuracy": 0.149015374481678,
      "num_tokens": 14160945.0,
      "step": 6450
    },
    {
      "entropy": 5.624499082565308,
      "epoch": 0.6406312028582771,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004966887755442421,
      "loss": 5.4147,
      "mean_token_accuracy": 0.14972492605447768,
      "num_tokens": 14172052.0,
      "step": 6455
    },
    {
      "entropy": 5.65600733757019,
      "epoch": 0.6411274315204446,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004966827165454456,
      "loss": 5.4815,
      "mean_token_accuracy": 0.1499800741672516,
      "num_tokens": 14181788.0,
      "step": 6460
    },
    {
      "entropy": 5.7158653259277346,
      "epoch": 0.6416236601826122,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004966766520493956,
      "loss": 5.5606,
      "mean_token_accuracy": 0.14471029564738275,
      "num_tokens": 14193668.0,
      "step": 6465
    },
    {
      "entropy": 5.702195644378662,
      "epoch": 0.6421198888447797,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004966705820562422,
      "loss": 5.4357,
      "mean_token_accuracy": 0.15261470079421996,
      "num_tokens": 14203665.0,
      "step": 6470
    },
    {
      "entropy": 5.655112648010254,
      "epoch": 0.6426161175069472,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004966645065661361,
      "loss": 5.4806,
      "mean_token_accuracy": 0.14383687674999238,
      "num_tokens": 14214213.0,
      "step": 6475
    },
    {
      "entropy": 5.683132743835449,
      "epoch": 0.6431123461691147,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004966584255792279,
      "loss": 5.5541,
      "mean_token_accuracy": 0.14166881144046783,
      "num_tokens": 14225754.0,
      "step": 6480
    },
    {
      "entropy": 5.671842384338379,
      "epoch": 0.6436085748312823,
      "grad_norm": 0.875,
      "learning_rate": 0.0004966523390956683,
      "loss": 5.4547,
      "mean_token_accuracy": 0.1519224651157856,
      "num_tokens": 14236508.0,
      "step": 6485
    },
    {
      "entropy": 5.710225915908813,
      "epoch": 0.6441048034934498,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004966462471156083,
      "loss": 5.4941,
      "mean_token_accuracy": 0.15028066784143448,
      "num_tokens": 14247719.0,
      "step": 6490
    },
    {
      "entropy": 5.667965984344482,
      "epoch": 0.6446010321556173,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004966401496391989,
      "loss": 5.5546,
      "mean_token_accuracy": 0.13857395723462104,
      "num_tokens": 14258667.0,
      "step": 6495
    },
    {
      "entropy": 5.68690299987793,
      "epoch": 0.6450972608177848,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004966340466665915,
      "loss": 5.5441,
      "mean_token_accuracy": 0.14910787716507912,
      "num_tokens": 14269666.0,
      "step": 6500
    },
    {
      "entropy": 5.591238164901734,
      "epoch": 0.6455934894799523,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004966279381979373,
      "loss": 5.4538,
      "mean_token_accuracy": 0.1470872327685356,
      "num_tokens": 14280862.0,
      "step": 6505
    },
    {
      "entropy": 5.715922689437866,
      "epoch": 0.6460897181421199,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004966218242333879,
      "loss": 5.5827,
      "mean_token_accuracy": 0.13394412621855736,
      "num_tokens": 14292505.0,
      "step": 6510
    },
    {
      "entropy": 5.672628402709961,
      "epoch": 0.6465859468042874,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004966157047730945,
      "loss": 5.3881,
      "mean_token_accuracy": 0.1511433869600296,
      "num_tokens": 14302547.0,
      "step": 6515
    },
    {
      "entropy": 5.617307519912719,
      "epoch": 0.647082175466455,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004966095798172095,
      "loss": 5.5199,
      "mean_token_accuracy": 0.1442197382450104,
      "num_tokens": 14313585.0,
      "step": 6520
    },
    {
      "entropy": 5.6147984027862545,
      "epoch": 0.6475784041286224,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004966034493658843,
      "loss": 5.4025,
      "mean_token_accuracy": 0.14724294617772102,
      "num_tokens": 14323665.0,
      "step": 6525
    },
    {
      "entropy": 5.619630432128906,
      "epoch": 0.64807463279079,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004965973134192711,
      "loss": 5.43,
      "mean_token_accuracy": 0.15175018832087517,
      "num_tokens": 14334257.0,
      "step": 6530
    },
    {
      "entropy": 5.702003145217896,
      "epoch": 0.6485708614529575,
      "grad_norm": 0.9375,
      "learning_rate": 0.000496591171977522,
      "loss": 5.4548,
      "mean_token_accuracy": 0.15189530849456787,
      "num_tokens": 14345350.0,
      "step": 6535
    },
    {
      "entropy": 5.690001726150513,
      "epoch": 0.6490670901151251,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004965850250407893,
      "loss": 5.4552,
      "mean_token_accuracy": 0.15488422363996507,
      "num_tokens": 14355593.0,
      "step": 6540
    },
    {
      "entropy": 5.604142856597901,
      "epoch": 0.6495633187772926,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004965788726092256,
      "loss": 5.3459,
      "mean_token_accuracy": 0.1584514856338501,
      "num_tokens": 14366405.0,
      "step": 6545
    },
    {
      "entropy": 5.601662302017212,
      "epoch": 0.6500595474394602,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004965727146829832,
      "loss": 5.4235,
      "mean_token_accuracy": 0.14971744865179062,
      "num_tokens": 14377228.0,
      "step": 6550
    },
    {
      "entropy": 5.766236972808838,
      "epoch": 0.6505557761016276,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004965665512622148,
      "loss": 5.576,
      "mean_token_accuracy": 0.14476309791207315,
      "num_tokens": 14387113.0,
      "step": 6555
    },
    {
      "entropy": 5.6732823848724365,
      "epoch": 0.6510520047637952,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004965603823470735,
      "loss": 5.46,
      "mean_token_accuracy": 0.15242564380168916,
      "num_tokens": 14397980.0,
      "step": 6560
    },
    {
      "entropy": 5.6098064422607425,
      "epoch": 0.6515482334259627,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004965542079377122,
      "loss": 5.3773,
      "mean_token_accuracy": 0.15433671772480012,
      "num_tokens": 14409063.0,
      "step": 6565
    },
    {
      "entropy": 5.568036031723023,
      "epoch": 0.6520444620881302,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004965480280342839,
      "loss": 5.364,
      "mean_token_accuracy": 0.1615317314863205,
      "num_tokens": 14421056.0,
      "step": 6570
    },
    {
      "entropy": 5.611463165283203,
      "epoch": 0.6525406907502977,
      "grad_norm": 0.875,
      "learning_rate": 0.0004965418426369419,
      "loss": 5.353,
      "mean_token_accuracy": 0.15946202576160431,
      "num_tokens": 14430778.0,
      "step": 6575
    },
    {
      "entropy": 5.614991188049316,
      "epoch": 0.6530369194124652,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004965356517458395,
      "loss": 5.3768,
      "mean_token_accuracy": 0.15489154756069184,
      "num_tokens": 14441544.0,
      "step": 6580
    },
    {
      "entropy": 5.607758903503418,
      "epoch": 0.6535331480746328,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004965294553611303,
      "loss": 5.5202,
      "mean_token_accuracy": 0.15016790255904197,
      "num_tokens": 14453062.0,
      "step": 6585
    },
    {
      "entropy": 5.680311822891236,
      "epoch": 0.6540293767368003,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000496523253482968,
      "loss": 5.4446,
      "mean_token_accuracy": 0.15107833594083786,
      "num_tokens": 14464402.0,
      "step": 6590
    },
    {
      "entropy": 5.682463645935059,
      "epoch": 0.6545256053989679,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004965170461115065,
      "loss": 5.536,
      "mean_token_accuracy": 0.1490773156285286,
      "num_tokens": 14476595.0,
      "step": 6595
    },
    {
      "entropy": 5.70050892829895,
      "epoch": 0.6550218340611353,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004965108332468994,
      "loss": 5.5022,
      "mean_token_accuracy": 0.1478926882147789,
      "num_tokens": 14487298.0,
      "step": 6600
    },
    {
      "entropy": 5.653990840911865,
      "epoch": 0.6555180627233029,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004965046148893011,
      "loss": 5.4201,
      "mean_token_accuracy": 0.15270100384950638,
      "num_tokens": 14499708.0,
      "step": 6605
    },
    {
      "entropy": 5.637703609466553,
      "epoch": 0.6560142913854704,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004964983910388654,
      "loss": 5.4779,
      "mean_token_accuracy": 0.14862264692783356,
      "num_tokens": 14511773.0,
      "step": 6610
    },
    {
      "entropy": 5.576990079879761,
      "epoch": 0.656510520047638,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004964921616957471,
      "loss": 5.3809,
      "mean_token_accuracy": 0.14810700714588165,
      "num_tokens": 14523024.0,
      "step": 6615
    },
    {
      "entropy": 5.658850431442261,
      "epoch": 0.6570067487098055,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004964859268601005,
      "loss": 5.4527,
      "mean_token_accuracy": 0.15175904259085654,
      "num_tokens": 14532915.0,
      "step": 6620
    },
    {
      "entropy": 5.683317947387695,
      "epoch": 0.657502977371973,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004964796865320801,
      "loss": 5.4417,
      "mean_token_accuracy": 0.1551850751042366,
      "num_tokens": 14543627.0,
      "step": 6625
    },
    {
      "entropy": 5.5610107421875,
      "epoch": 0.6579992060341405,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0004964734407118406,
      "loss": 5.3949,
      "mean_token_accuracy": 0.15504732429981233,
      "num_tokens": 14555181.0,
      "step": 6630
    },
    {
      "entropy": 5.605614471435547,
      "epoch": 0.6584954346963081,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004964671893995372,
      "loss": 5.4025,
      "mean_token_accuracy": 0.1500137612223625,
      "num_tokens": 14565542.0,
      "step": 6635
    },
    {
      "entropy": 5.685685634613037,
      "epoch": 0.6589916633584756,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004964609325953245,
      "loss": 5.4629,
      "mean_token_accuracy": 0.15267691314220427,
      "num_tokens": 14575395.0,
      "step": 6640
    },
    {
      "entropy": 5.671268796920776,
      "epoch": 0.6594878920206431,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0004964546702993581,
      "loss": 5.499,
      "mean_token_accuracy": 0.15098451375961303,
      "num_tokens": 14587206.0,
      "step": 6645
    },
    {
      "entropy": 5.686512517929077,
      "epoch": 0.6599841206828106,
      "grad_norm": 0.7734375,
      "learning_rate": 0.000496448402511793,
      "loss": 5.505,
      "mean_token_accuracy": 0.15307042002677917,
      "num_tokens": 14599048.0,
      "step": 6650
    },
    {
      "entropy": 5.68162202835083,
      "epoch": 0.6604803493449781,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004964421292327847,
      "loss": 5.4835,
      "mean_token_accuracy": 0.1511673018336296,
      "num_tokens": 14609633.0,
      "step": 6655
    },
    {
      "entropy": 5.630779695510864,
      "epoch": 0.6609765780071457,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004964358504624887,
      "loss": 5.381,
      "mean_token_accuracy": 0.15312047600746154,
      "num_tokens": 14620373.0,
      "step": 6660
    },
    {
      "entropy": 5.623717641830444,
      "epoch": 0.6614728066693132,
      "grad_norm": 0.875,
      "learning_rate": 0.0004964295662010608,
      "loss": 5.3234,
      "mean_token_accuracy": 0.15719757080078126,
      "num_tokens": 14630463.0,
      "step": 6665
    },
    {
      "entropy": 5.7358582496643065,
      "epoch": 0.6619690353314808,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004964232764486568,
      "loss": 5.5689,
      "mean_token_accuracy": 0.1434595428407192,
      "num_tokens": 14641523.0,
      "step": 6670
    },
    {
      "entropy": 5.620292282104492,
      "epoch": 0.6624652639936482,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004964169812054326,
      "loss": 5.398,
      "mean_token_accuracy": 0.1509662389755249,
      "num_tokens": 14651812.0,
      "step": 6675
    },
    {
      "entropy": 5.62327766418457,
      "epoch": 0.6629614926558158,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004964106804715444,
      "loss": 5.3856,
      "mean_token_accuracy": 0.15855159610509872,
      "num_tokens": 14661982.0,
      "step": 6680
    },
    {
      "entropy": 5.599073123931885,
      "epoch": 0.6634577213179833,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004964043742471486,
      "loss": 5.4146,
      "mean_token_accuracy": 0.15977433919906617,
      "num_tokens": 14671981.0,
      "step": 6685
    },
    {
      "entropy": 5.659870243072509,
      "epoch": 0.6639539499801509,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004963980625324012,
      "loss": 5.4266,
      "mean_token_accuracy": 0.1492961347103119,
      "num_tokens": 14682478.0,
      "step": 6690
    },
    {
      "entropy": 5.6607927799224855,
      "epoch": 0.6644501786423184,
      "grad_norm": 0.78515625,
      "learning_rate": 0.000496391745327459,
      "loss": 5.4959,
      "mean_token_accuracy": 0.1413546122610569,
      "num_tokens": 14695197.0,
      "step": 6695
    },
    {
      "entropy": 5.6239784240722654,
      "epoch": 0.664946407304486,
      "grad_norm": 0.75390625,
      "learning_rate": 0.0004963854226324785,
      "loss": 5.4102,
      "mean_token_accuracy": 0.15093137323856354,
      "num_tokens": 14705702.0,
      "step": 6700
    },
    {
      "entropy": 5.663140916824341,
      "epoch": 0.6654426359666534,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004963790944476167,
      "loss": 5.4238,
      "mean_token_accuracy": 0.15778421014547347,
      "num_tokens": 14715370.0,
      "step": 6705
    },
    {
      "entropy": 5.676147842407227,
      "epoch": 0.665938864628821,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004963727607730303,
      "loss": 5.4436,
      "mean_token_accuracy": 0.15048911571502685,
      "num_tokens": 14725481.0,
      "step": 6710
    },
    {
      "entropy": 5.605512475967407,
      "epoch": 0.6664350932909885,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004963664216088764,
      "loss": 5.3706,
      "mean_token_accuracy": 0.1536024570465088,
      "num_tokens": 14735250.0,
      "step": 6715
    },
    {
      "entropy": 5.615908527374268,
      "epoch": 0.666931321953156,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004963600769553124,
      "loss": 5.3774,
      "mean_token_accuracy": 0.15871376693248748,
      "num_tokens": 14744741.0,
      "step": 6720
    },
    {
      "entropy": 5.659100103378296,
      "epoch": 0.6674275506153236,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004963537268124953,
      "loss": 5.409,
      "mean_token_accuracy": 0.15116407573223115,
      "num_tokens": 14755865.0,
      "step": 6725
    },
    {
      "entropy": 5.574926137924194,
      "epoch": 0.667923779277491,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004963473711805829,
      "loss": 5.3635,
      "mean_token_accuracy": 0.1572342410683632,
      "num_tokens": 14766154.0,
      "step": 6730
    },
    {
      "entropy": 5.636857652664185,
      "epoch": 0.6684200079396586,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004963410100597326,
      "loss": 5.4038,
      "mean_token_accuracy": 0.15145311057567595,
      "num_tokens": 14776940.0,
      "step": 6735
    },
    {
      "entropy": 5.624228191375733,
      "epoch": 0.6689162366018261,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004963346434501021,
      "loss": 5.5153,
      "mean_token_accuracy": 0.14711894020438193,
      "num_tokens": 14788367.0,
      "step": 6740
    },
    {
      "entropy": 5.558290672302246,
      "epoch": 0.6694124652639937,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004963282713518495,
      "loss": 5.427,
      "mean_token_accuracy": 0.15619889050722122,
      "num_tokens": 14798751.0,
      "step": 6745
    },
    {
      "entropy": 5.571211624145508,
      "epoch": 0.6699086939261611,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004963218937651326,
      "loss": 5.3549,
      "mean_token_accuracy": 0.15779270827770234,
      "num_tokens": 14808748.0,
      "step": 6750
    },
    {
      "entropy": 5.694025945663452,
      "epoch": 0.6704049225883287,
      "grad_norm": 0.77734375,
      "learning_rate": 0.0004963155106901098,
      "loss": 5.5211,
      "mean_token_accuracy": 0.14606630355119704,
      "num_tokens": 14820524.0,
      "step": 6755
    },
    {
      "entropy": 5.737807846069336,
      "epoch": 0.6709011512504962,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004963091221269391,
      "loss": 5.5529,
      "mean_token_accuracy": 0.1397723875939846,
      "num_tokens": 14832485.0,
      "step": 6760
    },
    {
      "entropy": 5.638651561737061,
      "epoch": 0.6713973799126638,
      "grad_norm": 0.828125,
      "learning_rate": 0.000496302728075779,
      "loss": 5.3592,
      "mean_token_accuracy": 0.1506928727030754,
      "num_tokens": 14843951.0,
      "step": 6765
    },
    {
      "entropy": 5.599717283248902,
      "epoch": 0.6718936085748313,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004962963285367882,
      "loss": 5.4688,
      "mean_token_accuracy": 0.1526104286313057,
      "num_tokens": 14855502.0,
      "step": 6770
    },
    {
      "entropy": 5.599081659317017,
      "epoch": 0.6723898372369989,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004962899235101253,
      "loss": 5.3763,
      "mean_token_accuracy": 0.15440474897623063,
      "num_tokens": 14866305.0,
      "step": 6775
    },
    {
      "entropy": 5.632352066040039,
      "epoch": 0.6728860658991663,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004962835129959491,
      "loss": 5.3947,
      "mean_token_accuracy": 0.14619243144989014,
      "num_tokens": 14877222.0,
      "step": 6780
    },
    {
      "entropy": 5.59601092338562,
      "epoch": 0.6733822945613339,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004962770969944186,
      "loss": 5.4778,
      "mean_token_accuracy": 0.14912765175104142,
      "num_tokens": 14888099.0,
      "step": 6785
    },
    {
      "entropy": 5.617529344558716,
      "epoch": 0.6738785232235014,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000496270675505693,
      "loss": 5.3948,
      "mean_token_accuracy": 0.15923409014940262,
      "num_tokens": 14899042.0,
      "step": 6790
    },
    {
      "entropy": 5.664507579803467,
      "epoch": 0.6743747518856689,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004962642485299314,
      "loss": 5.4518,
      "mean_token_accuracy": 0.14887820482254027,
      "num_tokens": 14908837.0,
      "step": 6795
    },
    {
      "entropy": 5.581243324279785,
      "epoch": 0.6748709805478365,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004962578160672931,
      "loss": 5.3936,
      "mean_token_accuracy": 0.15571052581071854,
      "num_tokens": 14920773.0,
      "step": 6800
    },
    {
      "entropy": 5.607238864898681,
      "epoch": 0.6753672092100039,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004962513781179379,
      "loss": 5.3926,
      "mean_token_accuracy": 0.15196526497602464,
      "num_tokens": 14930970.0,
      "step": 6805
    },
    {
      "entropy": 5.623500776290894,
      "epoch": 0.6758634378721715,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004962449346820251,
      "loss": 5.4507,
      "mean_token_accuracy": 0.148509182035923,
      "num_tokens": 14942378.0,
      "step": 6810
    },
    {
      "entropy": 5.616042423248291,
      "epoch": 0.676359666534339,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004962384857597148,
      "loss": 5.4876,
      "mean_token_accuracy": 0.14774132370948792,
      "num_tokens": 14953020.0,
      "step": 6815
    },
    {
      "entropy": 5.712278175354004,
      "epoch": 0.6768558951965066,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004962320313511668,
      "loss": 5.5508,
      "mean_token_accuracy": 0.1478928282856941,
      "num_tokens": 14962340.0,
      "step": 6820
    },
    {
      "entropy": 5.709672260284424,
      "epoch": 0.677352123858674,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004962255714565409,
      "loss": 5.4271,
      "mean_token_accuracy": 0.15248478353023528,
      "num_tokens": 14972792.0,
      "step": 6825
    },
    {
      "entropy": 5.645695447921753,
      "epoch": 0.6778483525208416,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004962191060759978,
      "loss": 5.425,
      "mean_token_accuracy": 0.15230261087417601,
      "num_tokens": 14983698.0,
      "step": 6830
    },
    {
      "entropy": 5.651081132888794,
      "epoch": 0.6783445811830091,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004962126352096973,
      "loss": 5.3993,
      "mean_token_accuracy": 0.151994052529335,
      "num_tokens": 14994583.0,
      "step": 6835
    },
    {
      "entropy": 5.557571125030518,
      "epoch": 0.6788408098451767,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004962061588578004,
      "loss": 5.3352,
      "mean_token_accuracy": 0.15464683324098588,
      "num_tokens": 15005325.0,
      "step": 6840
    },
    {
      "entropy": 5.658808660507202,
      "epoch": 0.6793370385073442,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004961996770204673,
      "loss": 5.4081,
      "mean_token_accuracy": 0.1543305680155754,
      "num_tokens": 15015155.0,
      "step": 6845
    },
    {
      "entropy": 5.643760108947754,
      "epoch": 0.6798332671695118,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004961931896978588,
      "loss": 5.472,
      "mean_token_accuracy": 0.14792029708623886,
      "num_tokens": 15026529.0,
      "step": 6850
    },
    {
      "entropy": 5.5619231224060055,
      "epoch": 0.6803294958316792,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004961866968901359,
      "loss": 5.3866,
      "mean_token_accuracy": 0.15590434223413469,
      "num_tokens": 15037632.0,
      "step": 6855
    },
    {
      "entropy": 5.61463098526001,
      "epoch": 0.6808257244938468,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004961801985974595,
      "loss": 5.4668,
      "mean_token_accuracy": 0.1518971860408783,
      "num_tokens": 15050397.0,
      "step": 6860
    },
    {
      "entropy": 5.609272766113281,
      "epoch": 0.6813219531560143,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004961736948199907,
      "loss": 5.3782,
      "mean_token_accuracy": 0.1546167850494385,
      "num_tokens": 15061505.0,
      "step": 6865
    },
    {
      "entropy": 5.7006994724273685,
      "epoch": 0.6818181818181818,
      "grad_norm": 0.875,
      "learning_rate": 0.0004961671855578909,
      "loss": 5.5625,
      "mean_token_accuracy": 0.147130286693573,
      "num_tokens": 15073752.0,
      "step": 6870
    },
    {
      "entropy": 5.666177415847779,
      "epoch": 0.6823144104803494,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004961606708113216,
      "loss": 5.4674,
      "mean_token_accuracy": 0.14928337261080743,
      "num_tokens": 15084730.0,
      "step": 6875
    },
    {
      "entropy": 5.645604085922241,
      "epoch": 0.6828106391425168,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004961541505804441,
      "loss": 5.371,
      "mean_token_accuracy": 0.15932557433843614,
      "num_tokens": 15095811.0,
      "step": 6880
    },
    {
      "entropy": 5.6254149913787845,
      "epoch": 0.6833068678046844,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004961476248654202,
      "loss": 5.3637,
      "mean_token_accuracy": 0.1530770927667618,
      "num_tokens": 15106102.0,
      "step": 6885
    },
    {
      "entropy": 5.635340595245362,
      "epoch": 0.6838030964668519,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004961410936664118,
      "loss": 5.4507,
      "mean_token_accuracy": 0.14996094182133674,
      "num_tokens": 15115907.0,
      "step": 6890
    },
    {
      "entropy": 5.613401889801025,
      "epoch": 0.6842993251290195,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004961345569835807,
      "loss": 5.4362,
      "mean_token_accuracy": 0.14937734454870225,
      "num_tokens": 15126788.0,
      "step": 6895
    },
    {
      "entropy": 5.6303540706634525,
      "epoch": 0.684795553791187,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004961280148170892,
      "loss": 5.3864,
      "mean_token_accuracy": 0.1516811415553093,
      "num_tokens": 15137814.0,
      "step": 6900
    },
    {
      "entropy": 5.699887800216675,
      "epoch": 0.6852917824533545,
      "grad_norm": 1.0,
      "learning_rate": 0.0004961214671670994,
      "loss": 5.6138,
      "mean_token_accuracy": 0.1388474315404892,
      "num_tokens": 15148597.0,
      "step": 6905
    },
    {
      "entropy": 5.662448787689209,
      "epoch": 0.685788011115522,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004961149140337737,
      "loss": 5.4277,
      "mean_token_accuracy": 0.1468920439481735,
      "num_tokens": 15158985.0,
      "step": 6910
    },
    {
      "entropy": 5.618693494796753,
      "epoch": 0.6862842397776896,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004961083554172745,
      "loss": 5.4185,
      "mean_token_accuracy": 0.15202739387750625,
      "num_tokens": 15170242.0,
      "step": 6915
    },
    {
      "entropy": 5.596062803268433,
      "epoch": 0.6867804684398571,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004961017913177646,
      "loss": 5.3707,
      "mean_token_accuracy": 0.1583961695432663,
      "num_tokens": 15180353.0,
      "step": 6920
    },
    {
      "entropy": 5.544697618484497,
      "epoch": 0.6872766971020247,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004960952217354066,
      "loss": 5.3256,
      "mean_token_accuracy": 0.16041142493486404,
      "num_tokens": 15191955.0,
      "step": 6925
    },
    {
      "entropy": 5.5882710933685305,
      "epoch": 0.6877729257641921,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004960886466703636,
      "loss": 5.411,
      "mean_token_accuracy": 0.15261018127202988,
      "num_tokens": 15202088.0,
      "step": 6930
    },
    {
      "entropy": 5.533545112609863,
      "epoch": 0.6882691544263597,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004960820661227987,
      "loss": 5.3694,
      "mean_token_accuracy": 0.15650452077388763,
      "num_tokens": 15212291.0,
      "step": 6935
    },
    {
      "entropy": 5.591097450256347,
      "epoch": 0.6887653830885272,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004960754800928748,
      "loss": 5.4021,
      "mean_token_accuracy": 0.15690786093473436,
      "num_tokens": 15223315.0,
      "step": 6940
    },
    {
      "entropy": 5.718151521682739,
      "epoch": 0.6892616117506947,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004960688885807553,
      "loss": 5.52,
      "mean_token_accuracy": 0.15194137990474701,
      "num_tokens": 15233416.0,
      "step": 6945
    },
    {
      "entropy": 5.639258575439453,
      "epoch": 0.6897578404128623,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004960622915866038,
      "loss": 5.3451,
      "mean_token_accuracy": 0.16188562661409378,
      "num_tokens": 15242951.0,
      "step": 6950
    },
    {
      "entropy": 5.5644447803497314,
      "epoch": 0.6902540690750297,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004960556891105839,
      "loss": 5.4409,
      "mean_token_accuracy": 0.1520864710211754,
      "num_tokens": 15253145.0,
      "step": 6955
    },
    {
      "entropy": 5.5503040790557865,
      "epoch": 0.6907502977371973,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004960490811528591,
      "loss": 5.3893,
      "mean_token_accuracy": 0.15682558864355087,
      "num_tokens": 15263712.0,
      "step": 6960
    },
    {
      "entropy": 5.753786373138428,
      "epoch": 0.6912465263993648,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004960424677135935,
      "loss": 5.5061,
      "mean_token_accuracy": 0.15282690972089769,
      "num_tokens": 15273609.0,
      "step": 6965
    },
    {
      "entropy": 5.620067453384399,
      "epoch": 0.6917427550615324,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004960358487929511,
      "loss": 5.3677,
      "mean_token_accuracy": 0.16024328172206878,
      "num_tokens": 15283934.0,
      "step": 6970
    },
    {
      "entropy": 5.573494482040405,
      "epoch": 0.6922389837236999,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004960292243910958,
      "loss": 5.4345,
      "mean_token_accuracy": 0.1531072050333023,
      "num_tokens": 15294394.0,
      "step": 6975
    },
    {
      "entropy": 5.604720544815064,
      "epoch": 0.6927352123858674,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000496022594508192,
      "loss": 5.3812,
      "mean_token_accuracy": 0.16022567003965377,
      "num_tokens": 15304622.0,
      "step": 6980
    },
    {
      "entropy": 5.60088357925415,
      "epoch": 0.6932314410480349,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004960159591444041,
      "loss": 5.2983,
      "mean_token_accuracy": 0.16009942144155503,
      "num_tokens": 15314094.0,
      "step": 6985
    },
    {
      "entropy": 5.567172193527222,
      "epoch": 0.6937276697102025,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004960093182998968,
      "loss": 5.4139,
      "mean_token_accuracy": 0.15786027461290358,
      "num_tokens": 15324313.0,
      "step": 6990
    },
    {
      "entropy": 5.530042266845703,
      "epoch": 0.69422389837237,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004960026719748346,
      "loss": 5.2982,
      "mean_token_accuracy": 0.17230096459388733,
      "num_tokens": 15335195.0,
      "step": 6995
    },
    {
      "entropy": 5.735844707489013,
      "epoch": 0.6947201270345376,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004959960201693822,
      "loss": 5.5125,
      "mean_token_accuracy": 0.15034665167331696,
      "num_tokens": 15345798.0,
      "step": 7000
    },
    {
      "entropy": 5.632916975021362,
      "epoch": 0.695216355696705,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004959893628837048,
      "loss": 5.3691,
      "mean_token_accuracy": 0.15174235850572587,
      "num_tokens": 15356340.0,
      "step": 7005
    },
    {
      "entropy": 5.639342164993286,
      "epoch": 0.6957125843588726,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004959827001179675,
      "loss": 5.4657,
      "mean_token_accuracy": 0.14688759297132492,
      "num_tokens": 15367795.0,
      "step": 7010
    },
    {
      "entropy": 5.632861709594726,
      "epoch": 0.6962088130210401,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004959760318723353,
      "loss": 5.5909,
      "mean_token_accuracy": 0.1438564896583557,
      "num_tokens": 15379214.0,
      "step": 7015
    },
    {
      "entropy": 5.636499547958374,
      "epoch": 0.6967050416832076,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004959693581469736,
      "loss": 5.4438,
      "mean_token_accuracy": 0.1585008829832077,
      "num_tokens": 15390242.0,
      "step": 7020
    },
    {
      "entropy": 5.600987768173217,
      "epoch": 0.6972012703453752,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004959626789420481,
      "loss": 5.3733,
      "mean_token_accuracy": 0.1531446322798729,
      "num_tokens": 15402600.0,
      "step": 7025
    },
    {
      "entropy": 5.60857572555542,
      "epoch": 0.6976974990075426,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004959559942577243,
      "loss": 5.4076,
      "mean_token_accuracy": 0.15517537593841552,
      "num_tokens": 15412883.0,
      "step": 7030
    },
    {
      "entropy": 5.653280925750733,
      "epoch": 0.6981937276697102,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004959493040941678,
      "loss": 5.4398,
      "mean_token_accuracy": 0.149055640399456,
      "num_tokens": 15425919.0,
      "step": 7035
    },
    {
      "entropy": 5.598724746704102,
      "epoch": 0.6986899563318777,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004959426084515448,
      "loss": 5.3766,
      "mean_token_accuracy": 0.15377127379179,
      "num_tokens": 15437604.0,
      "step": 7040
    },
    {
      "entropy": 5.588111400604248,
      "epoch": 0.6991861849940453,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004959359073300212,
      "loss": 5.442,
      "mean_token_accuracy": 0.1484801024198532,
      "num_tokens": 15449023.0,
      "step": 7045
    },
    {
      "entropy": 5.67562313079834,
      "epoch": 0.6996824136562128,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004959292007297631,
      "loss": 5.4976,
      "mean_token_accuracy": 0.15364296585321427,
      "num_tokens": 15459567.0,
      "step": 7050
    },
    {
      "entropy": 5.648920297622681,
      "epoch": 0.7001786423183803,
      "grad_norm": 0.8046875,
      "learning_rate": 0.000495922488650937,
      "loss": 5.409,
      "mean_token_accuracy": 0.156674388051033,
      "num_tokens": 15470950.0,
      "step": 7055
    },
    {
      "entropy": 5.588835096359253,
      "epoch": 0.7006748709805478,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004959157710937092,
      "loss": 5.4154,
      "mean_token_accuracy": 0.15452724248170852,
      "num_tokens": 15481258.0,
      "step": 7060
    },
    {
      "entropy": 5.636629056930542,
      "epoch": 0.7011710996427154,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004959090480582462,
      "loss": 5.3459,
      "mean_token_accuracy": 0.15606230050325393,
      "num_tokens": 15491672.0,
      "step": 7065
    },
    {
      "entropy": 5.559443807601928,
      "epoch": 0.7016673283048829,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004959023195447148,
      "loss": 5.4015,
      "mean_token_accuracy": 0.15582506060600282,
      "num_tokens": 15502342.0,
      "step": 7070
    },
    {
      "entropy": 5.6122541427612305,
      "epoch": 0.7021635569670505,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000495895585553282,
      "loss": 5.4419,
      "mean_token_accuracy": 0.1493886798620224,
      "num_tokens": 15513430.0,
      "step": 7075
    },
    {
      "entropy": 5.807034015655518,
      "epoch": 0.7026597856292179,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004958888460841147,
      "loss": 5.543,
      "mean_token_accuracy": 0.1481260783970356,
      "num_tokens": 15524889.0,
      "step": 7080
    },
    {
      "entropy": 5.5820581912994385,
      "epoch": 0.7031560142913855,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004958821011373798,
      "loss": 5.3579,
      "mean_token_accuracy": 0.15555054545402527,
      "num_tokens": 15534998.0,
      "step": 7085
    },
    {
      "entropy": 5.565367937088013,
      "epoch": 0.703652242953553,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004958753507132449,
      "loss": 5.3706,
      "mean_token_accuracy": 0.15571859925985337,
      "num_tokens": 15544752.0,
      "step": 7090
    },
    {
      "entropy": 5.614150714874268,
      "epoch": 0.7041484716157205,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004958685948118772,
      "loss": 5.3921,
      "mean_token_accuracy": 0.15620832443237304,
      "num_tokens": 15555490.0,
      "step": 7095
    },
    {
      "entropy": 5.582898569107056,
      "epoch": 0.7046447002778881,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004958618334334443,
      "loss": 5.3252,
      "mean_token_accuracy": 0.1558916062116623,
      "num_tokens": 15565752.0,
      "step": 7100
    },
    {
      "entropy": 5.594721269607544,
      "epoch": 0.7051409289400555,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004958550665781138,
      "loss": 5.4522,
      "mean_token_accuracy": 0.15302032232284546,
      "num_tokens": 15576368.0,
      "step": 7105
    },
    {
      "entropy": 5.696494007110596,
      "epoch": 0.7056371576022231,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004958482942460536,
      "loss": 5.4426,
      "mean_token_accuracy": 0.15190984383225442,
      "num_tokens": 15587972.0,
      "step": 7110
    },
    {
      "entropy": 5.705682229995728,
      "epoch": 0.7061333862643906,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004958415164374316,
      "loss": 5.4789,
      "mean_token_accuracy": 0.15410581678152085,
      "num_tokens": 15599951.0,
      "step": 7115
    },
    {
      "entropy": 5.595183038711548,
      "epoch": 0.7066296149265582,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004958347331524158,
      "loss": 5.4646,
      "mean_token_accuracy": 0.1483406215906143,
      "num_tokens": 15610183.0,
      "step": 7120
    },
    {
      "entropy": 5.609962606430054,
      "epoch": 0.7071258435887257,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004958279443911745,
      "loss": 5.484,
      "mean_token_accuracy": 0.14564593136310577,
      "num_tokens": 15621074.0,
      "step": 7125
    },
    {
      "entropy": 5.703958988189697,
      "epoch": 0.7076220722508932,
      "grad_norm": 0.890625,
      "learning_rate": 0.000495821150153876,
      "loss": 5.48,
      "mean_token_accuracy": 0.14798002392053605,
      "num_tokens": 15632752.0,
      "step": 7130
    },
    {
      "entropy": 5.608168315887451,
      "epoch": 0.7081183009130607,
      "grad_norm": 0.9375,
      "learning_rate": 0.000495814350440689,
      "loss": 5.4443,
      "mean_token_accuracy": 0.14783835858106614,
      "num_tokens": 15644703.0,
      "step": 7135
    },
    {
      "entropy": 5.590032243728638,
      "epoch": 0.7086145295752283,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004958075452517816,
      "loss": 5.4155,
      "mean_token_accuracy": 0.15188954621553422,
      "num_tokens": 15655278.0,
      "step": 7140
    },
    {
      "entropy": 5.609332180023193,
      "epoch": 0.7091107582373958,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004958007345873232,
      "loss": 5.4376,
      "mean_token_accuracy": 0.1562029853463173,
      "num_tokens": 15665998.0,
      "step": 7145
    },
    {
      "entropy": 5.614063835144043,
      "epoch": 0.7096069868995634,
      "grad_norm": 0.875,
      "learning_rate": 0.0004957939184474821,
      "loss": 5.3326,
      "mean_token_accuracy": 0.15620039105415345,
      "num_tokens": 15676343.0,
      "step": 7150
    },
    {
      "entropy": 5.6093193054199215,
      "epoch": 0.7101032155617308,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004957870968324277,
      "loss": 5.4335,
      "mean_token_accuracy": 0.14923603609204292,
      "num_tokens": 15686520.0,
      "step": 7155
    },
    {
      "entropy": 5.624302196502685,
      "epoch": 0.7105994442238984,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004957802697423291,
      "loss": 5.5335,
      "mean_token_accuracy": 0.14949338436126708,
      "num_tokens": 15698302.0,
      "step": 7160
    },
    {
      "entropy": 5.577978181838989,
      "epoch": 0.7110956728860659,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004957734371773554,
      "loss": 5.2967,
      "mean_token_accuracy": 0.15855092704296112,
      "num_tokens": 15708462.0,
      "step": 7165
    },
    {
      "entropy": 5.615873384475708,
      "epoch": 0.7115919015482334,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004957665991376763,
      "loss": 5.4112,
      "mean_token_accuracy": 0.15870108604431152,
      "num_tokens": 15719234.0,
      "step": 7170
    },
    {
      "entropy": 5.5790088176727295,
      "epoch": 0.712088130210401,
      "grad_norm": 0.76953125,
      "learning_rate": 0.0004957597556234613,
      "loss": 5.4382,
      "mean_token_accuracy": 0.14830014258623123,
      "num_tokens": 15732245.0,
      "step": 7175
    },
    {
      "entropy": 5.621543312072754,
      "epoch": 0.7125843588725684,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004957529066348799,
      "loss": 5.4147,
      "mean_token_accuracy": 0.15519314706325532,
      "num_tokens": 15742403.0,
      "step": 7180
    },
    {
      "entropy": 5.536786890029907,
      "epoch": 0.713080587534736,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004957460521721022,
      "loss": 5.3183,
      "mean_token_accuracy": 0.16017779111862182,
      "num_tokens": 15752100.0,
      "step": 7185
    },
    {
      "entropy": 5.575576448440552,
      "epoch": 0.7135768161969035,
      "grad_norm": 0.890625,
      "learning_rate": 0.000495739192235298,
      "loss": 5.4034,
      "mean_token_accuracy": 0.15700195580720902,
      "num_tokens": 15762315.0,
      "step": 7190
    },
    {
      "entropy": 5.581356811523437,
      "epoch": 0.7140730448590711,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004957323268246376,
      "loss": 5.4223,
      "mean_token_accuracy": 0.15435817688703538,
      "num_tokens": 15773841.0,
      "step": 7195
    },
    {
      "entropy": 5.6099649429321286,
      "epoch": 0.7145692735212386,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004957254559402911,
      "loss": 5.3688,
      "mean_token_accuracy": 0.15388270914554597,
      "num_tokens": 15785511.0,
      "step": 7200
    },
    {
      "entropy": 5.574761009216308,
      "epoch": 0.7150655021834061,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004957185795824288,
      "loss": 5.4027,
      "mean_token_accuracy": 0.14808241724967958,
      "num_tokens": 15795394.0,
      "step": 7205
    },
    {
      "entropy": 5.578614854812622,
      "epoch": 0.7155617308455736,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004957116977512214,
      "loss": 5.3716,
      "mean_token_accuracy": 0.15402503162622452,
      "num_tokens": 15805894.0,
      "step": 7210
    },
    {
      "entropy": 5.67041802406311,
      "epoch": 0.7160579595077412,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004957048104468395,
      "loss": 5.4228,
      "mean_token_accuracy": 0.14573368281126023,
      "num_tokens": 15816595.0,
      "step": 7215
    },
    {
      "entropy": 5.591230201721191,
      "epoch": 0.7165541881699087,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004956979176694539,
      "loss": 5.3706,
      "mean_token_accuracy": 0.1590556561946869,
      "num_tokens": 15827638.0,
      "step": 7220
    },
    {
      "entropy": 5.582176876068115,
      "epoch": 0.7170504168320763,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004956910194192355,
      "loss": 5.4427,
      "mean_token_accuracy": 0.15021601766347886,
      "num_tokens": 15838437.0,
      "step": 7225
    },
    {
      "entropy": 5.712309551239014,
      "epoch": 0.7175466454942437,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004956841156963553,
      "loss": 5.5378,
      "mean_token_accuracy": 0.15349817723035813,
      "num_tokens": 15850110.0,
      "step": 7230
    },
    {
      "entropy": 5.600798845291138,
      "epoch": 0.7180428741564113,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004956772065009846,
      "loss": 5.4096,
      "mean_token_accuracy": 0.14814347252249718,
      "num_tokens": 15861786.0,
      "step": 7235
    },
    {
      "entropy": 5.612291049957276,
      "epoch": 0.7185391028185788,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004956702918332948,
      "loss": 5.3802,
      "mean_token_accuracy": 0.15146298035979272,
      "num_tokens": 15871574.0,
      "step": 7240
    },
    {
      "entropy": 5.6134391784667965,
      "epoch": 0.7190353314807463,
      "grad_norm": 0.875,
      "learning_rate": 0.0004956633716934572,
      "loss": 5.3772,
      "mean_token_accuracy": 0.15983042120933533,
      "num_tokens": 15881944.0,
      "step": 7245
    },
    {
      "entropy": 5.58657751083374,
      "epoch": 0.7195315601429139,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004956564460816433,
      "loss": 5.3772,
      "mean_token_accuracy": 0.1547867901623249,
      "num_tokens": 15892511.0,
      "step": 7250
    },
    {
      "entropy": 5.601710176467895,
      "epoch": 0.7200277888050813,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004956495149980252,
      "loss": 5.3936,
      "mean_token_accuracy": 0.14862362146377564,
      "num_tokens": 15903387.0,
      "step": 7255
    },
    {
      "entropy": 5.620168542861938,
      "epoch": 0.7205240174672489,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004956425784427745,
      "loss": 5.437,
      "mean_token_accuracy": 0.15066849887371064,
      "num_tokens": 15915117.0,
      "step": 7260
    },
    {
      "entropy": 5.559117650985717,
      "epoch": 0.7210202461294164,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004956356364160634,
      "loss": 5.3937,
      "mean_token_accuracy": 0.1504781186580658,
      "num_tokens": 15925163.0,
      "step": 7265
    },
    {
      "entropy": 5.514855098724365,
      "epoch": 0.721516474791584,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004956286889180639,
      "loss": 5.2844,
      "mean_token_accuracy": 0.16543339639902116,
      "num_tokens": 15935021.0,
      "step": 7270
    },
    {
      "entropy": 5.568548679351807,
      "epoch": 0.7220127034537515,
      "grad_norm": 0.875,
      "learning_rate": 0.0004956217359489483,
      "loss": 5.407,
      "mean_token_accuracy": 0.14721074551343918,
      "num_tokens": 15945416.0,
      "step": 7275
    },
    {
      "entropy": 5.690412378311157,
      "epoch": 0.722508932115919,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000495614777508889,
      "loss": 5.5183,
      "mean_token_accuracy": 0.14687851220369338,
      "num_tokens": 15955388.0,
      "step": 7280
    },
    {
      "entropy": 5.701138687133789,
      "epoch": 0.7230051607780865,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004956078135980586,
      "loss": 5.5243,
      "mean_token_accuracy": 0.14872784465551375,
      "num_tokens": 15966721.0,
      "step": 7285
    },
    {
      "entropy": 5.627881479263306,
      "epoch": 0.7235013894402541,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004956008442166299,
      "loss": 5.4569,
      "mean_token_accuracy": 0.15040554478764534,
      "num_tokens": 15976388.0,
      "step": 7290
    },
    {
      "entropy": 5.589539098739624,
      "epoch": 0.7239976181024216,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004955938693647756,
      "loss": 5.3362,
      "mean_token_accuracy": 0.1562737390398979,
      "num_tokens": 15986632.0,
      "step": 7295
    },
    {
      "entropy": 5.5868932723999025,
      "epoch": 0.7244938467645892,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004955868890426686,
      "loss": 5.3907,
      "mean_token_accuracy": 0.14940914288163185,
      "num_tokens": 15996626.0,
      "step": 7300
    },
    {
      "entropy": 5.561647605895996,
      "epoch": 0.7249900754267566,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004955799032504821,
      "loss": 5.3716,
      "mean_token_accuracy": 0.15110146403312683,
      "num_tokens": 16007055.0,
      "step": 7305
    },
    {
      "entropy": 5.682785081863403,
      "epoch": 0.7254863040889242,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004955729119883893,
      "loss": 5.4265,
      "mean_token_accuracy": 0.15077741891145707,
      "num_tokens": 16019087.0,
      "step": 7310
    },
    {
      "entropy": 5.6372357368469235,
      "epoch": 0.7259825327510917,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004955659152565636,
      "loss": 5.433,
      "mean_token_accuracy": 0.14655008614063264,
      "num_tokens": 16029980.0,
      "step": 7315
    },
    {
      "entropy": 5.614037990570068,
      "epoch": 0.7264787614132592,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004955589130551784,
      "loss": 5.4099,
      "mean_token_accuracy": 0.15711937397718428,
      "num_tokens": 16040932.0,
      "step": 7320
    },
    {
      "entropy": 5.526982545852661,
      "epoch": 0.7269749900754268,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004955519053844076,
      "loss": 5.3706,
      "mean_token_accuracy": 0.1539077416062355,
      "num_tokens": 16051809.0,
      "step": 7325
    },
    {
      "entropy": 5.629569101333618,
      "epoch": 0.7274712187375942,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004955448922444246,
      "loss": 5.4086,
      "mean_token_accuracy": 0.15429458543658256,
      "num_tokens": 16062957.0,
      "step": 7330
    },
    {
      "entropy": 5.649728870391845,
      "epoch": 0.7279674473997618,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0004955378736354036,
      "loss": 5.4582,
      "mean_token_accuracy": 0.15688300281763076,
      "num_tokens": 16073592.0,
      "step": 7335
    },
    {
      "entropy": 5.5612717628479,
      "epoch": 0.7284636760619293,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004955308495575185,
      "loss": 5.3993,
      "mean_token_accuracy": 0.15492410361766815,
      "num_tokens": 16085054.0,
      "step": 7340
    },
    {
      "entropy": 5.658267307281494,
      "epoch": 0.7289599047240969,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004955238200109436,
      "loss": 5.484,
      "mean_token_accuracy": 0.14988485276699065,
      "num_tokens": 16096936.0,
      "step": 7345
    },
    {
      "entropy": 5.668556070327758,
      "epoch": 0.7294561333862644,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004955167849958531,
      "loss": 5.4861,
      "mean_token_accuracy": 0.14705898761749267,
      "num_tokens": 16107673.0,
      "step": 7350
    },
    {
      "entropy": 5.624842739105224,
      "epoch": 0.729952362048432,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0004955097445124215,
      "loss": 5.3686,
      "mean_token_accuracy": 0.155429670214653,
      "num_tokens": 16117941.0,
      "step": 7355
    },
    {
      "entropy": 5.587152671813965,
      "epoch": 0.7304485907105994,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0004955026985608235,
      "loss": 5.3349,
      "mean_token_accuracy": 0.15884690582752228,
      "num_tokens": 16128737.0,
      "step": 7360
    },
    {
      "entropy": 5.716961622238159,
      "epoch": 0.730944819372767,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004954956471412336,
      "loss": 5.4776,
      "mean_token_accuracy": 0.15179453343153,
      "num_tokens": 16139599.0,
      "step": 7365
    },
    {
      "entropy": 5.514989519119263,
      "epoch": 0.7314410480349345,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004954885902538269,
      "loss": 5.3162,
      "mean_token_accuracy": 0.14936128705739976,
      "num_tokens": 16150796.0,
      "step": 7370
    },
    {
      "entropy": 5.581525039672852,
      "epoch": 0.7319372766971021,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004954815278987781,
      "loss": 5.3853,
      "mean_token_accuracy": 0.157125623524189,
      "num_tokens": 16161236.0,
      "step": 7375
    },
    {
      "entropy": 5.63885269165039,
      "epoch": 0.7324335053592695,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004954744600762627,
      "loss": 5.4908,
      "mean_token_accuracy": 0.15243475139141083,
      "num_tokens": 16172210.0,
      "step": 7380
    },
    {
      "entropy": 5.567775249481201,
      "epoch": 0.7329297340214371,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004954673867864555,
      "loss": 5.3399,
      "mean_token_accuracy": 0.16127611994743346,
      "num_tokens": 16182950.0,
      "step": 7385
    },
    {
      "entropy": 5.563462209701538,
      "epoch": 0.7334259626836046,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004954603080295323,
      "loss": 5.2683,
      "mean_token_accuracy": 0.16479185819625855,
      "num_tokens": 16193971.0,
      "step": 7390
    },
    {
      "entropy": 5.576210641860962,
      "epoch": 0.7339221913457721,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004954532238056686,
      "loss": 5.3896,
      "mean_token_accuracy": 0.1563618466258049,
      "num_tokens": 16205774.0,
      "step": 7395
    },
    {
      "entropy": 5.580049991607666,
      "epoch": 0.7344184200079397,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004954461341150399,
      "loss": 5.3774,
      "mean_token_accuracy": 0.15467396080493928,
      "num_tokens": 16216626.0,
      "step": 7400
    },
    {
      "entropy": 5.616230535507202,
      "epoch": 0.7349146486701071,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004954390389578222,
      "loss": 5.4602,
      "mean_token_accuracy": 0.14845651984214783,
      "num_tokens": 16227748.0,
      "step": 7405
    },
    {
      "entropy": 5.667726373672485,
      "epoch": 0.7354108773322747,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004954319383341912,
      "loss": 5.4267,
      "mean_token_accuracy": 0.1550072394311428,
      "num_tokens": 16240390.0,
      "step": 7410
    },
    {
      "entropy": 5.603915071487426,
      "epoch": 0.7359071059944422,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004954248322443232,
      "loss": 5.3715,
      "mean_token_accuracy": 0.1643430009484291,
      "num_tokens": 16250563.0,
      "step": 7415
    },
    {
      "entropy": 5.588047456741333,
      "epoch": 0.7364033346566098,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004954177206883943,
      "loss": 5.4464,
      "mean_token_accuracy": 0.14566802456974984,
      "num_tokens": 16260722.0,
      "step": 7420
    },
    {
      "entropy": 5.731152868270874,
      "epoch": 0.7368995633187773,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004954106036665809,
      "loss": 5.4487,
      "mean_token_accuracy": 0.1545932948589325,
      "num_tokens": 16270258.0,
      "step": 7425
    },
    {
      "entropy": 5.6218890190124515,
      "epoch": 0.7373957919809448,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004954034811790594,
      "loss": 5.3149,
      "mean_token_accuracy": 0.16307952105998993,
      "num_tokens": 16279504.0,
      "step": 7430
    },
    {
      "entropy": 5.524782609939575,
      "epoch": 0.7378920206431123,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004953963532260067,
      "loss": 5.3344,
      "mean_token_accuracy": 0.1574219733476639,
      "num_tokens": 16291327.0,
      "step": 7435
    },
    {
      "entropy": 5.609736680984497,
      "epoch": 0.7383882493052799,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004953892198075991,
      "loss": 5.4628,
      "mean_token_accuracy": 0.1531672716140747,
      "num_tokens": 16302778.0,
      "step": 7440
    },
    {
      "entropy": 5.65271110534668,
      "epoch": 0.7388844779674474,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004953820809240138,
      "loss": 5.4211,
      "mean_token_accuracy": 0.14703194200992584,
      "num_tokens": 16314373.0,
      "step": 7445
    },
    {
      "entropy": 5.618282842636108,
      "epoch": 0.739380706629615,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004953749365754279,
      "loss": 5.3871,
      "mean_token_accuracy": 0.15660699158906938,
      "num_tokens": 16324518.0,
      "step": 7450
    },
    {
      "entropy": 5.589332103729248,
      "epoch": 0.7398769352917824,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004953677867620184,
      "loss": 5.3121,
      "mean_token_accuracy": 0.15828358829021455,
      "num_tokens": 16335657.0,
      "step": 7455
    },
    {
      "entropy": 5.636603212356567,
      "epoch": 0.74037316395395,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004953606314839626,
      "loss": 5.5165,
      "mean_token_accuracy": 0.14771938398480416,
      "num_tokens": 16346497.0,
      "step": 7460
    },
    {
      "entropy": 5.570276975631714,
      "epoch": 0.7408693926161175,
      "grad_norm": 0.8125,
      "learning_rate": 0.000495353470741438,
      "loss": 5.3815,
      "mean_token_accuracy": 0.15153775215148926,
      "num_tokens": 16358175.0,
      "step": 7465
    },
    {
      "entropy": 5.668294286727905,
      "epoch": 0.741365621278285,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000495346304534622,
      "loss": 5.5129,
      "mean_token_accuracy": 0.14579920768737792,
      "num_tokens": 16369803.0,
      "step": 7470
    },
    {
      "entropy": 5.545563459396362,
      "epoch": 0.7418618499404526,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004953391328636926,
      "loss": 5.2069,
      "mean_token_accuracy": 0.16389822214841843,
      "num_tokens": 16380535.0,
      "step": 7475
    },
    {
      "entropy": 5.540350866317749,
      "epoch": 0.74235807860262,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004953319557288275,
      "loss": 5.3862,
      "mean_token_accuracy": 0.15173779129981996,
      "num_tokens": 16390518.0,
      "step": 7480
    },
    {
      "entropy": 5.594494771957398,
      "epoch": 0.7428543072647876,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004953247731302046,
      "loss": 5.4005,
      "mean_token_accuracy": 0.14989829510450364,
      "num_tokens": 16400627.0,
      "step": 7485
    },
    {
      "entropy": 5.635437822341919,
      "epoch": 0.7433505359269551,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004953175850680022,
      "loss": 5.4593,
      "mean_token_accuracy": 0.15600887238979338,
      "num_tokens": 16411136.0,
      "step": 7490
    },
    {
      "entropy": 5.661326122283936,
      "epoch": 0.7438467645891227,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004953103915423983,
      "loss": 5.3886,
      "mean_token_accuracy": 0.15466448217630385,
      "num_tokens": 16422830.0,
      "step": 7495
    },
    {
      "entropy": 5.560624313354492,
      "epoch": 0.7443429932512902,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004953031925535715,
      "loss": 5.3382,
      "mean_token_accuracy": 0.16099384129047395,
      "num_tokens": 16434158.0,
      "step": 7500
    },
    {
      "entropy": 5.588740110397339,
      "epoch": 0.7448392219134577,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004952959881017003,
      "loss": 5.3764,
      "mean_token_accuracy": 0.1568423628807068,
      "num_tokens": 16444562.0,
      "step": 7505
    },
    {
      "entropy": 5.597737884521484,
      "epoch": 0.7453354505756252,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004952887781869631,
      "loss": 5.411,
      "mean_token_accuracy": 0.15375222265720367,
      "num_tokens": 16456353.0,
      "step": 7510
    },
    {
      "entropy": 5.634686660766602,
      "epoch": 0.7458316792377928,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004952815628095391,
      "loss": 5.431,
      "mean_token_accuracy": 0.14708781838417054,
      "num_tokens": 16465756.0,
      "step": 7515
    },
    {
      "entropy": 5.600973129272461,
      "epoch": 0.7463279078999603,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004952743419696069,
      "loss": 5.3766,
      "mean_token_accuracy": 0.15321100056171416,
      "num_tokens": 16477562.0,
      "step": 7520
    },
    {
      "entropy": 5.621348571777344,
      "epoch": 0.7468241365621279,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004952671156673456,
      "loss": 5.4081,
      "mean_token_accuracy": 0.1518675997853279,
      "num_tokens": 16488134.0,
      "step": 7525
    },
    {
      "entropy": 5.649428272247315,
      "epoch": 0.7473203652242953,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004952598839029345,
      "loss": 5.396,
      "mean_token_accuracy": 0.15431894212961197,
      "num_tokens": 16499146.0,
      "step": 7530
    },
    {
      "entropy": 5.630555915832519,
      "epoch": 0.7478165938864629,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004952526466765529,
      "loss": 5.4354,
      "mean_token_accuracy": 0.15253461599349977,
      "num_tokens": 16510334.0,
      "step": 7535
    },
    {
      "entropy": 5.563698196411133,
      "epoch": 0.7483128225486304,
      "grad_norm": 0.875,
      "learning_rate": 0.0004952454039883804,
      "loss": 5.3178,
      "mean_token_accuracy": 0.1540248841047287,
      "num_tokens": 16520574.0,
      "step": 7540
    },
    {
      "entropy": 5.65339503288269,
      "epoch": 0.7488090512107979,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004952381558385963,
      "loss": 5.4607,
      "mean_token_accuracy": 0.1469207674264908,
      "num_tokens": 16532246.0,
      "step": 7545
    },
    {
      "entropy": 5.731358623504638,
      "epoch": 0.7493052798729655,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004952309022273805,
      "loss": 5.4305,
      "mean_token_accuracy": 0.14792006388306617,
      "num_tokens": 16541756.0,
      "step": 7550
    },
    {
      "entropy": 5.617871379852295,
      "epoch": 0.7498015085351329,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004952236431549129,
      "loss": 5.4353,
      "mean_token_accuracy": 0.14994454085826875,
      "num_tokens": 16551061.0,
      "step": 7555
    },
    {
      "entropy": 5.5954522609710695,
      "epoch": 0.7502977371973005,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004952163786213735,
      "loss": 5.3906,
      "mean_token_accuracy": 0.14958487153053285,
      "num_tokens": 16561155.0,
      "step": 7560
    },
    {
      "entropy": 5.601016807556152,
      "epoch": 0.750793965859468,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004952091086269425,
      "loss": 5.3398,
      "mean_token_accuracy": 0.15005064383149147,
      "num_tokens": 16571843.0,
      "step": 7565
    },
    {
      "entropy": 5.583734369277954,
      "epoch": 0.7512901945216356,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004952018331718001,
      "loss": 5.4233,
      "mean_token_accuracy": 0.15418492630124092,
      "num_tokens": 16583395.0,
      "step": 7570
    },
    {
      "entropy": 5.698184967041016,
      "epoch": 0.7517864231838031,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004951945522561268,
      "loss": 5.4716,
      "mean_token_accuracy": 0.15085942968726157,
      "num_tokens": 16594545.0,
      "step": 7575
    },
    {
      "entropy": 5.723637819290161,
      "epoch": 0.7522826518459707,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004951872658801029,
      "loss": 5.5345,
      "mean_token_accuracy": 0.1519157476723194,
      "num_tokens": 16605901.0,
      "step": 7580
    },
    {
      "entropy": 5.634394741058349,
      "epoch": 0.7527788805081381,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004951799740439093,
      "loss": 5.4775,
      "mean_token_accuracy": 0.1472939819097519,
      "num_tokens": 16616783.0,
      "step": 7585
    },
    {
      "entropy": 5.624340486526489,
      "epoch": 0.7532751091703057,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000495172676747727,
      "loss": 5.4087,
      "mean_token_accuracy": 0.1570548430085182,
      "num_tokens": 16628655.0,
      "step": 7590
    },
    {
      "entropy": 5.643128490447998,
      "epoch": 0.7537713378324732,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004951653739917366,
      "loss": 5.427,
      "mean_token_accuracy": 0.15031962543725969,
      "num_tokens": 16639403.0,
      "step": 7595
    },
    {
      "entropy": 5.658723402023315,
      "epoch": 0.7542675664946408,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004951580657761192,
      "loss": 5.4345,
      "mean_token_accuracy": 0.15885126292705537,
      "num_tokens": 16649937.0,
      "step": 7600
    },
    {
      "entropy": 5.52188630104065,
      "epoch": 0.7547637951568082,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004951507521010563,
      "loss": 5.322,
      "mean_token_accuracy": 0.15973489731550217,
      "num_tokens": 16661466.0,
      "step": 7605
    },
    {
      "entropy": 5.564820241928101,
      "epoch": 0.7552600238189758,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004951434329667293,
      "loss": 5.3423,
      "mean_token_accuracy": 0.15731873214244843,
      "num_tokens": 16673362.0,
      "step": 7610
    },
    {
      "entropy": 5.585936450958252,
      "epoch": 0.7557562524811433,
      "grad_norm": 0.7734375,
      "learning_rate": 0.0004951361083733192,
      "loss": 5.3876,
      "mean_token_accuracy": 0.15195401906967163,
      "num_tokens": 16685555.0,
      "step": 7615
    },
    {
      "entropy": 5.596336078643799,
      "epoch": 0.7562524811433108,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004951287783210082,
      "loss": 5.365,
      "mean_token_accuracy": 0.15113281160593034,
      "num_tokens": 16696006.0,
      "step": 7620
    },
    {
      "entropy": 5.592251348495483,
      "epoch": 0.7567487098054784,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004951214428099778,
      "loss": 5.3806,
      "mean_token_accuracy": 0.15848406106233598,
      "num_tokens": 16706784.0,
      "step": 7625
    },
    {
      "entropy": 5.532399702072143,
      "epoch": 0.7572449384676458,
      "grad_norm": 0.875,
      "learning_rate": 0.0004951141018404099,
      "loss": 5.3461,
      "mean_token_accuracy": 0.16066349446773528,
      "num_tokens": 16718492.0,
      "step": 7630
    },
    {
      "entropy": 5.607969093322754,
      "epoch": 0.7577411671298134,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004951067554124865,
      "loss": 5.4142,
      "mean_token_accuracy": 0.1540622740983963,
      "num_tokens": 16729711.0,
      "step": 7635
    },
    {
      "entropy": 5.604937648773193,
      "epoch": 0.7582373957919809,
      "grad_norm": 0.921875,
      "learning_rate": 0.00049509940352639,
      "loss": 5.3967,
      "mean_token_accuracy": 0.1564691573381424,
      "num_tokens": 16739913.0,
      "step": 7640
    },
    {
      "entropy": 5.666071844100952,
      "epoch": 0.7587336244541485,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004950920461823024,
      "loss": 5.4327,
      "mean_token_accuracy": 0.14718462899327278,
      "num_tokens": 16752273.0,
      "step": 7645
    },
    {
      "entropy": 5.584566307067871,
      "epoch": 0.759229853116316,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004950846833804064,
      "loss": 5.3707,
      "mean_token_accuracy": 0.15428989231586457,
      "num_tokens": 16762050.0,
      "step": 7650
    },
    {
      "entropy": 5.55289249420166,
      "epoch": 0.7597260817784836,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004950773151208845,
      "loss": 5.4317,
      "mean_token_accuracy": 0.15732259303331375,
      "num_tokens": 16772376.0,
      "step": 7655
    },
    {
      "entropy": 5.606362676620483,
      "epoch": 0.760222310440651,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004950699414039194,
      "loss": 5.3972,
      "mean_token_accuracy": 0.15157591700553893,
      "num_tokens": 16783307.0,
      "step": 7660
    },
    {
      "entropy": 5.606588840484619,
      "epoch": 0.7607185391028186,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000495062562229694,
      "loss": 5.2773,
      "mean_token_accuracy": 0.15962879210710526,
      "num_tokens": 16793582.0,
      "step": 7665
    },
    {
      "entropy": 5.545598363876342,
      "epoch": 0.7612147677649861,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004950551775983912,
      "loss": 5.3762,
      "mean_token_accuracy": 0.15512028336524963,
      "num_tokens": 16805160.0,
      "step": 7670
    },
    {
      "entropy": 5.589273405075073,
      "epoch": 0.7617109964271537,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004950477875101942,
      "loss": 5.4339,
      "mean_token_accuracy": 0.15162828862667083,
      "num_tokens": 16816108.0,
      "step": 7675
    },
    {
      "entropy": 5.565524578094482,
      "epoch": 0.7622072250893211,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004950403919652863,
      "loss": 5.3094,
      "mean_token_accuracy": 0.15955265760421752,
      "num_tokens": 16828126.0,
      "step": 7680
    },
    {
      "entropy": 5.60766282081604,
      "epoch": 0.7627034537514887,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004950329909638507,
      "loss": 5.3882,
      "mean_token_accuracy": 0.15282055735588074,
      "num_tokens": 16838186.0,
      "step": 7685
    },
    {
      "entropy": 5.576896286010742,
      "epoch": 0.7631996824136562,
      "grad_norm": 0.78515625,
      "learning_rate": 0.000495025584506071,
      "loss": 5.3379,
      "mean_token_accuracy": 0.15614795535802842,
      "num_tokens": 16848588.0,
      "step": 7690
    },
    {
      "entropy": 5.615791654586792,
      "epoch": 0.7636959110758237,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0004950181725921311,
      "loss": 5.4474,
      "mean_token_accuracy": 0.15467110872268677,
      "num_tokens": 16860269.0,
      "step": 7695
    },
    {
      "entropy": 5.623138427734375,
      "epoch": 0.7641921397379913,
      "grad_norm": 0.76953125,
      "learning_rate": 0.0004950107552222145,
      "loss": 5.4568,
      "mean_token_accuracy": 0.152884541451931,
      "num_tokens": 16872221.0,
      "step": 7700
    },
    {
      "entropy": 5.584514093399048,
      "epoch": 0.7646883684001587,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004950033323965052,
      "loss": 5.3396,
      "mean_token_accuracy": 0.14964289963245392,
      "num_tokens": 16882588.0,
      "step": 7705
    },
    {
      "entropy": 5.637409830093384,
      "epoch": 0.7651845970623263,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004949959041151875,
      "loss": 5.3837,
      "mean_token_accuracy": 0.15608891993761062,
      "num_tokens": 16894278.0,
      "step": 7710
    },
    {
      "entropy": 5.625741958618164,
      "epoch": 0.7656808257244938,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004949884703784453,
      "loss": 5.3701,
      "mean_token_accuracy": 0.15537020415067673,
      "num_tokens": 16905448.0,
      "step": 7715
    },
    {
      "entropy": 5.576999235153198,
      "epoch": 0.7661770543866614,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000494981031186463,
      "loss": 5.4053,
      "mean_token_accuracy": 0.15578350722789763,
      "num_tokens": 16914960.0,
      "step": 7720
    },
    {
      "entropy": 5.601236581802368,
      "epoch": 0.7666732830488289,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004949735865394253,
      "loss": 5.4705,
      "mean_token_accuracy": 0.15467178598046302,
      "num_tokens": 16926839.0,
      "step": 7725
    },
    {
      "entropy": 5.6161126613616945,
      "epoch": 0.7671695117109965,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004949661364375167,
      "loss": 5.3373,
      "mean_token_accuracy": 0.16146927922964097,
      "num_tokens": 16938553.0,
      "step": 7730
    },
    {
      "entropy": 5.628966236114502,
      "epoch": 0.7676657403731639,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004949586808809217,
      "loss": 5.4079,
      "mean_token_accuracy": 0.1493507981300354,
      "num_tokens": 16947987.0,
      "step": 7735
    },
    {
      "entropy": 5.5130181312561035,
      "epoch": 0.7681619690353315,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004949512198698256,
      "loss": 5.3523,
      "mean_token_accuracy": 0.15001057609915733,
      "num_tokens": 16959359.0,
      "step": 7740
    },
    {
      "entropy": 5.6035068988800045,
      "epoch": 0.768658197697499,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000494943753404413,
      "loss": 5.406,
      "mean_token_accuracy": 0.15714000016450883,
      "num_tokens": 16970731.0,
      "step": 7745
    },
    {
      "entropy": 5.550115060806275,
      "epoch": 0.7691544263596666,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004949362814848693,
      "loss": 5.3534,
      "mean_token_accuracy": 0.15779167264699936,
      "num_tokens": 16981868.0,
      "step": 7750
    },
    {
      "entropy": 5.617550277709961,
      "epoch": 0.769650655021834,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004949288041113799,
      "loss": 5.398,
      "mean_token_accuracy": 0.15382955670356752,
      "num_tokens": 16994639.0,
      "step": 7755
    },
    {
      "entropy": 5.579612159729004,
      "epoch": 0.7701468836840016,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004949213212841299,
      "loss": 5.3921,
      "mean_token_accuracy": 0.15306416451931,
      "num_tokens": 17005067.0,
      "step": 7760
    },
    {
      "entropy": 5.584072113037109,
      "epoch": 0.7706431123461691,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000494913833003305,
      "loss": 5.3078,
      "mean_token_accuracy": 0.16076150983572007,
      "num_tokens": 17015102.0,
      "step": 7765
    },
    {
      "entropy": 5.595542287826538,
      "epoch": 0.7711393410083366,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000494906339269091,
      "loss": 5.3431,
      "mean_token_accuracy": 0.1601025342941284,
      "num_tokens": 17026290.0,
      "step": 7770
    },
    {
      "entropy": 5.544939613342285,
      "epoch": 0.7716355696705042,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004948988400816736,
      "loss": 5.304,
      "mean_token_accuracy": 0.1543511003255844,
      "num_tokens": 17035445.0,
      "step": 7775
    },
    {
      "entropy": 5.618521976470947,
      "epoch": 0.7721317983326716,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004948913354412388,
      "loss": 5.4166,
      "mean_token_accuracy": 0.1582753151655197,
      "num_tokens": 17046515.0,
      "step": 7780
    },
    {
      "entropy": 5.6145752429962155,
      "epoch": 0.7726280269948392,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004948838253479727,
      "loss": 5.3804,
      "mean_token_accuracy": 0.15358736515045165,
      "num_tokens": 17056988.0,
      "step": 7785
    },
    {
      "entropy": 5.623727321624756,
      "epoch": 0.7731242556570067,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004948763098020616,
      "loss": 5.4372,
      "mean_token_accuracy": 0.14646037817001342,
      "num_tokens": 17068304.0,
      "step": 7790
    },
    {
      "entropy": 5.576439571380615,
      "epoch": 0.7736204843191743,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004948687888036918,
      "loss": 5.3575,
      "mean_token_accuracy": 0.15362756103277206,
      "num_tokens": 17078405.0,
      "step": 7795
    },
    {
      "entropy": 5.527836894989013,
      "epoch": 0.7741167129813418,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004948612623530497,
      "loss": 5.2609,
      "mean_token_accuracy": 0.1611561506986618,
      "num_tokens": 17087839.0,
      "step": 7800
    },
    {
      "entropy": 5.504787445068359,
      "epoch": 0.7746129416435094,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004948537304503221,
      "loss": 5.3644,
      "mean_token_accuracy": 0.1575997367501259,
      "num_tokens": 17099386.0,
      "step": 7805
    },
    {
      "entropy": 5.627182912826538,
      "epoch": 0.7751091703056768,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004948461930956959,
      "loss": 5.4202,
      "mean_token_accuracy": 0.14962111711502074,
      "num_tokens": 17111754.0,
      "step": 7810
    },
    {
      "entropy": 5.579688787460327,
      "epoch": 0.7756053989678444,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004948386502893577,
      "loss": 5.3019,
      "mean_token_accuracy": 0.1589604452252388,
      "num_tokens": 17122882.0,
      "step": 7815
    },
    {
      "entropy": 5.5752345561981205,
      "epoch": 0.7761016276300119,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004948311020314946,
      "loss": 5.4222,
      "mean_token_accuracy": 0.14875626266002656,
      "num_tokens": 17133986.0,
      "step": 7820
    },
    {
      "entropy": 5.595411729812622,
      "epoch": 0.7765978562921795,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004948235483222941,
      "loss": 5.3939,
      "mean_token_accuracy": 0.157666178047657,
      "num_tokens": 17144784.0,
      "step": 7825
    },
    {
      "entropy": 5.6028642654418945,
      "epoch": 0.777094084954347,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004948159891619431,
      "loss": 5.4863,
      "mean_token_accuracy": 0.14774239808321,
      "num_tokens": 17156328.0,
      "step": 7830
    },
    {
      "entropy": 5.66780309677124,
      "epoch": 0.7775903136165145,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0004948084245506293,
      "loss": 5.3128,
      "mean_token_accuracy": 0.15449730008840562,
      "num_tokens": 17168062.0,
      "step": 7835
    },
    {
      "entropy": 5.607169485092163,
      "epoch": 0.778086542278682,
      "grad_norm": 0.859375,
      "learning_rate": 0.00049480085448854,
      "loss": 5.3554,
      "mean_token_accuracy": 0.15255675166845323,
      "num_tokens": 17178872.0,
      "step": 7840
    },
    {
      "entropy": 5.571358346939087,
      "epoch": 0.7785827709408495,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004947932789758635,
      "loss": 5.3901,
      "mean_token_accuracy": 0.15869444906711577,
      "num_tokens": 17189445.0,
      "step": 7845
    },
    {
      "entropy": 5.55828332901001,
      "epoch": 0.7790789996030171,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000494785698012787,
      "loss": 5.3445,
      "mean_token_accuracy": 0.15926823616027833,
      "num_tokens": 17199454.0,
      "step": 7850
    },
    {
      "entropy": 5.600518465042114,
      "epoch": 0.7795752282651845,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004947781115994989,
      "loss": 5.4097,
      "mean_token_accuracy": 0.1518603652715683,
      "num_tokens": 17210729.0,
      "step": 7855
    },
    {
      "entropy": 5.604863548278809,
      "epoch": 0.7800714569273521,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004947705197361872,
      "loss": 5.4599,
      "mean_token_accuracy": 0.15032342523336412,
      "num_tokens": 17222961.0,
      "step": 7860
    },
    {
      "entropy": 5.564748811721802,
      "epoch": 0.7805676855895196,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004947629224230401,
      "loss": 5.3427,
      "mean_token_accuracy": 0.1614825740456581,
      "num_tokens": 17234026.0,
      "step": 7865
    },
    {
      "entropy": 5.655606985092163,
      "epoch": 0.7810639142516872,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004947553196602461,
      "loss": 5.4654,
      "mean_token_accuracy": 0.15531426221132277,
      "num_tokens": 17244643.0,
      "step": 7870
    },
    {
      "entropy": 5.649674892425537,
      "epoch": 0.7815601429138547,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004947477114479939,
      "loss": 5.4009,
      "mean_token_accuracy": 0.15609394907951354,
      "num_tokens": 17255382.0,
      "step": 7875
    },
    {
      "entropy": 5.534345436096191,
      "epoch": 0.7820563715760223,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004947400977864718,
      "loss": 5.3716,
      "mean_token_accuracy": 0.1518067568540573,
      "num_tokens": 17265860.0,
      "step": 7880
    },
    {
      "entropy": 5.606201410293579,
      "epoch": 0.7825526002381897,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004947324786758688,
      "loss": 5.4483,
      "mean_token_accuracy": 0.15152227655053138,
      "num_tokens": 17277570.0,
      "step": 7885
    },
    {
      "entropy": 5.7140497207641605,
      "epoch": 0.7830488289003573,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004947248541163738,
      "loss": 5.5029,
      "mean_token_accuracy": 0.1465599298477173,
      "num_tokens": 17288684.0,
      "step": 7890
    },
    {
      "entropy": 5.599674224853516,
      "epoch": 0.7835450575625248,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004947172241081759,
      "loss": 5.3324,
      "mean_token_accuracy": 0.1590951293706894,
      "num_tokens": 17299541.0,
      "step": 7895
    },
    {
      "entropy": 5.610669946670532,
      "epoch": 0.7840412862246924,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004947095886514643,
      "loss": 5.2983,
      "mean_token_accuracy": 0.1621933788061142,
      "num_tokens": 17310243.0,
      "step": 7900
    },
    {
      "entropy": 5.6187296390533445,
      "epoch": 0.7845375148868599,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004947019477464283,
      "loss": 5.409,
      "mean_token_accuracy": 0.152819262444973,
      "num_tokens": 17321167.0,
      "step": 7905
    },
    {
      "entropy": 5.508659219741821,
      "epoch": 0.7850337435490274,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004946943013932574,
      "loss": 5.3912,
      "mean_token_accuracy": 0.15561124607920646,
      "num_tokens": 17331076.0,
      "step": 7910
    },
    {
      "entropy": 5.622127485275269,
      "epoch": 0.7855299722111949,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004946866495921413,
      "loss": 5.3515,
      "mean_token_accuracy": 0.15189367532730103,
      "num_tokens": 17341337.0,
      "step": 7915
    },
    {
      "entropy": 5.627335453033448,
      "epoch": 0.7860262008733624,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004946789923432695,
      "loss": 5.3794,
      "mean_token_accuracy": 0.15439552217721939,
      "num_tokens": 17353055.0,
      "step": 7920
    },
    {
      "entropy": 5.504520225524902,
      "epoch": 0.78652242953553,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004946713296468322,
      "loss": 5.2988,
      "mean_token_accuracy": 0.15451223403215408,
      "num_tokens": 17364442.0,
      "step": 7925
    },
    {
      "entropy": 5.581909465789795,
      "epoch": 0.7870186581976975,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004946636615030192,
      "loss": 5.403,
      "mean_token_accuracy": 0.15277259051799774,
      "num_tokens": 17376237.0,
      "step": 7930
    },
    {
      "entropy": 5.62790265083313,
      "epoch": 0.787514886859865,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004946559879120208,
      "loss": 5.3441,
      "mean_token_accuracy": 0.15403714030981064,
      "num_tokens": 17386188.0,
      "step": 7935
    },
    {
      "entropy": 5.498153209686279,
      "epoch": 0.7880111155220325,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004946483088740273,
      "loss": 5.3213,
      "mean_token_accuracy": 0.1592441275715828,
      "num_tokens": 17397058.0,
      "step": 7940
    },
    {
      "entropy": 5.653704309463501,
      "epoch": 0.7885073441842001,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004946406243892289,
      "loss": 5.3937,
      "mean_token_accuracy": 0.1570437327027321,
      "num_tokens": 17408191.0,
      "step": 7945
    },
    {
      "entropy": 5.627149391174316,
      "epoch": 0.7890035728463676,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004946329344578163,
      "loss": 5.3575,
      "mean_token_accuracy": 0.15499195754528045,
      "num_tokens": 17418472.0,
      "step": 7950
    },
    {
      "entropy": 5.56838550567627,
      "epoch": 0.7894998015085352,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004946252390799803,
      "loss": 5.2977,
      "mean_token_accuracy": 0.15691063553094864,
      "num_tokens": 17429556.0,
      "step": 7955
    },
    {
      "entropy": 5.6031248569488525,
      "epoch": 0.7899960301707026,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004946175382559116,
      "loss": 5.4561,
      "mean_token_accuracy": 0.15173669755458832,
      "num_tokens": 17440950.0,
      "step": 7960
    },
    {
      "entropy": 5.548545217514038,
      "epoch": 0.7904922588328702,
      "grad_norm": 0.83203125,
      "learning_rate": 0.000494609831985801,
      "loss": 5.3304,
      "mean_token_accuracy": 0.15893898159265518,
      "num_tokens": 17453724.0,
      "step": 7965
    },
    {
      "entropy": 5.588298702239991,
      "epoch": 0.7909884874950377,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00049460212026984,
      "loss": 5.3704,
      "mean_token_accuracy": 0.1552225574851036,
      "num_tokens": 17464623.0,
      "step": 7970
    },
    {
      "entropy": 5.584340000152588,
      "epoch": 0.7914847161572053,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004945944031082195,
      "loss": 5.3725,
      "mean_token_accuracy": 0.15697754621505738,
      "num_tokens": 17474919.0,
      "step": 7975
    },
    {
      "entropy": 5.652498531341553,
      "epoch": 0.7919809448193728,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000494586680501131,
      "loss": 5.3671,
      "mean_token_accuracy": 0.1519192174077034,
      "num_tokens": 17484747.0,
      "step": 7980
    },
    {
      "entropy": 5.587380027770996,
      "epoch": 0.7924771734815403,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000494578952448766,
      "loss": 5.3069,
      "mean_token_accuracy": 0.15586603283882142,
      "num_tokens": 17496299.0,
      "step": 7985
    },
    {
      "entropy": 5.640566015243531,
      "epoch": 0.7929734021437078,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004945712189513161,
      "loss": 5.4597,
      "mean_token_accuracy": 0.14641395658254625,
      "num_tokens": 17507448.0,
      "step": 7990
    },
    {
      "entropy": 5.677541446685791,
      "epoch": 0.7934696308058753,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004945634800089733,
      "loss": 5.4385,
      "mean_token_accuracy": 0.149424247443676,
      "num_tokens": 17520112.0,
      "step": 7995
    },
    {
      "entropy": 5.625305986404419,
      "epoch": 0.7939658594680429,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004945557356219291,
      "loss": 5.4249,
      "mean_token_accuracy": 0.14955083578824996,
      "num_tokens": 17531368.0,
      "step": 8000
    },
    {
      "entropy": 5.523750829696655,
      "epoch": 0.7944620881302104,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004945479857903758,
      "loss": 5.3778,
      "mean_token_accuracy": 0.1584137961268425,
      "num_tokens": 17542082.0,
      "step": 8005
    },
    {
      "entropy": 5.6196800708770756,
      "epoch": 0.7949583167923779,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004945402305145056,
      "loss": 5.3994,
      "mean_token_accuracy": 0.15377297550439833,
      "num_tokens": 17555518.0,
      "step": 8010
    },
    {
      "entropy": 5.638475561141968,
      "epoch": 0.7954545454545454,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004945324697945107,
      "loss": 5.3718,
      "mean_token_accuracy": 0.15810612440109253,
      "num_tokens": 17566110.0,
      "step": 8015
    },
    {
      "entropy": 5.51535472869873,
      "epoch": 0.795950774116713,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004945247036305836,
      "loss": 5.2934,
      "mean_token_accuracy": 0.15774129033088685,
      "num_tokens": 17575942.0,
      "step": 8020
    },
    {
      "entropy": 5.557010507583618,
      "epoch": 0.7964470027788805,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004945169320229169,
      "loss": 5.3712,
      "mean_token_accuracy": 0.16034497767686845,
      "num_tokens": 17586001.0,
      "step": 8025
    },
    {
      "entropy": 5.682870721817016,
      "epoch": 0.7969432314410481,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004945091549717033,
      "loss": 5.5149,
      "mean_token_accuracy": 0.1454322263598442,
      "num_tokens": 17596774.0,
      "step": 8030
    },
    {
      "entropy": 5.618385601043701,
      "epoch": 0.7974394601032155,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004945013724771356,
      "loss": 5.3812,
      "mean_token_accuracy": 0.15872787684202194,
      "num_tokens": 17606557.0,
      "step": 8035
    },
    {
      "entropy": 5.497632646560669,
      "epoch": 0.7979356887653831,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004944935845394069,
      "loss": 5.2506,
      "mean_token_accuracy": 0.15902817547321318,
      "num_tokens": 17616465.0,
      "step": 8040
    },
    {
      "entropy": 5.638848829269409,
      "epoch": 0.7984319174275506,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004944857911587103,
      "loss": 5.4258,
      "mean_token_accuracy": 0.15399444103240967,
      "num_tokens": 17626907.0,
      "step": 8045
    },
    {
      "entropy": 5.685809516906739,
      "epoch": 0.7989281460897182,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004944779923352389,
      "loss": 5.4188,
      "mean_token_accuracy": 0.1529553696513176,
      "num_tokens": 17638660.0,
      "step": 8050
    },
    {
      "entropy": 5.539503812789917,
      "epoch": 0.7994243747518857,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004944701880691864,
      "loss": 5.3393,
      "mean_token_accuracy": 0.16294268518686295,
      "num_tokens": 17650135.0,
      "step": 8055
    },
    {
      "entropy": 5.5205878734588625,
      "epoch": 0.7999206034140532,
      "grad_norm": 0.9375,
      "learning_rate": 0.000494462378360746,
      "loss": 5.3665,
      "mean_token_accuracy": 0.15941154807806016,
      "num_tokens": 17660987.0,
      "step": 8060
    },
    {
      "entropy": 5.620557451248169,
      "epoch": 0.8004168320762207,
      "grad_norm": 0.75,
      "learning_rate": 0.0004944545632101117,
      "loss": 5.2929,
      "mean_token_accuracy": 0.15972078293561937,
      "num_tokens": 17672520.0,
      "step": 8065
    },
    {
      "entropy": 5.607070207595825,
      "epoch": 0.8009130607383882,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004944467426174769,
      "loss": 5.3607,
      "mean_token_accuracy": 0.156748728454113,
      "num_tokens": 17683688.0,
      "step": 8070
    },
    {
      "entropy": 5.535247707366944,
      "epoch": 0.8014092894005558,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000494438916583036,
      "loss": 5.2892,
      "mean_token_accuracy": 0.1605285659432411,
      "num_tokens": 17695213.0,
      "step": 8075
    },
    {
      "entropy": 5.575884532928467,
      "epoch": 0.8019055180627233,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004944310851069826,
      "loss": 5.38,
      "mean_token_accuracy": 0.1581933856010437,
      "num_tokens": 17705545.0,
      "step": 8080
    },
    {
      "entropy": 5.591823387145996,
      "epoch": 0.8024017467248908,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004944232481895112,
      "loss": 5.3581,
      "mean_token_accuracy": 0.16412023603916168,
      "num_tokens": 17715855.0,
      "step": 8085
    },
    {
      "entropy": 5.627885055541992,
      "epoch": 0.8028979753870583,
      "grad_norm": 0.80859375,
      "learning_rate": 0.000494415405830816,
      "loss": 5.4226,
      "mean_token_accuracy": 0.15787064135074616,
      "num_tokens": 17727764.0,
      "step": 8090
    },
    {
      "entropy": 5.612554740905762,
      "epoch": 0.8033942040492259,
      "grad_norm": 0.875,
      "learning_rate": 0.0004944075580310916,
      "loss": 5.4117,
      "mean_token_accuracy": 0.15614206790924073,
      "num_tokens": 17738464.0,
      "step": 8095
    },
    {
      "entropy": 5.606903982162476,
      "epoch": 0.8038904327113934,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004943997047905325,
      "loss": 5.3996,
      "mean_token_accuracy": 0.16208686232566832,
      "num_tokens": 17747502.0,
      "step": 8100
    },
    {
      "entropy": 5.565347719192505,
      "epoch": 0.804386661373561,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004943918461093335,
      "loss": 5.3872,
      "mean_token_accuracy": 0.15283362567424774,
      "num_tokens": 17759520.0,
      "step": 8105
    },
    {
      "entropy": 5.5941112518310545,
      "epoch": 0.8048828900357284,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004943839819876894,
      "loss": 5.3623,
      "mean_token_accuracy": 0.15686660557985305,
      "num_tokens": 17770631.0,
      "step": 8110
    },
    {
      "entropy": 5.634806537628174,
      "epoch": 0.805379118697896,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004943761124257954,
      "loss": 5.3226,
      "mean_token_accuracy": 0.15570886433124542,
      "num_tokens": 17781320.0,
      "step": 8115
    },
    {
      "entropy": 5.596116304397583,
      "epoch": 0.8058753473600635,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004943682374238464,
      "loss": 5.379,
      "mean_token_accuracy": 0.15654406771063806,
      "num_tokens": 17792627.0,
      "step": 8120
    },
    {
      "entropy": 5.584414863586426,
      "epoch": 0.8063715760222311,
      "grad_norm": 0.83203125,
      "learning_rate": 0.000494360356982038,
      "loss": 5.4261,
      "mean_token_accuracy": 0.15325156450271607,
      "num_tokens": 17803286.0,
      "step": 8125
    },
    {
      "entropy": 5.517915105819702,
      "epoch": 0.8068678046843986,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004943524711005653,
      "loss": 5.2346,
      "mean_token_accuracy": 0.15880848467350006,
      "num_tokens": 17813631.0,
      "step": 8130
    },
    {
      "entropy": 5.547467565536499,
      "epoch": 0.8073640333465661,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000494344579779624,
      "loss": 5.3388,
      "mean_token_accuracy": 0.15715861767530442,
      "num_tokens": 17823467.0,
      "step": 8135
    },
    {
      "entropy": 5.576457691192627,
      "epoch": 0.8078602620087336,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004943366830194098,
      "loss": 5.3278,
      "mean_token_accuracy": 0.1577102318406105,
      "num_tokens": 17833600.0,
      "step": 8140
    },
    {
      "entropy": 5.565642213821411,
      "epoch": 0.8083564906709012,
      "grad_norm": 1.0,
      "learning_rate": 0.0004943287808201183,
      "loss": 5.3374,
      "mean_token_accuracy": 0.1486472576856613,
      "num_tokens": 17843112.0,
      "step": 8145
    },
    {
      "entropy": 5.606211566925049,
      "epoch": 0.8088527193330687,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004943208731819459,
      "loss": 5.3357,
      "mean_token_accuracy": 0.15949689447879792,
      "num_tokens": 17853678.0,
      "step": 8150
    },
    {
      "entropy": 5.571245861053467,
      "epoch": 0.8093489479952362,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004943129601050883,
      "loss": 5.3215,
      "mean_token_accuracy": 0.16115986406803132,
      "num_tokens": 17863537.0,
      "step": 8155
    },
    {
      "entropy": 5.566862392425537,
      "epoch": 0.8098451766574037,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000494305041589742,
      "loss": 5.3981,
      "mean_token_accuracy": 0.15240979045629502,
      "num_tokens": 17873511.0,
      "step": 8160
    },
    {
      "entropy": 5.5311808586120605,
      "epoch": 0.8103414053195712,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004942971176361031,
      "loss": 5.32,
      "mean_token_accuracy": 0.16108133643865585,
      "num_tokens": 17884546.0,
      "step": 8165
    },
    {
      "entropy": 5.535860157012939,
      "epoch": 0.8108376339817388,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004942891882443682,
      "loss": 5.2846,
      "mean_token_accuracy": 0.15900460928678511,
      "num_tokens": 17895331.0,
      "step": 8170
    },
    {
      "entropy": 5.60030689239502,
      "epoch": 0.8113338626439063,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000494281253414734,
      "loss": 5.4444,
      "mean_token_accuracy": 0.152166049182415,
      "num_tokens": 17906465.0,
      "step": 8175
    },
    {
      "entropy": 5.5684957027435305,
      "epoch": 0.8118300913060739,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004942733131473973,
      "loss": 5.3821,
      "mean_token_accuracy": 0.1505785420536995,
      "num_tokens": 17916936.0,
      "step": 8180
    },
    {
      "entropy": 5.655961799621582,
      "epoch": 0.8123263199682413,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004942653674425548,
      "loss": 5.3866,
      "mean_token_accuracy": 0.15095711797475814,
      "num_tokens": 17927522.0,
      "step": 8185
    },
    {
      "entropy": 5.614254665374756,
      "epoch": 0.8128225486304089,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004942574163004038,
      "loss": 5.4485,
      "mean_token_accuracy": 0.15291159600019455,
      "num_tokens": 17938216.0,
      "step": 8190
    },
    {
      "entropy": 5.6159416198730465,
      "epoch": 0.8133187772925764,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004942494597211412,
      "loss": 5.3388,
      "mean_token_accuracy": 0.155377659201622,
      "num_tokens": 17949333.0,
      "step": 8195
    },
    {
      "entropy": 5.63876485824585,
      "epoch": 0.813815005954744,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004942414977049646,
      "loss": 5.3974,
      "mean_token_accuracy": 0.1446226179599762,
      "num_tokens": 17960178.0,
      "step": 8200
    },
    {
      "entropy": 5.559086084365845,
      "epoch": 0.8143112346169115,
      "grad_norm": 0.83203125,
      "learning_rate": 0.000494233530252071,
      "loss": 5.3627,
      "mean_token_accuracy": 0.15392488539218901,
      "num_tokens": 17971180.0,
      "step": 8205
    },
    {
      "entropy": 5.58487868309021,
      "epoch": 0.814807463279079,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004942255573626585,
      "loss": 5.3904,
      "mean_token_accuracy": 0.15401293933391572,
      "num_tokens": 17982572.0,
      "step": 8210
    },
    {
      "entropy": 5.636579418182373,
      "epoch": 0.8153036919412465,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004942175790369244,
      "loss": 5.3313,
      "mean_token_accuracy": 0.15755854845046996,
      "num_tokens": 17993236.0,
      "step": 8215
    },
    {
      "entropy": 5.603274631500244,
      "epoch": 0.8157999206034141,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004942095952750669,
      "loss": 5.4466,
      "mean_token_accuracy": 0.15470776259899138,
      "num_tokens": 18004215.0,
      "step": 8220
    },
    {
      "entropy": 5.5473466396331785,
      "epoch": 0.8162961492655816,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004942016060772837,
      "loss": 5.3344,
      "mean_token_accuracy": 0.1581986740231514,
      "num_tokens": 18013880.0,
      "step": 8225
    },
    {
      "entropy": 5.577635002136231,
      "epoch": 0.8167923779277491,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004941936114437731,
      "loss": 5.3905,
      "mean_token_accuracy": 0.15430509746074678,
      "num_tokens": 18024928.0,
      "step": 8230
    },
    {
      "entropy": 5.578258991241455,
      "epoch": 0.8172886065899166,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004941856113747331,
      "loss": 5.3504,
      "mean_token_accuracy": 0.15071632713079453,
      "num_tokens": 18035857.0,
      "step": 8235
    },
    {
      "entropy": 5.641419172286987,
      "epoch": 0.8177848352520841,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004941776058703623,
      "loss": 5.4625,
      "mean_token_accuracy": 0.1520865350961685,
      "num_tokens": 18046776.0,
      "step": 8240
    },
    {
      "entropy": 5.623132038116455,
      "epoch": 0.8182810639142517,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004941695949308593,
      "loss": 5.409,
      "mean_token_accuracy": 0.15991352796554564,
      "num_tokens": 18057167.0,
      "step": 8245
    },
    {
      "entropy": 5.624783039093018,
      "epoch": 0.8187772925764192,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004941615785564224,
      "loss": 5.4301,
      "mean_token_accuracy": 0.14956261664628984,
      "num_tokens": 18066953.0,
      "step": 8250
    },
    {
      "entropy": 5.58717827796936,
      "epoch": 0.8192735212385868,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004941535567472506,
      "loss": 5.2657,
      "mean_token_accuracy": 0.15928489118814468,
      "num_tokens": 18077977.0,
      "step": 8255
    },
    {
      "entropy": 5.5626789093017575,
      "epoch": 0.8197697499007542,
      "grad_norm": 0.9375,
      "learning_rate": 0.000494145529503543,
      "loss": 5.3594,
      "mean_token_accuracy": 0.15167167633771897,
      "num_tokens": 18087581.0,
      "step": 8260
    },
    {
      "entropy": 5.519986772537232,
      "epoch": 0.8202659785629218,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004941374968254985,
      "loss": 5.2653,
      "mean_token_accuracy": 0.15741712152957915,
      "num_tokens": 18098193.0,
      "step": 8265
    },
    {
      "entropy": 5.584941959381103,
      "epoch": 0.8207622072250893,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004941294587133161,
      "loss": 5.3791,
      "mean_token_accuracy": 0.1589309900999069,
      "num_tokens": 18110830.0,
      "step": 8270
    },
    {
      "entropy": 5.654639673233032,
      "epoch": 0.8212584358872569,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004941214151671954,
      "loss": 5.4134,
      "mean_token_accuracy": 0.15107691287994385,
      "num_tokens": 18122073.0,
      "step": 8275
    },
    {
      "entropy": 5.548416185379028,
      "epoch": 0.8217546645494244,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004941133661873357,
      "loss": 5.3926,
      "mean_token_accuracy": 0.15540233254432678,
      "num_tokens": 18132777.0,
      "step": 8280
    },
    {
      "entropy": 5.526298189163208,
      "epoch": 0.822250893211592,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004941053117739367,
      "loss": 5.3013,
      "mean_token_accuracy": 0.15420641452074052,
      "num_tokens": 18143232.0,
      "step": 8285
    },
    {
      "entropy": 5.59205904006958,
      "epoch": 0.8227471218737594,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004940972519271982,
      "loss": 5.4754,
      "mean_token_accuracy": 0.1535163328051567,
      "num_tokens": 18155225.0,
      "step": 8290
    },
    {
      "entropy": 5.646539306640625,
      "epoch": 0.823243350535927,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004940891866473199,
      "loss": 5.3967,
      "mean_token_accuracy": 0.1553361311554909,
      "num_tokens": 18165945.0,
      "step": 8295
    },
    {
      "entropy": 5.665407943725586,
      "epoch": 0.8237395791980945,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004940811159345017,
      "loss": 5.4304,
      "mean_token_accuracy": 0.14958547353744506,
      "num_tokens": 18177271.0,
      "step": 8300
    },
    {
      "entropy": 5.663650274276733,
      "epoch": 0.824235807860262,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004940730397889441,
      "loss": 5.3737,
      "mean_token_accuracy": 0.15421211272478103,
      "num_tokens": 18189732.0,
      "step": 8305
    },
    {
      "entropy": 5.594327116012574,
      "epoch": 0.8247320365224295,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004940649582108472,
      "loss": 5.4713,
      "mean_token_accuracy": 0.1524760343134403,
      "num_tokens": 18200731.0,
      "step": 8310
    },
    {
      "entropy": 5.6143065929412845,
      "epoch": 0.825228265184597,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004940568712004113,
      "loss": 5.3943,
      "mean_token_accuracy": 0.1596961349248886,
      "num_tokens": 18210873.0,
      "step": 8315
    },
    {
      "entropy": 5.679521274566651,
      "epoch": 0.8257244938467646,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004940487787578369,
      "loss": 5.3819,
      "mean_token_accuracy": 0.15673448145389557,
      "num_tokens": 18222072.0,
      "step": 8320
    },
    {
      "entropy": 5.530484390258789,
      "epoch": 0.8262207225089321,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000494040680883325,
      "loss": 5.2823,
      "mean_token_accuracy": 0.15844718366861343,
      "num_tokens": 18232522.0,
      "step": 8325
    },
    {
      "entropy": 5.625659942626953,
      "epoch": 0.8267169511710997,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004940325775770762,
      "loss": 5.3908,
      "mean_token_accuracy": 0.14972957149147986,
      "num_tokens": 18243587.0,
      "step": 8330
    },
    {
      "entropy": 5.575936985015869,
      "epoch": 0.8272131798332671,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004940244688392913,
      "loss": 5.2986,
      "mean_token_accuracy": 0.16298762559890748,
      "num_tokens": 18254965.0,
      "step": 8335
    },
    {
      "entropy": 5.571698570251465,
      "epoch": 0.8277094084954347,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004940163546701716,
      "loss": 5.32,
      "mean_token_accuracy": 0.15952432304620742,
      "num_tokens": 18265892.0,
      "step": 8340
    },
    {
      "entropy": 5.585426902770996,
      "epoch": 0.8282056371576022,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004940082350699183,
      "loss": 5.3184,
      "mean_token_accuracy": 0.15753851234912872,
      "num_tokens": 18276713.0,
      "step": 8345
    },
    {
      "entropy": 5.659633779525757,
      "epoch": 0.8287018658197698,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004940001100387325,
      "loss": 5.3726,
      "mean_token_accuracy": 0.14980877786874772,
      "num_tokens": 18288846.0,
      "step": 8350
    },
    {
      "entropy": 5.639483165740967,
      "epoch": 0.8291980944819373,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004939919795768161,
      "loss": 5.3987,
      "mean_token_accuracy": 0.15961014479398727,
      "num_tokens": 18298750.0,
      "step": 8355
    },
    {
      "entropy": 5.580147314071655,
      "epoch": 0.8296943231441049,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004939838436843704,
      "loss": 5.3744,
      "mean_token_accuracy": 0.15533585399389266,
      "num_tokens": 18310065.0,
      "step": 8360
    },
    {
      "entropy": 5.584295177459717,
      "epoch": 0.8301905518062723,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004939757023615973,
      "loss": 5.3345,
      "mean_token_accuracy": 0.15667842626571654,
      "num_tokens": 18319498.0,
      "step": 8365
    },
    {
      "entropy": 5.558476495742798,
      "epoch": 0.8306867804684399,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004939675556086986,
      "loss": 5.2902,
      "mean_token_accuracy": 0.16425713449716567,
      "num_tokens": 18329718.0,
      "step": 8370
    },
    {
      "entropy": 5.550742244720459,
      "epoch": 0.8311830091306074,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004939594034258763,
      "loss": 5.3006,
      "mean_token_accuracy": 0.16241085529327393,
      "num_tokens": 18341628.0,
      "step": 8375
    },
    {
      "entropy": 5.5619243621826175,
      "epoch": 0.8316792377927749,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004939512458133326,
      "loss": 5.3473,
      "mean_token_accuracy": 0.16096800416707993,
      "num_tokens": 18352869.0,
      "step": 8380
    },
    {
      "entropy": 5.5964590549469,
      "epoch": 0.8321754664549424,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004939430827712698,
      "loss": 5.362,
      "mean_token_accuracy": 0.15447923466563224,
      "num_tokens": 18363664.0,
      "step": 8385
    },
    {
      "entropy": 5.630773496627808,
      "epoch": 0.8326716951171099,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004939349142998905,
      "loss": 5.3507,
      "mean_token_accuracy": 0.1544475883245468,
      "num_tokens": 18374325.0,
      "step": 8390
    },
    {
      "entropy": 5.584388971328735,
      "epoch": 0.8331679237792775,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004939267403993968,
      "loss": 5.3453,
      "mean_token_accuracy": 0.15908517688512802,
      "num_tokens": 18384547.0,
      "step": 8395
    },
    {
      "entropy": 5.596279573440552,
      "epoch": 0.833664152441445,
      "grad_norm": 0.921875,
      "learning_rate": 0.000493918561069992,
      "loss": 5.3211,
      "mean_token_accuracy": 0.15538033843040466,
      "num_tokens": 18395141.0,
      "step": 8400
    },
    {
      "entropy": 5.537204551696777,
      "epoch": 0.8341603811036126,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004939103763118784,
      "loss": 5.3197,
      "mean_token_accuracy": 0.15610468983650208,
      "num_tokens": 18406326.0,
      "step": 8405
    },
    {
      "entropy": 5.60677318572998,
      "epoch": 0.83465660976578,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004939021861252593,
      "loss": 5.3866,
      "mean_token_accuracy": 0.15724660605192184,
      "num_tokens": 18417521.0,
      "step": 8410
    },
    {
      "entropy": 5.616010284423828,
      "epoch": 0.8351528384279476,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004938939905103376,
      "loss": 5.3199,
      "mean_token_accuracy": 0.15721606761217116,
      "num_tokens": 18426813.0,
      "step": 8415
    },
    {
      "entropy": 5.561202621459961,
      "epoch": 0.8356490670901151,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004938857894673166,
      "loss": 5.2708,
      "mean_token_accuracy": 0.15798320472240449,
      "num_tokens": 18438227.0,
      "step": 8420
    },
    {
      "entropy": 5.571479034423828,
      "epoch": 0.8361452957522827,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004938775829963997,
      "loss": 5.4019,
      "mean_token_accuracy": 0.15177630186080932,
      "num_tokens": 18448915.0,
      "step": 8425
    },
    {
      "entropy": 5.604960298538208,
      "epoch": 0.8366415244144502,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004938693710977904,
      "loss": 5.2908,
      "mean_token_accuracy": 0.15932296812534333,
      "num_tokens": 18459330.0,
      "step": 8430
    },
    {
      "entropy": 5.559336471557617,
      "epoch": 0.8371377530766178,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004938611537716922,
      "loss": 5.2693,
      "mean_token_accuracy": 0.16089767962694168,
      "num_tokens": 18469813.0,
      "step": 8435
    },
    {
      "entropy": 5.538674449920654,
      "epoch": 0.8376339817387852,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004938529310183091,
      "loss": 5.3194,
      "mean_token_accuracy": 0.15382401794195175,
      "num_tokens": 18479783.0,
      "step": 8440
    },
    {
      "entropy": 5.53818678855896,
      "epoch": 0.8381302104009528,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000493844702837845,
      "loss": 5.3445,
      "mean_token_accuracy": 0.1606591448187828,
      "num_tokens": 18489640.0,
      "step": 8445
    },
    {
      "entropy": 5.638463020324707,
      "epoch": 0.8386264390631203,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004938364692305036,
      "loss": 5.4002,
      "mean_token_accuracy": 0.15992128252983093,
      "num_tokens": 18500744.0,
      "step": 8450
    },
    {
      "entropy": 5.617222642898559,
      "epoch": 0.8391226677252878,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004938282301964896,
      "loss": 5.3049,
      "mean_token_accuracy": 0.15640706121921538,
      "num_tokens": 18513286.0,
      "step": 8455
    },
    {
      "entropy": 5.613192081451416,
      "epoch": 0.8396188963874553,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004938199857360068,
      "loss": 5.2865,
      "mean_token_accuracy": 0.15690700858831405,
      "num_tokens": 18522934.0,
      "step": 8460
    },
    {
      "entropy": 5.5160599708557125,
      "epoch": 0.8401151250496228,
      "grad_norm": 0.90625,
      "learning_rate": 0.00049381173584926,
      "loss": 5.3489,
      "mean_token_accuracy": 0.15866285264492036,
      "num_tokens": 18534828.0,
      "step": 8465
    },
    {
      "entropy": 5.553909063339233,
      "epoch": 0.8406113537117904,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004938034805364535,
      "loss": 5.3089,
      "mean_token_accuracy": 0.1571136400103569,
      "num_tokens": 18544733.0,
      "step": 8470
    },
    {
      "entropy": 5.582355403900147,
      "epoch": 0.8411075823739579,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004937952197977923,
      "loss": 5.3596,
      "mean_token_accuracy": 0.15624982565641404,
      "num_tokens": 18554583.0,
      "step": 8475
    },
    {
      "entropy": 5.679518318176269,
      "epoch": 0.8416038110361255,
      "grad_norm": 0.7890625,
      "learning_rate": 0.000493786953633481,
      "loss": 5.4003,
      "mean_token_accuracy": 0.1517256587743759,
      "num_tokens": 18566503.0,
      "step": 8480
    },
    {
      "entropy": 5.59273943901062,
      "epoch": 0.842100039698293,
      "grad_norm": 0.73828125,
      "learning_rate": 0.0004937786820437248,
      "loss": 5.3433,
      "mean_token_accuracy": 0.15839931070804597,
      "num_tokens": 18578161.0,
      "step": 8485
    },
    {
      "entropy": 5.565155076980591,
      "epoch": 0.8425962683604605,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004937704050287286,
      "loss": 5.4025,
      "mean_token_accuracy": 0.1599843755364418,
      "num_tokens": 18588212.0,
      "step": 8490
    },
    {
      "entropy": 5.683433914184571,
      "epoch": 0.843092497022628,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004937621225886978,
      "loss": 5.3947,
      "mean_token_accuracy": 0.14981088191270828,
      "num_tokens": 18601293.0,
      "step": 8495
    },
    {
      "entropy": 5.619963788986206,
      "epoch": 0.8435887256847956,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004937538347238378,
      "loss": 5.443,
      "mean_token_accuracy": 0.15082540810108186,
      "num_tokens": 18613316.0,
      "step": 8500
    },
    {
      "entropy": 5.584312963485718,
      "epoch": 0.8440849543469631,
      "grad_norm": 0.875,
      "learning_rate": 0.000493745541434354,
      "loss": 5.3798,
      "mean_token_accuracy": 0.1587498351931572,
      "num_tokens": 18624461.0,
      "step": 8505
    },
    {
      "entropy": 5.616329383850098,
      "epoch": 0.8445811830091307,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004937372427204522,
      "loss": 5.286,
      "mean_token_accuracy": 0.15837130844593048,
      "num_tokens": 18635165.0,
      "step": 8510
    },
    {
      "entropy": 5.602617645263672,
      "epoch": 0.8450774116712981,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004937289385823382,
      "loss": 5.3381,
      "mean_token_accuracy": 0.16230068504810333,
      "num_tokens": 18646649.0,
      "step": 8515
    },
    {
      "entropy": 5.515107536315918,
      "epoch": 0.8455736403334657,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004937206290202176,
      "loss": 5.3213,
      "mean_token_accuracy": 0.15529693216085433,
      "num_tokens": 18658527.0,
      "step": 8520
    },
    {
      "entropy": 5.590562391281128,
      "epoch": 0.8460698689956332,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004937123140342969,
      "loss": 5.3092,
      "mean_token_accuracy": 0.1559583380818367,
      "num_tokens": 18669463.0,
      "step": 8525
    },
    {
      "entropy": 5.620056390762329,
      "epoch": 0.8465660976578007,
      "grad_norm": 0.9375,
      "learning_rate": 0.000493703993624782,
      "loss": 5.3604,
      "mean_token_accuracy": 0.15542348027229308,
      "num_tokens": 18680806.0,
      "step": 8530
    },
    {
      "entropy": 5.5720561981201175,
      "epoch": 0.8470623263199683,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004936956677918794,
      "loss": 5.3553,
      "mean_token_accuracy": 0.15351967364549637,
      "num_tokens": 18692124.0,
      "step": 8535
    },
    {
      "entropy": 5.618325567245483,
      "epoch": 0.8475585549821357,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004936873365357955,
      "loss": 5.3882,
      "mean_token_accuracy": 0.1555369660258293,
      "num_tokens": 18702017.0,
      "step": 8540
    },
    {
      "entropy": 5.524436569213867,
      "epoch": 0.8480547836443033,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004936789998567368,
      "loss": 5.2841,
      "mean_token_accuracy": 0.16143308579921722,
      "num_tokens": 18713205.0,
      "step": 8545
    },
    {
      "entropy": 5.585800981521606,
      "epoch": 0.8485510123064708,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004936706577549101,
      "loss": 5.2979,
      "mean_token_accuracy": 0.15933337211608886,
      "num_tokens": 18724165.0,
      "step": 8550
    },
    {
      "entropy": 5.595956087112427,
      "epoch": 0.8490472409686384,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004936623102305223,
      "loss": 5.3747,
      "mean_token_accuracy": 0.1568777233362198,
      "num_tokens": 18735155.0,
      "step": 8555
    },
    {
      "entropy": 5.600664472579956,
      "epoch": 0.8495434696308058,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004936539572837804,
      "loss": 5.4374,
      "mean_token_accuracy": 0.1557487815618515,
      "num_tokens": 18746228.0,
      "step": 8560
    },
    {
      "entropy": 5.60756025314331,
      "epoch": 0.8500396982929734,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004936455989148916,
      "loss": 5.3961,
      "mean_token_accuracy": 0.15620139092206956,
      "num_tokens": 18756638.0,
      "step": 8565
    },
    {
      "entropy": 5.643007850646972,
      "epoch": 0.8505359269551409,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004936372351240629,
      "loss": 5.3628,
      "mean_token_accuracy": 0.15929753482341766,
      "num_tokens": 18768651.0,
      "step": 8570
    },
    {
      "entropy": 5.695384788513183,
      "epoch": 0.8510321556173085,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000493628865911502,
      "loss": 5.4583,
      "mean_token_accuracy": 0.15153556019067765,
      "num_tokens": 18780399.0,
      "step": 8575
    },
    {
      "entropy": 5.554955005645752,
      "epoch": 0.851528384279476,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004936204912774163,
      "loss": 5.3225,
      "mean_token_accuracy": 0.15521062090992926,
      "num_tokens": 18791927.0,
      "step": 8580
    },
    {
      "entropy": 5.536719274520874,
      "epoch": 0.8520246129416436,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004936121112220135,
      "loss": 5.313,
      "mean_token_accuracy": 0.15647718459367752,
      "num_tokens": 18802779.0,
      "step": 8585
    },
    {
      "entropy": 5.569678592681885,
      "epoch": 0.852520841603811,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004936037257455015,
      "loss": 5.3167,
      "mean_token_accuracy": 0.15982182323932648,
      "num_tokens": 18813123.0,
      "step": 8590
    },
    {
      "entropy": 5.558087778091431,
      "epoch": 0.8530170702659786,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000493595334848088,
      "loss": 5.3132,
      "mean_token_accuracy": 0.16583057940006257,
      "num_tokens": 18823699.0,
      "step": 8595
    },
    {
      "entropy": 5.650063419342041,
      "epoch": 0.8535132989281461,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004935869385299813,
      "loss": 5.4673,
      "mean_token_accuracy": 0.14959802478551865,
      "num_tokens": 18835450.0,
      "step": 8600
    },
    {
      "entropy": 5.563158893585205,
      "epoch": 0.8540095275903136,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004935785367913896,
      "loss": 5.3156,
      "mean_token_accuracy": 0.15441120266914368,
      "num_tokens": 18845919.0,
      "step": 8605
    },
    {
      "entropy": 5.624350547790527,
      "epoch": 0.8545057562524812,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004935701296325212,
      "loss": 5.4302,
      "mean_token_accuracy": 0.1497267834842205,
      "num_tokens": 18858376.0,
      "step": 8610
    },
    {
      "entropy": 5.6011314868927,
      "epoch": 0.8550019849146486,
      "grad_norm": 0.7578125,
      "learning_rate": 0.0004935617170535845,
      "loss": 5.3874,
      "mean_token_accuracy": 0.1607848286628723,
      "num_tokens": 18870456.0,
      "step": 8615
    },
    {
      "entropy": 5.594962406158447,
      "epoch": 0.8554982135768162,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004935532990547882,
      "loss": 5.3963,
      "mean_token_accuracy": 0.157123002409935,
      "num_tokens": 18882174.0,
      "step": 8620
    },
    {
      "entropy": 5.691591024398804,
      "epoch": 0.8559944422389837,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004935448756363409,
      "loss": 5.5157,
      "mean_token_accuracy": 0.15398339480161666,
      "num_tokens": 18892982.0,
      "step": 8625
    },
    {
      "entropy": 5.6582964897155765,
      "epoch": 0.8564906709011513,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004935364467984518,
      "loss": 5.3306,
      "mean_token_accuracy": 0.16034651547670364,
      "num_tokens": 18903557.0,
      "step": 8630
    },
    {
      "entropy": 5.655878257751465,
      "epoch": 0.8569868995633187,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004935280125413295,
      "loss": 5.4761,
      "mean_token_accuracy": 0.15420986860990524,
      "num_tokens": 18914912.0,
      "step": 8635
    },
    {
      "entropy": 5.65370626449585,
      "epoch": 0.8574831282254863,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0004935195728651836,
      "loss": 5.4052,
      "mean_token_accuracy": 0.1553529679775238,
      "num_tokens": 18927773.0,
      "step": 8640
    },
    {
      "entropy": 5.596816492080689,
      "epoch": 0.8579793568876538,
      "grad_norm": 0.9375,
      "learning_rate": 0.000493511127770223,
      "loss": 5.3631,
      "mean_token_accuracy": 0.15304585546255112,
      "num_tokens": 18938313.0,
      "step": 8645
    },
    {
      "entropy": 5.556548643112182,
      "epoch": 0.8584755855498214,
      "grad_norm": 0.875,
      "learning_rate": 0.0004935026772566574,
      "loss": 5.3083,
      "mean_token_accuracy": 0.16367752850055695,
      "num_tokens": 18948600.0,
      "step": 8650
    },
    {
      "entropy": 5.638124942779541,
      "epoch": 0.8589718142119889,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004934942213246961,
      "loss": 5.3981,
      "mean_token_accuracy": 0.1521536260843277,
      "num_tokens": 18958927.0,
      "step": 8655
    },
    {
      "entropy": 5.583178329467773,
      "epoch": 0.8594680428741565,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000493485759974549,
      "loss": 5.331,
      "mean_token_accuracy": 0.15029305815696717,
      "num_tokens": 18968625.0,
      "step": 8660
    },
    {
      "entropy": 5.485573816299438,
      "epoch": 0.8599642715363239,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004934772932064259,
      "loss": 5.2701,
      "mean_token_accuracy": 0.1593043938279152,
      "num_tokens": 18979772.0,
      "step": 8665
    },
    {
      "entropy": 5.617111444473267,
      "epoch": 0.8604605001984915,
      "grad_norm": 0.77734375,
      "learning_rate": 0.0004934688210205367,
      "loss": 5.4254,
      "mean_token_accuracy": 0.1574472039937973,
      "num_tokens": 18991522.0,
      "step": 8670
    },
    {
      "entropy": 5.635264205932617,
      "epoch": 0.860956728860659,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004934603434170913,
      "loss": 5.3878,
      "mean_token_accuracy": 0.15423630475997924,
      "num_tokens": 19002479.0,
      "step": 8675
    },
    {
      "entropy": 5.612432909011841,
      "epoch": 0.8614529575228265,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004934518603963003,
      "loss": 5.4496,
      "mean_token_accuracy": 0.15168481767177583,
      "num_tokens": 19013254.0,
      "step": 8680
    },
    {
      "entropy": 5.559790992736817,
      "epoch": 0.861949186184994,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004934433719583739,
      "loss": 5.2852,
      "mean_token_accuracy": 0.1610727444291115,
      "num_tokens": 19023943.0,
      "step": 8685
    },
    {
      "entropy": 5.655515956878662,
      "epoch": 0.8624454148471615,
      "grad_norm": 0.77734375,
      "learning_rate": 0.0004934348781035226,
      "loss": 5.4075,
      "mean_token_accuracy": 0.15645059943199158,
      "num_tokens": 19034672.0,
      "step": 8690
    },
    {
      "entropy": 5.66617374420166,
      "epoch": 0.8629416435093291,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000493426378831957,
      "loss": 5.4038,
      "mean_token_accuracy": 0.16310026049613952,
      "num_tokens": 19045758.0,
      "step": 8695
    },
    {
      "entropy": 5.554395055770874,
      "epoch": 0.8634378721714966,
      "grad_norm": 0.77734375,
      "learning_rate": 0.0004934178741438877,
      "loss": 5.3877,
      "mean_token_accuracy": 0.15989836752414704,
      "num_tokens": 19057526.0,
      "step": 8700
    },
    {
      "entropy": 5.52222466468811,
      "epoch": 0.8639341008336642,
      "grad_norm": 0.81640625,
      "learning_rate": 0.000493409364039526,
      "loss": 5.2491,
      "mean_token_accuracy": 0.15929249972105025,
      "num_tokens": 19069657.0,
      "step": 8705
    },
    {
      "entropy": 5.575068140029908,
      "epoch": 0.8644303294958317,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004934008485190825,
      "loss": 5.3201,
      "mean_token_accuracy": 0.15782753378152847,
      "num_tokens": 19080699.0,
      "step": 8710
    },
    {
      "entropy": 5.545238065719604,
      "epoch": 0.8649265581579992,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004933923275827687,
      "loss": 5.3345,
      "mean_token_accuracy": 0.15497382283210753,
      "num_tokens": 19093041.0,
      "step": 8715
    },
    {
      "entropy": 5.645983695983887,
      "epoch": 0.8654227868201667,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004933838012307957,
      "loss": 5.3964,
      "mean_token_accuracy": 0.15230150669813156,
      "num_tokens": 19104249.0,
      "step": 8720
    },
    {
      "entropy": 5.627114582061767,
      "epoch": 0.8659190154823343,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000493375269463375,
      "loss": 5.364,
      "mean_token_accuracy": 0.15343012809753417,
      "num_tokens": 19114434.0,
      "step": 8725
    },
    {
      "entropy": 5.577263927459716,
      "epoch": 0.8664152441445018,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004933667322807182,
      "loss": 5.3651,
      "mean_token_accuracy": 0.15154905319213868,
      "num_tokens": 19124814.0,
      "step": 8730
    },
    {
      "entropy": 5.59411849975586,
      "epoch": 0.8669114728066694,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000493358189683037,
      "loss": 5.3994,
      "mean_token_accuracy": 0.15635800510644912,
      "num_tokens": 19135205.0,
      "step": 8735
    },
    {
      "entropy": 5.556440114974976,
      "epoch": 0.8674077014688368,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004933496416705432,
      "loss": 5.2619,
      "mean_token_accuracy": 0.16255960166454314,
      "num_tokens": 19147117.0,
      "step": 8740
    },
    {
      "entropy": 5.572934341430664,
      "epoch": 0.8679039301310044,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004933410882434488,
      "loss": 5.3638,
      "mean_token_accuracy": 0.15515206605196,
      "num_tokens": 19158131.0,
      "step": 8745
    },
    {
      "entropy": 5.547046804428101,
      "epoch": 0.8684001587931719,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004933325294019659,
      "loss": 5.2526,
      "mean_token_accuracy": 0.1608527883887291,
      "num_tokens": 19168820.0,
      "step": 8750
    },
    {
      "entropy": 5.623052787780762,
      "epoch": 0.8688963874553394,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004933239651463068,
      "loss": 5.3474,
      "mean_token_accuracy": 0.16261591613292695,
      "num_tokens": 19178845.0,
      "step": 8755
    },
    {
      "entropy": 5.588670635223389,
      "epoch": 0.869392616117507,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004933153954766837,
      "loss": 5.3112,
      "mean_token_accuracy": 0.15922656804323196,
      "num_tokens": 19190830.0,
      "step": 8760
    },
    {
      "entropy": 5.448131847381592,
      "epoch": 0.8698888447796744,
      "grad_norm": 0.875,
      "learning_rate": 0.0004933068203933092,
      "loss": 5.1773,
      "mean_token_accuracy": 0.1624591201543808,
      "num_tokens": 19202056.0,
      "step": 8765
    },
    {
      "entropy": 5.514496946334839,
      "epoch": 0.870385073441842,
      "grad_norm": 0.81640625,
      "learning_rate": 0.000493298239896396,
      "loss": 5.3821,
      "mean_token_accuracy": 0.15466125309467316,
      "num_tokens": 19213497.0,
      "step": 8770
    },
    {
      "entropy": 5.659720468521118,
      "epoch": 0.8708813021040095,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0004932896539861568,
      "loss": 5.4068,
      "mean_token_accuracy": 0.15293225944042205,
      "num_tokens": 19226103.0,
      "step": 8775
    },
    {
      "entropy": 5.617606067657471,
      "epoch": 0.8713775307661771,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004932810626628047,
      "loss": 5.4197,
      "mean_token_accuracy": 0.1572378620505333,
      "num_tokens": 19237297.0,
      "step": 8780
    },
    {
      "entropy": 5.600645256042481,
      "epoch": 0.8718737594283446,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004932724659265526,
      "loss": 5.3421,
      "mean_token_accuracy": 0.15571752786636353,
      "num_tokens": 19247537.0,
      "step": 8785
    },
    {
      "entropy": 5.508291578292846,
      "epoch": 0.8723699880905121,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004932638637776134,
      "loss": 5.2609,
      "mean_token_accuracy": 0.16021536141633988,
      "num_tokens": 19260465.0,
      "step": 8790
    },
    {
      "entropy": 5.590749645233155,
      "epoch": 0.8728662167526796,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000493255256216201,
      "loss": 5.3521,
      "mean_token_accuracy": 0.15528776496648788,
      "num_tokens": 19270534.0,
      "step": 8795
    },
    {
      "entropy": 5.553844165802002,
      "epoch": 0.8733624454148472,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004932466432425284,
      "loss": 5.3579,
      "mean_token_accuracy": 0.15564285218715668,
      "num_tokens": 19281551.0,
      "step": 8800
    },
    {
      "entropy": 5.538980484008789,
      "epoch": 0.8738586740770147,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004932380248568094,
      "loss": 5.2982,
      "mean_token_accuracy": 0.15979526042938233,
      "num_tokens": 19293300.0,
      "step": 8805
    },
    {
      "entropy": 5.5843109607696535,
      "epoch": 0.8743549027391823,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004932294010592575,
      "loss": 5.2429,
      "mean_token_accuracy": 0.1628882959485054,
      "num_tokens": 19302701.0,
      "step": 8810
    },
    {
      "entropy": 5.597626638412476,
      "epoch": 0.8748511314013497,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004932207718500868,
      "loss": 5.3214,
      "mean_token_accuracy": 0.15882567018270494,
      "num_tokens": 19313024.0,
      "step": 8815
    },
    {
      "entropy": 5.588980340957642,
      "epoch": 0.8753473600635173,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004932121372295112,
      "loss": 5.3338,
      "mean_token_accuracy": 0.15698196291923522,
      "num_tokens": 19324014.0,
      "step": 8820
    },
    {
      "entropy": 5.558774614334107,
      "epoch": 0.8758435887256848,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004932034971977448,
      "loss": 5.3609,
      "mean_token_accuracy": 0.15866709500551224,
      "num_tokens": 19334226.0,
      "step": 8825
    },
    {
      "entropy": 5.643746709823608,
      "epoch": 0.8763398173878523,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004931948517550019,
      "loss": 5.3976,
      "mean_token_accuracy": 0.15261635184288025,
      "num_tokens": 19344363.0,
      "step": 8830
    },
    {
      "entropy": 5.588718843460083,
      "epoch": 0.8768360460500199,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004931862009014968,
      "loss": 5.4166,
      "mean_token_accuracy": 0.15612666457891464,
      "num_tokens": 19354915.0,
      "step": 8835
    },
    {
      "entropy": 5.640223169326783,
      "epoch": 0.8773322747121873,
      "grad_norm": 0.82421875,
      "learning_rate": 0.000493177544637444,
      "loss": 5.4139,
      "mean_token_accuracy": 0.15273035913705826,
      "num_tokens": 19365570.0,
      "step": 8840
    },
    {
      "entropy": 5.627228689193726,
      "epoch": 0.8778285033743549,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004931688829630582,
      "loss": 5.3306,
      "mean_token_accuracy": 0.16338916271924972,
      "num_tokens": 19375731.0,
      "step": 8845
    },
    {
      "entropy": 5.562938117980957,
      "epoch": 0.8783247320365224,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004931602158785544,
      "loss": 5.2857,
      "mean_token_accuracy": 0.16078079342842103,
      "num_tokens": 19386248.0,
      "step": 8850
    },
    {
      "entropy": 5.56929874420166,
      "epoch": 0.87882096069869,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004931515433841472,
      "loss": 5.298,
      "mean_token_accuracy": 0.163205586373806,
      "num_tokens": 19397593.0,
      "step": 8855
    },
    {
      "entropy": 5.600165939331054,
      "epoch": 0.8793171893608575,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004931428654800518,
      "loss": 5.3875,
      "mean_token_accuracy": 0.15078756362199783,
      "num_tokens": 19409701.0,
      "step": 8860
    },
    {
      "entropy": 5.584780073165893,
      "epoch": 0.879813418023025,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004931341821664834,
      "loss": 5.3642,
      "mean_token_accuracy": 0.15765655636787415,
      "num_tokens": 19420280.0,
      "step": 8865
    },
    {
      "entropy": 5.602287340164184,
      "epoch": 0.8803096466851925,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004931254934436573,
      "loss": 5.3432,
      "mean_token_accuracy": 0.15591977834701537,
      "num_tokens": 19432208.0,
      "step": 8870
    },
    {
      "entropy": 5.682718276977539,
      "epoch": 0.8808058753473601,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004931167993117891,
      "loss": 5.453,
      "mean_token_accuracy": 0.15500149577856065,
      "num_tokens": 19443333.0,
      "step": 8875
    },
    {
      "entropy": 5.497708368301391,
      "epoch": 0.8813021040095276,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004931080997710942,
      "loss": 5.2273,
      "mean_token_accuracy": 0.16059327721595765,
      "num_tokens": 19452867.0,
      "step": 8880
    },
    {
      "entropy": 5.498044013977051,
      "epoch": 0.8817983326716952,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004930993948217884,
      "loss": 5.2697,
      "mean_token_accuracy": 0.16537645161151887,
      "num_tokens": 19463021.0,
      "step": 8885
    },
    {
      "entropy": 5.6087535381317135,
      "epoch": 0.8822945613338626,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004930906844640876,
      "loss": 5.3973,
      "mean_token_accuracy": 0.1516403153538704,
      "num_tokens": 19474092.0,
      "step": 8890
    },
    {
      "entropy": 5.563568878173828,
      "epoch": 0.8827907899960302,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004930819686982077,
      "loss": 5.367,
      "mean_token_accuracy": 0.15222940146923064,
      "num_tokens": 19486035.0,
      "step": 8895
    },
    {
      "entropy": 5.539292335510254,
      "epoch": 0.8832870186581977,
      "grad_norm": 0.80859375,
      "learning_rate": 0.000493073247524365,
      "loss": 5.293,
      "mean_token_accuracy": 0.1607646718621254,
      "num_tokens": 19497484.0,
      "step": 8900
    },
    {
      "entropy": 5.5990314960479735,
      "epoch": 0.8837832473203652,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004930645209427756,
      "loss": 5.4007,
      "mean_token_accuracy": 0.15190002620220183,
      "num_tokens": 19508463.0,
      "step": 8905
    },
    {
      "entropy": 5.594696378707885,
      "epoch": 0.8842794759825328,
      "grad_norm": 0.81640625,
      "learning_rate": 0.000493055788953656,
      "loss": 5.3361,
      "mean_token_accuracy": 0.15622231513261794,
      "num_tokens": 19520260.0,
      "step": 8910
    },
    {
      "entropy": 5.547523164749146,
      "epoch": 0.8847757046447002,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004930470515572227,
      "loss": 5.2948,
      "mean_token_accuracy": 0.15937747359275817,
      "num_tokens": 19531621.0,
      "step": 8915
    },
    {
      "entropy": 5.589425563812256,
      "epoch": 0.8852719333068678,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004930383087536923,
      "loss": 5.2709,
      "mean_token_accuracy": 0.15654692500829698,
      "num_tokens": 19543271.0,
      "step": 8920
    },
    {
      "entropy": 5.649059867858886,
      "epoch": 0.8857681619690353,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004930295605432818,
      "loss": 5.4347,
      "mean_token_accuracy": 0.16115086674690246,
      "num_tokens": 19555178.0,
      "step": 8925
    },
    {
      "entropy": 5.625226259231567,
      "epoch": 0.8862643906312029,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004930208069262079,
      "loss": 5.3564,
      "mean_token_accuracy": 0.15755318850278854,
      "num_tokens": 19565153.0,
      "step": 8930
    },
    {
      "entropy": 5.525291490554809,
      "epoch": 0.8867606192933704,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004930120479026878,
      "loss": 5.2907,
      "mean_token_accuracy": 0.16522422730922698,
      "num_tokens": 19575608.0,
      "step": 8935
    },
    {
      "entropy": 5.594865846633911,
      "epoch": 0.8872568479555379,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004930032834729388,
      "loss": 5.25,
      "mean_token_accuracy": 0.16316999047994613,
      "num_tokens": 19587556.0,
      "step": 8940
    },
    {
      "entropy": 5.513002729415893,
      "epoch": 0.8877530766177054,
      "grad_norm": 0.859375,
      "learning_rate": 0.000492994513637178,
      "loss": 5.3003,
      "mean_token_accuracy": 0.15882664620876313,
      "num_tokens": 19597341.0,
      "step": 8945
    },
    {
      "entropy": 5.5407734394073485,
      "epoch": 0.888249305279873,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004929857383956229,
      "loss": 5.3284,
      "mean_token_accuracy": 0.15448749512434007,
      "num_tokens": 19608431.0,
      "step": 8950
    },
    {
      "entropy": 5.565536546707153,
      "epoch": 0.8887455339420405,
      "grad_norm": 1.0,
      "learning_rate": 0.0004929769577484914,
      "loss": 5.3106,
      "mean_token_accuracy": 0.16429621279239653,
      "num_tokens": 19619645.0,
      "step": 8955
    },
    {
      "entropy": 5.532659196853638,
      "epoch": 0.8892417626042081,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000492968171696001,
      "loss": 5.2675,
      "mean_token_accuracy": 0.16849614828824996,
      "num_tokens": 19630412.0,
      "step": 8960
    },
    {
      "entropy": 5.61044225692749,
      "epoch": 0.8897379912663755,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004929593802383696,
      "loss": 5.3337,
      "mean_token_accuracy": 0.15720237493515016,
      "num_tokens": 19640055.0,
      "step": 8965
    },
    {
      "entropy": 5.5500359535217285,
      "epoch": 0.8902342199285431,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004929505833758152,
      "loss": 5.3467,
      "mean_token_accuracy": 0.1521917000412941,
      "num_tokens": 19650814.0,
      "step": 8970
    },
    {
      "entropy": 5.59323296546936,
      "epoch": 0.8907304485907106,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004929417811085561,
      "loss": 5.3229,
      "mean_token_accuracy": 0.16198962926864624,
      "num_tokens": 19663669.0,
      "step": 8975
    },
    {
      "entropy": 5.661463022232056,
      "epoch": 0.8912266772528781,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004929329734368103,
      "loss": 5.4089,
      "mean_token_accuracy": 0.15372903198003768,
      "num_tokens": 19675851.0,
      "step": 8980
    },
    {
      "entropy": 5.586727428436279,
      "epoch": 0.8917229059150457,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004929241603607964,
      "loss": 5.292,
      "mean_token_accuracy": 0.16683239191770555,
      "num_tokens": 19686613.0,
      "step": 8985
    },
    {
      "entropy": 5.494939517974854,
      "epoch": 0.8922191345772131,
      "grad_norm": 0.875,
      "learning_rate": 0.000492915341880733,
      "loss": 5.2283,
      "mean_token_accuracy": 0.15991657227277756,
      "num_tokens": 19696419.0,
      "step": 8990
    },
    {
      "entropy": 5.560398435592651,
      "epoch": 0.8927153632393807,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004929065179968387,
      "loss": 5.3245,
      "mean_token_accuracy": 0.15727977454662323,
      "num_tokens": 19707269.0,
      "step": 8995
    },
    {
      "entropy": 5.645032024383545,
      "epoch": 0.8932115919015482,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004928976887093321,
      "loss": 5.3703,
      "mean_token_accuracy": 0.15670965164899825,
      "num_tokens": 19716099.0,
      "step": 9000
    },
    {
      "epoch": 0.8932115919015482,
      "eval_entropy": 5.385663988471853,
      "eval_loss": 5.402402400970459,
      "eval_mean_token_accuracy": 0.159667709009024,
      "eval_num_tokens": 19716099.0,
      "eval_runtime": 20.4365,
      "eval_samples_per_second": 1647.492,
      "eval_steps_per_second": 205.955,
      "step": 9000
    },
    {
      "entropy": 5.568923807144165,
      "epoch": 0.8937078205637158,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004928888540184325,
      "loss": 5.3618,
      "mean_token_accuracy": 0.15348413735628127,
      "num_tokens": 19726632.0,
      "step": 9005
    },
    {
      "entropy": 5.54155421257019,
      "epoch": 0.8942040492258833,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004928800139243587,
      "loss": 5.2618,
      "mean_token_accuracy": 0.162228325009346,
      "num_tokens": 19737929.0,
      "step": 9010
    },
    {
      "entropy": 5.564952373504639,
      "epoch": 0.8947002778880508,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004928711684273302,
      "loss": 5.2871,
      "mean_token_accuracy": 0.15866970866918564,
      "num_tokens": 19748191.0,
      "step": 9015
    },
    {
      "entropy": 5.584346628189087,
      "epoch": 0.8951965065502183,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004928623175275661,
      "loss": 5.3389,
      "mean_token_accuracy": 0.16422507613897325,
      "num_tokens": 19758892.0,
      "step": 9020
    },
    {
      "entropy": 5.53707799911499,
      "epoch": 0.8956927352123859,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004928534612252862,
      "loss": 5.2234,
      "mean_token_accuracy": 0.16193969398736954,
      "num_tokens": 19769120.0,
      "step": 9025
    },
    {
      "entropy": 5.522628831863403,
      "epoch": 0.8961889638745534,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004928445995207095,
      "loss": 5.2856,
      "mean_token_accuracy": 0.16119020879268647,
      "num_tokens": 19779484.0,
      "step": 9030
    },
    {
      "entropy": 5.590908098220825,
      "epoch": 0.896685192536721,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004928357324140564,
      "loss": 5.319,
      "mean_token_accuracy": 0.15963810235261916,
      "num_tokens": 19790438.0,
      "step": 9035
    },
    {
      "entropy": 5.64152512550354,
      "epoch": 0.8971814211988884,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004928268599055465,
      "loss": 5.4056,
      "mean_token_accuracy": 0.15609167665243148,
      "num_tokens": 19801646.0,
      "step": 9040
    },
    {
      "entropy": 5.561574840545655,
      "epoch": 0.897677649861056,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004928179819953998,
      "loss": 5.2805,
      "mean_token_accuracy": 0.1638476550579071,
      "num_tokens": 19812887.0,
      "step": 9045
    },
    {
      "entropy": 5.5434667587280275,
      "epoch": 0.8981738785232235,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004928090986838366,
      "loss": 5.3475,
      "mean_token_accuracy": 0.1530725747346878,
      "num_tokens": 19822938.0,
      "step": 9050
    },
    {
      "entropy": 5.565565156936645,
      "epoch": 0.898670107185391,
      "grad_norm": 1.046875,
      "learning_rate": 0.000492800209971077,
      "loss": 5.2934,
      "mean_token_accuracy": 0.16092495620250702,
      "num_tokens": 19833988.0,
      "step": 9055
    },
    {
      "entropy": 5.635081243515015,
      "epoch": 0.8991663358475586,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004927913158573416,
      "loss": 5.3763,
      "mean_token_accuracy": 0.15459856390953064,
      "num_tokens": 19845112.0,
      "step": 9060
    },
    {
      "entropy": 5.5673318862915036,
      "epoch": 0.899662564509726,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004927824163428508,
      "loss": 5.3355,
      "mean_token_accuracy": 0.15746665894985198,
      "num_tokens": 19856304.0,
      "step": 9065
    },
    {
      "entropy": 5.552544832229614,
      "epoch": 0.9001587931718936,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004927735114278254,
      "loss": 5.3159,
      "mean_token_accuracy": 0.15865050554275512,
      "num_tokens": 19866776.0,
      "step": 9070
    },
    {
      "entropy": 5.585403299331665,
      "epoch": 0.9006550218340611,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004927646011124861,
      "loss": 5.2922,
      "mean_token_accuracy": 0.1570994071662426,
      "num_tokens": 19876392.0,
      "step": 9075
    },
    {
      "entropy": 5.599952363967896,
      "epoch": 0.9011512504962287,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004927556853970539,
      "loss": 5.3235,
      "mean_token_accuracy": 0.15793648958206177,
      "num_tokens": 19887362.0,
      "step": 9080
    },
    {
      "entropy": 5.556565523147583,
      "epoch": 0.9016474791583962,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004927467642817499,
      "loss": 5.3017,
      "mean_token_accuracy": 0.16025232523679733,
      "num_tokens": 19898989.0,
      "step": 9085
    },
    {
      "entropy": 5.594432783126831,
      "epoch": 0.9021437078205637,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004927378377667955,
      "loss": 5.314,
      "mean_token_accuracy": 0.15048023015260698,
      "num_tokens": 19910112.0,
      "step": 9090
    },
    {
      "entropy": 5.634715747833252,
      "epoch": 0.9026399364827312,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000492728905852412,
      "loss": 5.4243,
      "mean_token_accuracy": 0.15499950647354127,
      "num_tokens": 19922601.0,
      "step": 9095
    },
    {
      "entropy": 5.572630405426025,
      "epoch": 0.9031361651448988,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004927199685388205,
      "loss": 5.3382,
      "mean_token_accuracy": 0.16147261261940002,
      "num_tokens": 19934153.0,
      "step": 9100
    },
    {
      "entropy": 5.596471452713013,
      "epoch": 0.9036323938070663,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000492711025826243,
      "loss": 5.2962,
      "mean_token_accuracy": 0.16143666207790375,
      "num_tokens": 19944572.0,
      "step": 9105
    },
    {
      "entropy": 5.578424310684204,
      "epoch": 0.9041286224692339,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004927020777149014,
      "loss": 5.4103,
      "mean_token_accuracy": 0.15403326451778412,
      "num_tokens": 19956431.0,
      "step": 9110
    },
    {
      "entropy": 5.610340118408203,
      "epoch": 0.9046248511314013,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004926931242050171,
      "loss": 5.3319,
      "mean_token_accuracy": 0.15525012612342834,
      "num_tokens": 19968118.0,
      "step": 9115
    },
    {
      "entropy": 5.596913480758667,
      "epoch": 0.9051210797935689,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004926841652968125,
      "loss": 5.3947,
      "mean_token_accuracy": 0.16022013872861862,
      "num_tokens": 19979545.0,
      "step": 9120
    },
    {
      "entropy": 5.583328580856323,
      "epoch": 0.9056173084557364,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004926752009905099,
      "loss": 5.3563,
      "mean_token_accuracy": 0.15155109912157058,
      "num_tokens": 19989832.0,
      "step": 9125
    },
    {
      "entropy": 5.5760022640228275,
      "epoch": 0.9061135371179039,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004926662312863311,
      "loss": 5.3614,
      "mean_token_accuracy": 0.15715830028057098,
      "num_tokens": 20000935.0,
      "step": 9130
    },
    {
      "entropy": 5.5620228290557865,
      "epoch": 0.9066097657800715,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004926572561844988,
      "loss": 5.3679,
      "mean_token_accuracy": 0.15192281007766723,
      "num_tokens": 20011485.0,
      "step": 9135
    },
    {
      "entropy": 5.557519960403442,
      "epoch": 0.9071059944422389,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004926482756852356,
      "loss": 5.2233,
      "mean_token_accuracy": 0.16481838822364808,
      "num_tokens": 20023043.0,
      "step": 9140
    },
    {
      "entropy": 5.61761155128479,
      "epoch": 0.9076022231044065,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004926392897887642,
      "loss": 5.3365,
      "mean_token_accuracy": 0.164206799864769,
      "num_tokens": 20032437.0,
      "step": 9145
    },
    {
      "entropy": 5.577384328842163,
      "epoch": 0.908098451766574,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004926302984953073,
      "loss": 5.3653,
      "mean_token_accuracy": 0.15488700121641158,
      "num_tokens": 20044662.0,
      "step": 9150
    },
    {
      "entropy": 5.567062854766846,
      "epoch": 0.9085946804287416,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004926213018050881,
      "loss": 5.3657,
      "mean_token_accuracy": 0.15862423181533813,
      "num_tokens": 20055924.0,
      "step": 9155
    },
    {
      "entropy": 5.542328882217407,
      "epoch": 0.9090909090909091,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004926122997183295,
      "loss": 5.3124,
      "mean_token_accuracy": 0.15485553443431854,
      "num_tokens": 20067909.0,
      "step": 9160
    },
    {
      "entropy": 5.589487886428833,
      "epoch": 0.9095871377530766,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004926032922352546,
      "loss": 5.2896,
      "mean_token_accuracy": 0.1649667114019394,
      "num_tokens": 20079333.0,
      "step": 9165
    },
    {
      "entropy": 5.575442457199097,
      "epoch": 0.9100833664152441,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004925942793560872,
      "loss": 5.2694,
      "mean_token_accuracy": 0.1613793909549713,
      "num_tokens": 20089741.0,
      "step": 9170
    },
    {
      "entropy": 5.530794143676758,
      "epoch": 0.9105795950774117,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004925852610810504,
      "loss": 5.3042,
      "mean_token_accuracy": 0.15670829117298127,
      "num_tokens": 20100267.0,
      "step": 9175
    },
    {
      "entropy": 5.515627527236939,
      "epoch": 0.9110758237395792,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000492576237410368,
      "loss": 5.2667,
      "mean_token_accuracy": 0.1595083385705948,
      "num_tokens": 20111964.0,
      "step": 9180
    },
    {
      "entropy": 5.59562497138977,
      "epoch": 0.9115720524017468,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004925672083442638,
      "loss": 5.3399,
      "mean_token_accuracy": 0.15084489285945893,
      "num_tokens": 20123132.0,
      "step": 9185
    },
    {
      "entropy": 5.54755129814148,
      "epoch": 0.9120682810639142,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004925581738829616,
      "loss": 5.243,
      "mean_token_accuracy": 0.1625302255153656,
      "num_tokens": 20134050.0,
      "step": 9190
    },
    {
      "entropy": 5.547417402267456,
      "epoch": 0.9125645097260818,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004925491340266855,
      "loss": 5.3128,
      "mean_token_accuracy": 0.16251576989889144,
      "num_tokens": 20143814.0,
      "step": 9195
    },
    {
      "entropy": 5.597006225585938,
      "epoch": 0.9130607383882493,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0004925400887756596,
      "loss": 5.3789,
      "mean_token_accuracy": 0.1628500372171402,
      "num_tokens": 20157647.0,
      "step": 9200
    },
    {
      "entropy": 5.598450517654419,
      "epoch": 0.9135569670504168,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004925310381301083,
      "loss": 5.3451,
      "mean_token_accuracy": 0.1600103721022606,
      "num_tokens": 20168366.0,
      "step": 9205
    },
    {
      "entropy": 5.6027578830719,
      "epoch": 0.9140531957125844,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000492521982090256,
      "loss": 5.3884,
      "mean_token_accuracy": 0.1589035838842392,
      "num_tokens": 20179362.0,
      "step": 9210
    },
    {
      "entropy": 5.5816926002502445,
      "epoch": 0.9145494243747518,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004925129206563273,
      "loss": 5.2859,
      "mean_token_accuracy": 0.1607089161872864,
      "num_tokens": 20190320.0,
      "step": 9215
    },
    {
      "entropy": 5.527514171600342,
      "epoch": 0.9150456530369194,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004925038538285467,
      "loss": 5.2998,
      "mean_token_accuracy": 0.15989586859941482,
      "num_tokens": 20201114.0,
      "step": 9220
    },
    {
      "entropy": 5.6007684707641605,
      "epoch": 0.9155418816990869,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004924947816071394,
      "loss": 5.3214,
      "mean_token_accuracy": 0.16533045768737792,
      "num_tokens": 20213351.0,
      "step": 9225
    },
    {
      "entropy": 5.560990667343139,
      "epoch": 0.9160381103612545,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00049248570399233,
      "loss": 5.2319,
      "mean_token_accuracy": 0.16471796929836274,
      "num_tokens": 20223376.0,
      "step": 9230
    },
    {
      "entropy": 5.603248834609985,
      "epoch": 0.916534339023422,
      "grad_norm": 0.90625,
      "learning_rate": 0.000492476620984344,
      "loss": 5.3694,
      "mean_token_accuracy": 0.15361211448907852,
      "num_tokens": 20234248.0,
      "step": 9235
    },
    {
      "entropy": 5.5417145729064945,
      "epoch": 0.9170305676855895,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004924675325834063,
      "loss": 5.2953,
      "mean_token_accuracy": 0.1586267977952957,
      "num_tokens": 20245823.0,
      "step": 9240
    },
    {
      "entropy": 5.612440586090088,
      "epoch": 0.917526796347757,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004924584387897424,
      "loss": 5.3455,
      "mean_token_accuracy": 0.1560446783900261,
      "num_tokens": 20257003.0,
      "step": 9245
    },
    {
      "entropy": 5.644407939910889,
      "epoch": 0.9180230250099246,
      "grad_norm": 0.875,
      "learning_rate": 0.0004924493396035777,
      "loss": 5.4769,
      "mean_token_accuracy": 0.1485078290104866,
      "num_tokens": 20268500.0,
      "step": 9250
    },
    {
      "entropy": 5.640565633773804,
      "epoch": 0.9185192536720921,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000492440235025138,
      "loss": 5.3488,
      "mean_token_accuracy": 0.1620658427476883,
      "num_tokens": 20279301.0,
      "step": 9255
    },
    {
      "entropy": 5.704019403457641,
      "epoch": 0.9190154823342597,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004924311250546491,
      "loss": 5.4476,
      "mean_token_accuracy": 0.1500500410795212,
      "num_tokens": 20291263.0,
      "step": 9260
    },
    {
      "entropy": 5.563519525527954,
      "epoch": 0.9195117109964271,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004924220096923367,
      "loss": 5.3323,
      "mean_token_accuracy": 0.16030509471893312,
      "num_tokens": 20301554.0,
      "step": 9265
    },
    {
      "entropy": 5.636999130249023,
      "epoch": 0.9200079396585947,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004924128889384271,
      "loss": 5.3545,
      "mean_token_accuracy": 0.1548002302646637,
      "num_tokens": 20314299.0,
      "step": 9270
    },
    {
      "entropy": 5.541638326644898,
      "epoch": 0.9205041683207622,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004924037627931463,
      "loss": 5.3274,
      "mean_token_accuracy": 0.15818847119808196,
      "num_tokens": 20324595.0,
      "step": 9275
    },
    {
      "entropy": 5.573790740966797,
      "epoch": 0.9210003969829297,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004923946312567206,
      "loss": 5.2908,
      "mean_token_accuracy": 0.15841273218393326,
      "num_tokens": 20335744.0,
      "step": 9280
    },
    {
      "entropy": 5.513968849182129,
      "epoch": 0.9214966256450973,
      "grad_norm": 0.875,
      "learning_rate": 0.0004923854943293765,
      "loss": 5.2173,
      "mean_token_accuracy": 0.15823314040899278,
      "num_tokens": 20346072.0,
      "step": 9285
    },
    {
      "entropy": 5.556331920623779,
      "epoch": 0.9219928543072647,
      "grad_norm": 0.7578125,
      "learning_rate": 0.0004923763520113406,
      "loss": 5.2921,
      "mean_token_accuracy": 0.16727010756731034,
      "num_tokens": 20357747.0,
      "step": 9290
    },
    {
      "entropy": 5.60493221282959,
      "epoch": 0.9224890829694323,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004923672043028395,
      "loss": 5.2813,
      "mean_token_accuracy": 0.1622532695531845,
      "num_tokens": 20368838.0,
      "step": 9295
    },
    {
      "entropy": 5.570999622344971,
      "epoch": 0.9229853116315998,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004923580512041001,
      "loss": 5.3543,
      "mean_token_accuracy": 0.15649488866329192,
      "num_tokens": 20380844.0,
      "step": 9300
    },
    {
      "entropy": 5.5870014190673825,
      "epoch": 0.9234815402937674,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004923488927153495,
      "loss": 5.3599,
      "mean_token_accuracy": 0.15315917357802392,
      "num_tokens": 20391739.0,
      "step": 9305
    },
    {
      "entropy": 5.597708320617675,
      "epoch": 0.9239777689559349,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004923397288368146,
      "loss": 5.2452,
      "mean_token_accuracy": 0.16886815428733826,
      "num_tokens": 20402571.0,
      "step": 9310
    },
    {
      "entropy": 5.564769697189331,
      "epoch": 0.9244739976181024,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004923305595687229,
      "loss": 5.2931,
      "mean_token_accuracy": 0.1555173598229885,
      "num_tokens": 20414766.0,
      "step": 9315
    },
    {
      "entropy": 5.518140077590942,
      "epoch": 0.9249702262802699,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004923213849113015,
      "loss": 5.2667,
      "mean_token_accuracy": 0.15986013263463975,
      "num_tokens": 20425508.0,
      "step": 9320
    },
    {
      "entropy": 5.535608434677124,
      "epoch": 0.9254664549424375,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004923122048647781,
      "loss": 5.2539,
      "mean_token_accuracy": 0.16222997456789018,
      "num_tokens": 20436128.0,
      "step": 9325
    },
    {
      "entropy": 5.5474447250366214,
      "epoch": 0.925962683604605,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004923030194293802,
      "loss": 5.3542,
      "mean_token_accuracy": 0.15925963670015336,
      "num_tokens": 20447212.0,
      "step": 9330
    },
    {
      "entropy": 5.557111740112305,
      "epoch": 0.9264589122667726,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004922938286053358,
      "loss": 5.2775,
      "mean_token_accuracy": 0.16774797141551973,
      "num_tokens": 20457770.0,
      "step": 9335
    },
    {
      "entropy": 5.611376190185547,
      "epoch": 0.92695514092894,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004922846323928726,
      "loss": 5.3509,
      "mean_token_accuracy": 0.15740771889686583,
      "num_tokens": 20468172.0,
      "step": 9340
    },
    {
      "entropy": 5.600006580352783,
      "epoch": 0.9274513695911076,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004922754307922188,
      "loss": 5.2834,
      "mean_token_accuracy": 0.1612941011786461,
      "num_tokens": 20479429.0,
      "step": 9345
    },
    {
      "entropy": 5.548315954208374,
      "epoch": 0.9279475982532751,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004922662238036024,
      "loss": 5.2487,
      "mean_token_accuracy": 0.1618860051035881,
      "num_tokens": 20490373.0,
      "step": 9350
    },
    {
      "entropy": 5.612733983993531,
      "epoch": 0.9284438269154426,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004922570114272518,
      "loss": 5.3664,
      "mean_token_accuracy": 0.1572950229048729,
      "num_tokens": 20500002.0,
      "step": 9355
    },
    {
      "entropy": 5.613262462615967,
      "epoch": 0.9289400555776102,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004922477936633955,
      "loss": 5.4121,
      "mean_token_accuracy": 0.1551825225353241,
      "num_tokens": 20512990.0,
      "step": 9360
    },
    {
      "entropy": 5.530987501144409,
      "epoch": 0.9294362842397776,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004922385705122621,
      "loss": 5.3286,
      "mean_token_accuracy": 0.16080232709646225,
      "num_tokens": 20522979.0,
      "step": 9365
    },
    {
      "entropy": 5.543492984771729,
      "epoch": 0.9299325129019452,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004922293419740802,
      "loss": 5.2089,
      "mean_token_accuracy": 0.16384829133749007,
      "num_tokens": 20534620.0,
      "step": 9370
    },
    {
      "entropy": 5.591291666030884,
      "epoch": 0.9304287415641127,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004922201080490787,
      "loss": 5.3088,
      "mean_token_accuracy": 0.15611242055892943,
      "num_tokens": 20545526.0,
      "step": 9375
    },
    {
      "entropy": 5.628178215026855,
      "epoch": 0.9309249702262803,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004922108687374867,
      "loss": 5.4732,
      "mean_token_accuracy": 0.15207466185092927,
      "num_tokens": 20556665.0,
      "step": 9380
    },
    {
      "entropy": 5.604941940307617,
      "epoch": 0.9314211988884478,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000492201624039533,
      "loss": 5.3144,
      "mean_token_accuracy": 0.1533210664987564,
      "num_tokens": 20566054.0,
      "step": 9385
    },
    {
      "entropy": 5.66755747795105,
      "epoch": 0.9319174275506154,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004921923739554472,
      "loss": 5.3927,
      "mean_token_accuracy": 0.15216393172740936,
      "num_tokens": 20577922.0,
      "step": 9390
    },
    {
      "entropy": 5.513110399246216,
      "epoch": 0.9324136562127828,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004921831184854585,
      "loss": 5.1973,
      "mean_token_accuracy": 0.16484288424253463,
      "num_tokens": 20588473.0,
      "step": 9395
    },
    {
      "entropy": 5.553794431686401,
      "epoch": 0.9329098848749504,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004921738576297965,
      "loss": 5.3837,
      "mean_token_accuracy": 0.158964903652668,
      "num_tokens": 20599371.0,
      "step": 9400
    },
    {
      "entropy": 5.541663026809692,
      "epoch": 0.9334061135371179,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004921645913886907,
      "loss": 5.2684,
      "mean_token_accuracy": 0.15712719410657883,
      "num_tokens": 20610066.0,
      "step": 9405
    },
    {
      "entropy": 5.582781600952148,
      "epoch": 0.9339023421992855,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004921553197623711,
      "loss": 5.373,
      "mean_token_accuracy": 0.1548023119568825,
      "num_tokens": 20621804.0,
      "step": 9410
    },
    {
      "entropy": 5.557465696334839,
      "epoch": 0.934398570861453,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004921460427510674,
      "loss": 5.2563,
      "mean_token_accuracy": 0.16722989529371263,
      "num_tokens": 20632706.0,
      "step": 9415
    },
    {
      "entropy": 5.557167625427246,
      "epoch": 0.9348947995236205,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004921367603550098,
      "loss": 5.3667,
      "mean_token_accuracy": 0.1509762555360794,
      "num_tokens": 20643303.0,
      "step": 9420
    },
    {
      "entropy": 5.596539688110352,
      "epoch": 0.935391028185788,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004921274725744283,
      "loss": 5.3155,
      "mean_token_accuracy": 0.16477171033620835,
      "num_tokens": 20653681.0,
      "step": 9425
    },
    {
      "entropy": 5.583985042572022,
      "epoch": 0.9358872568479555,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004921181794095536,
      "loss": 5.3219,
      "mean_token_accuracy": 0.15845015048980712,
      "num_tokens": 20665599.0,
      "step": 9430
    },
    {
      "entropy": 5.496887588500977,
      "epoch": 0.9363834855101231,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004921088808606157,
      "loss": 5.225,
      "mean_token_accuracy": 0.16497879475355148,
      "num_tokens": 20675726.0,
      "step": 9435
    },
    {
      "entropy": 5.586679458618164,
      "epoch": 0.9368797141722905,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004920995769278454,
      "loss": 5.4173,
      "mean_token_accuracy": 0.1540427625179291,
      "num_tokens": 20688083.0,
      "step": 9440
    },
    {
      "entropy": 5.63616795539856,
      "epoch": 0.9373759428344581,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004920902676114734,
      "loss": 5.2845,
      "mean_token_accuracy": 0.16210053712129593,
      "num_tokens": 20697642.0,
      "step": 9445
    },
    {
      "entropy": 5.618163156509399,
      "epoch": 0.9378721714966256,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004920809529117306,
      "loss": 5.3632,
      "mean_token_accuracy": 0.15220556557178497,
      "num_tokens": 20709172.0,
      "step": 9450
    },
    {
      "entropy": 5.606362199783325,
      "epoch": 0.9383684001587932,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004920716328288479,
      "loss": 5.4067,
      "mean_token_accuracy": 0.15718362331390381,
      "num_tokens": 20720563.0,
      "step": 9455
    },
    {
      "entropy": 5.5817773818969725,
      "epoch": 0.9388646288209607,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004920623073630566,
      "loss": 5.2902,
      "mean_token_accuracy": 0.16570660322904587,
      "num_tokens": 20730943.0,
      "step": 9460
    },
    {
      "entropy": 5.558560037612915,
      "epoch": 0.9393608574831283,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004920529765145876,
      "loss": 5.2801,
      "mean_token_accuracy": 0.1606566086411476,
      "num_tokens": 20740903.0,
      "step": 9465
    },
    {
      "entropy": 5.578950977325439,
      "epoch": 0.9398570861452957,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004920436402836726,
      "loss": 5.3513,
      "mean_token_accuracy": 0.1639459028840065,
      "num_tokens": 20750580.0,
      "step": 9470
    },
    {
      "entropy": 5.55942850112915,
      "epoch": 0.9403533148074633,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000492034298670543,
      "loss": 5.2823,
      "mean_token_accuracy": 0.15864358246326446,
      "num_tokens": 20761357.0,
      "step": 9475
    },
    {
      "entropy": 5.63238263130188,
      "epoch": 0.9408495434696308,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004920249516754304,
      "loss": 5.4247,
      "mean_token_accuracy": 0.1501690626144409,
      "num_tokens": 20773273.0,
      "step": 9480
    },
    {
      "entropy": 5.5499388694763185,
      "epoch": 0.9413457721317984,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004920155992985667,
      "loss": 5.2721,
      "mean_token_accuracy": 0.15872172266244888,
      "num_tokens": 20784290.0,
      "step": 9485
    },
    {
      "entropy": 5.595738077163697,
      "epoch": 0.9418420007939659,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004920062415401837,
      "loss": 5.2966,
      "mean_token_accuracy": 0.16069611459970473,
      "num_tokens": 20794764.0,
      "step": 9490
    },
    {
      "entropy": 5.63782000541687,
      "epoch": 0.9423382294561334,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004919968784005136,
      "loss": 5.3506,
      "mean_token_accuracy": 0.151334947347641,
      "num_tokens": 20804863.0,
      "step": 9495
    },
    {
      "entropy": 5.649145221710205,
      "epoch": 0.9428344581183009,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004919875098797886,
      "loss": 5.4244,
      "mean_token_accuracy": 0.15182149410247803,
      "num_tokens": 20816445.0,
      "step": 9500
    },
    {
      "entropy": 5.596583557128906,
      "epoch": 0.9433306867804684,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004919781359782407,
      "loss": 5.2546,
      "mean_token_accuracy": 0.16618213802576065,
      "num_tokens": 20827141.0,
      "step": 9505
    },
    {
      "entropy": 5.553158521652222,
      "epoch": 0.943826915442636,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004919687566961027,
      "loss": 5.3127,
      "mean_token_accuracy": 0.15420710295438766,
      "num_tokens": 20838291.0,
      "step": 9510
    },
    {
      "entropy": 5.6420862674713135,
      "epoch": 0.9443231441048034,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000491959372033607,
      "loss": 5.3006,
      "mean_token_accuracy": 0.16112050563097,
      "num_tokens": 20848839.0,
      "step": 9515
    },
    {
      "entropy": 5.588303565979004,
      "epoch": 0.944819372766971,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004919499819909865,
      "loss": 5.2852,
      "mean_token_accuracy": 0.16203968077898026,
      "num_tokens": 20859961.0,
      "step": 9520
    },
    {
      "entropy": 5.488934516906738,
      "epoch": 0.9453156014291385,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004919405865684738,
      "loss": 5.2296,
      "mean_token_accuracy": 0.16484673768281938,
      "num_tokens": 20869720.0,
      "step": 9525
    },
    {
      "entropy": 5.56879243850708,
      "epoch": 0.9458118300913061,
      "grad_norm": 0.953125,
      "learning_rate": 0.000491931185766302,
      "loss": 5.3764,
      "mean_token_accuracy": 0.16189633160829545,
      "num_tokens": 20880513.0,
      "step": 9530
    },
    {
      "entropy": 5.624775218963623,
      "epoch": 0.9463080587534736,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0004919217795847044,
      "loss": 5.2941,
      "mean_token_accuracy": 0.16495498567819594,
      "num_tokens": 20892456.0,
      "step": 9535
    },
    {
      "entropy": 5.552755737304688,
      "epoch": 0.9468042874156412,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004919123680239139,
      "loss": 5.293,
      "mean_token_accuracy": 0.15933222323656082,
      "num_tokens": 20904500.0,
      "step": 9540
    },
    {
      "entropy": 5.602396154403687,
      "epoch": 0.9473005160778086,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004919029510841641,
      "loss": 5.2781,
      "mean_token_accuracy": 0.1620924264192581,
      "num_tokens": 20915133.0,
      "step": 9545
    },
    {
      "entropy": 5.514210557937622,
      "epoch": 0.9477967447399762,
      "grad_norm": 1.0,
      "learning_rate": 0.0004918935287656887,
      "loss": 5.2799,
      "mean_token_accuracy": 0.16024440079927443,
      "num_tokens": 20924779.0,
      "step": 9550
    },
    {
      "entropy": 5.482671308517456,
      "epoch": 0.9482929734021437,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000491884101068721,
      "loss": 5.2177,
      "mean_token_accuracy": 0.1628309369087219,
      "num_tokens": 20935387.0,
      "step": 9555
    },
    {
      "entropy": 5.635864973068237,
      "epoch": 0.9487892020643113,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004918746679934949,
      "loss": 5.3922,
      "mean_token_accuracy": 0.1568456321954727,
      "num_tokens": 20945432.0,
      "step": 9560
    },
    {
      "entropy": 5.56051926612854,
      "epoch": 0.9492854307264788,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004918652295402444,
      "loss": 5.2715,
      "mean_token_accuracy": 0.16505309492349624,
      "num_tokens": 20956777.0,
      "step": 9565
    },
    {
      "entropy": 5.552857446670532,
      "epoch": 0.9497816593886463,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004918557857092036,
      "loss": 5.3099,
      "mean_token_accuracy": 0.15762278735637664,
      "num_tokens": 20966057.0,
      "step": 9570
    },
    {
      "entropy": 5.595779657363892,
      "epoch": 0.9502778880508138,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004918463365006065,
      "loss": 5.4026,
      "mean_token_accuracy": 0.1553104653954506,
      "num_tokens": 20976838.0,
      "step": 9575
    },
    {
      "entropy": 5.643113613128662,
      "epoch": 0.9507741167129813,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004918368819146875,
      "loss": 5.3239,
      "mean_token_accuracy": 0.15752324014902114,
      "num_tokens": 20988553.0,
      "step": 9580
    },
    {
      "entropy": 5.56094331741333,
      "epoch": 0.9512703453751489,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004918274219516811,
      "loss": 5.3029,
      "mean_token_accuracy": 0.15705730617046357,
      "num_tokens": 20998983.0,
      "step": 9585
    },
    {
      "entropy": 5.6389594078063965,
      "epoch": 0.9517665740373163,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004918179566118218,
      "loss": 5.4099,
      "mean_token_accuracy": 0.15401346534490584,
      "num_tokens": 21010008.0,
      "step": 9590
    },
    {
      "entropy": 5.597928380966186,
      "epoch": 0.9522628026994839,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004918084858953445,
      "loss": 5.378,
      "mean_token_accuracy": 0.16430667340755462,
      "num_tokens": 21022826.0,
      "step": 9595
    },
    {
      "entropy": 5.593146514892578,
      "epoch": 0.9527590313616514,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004917990098024839,
      "loss": 5.3065,
      "mean_token_accuracy": 0.16070859134197235,
      "num_tokens": 21033457.0,
      "step": 9600
    },
    {
      "entropy": 5.5332756519317625,
      "epoch": 0.953255260023819,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004917895283334749,
      "loss": 5.3259,
      "mean_token_accuracy": 0.15825341045856475,
      "num_tokens": 21044640.0,
      "step": 9605
    },
    {
      "entropy": 5.59820818901062,
      "epoch": 0.9537514886859865,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000491780041488553,
      "loss": 5.3125,
      "mean_token_accuracy": 0.16587773263454436,
      "num_tokens": 21055728.0,
      "step": 9610
    },
    {
      "entropy": 5.612208414077759,
      "epoch": 0.9542477173481541,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004917705492679529,
      "loss": 5.3012,
      "mean_token_accuracy": 0.15372338593006135,
      "num_tokens": 21066057.0,
      "step": 9615
    },
    {
      "entropy": 5.598805236816406,
      "epoch": 0.9547439460103215,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004917610516719104,
      "loss": 5.3228,
      "mean_token_accuracy": 0.15715788900852204,
      "num_tokens": 21075717.0,
      "step": 9620
    },
    {
      "entropy": 5.662686014175415,
      "epoch": 0.9552401746724891,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004917515487006609,
      "loss": 5.4111,
      "mean_token_accuracy": 0.1507570505142212,
      "num_tokens": 21087120.0,
      "step": 9625
    },
    {
      "entropy": 5.621013593673706,
      "epoch": 0.9557364033346566,
      "grad_norm": 0.796875,
      "learning_rate": 0.00049174204035444,
      "loss": 5.4172,
      "mean_token_accuracy": 0.1534568190574646,
      "num_tokens": 21100110.0,
      "step": 9630
    },
    {
      "entropy": 5.670272350311279,
      "epoch": 0.9562326319968242,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004917325266334836,
      "loss": 5.3487,
      "mean_token_accuracy": 0.16352131813764573,
      "num_tokens": 21110411.0,
      "step": 9635
    },
    {
      "entropy": 5.5196967124938965,
      "epoch": 0.9567288606589917,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004917230075380276,
      "loss": 5.2452,
      "mean_token_accuracy": 0.1625675991177559,
      "num_tokens": 21120303.0,
      "step": 9640
    },
    {
      "entropy": 5.565840482711792,
      "epoch": 0.9572250893211592,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004917134830683079,
      "loss": 5.3465,
      "mean_token_accuracy": 0.1565359801054001,
      "num_tokens": 21132312.0,
      "step": 9645
    },
    {
      "entropy": 5.641309928894043,
      "epoch": 0.9577213179833267,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004917039532245608,
      "loss": 5.3128,
      "mean_token_accuracy": 0.16016699820756913,
      "num_tokens": 21142472.0,
      "step": 9650
    },
    {
      "entropy": 5.602419042587281,
      "epoch": 0.9582175466454942,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004916944180070226,
      "loss": 5.3246,
      "mean_token_accuracy": 0.15507960319519043,
      "num_tokens": 21153306.0,
      "step": 9655
    },
    {
      "entropy": 5.651671934127807,
      "epoch": 0.9587137753076618,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004916848774159299,
      "loss": 5.3673,
      "mean_token_accuracy": 0.15087900310754776,
      "num_tokens": 21165606.0,
      "step": 9660
    },
    {
      "entropy": 5.605053329467774,
      "epoch": 0.9592100039698293,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000491675331451519,
      "loss": 5.2977,
      "mean_token_accuracy": 0.1654946967959404,
      "num_tokens": 21176734.0,
      "step": 9665
    },
    {
      "entropy": 5.522690868377685,
      "epoch": 0.9597062326319968,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004916657801140269,
      "loss": 5.2466,
      "mean_token_accuracy": 0.16344843357801436,
      "num_tokens": 21187109.0,
      "step": 9670
    },
    {
      "entropy": 5.559283876419068,
      "epoch": 0.9602024612941643,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004916562234036903,
      "loss": 5.299,
      "mean_token_accuracy": 0.16285582780838012,
      "num_tokens": 21197281.0,
      "step": 9675
    },
    {
      "entropy": 5.571779775619507,
      "epoch": 0.9606986899563319,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004916466613207462,
      "loss": 5.3347,
      "mean_token_accuracy": 0.16179765909910201,
      "num_tokens": 21208767.0,
      "step": 9680
    },
    {
      "entropy": 5.5672523975372314,
      "epoch": 0.9611949186184994,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004916370938654317,
      "loss": 5.3166,
      "mean_token_accuracy": 0.1586903601884842,
      "num_tokens": 21219328.0,
      "step": 9685
    },
    {
      "entropy": 5.642516803741455,
      "epoch": 0.961691147280667,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004916275210379841,
      "loss": 5.4304,
      "mean_token_accuracy": 0.15577488243579865,
      "num_tokens": 21231476.0,
      "step": 9690
    },
    {
      "entropy": 5.667134475708008,
      "epoch": 0.9621873759428344,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004916179428386408,
      "loss": 5.3833,
      "mean_token_accuracy": 0.1607624039053917,
      "num_tokens": 21242445.0,
      "step": 9695
    },
    {
      "entropy": 5.540036630630493,
      "epoch": 0.962683604605002,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004916083592676392,
      "loss": 5.1989,
      "mean_token_accuracy": 0.16129746437072753,
      "num_tokens": 21253179.0,
      "step": 9700
    },
    {
      "entropy": 5.554250860214234,
      "epoch": 0.9631798332671695,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004915987703252171,
      "loss": 5.2254,
      "mean_token_accuracy": 0.1666320875287056,
      "num_tokens": 21263460.0,
      "step": 9705
    },
    {
      "entropy": 5.489849710464478,
      "epoch": 0.9636760619293371,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004915891760116123,
      "loss": 5.2927,
      "mean_token_accuracy": 0.1543031245470047,
      "num_tokens": 21274819.0,
      "step": 9710
    },
    {
      "entropy": 5.586282253265381,
      "epoch": 0.9641722905915046,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004915795763270626,
      "loss": 5.2564,
      "mean_token_accuracy": 0.1650063179433346,
      "num_tokens": 21286215.0,
      "step": 9715
    },
    {
      "entropy": 5.587061882019043,
      "epoch": 0.9646685192536721,
      "grad_norm": 0.9375,
      "learning_rate": 0.000491569971271806,
      "loss": 5.2575,
      "mean_token_accuracy": 0.16380464881658555,
      "num_tokens": 21297033.0,
      "step": 9720
    },
    {
      "entropy": 5.498089981079102,
      "epoch": 0.9651647479158396,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004915603608460809,
      "loss": 5.2582,
      "mean_token_accuracy": 0.15726491957902908,
      "num_tokens": 21308019.0,
      "step": 9725
    },
    {
      "entropy": 5.559009122848511,
      "epoch": 0.9656609765780071,
      "grad_norm": 0.7734375,
      "learning_rate": 0.0004915507450501254,
      "loss": 5.3865,
      "mean_token_accuracy": 0.15224374756217002,
      "num_tokens": 21319313.0,
      "step": 9730
    },
    {
      "entropy": 5.576496982574463,
      "epoch": 0.9661572052401747,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000491541123884178,
      "loss": 5.2896,
      "mean_token_accuracy": 0.1618467316031456,
      "num_tokens": 21329950.0,
      "step": 9735
    },
    {
      "entropy": 5.590175819396973,
      "epoch": 0.9666534339023422,
      "grad_norm": 1.0,
      "learning_rate": 0.0004915314973484775,
      "loss": 5.2652,
      "mean_token_accuracy": 0.16106131225824355,
      "num_tokens": 21339061.0,
      "step": 9740
    },
    {
      "entropy": 5.539566326141357,
      "epoch": 0.9671496625645097,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004915218654432624,
      "loss": 5.3507,
      "mean_token_accuracy": 0.15383373200893402,
      "num_tokens": 21350114.0,
      "step": 9745
    },
    {
      "entropy": 5.566125679016113,
      "epoch": 0.9676458912266772,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004915122281687716,
      "loss": 5.2697,
      "mean_token_accuracy": 0.16104603707790374,
      "num_tokens": 21361219.0,
      "step": 9750
    },
    {
      "entropy": 5.535313653945923,
      "epoch": 0.9681421198888448,
      "grad_norm": 0.875,
      "learning_rate": 0.0004915025855252441,
      "loss": 5.273,
      "mean_token_accuracy": 0.16092607527971267,
      "num_tokens": 21372722.0,
      "step": 9755
    },
    {
      "entropy": 5.600295639038086,
      "epoch": 0.9686383485510123,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004914929375129188,
      "loss": 5.3071,
      "mean_token_accuracy": 0.1598965957760811,
      "num_tokens": 21383350.0,
      "step": 9760
    },
    {
      "entropy": 5.589402151107788,
      "epoch": 0.9691345772131799,
      "grad_norm": 0.7734375,
      "learning_rate": 0.0004914832841320353,
      "loss": 5.3418,
      "mean_token_accuracy": 0.16007469594478607,
      "num_tokens": 21394656.0,
      "step": 9765
    },
    {
      "entropy": 5.592403602600098,
      "epoch": 0.9696308058753473,
      "grad_norm": 0.79296875,
      "learning_rate": 0.000491473625382833,
      "loss": 5.3327,
      "mean_token_accuracy": 0.15597164034843444,
      "num_tokens": 21406163.0,
      "step": 9770
    },
    {
      "entropy": 5.495397901535034,
      "epoch": 0.9701270345375149,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000491463961265551,
      "loss": 5.228,
      "mean_token_accuracy": 0.16503653675317764,
      "num_tokens": 21418703.0,
      "step": 9775
    },
    {
      "entropy": 5.528452920913696,
      "epoch": 0.9706232631996824,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004914542917804292,
      "loss": 5.2247,
      "mean_token_accuracy": 0.1663466528058052,
      "num_tokens": 21428511.0,
      "step": 9780
    },
    {
      "entropy": 5.586885643005371,
      "epoch": 0.97111949186185,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004914446169277075,
      "loss": 5.2642,
      "mean_token_accuracy": 0.16102582216262817,
      "num_tokens": 21440350.0,
      "step": 9785
    },
    {
      "entropy": 5.656343078613281,
      "epoch": 0.9716157205240175,
      "grad_norm": 1.0,
      "learning_rate": 0.0004914349367076258,
      "loss": 5.378,
      "mean_token_accuracy": 0.1604055270552635,
      "num_tokens": 21451262.0,
      "step": 9790
    },
    {
      "entropy": 5.59821195602417,
      "epoch": 0.972111949186185,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004914252511204238,
      "loss": 5.3269,
      "mean_token_accuracy": 0.16023993492126465,
      "num_tokens": 21462587.0,
      "step": 9795
    },
    {
      "entropy": 5.590718078613281,
      "epoch": 0.9726081778483525,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004914155601663422,
      "loss": 5.2921,
      "mean_token_accuracy": 0.16131409704685212,
      "num_tokens": 21473452.0,
      "step": 9800
    },
    {
      "entropy": 5.5845705509185795,
      "epoch": 0.97310440651052,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004914058638456208,
      "loss": 5.2577,
      "mean_token_accuracy": 0.16089966893196106,
      "num_tokens": 21484011.0,
      "step": 9805
    },
    {
      "entropy": 5.459889936447143,
      "epoch": 0.9736006351726876,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004913961621585003,
      "loss": 5.2445,
      "mean_token_accuracy": 0.1632186383008957,
      "num_tokens": 21495959.0,
      "step": 9810
    },
    {
      "entropy": 5.507400798797607,
      "epoch": 0.974096863834855,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004913864551052214,
      "loss": 5.2857,
      "mean_token_accuracy": 0.16949345022439957,
      "num_tokens": 21507496.0,
      "step": 9815
    },
    {
      "entropy": 5.653198719024658,
      "epoch": 0.9745930924970226,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004913767426860246,
      "loss": 5.3691,
      "mean_token_accuracy": 0.15034950897097588,
      "num_tokens": 21517532.0,
      "step": 9820
    },
    {
      "entropy": 5.555367803573608,
      "epoch": 0.9750893211591901,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004913670249011508,
      "loss": 5.1951,
      "mean_token_accuracy": 0.163184355199337,
      "num_tokens": 21527969.0,
      "step": 9825
    },
    {
      "entropy": 5.525680065155029,
      "epoch": 0.9755855498213577,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004913573017508411,
      "loss": 5.309,
      "mean_token_accuracy": 0.15447816699743272,
      "num_tokens": 21538243.0,
      "step": 9830
    },
    {
      "entropy": 5.586614418029785,
      "epoch": 0.9760817784835252,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004913475732353365,
      "loss": 5.3026,
      "mean_token_accuracy": 0.1621467411518097,
      "num_tokens": 21548702.0,
      "step": 9835
    },
    {
      "entropy": 5.5326902866363525,
      "epoch": 0.9765780071456928,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004913378393548784,
      "loss": 5.2621,
      "mean_token_accuracy": 0.16488298326730727,
      "num_tokens": 21558583.0,
      "step": 9840
    },
    {
      "entropy": 5.57383246421814,
      "epoch": 0.9770742358078602,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004913281001097078,
      "loss": 5.2566,
      "mean_token_accuracy": 0.16477205902338027,
      "num_tokens": 21569845.0,
      "step": 9845
    },
    {
      "entropy": 5.626986360549926,
      "epoch": 0.9775704644700278,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004913183555000667,
      "loss": 5.4093,
      "mean_token_accuracy": 0.1548849605023861,
      "num_tokens": 21580540.0,
      "step": 9850
    },
    {
      "entropy": 5.584823036193848,
      "epoch": 0.9780666931321953,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004913086055261963,
      "loss": 5.2608,
      "mean_token_accuracy": 0.1617264226078987,
      "num_tokens": 21590982.0,
      "step": 9855
    },
    {
      "entropy": 5.542820787429809,
      "epoch": 0.9785629217943629,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004912988501883387,
      "loss": 5.2851,
      "mean_token_accuracy": 0.16189356297254562,
      "num_tokens": 21602817.0,
      "step": 9860
    },
    {
      "entropy": 5.594162940979004,
      "epoch": 0.9790591504565304,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004912890894867357,
      "loss": 5.3345,
      "mean_token_accuracy": 0.15913430005311965,
      "num_tokens": 21613422.0,
      "step": 9865
    },
    {
      "entropy": 5.548657035827636,
      "epoch": 0.9795553791186979,
      "grad_norm": 1.0,
      "learning_rate": 0.0004912793234216293,
      "loss": 5.314,
      "mean_token_accuracy": 0.15511717051267623,
      "num_tokens": 21624241.0,
      "step": 9870
    },
    {
      "entropy": 5.530315399169922,
      "epoch": 0.9800516077808654,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004912695519932618,
      "loss": 5.3152,
      "mean_token_accuracy": 0.16225161999464036,
      "num_tokens": 21636243.0,
      "step": 9875
    },
    {
      "entropy": 5.573291540145874,
      "epoch": 0.9805478364430329,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004912597752018753,
      "loss": 5.3316,
      "mean_token_accuracy": 0.1610553279519081,
      "num_tokens": 21647918.0,
      "step": 9880
    },
    {
      "entropy": 5.61486291885376,
      "epoch": 0.9810440651052005,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004912499930477126,
      "loss": 5.3078,
      "mean_token_accuracy": 0.15688589364290237,
      "num_tokens": 21659236.0,
      "step": 9885
    },
    {
      "entropy": 5.566563987731934,
      "epoch": 0.981540293767368,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004912402055310158,
      "loss": 5.3055,
      "mean_token_accuracy": 0.16469915062189103,
      "num_tokens": 21669647.0,
      "step": 9890
    },
    {
      "entropy": 5.535579442977905,
      "epoch": 0.9820365224295355,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004912304126520279,
      "loss": 5.3494,
      "mean_token_accuracy": 0.15597899705171586,
      "num_tokens": 21680369.0,
      "step": 9895
    },
    {
      "entropy": 5.631029987335205,
      "epoch": 0.982532751091703,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004912206144109917,
      "loss": 5.3534,
      "mean_token_accuracy": 0.15765170902013778,
      "num_tokens": 21690942.0,
      "step": 9900
    },
    {
      "entropy": 5.558161735534668,
      "epoch": 0.9830289797538706,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004912108108081503,
      "loss": 5.2878,
      "mean_token_accuracy": 0.16257334798574447,
      "num_tokens": 21701945.0,
      "step": 9905
    },
    {
      "entropy": 5.6102015495300295,
      "epoch": 0.9835252084160381,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004912010018437465,
      "loss": 5.3193,
      "mean_token_accuracy": 0.15981239974498748,
      "num_tokens": 21712246.0,
      "step": 9910
    },
    {
      "entropy": 5.678878927230835,
      "epoch": 0.9840214370782057,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000491191187518024,
      "loss": 5.317,
      "mean_token_accuracy": 0.16021854281425477,
      "num_tokens": 21721660.0,
      "step": 9915
    },
    {
      "entropy": 5.551132726669311,
      "epoch": 0.9845176657403731,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004911813678312257,
      "loss": 5.2036,
      "mean_token_accuracy": 0.16454039365053177,
      "num_tokens": 21732157.0,
      "step": 9920
    },
    {
      "entropy": 5.543543910980224,
      "epoch": 0.9850138944025407,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004911715427835953,
      "loss": 5.3336,
      "mean_token_accuracy": 0.15899271741509438,
      "num_tokens": 21741727.0,
      "step": 9925
    },
    {
      "entropy": 5.574287223815918,
      "epoch": 0.9855101230647082,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004911617123753765,
      "loss": 5.27,
      "mean_token_accuracy": 0.16396869122982025,
      "num_tokens": 21752596.0,
      "step": 9930
    },
    {
      "entropy": 5.515794897079468,
      "epoch": 0.9860063517268758,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004911518766068131,
      "loss": 5.1856,
      "mean_token_accuracy": 0.16887888461351394,
      "num_tokens": 21763978.0,
      "step": 9935
    },
    {
      "entropy": 5.537166929244995,
      "epoch": 0.9865025803890433,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004911420354781487,
      "loss": 5.3282,
      "mean_token_accuracy": 0.15229359567165374,
      "num_tokens": 21774961.0,
      "step": 9940
    },
    {
      "entropy": 5.562363862991333,
      "epoch": 0.9869988090512108,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004911321889896278,
      "loss": 5.2447,
      "mean_token_accuracy": 0.166769939661026,
      "num_tokens": 21785387.0,
      "step": 9945
    },
    {
      "entropy": 5.573436832427978,
      "epoch": 0.9874950377133783,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004911223371414943,
      "loss": 5.294,
      "mean_token_accuracy": 0.1654672920703888,
      "num_tokens": 21795982.0,
      "step": 9950
    },
    {
      "entropy": 5.5942527770996096,
      "epoch": 0.9879912663755459,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004911124799339924,
      "loss": 5.3208,
      "mean_token_accuracy": 0.15434270054101945,
      "num_tokens": 21806768.0,
      "step": 9955
    },
    {
      "entropy": 5.58071460723877,
      "epoch": 0.9884874950377134,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004911026173673667,
      "loss": 5.2859,
      "mean_token_accuracy": 0.16355279982089996,
      "num_tokens": 21818563.0,
      "step": 9960
    },
    {
      "entropy": 5.596285247802735,
      "epoch": 0.9889837236998809,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004910927494418619,
      "loss": 5.3664,
      "mean_token_accuracy": 0.1590631991624832,
      "num_tokens": 21830253.0,
      "step": 9965
    },
    {
      "entropy": 5.6451211929321286,
      "epoch": 0.9894799523620484,
      "grad_norm": 0.7578125,
      "learning_rate": 0.0004910828761577224,
      "loss": 5.3322,
      "mean_token_accuracy": 0.16312540024518968,
      "num_tokens": 21842353.0,
      "step": 9970
    },
    {
      "entropy": 5.575357866287232,
      "epoch": 0.9899761810242159,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004910729975151933,
      "loss": 5.2749,
      "mean_token_accuracy": 0.16515064537525176,
      "num_tokens": 21852066.0,
      "step": 9975
    },
    {
      "entropy": 5.517587375640869,
      "epoch": 0.9904724096863835,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004910631135145194,
      "loss": 5.311,
      "mean_token_accuracy": 0.1654649168252945,
      "num_tokens": 21862754.0,
      "step": 9980
    },
    {
      "entropy": 5.467855453491211,
      "epoch": 0.990968638348551,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004910532241559458,
      "loss": 5.2208,
      "mean_token_accuracy": 0.17459090799093246,
      "num_tokens": 21873531.0,
      "step": 9985
    },
    {
      "entropy": 5.472586536407471,
      "epoch": 0.9914648670107186,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004910433294397179,
      "loss": 5.1661,
      "mean_token_accuracy": 0.16823934763669968,
      "num_tokens": 21884339.0,
      "step": 9990
    },
    {
      "entropy": 5.521728467941284,
      "epoch": 0.991961095672886,
      "grad_norm": 0.890625,
      "learning_rate": 0.000491033429366081,
      "loss": 5.2675,
      "mean_token_accuracy": 0.15773390531539916,
      "num_tokens": 21895489.0,
      "step": 9995
    },
    {
      "entropy": 5.539831876754761,
      "epoch": 0.9924573243350536,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004910235239352804,
      "loss": 5.2592,
      "mean_token_accuracy": 0.16699083596467973,
      "num_tokens": 21906832.0,
      "step": 10000
    },
    {
      "entropy": 5.558729648590088,
      "epoch": 0.9929535529972211,
      "grad_norm": 0.859375,
      "learning_rate": 0.000491013613147562,
      "loss": 5.2973,
      "mean_token_accuracy": 0.159033165872097,
      "num_tokens": 21916979.0,
      "step": 10005
    },
    {
      "entropy": 5.547304725646972,
      "epoch": 0.9934497816593887,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004910036970031715,
      "loss": 5.2002,
      "mean_token_accuracy": 0.16779321283102036,
      "num_tokens": 21927352.0,
      "step": 10010
    },
    {
      "entropy": 5.543372678756714,
      "epoch": 0.9939460103215562,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004909937755023547,
      "loss": 5.3249,
      "mean_token_accuracy": 0.15822510719299315,
      "num_tokens": 21937765.0,
      "step": 10015
    },
    {
      "entropy": 5.495730400085449,
      "epoch": 0.9944422389837237,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004909838486453577,
      "loss": 5.2208,
      "mean_token_accuracy": 0.17041514962911605,
      "num_tokens": 21948659.0,
      "step": 10020
    },
    {
      "entropy": 5.5643548488616945,
      "epoch": 0.9949384676458912,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004909739164324267,
      "loss": 5.3328,
      "mean_token_accuracy": 0.1573019579052925,
      "num_tokens": 21959356.0,
      "step": 10025
    },
    {
      "entropy": 5.721648168563843,
      "epoch": 0.9954346963080588,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004909639788638079,
      "loss": 5.4387,
      "mean_token_accuracy": 0.14892093241214752,
      "num_tokens": 21971321.0,
      "step": 10030
    },
    {
      "entropy": 5.552315521240234,
      "epoch": 0.9959309249702263,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004909540359397477,
      "loss": 5.3001,
      "mean_token_accuracy": 0.15871900767087938,
      "num_tokens": 21981252.0,
      "step": 10035
    },
    {
      "entropy": 5.56937460899353,
      "epoch": 0.9964271536323938,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004909440876604929,
      "loss": 5.3107,
      "mean_token_accuracy": 0.15274860709905624,
      "num_tokens": 21991820.0,
      "step": 10040
    },
    {
      "entropy": 5.545371580123901,
      "epoch": 0.9969233822945613,
      "grad_norm": 0.8359375,
      "learning_rate": 0.00049093413402629,
      "loss": 5.2753,
      "mean_token_accuracy": 0.16395441740751265,
      "num_tokens": 22003619.0,
      "step": 10045
    },
    {
      "entropy": 5.557806158065796,
      "epoch": 0.9974196109567288,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004909241750373858,
      "loss": 5.3071,
      "mean_token_accuracy": 0.15856507569551467,
      "num_tokens": 22016116.0,
      "step": 10050
    },
    {
      "entropy": 5.6003748893737795,
      "epoch": 0.9979158396188964,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004909142106940275,
      "loss": 5.2478,
      "mean_token_accuracy": 0.16244348660111427,
      "num_tokens": 22026115.0,
      "step": 10055
    },
    {
      "entropy": 5.572235727310181,
      "epoch": 0.9984120682810639,
      "grad_norm": 0.890625,
      "learning_rate": 0.000490904240996462,
      "loss": 5.3147,
      "mean_token_accuracy": 0.16750263571739196,
      "num_tokens": 22038889.0,
      "step": 10060
    },
    {
      "entropy": 5.5575329780578615,
      "epoch": 0.9989082969432315,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004908942659449364,
      "loss": 5.2853,
      "mean_token_accuracy": 0.16204715073108672,
      "num_tokens": 22050773.0,
      "step": 10065
    },
    {
      "entropy": 5.546312522888184,
      "epoch": 0.9994045256053989,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004908842855396982,
      "loss": 5.2832,
      "mean_token_accuracy": 0.1552201345562935,
      "num_tokens": 22062869.0,
      "step": 10070
    },
    {
      "entropy": 5.548929119110108,
      "epoch": 0.9999007542675665,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004908742997809951,
      "loss": 5.3017,
      "mean_token_accuracy": 0.1592031702399254,
      "num_tokens": 22073898.0,
      "step": 10075
    },
    {
      "entropy": 5.624472570419312,
      "epoch": 1.0003969829297341,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004908643086690744,
      "loss": 5.2118,
      "mean_token_accuracy": 0.15984300523996353,
      "num_tokens": 22084383.0,
      "step": 10080
    },
    {
      "entropy": 5.565525245666504,
      "epoch": 1.0008932115919016,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000490854312204184,
      "loss": 5.2519,
      "mean_token_accuracy": 0.1635912373661995,
      "num_tokens": 22094990.0,
      "step": 10085
    },
    {
      "entropy": 5.548325061798096,
      "epoch": 1.001389440254069,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004908443103865719,
      "loss": 5.2839,
      "mean_token_accuracy": 0.1550723358988762,
      "num_tokens": 22105259.0,
      "step": 10090
    },
    {
      "entropy": 5.555172777175903,
      "epoch": 1.0018856689162365,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004908343032164859,
      "loss": 5.2474,
      "mean_token_accuracy": 0.1596066027879715,
      "num_tokens": 22116348.0,
      "step": 10095
    },
    {
      "entropy": 5.595499801635742,
      "epoch": 1.0023818975784042,
      "grad_norm": 1.0,
      "learning_rate": 0.0004908242906941743,
      "loss": 5.2351,
      "mean_token_accuracy": 0.16683860272169113,
      "num_tokens": 22127311.0,
      "step": 10100
    },
    {
      "entropy": 5.601737976074219,
      "epoch": 1.0028781262405717,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004908142728198853,
      "loss": 5.2655,
      "mean_token_accuracy": 0.15398730337619781,
      "num_tokens": 22137929.0,
      "step": 10105
    },
    {
      "entropy": 5.657776546478272,
      "epoch": 1.0033743549027392,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004908042495938673,
      "loss": 5.3697,
      "mean_token_accuracy": 0.15462761223316193,
      "num_tokens": 22148202.0,
      "step": 10110
    },
    {
      "entropy": 5.566437387466431,
      "epoch": 1.0038705835649067,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004907942210163692,
      "loss": 5.1538,
      "mean_token_accuracy": 0.16509081423282623,
      "num_tokens": 22158955.0,
      "step": 10115
    },
    {
      "entropy": 5.543176460266113,
      "epoch": 1.0043668122270741,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004907841870876392,
      "loss": 5.2014,
      "mean_token_accuracy": 0.16200576722621918,
      "num_tokens": 22169569.0,
      "step": 10120
    },
    {
      "entropy": 5.5353536128997805,
      "epoch": 1.0048630408892418,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004907741478079263,
      "loss": 5.1761,
      "mean_token_accuracy": 0.15873774886131287,
      "num_tokens": 22180171.0,
      "step": 10125
    },
    {
      "entropy": 5.559956693649292,
      "epoch": 1.0053592695514093,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004907641031774796,
      "loss": 5.2753,
      "mean_token_accuracy": 0.16398740112781524,
      "num_tokens": 22190531.0,
      "step": 10130
    },
    {
      "entropy": 5.545996809005738,
      "epoch": 1.0058554982135768,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000490754053196548,
      "loss": 5.1803,
      "mean_token_accuracy": 0.16567456275224685,
      "num_tokens": 22201495.0,
      "step": 10135
    },
    {
      "entropy": 5.619088268280029,
      "epoch": 1.0063517268757443,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004907439978653809,
      "loss": 5.2618,
      "mean_token_accuracy": 0.15639659762382507,
      "num_tokens": 22212941.0,
      "step": 10140
    },
    {
      "entropy": 5.636896085739136,
      "epoch": 1.006847955537912,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004907339371842275,
      "loss": 5.2993,
      "mean_token_accuracy": 0.1600252002477646,
      "num_tokens": 22225212.0,
      "step": 10145
    },
    {
      "entropy": 5.5214286804199215,
      "epoch": 1.0073441842000794,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004907238711533373,
      "loss": 5.2541,
      "mean_token_accuracy": 0.1617978185415268,
      "num_tokens": 22236370.0,
      "step": 10150
    },
    {
      "entropy": 5.557128143310547,
      "epoch": 1.007840412862247,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004907137997729599,
      "loss": 5.213,
      "mean_token_accuracy": 0.1678090587258339,
      "num_tokens": 22247641.0,
      "step": 10155
    },
    {
      "entropy": 5.515650749206543,
      "epoch": 1.0083366415244144,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000490703723043345,
      "loss": 5.2724,
      "mean_token_accuracy": 0.16332590878009795,
      "num_tokens": 22258599.0,
      "step": 10160
    },
    {
      "entropy": 5.535066986083985,
      "epoch": 1.008832870186582,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004906936409647428,
      "loss": 5.1949,
      "mean_token_accuracy": 0.1628037601709366,
      "num_tokens": 22270913.0,
      "step": 10165
    },
    {
      "entropy": 5.539957666397095,
      "epoch": 1.0093290988487496,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000490683553537403,
      "loss": 5.1037,
      "mean_token_accuracy": 0.16978239566087722,
      "num_tokens": 22281053.0,
      "step": 10170
    },
    {
      "entropy": 5.526017761230468,
      "epoch": 1.009825327510917,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004906734607615758,
      "loss": 5.1269,
      "mean_token_accuracy": 0.1676328420639038,
      "num_tokens": 22292570.0,
      "step": 10175
    },
    {
      "entropy": 5.509777212142945,
      "epoch": 1.0103215561730845,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004906633626375114,
      "loss": 5.2055,
      "mean_token_accuracy": 0.16312919110059737,
      "num_tokens": 22303548.0,
      "step": 10180
    },
    {
      "entropy": 5.563418865203857,
      "epoch": 1.010817784835252,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004906532591654604,
      "loss": 5.2281,
      "mean_token_accuracy": 0.16960761696100235,
      "num_tokens": 22314133.0,
      "step": 10185
    },
    {
      "entropy": 5.581880044937134,
      "epoch": 1.0113140134974197,
      "grad_norm": 0.76953125,
      "learning_rate": 0.0004906431503456733,
      "loss": 5.2008,
      "mean_token_accuracy": 0.1627922311425209,
      "num_tokens": 22326431.0,
      "step": 10190
    },
    {
      "entropy": 5.540881824493408,
      "epoch": 1.0118102421595871,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004906330361784007,
      "loss": 5.239,
      "mean_token_accuracy": 0.16488454043865203,
      "num_tokens": 22338507.0,
      "step": 10195
    },
    {
      "entropy": 5.517429780960083,
      "epoch": 1.0123064708217546,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004906229166638936,
      "loss": 5.2111,
      "mean_token_accuracy": 0.16140173077583314,
      "num_tokens": 22348863.0,
      "step": 10200
    },
    {
      "entropy": 5.484919691085816,
      "epoch": 1.012802699483922,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004906127918024026,
      "loss": 5.1588,
      "mean_token_accuracy": 0.16300449669361114,
      "num_tokens": 22359000.0,
      "step": 10205
    },
    {
      "entropy": 5.5954266548156735,
      "epoch": 1.0132989281460898,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004906026615941791,
      "loss": 5.2945,
      "mean_token_accuracy": 0.157001593708992,
      "num_tokens": 22369539.0,
      "step": 10210
    },
    {
      "entropy": 5.578581285476685,
      "epoch": 1.0137951568082573,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004905925260394741,
      "loss": 5.2615,
      "mean_token_accuracy": 0.16177936643362045,
      "num_tokens": 22380700.0,
      "step": 10215
    },
    {
      "entropy": 5.537117147445679,
      "epoch": 1.0142913854704247,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000490582385138539,
      "loss": 5.2368,
      "mean_token_accuracy": 0.1548960104584694,
      "num_tokens": 22392151.0,
      "step": 10220
    },
    {
      "entropy": 5.526557683944702,
      "epoch": 1.0147876141325922,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004905722388916254,
      "loss": 5.212,
      "mean_token_accuracy": 0.166645385324955,
      "num_tokens": 22401735.0,
      "step": 10225
    },
    {
      "entropy": 5.570654678344726,
      "epoch": 1.01528384279476,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004905620872989846,
      "loss": 5.2798,
      "mean_token_accuracy": 0.1610209047794342,
      "num_tokens": 22412549.0,
      "step": 10230
    },
    {
      "entropy": 5.614617395401001,
      "epoch": 1.0157800714569274,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004905519303608687,
      "loss": 5.2633,
      "mean_token_accuracy": 0.16019045561552048,
      "num_tokens": 22424728.0,
      "step": 10235
    },
    {
      "entropy": 5.491592454910278,
      "epoch": 1.0162763001190949,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004905417680775294,
      "loss": 5.1436,
      "mean_token_accuracy": 0.16299115121364594,
      "num_tokens": 22434197.0,
      "step": 10240
    },
    {
      "entropy": 5.610570001602173,
      "epoch": 1.0167725287812623,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004905316004492186,
      "loss": 5.3314,
      "mean_token_accuracy": 0.1562383472919464,
      "num_tokens": 22444671.0,
      "step": 10245
    },
    {
      "entropy": 5.529538679122925,
      "epoch": 1.01726875744343,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004905214274761885,
      "loss": 5.1939,
      "mean_token_accuracy": 0.16395601779222488,
      "num_tokens": 22456636.0,
      "step": 10250
    },
    {
      "entropy": 5.4973588466644285,
      "epoch": 1.0177649861055975,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004905112491586915,
      "loss": 5.1375,
      "mean_token_accuracy": 0.170780511200428,
      "num_tokens": 22467011.0,
      "step": 10255
    },
    {
      "entropy": 5.523624897003174,
      "epoch": 1.018261214767765,
      "grad_norm": 0.96875,
      "learning_rate": 0.00049050106549698,
      "loss": 5.2321,
      "mean_token_accuracy": 0.16345654129981996,
      "num_tokens": 22477630.0,
      "step": 10260
    },
    {
      "entropy": 5.512229251861572,
      "epoch": 1.0187574434299325,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004904908764913063,
      "loss": 5.1867,
      "mean_token_accuracy": 0.16930550932884217,
      "num_tokens": 22488132.0,
      "step": 10265
    },
    {
      "entropy": 5.477755403518676,
      "epoch": 1.0192536720921,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004904806821419233,
      "loss": 5.2339,
      "mean_token_accuracy": 0.15899064689874648,
      "num_tokens": 22499430.0,
      "step": 10270
    },
    {
      "entropy": 5.541066360473633,
      "epoch": 1.0197499007542676,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004904704824490835,
      "loss": 5.1672,
      "mean_token_accuracy": 0.1635948821902275,
      "num_tokens": 22509728.0,
      "step": 10275
    },
    {
      "entropy": 5.577764987945557,
      "epoch": 1.020246129416435,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004904602774130402,
      "loss": 5.2761,
      "mean_token_accuracy": 0.15536259561777116,
      "num_tokens": 22520955.0,
      "step": 10280
    },
    {
      "entropy": 5.629702186584472,
      "epoch": 1.0207423580786026,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004904500670340463,
      "loss": 5.2964,
      "mean_token_accuracy": 0.15821094959974288,
      "num_tokens": 22530440.0,
      "step": 10285
    },
    {
      "entropy": 5.540016794204712,
      "epoch": 1.02123858674077,
      "grad_norm": 0.875,
      "learning_rate": 0.000490439851312355,
      "loss": 5.2475,
      "mean_token_accuracy": 0.171015864610672,
      "num_tokens": 22541539.0,
      "step": 10290
    },
    {
      "entropy": 5.544416809082032,
      "epoch": 1.0217348154029378,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004904296302482196,
      "loss": 5.2322,
      "mean_token_accuracy": 0.16253914684057236,
      "num_tokens": 22555717.0,
      "step": 10295
    },
    {
      "entropy": 5.613153886795044,
      "epoch": 1.0222310440651052,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004904194038418935,
      "loss": 5.3377,
      "mean_token_accuracy": 0.1595346972346306,
      "num_tokens": 22566938.0,
      "step": 10300
    },
    {
      "entropy": 5.665261459350586,
      "epoch": 1.0227272727272727,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004904091720936305,
      "loss": 5.2044,
      "mean_token_accuracy": 0.1597888469696045,
      "num_tokens": 22578725.0,
      "step": 10305
    },
    {
      "entropy": 5.555477476119995,
      "epoch": 1.0232235013894402,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004903989350036841,
      "loss": 5.133,
      "mean_token_accuracy": 0.17128655016422273,
      "num_tokens": 22588476.0,
      "step": 10310
    },
    {
      "entropy": 5.5689750671386715,
      "epoch": 1.0237197300516079,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004903886925723083,
      "loss": 5.3335,
      "mean_token_accuracy": 0.1574925050139427,
      "num_tokens": 22600294.0,
      "step": 10315
    },
    {
      "entropy": 5.60640025138855,
      "epoch": 1.0242159587137754,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000490378444799757,
      "loss": 5.2219,
      "mean_token_accuracy": 0.16590719819068908,
      "num_tokens": 22610578.0,
      "step": 10320
    },
    {
      "entropy": 5.616174936294556,
      "epoch": 1.0247121873759428,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004903681916862844,
      "loss": 5.2451,
      "mean_token_accuracy": 0.16193822175264358,
      "num_tokens": 22620988.0,
      "step": 10325
    },
    {
      "entropy": 5.537631797790527,
      "epoch": 1.0252084160381103,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004903579332321447,
      "loss": 5.2202,
      "mean_token_accuracy": 0.16445315182209014,
      "num_tokens": 22632198.0,
      "step": 10330
    },
    {
      "entropy": 5.5386731147766115,
      "epoch": 1.0257046447002778,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004903476694375924,
      "loss": 5.2245,
      "mean_token_accuracy": 0.17040131539106368,
      "num_tokens": 22642543.0,
      "step": 10335
    },
    {
      "entropy": 5.565079832077027,
      "epoch": 1.0262008733624455,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004903374003028819,
      "loss": 5.2179,
      "mean_token_accuracy": 0.16459809094667435,
      "num_tokens": 22655453.0,
      "step": 10340
    },
    {
      "entropy": 5.510003232955933,
      "epoch": 1.026697102024613,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004903271258282678,
      "loss": 5.1604,
      "mean_token_accuracy": 0.1727517679333687,
      "num_tokens": 22667342.0,
      "step": 10345
    },
    {
      "entropy": 5.616212701797485,
      "epoch": 1.0271933306867804,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004903168460140049,
      "loss": 5.2466,
      "mean_token_accuracy": 0.16790675669908522,
      "num_tokens": 22676890.0,
      "step": 10350
    },
    {
      "entropy": 5.553227472305298,
      "epoch": 1.027689559348948,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004903065608603484,
      "loss": 5.239,
      "mean_token_accuracy": 0.16413081735372542,
      "num_tokens": 22687809.0,
      "step": 10355
    },
    {
      "entropy": 5.509211540222168,
      "epoch": 1.0281857880111156,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000490296270367553,
      "loss": 5.2487,
      "mean_token_accuracy": 0.16225795596837997,
      "num_tokens": 22697938.0,
      "step": 10360
    },
    {
      "entropy": 5.625287103652954,
      "epoch": 1.028682016673283,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004902859745358741,
      "loss": 5.2398,
      "mean_token_accuracy": 0.16604799926280975,
      "num_tokens": 22708303.0,
      "step": 10365
    },
    {
      "entropy": 5.5169538974761965,
      "epoch": 1.0291782453354505,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004902756733655668,
      "loss": 5.1732,
      "mean_token_accuracy": 0.16808167099952698,
      "num_tokens": 22717987.0,
      "step": 10370
    },
    {
      "entropy": 5.515469884872436,
      "epoch": 1.029674473997618,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004902653668568868,
      "loss": 5.25,
      "mean_token_accuracy": 0.16545470654964448,
      "num_tokens": 22729672.0,
      "step": 10375
    },
    {
      "entropy": 5.574868202209473,
      "epoch": 1.0301707026597857,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004902550550100896,
      "loss": 5.2082,
      "mean_token_accuracy": 0.17231257408857345,
      "num_tokens": 22741296.0,
      "step": 10380
    },
    {
      "entropy": 5.56593770980835,
      "epoch": 1.0306669313219532,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004902447378254307,
      "loss": 5.2163,
      "mean_token_accuracy": 0.16785264909267425,
      "num_tokens": 22752337.0,
      "step": 10385
    },
    {
      "entropy": 5.578250932693481,
      "epoch": 1.0311631599841207,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004902344153031663,
      "loss": 5.2412,
      "mean_token_accuracy": 0.16290030479431153,
      "num_tokens": 22762681.0,
      "step": 10390
    },
    {
      "entropy": 5.6123497009277346,
      "epoch": 1.0316593886462881,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000490224087443552,
      "loss": 5.283,
      "mean_token_accuracy": 0.15981359630823136,
      "num_tokens": 22774275.0,
      "step": 10395
    },
    {
      "entropy": 5.511215019226074,
      "epoch": 1.0321556173084558,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004902137542468442,
      "loss": 5.1508,
      "mean_token_accuracy": 0.1674342632293701,
      "num_tokens": 22785944.0,
      "step": 10400
    },
    {
      "entropy": 5.499436521530152,
      "epoch": 1.0326518459706233,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004902034157132991,
      "loss": 5.0925,
      "mean_token_accuracy": 0.16678470075130464,
      "num_tokens": 22796543.0,
      "step": 10405
    },
    {
      "entropy": 5.503799533843994,
      "epoch": 1.0331480746327908,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004901930718431729,
      "loss": 5.2084,
      "mean_token_accuracy": 0.1636475369334221,
      "num_tokens": 22807957.0,
      "step": 10410
    },
    {
      "entropy": 5.589832401275634,
      "epoch": 1.0336443032949583,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004901827226367223,
      "loss": 5.3044,
      "mean_token_accuracy": 0.16381883323192598,
      "num_tokens": 22818903.0,
      "step": 10415
    },
    {
      "entropy": 5.607730150222778,
      "epoch": 1.0341405319571257,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004901723680942038,
      "loss": 5.3367,
      "mean_token_accuracy": 0.16103588938713073,
      "num_tokens": 22829286.0,
      "step": 10420
    },
    {
      "entropy": 5.597163105010987,
      "epoch": 1.0346367606192934,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004901620082158742,
      "loss": 5.2421,
      "mean_token_accuracy": 0.1619735226035118,
      "num_tokens": 22839860.0,
      "step": 10425
    },
    {
      "entropy": 5.607494163513183,
      "epoch": 1.035132989281461,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004901516430019905,
      "loss": 5.2862,
      "mean_token_accuracy": 0.15376426428556442,
      "num_tokens": 22852238.0,
      "step": 10430
    },
    {
      "entropy": 5.581726598739624,
      "epoch": 1.0356292179436284,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004901412724528097,
      "loss": 5.2666,
      "mean_token_accuracy": 0.1640608638525009,
      "num_tokens": 22863013.0,
      "step": 10435
    },
    {
      "entropy": 5.523475646972656,
      "epoch": 1.0361254466057959,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004901308965685889,
      "loss": 5.1225,
      "mean_token_accuracy": 0.16678005307912827,
      "num_tokens": 22874233.0,
      "step": 10440
    },
    {
      "entropy": 5.572403764724731,
      "epoch": 1.0366216752679636,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004901205153495853,
      "loss": 5.2973,
      "mean_token_accuracy": 0.15935328900814055,
      "num_tokens": 22885313.0,
      "step": 10445
    },
    {
      "entropy": 5.534413146972656,
      "epoch": 1.037117903930131,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004901101287960566,
      "loss": 5.1848,
      "mean_token_accuracy": 0.16433444768190383,
      "num_tokens": 22895365.0,
      "step": 10450
    },
    {
      "entropy": 5.500718212127685,
      "epoch": 1.0376141325922985,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004900997369082602,
      "loss": 5.1861,
      "mean_token_accuracy": 0.15906585305929183,
      "num_tokens": 22906392.0,
      "step": 10455
    },
    {
      "entropy": 5.48196005821228,
      "epoch": 1.038110361254466,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004900893396864539,
      "loss": 5.1424,
      "mean_token_accuracy": 0.16976341009140014,
      "num_tokens": 22916760.0,
      "step": 10460
    },
    {
      "entropy": 5.570258903503418,
      "epoch": 1.0386065899166337,
      "grad_norm": 0.875,
      "learning_rate": 0.0004900789371308953,
      "loss": 5.2058,
      "mean_token_accuracy": 0.16422784626483916,
      "num_tokens": 22928070.0,
      "step": 10465
    },
    {
      "entropy": 5.470105075836182,
      "epoch": 1.0391028185788012,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004900685292418425,
      "loss": 5.1797,
      "mean_token_accuracy": 0.1650763064622879,
      "num_tokens": 22938815.0,
      "step": 10470
    },
    {
      "entropy": 5.53947172164917,
      "epoch": 1.0395990472409686,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004900581160195537,
      "loss": 5.2512,
      "mean_token_accuracy": 0.15671544075012206,
      "num_tokens": 22949806.0,
      "step": 10475
    },
    {
      "entropy": 5.5062002658844,
      "epoch": 1.040095275903136,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000490047697464287,
      "loss": 5.0866,
      "mean_token_accuracy": 0.16767628192901612,
      "num_tokens": 22960880.0,
      "step": 10480
    },
    {
      "entropy": 5.567434787750244,
      "epoch": 1.0405915045653038,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004900372735763009,
      "loss": 5.2358,
      "mean_token_accuracy": 0.16065094918012618,
      "num_tokens": 22972828.0,
      "step": 10485
    },
    {
      "entropy": 5.480893564224243,
      "epoch": 1.0410877332274713,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004900268443558536,
      "loss": 5.1356,
      "mean_token_accuracy": 0.1689052939414978,
      "num_tokens": 22981869.0,
      "step": 10490
    },
    {
      "entropy": 5.588772296905518,
      "epoch": 1.0415839618896388,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000490016409803204,
      "loss": 5.2549,
      "mean_token_accuracy": 0.1603129178285599,
      "num_tokens": 22993111.0,
      "step": 10495
    },
    {
      "entropy": 5.554627847671509,
      "epoch": 1.0420801905518062,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004900059699186108,
      "loss": 5.2244,
      "mean_token_accuracy": 0.16340605169534683,
      "num_tokens": 23003115.0,
      "step": 10500
    },
    {
      "entropy": 5.618977880477905,
      "epoch": 1.0425764192139737,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004899955247023328,
      "loss": 5.2683,
      "mean_token_accuracy": 0.16093970984220504,
      "num_tokens": 23012587.0,
      "step": 10505
    },
    {
      "entropy": 5.584914302825927,
      "epoch": 1.0430726478761414,
      "grad_norm": 0.859375,
      "learning_rate": 0.000489985074154629,
      "loss": 5.2885,
      "mean_token_accuracy": 0.15495433360338212,
      "num_tokens": 23024455.0,
      "step": 10510
    },
    {
      "entropy": 5.559906768798828,
      "epoch": 1.0435688765383089,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004899746182757587,
      "loss": 5.2154,
      "mean_token_accuracy": 0.1636367544531822,
      "num_tokens": 23034430.0,
      "step": 10515
    },
    {
      "entropy": 5.573592901229858,
      "epoch": 1.0440651052004764,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004899641570659811,
      "loss": 5.2827,
      "mean_token_accuracy": 0.16297447830438613,
      "num_tokens": 23044811.0,
      "step": 10520
    },
    {
      "entropy": 5.583052492141723,
      "epoch": 1.0445613338626438,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004899536905255555,
      "loss": 5.3092,
      "mean_token_accuracy": 0.16133083999156952,
      "num_tokens": 23056179.0,
      "step": 10525
    },
    {
      "entropy": 5.5573999881744385,
      "epoch": 1.0450575625248115,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004899432186547416,
      "loss": 5.2249,
      "mean_token_accuracy": 0.16555059999227523,
      "num_tokens": 23068331.0,
      "step": 10530
    },
    {
      "entropy": 5.532347536087036,
      "epoch": 1.045553791186979,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004899327414537991,
      "loss": 5.2172,
      "mean_token_accuracy": 0.15943560004234314,
      "num_tokens": 23079213.0,
      "step": 10535
    },
    {
      "entropy": 5.639857959747315,
      "epoch": 1.0460500198491465,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004899222589229877,
      "loss": 5.3245,
      "mean_token_accuracy": 0.1575188159942627,
      "num_tokens": 23089860.0,
      "step": 10540
    },
    {
      "entropy": 5.526367950439453,
      "epoch": 1.046546248511314,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004899117710625674,
      "loss": 5.1883,
      "mean_token_accuracy": 0.17317425310611725,
      "num_tokens": 23099431.0,
      "step": 10545
    },
    {
      "entropy": 5.567355155944824,
      "epoch": 1.0470424771734816,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004899012778727983,
      "loss": 5.2666,
      "mean_token_accuracy": 0.1594635382294655,
      "num_tokens": 23110947.0,
      "step": 10550
    },
    {
      "entropy": 5.594871854782104,
      "epoch": 1.0475387058356491,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004898907793539406,
      "loss": 5.3477,
      "mean_token_accuracy": 0.16127041429281236,
      "num_tokens": 23122782.0,
      "step": 10555
    },
    {
      "entropy": 5.639073514938355,
      "epoch": 1.0480349344978166,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004898802755062546,
      "loss": 5.2485,
      "mean_token_accuracy": 0.16379223763942719,
      "num_tokens": 23134254.0,
      "step": 10560
    },
    {
      "entropy": 5.5020448684692385,
      "epoch": 1.048531163159984,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004898697663300007,
      "loss": 5.0902,
      "mean_token_accuracy": 0.17229089438915252,
      "num_tokens": 23143851.0,
      "step": 10565
    },
    {
      "entropy": 5.543505001068115,
      "epoch": 1.0490273918221515,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004898592518254398,
      "loss": 5.2111,
      "mean_token_accuracy": 0.1649421662092209,
      "num_tokens": 23155966.0,
      "step": 10570
    },
    {
      "entropy": 5.5180806636810305,
      "epoch": 1.0495236204843192,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004898487319928322,
      "loss": 5.2171,
      "mean_token_accuracy": 0.16143583357334138,
      "num_tokens": 23165896.0,
      "step": 10575
    },
    {
      "entropy": 5.549243545532226,
      "epoch": 1.0500198491464867,
      "grad_norm": 0.7578125,
      "learning_rate": 0.0004898382068324392,
      "loss": 5.1975,
      "mean_token_accuracy": 0.16661687195301056,
      "num_tokens": 23177821.0,
      "step": 10580
    },
    {
      "entropy": 5.575086975097657,
      "epoch": 1.0505160778086542,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004898276763445215,
      "loss": 5.2779,
      "mean_token_accuracy": 0.16044059693813323,
      "num_tokens": 23188916.0,
      "step": 10585
    },
    {
      "entropy": 5.6137364387512205,
      "epoch": 1.0510123064708217,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004898171405293404,
      "loss": 5.2622,
      "mean_token_accuracy": 0.16863634288311005,
      "num_tokens": 23200490.0,
      "step": 10590
    },
    {
      "entropy": 5.5093831539154055,
      "epoch": 1.0515085351329894,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004898065993871571,
      "loss": 5.1958,
      "mean_token_accuracy": 0.16129535883665086,
      "num_tokens": 23212991.0,
      "step": 10595
    },
    {
      "entropy": 5.622107601165771,
      "epoch": 1.0520047637951568,
      "grad_norm": 0.875,
      "learning_rate": 0.000489796052918233,
      "loss": 5.3079,
      "mean_token_accuracy": 0.15788715630769729,
      "num_tokens": 23223361.0,
      "step": 10600
    },
    {
      "entropy": 5.596247768402099,
      "epoch": 1.0525009924573243,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004897855011228295,
      "loss": 5.2133,
      "mean_token_accuracy": 0.16734395325183868,
      "num_tokens": 23233765.0,
      "step": 10605
    },
    {
      "entropy": 5.5176421165466305,
      "epoch": 1.0529972211194918,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004897749440012085,
      "loss": 5.1477,
      "mean_token_accuracy": 0.1680014982819557,
      "num_tokens": 23246562.0,
      "step": 10610
    },
    {
      "entropy": 5.547542333602905,
      "epoch": 1.0534934497816595,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004897643815536316,
      "loss": 5.1867,
      "mean_token_accuracy": 0.16317762434482574,
      "num_tokens": 23256648.0,
      "step": 10615
    },
    {
      "entropy": 5.580404472351074,
      "epoch": 1.053989678443827,
      "grad_norm": 0.80859375,
      "learning_rate": 0.000489753813780361,
      "loss": 5.2331,
      "mean_token_accuracy": 0.1619618281722069,
      "num_tokens": 23267428.0,
      "step": 10620
    },
    {
      "entropy": 5.5343564510345455,
      "epoch": 1.0544859071059944,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004897432406816585,
      "loss": 5.1703,
      "mean_token_accuracy": 0.16505475640296935,
      "num_tokens": 23278440.0,
      "step": 10625
    },
    {
      "entropy": 5.544847297668457,
      "epoch": 1.054982135768162,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004897326622577863,
      "loss": 5.212,
      "mean_token_accuracy": 0.1572507679462433,
      "num_tokens": 23290571.0,
      "step": 10630
    },
    {
      "entropy": 5.620430898666382,
      "epoch": 1.0554783644303294,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004897220785090067,
      "loss": 5.3031,
      "mean_token_accuracy": 0.15959957987070084,
      "num_tokens": 23301634.0,
      "step": 10635
    },
    {
      "entropy": 5.508328247070312,
      "epoch": 1.055974593092497,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004897114894355824,
      "loss": 5.1701,
      "mean_token_accuracy": 0.16848744302988053,
      "num_tokens": 23313125.0,
      "step": 10640
    },
    {
      "entropy": 5.520039892196655,
      "epoch": 1.0564708217546646,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004897008950377758,
      "loss": 5.1828,
      "mean_token_accuracy": 0.16111970990896224,
      "num_tokens": 23323610.0,
      "step": 10645
    },
    {
      "entropy": 5.505582284927368,
      "epoch": 1.056967050416832,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004896902953158497,
      "loss": 5.1596,
      "mean_token_accuracy": 0.16464336961507797,
      "num_tokens": 23333652.0,
      "step": 10650
    },
    {
      "entropy": 5.500342750549317,
      "epoch": 1.0574632790789995,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004896796902700667,
      "loss": 5.2099,
      "mean_token_accuracy": 0.16780153959989547,
      "num_tokens": 23345954.0,
      "step": 10655
    },
    {
      "entropy": 5.5207911968231205,
      "epoch": 1.0579595077411672,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004896690799006901,
      "loss": 5.1939,
      "mean_token_accuracy": 0.16783688515424727,
      "num_tokens": 23357052.0,
      "step": 10660
    },
    {
      "entropy": 5.567644262313843,
      "epoch": 1.0584557364033347,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004896584642079828,
      "loss": 5.1852,
      "mean_token_accuracy": 0.17200947701931,
      "num_tokens": 23368837.0,
      "step": 10665
    },
    {
      "entropy": 5.510286521911621,
      "epoch": 1.0589519650655022,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004896478431922083,
      "loss": 5.1754,
      "mean_token_accuracy": 0.16501798629760742,
      "num_tokens": 23378608.0,
      "step": 10670
    },
    {
      "entropy": 5.577546834945679,
      "epoch": 1.0594481937276696,
      "grad_norm": 0.875,
      "learning_rate": 0.0004896372168536297,
      "loss": 5.2464,
      "mean_token_accuracy": 0.16384269744157792,
      "num_tokens": 23390094.0,
      "step": 10675
    },
    {
      "entropy": 5.521884918212891,
      "epoch": 1.0599444223898373,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004896265851925107,
      "loss": 5.1273,
      "mean_token_accuracy": 0.16916622072458268,
      "num_tokens": 23399991.0,
      "step": 10680
    },
    {
      "entropy": 5.565446949005127,
      "epoch": 1.0604406510520048,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004896159482091148,
      "loss": 5.3217,
      "mean_token_accuracy": 0.15393658131361007,
      "num_tokens": 23410705.0,
      "step": 10685
    },
    {
      "entropy": 5.490183115005493,
      "epoch": 1.0609368797141723,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004896053059037058,
      "loss": 5.1054,
      "mean_token_accuracy": 0.17834767401218415,
      "num_tokens": 23421939.0,
      "step": 10690
    },
    {
      "entropy": 5.547630167007446,
      "epoch": 1.0614331083763398,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004895946582765477,
      "loss": 5.2403,
      "mean_token_accuracy": 0.1605248786509037,
      "num_tokens": 23433252.0,
      "step": 10695
    },
    {
      "entropy": 5.649627161026001,
      "epoch": 1.0619293370385074,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004895840053279045,
      "loss": 5.3047,
      "mean_token_accuracy": 0.1565993219614029,
      "num_tokens": 23445582.0,
      "step": 10700
    },
    {
      "entropy": 5.60404782295227,
      "epoch": 1.062425565700675,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004895733470580404,
      "loss": 5.269,
      "mean_token_accuracy": 0.155484139919281,
      "num_tokens": 23456360.0,
      "step": 10705
    },
    {
      "entropy": 5.570484304428101,
      "epoch": 1.0629217943628424,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0004895626834672196,
      "loss": 5.1499,
      "mean_token_accuracy": 0.16302104145288468,
      "num_tokens": 23466576.0,
      "step": 10710
    },
    {
      "entropy": 5.469333744049072,
      "epoch": 1.0634180230250099,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004895520145557066,
      "loss": 5.1063,
      "mean_token_accuracy": 0.17305240035057068,
      "num_tokens": 23476200.0,
      "step": 10715
    },
    {
      "entropy": 5.577377367019653,
      "epoch": 1.0639142516871773,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004895413403237659,
      "loss": 5.2998,
      "mean_token_accuracy": 0.1634572774171829,
      "num_tokens": 23487776.0,
      "step": 10720
    },
    {
      "entropy": 5.531701850891113,
      "epoch": 1.064410480349345,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004895306607716624,
      "loss": 5.1623,
      "mean_token_accuracy": 0.17161939442157745,
      "num_tokens": 23497517.0,
      "step": 10725
    },
    {
      "entropy": 5.566427087783813,
      "epoch": 1.0649067090115125,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004895199758996607,
      "loss": 5.186,
      "mean_token_accuracy": 0.1692622035741806,
      "num_tokens": 23507436.0,
      "step": 10730
    },
    {
      "entropy": 5.547843360900879,
      "epoch": 1.06540293767368,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004895092857080258,
      "loss": 5.1391,
      "mean_token_accuracy": 0.15670267716050149,
      "num_tokens": 23517170.0,
      "step": 10735
    },
    {
      "entropy": 5.523667716979981,
      "epoch": 1.0658991663358475,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004894985901970229,
      "loss": 5.1461,
      "mean_token_accuracy": 0.17285354286432267,
      "num_tokens": 23527329.0,
      "step": 10740
    },
    {
      "entropy": 5.546849012374878,
      "epoch": 1.0663953949980152,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004894878893669171,
      "loss": 5.2449,
      "mean_token_accuracy": 0.1577153228223324,
      "num_tokens": 23538341.0,
      "step": 10745
    },
    {
      "entropy": 5.55429220199585,
      "epoch": 1.0668916236601826,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004894771832179739,
      "loss": 5.1759,
      "mean_token_accuracy": 0.1622370421886444,
      "num_tokens": 23549522.0,
      "step": 10750
    },
    {
      "entropy": 5.601254224777222,
      "epoch": 1.0673878523223501,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004894664717504587,
      "loss": 5.212,
      "mean_token_accuracy": 0.1650427684187889,
      "num_tokens": 23559274.0,
      "step": 10755
    },
    {
      "entropy": 5.543829917907715,
      "epoch": 1.0678840809845176,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004894557549646371,
      "loss": 5.1676,
      "mean_token_accuracy": 0.16680517047643661,
      "num_tokens": 23570852.0,
      "step": 10760
    },
    {
      "entropy": 5.522302579879761,
      "epoch": 1.0683803096466853,
      "grad_norm": 0.875,
      "learning_rate": 0.0004894450328607749,
      "loss": 5.1753,
      "mean_token_accuracy": 0.16622244268655778,
      "num_tokens": 23582918.0,
      "step": 10765
    },
    {
      "entropy": 5.489535284042359,
      "epoch": 1.0688765383088528,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004894343054391379,
      "loss": 5.2023,
      "mean_token_accuracy": 0.17074786126613617,
      "num_tokens": 23594177.0,
      "step": 10770
    },
    {
      "entropy": 5.546436786651611,
      "epoch": 1.0693727669710202,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004894235726999924,
      "loss": 5.1724,
      "mean_token_accuracy": 0.1656903401017189,
      "num_tokens": 23605901.0,
      "step": 10775
    },
    {
      "entropy": 5.521238327026367,
      "epoch": 1.0698689956331877,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004894128346436042,
      "loss": 5.2198,
      "mean_token_accuracy": 0.16673441082239152,
      "num_tokens": 23617354.0,
      "step": 10780
    },
    {
      "entropy": 5.568436861038208,
      "epoch": 1.0703652242953554,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004894020912702397,
      "loss": 5.241,
      "mean_token_accuracy": 0.16437995582818984,
      "num_tokens": 23627603.0,
      "step": 10785
    },
    {
      "entropy": 5.4602922916412355,
      "epoch": 1.0708614529575229,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004893913425801653,
      "loss": 5.0669,
      "mean_token_accuracy": 0.17634545862674714,
      "num_tokens": 23637280.0,
      "step": 10790
    },
    {
      "entropy": 5.541663074493409,
      "epoch": 1.0713576816196904,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004893805885736477,
      "loss": 5.2737,
      "mean_token_accuracy": 0.16722501814365387,
      "num_tokens": 23649326.0,
      "step": 10795
    },
    {
      "entropy": 5.614739513397216,
      "epoch": 1.0718539102818578,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004893698292509534,
      "loss": 5.2817,
      "mean_token_accuracy": 0.15696632862091064,
      "num_tokens": 23659561.0,
      "step": 10800
    },
    {
      "entropy": 5.549629878997803,
      "epoch": 1.0723501389440253,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004893590646123492,
      "loss": 5.1827,
      "mean_token_accuracy": 0.1700023129582405,
      "num_tokens": 23673206.0,
      "step": 10805
    },
    {
      "entropy": 5.521047687530517,
      "epoch": 1.072846367606193,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004893482946581022,
      "loss": 5.231,
      "mean_token_accuracy": 0.15918735712766646,
      "num_tokens": 23683920.0,
      "step": 10810
    },
    {
      "entropy": 5.5158453464508055,
      "epoch": 1.0733425962683605,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004893375193884793,
      "loss": 5.1361,
      "mean_token_accuracy": 0.1718864232301712,
      "num_tokens": 23693990.0,
      "step": 10815
    },
    {
      "entropy": 5.59973783493042,
      "epoch": 1.073838824930528,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004893267388037478,
      "loss": 5.2571,
      "mean_token_accuracy": 0.15837550908327103,
      "num_tokens": 23704870.0,
      "step": 10820
    },
    {
      "entropy": 5.536206817626953,
      "epoch": 1.0743350535926954,
      "grad_norm": 0.921875,
      "learning_rate": 0.000489315952904175,
      "loss": 5.2354,
      "mean_token_accuracy": 0.16550104022026063,
      "num_tokens": 23715159.0,
      "step": 10825
    },
    {
      "entropy": 5.517267560958862,
      "epoch": 1.0748312822548631,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004893051616900283,
      "loss": 5.1658,
      "mean_token_accuracy": 0.16473808735609055,
      "num_tokens": 23725252.0,
      "step": 10830
    },
    {
      "entropy": 5.476993989944458,
      "epoch": 1.0753275109170306,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004892943651615756,
      "loss": 5.1883,
      "mean_token_accuracy": 0.16928842812776565,
      "num_tokens": 23736472.0,
      "step": 10835
    },
    {
      "entropy": 5.572298336029053,
      "epoch": 1.075823739579198,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004892835633190843,
      "loss": 5.1026,
      "mean_token_accuracy": 0.1639563947916031,
      "num_tokens": 23747565.0,
      "step": 10840
    },
    {
      "entropy": 5.533527135848999,
      "epoch": 1.0763199682413656,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004892727561628223,
      "loss": 5.19,
      "mean_token_accuracy": 0.16483999490737916,
      "num_tokens": 23758201.0,
      "step": 10845
    },
    {
      "entropy": 5.554649829864502,
      "epoch": 1.0768161969035333,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004892619436930577,
      "loss": 5.2442,
      "mean_token_accuracy": 0.16440054923295974,
      "num_tokens": 23769087.0,
      "step": 10850
    },
    {
      "entropy": 5.540117359161377,
      "epoch": 1.0773124255657007,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004892511259100585,
      "loss": 5.2481,
      "mean_token_accuracy": 0.16564065217971802,
      "num_tokens": 23780598.0,
      "step": 10855
    },
    {
      "entropy": 5.536248350143433,
      "epoch": 1.0778086542278682,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004892403028140932,
      "loss": 5.1703,
      "mean_token_accuracy": 0.15898965299129486,
      "num_tokens": 23791449.0,
      "step": 10860
    },
    {
      "entropy": 5.516120576858521,
      "epoch": 1.0783048828900357,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00048922947440543,
      "loss": 5.1474,
      "mean_token_accuracy": 0.1687377378344536,
      "num_tokens": 23803071.0,
      "step": 10865
    },
    {
      "entropy": 5.576876544952393,
      "epoch": 1.0788011115522032,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004892186406843374,
      "loss": 5.2267,
      "mean_token_accuracy": 0.1598200261592865,
      "num_tokens": 23813372.0,
      "step": 10870
    },
    {
      "entropy": 5.543317222595215,
      "epoch": 1.0792973402143708,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004892078016510841,
      "loss": 5.2064,
      "mean_token_accuracy": 0.16511232703924178,
      "num_tokens": 23824979.0,
      "step": 10875
    },
    {
      "entropy": 5.532471513748169,
      "epoch": 1.0797935688765383,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000489196957305939,
      "loss": 5.1873,
      "mean_token_accuracy": 0.16898007094860076,
      "num_tokens": 23836335.0,
      "step": 10880
    },
    {
      "entropy": 5.516027307510376,
      "epoch": 1.0802897975387058,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004891861076491709,
      "loss": 5.1763,
      "mean_token_accuracy": 0.16643766015768052,
      "num_tokens": 23848147.0,
      "step": 10885
    },
    {
      "entropy": 5.485627222061157,
      "epoch": 1.0807860262008733,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004891752526810488,
      "loss": 5.1257,
      "mean_token_accuracy": 0.17008547633886337,
      "num_tokens": 23859502.0,
      "step": 10890
    },
    {
      "entropy": 5.558950901031494,
      "epoch": 1.081282254863041,
      "grad_norm": 0.80859375,
      "learning_rate": 0.000489164392401842,
      "loss": 5.2804,
      "mean_token_accuracy": 0.16260500401258468,
      "num_tokens": 23870433.0,
      "step": 10895
    },
    {
      "entropy": 5.533802032470703,
      "epoch": 1.0817784835252084,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004891535268118196,
      "loss": 5.1143,
      "mean_token_accuracy": 0.17327011674642562,
      "num_tokens": 23879581.0,
      "step": 10900
    },
    {
      "entropy": 5.481846332550049,
      "epoch": 1.082274712187376,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004891426559112514,
      "loss": 5.1806,
      "mean_token_accuracy": 0.17126995623111724,
      "num_tokens": 23890889.0,
      "step": 10905
    },
    {
      "entropy": 5.5440630435943605,
      "epoch": 1.0827709408495434,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004891317797004066,
      "loss": 5.1636,
      "mean_token_accuracy": 0.1686042696237564,
      "num_tokens": 23901577.0,
      "step": 10910
    },
    {
      "entropy": 5.522806215286255,
      "epoch": 1.083267169511711,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004891208981795552,
      "loss": 5.2405,
      "mean_token_accuracy": 0.16895045787096025,
      "num_tokens": 23912698.0,
      "step": 10915
    },
    {
      "entropy": 5.448703098297119,
      "epoch": 1.0837633981738786,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004891100113489669,
      "loss": 5.1132,
      "mean_token_accuracy": 0.17265091836452484,
      "num_tokens": 23923056.0,
      "step": 10920
    },
    {
      "entropy": 5.51653528213501,
      "epoch": 1.084259626836046,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004890991192089116,
      "loss": 5.2445,
      "mean_token_accuracy": 0.16402831077575683,
      "num_tokens": 23933998.0,
      "step": 10925
    },
    {
      "entropy": 5.573494100570679,
      "epoch": 1.0847558554982135,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004890882217596595,
      "loss": 5.1456,
      "mean_token_accuracy": 0.17067973613739013,
      "num_tokens": 23943980.0,
      "step": 10930
    },
    {
      "entropy": 5.516656303405762,
      "epoch": 1.085252084160381,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004890773190014809,
      "loss": 5.2213,
      "mean_token_accuracy": 0.16332485824823378,
      "num_tokens": 23955646.0,
      "step": 10935
    },
    {
      "entropy": 5.47779278755188,
      "epoch": 1.0857483128225487,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000489066410934646,
      "loss": 5.1906,
      "mean_token_accuracy": 0.17210307717323303,
      "num_tokens": 23966381.0,
      "step": 10940
    },
    {
      "entropy": 5.61372389793396,
      "epoch": 1.0862445414847162,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004890554975594255,
      "loss": 5.251,
      "mean_token_accuracy": 0.15600233227014543,
      "num_tokens": 23978669.0,
      "step": 10945
    },
    {
      "entropy": 5.673389863967896,
      "epoch": 1.0867407701468836,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004890445788760898,
      "loss": 5.3326,
      "mean_token_accuracy": 0.15747251212596894,
      "num_tokens": 23989467.0,
      "step": 10950
    },
    {
      "entropy": 5.512957191467285,
      "epoch": 1.0872369988090511,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004890336548849098,
      "loss": 5.1343,
      "mean_token_accuracy": 0.1714241921901703,
      "num_tokens": 24000797.0,
      "step": 10955
    },
    {
      "entropy": 5.607582521438599,
      "epoch": 1.0877332274712188,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004890227255861564,
      "loss": 5.2898,
      "mean_token_accuracy": 0.16513597071170807,
      "num_tokens": 24013182.0,
      "step": 10960
    },
    {
      "entropy": 5.543015098571777,
      "epoch": 1.0882294561333863,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004890117909801005,
      "loss": 5.18,
      "mean_token_accuracy": 0.17219192832708358,
      "num_tokens": 24024005.0,
      "step": 10965
    },
    {
      "entropy": 5.516589260101318,
      "epoch": 1.0887256847955538,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004890008510670134,
      "loss": 5.1614,
      "mean_token_accuracy": 0.1672101154923439,
      "num_tokens": 24034524.0,
      "step": 10970
    },
    {
      "entropy": 5.5888190269470215,
      "epoch": 1.0892219134577212,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004889899058471663,
      "loss": 5.2204,
      "mean_token_accuracy": 0.1629112109541893,
      "num_tokens": 24045944.0,
      "step": 10975
    },
    {
      "entropy": 5.60032696723938,
      "epoch": 1.089718142119889,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004889789553208306,
      "loss": 5.2917,
      "mean_token_accuracy": 0.16396206766366958,
      "num_tokens": 24058612.0,
      "step": 10980
    },
    {
      "entropy": 5.54950361251831,
      "epoch": 1.0902143707820564,
      "grad_norm": 1.03125,
      "learning_rate": 0.000488967999488278,
      "loss": 5.1604,
      "mean_token_accuracy": 0.16761168986558914,
      "num_tokens": 24068912.0,
      "step": 10985
    },
    {
      "entropy": 5.464646720886231,
      "epoch": 1.0907105994442239,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048895703834978,
      "loss": 5.162,
      "mean_token_accuracy": 0.16602682024240495,
      "num_tokens": 24079879.0,
      "step": 10990
    },
    {
      "entropy": 5.52567458152771,
      "epoch": 1.0912068281063914,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004889460719056085,
      "loss": 5.1914,
      "mean_token_accuracy": 0.16645094007253647,
      "num_tokens": 24091268.0,
      "step": 10995
    },
    {
      "entropy": 5.5216775894165036,
      "epoch": 1.091703056768559,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004889351001560354,
      "loss": 5.2195,
      "mean_token_accuracy": 0.166425521671772,
      "num_tokens": 24101428.0,
      "step": 11000
    },
    {
      "entropy": 5.5186601161956785,
      "epoch": 1.0921992854307265,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004889241231013329,
      "loss": 5.1879,
      "mean_token_accuracy": 0.16922087222337723,
      "num_tokens": 24112920.0,
      "step": 11005
    },
    {
      "entropy": 5.474888849258423,
      "epoch": 1.092695514092894,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000488913140741773,
      "loss": 5.1195,
      "mean_token_accuracy": 0.1672017127275467,
      "num_tokens": 24125731.0,
      "step": 11010
    },
    {
      "entropy": 5.485093688964843,
      "epoch": 1.0931917427550615,
      "grad_norm": 0.875,
      "learning_rate": 0.0004889021530776282,
      "loss": 5.1339,
      "mean_token_accuracy": 0.16532045900821685,
      "num_tokens": 24136606.0,
      "step": 11015
    },
    {
      "entropy": 5.549827671051025,
      "epoch": 1.093687971417229,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004888911601091709,
      "loss": 5.1493,
      "mean_token_accuracy": 0.16973443180322648,
      "num_tokens": 24146783.0,
      "step": 11020
    },
    {
      "entropy": 5.5369971752166744,
      "epoch": 1.0941842000793967,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004888801618366737,
      "loss": 5.2444,
      "mean_token_accuracy": 0.16227903217077255,
      "num_tokens": 24157412.0,
      "step": 11025
    },
    {
      "entropy": 5.546850252151489,
      "epoch": 1.0946804287415641,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004888691582604095,
      "loss": 5.2332,
      "mean_token_accuracy": 0.16568017452955247,
      "num_tokens": 24168077.0,
      "step": 11030
    },
    {
      "entropy": 5.568090295791626,
      "epoch": 1.0951766574037316,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000488858149380651,
      "loss": 5.2719,
      "mean_token_accuracy": 0.16354267597198485,
      "num_tokens": 24178913.0,
      "step": 11035
    },
    {
      "entropy": 5.637010669708252,
      "epoch": 1.095672886065899,
      "grad_norm": 0.921875,
      "learning_rate": 0.000488847135197671,
      "loss": 5.3148,
      "mean_token_accuracy": 0.15793400704860688,
      "num_tokens": 24190356.0,
      "step": 11040
    },
    {
      "entropy": 5.6193829536437985,
      "epoch": 1.0961691147280668,
      "grad_norm": 0.984375,
      "learning_rate": 0.000488836115711743,
      "loss": 5.2933,
      "mean_token_accuracy": 0.15517248809337617,
      "num_tokens": 24199772.0,
      "step": 11045
    },
    {
      "entropy": 5.565176010131836,
      "epoch": 1.0966653433902342,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004888250909231401,
      "loss": 5.1992,
      "mean_token_accuracy": 0.16370121985673905,
      "num_tokens": 24211807.0,
      "step": 11050
    },
    {
      "entropy": 5.500342893600464,
      "epoch": 1.0971615720524017,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004888140608321357,
      "loss": 5.1746,
      "mean_token_accuracy": 0.170803065598011,
      "num_tokens": 24223156.0,
      "step": 11055
    },
    {
      "entropy": 5.607478904724121,
      "epoch": 1.0976578007145692,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004888030254390033,
      "loss": 5.2322,
      "mean_token_accuracy": 0.16016623750329018,
      "num_tokens": 24234352.0,
      "step": 11060
    },
    {
      "entropy": 5.528764867782593,
      "epoch": 1.098154029376737,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004887919847440166,
      "loss": 5.1604,
      "mean_token_accuracy": 0.17382446974515914,
      "num_tokens": 24244137.0,
      "step": 11065
    },
    {
      "entropy": 5.478833484649658,
      "epoch": 1.0986502580389044,
      "grad_norm": 1.0,
      "learning_rate": 0.0004887809387474494,
      "loss": 5.1528,
      "mean_token_accuracy": 0.16616191267967223,
      "num_tokens": 24253250.0,
      "step": 11070
    },
    {
      "entropy": 5.570534372329712,
      "epoch": 1.0991464867010718,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004887698874495755,
      "loss": 5.2731,
      "mean_token_accuracy": 0.16532145887613298,
      "num_tokens": 24264551.0,
      "step": 11075
    },
    {
      "entropy": 5.548721599578857,
      "epoch": 1.0996427153632393,
      "grad_norm": 0.82421875,
      "learning_rate": 0.000488758830850669,
      "loss": 5.094,
      "mean_token_accuracy": 0.174309666454792,
      "num_tokens": 24275522.0,
      "step": 11080
    },
    {
      "entropy": 5.4962043285369875,
      "epoch": 1.100138944025407,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004887477689510042,
      "loss": 5.2571,
      "mean_token_accuracy": 0.16220258325338363,
      "num_tokens": 24286333.0,
      "step": 11085
    },
    {
      "entropy": 5.492782783508301,
      "epoch": 1.1006351726875745,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004887367017508553,
      "loss": 5.2564,
      "mean_token_accuracy": 0.16658498495817184,
      "num_tokens": 24297627.0,
      "step": 11090
    },
    {
      "entropy": 5.507976388931274,
      "epoch": 1.101131401349742,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004887256292504967,
      "loss": 5.2001,
      "mean_token_accuracy": 0.1654631868004799,
      "num_tokens": 24308246.0,
      "step": 11095
    },
    {
      "entropy": 5.56145453453064,
      "epoch": 1.1016276300119094,
      "grad_norm": 1.046875,
      "learning_rate": 0.000488714551450203,
      "loss": 5.2535,
      "mean_token_accuracy": 0.15819311887025833,
      "num_tokens": 24317718.0,
      "step": 11100
    },
    {
      "entropy": 5.531434106826782,
      "epoch": 1.102123858674077,
      "grad_norm": 0.890625,
      "learning_rate": 0.000488703468350249,
      "loss": 5.1786,
      "mean_token_accuracy": 0.16667785942554475,
      "num_tokens": 24329242.0,
      "step": 11105
    },
    {
      "entropy": 5.640681743621826,
      "epoch": 1.1026200873362446,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004886923799509095,
      "loss": 5.3281,
      "mean_token_accuracy": 0.15917756482958795,
      "num_tokens": 24340146.0,
      "step": 11110
    },
    {
      "entropy": 5.588403940200806,
      "epoch": 1.103116315998412,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004886812862524595,
      "loss": 5.2353,
      "mean_token_accuracy": 0.16205300241708756,
      "num_tokens": 24349370.0,
      "step": 11115
    },
    {
      "entropy": 5.609292650222779,
      "epoch": 1.1036125446605796,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004886701872551739,
      "loss": 5.1904,
      "mean_token_accuracy": 0.16353601068258286,
      "num_tokens": 24359770.0,
      "step": 11120
    },
    {
      "entropy": 5.524088191986084,
      "epoch": 1.104108773322747,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004886590829593281,
      "loss": 5.221,
      "mean_token_accuracy": 0.16579868793487548,
      "num_tokens": 24370974.0,
      "step": 11125
    },
    {
      "entropy": 5.527943181991577,
      "epoch": 1.1046050019849147,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004886479733651974,
      "loss": 5.2428,
      "mean_token_accuracy": 0.16564814150333404,
      "num_tokens": 24381771.0,
      "step": 11130
    },
    {
      "entropy": 5.503493356704712,
      "epoch": 1.1051012306470822,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004886368584730575,
      "loss": 5.1979,
      "mean_token_accuracy": 0.1624273732304573,
      "num_tokens": 24393067.0,
      "step": 11135
    },
    {
      "entropy": 5.547466897964478,
      "epoch": 1.1055974593092497,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004886257382831837,
      "loss": 5.2638,
      "mean_token_accuracy": 0.15839179530739783,
      "num_tokens": 24402866.0,
      "step": 11140
    },
    {
      "entropy": 5.600502014160156,
      "epoch": 1.1060936879714172,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000488614612795852,
      "loss": 5.2083,
      "mean_token_accuracy": 0.1660041555762291,
      "num_tokens": 24412956.0,
      "step": 11145
    },
    {
      "entropy": 5.5997397899627686,
      "epoch": 1.1065899166335849,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004886034820113382,
      "loss": 5.2651,
      "mean_token_accuracy": 0.16328827440738677,
      "num_tokens": 24424650.0,
      "step": 11150
    },
    {
      "entropy": 5.566668844223022,
      "epoch": 1.1070861452957523,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004885923459299182,
      "loss": 5.2725,
      "mean_token_accuracy": 0.1630299359560013,
      "num_tokens": 24436541.0,
      "step": 11155
    },
    {
      "entropy": 5.565120887756348,
      "epoch": 1.1075823739579198,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004885812045518685,
      "loss": 5.2507,
      "mean_token_accuracy": 0.15863975435495375,
      "num_tokens": 24447588.0,
      "step": 11160
    },
    {
      "entropy": 5.581294250488281,
      "epoch": 1.1080786026200873,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000488570057877465,
      "loss": 5.1758,
      "mean_token_accuracy": 0.16948388814926146,
      "num_tokens": 24458854.0,
      "step": 11165
    },
    {
      "entropy": 5.57621202468872,
      "epoch": 1.108574831282255,
      "grad_norm": 1.0,
      "learning_rate": 0.0004885589059069842,
      "loss": 5.25,
      "mean_token_accuracy": 0.1678898736834526,
      "num_tokens": 24469427.0,
      "step": 11170
    },
    {
      "entropy": 5.500083971023559,
      "epoch": 1.1090710599444225,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004885477486407028,
      "loss": 5.2404,
      "mean_token_accuracy": 0.15860867500305176,
      "num_tokens": 24479352.0,
      "step": 11175
    },
    {
      "entropy": 5.61491756439209,
      "epoch": 1.10956728860659,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004885365860788974,
      "loss": 5.2464,
      "mean_token_accuracy": 0.1635689303278923,
      "num_tokens": 24489580.0,
      "step": 11180
    },
    {
      "entropy": 5.545247268676758,
      "epoch": 1.1100635172687574,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004885254182218448,
      "loss": 5.1802,
      "mean_token_accuracy": 0.16670125126838684,
      "num_tokens": 24501141.0,
      "step": 11185
    },
    {
      "entropy": 5.49181900024414,
      "epoch": 1.1105597459309249,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004885142450698219,
      "loss": 5.1794,
      "mean_token_accuracy": 0.1701214224100113,
      "num_tokens": 24512079.0,
      "step": 11190
    },
    {
      "entropy": 5.589020252227783,
      "epoch": 1.1110559745930926,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004885030666231057,
      "loss": 5.2397,
      "mean_token_accuracy": 0.16306669265031815,
      "num_tokens": 24521839.0,
      "step": 11195
    },
    {
      "entropy": 5.60635404586792,
      "epoch": 1.11155220325526,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004884918828819738,
      "loss": 5.2757,
      "mean_token_accuracy": 0.16229153722524642,
      "num_tokens": 24532704.0,
      "step": 11200
    },
    {
      "entropy": 5.569793033599853,
      "epoch": 1.1120484319174275,
      "grad_norm": 0.859375,
      "learning_rate": 0.000488480693846703,
      "loss": 5.1815,
      "mean_token_accuracy": 0.17026060372591018,
      "num_tokens": 24543790.0,
      "step": 11205
    },
    {
      "entropy": 5.524898624420166,
      "epoch": 1.112544660579595,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004884694995175711,
      "loss": 5.2289,
      "mean_token_accuracy": 0.1626759022474289,
      "num_tokens": 24554765.0,
      "step": 11210
    },
    {
      "entropy": 5.662857723236084,
      "epoch": 1.1130408892417627,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004884582998948555,
      "loss": 5.3157,
      "mean_token_accuracy": 0.15305445492267608,
      "num_tokens": 24565888.0,
      "step": 11215
    },
    {
      "entropy": 5.523702001571655,
      "epoch": 1.1135371179039302,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004884470949788342,
      "loss": 5.1761,
      "mean_token_accuracy": 0.16873531937599182,
      "num_tokens": 24575482.0,
      "step": 11220
    },
    {
      "entropy": 5.596697044372559,
      "epoch": 1.1140333465660976,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004884358847697847,
      "loss": 5.2971,
      "mean_token_accuracy": 0.1532044753432274,
      "num_tokens": 24588249.0,
      "step": 11225
    },
    {
      "entropy": 5.625243711471557,
      "epoch": 1.1145295752282651,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004884246692679852,
      "loss": 5.2455,
      "mean_token_accuracy": 0.167199943959713,
      "num_tokens": 24599546.0,
      "step": 11230
    },
    {
      "entropy": 5.556321811676026,
      "epoch": 1.1150258038904326,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004884134484737139,
      "loss": 5.1773,
      "mean_token_accuracy": 0.1655934751033783,
      "num_tokens": 24611777.0,
      "step": 11235
    },
    {
      "entropy": 5.507972145080567,
      "epoch": 1.1155220325526003,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004884022223872489,
      "loss": 5.1588,
      "mean_token_accuracy": 0.17026867121458053,
      "num_tokens": 24623675.0,
      "step": 11240
    },
    {
      "entropy": 5.5565314292907715,
      "epoch": 1.1160182612147678,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004883909910088686,
      "loss": 5.1799,
      "mean_token_accuracy": 0.16695850640535354,
      "num_tokens": 24633929.0,
      "step": 11245
    },
    {
      "entropy": 5.533818960189819,
      "epoch": 1.1165144898769352,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004883797543388515,
      "loss": 5.1524,
      "mean_token_accuracy": 0.16628415733575821,
      "num_tokens": 24642800.0,
      "step": 11250
    },
    {
      "entropy": 5.5059174537658695,
      "epoch": 1.1170107185391027,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004883685123774764,
      "loss": 5.1531,
      "mean_token_accuracy": 0.17160961478948594,
      "num_tokens": 24653780.0,
      "step": 11255
    },
    {
      "entropy": 5.516745233535767,
      "epoch": 1.1175069472012704,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004883572651250218,
      "loss": 5.1827,
      "mean_token_accuracy": 0.17066242396831513,
      "num_tokens": 24663912.0,
      "step": 11260
    },
    {
      "entropy": 5.574978923797607,
      "epoch": 1.118003175863438,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004883460125817669,
      "loss": 5.2731,
      "mean_token_accuracy": 0.16199380457401275,
      "num_tokens": 24675222.0,
      "step": 11265
    },
    {
      "entropy": 5.577271175384522,
      "epoch": 1.1184994045256054,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004883347547479905,
      "loss": 5.276,
      "mean_token_accuracy": 0.16459541767835617,
      "num_tokens": 24686697.0,
      "step": 11270
    },
    {
      "entropy": 5.624046897888183,
      "epoch": 1.1189956331877728,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000488323491623972,
      "loss": 5.2002,
      "mean_token_accuracy": 0.1671757936477661,
      "num_tokens": 24697067.0,
      "step": 11275
    },
    {
      "entropy": 5.577747583389282,
      "epoch": 1.1194918618499405,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004883122232099905,
      "loss": 5.2006,
      "mean_token_accuracy": 0.16444696187973024,
      "num_tokens": 24707989.0,
      "step": 11280
    },
    {
      "entropy": 5.4619499206542965,
      "epoch": 1.119988090512108,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004883009495063256,
      "loss": 5.1408,
      "mean_token_accuracy": 0.17277270853519439,
      "num_tokens": 24718826.0,
      "step": 11285
    },
    {
      "entropy": 5.597539281845092,
      "epoch": 1.1204843191742755,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004882896705132566,
      "loss": 5.2532,
      "mean_token_accuracy": 0.1603585734963417,
      "num_tokens": 24729913.0,
      "step": 11290
    },
    {
      "entropy": 5.481209945678711,
      "epoch": 1.120980547836443,
      "grad_norm": 0.859375,
      "learning_rate": 0.00048827838623106355,
      "loss": 5.1073,
      "mean_token_accuracy": 0.16847607791423796,
      "num_tokens": 24741621.0,
      "step": 11295
    },
    {
      "entropy": 5.5243284702301025,
      "epoch": 1.1214767764986107,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000488267096660026,
      "loss": 5.1724,
      "mean_token_accuracy": 0.17016072571277618,
      "num_tokens": 24753427.0,
      "step": 11300
    },
    {
      "entropy": 5.558738565444946,
      "epoch": 1.1219730051607781,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048825580180042406,
      "loss": 5.2042,
      "mean_token_accuracy": 0.16348448991775513,
      "num_tokens": 24763580.0,
      "step": 11305
    },
    {
      "entropy": 5.497127676010132,
      "epoch": 1.1224692338229456,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004882445016525377,
      "loss": 5.2041,
      "mean_token_accuracy": 0.1653308317065239,
      "num_tokens": 24773955.0,
      "step": 11310
    },
    {
      "entropy": 5.513176345825196,
      "epoch": 1.122965462485113,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00048823319621664723,
      "loss": 5.1112,
      "mean_token_accuracy": 0.17569485306739807,
      "num_tokens": 24784039.0,
      "step": 11315
    },
    {
      "entropy": 5.587033414840699,
      "epoch": 1.1234616911472806,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00048822188549303296,
      "loss": 5.2307,
      "mean_token_accuracy": 0.16120216846466065,
      "num_tokens": 24794630.0,
      "step": 11320
    },
    {
      "entropy": 5.545216703414917,
      "epoch": 1.1239579198094483,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004882105694819754,
      "loss": 5.2779,
      "mean_token_accuracy": 0.15836898982524872,
      "num_tokens": 24805625.0,
      "step": 11325
    },
    {
      "entropy": 5.559854125976562,
      "epoch": 1.1244541484716157,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004881992481837551,
      "loss": 5.196,
      "mean_token_accuracy": 0.1735105663537979,
      "num_tokens": 24817356.0,
      "step": 11330
    },
    {
      "entropy": 5.561701059341431,
      "epoch": 1.1249503771337832,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004881879215986529,
      "loss": 5.1631,
      "mean_token_accuracy": 0.16098037213087082,
      "num_tokens": 24828349.0,
      "step": 11335
    },
    {
      "entropy": 5.554211711883545,
      "epoch": 1.1254466057959507,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048817658972694967,
      "loss": 5.2108,
      "mean_token_accuracy": 0.16408946812152864,
      "num_tokens": 24840138.0,
      "step": 11340
    },
    {
      "entropy": 5.532898616790772,
      "epoch": 1.1259428344581184,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004881652525689263,
      "loss": 5.1995,
      "mean_token_accuracy": 0.16382676213979722,
      "num_tokens": 24851570.0,
      "step": 11345
    },
    {
      "entropy": 5.441495895385742,
      "epoch": 1.1264390631202859,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000488153910124864,
      "loss": 5.1163,
      "mean_token_accuracy": 0.17039304226636887,
      "num_tokens": 24861396.0,
      "step": 11350
    },
    {
      "entropy": 5.562624883651734,
      "epoch": 1.1269352917824533,
      "grad_norm": 0.875,
      "learning_rate": 0.000488142562395044,
      "loss": 5.2213,
      "mean_token_accuracy": 0.1602090373635292,
      "num_tokens": 24873519.0,
      "step": 11355
    },
    {
      "entropy": 5.5660299301147464,
      "epoch": 1.1274315204446208,
      "grad_norm": 0.76953125,
      "learning_rate": 0.00048813120937974774,
      "loss": 5.2435,
      "mean_token_accuracy": 0.1628718212246895,
      "num_tokens": 24886856.0,
      "step": 11360
    },
    {
      "entropy": 5.513981866836548,
      "epoch": 1.1279277491067885,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00048811985107925684,
      "loss": 5.1994,
      "mean_token_accuracy": 0.16299989074468613,
      "num_tokens": 24897292.0,
      "step": 11365
    },
    {
      "entropy": 5.554580497741699,
      "epoch": 1.128423977768956,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048810848749385286,
      "loss": 5.1944,
      "mean_token_accuracy": 0.1647961512207985,
      "num_tokens": 24907757.0,
      "step": 11370
    },
    {
      "entropy": 5.463225793838501,
      "epoch": 1.1289202064311235,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00048809711862381753,
      "loss": 5.181,
      "mean_token_accuracy": 0.17165081948041916,
      "num_tokens": 24918071.0,
      "step": 11375
    },
    {
      "entropy": 5.507682466506958,
      "epoch": 1.129416435093291,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048808574446943274,
      "loss": 5.1673,
      "mean_token_accuracy": 0.1700497329235077,
      "num_tokens": 24927967.0,
      "step": 11380
    },
    {
      "entropy": 5.5919568061828615,
      "epoch": 1.1299126637554586,
      "grad_norm": 1.0,
      "learning_rate": 0.0004880743650309808,
      "loss": 5.2355,
      "mean_token_accuracy": 0.167525714635849,
      "num_tokens": 24939170.0,
      "step": 11385
    },
    {
      "entropy": 5.545287179946899,
      "epoch": 1.130408892417626,
      "grad_norm": 0.83984375,
      "learning_rate": 0.00048806298030874373,
      "loss": 5.2201,
      "mean_token_accuracy": 0.16491880267858505,
      "num_tokens": 24950937.0,
      "step": 11390
    },
    {
      "entropy": 5.542399215698242,
      "epoch": 1.1309051210797936,
      "grad_norm": 0.78515625,
      "learning_rate": 0.00048805159030300383,
      "loss": 5.2144,
      "mean_token_accuracy": 0.16268894523382188,
      "num_tokens": 24962813.0,
      "step": 11395
    },
    {
      "entropy": 5.517576360702515,
      "epoch": 1.131401349741961,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048804019501404357,
      "loss": 5.1137,
      "mean_token_accuracy": 0.17117615193128585,
      "num_tokens": 24974192.0,
      "step": 11400
    },
    {
      "entropy": 5.481450986862183,
      "epoch": 1.1318975784041285,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048802879444214553,
      "loss": 5.105,
      "mean_token_accuracy": 0.17104082107543944,
      "num_tokens": 24985353.0,
      "step": 11405
    },
    {
      "entropy": 5.536077547073364,
      "epoch": 1.1323938070662962,
      "grad_norm": 0.83984375,
      "learning_rate": 0.00048801738858759243,
      "loss": 5.2083,
      "mean_token_accuracy": 0.17019502222537994,
      "num_tokens": 24997259.0,
      "step": 11410
    },
    {
      "entropy": 5.536036491394043,
      "epoch": 1.1328900357284637,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004880059774506671,
      "loss": 5.1911,
      "mean_token_accuracy": 0.16489135324954987,
      "num_tokens": 25008222.0,
      "step": 11415
    },
    {
      "entropy": 5.502761554718018,
      "epoch": 1.1333862643906312,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004879945610316525,
      "loss": 5.2437,
      "mean_token_accuracy": 0.159339103102684,
      "num_tokens": 25020267.0,
      "step": 11420
    },
    {
      "entropy": 5.5328371047973635,
      "epoch": 1.1338824930527986,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00048798313933083175,
      "loss": 5.2061,
      "mean_token_accuracy": 0.16459888070821763,
      "num_tokens": 25031212.0,
      "step": 11425
    },
    {
      "entropy": 5.535136032104492,
      "epoch": 1.1343787217149663,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004879717123484882,
      "loss": 5.1783,
      "mean_token_accuracy": 0.15966017097234725,
      "num_tokens": 25042204.0,
      "step": 11430
    },
    {
      "entropy": 5.563981008529663,
      "epoch": 1.1348749503771338,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000487960280084905,
      "loss": 5.1957,
      "mean_token_accuracy": 0.16630644202232361,
      "num_tokens": 25052010.0,
      "step": 11435
    },
    {
      "entropy": 5.596749496459961,
      "epoch": 1.1353711790393013,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004879488425403657,
      "loss": 5.2621,
      "mean_token_accuracy": 0.16078420653939246,
      "num_tokens": 25062209.0,
      "step": 11440
    },
    {
      "entropy": 5.523932552337646,
      "epoch": 1.1358674077014688,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048793739971515407,
      "loss": 5.2328,
      "mean_token_accuracy": 0.1643170580267906,
      "num_tokens": 25072758.0,
      "step": 11445
    },
    {
      "entropy": 5.562268018722534,
      "epoch": 1.1363636363636362,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00048792595160955374,
      "loss": 5.2846,
      "mean_token_accuracy": 0.16242774128913878,
      "num_tokens": 25084318.0,
      "step": 11450
    },
    {
      "entropy": 5.569331455230713,
      "epoch": 1.136859865025804,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048791449822384856,
      "loss": 5.1761,
      "mean_token_accuracy": 0.17167195975780486,
      "num_tokens": 25095715.0,
      "step": 11455
    },
    {
      "entropy": 5.564904832839966,
      "epoch": 1.1373560936879714,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004879030395583226,
      "loss": 5.2061,
      "mean_token_accuracy": 0.16821445673704147,
      "num_tokens": 25106163.0,
      "step": 11460
    },
    {
      "entropy": 5.5954526424407955,
      "epoch": 1.137852322350139,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004878915756132601,
      "loss": 5.2656,
      "mean_token_accuracy": 0.16299975514411927,
      "num_tokens": 25118745.0,
      "step": 11465
    },
    {
      "entropy": 5.56226954460144,
      "epoch": 1.1383485510123066,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048788010638894523,
      "loss": 5.2574,
      "mean_token_accuracy": 0.1571410268545151,
      "num_tokens": 25130066.0,
      "step": 11470
    },
    {
      "entropy": 5.53004937171936,
      "epoch": 1.138844779674474,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004878686318856624,
      "loss": 5.103,
      "mean_token_accuracy": 0.16741354912519454,
      "num_tokens": 25139895.0,
      "step": 11475
    },
    {
      "entropy": 5.627090311050415,
      "epoch": 1.1393410083366415,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048785715210369635,
      "loss": 5.3192,
      "mean_token_accuracy": 0.15703511238098145,
      "num_tokens": 25150644.0,
      "step": 11480
    },
    {
      "entropy": 5.579454851150513,
      "epoch": 1.139837236998809,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004878456670433314,
      "loss": 5.2362,
      "mean_token_accuracy": 0.1619833841919899,
      "num_tokens": 25163038.0,
      "step": 11485
    },
    {
      "entropy": 5.523625373840332,
      "epoch": 1.1403334656609765,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004878341767048527,
      "loss": 5.153,
      "mean_token_accuracy": 0.16821326911449433,
      "num_tokens": 25174154.0,
      "step": 11490
    },
    {
      "entropy": 5.583241987228393,
      "epoch": 1.1408296943231442,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048782268108854494,
      "loss": 5.2622,
      "mean_token_accuracy": 0.15985383093357086,
      "num_tokens": 25185369.0,
      "step": 11495
    },
    {
      "entropy": 5.542490816116333,
      "epoch": 1.1413259229853117,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004878111801946933,
      "loss": 5.2562,
      "mean_token_accuracy": 0.16156903952360152,
      "num_tokens": 25197875.0,
      "step": 11500
    },
    {
      "entropy": 5.468142938613892,
      "epoch": 1.1418221516474791,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048779967402358296,
      "loss": 5.1433,
      "mean_token_accuracy": 0.17784634679555894,
      "num_tokens": 25208362.0,
      "step": 11505
    },
    {
      "entropy": 5.606678915023804,
      "epoch": 1.1423183803096466,
      "grad_norm": 0.7578125,
      "learning_rate": 0.0004877881625754993,
      "loss": 5.3466,
      "mean_token_accuracy": 0.1564700610935688,
      "num_tokens": 25220441.0,
      "step": 11510
    },
    {
      "entropy": 5.545219135284424,
      "epoch": 1.1428146089718143,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00048777664585072765,
      "loss": 5.1138,
      "mean_token_accuracy": 0.165564888715744,
      "num_tokens": 25231134.0,
      "step": 11515
    },
    {
      "entropy": 5.636066436767578,
      "epoch": 1.1433108376339818,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004877651238495537,
      "loss": 5.3777,
      "mean_token_accuracy": 0.1574774593114853,
      "num_tokens": 25242327.0,
      "step": 11520
    },
    {
      "entropy": 5.56852765083313,
      "epoch": 1.1438070662961493,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048775359657226316,
      "loss": 5.1979,
      "mean_token_accuracy": 0.16594454795122146,
      "num_tokens": 25253571.0,
      "step": 11525
    },
    {
      "entropy": 5.497541189193726,
      "epoch": 1.1443032949583167,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004877420640191418,
      "loss": 5.1438,
      "mean_token_accuracy": 0.1686657264828682,
      "num_tokens": 25264573.0,
      "step": 11530
    },
    {
      "entropy": 5.549622297286987,
      "epoch": 1.1447995236204842,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004877305261904758,
      "loss": 5.235,
      "mean_token_accuracy": 0.1609615996479988,
      "num_tokens": 25275197.0,
      "step": 11535
    },
    {
      "entropy": 5.514175796508789,
      "epoch": 1.145295752282652,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048771898308655107,
      "loss": 5.1227,
      "mean_token_accuracy": 0.17143765240907669,
      "num_tokens": 25286564.0,
      "step": 11540
    },
    {
      "entropy": 5.566322278976441,
      "epoch": 1.1457919809448194,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004877074347076539,
      "loss": 5.164,
      "mean_token_accuracy": 0.16721051335334777,
      "num_tokens": 25297874.0,
      "step": 11545
    },
    {
      "entropy": 5.590831518173218,
      "epoch": 1.1462882096069869,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048769588105407075,
      "loss": 5.2417,
      "mean_token_accuracy": 0.1635787934064865,
      "num_tokens": 25308047.0,
      "step": 11550
    },
    {
      "entropy": 5.626078748703003,
      "epoch": 1.1467844382691545,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048768432212608807,
      "loss": 5.279,
      "mean_token_accuracy": 0.15482100397348403,
      "num_tokens": 25318527.0,
      "step": 11555
    },
    {
      "entropy": 5.576137542724609,
      "epoch": 1.147280666931322,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004876727579239925,
      "loss": 5.1988,
      "mean_token_accuracy": 0.17010349482297898,
      "num_tokens": 25328242.0,
      "step": 11560
    },
    {
      "entropy": 5.528227472305298,
      "epoch": 1.1477768955934895,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004876611884480708,
      "loss": 5.1689,
      "mean_token_accuracy": 0.16878188848495485,
      "num_tokens": 25338214.0,
      "step": 11565
    },
    {
      "entropy": 5.5525308609008786,
      "epoch": 1.148273124255657,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048764961369860977,
      "loss": 5.226,
      "mean_token_accuracy": 0.16078789979219438,
      "num_tokens": 25349762.0,
      "step": 11570
    },
    {
      "entropy": 5.550897598266602,
      "epoch": 1.1487693529178244,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048763803367589663,
      "loss": 5.2265,
      "mean_token_accuracy": 0.16498002856969834,
      "num_tokens": 25361241.0,
      "step": 11575
    },
    {
      "entropy": 5.57495379447937,
      "epoch": 1.1492655815799921,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048762644838021843,
      "loss": 5.2522,
      "mean_token_accuracy": 0.16064665466547012,
      "num_tokens": 25373846.0,
      "step": 11580
    },
    {
      "entropy": 5.532123661041259,
      "epoch": 1.1497618102421596,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048761485781186243,
      "loss": 5.1808,
      "mean_token_accuracy": 0.1658075273036957,
      "num_tokens": 25385331.0,
      "step": 11585
    },
    {
      "entropy": 5.523156261444091,
      "epoch": 1.150258038904327,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004876032619711162,
      "loss": 5.1933,
      "mean_token_accuracy": 0.17255775928497313,
      "num_tokens": 25395140.0,
      "step": 11590
    },
    {
      "entropy": 5.60550479888916,
      "epoch": 1.1507542675664946,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048759166085826714,
      "loss": 5.2353,
      "mean_token_accuracy": 0.15751099735498428,
      "num_tokens": 25405455.0,
      "step": 11595
    },
    {
      "entropy": 5.561492872238159,
      "epoch": 1.1512504962286623,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000487580054473603,
      "loss": 5.2428,
      "mean_token_accuracy": 0.1647414207458496,
      "num_tokens": 25417000.0,
      "step": 11600
    },
    {
      "entropy": 5.473997592926025,
      "epoch": 1.1517467248908297,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048756844281741156,
      "loss": 5.1643,
      "mean_token_accuracy": 0.1671366423368454,
      "num_tokens": 25427808.0,
      "step": 11605
    },
    {
      "entropy": 5.541671848297119,
      "epoch": 1.1522429535529972,
      "grad_norm": 0.859375,
      "learning_rate": 0.00048755682588998085,
      "loss": 5.196,
      "mean_token_accuracy": 0.15793447494506835,
      "num_tokens": 25438302.0,
      "step": 11610
    },
    {
      "entropy": 5.457865190505982,
      "epoch": 1.1527391822151647,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004875452036915987,
      "loss": 5.0917,
      "mean_token_accuracy": 0.16701629757881165,
      "num_tokens": 25449335.0,
      "step": 11615
    },
    {
      "entropy": 5.621621704101562,
      "epoch": 1.1532354108773322,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004875335762225536,
      "loss": 5.2763,
      "mean_token_accuracy": 0.15871182680130005,
      "num_tokens": 25461173.0,
      "step": 11620
    },
    {
      "entropy": 5.59500823020935,
      "epoch": 1.1537316395394999,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048752194348313383,
      "loss": 5.1749,
      "mean_token_accuracy": 0.16683193892240525,
      "num_tokens": 25472538.0,
      "step": 11625
    },
    {
      "entropy": 5.507734775543213,
      "epoch": 1.1542278682016673,
      "grad_norm": 0.875,
      "learning_rate": 0.0004875103054736278,
      "loss": 5.2141,
      "mean_token_accuracy": 0.16634899973869324,
      "num_tokens": 25483386.0,
      "step": 11630
    },
    {
      "entropy": 5.572064542770386,
      "epoch": 1.1547240968638348,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004874986621943241,
      "loss": 5.2256,
      "mean_token_accuracy": 0.1629209354519844,
      "num_tokens": 25495505.0,
      "step": 11635
    },
    {
      "entropy": 5.56484956741333,
      "epoch": 1.1552203255260023,
      "grad_norm": 0.859375,
      "learning_rate": 0.00048748701364551147,
      "loss": 5.2173,
      "mean_token_accuracy": 0.1683715894818306,
      "num_tokens": 25506144.0,
      "step": 11640
    },
    {
      "entropy": 5.5736268043518065,
      "epoch": 1.15571655418817,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004874753598274787,
      "loss": 5.2145,
      "mean_token_accuracy": 0.16028875559568406,
      "num_tokens": 25516185.0,
      "step": 11645
    },
    {
      "entropy": 5.6265936374664305,
      "epoch": 1.1562127828503375,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004874637007405149,
      "loss": 5.2912,
      "mean_token_accuracy": 0.16133199632167816,
      "num_tokens": 25526657.0,
      "step": 11650
    },
    {
      "entropy": 5.510714387893676,
      "epoch": 1.156709011512505,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00048745203638490917,
      "loss": 5.151,
      "mean_token_accuracy": 0.17050716876983643,
      "num_tokens": 25537101.0,
      "step": 11655
    },
    {
      "entropy": 5.5446388721466064,
      "epoch": 1.1572052401746724,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048744036676095067,
      "loss": 5.2395,
      "mean_token_accuracy": 0.1617955043911934,
      "num_tokens": 25547672.0,
      "step": 11660
    },
    {
      "entropy": 5.631325149536133,
      "epoch": 1.15770146883684,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004874286918689289,
      "loss": 5.2818,
      "mean_token_accuracy": 0.16751983612775803,
      "num_tokens": 25558408.0,
      "step": 11665
    },
    {
      "entropy": 5.627182912826538,
      "epoch": 1.1581976974990076,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048741701170913326,
      "loss": 5.1884,
      "mean_token_accuracy": 0.16304911375045777,
      "num_tokens": 25569823.0,
      "step": 11670
    },
    {
      "entropy": 5.518573474884033,
      "epoch": 1.158693926161175,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048740532628185353,
      "loss": 5.2043,
      "mean_token_accuracy": 0.16901672780513763,
      "num_tokens": 25579705.0,
      "step": 11675
    },
    {
      "entropy": 5.5250725746154785,
      "epoch": 1.1591901548233425,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004873936355873793,
      "loss": 5.2614,
      "mean_token_accuracy": 0.1624846026301384,
      "num_tokens": 25590573.0,
      "step": 11680
    },
    {
      "entropy": 5.531937789916992,
      "epoch": 1.1596863834855102,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004873819396260007,
      "loss": 5.182,
      "mean_token_accuracy": 0.16720510870218278,
      "num_tokens": 25601418.0,
      "step": 11685
    },
    {
      "entropy": 5.5733130931854244,
      "epoch": 1.1601826121476777,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048737023839800754,
      "loss": 5.2549,
      "mean_token_accuracy": 0.1668507844209671,
      "num_tokens": 25612409.0,
      "step": 11690
    },
    {
      "entropy": 5.512886333465576,
      "epoch": 1.1606788408098452,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004873585319036901,
      "loss": 5.1582,
      "mean_token_accuracy": 0.16629520952701568,
      "num_tokens": 25623547.0,
      "step": 11695
    },
    {
      "entropy": 5.586042976379394,
      "epoch": 1.1611750694720127,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004873468201433387,
      "loss": 5.3246,
      "mean_token_accuracy": 0.1602830931544304,
      "num_tokens": 25636537.0,
      "step": 11700
    },
    {
      "entropy": 5.556938934326172,
      "epoch": 1.1616712981341801,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004873351031172437,
      "loss": 5.1818,
      "mean_token_accuracy": 0.17011043876409532,
      "num_tokens": 25646734.0,
      "step": 11705
    },
    {
      "entropy": 5.6243524074554445,
      "epoch": 1.1621675267963478,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048732338082569564,
      "loss": 5.2705,
      "mean_token_accuracy": 0.1712836816906929,
      "num_tokens": 25658058.0,
      "step": 11710
    },
    {
      "entropy": 5.507769775390625,
      "epoch": 1.1626637554585153,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004873116532689853,
      "loss": 5.1042,
      "mean_token_accuracy": 0.16782689243555068,
      "num_tokens": 25669630.0,
      "step": 11715
    },
    {
      "entropy": 5.463340520858765,
      "epoch": 1.1631599841206828,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004872999204474035,
      "loss": 5.1264,
      "mean_token_accuracy": 0.16765623539686203,
      "num_tokens": 25681928.0,
      "step": 11720
    },
    {
      "entropy": 5.452698564529419,
      "epoch": 1.1636562127828503,
      "grad_norm": 1.046875,
      "learning_rate": 0.000487288182361241,
      "loss": 5.1365,
      "mean_token_accuracy": 0.16379121989011763,
      "num_tokens": 25690653.0,
      "step": 11725
    },
    {
      "entropy": 5.582673168182373,
      "epoch": 1.164152441445018,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004872764390107891,
      "loss": 5.2413,
      "mean_token_accuracy": 0.15922019481658936,
      "num_tokens": 25701201.0,
      "step": 11730
    },
    {
      "entropy": 5.500560283660889,
      "epoch": 1.1646486701071854,
      "grad_norm": 0.875,
      "learning_rate": 0.00048726469039633895,
      "loss": 5.1521,
      "mean_token_accuracy": 0.1659471109509468,
      "num_tokens": 25710927.0,
      "step": 11735
    },
    {
      "entropy": 5.54877290725708,
      "epoch": 1.165144898769353,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048725293651818183,
      "loss": 5.2084,
      "mean_token_accuracy": 0.16513376086950302,
      "num_tokens": 25722136.0,
      "step": 11740
    },
    {
      "entropy": 5.517477798461914,
      "epoch": 1.1656411274315204,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004872411773766093,
      "loss": 5.0852,
      "mean_token_accuracy": 0.1738223686814308,
      "num_tokens": 25733386.0,
      "step": 11745
    },
    {
      "entropy": 5.5081868171691895,
      "epoch": 1.1661373560936878,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048722941297191284,
      "loss": 5.1887,
      "mean_token_accuracy": 0.16425832211971284,
      "num_tokens": 25742801.0,
      "step": 11750
    },
    {
      "entropy": 5.607944631576538,
      "epoch": 1.1666335847558555,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004872176433043843,
      "loss": 5.2764,
      "mean_token_accuracy": 0.1629337713122368,
      "num_tokens": 25755594.0,
      "step": 11755
    },
    {
      "entropy": 5.5814227104187015,
      "epoch": 1.167129813418023,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004872058683743155,
      "loss": 5.2251,
      "mean_token_accuracy": 0.16832107603549956,
      "num_tokens": 25766271.0,
      "step": 11760
    },
    {
      "entropy": 5.5676782608032225,
      "epoch": 1.1676260420801905,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048719408818199836,
      "loss": 5.2576,
      "mean_token_accuracy": 0.1625947266817093,
      "num_tokens": 25777338.0,
      "step": 11765
    },
    {
      "entropy": 5.538753366470337,
      "epoch": 1.1681222707423582,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00048718230272772517,
      "loss": 5.1361,
      "mean_token_accuracy": 0.16893320381641388,
      "num_tokens": 25787968.0,
      "step": 11770
    },
    {
      "entropy": 5.568584632873535,
      "epoch": 1.1686184994045257,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00048717051201178804,
      "loss": 5.2143,
      "mean_token_accuracy": 0.16679543107748032,
      "num_tokens": 25798274.0,
      "step": 11775
    },
    {
      "entropy": 5.491327142715454,
      "epoch": 1.1691147280666931,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004871587160344794,
      "loss": 5.1753,
      "mean_token_accuracy": 0.15845755189657212,
      "num_tokens": 25809140.0,
      "step": 11780
    },
    {
      "entropy": 5.569497537612915,
      "epoch": 1.1696109567288606,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004871469147960917,
      "loss": 5.2178,
      "mean_token_accuracy": 0.16689679473638536,
      "num_tokens": 25821502.0,
      "step": 11785
    },
    {
      "entropy": 5.560224294662476,
      "epoch": 1.170107185391028,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004871351082969177,
      "loss": 5.1036,
      "mean_token_accuracy": 0.174156291782856,
      "num_tokens": 25833231.0,
      "step": 11790
    },
    {
      "entropy": 5.60383448600769,
      "epoch": 1.1706034140531958,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004871232965372501,
      "loss": 5.2728,
      "mean_token_accuracy": 0.16255935728549958,
      "num_tokens": 25844867.0,
      "step": 11795
    },
    {
      "entropy": 5.549636602401733,
      "epoch": 1.1710996427153633,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004871114795173819,
      "loss": 5.2436,
      "mean_token_accuracy": 0.1703019618988037,
      "num_tokens": 25855959.0,
      "step": 11800
    },
    {
      "entropy": 5.57716703414917,
      "epoch": 1.1715958713775307,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004870996572376061,
      "loss": 5.3213,
      "mean_token_accuracy": 0.16501019299030303,
      "num_tokens": 25866926.0,
      "step": 11805
    },
    {
      "entropy": 5.618939924240112,
      "epoch": 1.1720921000396982,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004870878296982157,
      "loss": 5.254,
      "mean_token_accuracy": 0.15923433899879455,
      "num_tokens": 25877699.0,
      "step": 11810
    },
    {
      "entropy": 5.629858064651489,
      "epoch": 1.172588328701866,
      "grad_norm": 1.0,
      "learning_rate": 0.0004870759968995042,
      "loss": 5.2138,
      "mean_token_accuracy": 0.16843907088041304,
      "num_tokens": 25888640.0,
      "step": 11815
    },
    {
      "entropy": 5.577797985076904,
      "epoch": 1.1730845573640334,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00048706415884176493,
      "loss": 5.2014,
      "mean_token_accuracy": 0.16519638150930405,
      "num_tokens": 25900651.0,
      "step": 11820
    },
    {
      "entropy": 5.528300046920776,
      "epoch": 1.1735807860262009,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004870523155252915,
      "loss": 5.1725,
      "mean_token_accuracy": 0.16927069127559663,
      "num_tokens": 25911775.0,
      "step": 11825
    },
    {
      "entropy": 5.530733680725097,
      "epoch": 1.1740770146883683,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004870404669503776,
      "loss": 5.1528,
      "mean_token_accuracy": 0.16928931176662446,
      "num_tokens": 25922126.0,
      "step": 11830
    },
    {
      "entropy": 5.583652687072754,
      "epoch": 1.1745732433505358,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000487028613117317,
      "loss": 5.2067,
      "mean_token_accuracy": 0.16576076596975325,
      "num_tokens": 25932945.0,
      "step": 11835
    },
    {
      "entropy": 5.597814893722534,
      "epoch": 1.1750694720127035,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004870167540264037,
      "loss": 5.257,
      "mean_token_accuracy": 0.16133045703172683,
      "num_tokens": 25944277.0,
      "step": 11840
    },
    {
      "entropy": 5.496471834182739,
      "epoch": 1.175565700674871,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004870048896779317,
      "loss": 5.1171,
      "mean_token_accuracy": 0.17152974456548692,
      "num_tokens": 25956838.0,
      "step": 11845
    },
    {
      "entropy": 5.537053728103638,
      "epoch": 1.1760619293370385,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00048699302007219533,
      "loss": 5.1889,
      "mean_token_accuracy": 0.16208528131246566,
      "num_tokens": 25966639.0,
      "step": 11850
    },
    {
      "entropy": 5.531539678573608,
      "epoch": 1.1765581579992062,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004869811452094889,
      "loss": 5.2242,
      "mean_token_accuracy": 0.17045587301254272,
      "num_tokens": 25978209.0,
      "step": 11855
    },
    {
      "entropy": 5.51884183883667,
      "epoch": 1.1770543866613736,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048696926509010673,
      "loss": 5.1486,
      "mean_token_accuracy": 0.16951389610767365,
      "num_tokens": 25989306.0,
      "step": 11860
    },
    {
      "entropy": 5.5544044971466064,
      "epoch": 1.177550615323541,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004869573797143437,
      "loss": 5.2569,
      "mean_token_accuracy": 0.16379437744617462,
      "num_tokens": 26000235.0,
      "step": 11865
    },
    {
      "entropy": 5.524611091613769,
      "epoch": 1.1780468439857086,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004869454890824942,
      "loss": 5.1549,
      "mean_token_accuracy": 0.17138529419898987,
      "num_tokens": 26010616.0,
      "step": 11870
    },
    {
      "entropy": 5.553587245941162,
      "epoch": 1.178543072647876,
      "grad_norm": 0.83203125,
      "learning_rate": 0.00048693359319485337,
      "loss": 5.2271,
      "mean_token_accuracy": 0.1586771696805954,
      "num_tokens": 26021303.0,
      "step": 11875
    },
    {
      "entropy": 5.5444655418396,
      "epoch": 1.1790393013100438,
      "grad_norm": 0.875,
      "learning_rate": 0.0004869216920517161,
      "loss": 5.1621,
      "mean_token_accuracy": 0.16711566001176834,
      "num_tokens": 26032386.0,
      "step": 11880
    },
    {
      "entropy": 5.468569278717041,
      "epoch": 1.1795355299722112,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048690978565337753,
      "loss": 5.1296,
      "mean_token_accuracy": 0.1663554824888706,
      "num_tokens": 26042889.0,
      "step": 11885
    },
    {
      "entropy": 5.5141987800598145,
      "epoch": 1.1800317586343787,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048689787400013287,
      "loss": 5.2313,
      "mean_token_accuracy": 0.16644604802131652,
      "num_tokens": 26054299.0,
      "step": 11890
    },
    {
      "entropy": 5.612095355987549,
      "epoch": 1.1805279872965462,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004868859570922775,
      "loss": 5.2642,
      "mean_token_accuracy": 0.1618034929037094,
      "num_tokens": 26064587.0,
      "step": 11895
    },
    {
      "entropy": 5.624295282363891,
      "epoch": 1.1810242159587139,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004868740349301071,
      "loss": 5.1959,
      "mean_token_accuracy": 0.17469189912080765,
      "num_tokens": 26076495.0,
      "step": 11900
    },
    {
      "entropy": 5.5879435539245605,
      "epoch": 1.1815204446208813,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000486862107513917,
      "loss": 5.2002,
      "mean_token_accuracy": 0.16503895819187164,
      "num_tokens": 26086496.0,
      "step": 11905
    },
    {
      "entropy": 5.506546354293823,
      "epoch": 1.1820166732830488,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004868501748440033,
      "loss": 5.1837,
      "mean_token_accuracy": 0.16451774090528487,
      "num_tokens": 26097263.0,
      "step": 11910
    },
    {
      "entropy": 5.560457181930542,
      "epoch": 1.1825129019452163,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004868382369206617,
      "loss": 5.1795,
      "mean_token_accuracy": 0.16694652587175368,
      "num_tokens": 26108264.0,
      "step": 11915
    },
    {
      "entropy": 5.566349029541016,
      "epoch": 1.1830091306073838,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004868262937441883,
      "loss": 5.2521,
      "mean_token_accuracy": 0.17175290882587432,
      "num_tokens": 26119730.0,
      "step": 11920
    },
    {
      "entropy": 5.5816343307495115,
      "epoch": 1.1835053592695515,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00048681434531487925,
      "loss": 5.1533,
      "mean_token_accuracy": 0.16655018478631972,
      "num_tokens": 26131055.0,
      "step": 11925
    },
    {
      "entropy": 5.500956201553345,
      "epoch": 1.184001587931719,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048680239163303085,
      "loss": 5.1596,
      "mean_token_accuracy": 0.1692327693104744,
      "num_tokens": 26141577.0,
      "step": 11930
    },
    {
      "entropy": 5.512574529647827,
      "epoch": 1.1844978165938864,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004867904326989396,
      "loss": 5.2005,
      "mean_token_accuracy": 0.17093419134616852,
      "num_tokens": 26152733.0,
      "step": 11935
    },
    {
      "entropy": 5.577522802352905,
      "epoch": 1.184994045256054,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004867784685129019,
      "loss": 5.2401,
      "mean_token_accuracy": 0.16454808712005614,
      "num_tokens": 26164763.0,
      "step": 11940
    },
    {
      "entropy": 5.536622810363769,
      "epoch": 1.1854902739182216,
      "grad_norm": 0.859375,
      "learning_rate": 0.00048676649907521455,
      "loss": 5.1186,
      "mean_token_accuracy": 0.17210778146982192,
      "num_tokens": 26174391.0,
      "step": 11945
    },
    {
      "entropy": 5.442165374755859,
      "epoch": 1.185986502580389,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048675452438617434,
      "loss": 5.0972,
      "mean_token_accuracy": 0.1761639818549156,
      "num_tokens": 26184772.0,
      "step": 11950
    },
    {
      "entropy": 5.600083541870117,
      "epoch": 1.1864827312425565,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004867425444460782,
      "loss": 5.2579,
      "mean_token_accuracy": 0.15648220777511596,
      "num_tokens": 26195411.0,
      "step": 11955
    },
    {
      "entropy": 5.521206712722778,
      "epoch": 1.186978959904724,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004867305592552233,
      "loss": 5.1787,
      "mean_token_accuracy": 0.1663449838757515,
      "num_tokens": 26206719.0,
      "step": 11960
    },
    {
      "entropy": 5.496636438369751,
      "epoch": 1.1874751885668917,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004867185688139067,
      "loss": 5.122,
      "mean_token_accuracy": 0.1714530736207962,
      "num_tokens": 26217469.0,
      "step": 11965
    },
    {
      "entropy": 5.461162996292114,
      "epoch": 1.1879714172290592,
      "grad_norm": 1.0,
      "learning_rate": 0.0004867065731224258,
      "loss": 5.1429,
      "mean_token_accuracy": 0.16440949887037276,
      "num_tokens": 26228556.0,
      "step": 11970
    },
    {
      "entropy": 5.5642680644989015,
      "epoch": 1.1884676458912267,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004866945721810781,
      "loss": 5.2463,
      "mean_token_accuracy": 0.16796668469905854,
      "num_tokens": 26239891.0,
      "step": 11975
    },
    {
      "entropy": 5.570119047164917,
      "epoch": 1.1889638745533941,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004866825659901612,
      "loss": 5.1992,
      "mean_token_accuracy": 0.17129147946834564,
      "num_tokens": 26251793.0,
      "step": 11980
    },
    {
      "entropy": 5.473042297363281,
      "epoch": 1.1894601032155618,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048667055454997276,
      "loss": 5.0967,
      "mean_token_accuracy": 0.16642365604639053,
      "num_tokens": 26262367.0,
      "step": 11985
    },
    {
      "entropy": 5.500068712234497,
      "epoch": 1.1899563318777293,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048665853786081063,
      "loss": 5.1787,
      "mean_token_accuracy": 0.1641027122735977,
      "num_tokens": 26272681.0,
      "step": 11990
    },
    {
      "entropy": 5.54545578956604,
      "epoch": 1.1904525605398968,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048664651592297294,
      "loss": 5.249,
      "mean_token_accuracy": 0.1640736386179924,
      "num_tokens": 26281194.0,
      "step": 11995
    },
    {
      "entropy": 5.564527750015259,
      "epoch": 1.1909487892020643,
      "grad_norm": 0.859375,
      "learning_rate": 0.00048663448873675766,
      "loss": 5.1344,
      "mean_token_accuracy": 0.1650382623076439,
      "num_tokens": 26292682.0,
      "step": 12000
    },
    {
      "epoch": 1.1909487892020643,
      "eval_entropy": 5.415048342754847,
      "eval_loss": 5.320173740386963,
      "eval_mean_token_accuracy": 0.1656518037897599,
      "eval_num_tokens": 26292682.0,
      "eval_runtime": 20.3903,
      "eval_samples_per_second": 1651.228,
      "eval_steps_per_second": 206.422,
      "step": 12000
    },
    {
      "entropy": 5.578801870346069,
      "epoch": 1.1914450178642317,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004866224563024631,
      "loss": 5.2472,
      "mean_token_accuracy": 0.15969247370958328,
      "num_tokens": 26305500.0,
      "step": 12005
    },
    {
      "entropy": 5.550313234329224,
      "epoch": 1.1919412465263994,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004866104186203877,
      "loss": 5.144,
      "mean_token_accuracy": 0.16529141068458558,
      "num_tokens": 26317207.0,
      "step": 12010
    },
    {
      "entropy": 5.530046606063843,
      "epoch": 1.192437475188567,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048659837569082977,
      "loss": 5.1586,
      "mean_token_accuracy": 0.16852137297391892,
      "num_tokens": 26328728.0,
      "step": 12015
    },
    {
      "entropy": 5.606966304779053,
      "epoch": 1.1929337038507344,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004865863275140882,
      "loss": 5.1997,
      "mean_token_accuracy": 0.1659929111599922,
      "num_tokens": 26338622.0,
      "step": 12020
    },
    {
      "entropy": 5.561013460159302,
      "epoch": 1.1934299325129019,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004865742740904615,
      "loss": 5.1836,
      "mean_token_accuracy": 0.16556894183158874,
      "num_tokens": 26348807.0,
      "step": 12025
    },
    {
      "entropy": 5.472616577148438,
      "epoch": 1.1939261611750696,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004865622154202488,
      "loss": 5.15,
      "mean_token_accuracy": 0.16642577350139617,
      "num_tokens": 26358481.0,
      "step": 12030
    },
    {
      "entropy": 5.525966596603394,
      "epoch": 1.194422389837237,
      "grad_norm": 0.859375,
      "learning_rate": 0.00048655015150374904,
      "loss": 5.2154,
      "mean_token_accuracy": 0.1611705929040909,
      "num_tokens": 26370519.0,
      "step": 12035
    },
    {
      "entropy": 5.602569627761841,
      "epoch": 1.1949186184994045,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004865380823412613,
      "loss": 5.2153,
      "mean_token_accuracy": 0.16661960780620574,
      "num_tokens": 26380988.0,
      "step": 12040
    },
    {
      "entropy": 5.512216520309448,
      "epoch": 1.195414847161572,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000486526007933085,
      "loss": 5.1106,
      "mean_token_accuracy": 0.17212731540203094,
      "num_tokens": 26390308.0,
      "step": 12045
    },
    {
      "entropy": 5.556072664260864,
      "epoch": 1.1959110758237395,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048651392827951944,
      "loss": 5.2305,
      "mean_token_accuracy": 0.15660665333271026,
      "num_tokens": 26401145.0,
      "step": 12050
    },
    {
      "entropy": 5.550975894927978,
      "epoch": 1.1964073044859072,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00048650184338086436,
      "loss": 5.1638,
      "mean_token_accuracy": 0.168052639067173,
      "num_tokens": 26411930.0,
      "step": 12055
    },
    {
      "entropy": 5.516152286529541,
      "epoch": 1.1969035331480746,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048648975323741916,
      "loss": 5.1122,
      "mean_token_accuracy": 0.16455964744091034,
      "num_tokens": 26422320.0,
      "step": 12060
    },
    {
      "entropy": 5.615703058242798,
      "epoch": 1.197399761810242,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048647765784948395,
      "loss": 5.2378,
      "mean_token_accuracy": 0.16010700464248656,
      "num_tokens": 26433760.0,
      "step": 12065
    },
    {
      "entropy": 5.595371055603027,
      "epoch": 1.1978959904724098,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004864655572173584,
      "loss": 5.1572,
      "mean_token_accuracy": 0.1727421760559082,
      "num_tokens": 26444256.0,
      "step": 12070
    },
    {
      "entropy": 5.536590480804444,
      "epoch": 1.1983922191345773,
      "grad_norm": 0.84375,
      "learning_rate": 0.00048645345134134274,
      "loss": 5.2511,
      "mean_token_accuracy": 0.16822822988033295,
      "num_tokens": 26456433.0,
      "step": 12075
    },
    {
      "entropy": 5.569968271255493,
      "epoch": 1.1988884477967447,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048644134022173707,
      "loss": 5.2169,
      "mean_token_accuracy": 0.16848678588867189,
      "num_tokens": 26466909.0,
      "step": 12080
    },
    {
      "entropy": 5.589114999771118,
      "epoch": 1.1993846764589122,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004864292238588419,
      "loss": 5.2187,
      "mean_token_accuracy": 0.1713562101125717,
      "num_tokens": 26478785.0,
      "step": 12085
    },
    {
      "entropy": 5.500128746032715,
      "epoch": 1.1998809051210797,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048641710225295746,
      "loss": 5.2303,
      "mean_token_accuracy": 0.1657616525888443,
      "num_tokens": 26489596.0,
      "step": 12090
    },
    {
      "entropy": 5.528736543655396,
      "epoch": 1.2003771337832474,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048640497540438455,
      "loss": 5.1644,
      "mean_token_accuracy": 0.16949445754289627,
      "num_tokens": 26500376.0,
      "step": 12095
    },
    {
      "entropy": 5.508809995651245,
      "epoch": 1.2008733624454149,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004863928433134237,
      "loss": 5.0949,
      "mean_token_accuracy": 0.17248708307743071,
      "num_tokens": 26510339.0,
      "step": 12100
    },
    {
      "entropy": 5.55097508430481,
      "epoch": 1.2013695911075823,
      "grad_norm": 0.8359375,
      "learning_rate": 0.00048638070598037585,
      "loss": 5.1962,
      "mean_token_accuracy": 0.1656286373734474,
      "num_tokens": 26521359.0,
      "step": 12105
    },
    {
      "entropy": 5.607142925262451,
      "epoch": 1.2018658197697498,
      "grad_norm": 1.0,
      "learning_rate": 0.000486368563405542,
      "loss": 5.2709,
      "mean_token_accuracy": 0.16035255342721938,
      "num_tokens": 26532716.0,
      "step": 12110
    },
    {
      "entropy": 5.508362054824829,
      "epoch": 1.2023620484319175,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004863564155892232,
      "loss": 5.106,
      "mean_token_accuracy": 0.16817451715469361,
      "num_tokens": 26542624.0,
      "step": 12115
    },
    {
      "entropy": 5.583513975143433,
      "epoch": 1.202858277094085,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004863442625317208,
      "loss": 5.2313,
      "mean_token_accuracy": 0.16890258938074112,
      "num_tokens": 26552907.0,
      "step": 12120
    },
    {
      "entropy": 5.495972442626953,
      "epoch": 1.2033545057562525,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048633210423333607,
      "loss": 5.1739,
      "mean_token_accuracy": 0.16416290402412415,
      "num_tokens": 26564871.0,
      "step": 12125
    },
    {
      "entropy": 5.5188898086547855,
      "epoch": 1.20385073441842,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004863199406943705,
      "loss": 5.1736,
      "mean_token_accuracy": 0.16844612807035447,
      "num_tokens": 26576254.0,
      "step": 12130
    },
    {
      "entropy": 5.581023836135865,
      "epoch": 1.2043469630805874,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048630777191512584,
      "loss": 5.1916,
      "mean_token_accuracy": 0.17156504690647126,
      "num_tokens": 26587518.0,
      "step": 12135
    },
    {
      "entropy": 5.507179260253906,
      "epoch": 1.2048431917427551,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048629559789590366,
      "loss": 5.103,
      "mean_token_accuracy": 0.1670466184616089,
      "num_tokens": 26597485.0,
      "step": 12140
    },
    {
      "entropy": 5.483174276351929,
      "epoch": 1.2053394204049226,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000486283418637006,
      "loss": 5.1563,
      "mean_token_accuracy": 0.16611899733543395,
      "num_tokens": 26607696.0,
      "step": 12145
    },
    {
      "entropy": 5.561501216888428,
      "epoch": 1.20583564906709,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004862712341387348,
      "loss": 5.1882,
      "mean_token_accuracy": 0.16724651902914048,
      "num_tokens": 26618974.0,
      "step": 12150
    },
    {
      "entropy": 5.596035289764404,
      "epoch": 1.2063318777292578,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004862590444013923,
      "loss": 5.2305,
      "mean_token_accuracy": 0.16883024722337722,
      "num_tokens": 26629949.0,
      "step": 12155
    },
    {
      "entropy": 5.566052055358886,
      "epoch": 1.2068281063914252,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004862468494252806,
      "loss": 5.1598,
      "mean_token_accuracy": 0.16777434647083284,
      "num_tokens": 26641460.0,
      "step": 12160
    },
    {
      "entropy": 5.5983438968658445,
      "epoch": 1.2073243350535927,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004862346492107024,
      "loss": 5.2552,
      "mean_token_accuracy": 0.16443577408790588,
      "num_tokens": 26652553.0,
      "step": 12165
    },
    {
      "entropy": 5.518207979202271,
      "epoch": 1.2078205637157602,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048622244375796,
      "loss": 5.1733,
      "mean_token_accuracy": 0.16214610785245895,
      "num_tokens": 26662479.0,
      "step": 12170
    },
    {
      "entropy": 5.574397182464599,
      "epoch": 1.2083167923779277,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004862102330673561,
      "loss": 5.2118,
      "mean_token_accuracy": 0.16284110248088837,
      "num_tokens": 26673326.0,
      "step": 12175
    },
    {
      "entropy": 5.598153114318848,
      "epoch": 1.2088130210400954,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048619801713919357,
      "loss": 5.2643,
      "mean_token_accuracy": 0.159586501121521,
      "num_tokens": 26684189.0,
      "step": 12180
    },
    {
      "entropy": 5.6394611358642575,
      "epoch": 1.2093092497022628,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004861857959737753,
      "loss": 5.2794,
      "mean_token_accuracy": 0.16381492763757705,
      "num_tokens": 26695399.0,
      "step": 12185
    },
    {
      "entropy": 5.595186042785644,
      "epoch": 1.2098054783644303,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004861735695714044,
      "loss": 5.2092,
      "mean_token_accuracy": 0.1677875965833664,
      "num_tokens": 26705993.0,
      "step": 12190
    },
    {
      "entropy": 5.562430095672608,
      "epoch": 1.2103017070265978,
      "grad_norm": 1.0,
      "learning_rate": 0.00048616133793238394,
      "loss": 5.1838,
      "mean_token_accuracy": 0.1704581633210182,
      "num_tokens": 26716120.0,
      "step": 12195
    },
    {
      "entropy": 5.577006244659424,
      "epoch": 1.2107979356887655,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048614910105701737,
      "loss": 5.3138,
      "mean_token_accuracy": 0.15813263058662413,
      "num_tokens": 26727736.0,
      "step": 12200
    },
    {
      "entropy": 5.603042507171631,
      "epoch": 1.211294164350933,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004861368589456081,
      "loss": 5.1655,
      "mean_token_accuracy": 0.1643623873591423,
      "num_tokens": 26737825.0,
      "step": 12205
    },
    {
      "entropy": 5.643135738372803,
      "epoch": 1.2117903930131004,
      "grad_norm": 1.0,
      "learning_rate": 0.0004861246115984596,
      "loss": 5.2516,
      "mean_token_accuracy": 0.15876631587743759,
      "num_tokens": 26748035.0,
      "step": 12210
    },
    {
      "entropy": 5.604583644866944,
      "epoch": 1.212286621675268,
      "grad_norm": 0.80078125,
      "learning_rate": 0.00048611235901587575,
      "loss": 5.1746,
      "mean_token_accuracy": 0.1688556954264641,
      "num_tokens": 26759899.0,
      "step": 12215
    },
    {
      "entropy": 5.462530183792114,
      "epoch": 1.2127828503374354,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004861001011981602,
      "loss": 5.1236,
      "mean_token_accuracy": 0.17259134650230407,
      "num_tokens": 26771562.0,
      "step": 12220
    },
    {
      "entropy": 5.569944047927857,
      "epoch": 1.213279078999603,
      "grad_norm": 0.875,
      "learning_rate": 0.0004860878381456171,
      "loss": 5.2354,
      "mean_token_accuracy": 0.16821920722723008,
      "num_tokens": 26783002.0,
      "step": 12225
    },
    {
      "entropy": 5.575166654586792,
      "epoch": 1.2137753076617706,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048607556985855047,
      "loss": 5.2314,
      "mean_token_accuracy": 0.1648027256131172,
      "num_tokens": 26794136.0,
      "step": 12230
    },
    {
      "entropy": 5.460007524490356,
      "epoch": 1.214271536323938,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048606329633726447,
      "loss": 5.2025,
      "mean_token_accuracy": 0.16251839250326156,
      "num_tokens": 26804776.0,
      "step": 12235
    },
    {
      "entropy": 5.502743482589722,
      "epoch": 1.2147677649861055,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004860510175820636,
      "loss": 5.1362,
      "mean_token_accuracy": 0.16850659400224685,
      "num_tokens": 26816696.0,
      "step": 12240
    },
    {
      "entropy": 5.48726978302002,
      "epoch": 1.2152639936482732,
      "grad_norm": 0.875,
      "learning_rate": 0.0004860387335932522,
      "loss": 5.1801,
      "mean_token_accuracy": 0.1710625782608986,
      "num_tokens": 26827789.0,
      "step": 12245
    },
    {
      "entropy": 5.572609901428223,
      "epoch": 1.2157602223104407,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004860264443711351,
      "loss": 5.135,
      "mean_token_accuracy": 0.16863904297351837,
      "num_tokens": 26838763.0,
      "step": 12250
    },
    {
      "entropy": 5.502274799346924,
      "epoch": 1.2162564509726081,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004860141499160167,
      "loss": 5.1693,
      "mean_token_accuracy": 0.1614042803645134,
      "num_tokens": 26849304.0,
      "step": 12255
    },
    {
      "entropy": 5.53492226600647,
      "epoch": 1.2167526796347756,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004860018502282022,
      "loss": 5.2349,
      "mean_token_accuracy": 0.16254792138934135,
      "num_tokens": 26858777.0,
      "step": 12260
    },
    {
      "entropy": 5.595277976989746,
      "epoch": 1.2172489082969433,
      "grad_norm": 0.828125,
      "learning_rate": 0.00048598954530799644,
      "loss": 5.1879,
      "mean_token_accuracy": 0.16661825478076936,
      "num_tokens": 26870286.0,
      "step": 12265
    },
    {
      "entropy": 5.582540559768677,
      "epoch": 1.2177451369591108,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048597723515570465,
      "loss": 5.1736,
      "mean_token_accuracy": 0.1679397478699684,
      "num_tokens": 26880485.0,
      "step": 12270
    },
    {
      "entropy": 5.436429119110107,
      "epoch": 1.2182413656212783,
      "grad_norm": 0.81640625,
      "learning_rate": 0.000485964919771632,
      "loss": 5.0799,
      "mean_token_accuracy": 0.1702420324087143,
      "num_tokens": 26893556.0,
      "step": 12275
    },
    {
      "entropy": 5.519435024261474,
      "epoch": 1.2187375942834457,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048595259915608396,
      "loss": 5.1668,
      "mean_token_accuracy": 0.16302407830953597,
      "num_tokens": 26905699.0,
      "step": 12280
    },
    {
      "entropy": 5.570646667480469,
      "epoch": 1.2192338229456134,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048594027330936594,
      "loss": 5.1946,
      "mean_token_accuracy": 0.16270232051610947,
      "num_tokens": 26917188.0,
      "step": 12285
    },
    {
      "entropy": 5.552227020263672,
      "epoch": 1.219730051607781,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00048592794223178374,
      "loss": 5.1808,
      "mean_token_accuracy": 0.1670808508992195,
      "num_tokens": 26927905.0,
      "step": 12290
    },
    {
      "entropy": 5.465819072723389,
      "epoch": 1.2202262802699484,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048591560592364314,
      "loss": 5.1508,
      "mean_token_accuracy": 0.16758487820625306,
      "num_tokens": 26938665.0,
      "step": 12295
    },
    {
      "entropy": 5.534365844726563,
      "epoch": 1.2207225089321159,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004859032643852499,
      "loss": 5.253,
      "mean_token_accuracy": 0.16436808854341506,
      "num_tokens": 26950516.0,
      "step": 12300
    },
    {
      "entropy": 5.635159540176391,
      "epoch": 1.2212187375942833,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004858909176169102,
      "loss": 5.2465,
      "mean_token_accuracy": 0.17042109072208406,
      "num_tokens": 26960018.0,
      "step": 12305
    },
    {
      "entropy": 5.563248777389527,
      "epoch": 1.221714966256451,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048587856561893014,
      "loss": 5.1954,
      "mean_token_accuracy": 0.16420393139123918,
      "num_tokens": 26969825.0,
      "step": 12310
    },
    {
      "entropy": 5.587969255447388,
      "epoch": 1.2222111949186185,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004858662083916161,
      "loss": 5.2928,
      "mean_token_accuracy": 0.1615210860967636,
      "num_tokens": 26979873.0,
      "step": 12315
    },
    {
      "entropy": 5.62132625579834,
      "epoch": 1.222707423580786,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004858538459352744,
      "loss": 5.2605,
      "mean_token_accuracy": 0.16628895476460456,
      "num_tokens": 26990577.0,
      "step": 12320
    },
    {
      "entropy": 5.524898433685303,
      "epoch": 1.2232036522429535,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004858414782502117,
      "loss": 5.117,
      "mean_token_accuracy": 0.17091183364391327,
      "num_tokens": 27001598.0,
      "step": 12325
    },
    {
      "entropy": 5.49240665435791,
      "epoch": 1.2236998809051212,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048582910533673474,
      "loss": 5.1291,
      "mean_token_accuracy": 0.17308679819107056,
      "num_tokens": 27012535.0,
      "step": 12330
    },
    {
      "entropy": 5.531280088424682,
      "epoch": 1.2241961095672886,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004858167271951502,
      "loss": 5.2008,
      "mean_token_accuracy": 0.1611105963587761,
      "num_tokens": 27021867.0,
      "step": 12335
    },
    {
      "entropy": 5.474882888793945,
      "epoch": 1.2246923382294561,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004858043438257651,
      "loss": 5.1619,
      "mean_token_accuracy": 0.16459438651800157,
      "num_tokens": 27032540.0,
      "step": 12340
    },
    {
      "entropy": 5.458321809768677,
      "epoch": 1.2251885668916236,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004857919552288865,
      "loss": 5.1431,
      "mean_token_accuracy": 0.16965974122285843,
      "num_tokens": 27044008.0,
      "step": 12345
    },
    {
      "entropy": 5.583375597000122,
      "epoch": 1.225684795553791,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004857795614048217,
      "loss": 5.1638,
      "mean_token_accuracy": 0.1659913569688797,
      "num_tokens": 27053646.0,
      "step": 12350
    },
    {
      "entropy": 5.602629709243774,
      "epoch": 1.2261810242159588,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048576716235387786,
      "loss": 5.2772,
      "mean_token_accuracy": 0.16023823320865632,
      "num_tokens": 27065834.0,
      "step": 12355
    },
    {
      "entropy": 5.517317819595337,
      "epoch": 1.2266772528781262,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048575475807636267,
      "loss": 5.2164,
      "mean_token_accuracy": 0.16655384600162507,
      "num_tokens": 27077544.0,
      "step": 12360
    },
    {
      "entropy": 5.592941522598267,
      "epoch": 1.2271734815402937,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048574234857258356,
      "loss": 5.2657,
      "mean_token_accuracy": 0.1652846321463585,
      "num_tokens": 27088294.0,
      "step": 12365
    },
    {
      "entropy": 5.559949636459351,
      "epoch": 1.2276697102024614,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048572993384284833,
      "loss": 5.1976,
      "mean_token_accuracy": 0.16609010398387908,
      "num_tokens": 27098848.0,
      "step": 12370
    },
    {
      "entropy": 5.482949447631836,
      "epoch": 1.2281659388646289,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004857175138874648,
      "loss": 5.1588,
      "mean_token_accuracy": 0.1705593392252922,
      "num_tokens": 27112319.0,
      "step": 12375
    },
    {
      "entropy": 5.582285785675049,
      "epoch": 1.2286621675267964,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048570508870674097,
      "loss": 5.2312,
      "mean_token_accuracy": 0.1679324045777321,
      "num_tokens": 27121840.0,
      "step": 12380
    },
    {
      "entropy": 5.583931303024292,
      "epoch": 1.2291583961889638,
      "grad_norm": 0.8359375,
      "learning_rate": 0.00048569265830098493,
      "loss": 5.2412,
      "mean_token_accuracy": 0.16400841474533082,
      "num_tokens": 27133619.0,
      "step": 12385
    },
    {
      "entropy": 5.539572668075562,
      "epoch": 1.2296546248511313,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000485680222670505,
      "loss": 5.2212,
      "mean_token_accuracy": 0.1619301199913025,
      "num_tokens": 27144161.0,
      "step": 12390
    },
    {
      "entropy": 5.528607940673828,
      "epoch": 1.230150853513299,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004856677818156096,
      "loss": 5.1353,
      "mean_token_accuracy": 0.1702260345220566,
      "num_tokens": 27154906.0,
      "step": 12395
    },
    {
      "entropy": 5.52781310081482,
      "epoch": 1.2306470821754665,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048565533573660707,
      "loss": 5.16,
      "mean_token_accuracy": 0.16407609432935716,
      "num_tokens": 27165746.0,
      "step": 12400
    },
    {
      "entropy": 5.500742769241333,
      "epoch": 1.231143310837634,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004856428844338062,
      "loss": 5.1437,
      "mean_token_accuracy": 0.17028856575489043,
      "num_tokens": 27176108.0,
      "step": 12405
    },
    {
      "entropy": 5.5010072708129885,
      "epoch": 1.2316395394998014,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004856304279075156,
      "loss": 5.2306,
      "mean_token_accuracy": 0.16487469524145126,
      "num_tokens": 27186721.0,
      "step": 12410
    },
    {
      "entropy": 5.525478410720825,
      "epoch": 1.2321357681619691,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004856179661580444,
      "loss": 5.0979,
      "mean_token_accuracy": 0.17067277580499648,
      "num_tokens": 27196120.0,
      "step": 12415
    },
    {
      "entropy": 5.60766110420227,
      "epoch": 1.2326319968241366,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004856054991857013,
      "loss": 5.233,
      "mean_token_accuracy": 0.15933823734521865,
      "num_tokens": 27207431.0,
      "step": 12420
    },
    {
      "entropy": 5.519527626037598,
      "epoch": 1.233128225486304,
      "grad_norm": 0.78515625,
      "learning_rate": 0.00048559302699079576,
      "loss": 5.1169,
      "mean_token_accuracy": 0.17377728521823882,
      "num_tokens": 27218489.0,
      "step": 12425
    },
    {
      "entropy": 5.568860244750977,
      "epoch": 1.2336244541484715,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004855805495736369,
      "loss": 5.2375,
      "mean_token_accuracy": 0.16518362909555434,
      "num_tokens": 27230849.0,
      "step": 12430
    },
    {
      "entropy": 5.56860466003418,
      "epoch": 1.234120682810639,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004855680669345342,
      "loss": 5.1956,
      "mean_token_accuracy": 0.16450659632682801,
      "num_tokens": 27242838.0,
      "step": 12435
    },
    {
      "entropy": 5.55742564201355,
      "epoch": 1.2346169114728067,
      "grad_norm": 0.84375,
      "learning_rate": 0.00048555557907379714,
      "loss": 5.2193,
      "mean_token_accuracy": 0.1727436289191246,
      "num_tokens": 27253945.0,
      "step": 12440
    },
    {
      "entropy": 5.56047625541687,
      "epoch": 1.2351131401349742,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048554308599173544,
      "loss": 5.2804,
      "mean_token_accuracy": 0.16149513870477678,
      "num_tokens": 27263999.0,
      "step": 12445
    },
    {
      "entropy": 5.587174415588379,
      "epoch": 1.2356093687971417,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004855305876886589,
      "loss": 5.1343,
      "mean_token_accuracy": 0.1735944405198097,
      "num_tokens": 27274677.0,
      "step": 12450
    },
    {
      "entropy": 5.560019540786743,
      "epoch": 1.2361055974593094,
      "grad_norm": 0.80078125,
      "learning_rate": 0.00048551808416487743,
      "loss": 5.1995,
      "mean_token_accuracy": 0.16757825016975403,
      "num_tokens": 27286512.0,
      "step": 12455
    },
    {
      "entropy": 5.513830327987671,
      "epoch": 1.2366018261214768,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004855055754207011,
      "loss": 5.2137,
      "mean_token_accuracy": 0.16798983067274093,
      "num_tokens": 27297620.0,
      "step": 12460
    },
    {
      "entropy": 5.58218092918396,
      "epoch": 1.2370980547836443,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004854930614564401,
      "loss": 5.2136,
      "mean_token_accuracy": 0.1628502354025841,
      "num_tokens": 27309963.0,
      "step": 12465
    },
    {
      "entropy": 5.662808084487915,
      "epoch": 1.2375942834458118,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004854805422724047,
      "loss": 5.301,
      "mean_token_accuracy": 0.16280816197395326,
      "num_tokens": 27320740.0,
      "step": 12470
    },
    {
      "entropy": 5.523580741882324,
      "epoch": 1.2380905121079793,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004854680178689055,
      "loss": 5.1203,
      "mean_token_accuracy": 0.17369242906570434,
      "num_tokens": 27331543.0,
      "step": 12475
    },
    {
      "entropy": 5.468808984756469,
      "epoch": 1.238586740770147,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048545548824625287,
      "loss": 5.1065,
      "mean_token_accuracy": 0.16945707947015762,
      "num_tokens": 27343222.0,
      "step": 12480
    },
    {
      "entropy": 5.5376989364624025,
      "epoch": 1.2390829694323144,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004854429534047577,
      "loss": 5.1654,
      "mean_token_accuracy": 0.17010142505168915,
      "num_tokens": 27354694.0,
      "step": 12485
    },
    {
      "entropy": 5.51598424911499,
      "epoch": 1.239579198094482,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004854304133447307,
      "loss": 5.1617,
      "mean_token_accuracy": 0.1636533945798874,
      "num_tokens": 27365462.0,
      "step": 12490
    },
    {
      "entropy": 5.487661075592041,
      "epoch": 1.2400754267566494,
      "grad_norm": 0.84375,
      "learning_rate": 0.00048541786806648293,
      "loss": 5.0887,
      "mean_token_accuracy": 0.16731098741292955,
      "num_tokens": 27377548.0,
      "step": 12495
    },
    {
      "entropy": 5.589825820922852,
      "epoch": 1.240571655418817,
      "grad_norm": 0.875,
      "learning_rate": 0.0004854053175703254,
      "loss": 5.207,
      "mean_token_accuracy": 0.16157304495573044,
      "num_tokens": 27388221.0,
      "step": 12500
    },
    {
      "entropy": 5.537165689468384,
      "epoch": 1.2410678840809846,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004853927618565694,
      "loss": 5.1935,
      "mean_token_accuracy": 0.1684794694185257,
      "num_tokens": 27400497.0,
      "step": 12505
    },
    {
      "entropy": 5.506247568130493,
      "epoch": 1.241564112743152,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048538020092552626,
      "loss": 5.0914,
      "mean_token_accuracy": 0.16751380562782286,
      "num_tokens": 27410700.0,
      "step": 12510
    },
    {
      "entropy": 5.551796579360962,
      "epoch": 1.2420603414053195,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048536763477750747,
      "loss": 5.2258,
      "mean_token_accuracy": 0.1655466765165329,
      "num_tokens": 27421286.0,
      "step": 12515
    },
    {
      "entropy": 5.626036262512207,
      "epoch": 1.242556570067487,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004853550634128246,
      "loss": 5.2336,
      "mean_token_accuracy": 0.16408672034740449,
      "num_tokens": 27433922.0,
      "step": 12520
    },
    {
      "entropy": 5.542233991622925,
      "epoch": 1.2430527987296547,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004853424868317895,
      "loss": 5.1626,
      "mean_token_accuracy": 0.16384242922067643,
      "num_tokens": 27444684.0,
      "step": 12525
    },
    {
      "entropy": 5.5254229545593265,
      "epoch": 1.2435490273918222,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004853299050347139,
      "loss": 5.1593,
      "mean_token_accuracy": 0.16584260612726212,
      "num_tokens": 27455494.0,
      "step": 12530
    },
    {
      "entropy": 5.602838563919067,
      "epoch": 1.2440452560539896,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048531731802190994,
      "loss": 5.1972,
      "mean_token_accuracy": 0.1603814661502838,
      "num_tokens": 27464927.0,
      "step": 12535
    },
    {
      "entropy": 5.608969068527221,
      "epoch": 1.244541484716157,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004853047257936897,
      "loss": 5.1974,
      "mean_token_accuracy": 0.16270526647567748,
      "num_tokens": 27476228.0,
      "step": 12540
    },
    {
      "entropy": 5.559487533569336,
      "epoch": 1.2450377133783248,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004852921283503653,
      "loss": 5.1966,
      "mean_token_accuracy": 0.16996813714504241,
      "num_tokens": 27488835.0,
      "step": 12545
    },
    {
      "entropy": 5.550812721252441,
      "epoch": 1.2455339420404923,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004852795256922494,
      "loss": 5.1811,
      "mean_token_accuracy": 0.1639793783426285,
      "num_tokens": 27501481.0,
      "step": 12550
    },
    {
      "entropy": 5.571440935134888,
      "epoch": 1.2460301707026598,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048526691781965425,
      "loss": 5.1766,
      "mean_token_accuracy": 0.16824175864458085,
      "num_tokens": 27512234.0,
      "step": 12555
    },
    {
      "entropy": 5.6053447246551515,
      "epoch": 1.2465263993648272,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048525430473289274,
      "loss": 5.2434,
      "mean_token_accuracy": 0.1635236144065857,
      "num_tokens": 27523756.0,
      "step": 12560
    },
    {
      "entropy": 5.56353530883789,
      "epoch": 1.247022628026995,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048524168643227747,
      "loss": 5.2457,
      "mean_token_accuracy": 0.1614888444542885,
      "num_tokens": 27534579.0,
      "step": 12565
    },
    {
      "entropy": 5.551519727706909,
      "epoch": 1.2475188566891624,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00048522906291812145,
      "loss": 5.2725,
      "mean_token_accuracy": 0.16104435324668884,
      "num_tokens": 27546420.0,
      "step": 12570
    },
    {
      "entropy": 5.575799655914307,
      "epoch": 1.2480150853513299,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048521643419073763,
      "loss": 5.1446,
      "mean_token_accuracy": 0.16417765468358994,
      "num_tokens": 27557701.0,
      "step": 12575
    },
    {
      "entropy": 5.517214632034301,
      "epoch": 1.2485113140134974,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004852038002504392,
      "loss": 5.1611,
      "mean_token_accuracy": 0.1708637699484825,
      "num_tokens": 27568592.0,
      "step": 12580
    },
    {
      "entropy": 5.543105173110962,
      "epoch": 1.249007542675665,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048519116109753955,
      "loss": 5.2086,
      "mean_token_accuracy": 0.17284478098154069,
      "num_tokens": 27579286.0,
      "step": 12585
    },
    {
      "entropy": 5.531618976593018,
      "epoch": 1.2495037713378325,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048517851673235196,
      "loss": 5.1354,
      "mean_token_accuracy": 0.17142497897148132,
      "num_tokens": 27590031.0,
      "step": 12590
    },
    {
      "entropy": 5.552273797988891,
      "epoch": 1.25,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004851658671551901,
      "loss": 5.1788,
      "mean_token_accuracy": 0.16726864129304886,
      "num_tokens": 27600999.0,
      "step": 12595
    },
    {
      "entropy": 5.520488452911377,
      "epoch": 1.2504962286621675,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048515321236636754,
      "loss": 5.1668,
      "mean_token_accuracy": 0.16250496804714204,
      "num_tokens": 27613394.0,
      "step": 12600
    },
    {
      "entropy": 5.5385887145996096,
      "epoch": 1.250992457324335,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004851405523661982,
      "loss": 5.1414,
      "mean_token_accuracy": 0.1739915832877159,
      "num_tokens": 27624819.0,
      "step": 12605
    },
    {
      "entropy": 5.502660655975342,
      "epoch": 1.2514886859865026,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048512788715499597,
      "loss": 5.1651,
      "mean_token_accuracy": 0.17367679625749588,
      "num_tokens": 27635993.0,
      "step": 12610
    },
    {
      "entropy": 5.542938470840454,
      "epoch": 1.2519849146486701,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004851152167330749,
      "loss": 5.2446,
      "mean_token_accuracy": 0.16339800804853438,
      "num_tokens": 27647298.0,
      "step": 12615
    },
    {
      "entropy": 5.589235067367554,
      "epoch": 1.2524811433108376,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048510254110074923,
      "loss": 5.2128,
      "mean_token_accuracy": 0.16768902838230132,
      "num_tokens": 27659381.0,
      "step": 12620
    },
    {
      "entropy": 5.559755516052246,
      "epoch": 1.2529773719730053,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004850898602583333,
      "loss": 5.1878,
      "mean_token_accuracy": 0.1667603462934494,
      "num_tokens": 27671122.0,
      "step": 12625
    },
    {
      "entropy": 5.594545793533325,
      "epoch": 1.2534736006351728,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048507717420614153,
      "loss": 5.1907,
      "mean_token_accuracy": 0.17167871445417404,
      "num_tokens": 27683258.0,
      "step": 12630
    },
    {
      "entropy": 5.4619140625,
      "epoch": 1.2539698292973402,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004850644829444884,
      "loss": 5.0643,
      "mean_token_accuracy": 0.17000673711299896,
      "num_tokens": 27693116.0,
      "step": 12635
    },
    {
      "entropy": 5.4681346893310545,
      "epoch": 1.2544660579595077,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00048505178647368887,
      "loss": 5.1512,
      "mean_token_accuracy": 0.16957887262105942,
      "num_tokens": 27704859.0,
      "step": 12640
    },
    {
      "entropy": 5.50218620300293,
      "epoch": 1.2549622866216752,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004850390847940576,
      "loss": 5.1032,
      "mean_token_accuracy": 0.17045054137706755,
      "num_tokens": 27715854.0,
      "step": 12645
    },
    {
      "entropy": 5.553264617919922,
      "epoch": 1.2554585152838427,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004850263779059097,
      "loss": 5.1458,
      "mean_token_accuracy": 0.1721723720431328,
      "num_tokens": 27726398.0,
      "step": 12650
    },
    {
      "entropy": 5.552562665939331,
      "epoch": 1.2559547439460104,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004850136658095601,
      "loss": 5.171,
      "mean_token_accuracy": 0.17330212593078614,
      "num_tokens": 27737337.0,
      "step": 12655
    },
    {
      "entropy": 5.524002265930176,
      "epoch": 1.2564509726081778,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00048500094850532417,
      "loss": 5.1957,
      "mean_token_accuracy": 0.16753595620393752,
      "num_tokens": 27748593.0,
      "step": 12660
    },
    {
      "entropy": 5.563277959823608,
      "epoch": 1.2569472012703453,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004849882259935172,
      "loss": 5.1649,
      "mean_token_accuracy": 0.16561111956834793,
      "num_tokens": 27760908.0,
      "step": 12665
    },
    {
      "entropy": 5.589109563827515,
      "epoch": 1.257443429932513,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004849754982744548,
      "loss": 5.2399,
      "mean_token_accuracy": 0.16375775039196014,
      "num_tokens": 27772104.0,
      "step": 12670
    },
    {
      "entropy": 5.502607107162476,
      "epoch": 1.2579396585946805,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004849627653484524,
      "loss": 5.1899,
      "mean_token_accuracy": 0.1701417937874794,
      "num_tokens": 27781654.0,
      "step": 12675
    },
    {
      "entropy": 5.555015945434571,
      "epoch": 1.258435887256848,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004849500272158259,
      "loss": 5.1727,
      "mean_token_accuracy": 0.16652282774448396,
      "num_tokens": 27793026.0,
      "step": 12680
    },
    {
      "entropy": 5.548802280426026,
      "epoch": 1.2589321159190154,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004849372838768911,
      "loss": 5.1495,
      "mean_token_accuracy": 0.1729058787226677,
      "num_tokens": 27803447.0,
      "step": 12685
    },
    {
      "entropy": 5.615548849105835,
      "epoch": 1.259428344581183,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000484924535331964,
      "loss": 5.1546,
      "mean_token_accuracy": 0.1635882079601288,
      "num_tokens": 27814632.0,
      "step": 12690
    },
    {
      "entropy": 5.571075344085694,
      "epoch": 1.2599245732433506,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048491178158136086,
      "loss": 5.2367,
      "mean_token_accuracy": 0.1636323943734169,
      "num_tokens": 27826582.0,
      "step": 12695
    },
    {
      "entropy": 5.549906587600708,
      "epoch": 1.260420801905518,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004848990226253978,
      "loss": 5.0991,
      "mean_token_accuracy": 0.16885340809822083,
      "num_tokens": 27838065.0,
      "step": 12700
    },
    {
      "entropy": 5.4739526271820065,
      "epoch": 1.2609170305676856,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048488625846439135,
      "loss": 5.1731,
      "mean_token_accuracy": 0.1661983385682106,
      "num_tokens": 27848182.0,
      "step": 12705
    },
    {
      "entropy": 5.585486602783203,
      "epoch": 1.261413259229853,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004848734890986578,
      "loss": 5.1727,
      "mean_token_accuracy": 0.16330449134111405,
      "num_tokens": 27858524.0,
      "step": 12710
    },
    {
      "entropy": 5.545848560333252,
      "epoch": 1.2619094878920207,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004848607145285141,
      "loss": 5.0946,
      "mean_token_accuracy": 0.17027559280395507,
      "num_tokens": 27868588.0,
      "step": 12715
    },
    {
      "entropy": 5.47784686088562,
      "epoch": 1.2624057165541882,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004848479347542768,
      "loss": 5.0962,
      "mean_token_accuracy": 0.16958286762237548,
      "num_tokens": 27878829.0,
      "step": 12720
    },
    {
      "entropy": 5.513685274124145,
      "epoch": 1.2629019452163557,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000484835149776263,
      "loss": 5.1728,
      "mean_token_accuracy": 0.17348767071962357,
      "num_tokens": 27889872.0,
      "step": 12725
    },
    {
      "entropy": 5.5392392635345455,
      "epoch": 1.2633981738785232,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048482235959478953,
      "loss": 5.2392,
      "mean_token_accuracy": 0.16797590106725693,
      "num_tokens": 27900567.0,
      "step": 12730
    },
    {
      "entropy": 5.547548723220825,
      "epoch": 1.2638944025406906,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004848095642101737,
      "loss": 5.1187,
      "mean_token_accuracy": 0.16998957693576813,
      "num_tokens": 27911594.0,
      "step": 12735
    },
    {
      "entropy": 5.577099561691284,
      "epoch": 1.2643906312028583,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048479676362273277,
      "loss": 5.2251,
      "mean_token_accuracy": 0.158519646525383,
      "num_tokens": 27923694.0,
      "step": 12740
    },
    {
      "entropy": 5.561547088623047,
      "epoch": 1.2648868598650258,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048478395783278425,
      "loss": 5.2971,
      "mean_token_accuracy": 0.15378415286540986,
      "num_tokens": 27934168.0,
      "step": 12745
    },
    {
      "entropy": 5.597050571441651,
      "epoch": 1.2653830885271933,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004847711468406455,
      "loss": 5.1886,
      "mean_token_accuracy": 0.16538678854703903,
      "num_tokens": 27944401.0,
      "step": 12750
    },
    {
      "entropy": 5.499141550064087,
      "epoch": 1.265879317189361,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00048475833064663443,
      "loss": 5.0808,
      "mean_token_accuracy": 0.17664188891649246,
      "num_tokens": 27956219.0,
      "step": 12755
    },
    {
      "entropy": 5.547073936462402,
      "epoch": 1.2663755458515285,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004847455092510687,
      "loss": 5.1806,
      "mean_token_accuracy": 0.1704688251018524,
      "num_tokens": 27967457.0,
      "step": 12760
    },
    {
      "entropy": 5.5392418384552,
      "epoch": 1.266871774513696,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004847326826542663,
      "loss": 5.2006,
      "mean_token_accuracy": 0.16247204542160035,
      "num_tokens": 27978287.0,
      "step": 12765
    },
    {
      "entropy": 5.565428781509399,
      "epoch": 1.2673680031758634,
      "grad_norm": 0.875,
      "learning_rate": 0.00048471985085654537,
      "loss": 5.2134,
      "mean_token_accuracy": 0.17096083909273146,
      "num_tokens": 27989703.0,
      "step": 12770
    },
    {
      "entropy": 5.540671348571777,
      "epoch": 1.2678642318380309,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048470701385822394,
      "loss": 5.17,
      "mean_token_accuracy": 0.168598735332489,
      "num_tokens": 28002071.0,
      "step": 12775
    },
    {
      "entropy": 5.5525736808776855,
      "epoch": 1.2683604605001986,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048469417165962047,
      "loss": 5.2293,
      "mean_token_accuracy": 0.17004916071891785,
      "num_tokens": 28014517.0,
      "step": 12780
    },
    {
      "entropy": 5.554504776000977,
      "epoch": 1.268856689162366,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004846813242610535,
      "loss": 5.1979,
      "mean_token_accuracy": 0.1747070163488388,
      "num_tokens": 28025135.0,
      "step": 12785
    },
    {
      "entropy": 5.518036222457885,
      "epoch": 1.2693529178245335,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004846684716628414,
      "loss": 5.1215,
      "mean_token_accuracy": 0.17613378763198853,
      "num_tokens": 28036105.0,
      "step": 12790
    },
    {
      "entropy": 5.480977630615234,
      "epoch": 1.269849146486701,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000484655613865303,
      "loss": 5.1611,
      "mean_token_accuracy": 0.17135216891765595,
      "num_tokens": 28048825.0,
      "step": 12795
    },
    {
      "entropy": 5.545332288742065,
      "epoch": 1.2703453751488687,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00048464275086875717,
      "loss": 5.1196,
      "mean_token_accuracy": 0.1681775078177452,
      "num_tokens": 28060408.0,
      "step": 12800
    },
    {
      "entropy": 5.533175611495972,
      "epoch": 1.2708416038110362,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004846298826735229,
      "loss": 5.1877,
      "mean_token_accuracy": 0.16483981162309647,
      "num_tokens": 28072179.0,
      "step": 12805
    },
    {
      "entropy": 5.508754110336303,
      "epoch": 1.2713378324732036,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004846170092799192,
      "loss": 5.1751,
      "mean_token_accuracy": 0.16775323301553727,
      "num_tokens": 28083825.0,
      "step": 12810
    },
    {
      "entropy": 5.699008941650391,
      "epoch": 1.2718340611353711,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048460413068826536,
      "loss": 5.3208,
      "mean_token_accuracy": 0.16472983807325364,
      "num_tokens": 28095136.0,
      "step": 12815
    },
    {
      "entropy": 5.586997032165527,
      "epoch": 1.2723302897975386,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004845912468988807,
      "loss": 5.1461,
      "mean_token_accuracy": 0.1671767935156822,
      "num_tokens": 28104763.0,
      "step": 12820
    },
    {
      "entropy": 5.539486837387085,
      "epoch": 1.2728265184597063,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004845783579120849,
      "loss": 5.1947,
      "mean_token_accuracy": 0.17052305340766907,
      "num_tokens": 28115771.0,
      "step": 12825
    },
    {
      "entropy": 5.475801658630371,
      "epoch": 1.2733227471218738,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004845654637281973,
      "loss": 5.1209,
      "mean_token_accuracy": 0.16984303593635558,
      "num_tokens": 28128039.0,
      "step": 12830
    },
    {
      "entropy": 5.4931408882141115,
      "epoch": 1.2738189757840412,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00048455256434753774,
      "loss": 5.1431,
      "mean_token_accuracy": 0.17383018583059312,
      "num_tokens": 28139163.0,
      "step": 12835
    },
    {
      "entropy": 5.526537895202637,
      "epoch": 1.274315204446209,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004845396597704262,
      "loss": 5.1763,
      "mean_token_accuracy": 0.1646669551730156,
      "num_tokens": 28150625.0,
      "step": 12840
    },
    {
      "entropy": 5.5661365509033205,
      "epoch": 1.2748114331083764,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048452674999718257,
      "loss": 5.2506,
      "mean_token_accuracy": 0.16619838774204254,
      "num_tokens": 28163072.0,
      "step": 12845
    },
    {
      "entropy": 5.602521657943726,
      "epoch": 1.2753076617705439,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048451383502812705,
      "loss": 5.2701,
      "mean_token_accuracy": 0.162578547000885,
      "num_tokens": 28174192.0,
      "step": 12850
    },
    {
      "entropy": 5.526638650894165,
      "epoch": 1.2758038904327114,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048450091486357987,
      "loss": 5.141,
      "mean_token_accuracy": 0.1731074795126915,
      "num_tokens": 28184477.0,
      "step": 12855
    },
    {
      "entropy": 5.510159635543824,
      "epoch": 1.2763001190948788,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00048448798950386147,
      "loss": 5.053,
      "mean_token_accuracy": 0.16859374642372132,
      "num_tokens": 28194552.0,
      "step": 12860
    },
    {
      "entropy": 5.533937597274781,
      "epoch": 1.2767963477570463,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048447505894929233,
      "loss": 5.2184,
      "mean_token_accuracy": 0.168878173828125,
      "num_tokens": 28205712.0,
      "step": 12865
    },
    {
      "entropy": 5.55695071220398,
      "epoch": 1.277292576419214,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048446212320019306,
      "loss": 5.1473,
      "mean_token_accuracy": 0.17181674540042877,
      "num_tokens": 28217246.0,
      "step": 12870
    },
    {
      "entropy": 5.537813138961792,
      "epoch": 1.2777888050813815,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004844491822568846,
      "loss": 5.1716,
      "mean_token_accuracy": 0.16606325507164002,
      "num_tokens": 28227933.0,
      "step": 12875
    },
    {
      "entropy": 5.523345565795898,
      "epoch": 1.278285033743549,
      "grad_norm": 0.8203125,
      "learning_rate": 0.00048443623611968764,
      "loss": 5.1863,
      "mean_token_accuracy": 0.17389071732759476,
      "num_tokens": 28239692.0,
      "step": 12880
    },
    {
      "entropy": 5.496372127532959,
      "epoch": 1.2787812624057167,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048442328478892334,
      "loss": 5.1554,
      "mean_token_accuracy": 0.17384358793497084,
      "num_tokens": 28249815.0,
      "step": 12885
    },
    {
      "entropy": 5.605033826828003,
      "epoch": 1.2792774910678841,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004844103282649128,
      "loss": 5.1849,
      "mean_token_accuracy": 0.1656591400504112,
      "num_tokens": 28260385.0,
      "step": 12890
    },
    {
      "entropy": 5.528675222396851,
      "epoch": 1.2797737197300516,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004843973665479774,
      "loss": 5.2006,
      "mean_token_accuracy": 0.16269042789936067,
      "num_tokens": 28271858.0,
      "step": 12895
    },
    {
      "entropy": 5.5495401382446286,
      "epoch": 1.280269948392219,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048438439963843853,
      "loss": 5.1878,
      "mean_token_accuracy": 0.16631882786750793,
      "num_tokens": 28283988.0,
      "step": 12900
    },
    {
      "entropy": 5.554886674880981,
      "epoch": 1.2807661770543866,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048437142753661775,
      "loss": 5.2309,
      "mean_token_accuracy": 0.16862109303474426,
      "num_tokens": 28293281.0,
      "step": 12905
    },
    {
      "entropy": 5.606166076660156,
      "epoch": 1.2812624057165543,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004843584502428367,
      "loss": 5.1776,
      "mean_token_accuracy": 0.16311835944652558,
      "num_tokens": 28304468.0,
      "step": 12910
    },
    {
      "entropy": 5.464374637603759,
      "epoch": 1.2817586343787217,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004843454677574173,
      "loss": 5.1312,
      "mean_token_accuracy": 0.17349398732185364,
      "num_tokens": 28315864.0,
      "step": 12915
    },
    {
      "entropy": 5.518435478210449,
      "epoch": 1.2822548630408892,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00048433248008068123,
      "loss": 5.1815,
      "mean_token_accuracy": 0.170827279984951,
      "num_tokens": 28327212.0,
      "step": 12920
    },
    {
      "entropy": 5.581830406188965,
      "epoch": 1.282751091703057,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048431948721295085,
      "loss": 5.1907,
      "mean_token_accuracy": 0.17047163248062133,
      "num_tokens": 28339256.0,
      "step": 12925
    },
    {
      "entropy": 5.577413654327392,
      "epoch": 1.2832473203652244,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004843064891545482,
      "loss": 5.2143,
      "mean_token_accuracy": 0.16551536619663237,
      "num_tokens": 28350196.0,
      "step": 12930
    },
    {
      "entropy": 5.52842082977295,
      "epoch": 1.2837435490273919,
      "grad_norm": 0.875,
      "learning_rate": 0.0004842934859057956,
      "loss": 5.1686,
      "mean_token_accuracy": 0.16753441393375396,
      "num_tokens": 28361241.0,
      "step": 12935
    },
    {
      "entropy": 5.627494764328003,
      "epoch": 1.2842397776895593,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004842804774670156,
      "loss": 5.3024,
      "mean_token_accuracy": 0.16167117804288864,
      "num_tokens": 28373369.0,
      "step": 12940
    },
    {
      "entropy": 5.568782329559326,
      "epoch": 1.2847360063517268,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048426746383853066,
      "loss": 5.127,
      "mean_token_accuracy": 0.16794929206371306,
      "num_tokens": 28384970.0,
      "step": 12945
    },
    {
      "entropy": 5.568664836883545,
      "epoch": 1.2852322350138943,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048425444502066354,
      "loss": 5.1453,
      "mean_token_accuracy": 0.16711928248405455,
      "num_tokens": 28396469.0,
      "step": 12950
    },
    {
      "entropy": 5.528152179718018,
      "epoch": 1.285728463676062,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048424142101373706,
      "loss": 5.195,
      "mean_token_accuracy": 0.17144554257392883,
      "num_tokens": 28406975.0,
      "step": 12955
    },
    {
      "entropy": 5.466510486602783,
      "epoch": 1.2862246923382294,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048422839181807426,
      "loss": 5.0706,
      "mean_token_accuracy": 0.17831925451755523,
      "num_tokens": 28418285.0,
      "step": 12960
    },
    {
      "entropy": 5.513025188446045,
      "epoch": 1.286720921000397,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004842153574339982,
      "loss": 5.2206,
      "mean_token_accuracy": 0.16516928300261496,
      "num_tokens": 28429769.0,
      "step": 12965
    },
    {
      "entropy": 5.506083345413208,
      "epoch": 1.2872171496625646,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048420231786183216,
      "loss": 5.1595,
      "mean_token_accuracy": 0.1747983694076538,
      "num_tokens": 28441188.0,
      "step": 12970
    },
    {
      "entropy": 5.616161108016968,
      "epoch": 1.287713378324732,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004841892731018994,
      "loss": 5.2329,
      "mean_token_accuracy": 0.16167017966508865,
      "num_tokens": 28453504.0,
      "step": 12975
    },
    {
      "entropy": 5.578122282028199,
      "epoch": 1.2882096069868996,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004841762231545233,
      "loss": 5.1995,
      "mean_token_accuracy": 0.17572660893201827,
      "num_tokens": 28465930.0,
      "step": 12980
    },
    {
      "entropy": 5.494106912612915,
      "epoch": 1.288705835649067,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048416316802002776,
      "loss": 5.1294,
      "mean_token_accuracy": 0.1717277780175209,
      "num_tokens": 28477803.0,
      "step": 12985
    },
    {
      "entropy": 5.583841276168823,
      "epoch": 1.2892020643112345,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048415010769873625,
      "loss": 5.233,
      "mean_token_accuracy": 0.16571509540081025,
      "num_tokens": 28488556.0,
      "step": 12990
    },
    {
      "entropy": 5.595171785354614,
      "epoch": 1.2896982929734022,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004841370421909728,
      "loss": 5.1394,
      "mean_token_accuracy": 0.16938367933034898,
      "num_tokens": 28499275.0,
      "step": 12995
    },
    {
      "entropy": 5.572546529769897,
      "epoch": 1.2901945216355697,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00048412397149706133,
      "loss": 5.1668,
      "mean_token_accuracy": 0.1695139303803444,
      "num_tokens": 28511327.0,
      "step": 13000
    },
    {
      "entropy": 5.525767993927002,
      "epoch": 1.2906907502977372,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000484110895617326,
      "loss": 5.1943,
      "mean_token_accuracy": 0.1702712744474411,
      "num_tokens": 28523937.0,
      "step": 13005
    },
    {
      "entropy": 5.451292610168457,
      "epoch": 1.2911869789599046,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048409781455209104,
      "loss": 5.1005,
      "mean_token_accuracy": 0.17020800113677978,
      "num_tokens": 28534028.0,
      "step": 13010
    },
    {
      "entropy": 5.606750440597534,
      "epoch": 1.2916832076220723,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004840847283016809,
      "loss": 5.2013,
      "mean_token_accuracy": 0.15455452352762222,
      "num_tokens": 28545480.0,
      "step": 13015
    },
    {
      "entropy": 5.568125677108765,
      "epoch": 1.2921794362842398,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048407163686642,
      "loss": 5.1594,
      "mean_token_accuracy": 0.1739239364862442,
      "num_tokens": 28555401.0,
      "step": 13020
    },
    {
      "entropy": 5.554666662216187,
      "epoch": 1.2926756649464073,
      "grad_norm": 1.015625,
      "learning_rate": 0.000484058540246633,
      "loss": 5.2059,
      "mean_token_accuracy": 0.1698061391711235,
      "num_tokens": 28565497.0,
      "step": 13025
    },
    {
      "entropy": 5.57824854850769,
      "epoch": 1.2931718936085748,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004840454384426447,
      "loss": 5.2922,
      "mean_token_accuracy": 0.15998981446027755,
      "num_tokens": 28576186.0,
      "step": 13030
    },
    {
      "entropy": 5.581986141204834,
      "epoch": 1.2936681222707422,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004840323314547799,
      "loss": 5.1656,
      "mean_token_accuracy": 0.1692199230194092,
      "num_tokens": 28586202.0,
      "step": 13035
    },
    {
      "entropy": 5.521803569793701,
      "epoch": 1.29416435093291,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048401921928336385,
      "loss": 5.135,
      "mean_token_accuracy": 0.16795178800821303,
      "num_tokens": 28596486.0,
      "step": 13040
    },
    {
      "entropy": 5.572145986557007,
      "epoch": 1.2946605795950774,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004840061019287215,
      "loss": 5.1745,
      "mean_token_accuracy": 0.1702177494764328,
      "num_tokens": 28607934.0,
      "step": 13045
    },
    {
      "entropy": 5.610029458999634,
      "epoch": 1.2951568082572449,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048399297939117817,
      "loss": 5.21,
      "mean_token_accuracy": 0.16241537481546403,
      "num_tokens": 28617673.0,
      "step": 13050
    },
    {
      "entropy": 5.622319078445434,
      "epoch": 1.2956530369194126,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004839798516710593,
      "loss": 5.2013,
      "mean_token_accuracy": 0.16854730099439622,
      "num_tokens": 28629811.0,
      "step": 13055
    },
    {
      "entropy": 5.602078294754028,
      "epoch": 1.29614926558158,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004839667187686904,
      "loss": 5.287,
      "mean_token_accuracy": 0.16146073937416078,
      "num_tokens": 28640371.0,
      "step": 13060
    },
    {
      "entropy": 5.5573790073394775,
      "epoch": 1.2966454942437475,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004839535806843972,
      "loss": 5.1969,
      "mean_token_accuracy": 0.1657620057463646,
      "num_tokens": 28651904.0,
      "step": 13065
    },
    {
      "entropy": 5.514594411849975,
      "epoch": 1.297141722905915,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004839404374185054,
      "loss": 5.1405,
      "mean_token_accuracy": 0.1666380301117897,
      "num_tokens": 28662241.0,
      "step": 13070
    },
    {
      "entropy": 5.5007569789886475,
      "epoch": 1.2976379515680825,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000483927288971341,
      "loss": 5.099,
      "mean_token_accuracy": 0.17499319612979888,
      "num_tokens": 28672791.0,
      "step": 13075
    },
    {
      "entropy": 5.5438220500946045,
      "epoch": 1.2981341802302502,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004839141353432301,
      "loss": 5.2433,
      "mean_token_accuracy": 0.1647740289568901,
      "num_tokens": 28684218.0,
      "step": 13080
    },
    {
      "entropy": 5.5660436153411865,
      "epoch": 1.2986304088924177,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004839009765344987,
      "loss": 5.1839,
      "mean_token_accuracy": 0.16340476721525193,
      "num_tokens": 28694690.0,
      "step": 13085
    },
    {
      "entropy": 5.50096116065979,
      "epoch": 1.2991266375545851,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00048388781254547324,
      "loss": 5.1645,
      "mean_token_accuracy": 0.17385320067405702,
      "num_tokens": 28705890.0,
      "step": 13090
    },
    {
      "entropy": 5.550221252441406,
      "epoch": 1.2996228662167526,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048387464337648026,
      "loss": 5.1948,
      "mean_token_accuracy": 0.16657274216413498,
      "num_tokens": 28716935.0,
      "step": 13095
    },
    {
      "entropy": 5.6260992050170895,
      "epoch": 1.3001190948789203,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048386146902784606,
      "loss": 5.1926,
      "mean_token_accuracy": 0.16981116384267808,
      "num_tokens": 28728053.0,
      "step": 13100
    },
    {
      "entropy": 5.503963327407837,
      "epoch": 1.3006153235410878,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048384828949989757,
      "loss": 5.1349,
      "mean_token_accuracy": 0.17302269190549852,
      "num_tokens": 28738940.0,
      "step": 13105
    },
    {
      "entropy": 5.5394786357879635,
      "epoch": 1.3011115522032553,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048383510479296153,
      "loss": 5.2042,
      "mean_token_accuracy": 0.1705235943198204,
      "num_tokens": 28750259.0,
      "step": 13110
    },
    {
      "entropy": 5.572442293167114,
      "epoch": 1.3016077808654227,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004838219149073649,
      "loss": 5.1611,
      "mean_token_accuracy": 0.17305178046226502,
      "num_tokens": 28760119.0,
      "step": 13115
    },
    {
      "entropy": 5.591745042800904,
      "epoch": 1.3021040095275902,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004838087198434347,
      "loss": 5.1939,
      "mean_token_accuracy": 0.16606315076351166,
      "num_tokens": 28772022.0,
      "step": 13120
    },
    {
      "entropy": 5.582187938690185,
      "epoch": 1.302600238189758,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004837955196014982,
      "loss": 5.152,
      "mean_token_accuracy": 0.16164156645536423,
      "num_tokens": 28782935.0,
      "step": 13125
    },
    {
      "entropy": 5.586201763153076,
      "epoch": 1.3030964668519254,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048378231418188275,
      "loss": 5.2745,
      "mean_token_accuracy": 0.1642395079135895,
      "num_tokens": 28793604.0,
      "step": 13130
    },
    {
      "entropy": 5.56099305152893,
      "epoch": 1.3035926955140928,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048376910358491576,
      "loss": 5.2239,
      "mean_token_accuracy": 0.16122603565454482,
      "num_tokens": 28803494.0,
      "step": 13135
    },
    {
      "entropy": 5.557882404327392,
      "epoch": 1.3040889241762605,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004837558878109249,
      "loss": 5.1895,
      "mean_token_accuracy": 0.1622390329837799,
      "num_tokens": 28816232.0,
      "step": 13140
    },
    {
      "entropy": 5.549216604232788,
      "epoch": 1.304585152838428,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004837426668602378,
      "loss": 5.1566,
      "mean_token_accuracy": 0.17220099568367003,
      "num_tokens": 28827056.0,
      "step": 13145
    },
    {
      "entropy": 5.509700107574463,
      "epoch": 1.3050813815005955,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004837294407331825,
      "loss": 5.1413,
      "mean_token_accuracy": 0.17143516838550568,
      "num_tokens": 28837217.0,
      "step": 13150
    },
    {
      "entropy": 5.472987222671509,
      "epoch": 1.305577610162763,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004837162094300867,
      "loss": 5.0571,
      "mean_token_accuracy": 0.16928470134735107,
      "num_tokens": 28846838.0,
      "step": 13155
    },
    {
      "entropy": 5.521747398376465,
      "epoch": 1.3060738388249304,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004837029729512787,
      "loss": 5.0973,
      "mean_token_accuracy": 0.17382522225379943,
      "num_tokens": 28856269.0,
      "step": 13160
    },
    {
      "entropy": 5.534066820144654,
      "epoch": 1.306570067487098,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00048368973129708667,
      "loss": 5.1308,
      "mean_token_accuracy": 0.17246145457029344,
      "num_tokens": 28867142.0,
      "step": 13165
    },
    {
      "entropy": 5.549687194824219,
      "epoch": 1.3070662961492656,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048367648446783894,
      "loss": 5.1031,
      "mean_token_accuracy": 0.1684173509478569,
      "num_tokens": 28877346.0,
      "step": 13170
    },
    {
      "entropy": 5.536315250396728,
      "epoch": 1.307562524811433,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004836632324638642,
      "loss": 5.1142,
      "mean_token_accuracy": 0.16722597181797028,
      "num_tokens": 28888425.0,
      "step": 13175
    },
    {
      "entropy": 5.506155681610108,
      "epoch": 1.3080587534736006,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048364997528549077,
      "loss": 5.1404,
      "mean_token_accuracy": 0.16865494847297668,
      "num_tokens": 28899575.0,
      "step": 13180
    },
    {
      "entropy": 5.520729732513428,
      "epoch": 1.3085549821357683,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048363671293304763,
      "loss": 5.1346,
      "mean_token_accuracy": 0.17176464796066285,
      "num_tokens": 28910769.0,
      "step": 13185
    },
    {
      "entropy": 5.545792770385742,
      "epoch": 1.3090512107979357,
      "grad_norm": 1.0,
      "learning_rate": 0.0004836234454068636,
      "loss": 5.1282,
      "mean_token_accuracy": 0.17846204042434693,
      "num_tokens": 28921224.0,
      "step": 13190
    },
    {
      "entropy": 5.505349254608154,
      "epoch": 1.3095474394601032,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048361017270726764,
      "loss": 5.1524,
      "mean_token_accuracy": 0.16589900255203247,
      "num_tokens": 28932585.0,
      "step": 13195
    },
    {
      "entropy": 5.492430877685547,
      "epoch": 1.3100436681222707,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004835968948345889,
      "loss": 5.1721,
      "mean_token_accuracy": 0.170295087993145,
      "num_tokens": 28943471.0,
      "step": 13200
    },
    {
      "entropy": 5.488894414901734,
      "epoch": 1.3105398967844382,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00048358361178915666,
      "loss": 5.0959,
      "mean_token_accuracy": 0.1799687534570694,
      "num_tokens": 28956336.0,
      "step": 13205
    },
    {
      "entropy": 5.630616474151611,
      "epoch": 1.3110361254466059,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004835703235713003,
      "loss": 5.2232,
      "mean_token_accuracy": 0.16556530743837355,
      "num_tokens": 28966689.0,
      "step": 13210
    },
    {
      "entropy": 5.607818365097046,
      "epoch": 1.3115323541087733,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048355703018134935,
      "loss": 5.32,
      "mean_token_accuracy": 0.17158676236867904,
      "num_tokens": 28978548.0,
      "step": 13215
    },
    {
      "entropy": 5.579065275192261,
      "epoch": 1.3120285827709408,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004835437316196335,
      "loss": 5.229,
      "mean_token_accuracy": 0.16674552112817764,
      "num_tokens": 28988899.0,
      "step": 13220
    },
    {
      "entropy": 5.509582948684693,
      "epoch": 1.3125248114331085,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004835304278864824,
      "loss": 5.154,
      "mean_token_accuracy": 0.17324082553386688,
      "num_tokens": 28999666.0,
      "step": 13225
    },
    {
      "entropy": 5.594459486007691,
      "epoch": 1.313021040095276,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004835171189822261,
      "loss": 5.1625,
      "mean_token_accuracy": 0.16484105587005615,
      "num_tokens": 29010185.0,
      "step": 13230
    },
    {
      "entropy": 5.535682582855225,
      "epoch": 1.3135172687574435,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004835038049071945,
      "loss": 5.1682,
      "mean_token_accuracy": 0.16760471016168593,
      "num_tokens": 29020761.0,
      "step": 13235
    },
    {
      "entropy": 5.585552406311035,
      "epoch": 1.314013497419611,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004834904856617178,
      "loss": 5.3137,
      "mean_token_accuracy": 0.1671786919236183,
      "num_tokens": 29031844.0,
      "step": 13240
    },
    {
      "entropy": 5.6389909267425535,
      "epoch": 1.3145097260817784,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048347716124612643,
      "loss": 5.2712,
      "mean_token_accuracy": 0.16106192171573638,
      "num_tokens": 29043101.0,
      "step": 13245
    },
    {
      "entropy": 5.616788768768311,
      "epoch": 1.3150059547439459,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004834638316607506,
      "loss": 5.2044,
      "mean_token_accuracy": 0.17278918176889418,
      "num_tokens": 29054016.0,
      "step": 13250
    },
    {
      "entropy": 5.4927465438842775,
      "epoch": 1.3155021834061136,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00048345049690592105,
      "loss": 5.1901,
      "mean_token_accuracy": 0.16275459229946138,
      "num_tokens": 29065127.0,
      "step": 13255
    },
    {
      "entropy": 5.581573867797852,
      "epoch": 1.315998412068281,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048343715698196814,
      "loss": 5.1544,
      "mean_token_accuracy": 0.16777286380529405,
      "num_tokens": 29075130.0,
      "step": 13260
    },
    {
      "entropy": 5.635171556472779,
      "epoch": 1.3164946407304485,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004834238118892231,
      "loss": 5.265,
      "mean_token_accuracy": 0.16001363098621368,
      "num_tokens": 29086405.0,
      "step": 13265
    },
    {
      "entropy": 5.600570344924927,
      "epoch": 1.3169908693926162,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004834104616280165,
      "loss": 5.2262,
      "mean_token_accuracy": 0.1636667974293232,
      "num_tokens": 29096945.0,
      "step": 13270
    },
    {
      "entropy": 5.600618743896485,
      "epoch": 1.3174870980547837,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004833971061986796,
      "loss": 5.2881,
      "mean_token_accuracy": 0.1700609028339386,
      "num_tokens": 29108275.0,
      "step": 13275
    },
    {
      "entropy": 5.555669546127319,
      "epoch": 1.3179833267169512,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048338374560154344,
      "loss": 5.1227,
      "mean_token_accuracy": 0.1661107361316681,
      "num_tokens": 29118832.0,
      "step": 13280
    },
    {
      "entropy": 5.5101402759552,
      "epoch": 1.3184795553791187,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004833703798369395,
      "loss": 5.1365,
      "mean_token_accuracy": 0.17052176594734192,
      "num_tokens": 29130720.0,
      "step": 13285
    },
    {
      "entropy": 5.595566844940185,
      "epoch": 1.3189757840412861,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004833570089051992,
      "loss": 5.1475,
      "mean_token_accuracy": 0.17092374265193938,
      "num_tokens": 29140441.0,
      "step": 13290
    },
    {
      "entropy": 5.563398361206055,
      "epoch": 1.3194720127034538,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048334363280665393,
      "loss": 5.2091,
      "mean_token_accuracy": 0.16432921588420868,
      "num_tokens": 29151544.0,
      "step": 13295
    },
    {
      "entropy": 5.595821905136108,
      "epoch": 1.3199682413656213,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004833302515416355,
      "loss": 5.2628,
      "mean_token_accuracy": 0.16516421288251876,
      "num_tokens": 29165475.0,
      "step": 13300
    },
    {
      "entropy": 5.523571920394898,
      "epoch": 1.3204644700277888,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00048331686511047584,
      "loss": 5.1181,
      "mean_token_accuracy": 0.165736648440361,
      "num_tokens": 29177139.0,
      "step": 13305
    },
    {
      "entropy": 5.558342456817627,
      "epoch": 1.3209606986899565,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048330347351350676,
      "loss": 5.1956,
      "mean_token_accuracy": 0.17106725871562958,
      "num_tokens": 29187812.0,
      "step": 13310
    },
    {
      "entropy": 5.482630729675293,
      "epoch": 1.321456927352124,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048329007675106036,
      "loss": 5.0723,
      "mean_token_accuracy": 0.17400735318660737,
      "num_tokens": 29197932.0,
      "step": 13315
    },
    {
      "entropy": 5.491154193878174,
      "epoch": 1.3219531560142914,
      "grad_norm": 0.8125,
      "learning_rate": 0.00048327667482346895,
      "loss": 5.1023,
      "mean_token_accuracy": 0.17220928221940995,
      "num_tokens": 29209529.0,
      "step": 13320
    },
    {
      "entropy": 5.505966997146606,
      "epoch": 1.322449384676459,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004832632677310647,
      "loss": 5.1667,
      "mean_token_accuracy": 0.17061394900083543,
      "num_tokens": 29219027.0,
      "step": 13325
    },
    {
      "entropy": 5.546169328689575,
      "epoch": 1.3229456133386264,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048324985547418026,
      "loss": 5.1817,
      "mean_token_accuracy": 0.16532337069511413,
      "num_tokens": 29229647.0,
      "step": 13330
    },
    {
      "entropy": 5.588563585281372,
      "epoch": 1.3234418420007938,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004832364380531482,
      "loss": 5.2094,
      "mean_token_accuracy": 0.17632917761802674,
      "num_tokens": 29240263.0,
      "step": 13335
    },
    {
      "entropy": 5.598335027694702,
      "epoch": 1.3239380706629615,
      "grad_norm": 0.859375,
      "learning_rate": 0.00048322301546830103,
      "loss": 5.2309,
      "mean_token_accuracy": 0.16411226987838745,
      "num_tokens": 29251115.0,
      "step": 13340
    },
    {
      "entropy": 5.540830087661743,
      "epoch": 1.324434299325129,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004832095877199718,
      "loss": 5.1977,
      "mean_token_accuracy": 0.165518681704998,
      "num_tokens": 29261888.0,
      "step": 13345
    },
    {
      "entropy": 5.522991180419922,
      "epoch": 1.3249305279872965,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048319615480849347,
      "loss": 5.0953,
      "mean_token_accuracy": 0.17710457146167755,
      "num_tokens": 29271948.0,
      "step": 13350
    },
    {
      "entropy": 5.5838206768035885,
      "epoch": 1.3254267566494642,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048318271673419914,
      "loss": 5.1643,
      "mean_token_accuracy": 0.16720860749483107,
      "num_tokens": 29282449.0,
      "step": 13355
    },
    {
      "entropy": 5.556437206268311,
      "epoch": 1.3259229853116317,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048316927349742207,
      "loss": 5.168,
      "mean_token_accuracy": 0.16741278469562532,
      "num_tokens": 29294280.0,
      "step": 13360
    },
    {
      "entropy": 5.566699028015137,
      "epoch": 1.3264192139737991,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004831558250984955,
      "loss": 5.161,
      "mean_token_accuracy": 0.16971770375967027,
      "num_tokens": 29305143.0,
      "step": 13365
    },
    {
      "entropy": 5.532807159423828,
      "epoch": 1.3269154426359666,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048314237153775303,
      "loss": 5.0857,
      "mean_token_accuracy": 0.17136003226041793,
      "num_tokens": 29314697.0,
      "step": 13370
    },
    {
      "entropy": 5.483678770065308,
      "epoch": 1.327411671298134,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004831289128155283,
      "loss": 5.1114,
      "mean_token_accuracy": 0.16558410972356796,
      "num_tokens": 29325257.0,
      "step": 13375
    },
    {
      "entropy": 5.563903284072876,
      "epoch": 1.3279078999603018,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048311544893215496,
      "loss": 5.2156,
      "mean_token_accuracy": 0.16387782990932465,
      "num_tokens": 29335449.0,
      "step": 13380
    },
    {
      "entropy": 5.585507440567016,
      "epoch": 1.3284041286224693,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000483101979887967,
      "loss": 5.2078,
      "mean_token_accuracy": 0.16704815328121186,
      "num_tokens": 29346898.0,
      "step": 13385
    },
    {
      "entropy": 5.61611499786377,
      "epoch": 1.3289003572846367,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048308850568329825,
      "loss": 5.1664,
      "mean_token_accuracy": 0.16246731579303741,
      "num_tokens": 29356118.0,
      "step": 13390
    },
    {
      "entropy": 5.564300537109375,
      "epoch": 1.3293965859468042,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048307502631848306,
      "loss": 5.2394,
      "mean_token_accuracy": 0.15915516018867493,
      "num_tokens": 29367597.0,
      "step": 13395
    },
    {
      "entropy": 5.526433944702148,
      "epoch": 1.329892814608972,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048306154179385546,
      "loss": 5.1232,
      "mean_token_accuracy": 0.16847596168518067,
      "num_tokens": 29377829.0,
      "step": 13400
    },
    {
      "entropy": 5.53847107887268,
      "epoch": 1.3303890432711394,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004830480521097501,
      "loss": 5.076,
      "mean_token_accuracy": 0.17512914091348647,
      "num_tokens": 29387945.0,
      "step": 13405
    },
    {
      "entropy": 5.48411340713501,
      "epoch": 1.3308852719333069,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004830345572665012,
      "loss": 5.0532,
      "mean_token_accuracy": 0.17665792554616927,
      "num_tokens": 29398269.0,
      "step": 13410
    },
    {
      "entropy": 5.576151895523071,
      "epoch": 1.3313815005954743,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048302105726444366,
      "loss": 5.2323,
      "mean_token_accuracy": 0.16191613674163818,
      "num_tokens": 29410946.0,
      "step": 13415
    },
    {
      "entropy": 5.498389053344726,
      "epoch": 1.3318777292576418,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00048300755210391205,
      "loss": 5.0583,
      "mean_token_accuracy": 0.17289605289697646,
      "num_tokens": 29421770.0,
      "step": 13420
    },
    {
      "entropy": 5.563178682327271,
      "epoch": 1.3323739579198095,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004829940417852415,
      "loss": 5.2394,
      "mean_token_accuracy": 0.16597738564014436,
      "num_tokens": 29433717.0,
      "step": 13425
    },
    {
      "entropy": 5.485410499572754,
      "epoch": 1.332870186581977,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004829805263087668,
      "loss": 5.0662,
      "mean_token_accuracy": 0.17547925263643266,
      "num_tokens": 29444382.0,
      "step": 13430
    },
    {
      "entropy": 5.571080541610717,
      "epoch": 1.3333664152441445,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048296700567482324,
      "loss": 5.1981,
      "mean_token_accuracy": 0.16249519884586333,
      "num_tokens": 29454680.0,
      "step": 13435
    },
    {
      "entropy": 5.621443223953247,
      "epoch": 1.3338626439063122,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004829534798837461,
      "loss": 5.1994,
      "mean_token_accuracy": 0.16936905831098556,
      "num_tokens": 29465065.0,
      "step": 13440
    },
    {
      "entropy": 5.583467960357666,
      "epoch": 1.3343588725684796,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004829399489358707,
      "loss": 5.1311,
      "mean_token_accuracy": 0.16703284978866578,
      "num_tokens": 29475704.0,
      "step": 13445
    },
    {
      "entropy": 5.513696384429932,
      "epoch": 1.334855101230647,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004829264128315326,
      "loss": 5.1712,
      "mean_token_accuracy": 0.16665140837430953,
      "num_tokens": 29486418.0,
      "step": 13450
    },
    {
      "entropy": 5.606047058105469,
      "epoch": 1.3353513298928146,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004829128715710676,
      "loss": 5.2418,
      "mean_token_accuracy": 0.1586636483669281,
      "num_tokens": 29498939.0,
      "step": 13455
    },
    {
      "entropy": 5.550904178619385,
      "epoch": 1.335847558554982,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004828993251548114,
      "loss": 5.0857,
      "mean_token_accuracy": 0.17292490005493164,
      "num_tokens": 29508525.0,
      "step": 13460
    },
    {
      "entropy": 5.563429927825927,
      "epoch": 1.3363437872171495,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004828857735830999,
      "loss": 5.1955,
      "mean_token_accuracy": 0.17441526651382447,
      "num_tokens": 29519962.0,
      "step": 13465
    },
    {
      "entropy": 5.532764673233032,
      "epoch": 1.3368400158793172,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004828722168562691,
      "loss": 5.1508,
      "mean_token_accuracy": 0.16437328159809111,
      "num_tokens": 29529586.0,
      "step": 13470
    },
    {
      "entropy": 5.536051225662232,
      "epoch": 1.3373362445414847,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004828586549746553,
      "loss": 5.2257,
      "mean_token_accuracy": 0.16071384251117707,
      "num_tokens": 29540670.0,
      "step": 13475
    },
    {
      "entropy": 5.555367660522461,
      "epoch": 1.3378324732036522,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004828450879385947,
      "loss": 5.1919,
      "mean_token_accuracy": 0.16150649785995483,
      "num_tokens": 29551484.0,
      "step": 13480
    },
    {
      "entropy": 5.580913972854614,
      "epoch": 1.3383287018658199,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004828315157484239,
      "loss": 5.1321,
      "mean_token_accuracy": 0.16958364099264145,
      "num_tokens": 29561596.0,
      "step": 13485
    },
    {
      "entropy": 5.588375091552734,
      "epoch": 1.3388249305279873,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004828179384044793,
      "loss": 5.1529,
      "mean_token_accuracy": 0.16749670803546907,
      "num_tokens": 29572837.0,
      "step": 13490
    },
    {
      "entropy": 5.527883672714234,
      "epoch": 1.3393211591901548,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048280435590709756,
      "loss": 5.1158,
      "mean_token_accuracy": 0.16936204582452774,
      "num_tokens": 29584095.0,
      "step": 13495
    },
    {
      "entropy": 5.551968097686768,
      "epoch": 1.3398173878523223,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004827907682566157,
      "loss": 5.1733,
      "mean_token_accuracy": 0.16764944046735764,
      "num_tokens": 29594903.0,
      "step": 13500
    },
    {
      "entropy": 5.600491905212403,
      "epoch": 1.3403136165144898,
      "grad_norm": 0.875,
      "learning_rate": 0.0004827771754533704,
      "loss": 5.1266,
      "mean_token_accuracy": 0.17046167999505996,
      "num_tokens": 29605199.0,
      "step": 13505
    },
    {
      "entropy": 5.615497827529907,
      "epoch": 1.3408098451766575,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048276357749769895,
      "loss": 5.3054,
      "mean_token_accuracy": 0.16074269488453866,
      "num_tokens": 29617535.0,
      "step": 13510
    },
    {
      "entropy": 5.54388952255249,
      "epoch": 1.341306073838825,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048274997438993845,
      "loss": 5.1119,
      "mean_token_accuracy": 0.1747731849551201,
      "num_tokens": 29628227.0,
      "step": 13515
    },
    {
      "entropy": 5.582678985595703,
      "epoch": 1.3418023025009924,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048273636613042626,
      "loss": 5.2316,
      "mean_token_accuracy": 0.1667240098118782,
      "num_tokens": 29639112.0,
      "step": 13520
    },
    {
      "entropy": 5.500787687301636,
      "epoch": 1.3422985311631601,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004827227527194998,
      "loss": 5.1145,
      "mean_token_accuracy": 0.16373225003480912,
      "num_tokens": 29650768.0,
      "step": 13525
    },
    {
      "entropy": 5.563147640228271,
      "epoch": 1.3427947598253276,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004827091341574967,
      "loss": 5.2197,
      "mean_token_accuracy": 0.17597340792417526,
      "num_tokens": 29662388.0,
      "step": 13530
    },
    {
      "entropy": 5.548911285400391,
      "epoch": 1.343290988487495,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004826955104447546,
      "loss": 5.1121,
      "mean_token_accuracy": 0.1745942071080208,
      "num_tokens": 29672231.0,
      "step": 13535
    },
    {
      "entropy": 5.5248106002807615,
      "epoch": 1.3437872171496625,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004826818815816114,
      "loss": 5.1568,
      "mean_token_accuracy": 0.1730260968208313,
      "num_tokens": 29683019.0,
      "step": 13540
    },
    {
      "entropy": 5.5459236145019535,
      "epoch": 1.34428344581183,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048266824756840513,
      "loss": 5.1853,
      "mean_token_accuracy": 0.1690635859966278,
      "num_tokens": 29694233.0,
      "step": 13545
    },
    {
      "entropy": 5.594381952285767,
      "epoch": 1.3447796744739975,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004826546084054737,
      "loss": 5.2073,
      "mean_token_accuracy": 0.169435915350914,
      "num_tokens": 29704023.0,
      "step": 13550
    },
    {
      "entropy": 5.6499780178070065,
      "epoch": 1.3452759031361652,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004826409640931555,
      "loss": 5.302,
      "mean_token_accuracy": 0.16322341114282607,
      "num_tokens": 29715407.0,
      "step": 13555
    },
    {
      "entropy": 5.49336805343628,
      "epoch": 1.3457721317983327,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00048262731463178884,
      "loss": 5.1254,
      "mean_token_accuracy": 0.1686178997159004,
      "num_tokens": 29726223.0,
      "step": 13560
    },
    {
      "entropy": 5.51274528503418,
      "epoch": 1.3462683604605001,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004826136600217121,
      "loss": 5.0545,
      "mean_token_accuracy": 0.1747345432639122,
      "num_tokens": 29735850.0,
      "step": 13565
    },
    {
      "entropy": 5.562219142913818,
      "epoch": 1.3467645891226678,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000482600000263264,
      "loss": 5.2147,
      "mean_token_accuracy": 0.1668240875005722,
      "num_tokens": 29747381.0,
      "step": 13570
    },
    {
      "entropy": 5.613744258880615,
      "epoch": 1.3472608177848353,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004825863353567832,
      "loss": 5.2523,
      "mean_token_accuracy": 0.1672103449702263,
      "num_tokens": 29758440.0,
      "step": 13575
    },
    {
      "entropy": 5.51864185333252,
      "epoch": 1.3477570464470028,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048257266530260865,
      "loss": 5.1052,
      "mean_token_accuracy": 0.17141530215740203,
      "num_tokens": 29770606.0,
      "step": 13580
    },
    {
      "entropy": 5.505066061019898,
      "epoch": 1.3482532751091703,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004825589901010792,
      "loss": 5.1694,
      "mean_token_accuracy": 0.17213753908872603,
      "num_tokens": 29781469.0,
      "step": 13585
    },
    {
      "entropy": 5.534030342102051,
      "epoch": 1.3487495037713377,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004825453097525341,
      "loss": 5.2354,
      "mean_token_accuracy": 0.17037178575992584,
      "num_tokens": 29792847.0,
      "step": 13590
    },
    {
      "entropy": 5.569099855422974,
      "epoch": 1.3492457324335054,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048253162425731253,
      "loss": 5.1555,
      "mean_token_accuracy": 0.16667943447828293,
      "num_tokens": 29802784.0,
      "step": 13595
    },
    {
      "entropy": 5.613606882095337,
      "epoch": 1.349741961095673,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004825179336157539,
      "loss": 5.1659,
      "mean_token_accuracy": 0.16834889203310013,
      "num_tokens": 29812849.0,
      "step": 13600
    },
    {
      "entropy": 5.5284843921661375,
      "epoch": 1.3502381897578404,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048250423782819765,
      "loss": 5.1411,
      "mean_token_accuracy": 0.1682274267077446,
      "num_tokens": 29823495.0,
      "step": 13605
    },
    {
      "entropy": 5.552795934677124,
      "epoch": 1.350734418420008,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004824905368949834,
      "loss": 5.1903,
      "mean_token_accuracy": 0.16858471184968948,
      "num_tokens": 29835295.0,
      "step": 13610
    },
    {
      "entropy": 5.529150724411011,
      "epoch": 1.3512306470821756,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004824768308164509,
      "loss": 5.1193,
      "mean_token_accuracy": 0.17711340337991716,
      "num_tokens": 29845747.0,
      "step": 13615
    },
    {
      "entropy": 5.5408612251281735,
      "epoch": 1.351726875744343,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004824631195929402,
      "loss": 5.1786,
      "mean_token_accuracy": 0.17181971967220305,
      "num_tokens": 29857111.0,
      "step": 13620
    },
    {
      "entropy": 5.537961673736572,
      "epoch": 1.3522231044065105,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004824494032247911,
      "loss": 5.1826,
      "mean_token_accuracy": 0.1715356096625328,
      "num_tokens": 29866919.0,
      "step": 13625
    },
    {
      "entropy": 5.5413860321044925,
      "epoch": 1.352719333068678,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004824356817123439,
      "loss": 5.1216,
      "mean_token_accuracy": 0.16942113339900972,
      "num_tokens": 29878631.0,
      "step": 13630
    },
    {
      "entropy": 5.544751691818237,
      "epoch": 1.3532155617308455,
      "grad_norm": 0.875,
      "learning_rate": 0.00048242195505593865,
      "loss": 5.1539,
      "mean_token_accuracy": 0.16985422968864441,
      "num_tokens": 29890809.0,
      "step": 13635
    },
    {
      "entropy": 5.5777698993682865,
      "epoch": 1.3537117903930131,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048240822325591596,
      "loss": 5.2335,
      "mean_token_accuracy": 0.16819322258234023,
      "num_tokens": 29901789.0,
      "step": 13640
    },
    {
      "entropy": 5.528992795944214,
      "epoch": 1.3542080190551806,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004823944863126162,
      "loss": 5.152,
      "mean_token_accuracy": 0.16734090596437454,
      "num_tokens": 29912576.0,
      "step": 13645
    },
    {
      "entropy": 5.6027082920074465,
      "epoch": 1.354704247717348,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004823807442263801,
      "loss": 5.2427,
      "mean_token_accuracy": 0.1628077208995819,
      "num_tokens": 29924076.0,
      "step": 13650
    },
    {
      "entropy": 5.54388427734375,
      "epoch": 1.3552004763795158,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048236699699754845,
      "loss": 5.1475,
      "mean_token_accuracy": 0.16844340562820434,
      "num_tokens": 29935854.0,
      "step": 13655
    },
    {
      "entropy": 5.560317659378052,
      "epoch": 1.3556967050416833,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000482353244626462,
      "loss": 5.1113,
      "mean_token_accuracy": 0.1707082286477089,
      "num_tokens": 29946389.0,
      "step": 13660
    },
    {
      "entropy": 5.56740894317627,
      "epoch": 1.3561929337038507,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048233948711346196,
      "loss": 5.1661,
      "mean_token_accuracy": 0.16355991512537002,
      "num_tokens": 29958170.0,
      "step": 13665
    },
    {
      "entropy": 5.475707769393921,
      "epoch": 1.3566891623660182,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048232572445888945,
      "loss": 5.0695,
      "mean_token_accuracy": 0.16982268244028093,
      "num_tokens": 29969067.0,
      "step": 13670
    },
    {
      "entropy": 5.59179310798645,
      "epoch": 1.3571853910281857,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004823119566630857,
      "loss": 5.2087,
      "mean_token_accuracy": 0.16459728628396988,
      "num_tokens": 29979727.0,
      "step": 13675
    },
    {
      "entropy": 5.613336992263794,
      "epoch": 1.3576816196903534,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048229818372639216,
      "loss": 5.1716,
      "mean_token_accuracy": 0.17122318595647812,
      "num_tokens": 29989397.0,
      "step": 13680
    },
    {
      "entropy": 5.461426305770874,
      "epoch": 1.3581778483525209,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048228440564915026,
      "loss": 5.0893,
      "mean_token_accuracy": 0.16765320748090745,
      "num_tokens": 29999415.0,
      "step": 13685
    },
    {
      "entropy": 5.517765140533447,
      "epoch": 1.3586740770146883,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048227062243170186,
      "loss": 5.1292,
      "mean_token_accuracy": 0.17012468576431275,
      "num_tokens": 30010277.0,
      "step": 13690
    },
    {
      "entropy": 5.47207727432251,
      "epoch": 1.3591703056768558,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004822568340743886,
      "loss": 5.0397,
      "mean_token_accuracy": 0.17928745299577714,
      "num_tokens": 30020684.0,
      "step": 13695
    },
    {
      "entropy": 5.622174072265625,
      "epoch": 1.3596665343390235,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004822430405775525,
      "loss": 5.2378,
      "mean_token_accuracy": 0.16153559684753419,
      "num_tokens": 30030916.0,
      "step": 13700
    },
    {
      "entropy": 5.582497549057007,
      "epoch": 1.360162763001191,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004822292419415355,
      "loss": 5.1678,
      "mean_token_accuracy": 0.16640244275331498,
      "num_tokens": 30041140.0,
      "step": 13705
    },
    {
      "entropy": 5.500198459625244,
      "epoch": 1.3606589916633585,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00048221543816667984,
      "loss": 5.1297,
      "mean_token_accuracy": 0.17232650220394136,
      "num_tokens": 30052765.0,
      "step": 13710
    },
    {
      "entropy": 5.496519613265991,
      "epoch": 1.361155220325526,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048220162925332786,
      "loss": 5.15,
      "mean_token_accuracy": 0.1708048865199089,
      "num_tokens": 30062511.0,
      "step": 13715
    },
    {
      "entropy": 5.657969188690186,
      "epoch": 1.3616514489876934,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000482187815201822,
      "loss": 5.2364,
      "mean_token_accuracy": 0.1670539304614067,
      "num_tokens": 30073404.0,
      "step": 13720
    },
    {
      "entropy": 5.501927328109741,
      "epoch": 1.362147677649861,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00048217399601250464,
      "loss": 5.0816,
      "mean_token_accuracy": 0.17502516508102417,
      "num_tokens": 30084978.0,
      "step": 13725
    },
    {
      "entropy": 5.55900616645813,
      "epoch": 1.3626439063120286,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004821601716857187,
      "loss": 5.1526,
      "mean_token_accuracy": 0.17103137522935868,
      "num_tokens": 30094498.0,
      "step": 13730
    },
    {
      "entropy": 5.536233186721802,
      "epoch": 1.363140134974196,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048214634222180686,
      "loss": 5.1838,
      "mean_token_accuracy": 0.17339248061180115,
      "num_tokens": 30104545.0,
      "step": 13735
    },
    {
      "entropy": 5.584895992279053,
      "epoch": 1.3636363636363638,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004821325076211121,
      "loss": 5.2483,
      "mean_token_accuracy": 0.16983823627233505,
      "num_tokens": 30114863.0,
      "step": 13740
    },
    {
      "entropy": 5.584507989883423,
      "epoch": 1.3641325922985312,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048211866788397753,
      "loss": 5.1579,
      "mean_token_accuracy": 0.16757967472076415,
      "num_tokens": 30125101.0,
      "step": 13745
    },
    {
      "entropy": 5.562633943557739,
      "epoch": 1.3646288209606987,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004821048230107462,
      "loss": 5.2232,
      "mean_token_accuracy": 0.16689809411764145,
      "num_tokens": 30135081.0,
      "step": 13750
    },
    {
      "entropy": 5.663543605804444,
      "epoch": 1.3651250496228662,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004820909730017616,
      "loss": 5.2439,
      "mean_token_accuracy": 0.1603972226381302,
      "num_tokens": 30147169.0,
      "step": 13755
    },
    {
      "entropy": 5.575499391555786,
      "epoch": 1.3656212782850337,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048207711785736707,
      "loss": 5.0667,
      "mean_token_accuracy": 0.17057625353336334,
      "num_tokens": 30157523.0,
      "step": 13760
    },
    {
      "entropy": 5.469462966918945,
      "epoch": 1.3661175069472011,
      "grad_norm": 0.83984375,
      "learning_rate": 0.00048206325757790624,
      "loss": 5.1463,
      "mean_token_accuracy": 0.1721128538250923,
      "num_tokens": 30168313.0,
      "step": 13765
    },
    {
      "entropy": 5.427523851394653,
      "epoch": 1.3666137356093688,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048204939216372285,
      "loss": 5.1029,
      "mean_token_accuracy": 0.17009972929954528,
      "num_tokens": 30179173.0,
      "step": 13770
    },
    {
      "entropy": 5.586271905899048,
      "epoch": 1.3671099642715363,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004820355216151607,
      "loss": 5.1943,
      "mean_token_accuracy": 0.16727735698223115,
      "num_tokens": 30191666.0,
      "step": 13775
    },
    {
      "entropy": 5.627197122573852,
      "epoch": 1.3676061929337038,
      "grad_norm": 0.84375,
      "learning_rate": 0.00048202164593256363,
      "loss": 5.1465,
      "mean_token_accuracy": 0.16737007945775986,
      "num_tokens": 30202752.0,
      "step": 13780
    },
    {
      "entropy": 5.533715486526489,
      "epoch": 1.3681024215958715,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048200776511627585,
      "loss": 5.1701,
      "mean_token_accuracy": 0.17021922171115875,
      "num_tokens": 30213127.0,
      "step": 13785
    },
    {
      "entropy": 5.558139133453369,
      "epoch": 1.368598650258039,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004819938791666416,
      "loss": 5.2081,
      "mean_token_accuracy": 0.1704016551375389,
      "num_tokens": 30224782.0,
      "step": 13790
    },
    {
      "entropy": 5.539986085891724,
      "epoch": 1.3690948789202064,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048197998808400524,
      "loss": 5.2144,
      "mean_token_accuracy": 0.17109808027744294,
      "num_tokens": 30235129.0,
      "step": 13795
    },
    {
      "entropy": 5.587945127487183,
      "epoch": 1.369591107582374,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004819660918687111,
      "loss": 5.2157,
      "mean_token_accuracy": 0.16336526721715927,
      "num_tokens": 30247304.0,
      "step": 13800
    },
    {
      "entropy": 5.543242025375366,
      "epoch": 1.3700873362445414,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004819521905211039,
      "loss": 5.0962,
      "mean_token_accuracy": 0.17269397526979446,
      "num_tokens": 30257900.0,
      "step": 13805
    },
    {
      "entropy": 5.602170324325561,
      "epoch": 1.370583564906709,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00048193828404152826,
      "loss": 5.2192,
      "mean_token_accuracy": 0.1688413664698601,
      "num_tokens": 30269741.0,
      "step": 13810
    },
    {
      "entropy": 5.543507671356201,
      "epoch": 1.3710797935688765,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004819243724303291,
      "loss": 5.2156,
      "mean_token_accuracy": 0.16407763957977295,
      "num_tokens": 30279382.0,
      "step": 13815
    },
    {
      "entropy": 5.5421445846557615,
      "epoch": 1.371576022231044,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048191045568785144,
      "loss": 5.1928,
      "mean_token_accuracy": 0.1694382518529892,
      "num_tokens": 30289837.0,
      "step": 13820
    },
    {
      "entropy": 5.526802396774292,
      "epoch": 1.3720722508932117,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00048189653381444024,
      "loss": 5.1353,
      "mean_token_accuracy": 0.17473276555538178,
      "num_tokens": 30302138.0,
      "step": 13825
    },
    {
      "entropy": 5.57561764717102,
      "epoch": 1.3725684795553792,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004818826068104409,
      "loss": 5.2207,
      "mean_token_accuracy": 0.16866065412759781,
      "num_tokens": 30313623.0,
      "step": 13830
    },
    {
      "entropy": 5.564894104003907,
      "epoch": 1.3730647082175467,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004818686746761987,
      "loss": 5.1698,
      "mean_token_accuracy": 0.1662436842918396,
      "num_tokens": 30324533.0,
      "step": 13835
    },
    {
      "entropy": 5.571521854400634,
      "epoch": 1.3735609368797141,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004818547374120591,
      "loss": 5.265,
      "mean_token_accuracy": 0.15774538218975068,
      "num_tokens": 30335819.0,
      "step": 13840
    },
    {
      "entropy": 5.550901889801025,
      "epoch": 1.3740571655418816,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004818407950183678,
      "loss": 5.1465,
      "mean_token_accuracy": 0.1719643294811249,
      "num_tokens": 30345947.0,
      "step": 13845
    },
    {
      "entropy": 5.523833751678467,
      "epoch": 1.374553394204049,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004818268474954704,
      "loss": 5.1193,
      "mean_token_accuracy": 0.17062321901321412,
      "num_tokens": 30356972.0,
      "step": 13850
    },
    {
      "entropy": 5.599646186828613,
      "epoch": 1.3750496228662168,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004818128948437129,
      "loss": 5.2143,
      "mean_token_accuracy": 0.1654777243733406,
      "num_tokens": 30368414.0,
      "step": 13855
    },
    {
      "entropy": 5.581432962417603,
      "epoch": 1.3755458515283843,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004817989370634412,
      "loss": 5.142,
      "mean_token_accuracy": 0.17397802621126174,
      "num_tokens": 30379141.0,
      "step": 13860
    },
    {
      "entropy": 5.524407482147216,
      "epoch": 1.3760420801905517,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048178497415500153,
      "loss": 5.0933,
      "mean_token_accuracy": 0.1796994298696518,
      "num_tokens": 30388985.0,
      "step": 13865
    },
    {
      "entropy": 5.506087160110473,
      "epoch": 1.3765383088527194,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004817710061187401,
      "loss": 5.1407,
      "mean_token_accuracy": 0.17578205913305284,
      "num_tokens": 30399599.0,
      "step": 13870
    },
    {
      "entropy": 5.57473669052124,
      "epoch": 1.377034537514887,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00048175703295500323,
      "loss": 5.229,
      "mean_token_accuracy": 0.1658189907670021,
      "num_tokens": 30411143.0,
      "step": 13875
    },
    {
      "entropy": 5.628584384918213,
      "epoch": 1.3775307661770544,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048174305466413755,
      "loss": 5.1992,
      "mean_token_accuracy": 0.1683528795838356,
      "num_tokens": 30422339.0,
      "step": 13880
    },
    {
      "entropy": 5.548181581497192,
      "epoch": 1.3780269948392219,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048172907124648947,
      "loss": 5.1343,
      "mean_token_accuracy": 0.17108379751443864,
      "num_tokens": 30432588.0,
      "step": 13885
    },
    {
      "entropy": 5.568365097045898,
      "epoch": 1.3785232235013893,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004817150827024059,
      "loss": 5.1966,
      "mean_token_accuracy": 0.1727196380496025,
      "num_tokens": 30442272.0,
      "step": 13890
    },
    {
      "entropy": 5.5229079246521,
      "epoch": 1.379019452163557,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004817010890322338,
      "loss": 5.1108,
      "mean_token_accuracy": 0.1727242201566696,
      "num_tokens": 30453142.0,
      "step": 13895
    },
    {
      "entropy": 5.565318965911866,
      "epoch": 1.3795156808257245,
      "grad_norm": 1.0,
      "learning_rate": 0.00048168709023632006,
      "loss": 5.2039,
      "mean_token_accuracy": 0.16730950325727462,
      "num_tokens": 30465136.0,
      "step": 13900
    },
    {
      "entropy": 5.606248474121093,
      "epoch": 1.380011909487892,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048167308631501184,
      "loss": 5.1513,
      "mean_token_accuracy": 0.1690399691462517,
      "num_tokens": 30476511.0,
      "step": 13905
    },
    {
      "entropy": 5.560820484161377,
      "epoch": 1.3805081381500597,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048165907726865643,
      "loss": 5.2592,
      "mean_token_accuracy": 0.16012985557317733,
      "num_tokens": 30486547.0,
      "step": 13910
    },
    {
      "entropy": 5.591954755783081,
      "epoch": 1.3810043668122272,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004816450630976012,
      "loss": 5.2104,
      "mean_token_accuracy": 0.17533600628376006,
      "num_tokens": 30496163.0,
      "step": 13915
    },
    {
      "entropy": 5.546323442459107,
      "epoch": 1.3815005954743946,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004816310438021937,
      "loss": 5.2027,
      "mean_token_accuracy": 0.16715382486581803,
      "num_tokens": 30506979.0,
      "step": 13920
    },
    {
      "entropy": 5.561068534851074,
      "epoch": 1.381996824136562,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048161701938278156,
      "loss": 5.233,
      "mean_token_accuracy": 0.15811349898576738,
      "num_tokens": 30517492.0,
      "step": 13925
    },
    {
      "entropy": 5.510106897354126,
      "epoch": 1.3824930527987296,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048160298983971256,
      "loss": 5.1111,
      "mean_token_accuracy": 0.17669742405414582,
      "num_tokens": 30528614.0,
      "step": 13930
    },
    {
      "entropy": 5.551264476776123,
      "epoch": 1.382989281460897,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004815889551733346,
      "loss": 5.2056,
      "mean_token_accuracy": 0.16688290983438492,
      "num_tokens": 30539479.0,
      "step": 13935
    },
    {
      "entropy": 5.602884340286255,
      "epoch": 1.3834855101230648,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048157491538399573,
      "loss": 5.2234,
      "mean_token_accuracy": 0.16950736194849014,
      "num_tokens": 30549909.0,
      "step": 13940
    },
    {
      "entropy": 5.550446462631226,
      "epoch": 1.3839817387852322,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048156087047204407,
      "loss": 5.1272,
      "mean_token_accuracy": 0.17203201353549957,
      "num_tokens": 30562265.0,
      "step": 13945
    },
    {
      "entropy": 5.57637996673584,
      "epoch": 1.3844779674473997,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004815468204378279,
      "loss": 5.1589,
      "mean_token_accuracy": 0.16654592901468276,
      "num_tokens": 30572622.0,
      "step": 13950
    },
    {
      "entropy": 5.519306898117065,
      "epoch": 1.3849741961095674,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004815327652816957,
      "loss": 5.1355,
      "mean_token_accuracy": 0.17387963980436325,
      "num_tokens": 30583300.0,
      "step": 13955
    },
    {
      "entropy": 5.50309271812439,
      "epoch": 1.3854704247717349,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004815187050039959,
      "loss": 5.0863,
      "mean_token_accuracy": 0.17074283510446547,
      "num_tokens": 30594460.0,
      "step": 13960
    },
    {
      "entropy": 5.5312470436096195,
      "epoch": 1.3859666534339024,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048150463960507724,
      "loss": 5.1283,
      "mean_token_accuracy": 0.17126694470643997,
      "num_tokens": 30604658.0,
      "step": 13965
    },
    {
      "entropy": 5.569334316253662,
      "epoch": 1.3864628820960698,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00048149056908528855,
      "loss": 5.1333,
      "mean_token_accuracy": 0.16687559485435485,
      "num_tokens": 30616151.0,
      "step": 13970
    },
    {
      "entropy": 5.530208683013916,
      "epoch": 1.3869591107582373,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004814764934449786,
      "loss": 5.1825,
      "mean_token_accuracy": 0.17159994393587114,
      "num_tokens": 30627560.0,
      "step": 13975
    },
    {
      "entropy": 5.524506521224976,
      "epoch": 1.387455339420405,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048146241268449665,
      "loss": 5.163,
      "mean_token_accuracy": 0.16732659041881562,
      "num_tokens": 30638490.0,
      "step": 13980
    },
    {
      "entropy": 5.600838327407837,
      "epoch": 1.3879515680825725,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004814483268041917,
      "loss": 5.2644,
      "mean_token_accuracy": 0.16585466265678406,
      "num_tokens": 30649501.0,
      "step": 13985
    },
    {
      "entropy": 5.56956901550293,
      "epoch": 1.38844779674474,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004814342358044131,
      "loss": 5.1443,
      "mean_token_accuracy": 0.17039851397275924,
      "num_tokens": 30660479.0,
      "step": 13990
    },
    {
      "entropy": 5.6756126403808596,
      "epoch": 1.3889440254069074,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048142013968551026,
      "loss": 5.3193,
      "mean_token_accuracy": 0.15933064222335816,
      "num_tokens": 30672015.0,
      "step": 13995
    },
    {
      "entropy": 5.475689506530761,
      "epoch": 1.3894402540690751,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048140603844783273,
      "loss": 5.1746,
      "mean_token_accuracy": 0.16900611966848372,
      "num_tokens": 30682612.0,
      "step": 14000
    },
    {
      "entropy": 5.476071882247925,
      "epoch": 1.3899364827312426,
      "grad_norm": 0.875,
      "learning_rate": 0.0004813919320917303,
      "loss": 5.1163,
      "mean_token_accuracy": 0.16966806650161742,
      "num_tokens": 30693289.0,
      "step": 14005
    },
    {
      "entropy": 5.544860315322876,
      "epoch": 1.39043271139341,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048137782061755265,
      "loss": 5.1525,
      "mean_token_accuracy": 0.17297138124704362,
      "num_tokens": 30705346.0,
      "step": 14010
    },
    {
      "entropy": 5.522922992706299,
      "epoch": 1.3909289400555775,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048136370402564974,
      "loss": 5.1591,
      "mean_token_accuracy": 0.1646505579352379,
      "num_tokens": 30715968.0,
      "step": 14015
    },
    {
      "entropy": 5.53752064704895,
      "epoch": 1.391425168717745,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004813495823163716,
      "loss": 5.1106,
      "mean_token_accuracy": 0.1739235997200012,
      "num_tokens": 30726225.0,
      "step": 14020
    },
    {
      "entropy": 5.503029441833496,
      "epoch": 1.3919213973799127,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004813354554900685,
      "loss": 5.0886,
      "mean_token_accuracy": 0.17494098991155624,
      "num_tokens": 30735603.0,
      "step": 14025
    },
    {
      "entropy": 5.4948155879974365,
      "epoch": 1.3924176260420802,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004813213235470907,
      "loss": 5.1365,
      "mean_token_accuracy": 0.1694853588938713,
      "num_tokens": 30747414.0,
      "step": 14030
    },
    {
      "entropy": 5.570358467102051,
      "epoch": 1.3929138547042477,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004813071864877887,
      "loss": 5.1944,
      "mean_token_accuracy": 0.16979669034481049,
      "num_tokens": 30758389.0,
      "step": 14035
    },
    {
      "entropy": 5.578168058395386,
      "epoch": 1.3934100833664154,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000481293044312513,
      "loss": 5.2126,
      "mean_token_accuracy": 0.16620449870824813,
      "num_tokens": 30769484.0,
      "step": 14040
    },
    {
      "entropy": 5.5113927841186525,
      "epoch": 1.3939063120285828,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048127889702161435,
      "loss": 5.1298,
      "mean_token_accuracy": 0.17630065977573395,
      "num_tokens": 30779775.0,
      "step": 14045
    },
    {
      "entropy": 5.589673566818237,
      "epoch": 1.3944025406907503,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004812647446154435,
      "loss": 5.2281,
      "mean_token_accuracy": 0.16023280918598176,
      "num_tokens": 30791456.0,
      "step": 14050
    },
    {
      "entropy": 5.52322096824646,
      "epoch": 1.3948987693529178,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004812505870943515,
      "loss": 5.0514,
      "mean_token_accuracy": 0.1758187085390091,
      "num_tokens": 30801659.0,
      "step": 14055
    },
    {
      "entropy": 5.615298700332642,
      "epoch": 1.3953949980150853,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004812364244586893,
      "loss": 5.2849,
      "mean_token_accuracy": 0.16573704034090042,
      "num_tokens": 30813389.0,
      "step": 14060
    },
    {
      "entropy": 5.473172521591186,
      "epoch": 1.395891226677253,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004812222567088082,
      "loss": 5.0484,
      "mean_token_accuracy": 0.1681053012609482,
      "num_tokens": 30823725.0,
      "step": 14065
    },
    {
      "entropy": 5.481383562088013,
      "epoch": 1.3963874553394204,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004812080838450596,
      "loss": 5.056,
      "mean_token_accuracy": 0.17506962269544601,
      "num_tokens": 30834448.0,
      "step": 14070
    },
    {
      "entropy": 5.5361066341400145,
      "epoch": 1.396883684001588,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004811939058677947,
      "loss": 5.1252,
      "mean_token_accuracy": 0.16739902049303054,
      "num_tokens": 30845891.0,
      "step": 14075
    },
    {
      "entropy": 5.497674369812012,
      "epoch": 1.3973799126637554,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004811797227773654,
      "loss": 5.0618,
      "mean_token_accuracy": 0.17847796231508256,
      "num_tokens": 30855170.0,
      "step": 14080
    },
    {
      "entropy": 5.489937591552734,
      "epoch": 1.397876141325923,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048116553457412317,
      "loss": 5.1531,
      "mean_token_accuracy": 0.16629391461610793,
      "num_tokens": 30866527.0,
      "step": 14085
    },
    {
      "entropy": 5.516706848144532,
      "epoch": 1.3983723699880906,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048115134125841993,
      "loss": 5.0908,
      "mean_token_accuracy": 0.1756426826119423,
      "num_tokens": 30876292.0,
      "step": 14090
    },
    {
      "entropy": 5.534853887557984,
      "epoch": 1.398868598650258,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004811371428306076,
      "loss": 5.0916,
      "mean_token_accuracy": 0.16869684159755707,
      "num_tokens": 30887440.0,
      "step": 14095
    },
    {
      "entropy": 5.500794744491577,
      "epoch": 1.3993648273124255,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004811229392910384,
      "loss": 5.1372,
      "mean_token_accuracy": 0.17133892476558685,
      "num_tokens": 30897925.0,
      "step": 14100
    },
    {
      "entropy": 5.514557552337647,
      "epoch": 1.399861055974593,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004811087306400644,
      "loss": 5.1605,
      "mean_token_accuracy": 0.17588270008563994,
      "num_tokens": 30908413.0,
      "step": 14105
    },
    {
      "entropy": 5.584716844558716,
      "epoch": 1.4003572846367607,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048109451687803804,
      "loss": 5.2048,
      "mean_token_accuracy": 0.16762253493070603,
      "num_tokens": 30919149.0,
      "step": 14110
    },
    {
      "entropy": 5.5883903980255125,
      "epoch": 1.4008535132989282,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004810802980053117,
      "loss": 5.2168,
      "mean_token_accuracy": 0.17158247977495195,
      "num_tokens": 30929883.0,
      "step": 14115
    },
    {
      "entropy": 5.510629367828369,
      "epoch": 1.4013497419610956,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048106607402223815,
      "loss": 5.1281,
      "mean_token_accuracy": 0.16870941817760468,
      "num_tokens": 30941358.0,
      "step": 14120
    },
    {
      "entropy": 5.556718921661377,
      "epoch": 1.4018459706232633,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048105184492916995,
      "loss": 5.1314,
      "mean_token_accuracy": 0.16843397319316863,
      "num_tokens": 30953289.0,
      "step": 14125
    },
    {
      "entropy": 5.51023678779602,
      "epoch": 1.4023421992854308,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048103761072645993,
      "loss": 5.0589,
      "mean_token_accuracy": 0.17450836300849915,
      "num_tokens": 30964304.0,
      "step": 14130
    },
    {
      "entropy": 5.471431589126587,
      "epoch": 1.4028384279475983,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048102337141446116,
      "loss": 5.0987,
      "mean_token_accuracy": 0.17722263038158417,
      "num_tokens": 30974840.0,
      "step": 14135
    },
    {
      "entropy": 5.512514352798462,
      "epoch": 1.4033346566097658,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048100912699352666,
      "loss": 5.1543,
      "mean_token_accuracy": 0.17313008904457092,
      "num_tokens": 30985805.0,
      "step": 14140
    },
    {
      "entropy": 5.523301982879639,
      "epoch": 1.4038308852719332,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004809948774640098,
      "loss": 5.1589,
      "mean_token_accuracy": 0.17004996538162231,
      "num_tokens": 30998174.0,
      "step": 14145
    },
    {
      "entropy": 5.565105247497558,
      "epoch": 1.4043271139341007,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004809806228262639,
      "loss": 5.1708,
      "mean_token_accuracy": 0.16836102306842804,
      "num_tokens": 31010050.0,
      "step": 14150
    },
    {
      "entropy": 5.646402502059937,
      "epoch": 1.4048233425962684,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048096636308064236,
      "loss": 5.2904,
      "mean_token_accuracy": 0.16075037270784379,
      "num_tokens": 31023069.0,
      "step": 14155
    },
    {
      "entropy": 5.5874903202056885,
      "epoch": 1.4053195712584359,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004809520982274989,
      "loss": 5.0953,
      "mean_token_accuracy": 0.17480895817279815,
      "num_tokens": 31033512.0,
      "step": 14160
    },
    {
      "entropy": 5.53944730758667,
      "epoch": 1.4058157999206033,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048093782826718706,
      "loss": 5.1332,
      "mean_token_accuracy": 0.16568226963281632,
      "num_tokens": 31045731.0,
      "step": 14165
    },
    {
      "entropy": 5.488059997558594,
      "epoch": 1.406312028582771,
      "grad_norm": 0.875,
      "learning_rate": 0.00048092355320006086,
      "loss": 5.0563,
      "mean_token_accuracy": 0.17808159291744233,
      "num_tokens": 31055443.0,
      "step": 14170
    },
    {
      "entropy": 5.4675720691680905,
      "epoch": 1.4068082572449385,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004809092730264744,
      "loss": 5.1355,
      "mean_token_accuracy": 0.17498398423194886,
      "num_tokens": 31067575.0,
      "step": 14175
    },
    {
      "entropy": 5.55556354522705,
      "epoch": 1.407304485907106,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048089498774678155,
      "loss": 5.2358,
      "mean_token_accuracy": 0.16588225066661835,
      "num_tokens": 31078198.0,
      "step": 14180
    },
    {
      "entropy": 5.530391883850098,
      "epoch": 1.4078007145692735,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00048088069736133664,
      "loss": 5.1155,
      "mean_token_accuracy": 0.16887167096138,
      "num_tokens": 31089961.0,
      "step": 14185
    },
    {
      "entropy": 5.524352073669434,
      "epoch": 1.408296943231441,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004808664018704941,
      "loss": 5.1244,
      "mean_token_accuracy": 0.16945887804031373,
      "num_tokens": 31100534.0,
      "step": 14190
    },
    {
      "entropy": 5.625082921981812,
      "epoch": 1.4087931718936086,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004808521012746085,
      "loss": 5.1527,
      "mean_token_accuracy": 0.17784729450941086,
      "num_tokens": 31112016.0,
      "step": 14195
    },
    {
      "entropy": 5.5022547245025635,
      "epoch": 1.4092894005557761,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00048083779557403424,
      "loss": 5.1955,
      "mean_token_accuracy": 0.16965552270412446,
      "num_tokens": 31123659.0,
      "step": 14200
    },
    {
      "entropy": 5.548771905899048,
      "epoch": 1.4097856292179436,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004808234847691262,
      "loss": 5.1352,
      "mean_token_accuracy": 0.16782886236906053,
      "num_tokens": 31133872.0,
      "step": 14205
    },
    {
      "entropy": 5.587513971328735,
      "epoch": 1.4102818578801113,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004808091688602393,
      "loss": 5.1754,
      "mean_token_accuracy": 0.16750812232494355,
      "num_tokens": 31145834.0,
      "step": 14210
    },
    {
      "entropy": 5.561975383758545,
      "epoch": 1.4107780865422788,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048079484784772854,
      "loss": 5.2513,
      "mean_token_accuracy": 0.16297162771224977,
      "num_tokens": 31157037.0,
      "step": 14215
    },
    {
      "entropy": 5.539417314529419,
      "epoch": 1.4112743152044462,
      "grad_norm": 1.125,
      "learning_rate": 0.00048078052173194886,
      "loss": 5.1374,
      "mean_token_accuracy": 0.17294942885637282,
      "num_tokens": 31166812.0,
      "step": 14220
    },
    {
      "entropy": 5.551455640792847,
      "epoch": 1.4117705438666137,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004807661905132557,
      "loss": 5.1318,
      "mean_token_accuracy": 0.1633587971329689,
      "num_tokens": 31177162.0,
      "step": 14225
    },
    {
      "entropy": 5.552932548522949,
      "epoch": 1.4122667725287812,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004807518541920045,
      "loss": 5.1582,
      "mean_token_accuracy": 0.17218452095985412,
      "num_tokens": 31187333.0,
      "step": 14230
    },
    {
      "entropy": 5.545293998718262,
      "epoch": 1.4127630011909487,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048073751276855056,
      "loss": 5.1727,
      "mean_token_accuracy": 0.16789565086364747,
      "num_tokens": 31197869.0,
      "step": 14235
    },
    {
      "entropy": 5.540854740142822,
      "epoch": 1.4132592298531164,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004807231662432497,
      "loss": 5.1834,
      "mean_token_accuracy": 0.16220247000455856,
      "num_tokens": 31208199.0,
      "step": 14240
    },
    {
      "entropy": 5.563972520828247,
      "epoch": 1.4137554585152838,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004807088146164575,
      "loss": 5.1633,
      "mean_token_accuracy": 0.1621552437543869,
      "num_tokens": 31218903.0,
      "step": 14245
    },
    {
      "entropy": 5.5235213279724125,
      "epoch": 1.4142516871774513,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004806944578885301,
      "loss": 5.1668,
      "mean_token_accuracy": 0.16414968222379683,
      "num_tokens": 31231146.0,
      "step": 14250
    },
    {
      "entropy": 5.532286977767944,
      "epoch": 1.414747915839619,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048068009605982333,
      "loss": 5.095,
      "mean_token_accuracy": 0.17158334255218505,
      "num_tokens": 31240980.0,
      "step": 14255
    },
    {
      "entropy": 5.51092381477356,
      "epoch": 1.4152441445017865,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048066572913069334,
      "loss": 5.1316,
      "mean_token_accuracy": 0.1706465005874634,
      "num_tokens": 31252458.0,
      "step": 14260
    },
    {
      "entropy": 5.501100158691406,
      "epoch": 1.415740373163954,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004806513571014965,
      "loss": 5.1419,
      "mean_token_accuracy": 0.16705355495214463,
      "num_tokens": 31262979.0,
      "step": 14265
    },
    {
      "entropy": 5.523218441009521,
      "epoch": 1.4162366018261214,
      "grad_norm": 1.0,
      "learning_rate": 0.0004806369799725891,
      "loss": 5.1156,
      "mean_token_accuracy": 0.1683339312672615,
      "num_tokens": 31274811.0,
      "step": 14270
    },
    {
      "entropy": 5.510698366165161,
      "epoch": 1.416732830488289,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004806225977443277,
      "loss": 5.0926,
      "mean_token_accuracy": 0.1792379304766655,
      "num_tokens": 31285573.0,
      "step": 14275
    },
    {
      "entropy": 5.609202814102173,
      "epoch": 1.4172290591504566,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004806082104170689,
      "loss": 5.2536,
      "mean_token_accuracy": 0.15575346797704698,
      "num_tokens": 31296993.0,
      "step": 14280
    },
    {
      "entropy": 5.49761552810669,
      "epoch": 1.417725287812624,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00048059381799116967,
      "loss": 5.1388,
      "mean_token_accuracy": 0.1710622638463974,
      "num_tokens": 31307217.0,
      "step": 14285
    },
    {
      "entropy": 5.503528642654419,
      "epoch": 1.4182215164747916,
      "grad_norm": 1.0,
      "learning_rate": 0.0004805794204669867,
      "loss": 5.1675,
      "mean_token_accuracy": 0.16914174258708953,
      "num_tokens": 31317939.0,
      "step": 14290
    },
    {
      "entropy": 5.604456090927124,
      "epoch": 1.418717745136959,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048056501784487707,
      "loss": 5.1356,
      "mean_token_accuracy": 0.1669573113322258,
      "num_tokens": 31327731.0,
      "step": 14295
    },
    {
      "entropy": 5.688533210754395,
      "epoch": 1.4192139737991267,
      "grad_norm": 1.03125,
      "learning_rate": 0.000480550610125198,
      "loss": 5.3187,
      "mean_token_accuracy": 0.155430468916893,
      "num_tokens": 31339388.0,
      "step": 14300
    },
    {
      "entropy": 5.478641843795776,
      "epoch": 1.4197102024612942,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004805361973083066,
      "loss": 5.003,
      "mean_token_accuracy": 0.1770283043384552,
      "num_tokens": 31350186.0,
      "step": 14305
    },
    {
      "entropy": 5.53291425704956,
      "epoch": 1.4202064311234617,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004805217793945605,
      "loss": 5.1719,
      "mean_token_accuracy": 0.16603127866983414,
      "num_tokens": 31361219.0,
      "step": 14310
    },
    {
      "entropy": 5.526129961013794,
      "epoch": 1.4207026597856292,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004805073563843171,
      "loss": 5.1992,
      "mean_token_accuracy": 0.16637402176856994,
      "num_tokens": 31372820.0,
      "step": 14315
    },
    {
      "entropy": 5.534232807159424,
      "epoch": 1.4211988884477966,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048049292827793413,
      "loss": 5.0818,
      "mean_token_accuracy": 0.17799466252326965,
      "num_tokens": 31382779.0,
      "step": 14320
    },
    {
      "entropy": 5.611496067047119,
      "epoch": 1.4216951171099643,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004804784950757693,
      "loss": 5.3072,
      "mean_token_accuracy": 0.1644120469689369,
      "num_tokens": 31394872.0,
      "step": 14325
    },
    {
      "entropy": 5.580011940002441,
      "epoch": 1.4221913457721318,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048046405677818057,
      "loss": 5.1921,
      "mean_token_accuracy": 0.1673213794827461,
      "num_tokens": 31405767.0,
      "step": 14330
    },
    {
      "entropy": 5.563827133178711,
      "epoch": 1.4226875744342993,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048044961338552595,
      "loss": 5.1744,
      "mean_token_accuracy": 0.16784346848726273,
      "num_tokens": 31417255.0,
      "step": 14335
    },
    {
      "entropy": 5.54085488319397,
      "epoch": 1.423183803096467,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004804351648981636,
      "loss": 5.1488,
      "mean_token_accuracy": 0.17346104085445405,
      "num_tokens": 31428494.0,
      "step": 14340
    },
    {
      "entropy": 5.522696399688721,
      "epoch": 1.4236800317586344,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004804207113164519,
      "loss": 5.0535,
      "mean_token_accuracy": 0.17751450538635255,
      "num_tokens": 31439221.0,
      "step": 14345
    },
    {
      "entropy": 5.544555425643921,
      "epoch": 1.424176260420802,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00048040625264074915,
      "loss": 5.1371,
      "mean_token_accuracy": 0.1722538113594055,
      "num_tokens": 31450575.0,
      "step": 14350
    },
    {
      "entropy": 5.453340625762939,
      "epoch": 1.4246724890829694,
      "grad_norm": 0.875,
      "learning_rate": 0.00048039178887141404,
      "loss": 5.1356,
      "mean_token_accuracy": 0.17231809943914414,
      "num_tokens": 31462168.0,
      "step": 14355
    },
    {
      "entropy": 5.526081132888794,
      "epoch": 1.4251687177451369,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048037732000880503,
      "loss": 5.1599,
      "mean_token_accuracy": 0.17401762157678605,
      "num_tokens": 31472619.0,
      "step": 14360
    },
    {
      "entropy": 5.584140920639038,
      "epoch": 1.4256649464073046,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048036284605328106,
      "loss": 5.2268,
      "mean_token_accuracy": 0.16818679571151735,
      "num_tokens": 31485246.0,
      "step": 14365
    },
    {
      "entropy": 5.550671482086182,
      "epoch": 1.426161175069472,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004803483670052011,
      "loss": 5.1557,
      "mean_token_accuracy": 0.1648237943649292,
      "num_tokens": 31496258.0,
      "step": 14370
    },
    {
      "entropy": 5.537722778320313,
      "epoch": 1.4266574037316395,
      "grad_norm": 0.96875,
      "learning_rate": 0.000480333882864924,
      "loss": 5.1164,
      "mean_token_accuracy": 0.17127072364091872,
      "num_tokens": 31506671.0,
      "step": 14375
    },
    {
      "entropy": 5.574142408370972,
      "epoch": 1.427153632393807,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004803193936328092,
      "loss": 5.1543,
      "mean_token_accuracy": 0.16316839307546616,
      "num_tokens": 31517001.0,
      "step": 14380
    },
    {
      "entropy": 5.549087429046631,
      "epoch": 1.4276498610559747,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004803048993092158,
      "loss": 5.1566,
      "mean_token_accuracy": 0.1722284361720085,
      "num_tokens": 31529649.0,
      "step": 14385
    },
    {
      "entropy": 5.4671769618988035,
      "epoch": 1.4281460897181422,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048029039989450317,
      "loss": 5.0957,
      "mean_token_accuracy": 0.1762278273701668,
      "num_tokens": 31540083.0,
      "step": 14390
    },
    {
      "entropy": 5.577415323257446,
      "epoch": 1.4286423183803096,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048027589538903115,
      "loss": 5.1908,
      "mean_token_accuracy": 0.16957178115844726,
      "num_tokens": 31551400.0,
      "step": 14395
    },
    {
      "entropy": 5.592285776138306,
      "epoch": 1.4291385470424771,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004802613857931592,
      "loss": 5.2006,
      "mean_token_accuracy": 0.16429416537284852,
      "num_tokens": 31562430.0,
      "step": 14400
    },
    {
      "entropy": 5.598383378982544,
      "epoch": 1.4296347757046446,
      "grad_norm": 1.0,
      "learning_rate": 0.0004802468711072472,
      "loss": 5.2389,
      "mean_token_accuracy": 0.16840045303106307,
      "num_tokens": 31573173.0,
      "step": 14405
    },
    {
      "entropy": 5.570560121536255,
      "epoch": 1.4301310043668123,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000480232351331655,
      "loss": 5.1784,
      "mean_token_accuracy": 0.16820122450590133,
      "num_tokens": 31584072.0,
      "step": 14410
    },
    {
      "entropy": 5.487016344070435,
      "epoch": 1.4306272330289798,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048021782646674277,
      "loss": 5.1073,
      "mean_token_accuracy": 0.17149910926818848,
      "num_tokens": 31594993.0,
      "step": 14415
    },
    {
      "entropy": 5.5244057178497314,
      "epoch": 1.4311234616911472,
      "grad_norm": 1.0,
      "learning_rate": 0.00048020329651287064,
      "loss": 5.1434,
      "mean_token_accuracy": 0.17616817504167556,
      "num_tokens": 31604284.0,
      "step": 14420
    },
    {
      "entropy": 5.498825597763061,
      "epoch": 1.431619690353315,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000480188761470399,
      "loss": 5.2082,
      "mean_token_accuracy": 0.17070335745811463,
      "num_tokens": 31616188.0,
      "step": 14425
    },
    {
      "entropy": 5.595587110519409,
      "epoch": 1.4321159190154824,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004801742213396881,
      "loss": 5.1826,
      "mean_token_accuracy": 0.17196833193302155,
      "num_tokens": 31626606.0,
      "step": 14430
    },
    {
      "entropy": 5.502880573272705,
      "epoch": 1.4326121476776499,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048015967612109875,
      "loss": 5.0826,
      "mean_token_accuracy": 0.17779795527458192,
      "num_tokens": 31638050.0,
      "step": 14435
    },
    {
      "entropy": 5.54227499961853,
      "epoch": 1.4331083763398174,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048014512581499136,
      "loss": 5.1027,
      "mean_token_accuracy": 0.16992465257644654,
      "num_tokens": 31648451.0,
      "step": 14440
    },
    {
      "entropy": 5.509846973419189,
      "epoch": 1.4336046050019848,
      "grad_norm": 0.890625,
      "learning_rate": 0.000480130570421727,
      "loss": 5.1274,
      "mean_token_accuracy": 0.16722584366798401,
      "num_tokens": 31659310.0,
      "step": 14445
    },
    {
      "entropy": 5.508379030227661,
      "epoch": 1.4341008336641523,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004801160099416665,
      "loss": 5.1524,
      "mean_token_accuracy": 0.17036597728729247,
      "num_tokens": 31669594.0,
      "step": 14450
    },
    {
      "entropy": 5.557476282119751,
      "epoch": 1.43459706232632,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004801014443751709,
      "loss": 5.1491,
      "mean_token_accuracy": 0.1631366103887558,
      "num_tokens": 31682034.0,
      "step": 14455
    },
    {
      "entropy": 5.5781943798065186,
      "epoch": 1.4350932909884875,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004800868737226015,
      "loss": 5.1849,
      "mean_token_accuracy": 0.17017990052700044,
      "num_tokens": 31692390.0,
      "step": 14460
    },
    {
      "entropy": 5.553547954559326,
      "epoch": 1.435589519650655,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048007229798431946,
      "loss": 5.1581,
      "mean_token_accuracy": 0.17409306913614273,
      "num_tokens": 31703690.0,
      "step": 14465
    },
    {
      "entropy": 5.571557331085205,
      "epoch": 1.4360857483128227,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004800577171606864,
      "loss": 5.1657,
      "mean_token_accuracy": 0.16779595464468003,
      "num_tokens": 31714655.0,
      "step": 14470
    },
    {
      "entropy": 5.5033680438995365,
      "epoch": 1.4365819769749901,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004800431312520638,
      "loss": 5.0764,
      "mean_token_accuracy": 0.17734036296606065,
      "num_tokens": 31727022.0,
      "step": 14475
    },
    {
      "entropy": 5.552882957458496,
      "epoch": 1.4370782056371576,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004800285402588134,
      "loss": 5.1085,
      "mean_token_accuracy": 0.17871950417757035,
      "num_tokens": 31738071.0,
      "step": 14480
    },
    {
      "entropy": 5.588936042785645,
      "epoch": 1.437574434299325,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004800139441812969,
      "loss": 5.2334,
      "mean_token_accuracy": 0.16277743577957154,
      "num_tokens": 31749342.0,
      "step": 14485
    },
    {
      "entropy": 5.597983121871948,
      "epoch": 1.4380706629614926,
      "grad_norm": 0.90625,
      "learning_rate": 0.00047999934301987644,
      "loss": 5.2065,
      "mean_token_accuracy": 0.1663838341832161,
      "num_tokens": 31761218.0,
      "step": 14490
    },
    {
      "entropy": 5.555614423751831,
      "epoch": 1.4385668916236602,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000479984736774914,
      "loss": 5.2175,
      "mean_token_accuracy": 0.16957231163978576,
      "num_tokens": 31773111.0,
      "step": 14495
    },
    {
      "entropy": 5.590849876403809,
      "epoch": 1.4390631202858277,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047997012544677176,
      "loss": 5.1684,
      "mean_token_accuracy": 0.16931167244911194,
      "num_tokens": 31782948.0,
      "step": 14500
    },
    {
      "entropy": 5.618510818481445,
      "epoch": 1.4395593489479952,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004799555090358121,
      "loss": 5.153,
      "mean_token_accuracy": 0.1706443354487419,
      "num_tokens": 31793992.0,
      "step": 14505
    },
    {
      "entropy": 5.5874392032623295,
      "epoch": 1.440055577610163,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004799408875423975,
      "loss": 5.1657,
      "mean_token_accuracy": 0.16228770464658737,
      "num_tokens": 31804830.0,
      "step": 14510
    },
    {
      "entropy": 5.493667936325073,
      "epoch": 1.4405518062723304,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004799262609668904,
      "loss": 5.152,
      "mean_token_accuracy": 0.16971576064825059,
      "num_tokens": 31817200.0,
      "step": 14515
    },
    {
      "entropy": 5.541617822647095,
      "epoch": 1.4410480349344978,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004799116293096537,
      "loss": 5.1986,
      "mean_token_accuracy": 0.17282797545194625,
      "num_tokens": 31828226.0,
      "step": 14520
    },
    {
      "entropy": 5.529183864593506,
      "epoch": 1.4415442635966653,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004798969925710501,
      "loss": 5.0854,
      "mean_token_accuracy": 0.1796392396092415,
      "num_tokens": 31839635.0,
      "step": 14525
    },
    {
      "entropy": 5.54269700050354,
      "epoch": 1.4420404922588328,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047988235075144254,
      "loss": 5.1265,
      "mean_token_accuracy": 0.17201922684907914,
      "num_tokens": 31850333.0,
      "step": 14530
    },
    {
      "entropy": 5.543643379211426,
      "epoch": 1.4425367209210003,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004798677038511942,
      "loss": 5.1091,
      "mean_token_accuracy": 0.16935644000768663,
      "num_tokens": 31861002.0,
      "step": 14535
    },
    {
      "entropy": 5.570112562179565,
      "epoch": 1.443032949583168,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004798530518706683,
      "loss": 5.1883,
      "mean_token_accuracy": 0.1678013578057289,
      "num_tokens": 31871713.0,
      "step": 14540
    },
    {
      "entropy": 5.563977193832398,
      "epoch": 1.4435291782453354,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004798383948102281,
      "loss": 5.1084,
      "mean_token_accuracy": 0.17264219522476196,
      "num_tokens": 31882381.0,
      "step": 14545
    },
    {
      "entropy": 5.567434024810791,
      "epoch": 1.444025406907503,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004798237326702371,
      "loss": 5.1588,
      "mean_token_accuracy": 0.17382670044898987,
      "num_tokens": 31892347.0,
      "step": 14550
    },
    {
      "entropy": 5.614115762710571,
      "epoch": 1.4445216355696706,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047980906545105884,
      "loss": 5.2588,
      "mean_token_accuracy": 0.16123803406953813,
      "num_tokens": 31902751.0,
      "step": 14555
    },
    {
      "entropy": 5.529605722427368,
      "epoch": 1.445017864231838,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004797943931530571,
      "loss": 5.132,
      "mean_token_accuracy": 0.1714739426970482,
      "num_tokens": 31913294.0,
      "step": 14560
    },
    {
      "entropy": 5.526498317718506,
      "epoch": 1.4455140928940056,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004797797157765958,
      "loss": 5.2439,
      "mean_token_accuracy": 0.16959428191184997,
      "num_tokens": 31925678.0,
      "step": 14565
    },
    {
      "entropy": 5.505324792861939,
      "epoch": 1.446010321556173,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004797650333220387,
      "loss": 5.0682,
      "mean_token_accuracy": 0.17543325573205948,
      "num_tokens": 31934578.0,
      "step": 14570
    },
    {
      "entropy": 5.510114860534668,
      "epoch": 1.4465065502183405,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047975034578975006,
      "loss": 5.1093,
      "mean_token_accuracy": 0.17227471321821214,
      "num_tokens": 31945482.0,
      "step": 14575
    },
    {
      "entropy": 5.547610950469971,
      "epoch": 1.4470027788805082,
      "grad_norm": 0.984375,
      "learning_rate": 0.000479735653180094,
      "loss": 5.2171,
      "mean_token_accuracy": 0.17302812039852142,
      "num_tokens": 31957303.0,
      "step": 14580
    },
    {
      "entropy": 5.533083152770996,
      "epoch": 1.4474990075426757,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004797209554934349,
      "loss": 5.1343,
      "mean_token_accuracy": 0.16546977162361146,
      "num_tokens": 31968601.0,
      "step": 14585
    },
    {
      "entropy": 5.525278902053833,
      "epoch": 1.4479952362048432,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004797062527301373,
      "loss": 5.1689,
      "mean_token_accuracy": 0.17269700914621353,
      "num_tokens": 31978987.0,
      "step": 14590
    },
    {
      "entropy": 5.554373645782471,
      "epoch": 1.4484914648670106,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004796915448905657,
      "loss": 5.1443,
      "mean_token_accuracy": 0.16631865054368972,
      "num_tokens": 31990950.0,
      "step": 14595
    },
    {
      "entropy": 5.614176273345947,
      "epoch": 1.4489876935291783,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047967683197508486,
      "loss": 5.164,
      "mean_token_accuracy": 0.1636600226163864,
      "num_tokens": 32001795.0,
      "step": 14600
    },
    {
      "entropy": 5.5594127655029295,
      "epoch": 1.4494839221913458,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00047966211398405966,
      "loss": 5.1741,
      "mean_token_accuracy": 0.16618093103170395,
      "num_tokens": 32011775.0,
      "step": 14605
    },
    {
      "entropy": 5.65315523147583,
      "epoch": 1.4499801508535133,
      "grad_norm": 1.0,
      "learning_rate": 0.000479647390917855,
      "loss": 5.2741,
      "mean_token_accuracy": 0.15630142390727997,
      "num_tokens": 32022674.0,
      "step": 14610
    },
    {
      "entropy": 5.615009021759033,
      "epoch": 1.4504763795156808,
      "grad_norm": 1.0,
      "learning_rate": 0.000479632662776836,
      "loss": 5.21,
      "mean_token_accuracy": 0.17004247307777404,
      "num_tokens": 32033383.0,
      "step": 14615
    },
    {
      "entropy": 5.546019887924194,
      "epoch": 1.4509726081778482,
      "grad_norm": 0.921875,
      "learning_rate": 0.000479617929561368,
      "loss": 5.1144,
      "mean_token_accuracy": 0.17207654863595961,
      "num_tokens": 32044066.0,
      "step": 14620
    },
    {
      "entropy": 5.54558539390564,
      "epoch": 1.451468836840016,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004796031912718162,
      "loss": 5.2007,
      "mean_token_accuracy": 0.1614237979054451,
      "num_tokens": 32054650.0,
      "step": 14625
    },
    {
      "entropy": 5.634576606750488,
      "epoch": 1.4519650655021834,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047958844790854606,
      "loss": 5.1818,
      "mean_token_accuracy": 0.16604610979557038,
      "num_tokens": 32065786.0,
      "step": 14630
    },
    {
      "entropy": 5.60061731338501,
      "epoch": 1.4524612941643509,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047957369947192337,
      "loss": 5.1483,
      "mean_token_accuracy": 0.17453572899103165,
      "num_tokens": 32076326.0,
      "step": 14635
    },
    {
      "entropy": 5.499564456939697,
      "epoch": 1.4529575228265186,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004795589459623136,
      "loss": 5.0875,
      "mean_token_accuracy": 0.16856737136840821,
      "num_tokens": 32086915.0,
      "step": 14640
    },
    {
      "entropy": 5.539043855667114,
      "epoch": 1.453453751488686,
      "grad_norm": 0.90625,
      "learning_rate": 0.00047954418738008295,
      "loss": 5.0666,
      "mean_token_accuracy": 0.1736020877957344,
      "num_tokens": 32096614.0,
      "step": 14645
    },
    {
      "entropy": 5.540634393692017,
      "epoch": 1.4539499801508535,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004795294237255971,
      "loss": 5.158,
      "mean_token_accuracy": 0.174413201212883,
      "num_tokens": 32107721.0,
      "step": 14650
    },
    {
      "entropy": 5.539350366592407,
      "epoch": 1.454446208813021,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047951465499922234,
      "loss": 5.1699,
      "mean_token_accuracy": 0.16659918129444123,
      "num_tokens": 32116621.0,
      "step": 14655
    },
    {
      "entropy": 5.5818687915802006,
      "epoch": 1.4549424374751885,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004794998812013248,
      "loss": 5.1986,
      "mean_token_accuracy": 0.16622278988361358,
      "num_tokens": 32126536.0,
      "step": 14660
    },
    {
      "entropy": 5.593965196609497,
      "epoch": 1.4554386661373562,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00047948510233227084,
      "loss": 5.2495,
      "mean_token_accuracy": 0.15925149172544478,
      "num_tokens": 32137059.0,
      "step": 14665
    },
    {
      "entropy": 5.510193252563477,
      "epoch": 1.4559348947995236,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004794703183924269,
      "loss": 5.1836,
      "mean_token_accuracy": 0.17112441658973693,
      "num_tokens": 32148186.0,
      "step": 14670
    },
    {
      "entropy": 5.621796751022339,
      "epoch": 1.4564311234616911,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004794555293821597,
      "loss": 5.2336,
      "mean_token_accuracy": 0.15495994687080383,
      "num_tokens": 32158244.0,
      "step": 14675
    },
    {
      "entropy": 5.569933891296387,
      "epoch": 1.4569273521238586,
      "grad_norm": 0.953125,
      "learning_rate": 0.000479440735301836,
      "loss": 5.1135,
      "mean_token_accuracy": 0.17377600073814392,
      "num_tokens": 32168404.0,
      "step": 14680
    },
    {
      "entropy": 5.499804401397705,
      "epoch": 1.4574235807860263,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004794259361518226,
      "loss": 5.0805,
      "mean_token_accuracy": 0.17535601258277894,
      "num_tokens": 32178744.0,
      "step": 14685
    },
    {
      "entropy": 5.547453737258911,
      "epoch": 1.4579198094481938,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047941113193248643,
      "loss": 5.2116,
      "mean_token_accuracy": 0.16683421581983565,
      "num_tokens": 32189948.0,
      "step": 14690
    },
    {
      "entropy": 5.579397249221802,
      "epoch": 1.4584160381103612,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004793963226441947,
      "loss": 5.1889,
      "mean_token_accuracy": 0.16290004551410675,
      "num_tokens": 32200979.0,
      "step": 14695
    },
    {
      "entropy": 5.544812917709351,
      "epoch": 1.4589122667725287,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047938150828731455,
      "loss": 5.1645,
      "mean_token_accuracy": 0.16926503032445908,
      "num_tokens": 32211569.0,
      "step": 14700
    },
    {
      "entropy": 5.5758302211761475,
      "epoch": 1.4594084954346962,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004793666888622134,
      "loss": 5.1413,
      "mean_token_accuracy": 0.17045687288045883,
      "num_tokens": 32222429.0,
      "step": 14705
    },
    {
      "entropy": 5.553287363052368,
      "epoch": 1.459904724096864,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004793518643692587,
      "loss": 5.199,
      "mean_token_accuracy": 0.1621451273560524,
      "num_tokens": 32234332.0,
      "step": 14710
    },
    {
      "entropy": 5.617584609985352,
      "epoch": 1.4604009527590314,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004793370348088182,
      "loss": 5.251,
      "mean_token_accuracy": 0.16487257480621337,
      "num_tokens": 32244798.0,
      "step": 14715
    },
    {
      "entropy": 5.568008756637573,
      "epoch": 1.4608971814211988,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004793222001812595,
      "loss": 5.1389,
      "mean_token_accuracy": 0.17323154211044312,
      "num_tokens": 32257746.0,
      "step": 14720
    },
    {
      "entropy": 5.593483734130859,
      "epoch": 1.4613934100833665,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047930736048695045,
      "loss": 5.2366,
      "mean_token_accuracy": 0.16620125472545624,
      "num_tokens": 32270692.0,
      "step": 14725
    },
    {
      "entropy": 5.5526793003082275,
      "epoch": 1.461889638745534,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004792925157262592,
      "loss": 5.1368,
      "mean_token_accuracy": 0.16730573475360871,
      "num_tokens": 32281350.0,
      "step": 14730
    },
    {
      "entropy": 5.5537224292755125,
      "epoch": 1.4623858674077015,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004792776658995537,
      "loss": 5.1645,
      "mean_token_accuracy": 0.16899915635585785,
      "num_tokens": 32293182.0,
      "step": 14735
    },
    {
      "entropy": 5.549123287200928,
      "epoch": 1.462882096069869,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004792628110072023,
      "loss": 5.1452,
      "mean_token_accuracy": 0.16692745238542556,
      "num_tokens": 32304903.0,
      "step": 14740
    },
    {
      "entropy": 5.622548294067383,
      "epoch": 1.4633783247320364,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004792479510495734,
      "loss": 5.3181,
      "mean_token_accuracy": 0.164116707444191,
      "num_tokens": 32316074.0,
      "step": 14745
    },
    {
      "entropy": 5.557573318481445,
      "epoch": 1.463874553394204,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047923308602703535,
      "loss": 5.1317,
      "mean_token_accuracy": 0.17938311100006105,
      "num_tokens": 32327594.0,
      "step": 14750
    },
    {
      "entropy": 5.528957319259644,
      "epoch": 1.4643707820563716,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047921821593995684,
      "loss": 5.1812,
      "mean_token_accuracy": 0.17304366379976271,
      "num_tokens": 32339224.0,
      "step": 14755
    },
    {
      "entropy": 5.556363773345947,
      "epoch": 1.464867010718539,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004792033407887066,
      "loss": 5.1505,
      "mean_token_accuracy": 0.17215009927749633,
      "num_tokens": 32349451.0,
      "step": 14760
    },
    {
      "entropy": 5.677080297470093,
      "epoch": 1.4653632393807066,
      "grad_norm": 0.875,
      "learning_rate": 0.0004791884605736536,
      "loss": 5.227,
      "mean_token_accuracy": 0.16836720257997512,
      "num_tokens": 32361177.0,
      "step": 14765
    },
    {
      "entropy": 5.585278797149658,
      "epoch": 1.4658594680428743,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047917357529516676,
      "loss": 5.1441,
      "mean_token_accuracy": 0.16627820134162902,
      "num_tokens": 32372031.0,
      "step": 14770
    },
    {
      "entropy": 5.512536144256591,
      "epoch": 1.4663556967050417,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004791586849536152,
      "loss": 5.0603,
      "mean_token_accuracy": 0.17988549321889877,
      "num_tokens": 32383936.0,
      "step": 14775
    },
    {
      "entropy": 5.547338581085205,
      "epoch": 1.4668519253672092,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004791437895493682,
      "loss": 5.2051,
      "mean_token_accuracy": 0.17175210565328597,
      "num_tokens": 32394283.0,
      "step": 14780
    },
    {
      "entropy": 5.55866060256958,
      "epoch": 1.4673481540293767,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00047912888908279507,
      "loss": 5.1907,
      "mean_token_accuracy": 0.16381733417510985,
      "num_tokens": 32405549.0,
      "step": 14785
    },
    {
      "entropy": 5.535640096664428,
      "epoch": 1.4678443826915442,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004791139835542653,
      "loss": 5.1286,
      "mean_token_accuracy": 0.16895053386688233,
      "num_tokens": 32414896.0,
      "step": 14790
    },
    {
      "entropy": 5.560206079483033,
      "epoch": 1.4683406113537119,
      "grad_norm": 0.921875,
      "learning_rate": 0.00047909907296414864,
      "loss": 5.1815,
      "mean_token_accuracy": 0.17300113737583162,
      "num_tokens": 32426863.0,
      "step": 14795
    },
    {
      "entropy": 5.518635702133179,
      "epoch": 1.4688368400158793,
      "grad_norm": 0.90625,
      "learning_rate": 0.00047908415731281475,
      "loss": 5.1553,
      "mean_token_accuracy": 0.17030056118965148,
      "num_tokens": 32437703.0,
      "step": 14800
    },
    {
      "entropy": 5.582429456710815,
      "epoch": 1.4693330686780468,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047906923660063344,
      "loss": 5.2081,
      "mean_token_accuracy": 0.1715475559234619,
      "num_tokens": 32448517.0,
      "step": 14805
    },
    {
      "entropy": 5.537463235855102,
      "epoch": 1.4698292973402145,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000479054310827975,
      "loss": 5.1883,
      "mean_token_accuracy": 0.16616832762956618,
      "num_tokens": 32458782.0,
      "step": 14810
    },
    {
      "entropy": 5.5124444484710695,
      "epoch": 1.470325526002382,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047903937999520913,
      "loss": 5.0592,
      "mean_token_accuracy": 0.17402507066726686,
      "num_tokens": 32470113.0,
      "step": 14815
    },
    {
      "entropy": 5.562623453140259,
      "epoch": 1.4708217546645495,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004790244441027064,
      "loss": 5.1421,
      "mean_token_accuracy": 0.17384142577648162,
      "num_tokens": 32482455.0,
      "step": 14820
    },
    {
      "entropy": 5.583147239685059,
      "epoch": 1.471317983326717,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004790095031508371,
      "loss": 5.2012,
      "mean_token_accuracy": 0.17069411426782607,
      "num_tokens": 32493536.0,
      "step": 14825
    },
    {
      "entropy": 5.503097295761108,
      "epoch": 1.4718142119888844,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00047899455713997164,
      "loss": 5.1062,
      "mean_token_accuracy": 0.1720515802502632,
      "num_tokens": 32503991.0,
      "step": 14830
    },
    {
      "entropy": 5.5159343719482425,
      "epoch": 1.4723104406510519,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047897960607048084,
      "loss": 5.187,
      "mean_token_accuracy": 0.1663205772638321,
      "num_tokens": 32515474.0,
      "step": 14835
    },
    {
      "entropy": 5.499777364730835,
      "epoch": 1.4728066693132196,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004789646499427353,
      "loss": 5.0869,
      "mean_token_accuracy": 0.17534635812044144,
      "num_tokens": 32526751.0,
      "step": 14840
    },
    {
      "entropy": 5.633343601226807,
      "epoch": 1.473302897975387,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004789496887571059,
      "loss": 5.2573,
      "mean_token_accuracy": 0.16634935438632964,
      "num_tokens": 32538191.0,
      "step": 14845
    },
    {
      "entropy": 5.56784291267395,
      "epoch": 1.4737991266375545,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047893472251396373,
      "loss": 5.1872,
      "mean_token_accuracy": 0.17286429852247237,
      "num_tokens": 32548841.0,
      "step": 14850
    },
    {
      "entropy": 5.539698266983033,
      "epoch": 1.4742953552997222,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047891975121368,
      "loss": 5.13,
      "mean_token_accuracy": 0.16835729628801346,
      "num_tokens": 32558599.0,
      "step": 14855
    },
    {
      "entropy": 5.50932240486145,
      "epoch": 1.4747915839618897,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047890477485662563,
      "loss": 5.0753,
      "mean_token_accuracy": 0.17769239395856856,
      "num_tokens": 32568641.0,
      "step": 14860
    },
    {
      "entropy": 5.555522441864014,
      "epoch": 1.4752878126240572,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047888979344317236,
      "loss": 5.187,
      "mean_token_accuracy": 0.16348331421613693,
      "num_tokens": 32578924.0,
      "step": 14865
    },
    {
      "entropy": 5.60650634765625,
      "epoch": 1.4757840412862246,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047887480697369155,
      "loss": 5.145,
      "mean_token_accuracy": 0.1765582025051117,
      "num_tokens": 32590233.0,
      "step": 14870
    },
    {
      "entropy": 5.589346075057984,
      "epoch": 1.4762802699483921,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047885981544855475,
      "loss": 5.1782,
      "mean_token_accuracy": 0.1679176941514015,
      "num_tokens": 32600782.0,
      "step": 14875
    },
    {
      "entropy": 5.4977126121521,
      "epoch": 1.4767764986105598,
      "grad_norm": 0.875,
      "learning_rate": 0.0004788448188681339,
      "loss": 5.1495,
      "mean_token_accuracy": 0.17804412692785263,
      "num_tokens": 32612611.0,
      "step": 14880
    },
    {
      "entropy": 5.602090072631836,
      "epoch": 1.4772727272727273,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004788298172328007,
      "loss": 5.206,
      "mean_token_accuracy": 0.1728799596428871,
      "num_tokens": 32623436.0,
      "step": 14885
    },
    {
      "entropy": 5.582080698013305,
      "epoch": 1.4777689559348948,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047881481054292736,
      "loss": 5.0999,
      "mean_token_accuracy": 0.1705150306224823,
      "num_tokens": 32633356.0,
      "step": 14890
    },
    {
      "entropy": 5.5104601860046385,
      "epoch": 1.4782651845970625,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047879979879888584,
      "loss": 5.1367,
      "mean_token_accuracy": 0.1757357746362686,
      "num_tokens": 32643927.0,
      "step": 14895
    },
    {
      "entropy": 5.481229019165039,
      "epoch": 1.47876141325923,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047878478200104845,
      "loss": 5.11,
      "mean_token_accuracy": 0.17548717707395553,
      "num_tokens": 32655086.0,
      "step": 14900
    },
    {
      "entropy": 5.506432580947876,
      "epoch": 1.4792576419213974,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004787697601497876,
      "loss": 5.0577,
      "mean_token_accuracy": 0.17985972613096238,
      "num_tokens": 32665298.0,
      "step": 14905
    },
    {
      "entropy": 5.551363706588745,
      "epoch": 1.479753870583565,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004787547332454758,
      "loss": 5.1131,
      "mean_token_accuracy": 0.16782129257917405,
      "num_tokens": 32676717.0,
      "step": 14910
    },
    {
      "entropy": 5.5900932312011715,
      "epoch": 1.4802500992457324,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047873970128848567,
      "loss": 5.191,
      "mean_token_accuracy": 0.1645972102880478,
      "num_tokens": 32686886.0,
      "step": 14915
    },
    {
      "entropy": 5.505083513259888,
      "epoch": 1.4807463279078998,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004787246642791899,
      "loss": 5.1003,
      "mean_token_accuracy": 0.17709156572818757,
      "num_tokens": 32698631.0,
      "step": 14920
    },
    {
      "entropy": 5.498608922958374,
      "epoch": 1.4812425565700675,
      "grad_norm": 0.859375,
      "learning_rate": 0.00047870962221796147,
      "loss": 5.1289,
      "mean_token_accuracy": 0.16633621156215667,
      "num_tokens": 32710672.0,
      "step": 14925
    },
    {
      "entropy": 5.553397369384766,
      "epoch": 1.481738785232235,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004786945751051734,
      "loss": 5.1746,
      "mean_token_accuracy": 0.16914642006158828,
      "num_tokens": 32720858.0,
      "step": 14930
    },
    {
      "entropy": 5.509949350357056,
      "epoch": 1.4822350138944025,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004786795229411988,
      "loss": 5.1075,
      "mean_token_accuracy": 0.17026506662368773,
      "num_tokens": 32731592.0,
      "step": 14935
    },
    {
      "entropy": 5.61543083190918,
      "epoch": 1.4827312425565702,
      "grad_norm": 0.875,
      "learning_rate": 0.00047866446572641084,
      "loss": 5.1592,
      "mean_token_accuracy": 0.16700685471296312,
      "num_tokens": 32742777.0,
      "step": 14940
    },
    {
      "entropy": 5.5641392230987545,
      "epoch": 1.4832274712187377,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004786494034611831,
      "loss": 5.0983,
      "mean_token_accuracy": 0.1711895078420639,
      "num_tokens": 32753035.0,
      "step": 14945
    },
    {
      "entropy": 5.546930456161499,
      "epoch": 1.4837236998809051,
      "grad_norm": 1.0,
      "learning_rate": 0.000478634336145889,
      "loss": 5.1238,
      "mean_token_accuracy": 0.1683872476220131,
      "num_tokens": 32764318.0,
      "step": 14950
    },
    {
      "entropy": 5.55269865989685,
      "epoch": 1.4842199285430726,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047861926378090197,
      "loss": 5.1529,
      "mean_token_accuracy": 0.1649496704339981,
      "num_tokens": 32775521.0,
      "step": 14955
    },
    {
      "entropy": 5.6163740158081055,
      "epoch": 1.48471615720524,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004786041863665961,
      "loss": 5.1793,
      "mean_token_accuracy": 0.17126491218805312,
      "num_tokens": 32787101.0,
      "step": 14960
    },
    {
      "entropy": 5.5606588363647464,
      "epoch": 1.4852123858674078,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004785891039033451,
      "loss": 5.1716,
      "mean_token_accuracy": 0.16909338980913163,
      "num_tokens": 32798446.0,
      "step": 14965
    },
    {
      "entropy": 5.550089931488037,
      "epoch": 1.4857086145295753,
      "grad_norm": 0.9375,
      "learning_rate": 0.000478574016391523,
      "loss": 5.1777,
      "mean_token_accuracy": 0.1607951730489731,
      "num_tokens": 32809636.0,
      "step": 14970
    },
    {
      "entropy": 5.495729112625122,
      "epoch": 1.4862048431917427,
      "grad_norm": 0.84375,
      "learning_rate": 0.000478558923831504,
      "loss": 5.0742,
      "mean_token_accuracy": 0.1702270820736885,
      "num_tokens": 32820132.0,
      "step": 14975
    },
    {
      "entropy": 5.585244655609131,
      "epoch": 1.4867010718539102,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047854382622366235,
      "loss": 5.1572,
      "mean_token_accuracy": 0.1750050887465477,
      "num_tokens": 32830720.0,
      "step": 14980
    },
    {
      "entropy": 5.525561332702637,
      "epoch": 1.487197300516078,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047852872356837236,
      "loss": 5.11,
      "mean_token_accuracy": 0.17375846356153488,
      "num_tokens": 32841621.0,
      "step": 14985
    },
    {
      "entropy": 5.604021453857422,
      "epoch": 1.4876935291782454,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004785136158660085,
      "loss": 5.2504,
      "mean_token_accuracy": 0.16192498207092285,
      "num_tokens": 32852074.0,
      "step": 14990
    },
    {
      "entropy": 5.646837186813355,
      "epoch": 1.4881897578404129,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047849850311694567,
      "loss": 5.1974,
      "mean_token_accuracy": 0.1767324000597,
      "num_tokens": 32862556.0,
      "step": 14995
    },
    {
      "entropy": 5.563993263244629,
      "epoch": 1.4886859865025803,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00047848338532155833,
      "loss": 5.123,
      "mean_token_accuracy": 0.17461850196123124,
      "num_tokens": 32874407.0,
      "step": 15000
    },
    {
      "epoch": 1.4886859865025803,
      "eval_entropy": 5.35035786092862,
      "eval_loss": 5.255321979522705,
      "eval_mean_token_accuracy": 0.17079043645570158,
      "eval_num_tokens": 32874407.0,
      "eval_runtime": 21.0033,
      "eval_samples_per_second": 1603.034,
      "eval_steps_per_second": 200.397,
      "step": 15000
    },
    {
      "entropy": 5.554626750946045,
      "epoch": 1.4891822151647478,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047846826248022155,
      "loss": 5.1517,
      "mean_token_accuracy": 0.1650475338101387,
      "num_tokens": 32884986.0,
      "step": 15005
    },
    {
      "entropy": 5.5887526512146,
      "epoch": 1.4896784438269155,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00047845313459331034,
      "loss": 5.2083,
      "mean_token_accuracy": 0.16810910999774933,
      "num_tokens": 32896016.0,
      "step": 15010
    },
    {
      "entropy": 5.533790540695191,
      "epoch": 1.490174672489083,
      "grad_norm": 0.90625,
      "learning_rate": 0.00047843800166119974,
      "loss": 5.1051,
      "mean_token_accuracy": 0.175351320207119,
      "num_tokens": 32905648.0,
      "step": 15015
    },
    {
      "entropy": 5.568153953552246,
      "epoch": 1.4906709011512504,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000478422863684265,
      "loss": 5.1732,
      "mean_token_accuracy": 0.17154745608568192,
      "num_tokens": 32917103.0,
      "step": 15020
    },
    {
      "entropy": 5.585504627227783,
      "epoch": 1.4911671298134181,
      "grad_norm": 0.8359375,
      "learning_rate": 0.00047840772066288167,
      "loss": 5.1145,
      "mean_token_accuracy": 0.1708699107170105,
      "num_tokens": 32928741.0,
      "step": 15025
    },
    {
      "entropy": 5.547166776657105,
      "epoch": 1.4916633584755856,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004783925725974252,
      "loss": 5.1752,
      "mean_token_accuracy": 0.16674145460128784,
      "num_tokens": 32939292.0,
      "step": 15030
    },
    {
      "entropy": 5.569071197509766,
      "epoch": 1.492159587137753,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004783774194882712,
      "loss": 5.1652,
      "mean_token_accuracy": 0.16695128083229066,
      "num_tokens": 32949940.0,
      "step": 15035
    },
    {
      "entropy": 5.543909549713135,
      "epoch": 1.4926558157999206,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004783622613357954,
      "loss": 5.1315,
      "mean_token_accuracy": 0.17065854519605636,
      "num_tokens": 32961222.0,
      "step": 15040
    },
    {
      "entropy": 5.492350292205811,
      "epoch": 1.493152044462088,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004783470981403737,
      "loss": 5.1198,
      "mean_token_accuracy": 0.1739916533231735,
      "num_tokens": 32972095.0,
      "step": 15045
    },
    {
      "entropy": 5.572208499908447,
      "epoch": 1.4936482731242555,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004783319299023821,
      "loss": 5.1476,
      "mean_token_accuracy": 0.17626485526561736,
      "num_tokens": 32981834.0,
      "step": 15050
    },
    {
      "entropy": 5.674044942855835,
      "epoch": 1.4941445017864232,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004783167566221969,
      "loss": 5.2388,
      "mean_token_accuracy": 0.16360479593276978,
      "num_tokens": 32994347.0,
      "step": 15055
    },
    {
      "entropy": 5.651814270019531,
      "epoch": 1.4946407304485907,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047830157830019423,
      "loss": 5.2059,
      "mean_token_accuracy": 0.16891241669654847,
      "num_tokens": 33004780.0,
      "step": 15060
    },
    {
      "entropy": 5.572425079345703,
      "epoch": 1.4951369591107582,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004782863949367505,
      "loss": 5.1599,
      "mean_token_accuracy": 0.16840120255947114,
      "num_tokens": 33015697.0,
      "step": 15065
    },
    {
      "entropy": 5.503729009628296,
      "epoch": 1.4956331877729259,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004782712065322421,
      "loss": 5.1415,
      "mean_token_accuracy": 0.16683482825756074,
      "num_tokens": 33026563.0,
      "step": 15070
    },
    {
      "entropy": 5.588099575042724,
      "epoch": 1.4961294164350933,
      "grad_norm": 0.875,
      "learning_rate": 0.0004782560130870459,
      "loss": 5.1747,
      "mean_token_accuracy": 0.16130253672599792,
      "num_tokens": 33037290.0,
      "step": 15075
    },
    {
      "entropy": 5.628545093536377,
      "epoch": 1.4966256450972608,
      "grad_norm": 0.90625,
      "learning_rate": 0.00047824081460153846,
      "loss": 5.1043,
      "mean_token_accuracy": 0.17105162888765335,
      "num_tokens": 33047199.0,
      "step": 15080
    },
    {
      "entropy": 5.500002098083496,
      "epoch": 1.4971218737594283,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004782256110760969,
      "loss": 5.1172,
      "mean_token_accuracy": 0.1685728833079338,
      "num_tokens": 33057253.0,
      "step": 15085
    },
    {
      "entropy": 5.5785651206970215,
      "epoch": 1.4976181024215958,
      "grad_norm": 1.015625,
      "learning_rate": 0.000478210402511098,
      "loss": 5.2578,
      "mean_token_accuracy": 0.1641707092523575,
      "num_tokens": 33068228.0,
      "step": 15090
    },
    {
      "entropy": 5.622333240509033,
      "epoch": 1.4981143310837635,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004781951889069189,
      "loss": 5.2197,
      "mean_token_accuracy": 0.17025210857391357,
      "num_tokens": 33079285.0,
      "step": 15095
    },
    {
      "entropy": 5.536925792694092,
      "epoch": 1.498610559745931,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00047817997026393714,
      "loss": 5.0698,
      "mean_token_accuracy": 0.17296504974365234,
      "num_tokens": 33091253.0,
      "step": 15100
    },
    {
      "entropy": 5.499868535995484,
      "epoch": 1.4991067884080984,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00047816474658252985,
      "loss": 5.1011,
      "mean_token_accuracy": 0.17394308596849442,
      "num_tokens": 33102492.0,
      "step": 15105
    },
    {
      "entropy": 5.490017223358154,
      "epoch": 1.499603017070266,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047814951786307467,
      "loss": 5.1278,
      "mean_token_accuracy": 0.16923998445272445,
      "num_tokens": 33115038.0,
      "step": 15110
    },
    {
      "entropy": 5.558667469024658,
      "epoch": 1.5000992457324336,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004781342841059492,
      "loss": 5.1017,
      "mean_token_accuracy": 0.1734825849533081,
      "num_tokens": 33124470.0,
      "step": 15115
    },
    {
      "entropy": 5.541827583312989,
      "epoch": 1.500595474394601,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004781190453115311,
      "loss": 5.1193,
      "mean_token_accuracy": 0.17209340929985045,
      "num_tokens": 33135994.0,
      "step": 15120
    },
    {
      "entropy": 5.510991096496582,
      "epoch": 1.5010917030567685,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047810380148019847,
      "loss": 5.1294,
      "mean_token_accuracy": 0.17178987711668015,
      "num_tokens": 33145783.0,
      "step": 15125
    },
    {
      "entropy": 5.518322372436524,
      "epoch": 1.501587931718936,
      "grad_norm": 0.9375,
      "learning_rate": 0.00047808855261232915,
      "loss": 5.2334,
      "mean_token_accuracy": 0.17065344452857972,
      "num_tokens": 33156404.0,
      "step": 15130
    },
    {
      "entropy": 5.563249492645264,
      "epoch": 1.5020841603811035,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00047807329870830134,
      "loss": 5.2249,
      "mean_token_accuracy": 0.16813631504774093,
      "num_tokens": 33168226.0,
      "step": 15135
    },
    {
      "entropy": 5.559996652603149,
      "epoch": 1.5025803890432712,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004780580397684933,
      "loss": 5.1422,
      "mean_token_accuracy": 0.16867240965366365,
      "num_tokens": 33178265.0,
      "step": 15140
    },
    {
      "entropy": 5.528187274932861,
      "epoch": 1.5030766177054387,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004780427757932835,
      "loss": 5.1002,
      "mean_token_accuracy": 0.17924540638923644,
      "num_tokens": 33188992.0,
      "step": 15145
    },
    {
      "entropy": 5.542638301849365,
      "epoch": 1.5035728463676064,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004780275067830503,
      "loss": 5.1766,
      "mean_token_accuracy": 0.1707996666431427,
      "num_tokens": 33199424.0,
      "step": 15150
    },
    {
      "entropy": 5.531479692459106,
      "epoch": 1.5040690750297738,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004780122327381725,
      "loss": 5.0979,
      "mean_token_accuracy": 0.18169230818748475,
      "num_tokens": 33210900.0,
      "step": 15155
    },
    {
      "entropy": 5.592750024795532,
      "epoch": 1.5045653036919413,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00047799695365902865,
      "loss": 5.1743,
      "mean_token_accuracy": 0.16534914076328278,
      "num_tokens": 33220796.0,
      "step": 15160
    },
    {
      "entropy": 5.5601893901824955,
      "epoch": 1.5050615323541088,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004779816695459979,
      "loss": 5.1517,
      "mean_token_accuracy": 0.1704527273774147,
      "num_tokens": 33233454.0,
      "step": 15165
    },
    {
      "entropy": 5.566911077499389,
      "epoch": 1.5055577610162763,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004779663803994591,
      "loss": 5.0857,
      "mean_token_accuracy": 0.1749548941850662,
      "num_tokens": 33243574.0,
      "step": 15170
    },
    {
      "entropy": 5.550911569595337,
      "epoch": 1.5060539896784437,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004779510862197914,
      "loss": 5.1286,
      "mean_token_accuracy": 0.16912004500627517,
      "num_tokens": 33256298.0,
      "step": 15175
    },
    {
      "entropy": 5.588207387924195,
      "epoch": 1.5065502183406112,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004779357870073741,
      "loss": 5.2319,
      "mean_token_accuracy": 0.16805994808673858,
      "num_tokens": 33268089.0,
      "step": 15180
    },
    {
      "entropy": 5.528545761108399,
      "epoch": 1.507046447002779,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047792048276258653,
      "loss": 5.1488,
      "mean_token_accuracy": 0.16780902147293092,
      "num_tokens": 33279955.0,
      "step": 15185
    },
    {
      "entropy": 5.52419638633728,
      "epoch": 1.5075426756649464,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004779051734858083,
      "loss": 5.1292,
      "mean_token_accuracy": 0.16726892441511154,
      "num_tokens": 33291470.0,
      "step": 15190
    },
    {
      "entropy": 5.580616283416748,
      "epoch": 1.508038904327114,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047788985917741894,
      "loss": 5.1589,
      "mean_token_accuracy": 0.1763240337371826,
      "num_tokens": 33304169.0,
      "step": 15195
    },
    {
      "entropy": 5.479222393035888,
      "epoch": 1.5085351329892815,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047787453983779826,
      "loss": 5.0122,
      "mean_token_accuracy": 0.17872458398342134,
      "num_tokens": 33314436.0,
      "step": 15200
    },
    {
      "entropy": 5.530719041824341,
      "epoch": 1.509031361651449,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004778592154673262,
      "loss": 5.2258,
      "mean_token_accuracy": 0.16716734170913697,
      "num_tokens": 33325361.0,
      "step": 15205
    },
    {
      "entropy": 5.450767612457275,
      "epoch": 1.5095275903136165,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047784388606638263,
      "loss": 5.1015,
      "mean_token_accuracy": 0.16731740683317184,
      "num_tokens": 33336644.0,
      "step": 15210
    },
    {
      "entropy": 5.542257261276245,
      "epoch": 1.510023818975784,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004778285516353478,
      "loss": 5.153,
      "mean_token_accuracy": 0.16704134345054628,
      "num_tokens": 33348418.0,
      "step": 15215
    },
    {
      "entropy": 5.546857166290283,
      "epoch": 1.5105200476379514,
      "grad_norm": 0.921875,
      "learning_rate": 0.00047781321217460205,
      "loss": 5.1008,
      "mean_token_accuracy": 0.17379361242055893,
      "num_tokens": 33360615.0,
      "step": 15220
    },
    {
      "entropy": 5.5339882373809814,
      "epoch": 1.5110162763001191,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047779786768452554,
      "loss": 5.1044,
      "mean_token_accuracy": 0.1691182181239128,
      "num_tokens": 33372030.0,
      "step": 15225
    },
    {
      "entropy": 5.537009286880493,
      "epoch": 1.5115125049622866,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047778251816549895,
      "loss": 5.1229,
      "mean_token_accuracy": 0.16500473916530609,
      "num_tokens": 33382834.0,
      "step": 15230
    },
    {
      "entropy": 5.533806085586548,
      "epoch": 1.512008733624454,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00047776716361790284,
      "loss": 5.1379,
      "mean_token_accuracy": 0.17281102687120437,
      "num_tokens": 33393250.0,
      "step": 15235
    },
    {
      "entropy": 5.523517274856568,
      "epoch": 1.5125049622866218,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047775180404211804,
      "loss": 5.0835,
      "mean_token_accuracy": 0.17765696346759796,
      "num_tokens": 33402778.0,
      "step": 15240
    },
    {
      "entropy": 5.5844526290893555,
      "epoch": 1.5130011909487893,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004777364394385254,
      "loss": 5.1716,
      "mean_token_accuracy": 0.1688630297780037,
      "num_tokens": 33413565.0,
      "step": 15245
    },
    {
      "entropy": 5.597110939025879,
      "epoch": 1.5134974196109567,
      "grad_norm": 0.875,
      "learning_rate": 0.00047772106980750593,
      "loss": 5.1123,
      "mean_token_accuracy": 0.17104207575321198,
      "num_tokens": 33424108.0,
      "step": 15250
    },
    {
      "entropy": 5.535395956039428,
      "epoch": 1.5139936482731242,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004777056951494407,
      "loss": 5.1934,
      "mean_token_accuracy": 0.16497849225997924,
      "num_tokens": 33435683.0,
      "step": 15255
    },
    {
      "entropy": 5.522085094451905,
      "epoch": 1.5144898769352917,
      "grad_norm": 1.0625,
      "learning_rate": 0.000477690315464711,
      "loss": 5.1591,
      "mean_token_accuracy": 0.17248237878084183,
      "num_tokens": 33446267.0,
      "step": 15260
    },
    {
      "entropy": 5.515142154693604,
      "epoch": 1.5149861055974592,
      "grad_norm": 0.8359375,
      "learning_rate": 0.00047767493075369827,
      "loss": 5.0704,
      "mean_token_accuracy": 0.18734150230884553,
      "num_tokens": 33457901.0,
      "step": 15265
    },
    {
      "entropy": 5.587907075881958,
      "epoch": 1.5154823342596269,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000477659541016784,
      "loss": 5.2391,
      "mean_token_accuracy": 0.17015222162008287,
      "num_tokens": 33469215.0,
      "step": 15270
    },
    {
      "entropy": 5.560878896713257,
      "epoch": 1.5159785629217943,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047764414625434983,
      "loss": 5.1081,
      "mean_token_accuracy": 0.17409939616918563,
      "num_tokens": 33480384.0,
      "step": 15275
    },
    {
      "entropy": 5.552048778533935,
      "epoch": 1.516474791583962,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004776287464667774,
      "loss": 5.1014,
      "mean_token_accuracy": 0.17710302174091339,
      "num_tokens": 33491404.0,
      "step": 15280
    },
    {
      "entropy": 5.626476049423218,
      "epoch": 1.5169710202461295,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00047761334165444876,
      "loss": 5.2384,
      "mean_token_accuracy": 0.1626495823264122,
      "num_tokens": 33501888.0,
      "step": 15285
    },
    {
      "entropy": 5.567591142654419,
      "epoch": 1.517467248908297,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00047759793181774586,
      "loss": 5.1709,
      "mean_token_accuracy": 0.1713092014193535,
      "num_tokens": 33513274.0,
      "step": 15290
    },
    {
      "entropy": 5.552818870544433,
      "epoch": 1.5179634775704645,
      "grad_norm": 1.125,
      "learning_rate": 0.00047758251695705073,
      "loss": 5.1742,
      "mean_token_accuracy": 0.1670441672205925,
      "num_tokens": 33524188.0,
      "step": 15295
    },
    {
      "entropy": 5.563287162780762,
      "epoch": 1.518459706232632,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047756709707274576,
      "loss": 5.213,
      "mean_token_accuracy": 0.16973214894533156,
      "num_tokens": 33535207.0,
      "step": 15300
    },
    {
      "entropy": 5.610132312774658,
      "epoch": 1.5189559348947994,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004775516721652133,
      "loss": 5.164,
      "mean_token_accuracy": 0.16901655495166779,
      "num_tokens": 33546154.0,
      "step": 15305
    },
    {
      "entropy": 5.547743606567383,
      "epoch": 1.519452163556967,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004775362422348358,
      "loss": 5.0978,
      "mean_token_accuracy": 0.170930977165699,
      "num_tokens": 33555365.0,
      "step": 15310
    },
    {
      "entropy": 5.532332944869995,
      "epoch": 1.5199483922191346,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00047752080728199586,
      "loss": 5.1431,
      "mean_token_accuracy": 0.16855551451444625,
      "num_tokens": 33566316.0,
      "step": 15315
    },
    {
      "entropy": 5.538038635253907,
      "epoch": 1.520444620881302,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004775053673070764,
      "loss": 5.0765,
      "mean_token_accuracy": 0.1789124056696892,
      "num_tokens": 33576549.0,
      "step": 15320
    },
    {
      "entropy": 5.521228551864624,
      "epoch": 1.5209408495434698,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004774899223104601,
      "loss": 5.1379,
      "mean_token_accuracy": 0.1723025009036064,
      "num_tokens": 33586681.0,
      "step": 15325
    },
    {
      "entropy": 5.6190918445587155,
      "epoch": 1.5214370782056372,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004774744722925301,
      "loss": 5.2631,
      "mean_token_accuracy": 0.1627289354801178,
      "num_tokens": 33599201.0,
      "step": 15330
    },
    {
      "entropy": 5.591014432907104,
      "epoch": 1.5219333068678047,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004774590172536695,
      "loss": 5.1337,
      "mean_token_accuracy": 0.17084709405899048,
      "num_tokens": 33610217.0,
      "step": 15335
    },
    {
      "entropy": 5.544988012313842,
      "epoch": 1.5224295355299722,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047744355719426146,
      "loss": 5.1609,
      "mean_token_accuracy": 0.17455919235944747,
      "num_tokens": 33620586.0,
      "step": 15340
    },
    {
      "entropy": 5.532115173339844,
      "epoch": 1.5229257641921397,
      "grad_norm": 0.90625,
      "learning_rate": 0.00047742809211468953,
      "loss": 5.1381,
      "mean_token_accuracy": 0.17375372499227523,
      "num_tokens": 33630712.0,
      "step": 15345
    },
    {
      "entropy": 5.517926502227783,
      "epoch": 1.5234219928543071,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000477412622015337,
      "loss": 5.1047,
      "mean_token_accuracy": 0.17157377004623414,
      "num_tokens": 33640367.0,
      "step": 15350
    },
    {
      "entropy": 5.5922643661499025,
      "epoch": 1.5239182215164748,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004773971468965876,
      "loss": 5.1848,
      "mean_token_accuracy": 0.16498054713010787,
      "num_tokens": 33651816.0,
      "step": 15355
    },
    {
      "entropy": 5.550317668914795,
      "epoch": 1.5244144501786423,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047738166675882516,
      "loss": 5.0736,
      "mean_token_accuracy": 0.1746278539299965,
      "num_tokens": 33663661.0,
      "step": 15360
    },
    {
      "entropy": 5.45186939239502,
      "epoch": 1.52491067884081,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004773661816024335,
      "loss": 5.0022,
      "mean_token_accuracy": 0.1824497327208519,
      "num_tokens": 33673670.0,
      "step": 15365
    },
    {
      "entropy": 5.5002635478973385,
      "epoch": 1.5254069075029775,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004773506914277965,
      "loss": 5.219,
      "mean_token_accuracy": 0.1662943631410599,
      "num_tokens": 33685434.0,
      "step": 15370
    },
    {
      "entropy": 5.582822227478028,
      "epoch": 1.525903136165145,
      "grad_norm": 0.90625,
      "learning_rate": 0.00047733519623529853,
      "loss": 5.1726,
      "mean_token_accuracy": 0.17017682194709777,
      "num_tokens": 33696751.0,
      "step": 15375
    },
    {
      "entropy": 5.5826146602630615,
      "epoch": 1.5263993648273124,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004773196960253236,
      "loss": 5.1117,
      "mean_token_accuracy": 0.17130072116851808,
      "num_tokens": 33708141.0,
      "step": 15380
    },
    {
      "entropy": 5.448153877258301,
      "epoch": 1.52689559348948,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004773041907982561,
      "loss": 4.9879,
      "mean_token_accuracy": 0.179925836622715,
      "num_tokens": 33719219.0,
      "step": 15385
    },
    {
      "entropy": 5.546548080444336,
      "epoch": 1.5273918221516474,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00047728868055448064,
      "loss": 5.1375,
      "mean_token_accuracy": 0.1689855009317398,
      "num_tokens": 33730493.0,
      "step": 15390
    },
    {
      "entropy": 5.592306280136109,
      "epoch": 1.5278880508138148,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004772731652943818,
      "loss": 5.1986,
      "mean_token_accuracy": 0.16408584117889405,
      "num_tokens": 33740273.0,
      "step": 15395
    },
    {
      "entropy": 5.5661030292510985,
      "epoch": 1.5283842794759825,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004772576450183444,
      "loss": 5.1628,
      "mean_token_accuracy": 0.17227281183004378,
      "num_tokens": 33752162.0,
      "step": 15400
    },
    {
      "entropy": 5.559649562835693,
      "epoch": 1.52888050813815,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047724211972675315,
      "loss": 5.1089,
      "mean_token_accuracy": 0.17103544473648072,
      "num_tokens": 33763139.0,
      "step": 15405
    },
    {
      "entropy": 5.520846557617188,
      "epoch": 1.5293767368003177,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004772265894199931,
      "loss": 5.1577,
      "mean_token_accuracy": 0.1723170757293701,
      "num_tokens": 33774641.0,
      "step": 15410
    },
    {
      "entropy": 5.549260997772217,
      "epoch": 1.5298729654624852,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004772110540984494,
      "loss": 5.1115,
      "mean_token_accuracy": 0.18366141021251678,
      "num_tokens": 33785644.0,
      "step": 15415
    },
    {
      "entropy": 5.618649005889893,
      "epoch": 1.5303691941246527,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004771955137625073,
      "loss": 5.2728,
      "mean_token_accuracy": 0.15961731672286988,
      "num_tokens": 33797880.0,
      "step": 15420
    },
    {
      "entropy": 5.52158145904541,
      "epoch": 1.5308654227868201,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047717996841255224,
      "loss": 5.0784,
      "mean_token_accuracy": 0.17061331272125244,
      "num_tokens": 33808782.0,
      "step": 15425
    },
    {
      "entropy": 5.501331567764282,
      "epoch": 1.5313616514489876,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047716441804896955,
      "loss": 5.1086,
      "mean_token_accuracy": 0.17796216160058975,
      "num_tokens": 33818483.0,
      "step": 15430
    },
    {
      "entropy": 5.506070995330811,
      "epoch": 1.531857880111155,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004771488626721449,
      "loss": 5.1309,
      "mean_token_accuracy": 0.1669403940439224,
      "num_tokens": 33828590.0,
      "step": 15435
    },
    {
      "entropy": 5.591064739227295,
      "epoch": 1.5323541087733228,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047713330228246406,
      "loss": 5.2146,
      "mean_token_accuracy": 0.16426829248666763,
      "num_tokens": 33838978.0,
      "step": 15440
    },
    {
      "entropy": 5.546977138519287,
      "epoch": 1.5328503374354903,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004771177368803129,
      "loss": 5.1452,
      "mean_token_accuracy": 0.16866506189107894,
      "num_tokens": 33851244.0,
      "step": 15445
    },
    {
      "entropy": 5.5352836608886715,
      "epoch": 1.533346566097658,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047710216646607733,
      "loss": 5.0403,
      "mean_token_accuracy": 0.17950624078512192,
      "num_tokens": 33861666.0,
      "step": 15450
    },
    {
      "entropy": 5.551935148239136,
      "epoch": 1.5338427947598254,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047708659104014363,
      "loss": 5.1294,
      "mean_token_accuracy": 0.17123485654592513,
      "num_tokens": 33872416.0,
      "step": 15455
    },
    {
      "entropy": 5.51541919708252,
      "epoch": 1.534339023421993,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004770710106028978,
      "loss": 5.1169,
      "mean_token_accuracy": 0.17378889471292497,
      "num_tokens": 33881808.0,
      "step": 15460
    },
    {
      "entropy": 5.589226913452149,
      "epoch": 1.5348352520841604,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047705542515472647,
      "loss": 5.1926,
      "mean_token_accuracy": 0.17101919651031494,
      "num_tokens": 33892925.0,
      "step": 15465
    },
    {
      "entropy": 5.559091520309448,
      "epoch": 1.5353314807463279,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004770398346960159,
      "loss": 5.1301,
      "mean_token_accuracy": 0.16781868487596513,
      "num_tokens": 33903702.0,
      "step": 15470
    },
    {
      "entropy": 5.591299343109131,
      "epoch": 1.5358277094084953,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047702423922715284,
      "loss": 5.2128,
      "mean_token_accuracy": 0.17060076594352722,
      "num_tokens": 33917233.0,
      "step": 15475
    },
    {
      "entropy": 5.585413789749145,
      "epoch": 1.5363239380706628,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004770086387485239,
      "loss": 5.1747,
      "mean_token_accuracy": 0.17281627655029297,
      "num_tokens": 33928362.0,
      "step": 15480
    },
    {
      "entropy": 5.62237229347229,
      "epoch": 1.5368201667328305,
      "grad_norm": 0.953125,
      "learning_rate": 0.000476993033260516,
      "loss": 5.1919,
      "mean_token_accuracy": 0.1707867667078972,
      "num_tokens": 33938624.0,
      "step": 15485
    },
    {
      "entropy": 5.508676242828369,
      "epoch": 1.537316395394998,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004769774227635162,
      "loss": 5.095,
      "mean_token_accuracy": 0.1748700723052025,
      "num_tokens": 33950058.0,
      "step": 15490
    },
    {
      "entropy": 5.565647745132447,
      "epoch": 1.5378126240571657,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004769618072579115,
      "loss": 5.1887,
      "mean_token_accuracy": 0.16761232614517213,
      "num_tokens": 33962404.0,
      "step": 15495
    },
    {
      "entropy": 5.590378046035767,
      "epoch": 1.5383088527193332,
      "grad_norm": 0.859375,
      "learning_rate": 0.00047694618674408917,
      "loss": 5.1466,
      "mean_token_accuracy": 0.16801785230636596,
      "num_tokens": 33973095.0,
      "step": 15500
    },
    {
      "entropy": 5.547408533096314,
      "epoch": 1.5388050813815006,
      "grad_norm": 0.890625,
      "learning_rate": 0.00047693056122243655,
      "loss": 5.1556,
      "mean_token_accuracy": 0.16770165860652925,
      "num_tokens": 33984152.0,
      "step": 15505
    },
    {
      "entropy": 5.585107326507568,
      "epoch": 1.539301310043668,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047691493069334124,
      "loss": 5.1689,
      "mean_token_accuracy": 0.16799376159906387,
      "num_tokens": 33994854.0,
      "step": 15510
    },
    {
      "entropy": 5.5745879173278805,
      "epoch": 1.5397975387058356,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00047689929515719066,
      "loss": 5.1832,
      "mean_token_accuracy": 0.17225729525089264,
      "num_tokens": 34006573.0,
      "step": 15515
    },
    {
      "entropy": 5.552916860580444,
      "epoch": 1.540293767368003,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047688365461437263,
      "loss": 5.0989,
      "mean_token_accuracy": 0.16951732635498046,
      "num_tokens": 34016537.0,
      "step": 15520
    },
    {
      "entropy": 5.5335992813110355,
      "epoch": 1.5407899960301707,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047686800906527493,
      "loss": 5.1269,
      "mean_token_accuracy": 0.178988978266716,
      "num_tokens": 34026825.0,
      "step": 15525
    },
    {
      "entropy": 5.601338958740234,
      "epoch": 1.5412862246923382,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004768523585102856,
      "loss": 5.1285,
      "mean_token_accuracy": 0.17599419206380845,
      "num_tokens": 34036786.0,
      "step": 15530
    },
    {
      "entropy": 5.58551812171936,
      "epoch": 1.541782453354506,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004768367029497928,
      "loss": 5.1232,
      "mean_token_accuracy": 0.1749888151884079,
      "num_tokens": 34046784.0,
      "step": 15535
    },
    {
      "entropy": 5.576945066452026,
      "epoch": 1.5422786820166734,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047682104238418464,
      "loss": 5.1726,
      "mean_token_accuracy": 0.17262601256370544,
      "num_tokens": 34058250.0,
      "step": 15540
    },
    {
      "entropy": 5.4955487728118895,
      "epoch": 1.5427749106788409,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047680537681384956,
      "loss": 5.1112,
      "mean_token_accuracy": 0.17404531836509704,
      "num_tokens": 34069969.0,
      "step": 15545
    },
    {
      "entropy": 5.605706977844238,
      "epoch": 1.5432711393410083,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004767897062391759,
      "loss": 5.2115,
      "mean_token_accuracy": 0.1702331081032753,
      "num_tokens": 34080817.0,
      "step": 15550
    },
    {
      "entropy": 5.567223834991455,
      "epoch": 1.5437673680031758,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004767740306605524,
      "loss": 5.1505,
      "mean_token_accuracy": 0.1668059140443802,
      "num_tokens": 34091138.0,
      "step": 15555
    },
    {
      "entropy": 5.545876979827881,
      "epoch": 1.5442635966653433,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004767583500783677,
      "loss": 5.0837,
      "mean_token_accuracy": 0.18054493963718415,
      "num_tokens": 34102035.0,
      "step": 15560
    },
    {
      "entropy": 5.550889921188355,
      "epoch": 1.5447598253275108,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00047674266449301074,
      "loss": 5.1482,
      "mean_token_accuracy": 0.1734504833817482,
      "num_tokens": 34112533.0,
      "step": 15565
    },
    {
      "entropy": 5.541175174713135,
      "epoch": 1.5452560539896785,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00047672697390487034,
      "loss": 5.1444,
      "mean_token_accuracy": 0.17447938472032548,
      "num_tokens": 34124196.0,
      "step": 15570
    },
    {
      "entropy": 5.5504638195037845,
      "epoch": 1.545752282651846,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004767112783143357,
      "loss": 5.1229,
      "mean_token_accuracy": 0.17578904777765275,
      "num_tokens": 34135635.0,
      "step": 15575
    },
    {
      "entropy": 5.503779363632202,
      "epoch": 1.5462485113140136,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000476695577721796,
      "loss": 5.0579,
      "mean_token_accuracy": 0.1799115866422653,
      "num_tokens": 34146202.0,
      "step": 15580
    },
    {
      "entropy": 5.52239408493042,
      "epoch": 1.5467447399761811,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00047667987212764065,
      "loss": 5.1026,
      "mean_token_accuracy": 0.1759197324514389,
      "num_tokens": 34157354.0,
      "step": 15585
    },
    {
      "entropy": 5.554458808898926,
      "epoch": 1.5472409686383486,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00047666416153225894,
      "loss": 5.1742,
      "mean_token_accuracy": 0.16596896797418595,
      "num_tokens": 34168563.0,
      "step": 15590
    },
    {
      "entropy": 5.558117198944092,
      "epoch": 1.547737197300516,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004766484459360406,
      "loss": 5.0993,
      "mean_token_accuracy": 0.17693301886320115,
      "num_tokens": 34178604.0,
      "step": 15595
    },
    {
      "entropy": 5.56316556930542,
      "epoch": 1.5482334259626835,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047663272533937537,
      "loss": 5.1449,
      "mean_token_accuracy": 0.17188220471143723,
      "num_tokens": 34190321.0,
      "step": 15600
    },
    {
      "entropy": 5.5182630062103275,
      "epoch": 1.548729654624851,
      "grad_norm": 0.859375,
      "learning_rate": 0.00047661699974265294,
      "loss": 5.1527,
      "mean_token_accuracy": 0.16989970803260804,
      "num_tokens": 34201519.0,
      "step": 15605
    },
    {
      "entropy": 5.506045722961426,
      "epoch": 1.5492258832870187,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00047660126914626344,
      "loss": 5.0958,
      "mean_token_accuracy": 0.17159055024385453,
      "num_tokens": 34212710.0,
      "step": 15610
    },
    {
      "entropy": 5.592389535903931,
      "epoch": 1.5497221119491862,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004765855335505969,
      "loss": 5.2244,
      "mean_token_accuracy": 0.16767915785312654,
      "num_tokens": 34223187.0,
      "step": 15615
    },
    {
      "entropy": 5.572989273071289,
      "epoch": 1.5502183406113537,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004765697929560434,
      "loss": 5.1657,
      "mean_token_accuracy": 0.1772853121161461,
      "num_tokens": 34234530.0,
      "step": 15620
    },
    {
      "entropy": 5.592358255386353,
      "epoch": 1.5507145692735214,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00047655404736299354,
      "loss": 5.1705,
      "mean_token_accuracy": 0.16713092476129532,
      "num_tokens": 34245003.0,
      "step": 15625
    },
    {
      "entropy": 5.553425407409668,
      "epoch": 1.5512107979356888,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047653829677183745,
      "loss": 5.1407,
      "mean_token_accuracy": 0.17425866723060607,
      "num_tokens": 34255630.0,
      "step": 15630
    },
    {
      "entropy": 5.5319568634033205,
      "epoch": 1.5517070265978563,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00047652254118296596,
      "loss": 5.12,
      "mean_token_accuracy": 0.17600819021463393,
      "num_tokens": 34266716.0,
      "step": 15635
    },
    {
      "entropy": 5.573258018493652,
      "epoch": 1.5522032552600238,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004765067805967697,
      "loss": 5.1533,
      "mean_token_accuracy": 0.16941824555397034,
      "num_tokens": 34278905.0,
      "step": 15640
    },
    {
      "entropy": 5.557124185562134,
      "epoch": 1.5526994839221913,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004764910150136395,
      "loss": 5.1058,
      "mean_token_accuracy": 0.17113475352525712,
      "num_tokens": 34289688.0,
      "step": 15645
    },
    {
      "entropy": 5.540787410736084,
      "epoch": 1.5531957125843587,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047647524443396625,
      "loss": 5.1364,
      "mean_token_accuracy": 0.1701718896627426,
      "num_tokens": 34300964.0,
      "step": 15650
    },
    {
      "entropy": 5.548469209671021,
      "epoch": 1.5536919412465264,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047645946885814105,
      "loss": 5.1263,
      "mean_token_accuracy": 0.16975250989198684,
      "num_tokens": 34311435.0,
      "step": 15655
    },
    {
      "entropy": 5.588261842727661,
      "epoch": 1.554188169908694,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004764436882865552,
      "loss": 5.1426,
      "mean_token_accuracy": 0.16890763938426973,
      "num_tokens": 34321742.0,
      "step": 15660
    },
    {
      "entropy": 5.597425842285157,
      "epoch": 1.5546843985708616,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004764279027195999,
      "loss": 5.2113,
      "mean_token_accuracy": 0.16439826041460037,
      "num_tokens": 34333796.0,
      "step": 15665
    },
    {
      "entropy": 5.674645519256591,
      "epoch": 1.555180627233029,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047641211215766673,
      "loss": 5.2626,
      "mean_token_accuracy": 0.16649988889694214,
      "num_tokens": 34344521.0,
      "step": 15670
    },
    {
      "entropy": 5.524730157852173,
      "epoch": 1.5556768558951966,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004763963166011471,
      "loss": 5.1364,
      "mean_token_accuracy": 0.1763203114271164,
      "num_tokens": 34355629.0,
      "step": 15675
    },
    {
      "entropy": 5.555655670166016,
      "epoch": 1.556173084557364,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004763805160504328,
      "loss": 5.1958,
      "mean_token_accuracy": 0.17040666788816453,
      "num_tokens": 34366935.0,
      "step": 15680
    },
    {
      "entropy": 5.562149286270142,
      "epoch": 1.5566693132195315,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004763647105059157,
      "loss": 5.0966,
      "mean_token_accuracy": 0.16582603752613068,
      "num_tokens": 34378355.0,
      "step": 15685
    },
    {
      "entropy": 5.54614577293396,
      "epoch": 1.557165541881699,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004763488999679877,
      "loss": 5.0498,
      "mean_token_accuracy": 0.17221602499485017,
      "num_tokens": 34388811.0,
      "step": 15690
    },
    {
      "entropy": 5.61201057434082,
      "epoch": 1.5576617705438665,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00047633308443704076,
      "loss": 5.1823,
      "mean_token_accuracy": 0.1666204437613487,
      "num_tokens": 34399905.0,
      "step": 15695
    },
    {
      "entropy": 5.506280183792114,
      "epoch": 1.5581579992060341,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047631726391346717,
      "loss": 5.0428,
      "mean_token_accuracy": 0.17504944801330566,
      "num_tokens": 34410338.0,
      "step": 15700
    },
    {
      "entropy": 5.6072704792022705,
      "epoch": 1.5586542278682016,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004763014383976593,
      "loss": 5.2212,
      "mean_token_accuracy": 0.16766735315322875,
      "num_tokens": 34421204.0,
      "step": 15705
    },
    {
      "entropy": 5.510582160949707,
      "epoch": 1.5591504565303693,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047628560789000945,
      "loss": 5.0326,
      "mean_token_accuracy": 0.18137399554252626,
      "num_tokens": 34431002.0,
      "step": 15710
    },
    {
      "entropy": 5.511729669570923,
      "epoch": 1.5596466851925368,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00047626977239091027,
      "loss": 5.0854,
      "mean_token_accuracy": 0.17405832856893538,
      "num_tokens": 34442446.0,
      "step": 15715
    },
    {
      "entropy": 5.534721565246582,
      "epoch": 1.5601429138547043,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004762539319007545,
      "loss": 5.0568,
      "mean_token_accuracy": 0.17754935026168822,
      "num_tokens": 34453590.0,
      "step": 15720
    },
    {
      "entropy": 5.586336135864258,
      "epoch": 1.5606391425168717,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004762380864199348,
      "loss": 5.1893,
      "mean_token_accuracy": 0.16865919828414916,
      "num_tokens": 34465614.0,
      "step": 15725
    },
    {
      "entropy": 5.620033407211304,
      "epoch": 1.5611353711790392,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004762222359488443,
      "loss": 5.2927,
      "mean_token_accuracy": 0.1614787682890892,
      "num_tokens": 34476616.0,
      "step": 15730
    },
    {
      "entropy": 5.5874059200286865,
      "epoch": 1.5616315998412067,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004762063804878758,
      "loss": 5.0999,
      "mean_token_accuracy": 0.16776054203510285,
      "num_tokens": 34486260.0,
      "step": 15735
    },
    {
      "entropy": 5.547059774398804,
      "epoch": 1.5621278285033744,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00047619052003742277,
      "loss": 5.1529,
      "mean_token_accuracy": 0.17356517910957336,
      "num_tokens": 34496392.0,
      "step": 15740
    },
    {
      "entropy": 5.538840103149414,
      "epoch": 1.5626240571655419,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004761746545978782,
      "loss": 5.1381,
      "mean_token_accuracy": 0.17499603182077408,
      "num_tokens": 34507218.0,
      "step": 15745
    },
    {
      "entropy": 5.609614658355713,
      "epoch": 1.5631202858277096,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004761587841696358,
      "loss": 5.1907,
      "mean_token_accuracy": 0.17009020000696182,
      "num_tokens": 34518888.0,
      "step": 15750
    },
    {
      "entropy": 5.563820886611938,
      "epoch": 1.563616514489877,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000476142908753089,
      "loss": 5.1181,
      "mean_token_accuracy": 0.17389977276325225,
      "num_tokens": 34529865.0,
      "step": 15755
    },
    {
      "entropy": 5.570053434371948,
      "epoch": 1.5641127431520445,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004761270283486314,
      "loss": 5.183,
      "mean_token_accuracy": 0.1677352398633957,
      "num_tokens": 34542541.0,
      "step": 15760
    },
    {
      "entropy": 5.617619466781616,
      "epoch": 1.564608971814212,
      "grad_norm": 0.90625,
      "learning_rate": 0.00047611114295665697,
      "loss": 5.181,
      "mean_token_accuracy": 0.1722867727279663,
      "num_tokens": 34553056.0,
      "step": 15765
    },
    {
      "entropy": 5.655702590942383,
      "epoch": 1.5651052004763795,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047609525257755967,
      "loss": 5.1735,
      "mean_token_accuracy": 0.17132552117109298,
      "num_tokens": 34563344.0,
      "step": 15770
    },
    {
      "entropy": 5.595164585113525,
      "epoch": 1.565601429138547,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004760793572117333,
      "loss": 5.0738,
      "mean_token_accuracy": 0.17740073800086975,
      "num_tokens": 34574320.0,
      "step": 15775
    },
    {
      "entropy": 5.534933137893677,
      "epoch": 1.5660976578007144,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004760634568595721,
      "loss": 5.1453,
      "mean_token_accuracy": 0.17128768861293792,
      "num_tokens": 34584728.0,
      "step": 15780
    },
    {
      "entropy": 5.585668087005615,
      "epoch": 1.5665938864628821,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004760475515214704,
      "loss": 5.2388,
      "mean_token_accuracy": 0.1676257148385048,
      "num_tokens": 34596395.0,
      "step": 15785
    },
    {
      "entropy": 5.543160724639892,
      "epoch": 1.5670901151250496,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004760316411978228,
      "loss": 5.141,
      "mean_token_accuracy": 0.16829395294189453,
      "num_tokens": 34607950.0,
      "step": 15790
    },
    {
      "entropy": 5.562315177917481,
      "epoch": 1.5675863437872173,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047601572588902363,
      "loss": 5.1379,
      "mean_token_accuracy": 0.1778804324567318,
      "num_tokens": 34617640.0,
      "step": 15795
    },
    {
      "entropy": 5.543839597702027,
      "epoch": 1.5680825724493848,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00047599980559546757,
      "loss": 5.1845,
      "mean_token_accuracy": 0.16260297149419783,
      "num_tokens": 34629475.0,
      "step": 15800
    },
    {
      "entropy": 5.531244897842408,
      "epoch": 1.5685788011115522,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004759838803175495,
      "loss": 5.1135,
      "mean_token_accuracy": 0.16955649703741074,
      "num_tokens": 34641423.0,
      "step": 15805
    },
    {
      "entropy": 5.668246412277222,
      "epoch": 1.5690750297737197,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047596795005566423,
      "loss": 5.1744,
      "mean_token_accuracy": 0.17198489904403685,
      "num_tokens": 34651894.0,
      "step": 15810
    },
    {
      "entropy": 5.613889122009278,
      "epoch": 1.5695712584358872,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00047595201481020685,
      "loss": 5.1651,
      "mean_token_accuracy": 0.1670777052640915,
      "num_tokens": 34664177.0,
      "step": 15815
    },
    {
      "entropy": 5.546554803848267,
      "epoch": 1.5700674870980547,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00047593607458157257,
      "loss": 5.1488,
      "mean_token_accuracy": 0.17135327607393264,
      "num_tokens": 34674850.0,
      "step": 15820
    },
    {
      "entropy": 5.552460622787476,
      "epoch": 1.5705637157602224,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047592012937015647,
      "loss": 5.1708,
      "mean_token_accuracy": 0.17258865833282472,
      "num_tokens": 34685719.0,
      "step": 15825
    },
    {
      "entropy": 5.622128868103028,
      "epoch": 1.5710599444223898,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047590417917635423,
      "loss": 5.1587,
      "mean_token_accuracy": 0.16562519967556,
      "num_tokens": 34697261.0,
      "step": 15830
    },
    {
      "entropy": 5.600107097625733,
      "epoch": 1.5715561730845575,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004758882240005613,
      "loss": 5.1666,
      "mean_token_accuracy": 0.17027785032987594,
      "num_tokens": 34708196.0,
      "step": 15835
    },
    {
      "entropy": 5.6132386207580565,
      "epoch": 1.572052401746725,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047587226384317323,
      "loss": 5.1869,
      "mean_token_accuracy": 0.1689977005124092,
      "num_tokens": 34718855.0,
      "step": 15840
    },
    {
      "entropy": 5.559962892532349,
      "epoch": 1.5725486304088925,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00047585629870458584,
      "loss": 5.1121,
      "mean_token_accuracy": 0.17602272629737853,
      "num_tokens": 34731137.0,
      "step": 15845
    },
    {
      "entropy": 5.575714015960694,
      "epoch": 1.57304485907106,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004758403285851951,
      "loss": 5.0746,
      "mean_token_accuracy": 0.17832073271274568,
      "num_tokens": 34741628.0,
      "step": 15850
    },
    {
      "entropy": 5.512636184692383,
      "epoch": 1.5735410877332274,
      "grad_norm": 0.796875,
      "learning_rate": 0.00047582435348539695,
      "loss": 5.1501,
      "mean_token_accuracy": 0.17139003723859786,
      "num_tokens": 34752984.0,
      "step": 15855
    },
    {
      "entropy": 5.568330001831055,
      "epoch": 1.574037316395395,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004758083734055875,
      "loss": 5.2216,
      "mean_token_accuracy": 0.16662386804819107,
      "num_tokens": 34764363.0,
      "step": 15860
    },
    {
      "entropy": 5.583333253860474,
      "epoch": 1.5745335450575624,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004757923883461632,
      "loss": 5.1664,
      "mean_token_accuracy": 0.1679494321346283,
      "num_tokens": 34776289.0,
      "step": 15865
    },
    {
      "entropy": 5.623749017715454,
      "epoch": 1.57502977371973,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047577639830752034,
      "loss": 5.1386,
      "mean_token_accuracy": 0.17214005440473557,
      "num_tokens": 34786760.0,
      "step": 15870
    },
    {
      "entropy": 5.541008472442627,
      "epoch": 1.5755260023818976,
      "grad_norm": 1.0,
      "learning_rate": 0.00047576040329005536,
      "loss": 5.1049,
      "mean_token_accuracy": 0.17186620384454726,
      "num_tokens": 34797737.0,
      "step": 15875
    },
    {
      "entropy": 5.526557493209839,
      "epoch": 1.5760222310440652,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000475744403294165,
      "loss": 5.1657,
      "mean_token_accuracy": 0.17382304221391678,
      "num_tokens": 34809650.0,
      "step": 15880
    },
    {
      "entropy": 5.6612968921661375,
      "epoch": 1.5765184597062327,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047572839832024606,
      "loss": 5.1888,
      "mean_token_accuracy": 0.1654067099094391,
      "num_tokens": 34820171.0,
      "step": 15885
    },
    {
      "entropy": 5.527270126342773,
      "epoch": 1.5770146883684002,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00047571238836869537,
      "loss": 5.0427,
      "mean_token_accuracy": 0.17614986151456832,
      "num_tokens": 34831286.0,
      "step": 15890
    },
    {
      "entropy": 5.4632336616516115,
      "epoch": 1.5775109170305677,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004756963734399099,
      "loss": 5.0544,
      "mean_token_accuracy": 0.17737851440906524,
      "num_tokens": 34841767.0,
      "step": 15895
    },
    {
      "entropy": 5.536492156982422,
      "epoch": 1.5780071456927351,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047568035353428683,
      "loss": 5.1629,
      "mean_token_accuracy": 0.172687329351902,
      "num_tokens": 34852644.0,
      "step": 15900
    },
    {
      "entropy": 5.606753301620484,
      "epoch": 1.5785033743549026,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047566432865222343,
      "loss": 5.1718,
      "mean_token_accuracy": 0.17330268919467925,
      "num_tokens": 34863605.0,
      "step": 15905
    },
    {
      "entropy": 5.525470924377442,
      "epoch": 1.5789996030170703,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00047564829879411706,
      "loss": 5.1198,
      "mean_token_accuracy": 0.17635508477687836,
      "num_tokens": 34874594.0,
      "step": 15910
    },
    {
      "entropy": 5.561229991912842,
      "epoch": 1.5794958316792378,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004756322639603653,
      "loss": 5.2146,
      "mean_token_accuracy": 0.167847740650177,
      "num_tokens": 34884627.0,
      "step": 15915
    },
    {
      "entropy": 5.6274457454681395,
      "epoch": 1.5799920603414053,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00047561622415136565,
      "loss": 5.1755,
      "mean_token_accuracy": 0.1773965761065483,
      "num_tokens": 34895679.0,
      "step": 15920
    },
    {
      "entropy": 5.525401973724366,
      "epoch": 1.580488289003573,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047560017936751597,
      "loss": 5.105,
      "mean_token_accuracy": 0.17527058273553847,
      "num_tokens": 34906811.0,
      "step": 15925
    },
    {
      "entropy": 5.5443473815917965,
      "epoch": 1.5809845176657404,
      "grad_norm": 1.0,
      "learning_rate": 0.0004755841296092141,
      "loss": 5.1479,
      "mean_token_accuracy": 0.1667722150683403,
      "num_tokens": 34918655.0,
      "step": 15930
    },
    {
      "entropy": 5.533793354034424,
      "epoch": 1.581480746327908,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047556807487685797,
      "loss": 5.134,
      "mean_token_accuracy": 0.17009201943874358,
      "num_tokens": 34931034.0,
      "step": 15935
    },
    {
      "entropy": 5.596280765533447,
      "epoch": 1.5819769749900754,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004755520151708458,
      "loss": 5.0867,
      "mean_token_accuracy": 0.17267320454120635,
      "num_tokens": 34939707.0,
      "step": 15940
    },
    {
      "entropy": 5.55272331237793,
      "epoch": 1.5824732036522429,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004755359504915758,
      "loss": 5.1928,
      "mean_token_accuracy": 0.1639191046357155,
      "num_tokens": 34949237.0,
      "step": 15945
    },
    {
      "entropy": 5.605570697784424,
      "epoch": 1.5829694323144103,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004755198808394464,
      "loss": 5.2142,
      "mean_token_accuracy": 0.17125055491924285,
      "num_tokens": 34960380.0,
      "step": 15950
    },
    {
      "entropy": 5.591394853591919,
      "epoch": 1.583465660976578,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000475503806214856,
      "loss": 5.1725,
      "mean_token_accuracy": 0.16544706821441652,
      "num_tokens": 34971699.0,
      "step": 15955
    },
    {
      "entropy": 5.586516952514648,
      "epoch": 1.5839618896387455,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004754877266182033,
      "loss": 5.1375,
      "mean_token_accuracy": 0.17231732606887817,
      "num_tokens": 34982333.0,
      "step": 15960
    },
    {
      "entropy": 5.564710235595703,
      "epoch": 1.5844581183009132,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000475471642049887,
      "loss": 5.172,
      "mean_token_accuracy": 0.17079628705978395,
      "num_tokens": 34993277.0,
      "step": 15965
    },
    {
      "entropy": 5.522321701049805,
      "epoch": 1.5849543469630807,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004754555525103058,
      "loss": 5.1107,
      "mean_token_accuracy": 0.17557063400745393,
      "num_tokens": 35004252.0,
      "step": 15970
    },
    {
      "entropy": 5.49769139289856,
      "epoch": 1.5854505756252482,
      "grad_norm": 0.875,
      "learning_rate": 0.000475439457999859,
      "loss": 5.0911,
      "mean_token_accuracy": 0.18037165701389313,
      "num_tokens": 35016974.0,
      "step": 15975
    },
    {
      "entropy": 5.575330209732056,
      "epoch": 1.5859468042874156,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00047542335851894553,
      "loss": 5.1845,
      "mean_token_accuracy": 0.16851292848587035,
      "num_tokens": 35029017.0,
      "step": 15980
    },
    {
      "entropy": 5.620971441268921,
      "epoch": 1.586443032949583,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047540725406796454,
      "loss": 5.2186,
      "mean_token_accuracy": 0.16834655702114104,
      "num_tokens": 35040239.0,
      "step": 15985
    },
    {
      "entropy": 5.602098846435547,
      "epoch": 1.5869392616117506,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004753911446473156,
      "loss": 5.1213,
      "mean_token_accuracy": 0.16988668143749236,
      "num_tokens": 35051426.0,
      "step": 15990
    },
    {
      "entropy": 5.4708832740783695,
      "epoch": 1.587435490273918,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000475375030257398,
      "loss": 5.1022,
      "mean_token_accuracy": 0.16767460405826567,
      "num_tokens": 35061777.0,
      "step": 15995
    },
    {
      "entropy": 5.52356915473938,
      "epoch": 1.5879317189360858,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004753589108986114,
      "loss": 5.0498,
      "mean_token_accuracy": 0.18492615222930908,
      "num_tokens": 35071822.0,
      "step": 16000
    },
    {
      "entropy": 5.558615684509277,
      "epoch": 1.5884279475982532,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047534278657135565,
      "loss": 5.1437,
      "mean_token_accuracy": 0.17297255098819733,
      "num_tokens": 35081900.0,
      "step": 16005
    },
    {
      "entropy": 5.59310302734375,
      "epoch": 1.588924176260421,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004753266572760304,
      "loss": 5.1964,
      "mean_token_accuracy": 0.1688542127609253,
      "num_tokens": 35092866.0,
      "step": 16010
    },
    {
      "entropy": 5.734499120712281,
      "epoch": 1.5894204049225884,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047531052301303577,
      "loss": 5.281,
      "mean_token_accuracy": 0.16906654834747314,
      "num_tokens": 35103425.0,
      "step": 16015
    },
    {
      "entropy": 5.508281993865967,
      "epoch": 1.5899166335847559,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004752943837827718,
      "loss": 5.0661,
      "mean_token_accuracy": 0.17544660419225694,
      "num_tokens": 35114620.0,
      "step": 16020
    },
    {
      "entropy": 5.600801849365235,
      "epoch": 1.5904128622469234,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004752782395856387,
      "loss": 5.1724,
      "mean_token_accuracy": 0.16886257827281953,
      "num_tokens": 35125199.0,
      "step": 16025
    },
    {
      "entropy": 5.528537178039551,
      "epoch": 1.5909090909090908,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004752620904220368,
      "loss": 5.1347,
      "mean_token_accuracy": 0.17435365915298462,
      "num_tokens": 35135892.0,
      "step": 16030
    },
    {
      "entropy": 5.567734861373902,
      "epoch": 1.5914053195712583,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004752459362923666,
      "loss": 5.1463,
      "mean_token_accuracy": 0.1693684697151184,
      "num_tokens": 35146186.0,
      "step": 16035
    },
    {
      "entropy": 5.577980899810791,
      "epoch": 1.591901548233426,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004752297771970287,
      "loss": 5.1956,
      "mean_token_accuracy": 0.17215658277273177,
      "num_tokens": 35157261.0,
      "step": 16040
    },
    {
      "entropy": 5.606237268447876,
      "epoch": 1.5923977768955935,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00047521361313642373,
      "loss": 5.1446,
      "mean_token_accuracy": 0.17055224627256393,
      "num_tokens": 35168508.0,
      "step": 16045
    },
    {
      "entropy": 5.533114576339722,
      "epoch": 1.5928940055577612,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004751974441109526,
      "loss": 5.1103,
      "mean_token_accuracy": 0.16907101571559907,
      "num_tokens": 35179444.0,
      "step": 16050
    },
    {
      "entropy": 5.520071268081665,
      "epoch": 1.5933902342199286,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047518127012101624,
      "loss": 5.1108,
      "mean_token_accuracy": 0.1762010484933853,
      "num_tokens": 35189517.0,
      "step": 16055
    },
    {
      "entropy": 5.5501398086547855,
      "epoch": 1.5938864628820961,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004751650911670157,
      "loss": 5.1529,
      "mean_token_accuracy": 0.16915027648210526,
      "num_tokens": 35200191.0,
      "step": 16060
    },
    {
      "entropy": 5.604111051559448,
      "epoch": 1.5943826915442636,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00047514890724935236,
      "loss": 5.1661,
      "mean_token_accuracy": 0.16918279230594635,
      "num_tokens": 35211260.0,
      "step": 16065
    },
    {
      "entropy": 5.500099086761475,
      "epoch": 1.594878920206431,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004751327183684272,
      "loss": 5.1447,
      "mean_token_accuracy": 0.1680993989109993,
      "num_tokens": 35222408.0,
      "step": 16070
    },
    {
      "entropy": 5.605546140670777,
      "epoch": 1.5953751488685985,
      "grad_norm": 0.890625,
      "learning_rate": 0.00047511652452464196,
      "loss": 5.1854,
      "mean_token_accuracy": 0.16996719986200332,
      "num_tokens": 35232760.0,
      "step": 16075
    },
    {
      "entropy": 5.588399028778076,
      "epoch": 1.595871377530766,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047510032571839816,
      "loss": 5.1506,
      "mean_token_accuracy": 0.16914570480585098,
      "num_tokens": 35243849.0,
      "step": 16080
    },
    {
      "entropy": 5.588242149353027,
      "epoch": 1.5963676061929337,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004750841219500975,
      "loss": 5.1648,
      "mean_token_accuracy": 0.16913552582263947,
      "num_tokens": 35254962.0,
      "step": 16085
    },
    {
      "entropy": 5.583474016189575,
      "epoch": 1.5968638348551012,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004750679132201416,
      "loss": 5.2403,
      "mean_token_accuracy": 0.1665369153022766,
      "num_tokens": 35265895.0,
      "step": 16090
    },
    {
      "entropy": 5.69076132774353,
      "epoch": 1.597360063517269,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047505169952893276,
      "loss": 5.2322,
      "mean_token_accuracy": 0.17190774381160737,
      "num_tokens": 35277543.0,
      "step": 16095
    },
    {
      "entropy": 5.587152338027954,
      "epoch": 1.5978562921794364,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047503548087687267,
      "loss": 5.1317,
      "mean_token_accuracy": 0.17146954238414763,
      "num_tokens": 35289130.0,
      "step": 16100
    },
    {
      "entropy": 5.5185308933258055,
      "epoch": 1.5983525208416038,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047501925726436376,
      "loss": 5.0779,
      "mean_token_accuracy": 0.17090461403131485,
      "num_tokens": 35298414.0,
      "step": 16105
    },
    {
      "entropy": 5.543598985671997,
      "epoch": 1.5988487495037713,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004750030286918082,
      "loss": 5.1416,
      "mean_token_accuracy": 0.1716749131679535,
      "num_tokens": 35310629.0,
      "step": 16110
    },
    {
      "entropy": 5.6182928562164305,
      "epoch": 1.5993449781659388,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004749867951596086,
      "loss": 5.1818,
      "mean_token_accuracy": 0.16935135573148727,
      "num_tokens": 35321810.0,
      "step": 16115
    },
    {
      "entropy": 5.5639870166778564,
      "epoch": 1.5998412068281063,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047497055666816736,
      "loss": 5.0998,
      "mean_token_accuracy": 0.1760576844215393,
      "num_tokens": 35333517.0,
      "step": 16120
    },
    {
      "entropy": 5.496850061416626,
      "epoch": 1.600337435490274,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004749543132178873,
      "loss": 5.0415,
      "mean_token_accuracy": 0.17744467705488204,
      "num_tokens": 35344170.0,
      "step": 16125
    },
    {
      "entropy": 5.571909856796265,
      "epoch": 1.6008336641524414,
      "grad_norm": 0.96875,
      "learning_rate": 0.000474938064809171,
      "loss": 5.2249,
      "mean_token_accuracy": 0.16783830374479294,
      "num_tokens": 35354963.0,
      "step": 16130
    },
    {
      "entropy": 5.586218929290771,
      "epoch": 1.6013298928146091,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004749218114424216,
      "loss": 5.1861,
      "mean_token_accuracy": 0.17537025958299637,
      "num_tokens": 35365716.0,
      "step": 16135
    },
    {
      "entropy": 5.546600103378296,
      "epoch": 1.6018261214767766,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047490555311804206,
      "loss": 5.112,
      "mean_token_accuracy": 0.17386993765830994,
      "num_tokens": 35376113.0,
      "step": 16140
    },
    {
      "entropy": 5.51456880569458,
      "epoch": 1.602322350138944,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047488928983643565,
      "loss": 5.0889,
      "mean_token_accuracy": 0.1700906589627266,
      "num_tokens": 35386178.0,
      "step": 16145
    },
    {
      "entropy": 5.521318960189819,
      "epoch": 1.6028185788011116,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047487302159800545,
      "loss": 5.0097,
      "mean_token_accuracy": 0.17954554110765458,
      "num_tokens": 35396874.0,
      "step": 16150
    },
    {
      "entropy": 5.54805850982666,
      "epoch": 1.603314807463279,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000474856748403155,
      "loss": 5.2282,
      "mean_token_accuracy": 0.1678831994533539,
      "num_tokens": 35409405.0,
      "step": 16155
    },
    {
      "entropy": 5.616310119628906,
      "epoch": 1.6038110361254465,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004748404702522879,
      "loss": 5.1762,
      "mean_token_accuracy": 0.1686457484960556,
      "num_tokens": 35421449.0,
      "step": 16160
    },
    {
      "entropy": 5.597290849685669,
      "epoch": 1.604307264787614,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00047482418714580774,
      "loss": 5.2114,
      "mean_token_accuracy": 0.17077624499797822,
      "num_tokens": 35434612.0,
      "step": 16165
    },
    {
      "entropy": 5.533904409408569,
      "epoch": 1.6048034934497817,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004748078990841183,
      "loss": 5.0961,
      "mean_token_accuracy": 0.17794411033391952,
      "num_tokens": 35446016.0,
      "step": 16170
    },
    {
      "entropy": 5.617804098129272,
      "epoch": 1.6052997221119492,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004747916060676235,
      "loss": 5.1498,
      "mean_token_accuracy": 0.17460282891988754,
      "num_tokens": 35457405.0,
      "step": 16175
    },
    {
      "entropy": 5.593232297897339,
      "epoch": 1.6057959507741169,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00047477530809672747,
      "loss": 5.2112,
      "mean_token_accuracy": 0.16786461025476457,
      "num_tokens": 35468361.0,
      "step": 16180
    },
    {
      "entropy": 5.540446519851685,
      "epoch": 1.6062921794362843,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004747590051718342,
      "loss": 5.0968,
      "mean_token_accuracy": 0.17383848875761032,
      "num_tokens": 35480092.0,
      "step": 16185
    },
    {
      "entropy": 5.596573686599731,
      "epoch": 1.6067884080984518,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047474269729334806,
      "loss": 5.2194,
      "mean_token_accuracy": 0.1624014213681221,
      "num_tokens": 35492131.0,
      "step": 16190
    },
    {
      "entropy": 5.619562721252441,
      "epoch": 1.6072846367606193,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00047472638446167346,
      "loss": 5.169,
      "mean_token_accuracy": 0.1646542578935623,
      "num_tokens": 35503600.0,
      "step": 16195
    },
    {
      "entropy": 5.629235458374024,
      "epoch": 1.6077808654227868,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047471006667721484,
      "loss": 5.2073,
      "mean_token_accuracy": 0.16861433386802674,
      "num_tokens": 35514716.0,
      "step": 16200
    },
    {
      "entropy": 5.584695434570312,
      "epoch": 1.6082770940849542,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000474693743940377,
      "loss": 5.1366,
      "mean_token_accuracy": 0.17781931310892105,
      "num_tokens": 35525688.0,
      "step": 16205
    },
    {
      "entropy": 5.554493474960327,
      "epoch": 1.608773322747122,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00047467741625156453,
      "loss": 5.1422,
      "mean_token_accuracy": 0.17368559837341307,
      "num_tokens": 35536391.0,
      "step": 16210
    },
    {
      "entropy": 5.510008811950684,
      "epoch": 1.6092695514092894,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00047466108361118235,
      "loss": 5.0776,
      "mean_token_accuracy": 0.17908220440149308,
      "num_tokens": 35547016.0,
      "step": 16215
    },
    {
      "entropy": 5.568167591094971,
      "epoch": 1.6097657800714569,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047464474601963564,
      "loss": 5.1603,
      "mean_token_accuracy": 0.17011854499578477,
      "num_tokens": 35558402.0,
      "step": 16220
    },
    {
      "entropy": 5.585215377807617,
      "epoch": 1.6102620087336246,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004746284034773294,
      "loss": 5.1485,
      "mean_token_accuracy": 0.1724248558282852,
      "num_tokens": 35569779.0,
      "step": 16225
    },
    {
      "entropy": 5.5655535697937015,
      "epoch": 1.610758237395792,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004746120559846689,
      "loss": 5.1547,
      "mean_token_accuracy": 0.17170076668262482,
      "num_tokens": 35580044.0,
      "step": 16230
    },
    {
      "entropy": 5.600090169906617,
      "epoch": 1.6112544660579595,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00047459570354205954,
      "loss": 5.2357,
      "mean_token_accuracy": 0.17075808644294738,
      "num_tokens": 35591216.0,
      "step": 16235
    },
    {
      "entropy": 5.508529853820801,
      "epoch": 1.611750694720127,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047457934614990684,
      "loss": 5.0616,
      "mean_token_accuracy": 0.17528978139162063,
      "num_tokens": 35602376.0,
      "step": 16240
    },
    {
      "entropy": 5.5367542743682865,
      "epoch": 1.6122469233822945,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004745629838086164,
      "loss": 5.0379,
      "mean_token_accuracy": 0.17735781371593476,
      "num_tokens": 35612184.0,
      "step": 16245
    },
    {
      "entropy": 5.579059839248657,
      "epoch": 1.612743152044462,
      "grad_norm": 0.921875,
      "learning_rate": 0.00047454661651859394,
      "loss": 5.2048,
      "mean_token_accuracy": 0.17242882996797562,
      "num_tokens": 35624864.0,
      "step": 16250
    },
    {
      "entropy": 5.573807668685913,
      "epoch": 1.6132393807066296,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004745302442802454,
      "loss": 5.1435,
      "mean_token_accuracy": 0.1738520309329033,
      "num_tokens": 35636492.0,
      "step": 16255
    },
    {
      "entropy": 5.512031173706054,
      "epoch": 1.6137356093687971,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004745138670939768,
      "loss": 5.0852,
      "mean_token_accuracy": 0.17595444470643998,
      "num_tokens": 35646831.0,
      "step": 16260
    },
    {
      "entropy": 5.599237966537475,
      "epoch": 1.6142318380309648,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004744974849601941,
      "loss": 5.1966,
      "mean_token_accuracy": 0.17177168279886246,
      "num_tokens": 35657703.0,
      "step": 16265
    },
    {
      "entropy": 5.4731049060821535,
      "epoch": 1.6147280666931323,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004744810978793038,
      "loss": 5.023,
      "mean_token_accuracy": 0.17758560180664062,
      "num_tokens": 35668334.0,
      "step": 16270
    },
    {
      "entropy": 5.526353549957276,
      "epoch": 1.6152242953552998,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047446470585171206,
      "loss": 5.1726,
      "mean_token_accuracy": 0.16676280796527862,
      "num_tokens": 35679796.0,
      "step": 16275
    },
    {
      "entropy": 5.531010437011719,
      "epoch": 1.6157205240174672,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047444830887782537,
      "loss": 5.0604,
      "mean_token_accuracy": 0.17524588108062744,
      "num_tokens": 35690825.0,
      "step": 16280
    },
    {
      "entropy": 5.62803020477295,
      "epoch": 1.6162167526796347,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004744319069580504,
      "loss": 5.1849,
      "mean_token_accuracy": 0.17152267098426818,
      "num_tokens": 35701824.0,
      "step": 16285
    },
    {
      "entropy": 5.612047958374023,
      "epoch": 1.6167129813418022,
      "grad_norm": 0.859375,
      "learning_rate": 0.00047441550009279395,
      "loss": 5.1012,
      "mean_token_accuracy": 0.17642791122198104,
      "num_tokens": 35713250.0,
      "step": 16290
    },
    {
      "entropy": 5.565155410766602,
      "epoch": 1.6172092100039697,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004743990882824627,
      "loss": 5.1801,
      "mean_token_accuracy": 0.16946621239185333,
      "num_tokens": 35725049.0,
      "step": 16295
    },
    {
      "entropy": 5.494463205337524,
      "epoch": 1.6177054386661374,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047438267152746376,
      "loss": 5.1189,
      "mean_token_accuracy": 0.16613352298736572,
      "num_tokens": 35734676.0,
      "step": 16300
    },
    {
      "entropy": 5.568187665939331,
      "epoch": 1.6182016673283048,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047436624982820423,
      "loss": 5.1146,
      "mean_token_accuracy": 0.17054773420095443,
      "num_tokens": 35743953.0,
      "step": 16305
    },
    {
      "entropy": 5.530277061462402,
      "epoch": 1.6186978959904725,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004743498231850913,
      "loss": 5.0523,
      "mean_token_accuracy": 0.18202498704195022,
      "num_tokens": 35753149.0,
      "step": 16310
    },
    {
      "entropy": 5.491792345046997,
      "epoch": 1.61919412465264,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004743333915985323,
      "loss": 5.127,
      "mean_token_accuracy": 0.16944457590579987,
      "num_tokens": 35763783.0,
      "step": 16315
    },
    {
      "entropy": 5.484177207946777,
      "epoch": 1.6196903533148075,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047431695506893476,
      "loss": 5.0704,
      "mean_token_accuracy": 0.18196655958890914,
      "num_tokens": 35774663.0,
      "step": 16320
    },
    {
      "entropy": 5.574009418487549,
      "epoch": 1.620186581976975,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047430051359670616,
      "loss": 5.2252,
      "mean_token_accuracy": 0.16487312018871308,
      "num_tokens": 35786990.0,
      "step": 16325
    },
    {
      "entropy": 5.659722185134887,
      "epoch": 1.6206828106391424,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004742840671822542,
      "loss": 5.2263,
      "mean_token_accuracy": 0.1707298368215561,
      "num_tokens": 35798818.0,
      "step": 16330
    },
    {
      "entropy": 5.613869762420654,
      "epoch": 1.62117903930131,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000474267615825987,
      "loss": 5.1666,
      "mean_token_accuracy": 0.1698811709880829,
      "num_tokens": 35808813.0,
      "step": 16335
    },
    {
      "entropy": 5.531151485443115,
      "epoch": 1.6216752679634776,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047425115952831213,
      "loss": 5.146,
      "mean_token_accuracy": 0.1767699286341667,
      "num_tokens": 35818664.0,
      "step": 16340
    },
    {
      "entropy": 5.497013330459595,
      "epoch": 1.622171496625645,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000474234698289638,
      "loss": 5.0933,
      "mean_token_accuracy": 0.17992809414863586,
      "num_tokens": 35828975.0,
      "step": 16345
    },
    {
      "entropy": 5.494010591506958,
      "epoch": 1.6226677252878128,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004742182321103725,
      "loss": 5.1152,
      "mean_token_accuracy": 0.17443787306547165,
      "num_tokens": 35839437.0,
      "step": 16350
    },
    {
      "entropy": 5.543405532836914,
      "epoch": 1.6231639539499803,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004742017609909242,
      "loss": 5.0733,
      "mean_token_accuracy": 0.17784169763326646,
      "num_tokens": 35850413.0,
      "step": 16355
    },
    {
      "entropy": 5.596259117126465,
      "epoch": 1.6236601826121477,
      "grad_norm": 0.921875,
      "learning_rate": 0.00047418528493170157,
      "loss": 5.1427,
      "mean_token_accuracy": 0.1686675876379013,
      "num_tokens": 35862745.0,
      "step": 16360
    },
    {
      "entropy": 5.6065216064453125,
      "epoch": 1.6241564112743152,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004741688039331129,
      "loss": 5.13,
      "mean_token_accuracy": 0.1675193652510643,
      "num_tokens": 35873934.0,
      "step": 16365
    },
    {
      "entropy": 5.505111026763916,
      "epoch": 1.6246526399364827,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047415231799556725,
      "loss": 5.0541,
      "mean_token_accuracy": 0.1791858971118927,
      "num_tokens": 35884279.0,
      "step": 16370
    },
    {
      "entropy": 5.488245153427124,
      "epoch": 1.6251488685986502,
      "grad_norm": 0.9375,
      "learning_rate": 0.00047413582711947305,
      "loss": 5.1123,
      "mean_token_accuracy": 0.1776946723461151,
      "num_tokens": 35894144.0,
      "step": 16375
    },
    {
      "entropy": 5.5181357860565186,
      "epoch": 1.6256450972608176,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047411933130523954,
      "loss": 5.1133,
      "mean_token_accuracy": 0.17058235704898833,
      "num_tokens": 35903641.0,
      "step": 16380
    },
    {
      "entropy": 5.511588525772095,
      "epoch": 1.6261413259229853,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004741028305532757,
      "loss": 5.0773,
      "mean_token_accuracy": 0.17215801030397415,
      "num_tokens": 35913634.0,
      "step": 16385
    },
    {
      "entropy": 5.548446798324585,
      "epoch": 1.6266375545851528,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004740863248639907,
      "loss": 5.1221,
      "mean_token_accuracy": 0.17387913465499877,
      "num_tokens": 35923106.0,
      "step": 16390
    },
    {
      "entropy": 5.556178951263428,
      "epoch": 1.6271337832473205,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047406981423779376,
      "loss": 5.1118,
      "mean_token_accuracy": 0.17499779313802719,
      "num_tokens": 35933709.0,
      "step": 16395
    },
    {
      "entropy": 5.517813062667846,
      "epoch": 1.627630011909488,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047405329867509447,
      "loss": 5.1352,
      "mean_token_accuracy": 0.16638769805431367,
      "num_tokens": 35944734.0,
      "step": 16400
    },
    {
      "entropy": 5.53065299987793,
      "epoch": 1.6281262405716554,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047403677817630226,
      "loss": 5.0996,
      "mean_token_accuracy": 0.17241687923669816,
      "num_tokens": 35956513.0,
      "step": 16405
    },
    {
      "entropy": 5.495169544219971,
      "epoch": 1.628622469233823,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004740202527418269,
      "loss": 5.115,
      "mean_token_accuracy": 0.1735220193862915,
      "num_tokens": 35967096.0,
      "step": 16410
    },
    {
      "entropy": 5.566775226593018,
      "epoch": 1.6291186978959904,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000474003722372078,
      "loss": 5.1936,
      "mean_token_accuracy": 0.17107468843460083,
      "num_tokens": 35979010.0,
      "step": 16415
    },
    {
      "entropy": 5.52549147605896,
      "epoch": 1.6296149265581579,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047398718706746556,
      "loss": 5.0446,
      "mean_token_accuracy": 0.1739116832613945,
      "num_tokens": 35989253.0,
      "step": 16420
    },
    {
      "entropy": 5.518692302703857,
      "epoch": 1.6301111552203256,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004739706468283998,
      "loss": 5.0663,
      "mean_token_accuracy": 0.17488324493169785,
      "num_tokens": 35999659.0,
      "step": 16425
    },
    {
      "entropy": 5.507155132293701,
      "epoch": 1.630607383882493,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047395410165529075,
      "loss": 5.1269,
      "mean_token_accuracy": 0.16918691098690034,
      "num_tokens": 36010295.0,
      "step": 16430
    },
    {
      "entropy": 5.563255739212036,
      "epoch": 1.6311036125446607,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047393755154854855,
      "loss": 5.15,
      "mean_token_accuracy": 0.1705315187573433,
      "num_tokens": 36021575.0,
      "step": 16435
    },
    {
      "entropy": 5.58696722984314,
      "epoch": 1.6315998412068282,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004739209965085839,
      "loss": 5.1411,
      "mean_token_accuracy": 0.16836782693862914,
      "num_tokens": 36032764.0,
      "step": 16440
    },
    {
      "entropy": 5.540527439117431,
      "epoch": 1.6320960698689957,
      "grad_norm": 1.015625,
      "learning_rate": 0.000473904436535807,
      "loss": 5.0997,
      "mean_token_accuracy": 0.1794095903635025,
      "num_tokens": 36041864.0,
      "step": 16445
    },
    {
      "entropy": 5.476602792739868,
      "epoch": 1.6325922985311632,
      "grad_norm": 1.0,
      "learning_rate": 0.0004738878716306288,
      "loss": 5.0864,
      "mean_token_accuracy": 0.17238736450672149,
      "num_tokens": 36051321.0,
      "step": 16450
    },
    {
      "entropy": 5.504956436157227,
      "epoch": 1.6330885271933306,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047387130179345996,
      "loss": 5.0609,
      "mean_token_accuracy": 0.17036915570497513,
      "num_tokens": 36061448.0,
      "step": 16455
    },
    {
      "entropy": 5.558578538894653,
      "epoch": 1.6335847558554981,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047385472702471133,
      "loss": 5.1948,
      "mean_token_accuracy": 0.17376814037561417,
      "num_tokens": 36071832.0,
      "step": 16460
    },
    {
      "entropy": 5.545031309127808,
      "epoch": 1.6340809845176656,
      "grad_norm": 0.953125,
      "learning_rate": 0.000473838147324794,
      "loss": 5.1079,
      "mean_token_accuracy": 0.17153619229793549,
      "num_tokens": 36082103.0,
      "step": 16465
    },
    {
      "entropy": 5.593025016784668,
      "epoch": 1.6345772131798333,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000473821562694119,
      "loss": 5.1478,
      "mean_token_accuracy": 0.1702588453888893,
      "num_tokens": 36092063.0,
      "step": 16470
    },
    {
      "entropy": 5.569368028640747,
      "epoch": 1.6350734418420008,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004738049731330977,
      "loss": 5.1355,
      "mean_token_accuracy": 0.16376752853393556,
      "num_tokens": 36102573.0,
      "step": 16475
    },
    {
      "entropy": 5.599412155151367,
      "epoch": 1.6355696705041685,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004737883786421414,
      "loss": 5.1153,
      "mean_token_accuracy": 0.17183687835931777,
      "num_tokens": 36112677.0,
      "step": 16480
    },
    {
      "entropy": 5.562631511688233,
      "epoch": 1.636065899166336,
      "grad_norm": 0.890625,
      "learning_rate": 0.00047377177922166175,
      "loss": 5.1816,
      "mean_token_accuracy": 0.16841134577989578,
      "num_tokens": 36123337.0,
      "step": 16485
    },
    {
      "entropy": 5.598000764846802,
      "epoch": 1.6365621278285034,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00047375517487207024,
      "loss": 5.1493,
      "mean_token_accuracy": 0.173994217813015,
      "num_tokens": 36134503.0,
      "step": 16490
    },
    {
      "entropy": 5.64011869430542,
      "epoch": 1.6370583564906709,
      "grad_norm": 0.890625,
      "learning_rate": 0.00047373856559377876,
      "loss": 5.2186,
      "mean_token_accuracy": 0.16748932301998137,
      "num_tokens": 36145766.0,
      "step": 16495
    },
    {
      "entropy": 5.591852426528931,
      "epoch": 1.6375545851528384,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004737219513871989,
      "loss": 5.1654,
      "mean_token_accuracy": 0.17032527923583984,
      "num_tokens": 36156584.0,
      "step": 16500
    },
    {
      "entropy": 5.534194231033325,
      "epoch": 1.6380508138150058,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00047370533225274305,
      "loss": 5.147,
      "mean_token_accuracy": 0.17428812086582185,
      "num_tokens": 36166999.0,
      "step": 16505
    },
    {
      "entropy": 5.56323504447937,
      "epoch": 1.6385470424771735,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004736887081908231,
      "loss": 5.1166,
      "mean_token_accuracy": 0.17571138590574265,
      "num_tokens": 36178724.0,
      "step": 16510
    },
    {
      "entropy": 5.64340615272522,
      "epoch": 1.639043271139341,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004736720792018513,
      "loss": 5.2021,
      "mean_token_accuracy": 0.16817369908094407,
      "num_tokens": 36188912.0,
      "step": 16515
    },
    {
      "entropy": 5.541040754318237,
      "epoch": 1.6395394998015085,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004736554452862399,
      "loss": 5.0997,
      "mean_token_accuracy": 0.17425263971090316,
      "num_tokens": 36198525.0,
      "step": 16520
    },
    {
      "entropy": 5.508972072601319,
      "epoch": 1.6400357284636762,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047363880644440173,
      "loss": 5.0813,
      "mean_token_accuracy": 0.17363713532686234,
      "num_tokens": 36209384.0,
      "step": 16525
    },
    {
      "entropy": 5.598171281814575,
      "epoch": 1.6405319571258437,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047362216267674906,
      "loss": 5.1677,
      "mean_token_accuracy": 0.16821869164705278,
      "num_tokens": 36220597.0,
      "step": 16530
    },
    {
      "entropy": 5.626340723037719,
      "epoch": 1.6410281857880111,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004736055139836948,
      "loss": 5.2122,
      "mean_token_accuracy": 0.17054324597120285,
      "num_tokens": 36230317.0,
      "step": 16535
    },
    {
      "entropy": 5.686342430114746,
      "epoch": 1.6415244144501786,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047358886036565176,
      "loss": 5.2914,
      "mean_token_accuracy": 0.16591713726520538,
      "num_tokens": 36241645.0,
      "step": 16540
    },
    {
      "entropy": 5.599885749816894,
      "epoch": 1.642020643112346,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047357220182303285,
      "loss": 5.2165,
      "mean_token_accuracy": 0.1655209943652153,
      "num_tokens": 36253098.0,
      "step": 16545
    },
    {
      "entropy": 5.538655519485474,
      "epoch": 1.6425168717745136,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004735555383562513,
      "loss": 5.1113,
      "mean_token_accuracy": 0.17447598278522491,
      "num_tokens": 36264417.0,
      "step": 16550
    },
    {
      "entropy": 5.635347318649292,
      "epoch": 1.6430131004366813,
      "grad_norm": 0.828125,
      "learning_rate": 0.00047353886996572014,
      "loss": 5.2061,
      "mean_token_accuracy": 0.16311426609754562,
      "num_tokens": 36276589.0,
      "step": 16555
    },
    {
      "entropy": 5.577414226531983,
      "epoch": 1.6435093290988487,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047352219665185293,
      "loss": 5.1238,
      "mean_token_accuracy": 0.16760046929121017,
      "num_tokens": 36287308.0,
      "step": 16560
    },
    {
      "entropy": 5.541089010238648,
      "epoch": 1.6440055577610164,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004735055184150629,
      "loss": 5.1248,
      "mean_token_accuracy": 0.17476615756750108,
      "num_tokens": 36298966.0,
      "step": 16565
    },
    {
      "entropy": 5.509775972366333,
      "epoch": 1.644501786423184,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047348883525576385,
      "loss": 5.1361,
      "mean_token_accuracy": 0.16911161541938782,
      "num_tokens": 36310498.0,
      "step": 16570
    },
    {
      "entropy": 5.5917764663696286,
      "epoch": 1.6449980150853514,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004734721471743694,
      "loss": 5.2006,
      "mean_token_accuracy": 0.1728076919913292,
      "num_tokens": 36321589.0,
      "step": 16575
    },
    {
      "entropy": 5.565700817108154,
      "epoch": 1.6454942437475188,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004734554541712933,
      "loss": 5.1009,
      "mean_token_accuracy": 0.16892541497945784,
      "num_tokens": 36333041.0,
      "step": 16580
    },
    {
      "entropy": 5.602836275100708,
      "epoch": 1.6459904724096863,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004734387562469496,
      "loss": 5.176,
      "mean_token_accuracy": 0.16549454629421234,
      "num_tokens": 36344225.0,
      "step": 16585
    },
    {
      "entropy": 5.569616603851318,
      "epoch": 1.6464867010718538,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047342205340175235,
      "loss": 5.066,
      "mean_token_accuracy": 0.17604072988033295,
      "num_tokens": 36355524.0,
      "step": 16590
    },
    {
      "entropy": 5.583031368255615,
      "epoch": 1.6469829297340213,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047340534563611574,
      "loss": 5.1464,
      "mean_token_accuracy": 0.1758979856967926,
      "num_tokens": 36366495.0,
      "step": 16595
    },
    {
      "entropy": 5.566857099533081,
      "epoch": 1.647479158396189,
      "grad_norm": 0.9375,
      "learning_rate": 0.00047338863295045414,
      "loss": 5.1339,
      "mean_token_accuracy": 0.17026081532239914,
      "num_tokens": 36378316.0,
      "step": 16600
    },
    {
      "entropy": 5.605424594879151,
      "epoch": 1.6479753870583564,
      "grad_norm": 0.9375,
      "learning_rate": 0.00047337191534518186,
      "loss": 5.2239,
      "mean_token_accuracy": 0.16221691519021988,
      "num_tokens": 36390500.0,
      "step": 16605
    },
    {
      "entropy": 5.609649991989135,
      "epoch": 1.6484716157205241,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004733551928207135,
      "loss": 5.1606,
      "mean_token_accuracy": 0.1712506279349327,
      "num_tokens": 36403831.0,
      "step": 16610
    },
    {
      "entropy": 5.569226312637329,
      "epoch": 1.6489678443826916,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004733384653774639,
      "loss": 5.1242,
      "mean_token_accuracy": 0.17524966299533845,
      "num_tokens": 36414896.0,
      "step": 16615
    },
    {
      "entropy": 5.519056224822998,
      "epoch": 1.649464073044859,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047332173301584764,
      "loss": 5.0621,
      "mean_token_accuracy": 0.17273387908935547,
      "num_tokens": 36426260.0,
      "step": 16620
    },
    {
      "entropy": 5.48406753540039,
      "epoch": 1.6499603017070266,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047330499573627975,
      "loss": 5.0224,
      "mean_token_accuracy": 0.18159683048725128,
      "num_tokens": 36437125.0,
      "step": 16625
    },
    {
      "entropy": 5.542082023620606,
      "epoch": 1.650456530369194,
      "grad_norm": 1.0,
      "learning_rate": 0.00047328825353917526,
      "loss": 5.0679,
      "mean_token_accuracy": 0.1763029381632805,
      "num_tokens": 36447507.0,
      "step": 16630
    },
    {
      "entropy": 5.6174413681030275,
      "epoch": 1.6509527590313615,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047327150642494944,
      "loss": 5.2347,
      "mean_token_accuracy": 0.16983219236135483,
      "num_tokens": 36459134.0,
      "step": 16635
    },
    {
      "entropy": 5.606341028213501,
      "epoch": 1.6514489876935292,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004732547543940174,
      "loss": 5.1425,
      "mean_token_accuracy": 0.1669211804866791,
      "num_tokens": 36469594.0,
      "step": 16640
    },
    {
      "entropy": 5.553228759765625,
      "epoch": 1.6519452163556967,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004732379974467947,
      "loss": 5.1445,
      "mean_token_accuracy": 0.17724878042936326,
      "num_tokens": 36480477.0,
      "step": 16645
    },
    {
      "entropy": 5.52822093963623,
      "epoch": 1.6524414450178644,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004732212355836968,
      "loss": 5.1445,
      "mean_token_accuracy": 0.1737054780125618,
      "num_tokens": 36491304.0,
      "step": 16650
    },
    {
      "entropy": 5.6495777606964115,
      "epoch": 1.6529376736800319,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004732044688051395,
      "loss": 5.2132,
      "mean_token_accuracy": 0.1647946760058403,
      "num_tokens": 36502030.0,
      "step": 16655
    },
    {
      "entropy": 5.585239744186401,
      "epoch": 1.6534339023421993,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047318769711153833,
      "loss": 5.2391,
      "mean_token_accuracy": 0.16840274035930633,
      "num_tokens": 36514405.0,
      "step": 16660
    },
    {
      "entropy": 5.597030115127564,
      "epoch": 1.6539301310043668,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004731709205033094,
      "loss": 5.1104,
      "mean_token_accuracy": 0.17021507918834686,
      "num_tokens": 36524626.0,
      "step": 16665
    },
    {
      "entropy": 5.608535528182983,
      "epoch": 1.6544263596665343,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047315413898086863,
      "loss": 5.1091,
      "mean_token_accuracy": 0.1769067496061325,
      "num_tokens": 36535340.0,
      "step": 16670
    },
    {
      "entropy": 5.593055534362793,
      "epoch": 1.6549225883287018,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004731373525446322,
      "loss": 5.1108,
      "mean_token_accuracy": 0.17366884648799896,
      "num_tokens": 36547189.0,
      "step": 16675
    },
    {
      "entropy": 5.573965406417846,
      "epoch": 1.6554188169908692,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004731205611950164,
      "loss": 5.1947,
      "mean_token_accuracy": 0.16631772369146347,
      "num_tokens": 36558177.0,
      "step": 16680
    },
    {
      "entropy": 5.528534984588623,
      "epoch": 1.655915045653037,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047310376493243754,
      "loss": 5.0156,
      "mean_token_accuracy": 0.1760718896985054,
      "num_tokens": 36569559.0,
      "step": 16685
    },
    {
      "entropy": 5.567520809173584,
      "epoch": 1.6564112743152044,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004730869637573122,
      "loss": 5.0827,
      "mean_token_accuracy": 0.1765242785215378,
      "num_tokens": 36579617.0,
      "step": 16690
    },
    {
      "entropy": 5.549475860595703,
      "epoch": 1.656907502977372,
      "grad_norm": 0.83203125,
      "learning_rate": 0.00047307015767005705,
      "loss": 5.0752,
      "mean_token_accuracy": 0.17589021474123,
      "num_tokens": 36590427.0,
      "step": 16695
    },
    {
      "entropy": 5.52656102180481,
      "epoch": 1.6574037316395396,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004730533466710888,
      "loss": 5.0776,
      "mean_token_accuracy": 0.17531200647354125,
      "num_tokens": 36602313.0,
      "step": 16700
    },
    {
      "entropy": 5.605986881256103,
      "epoch": 1.657899960301707,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004730365307608242,
      "loss": 5.1765,
      "mean_token_accuracy": 0.17297641038894654,
      "num_tokens": 36615191.0,
      "step": 16705
    },
    {
      "entropy": 5.565281629562378,
      "epoch": 1.6583961889638745,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047301970993968054,
      "loss": 5.1021,
      "mean_token_accuracy": 0.1765014037489891,
      "num_tokens": 36624692.0,
      "step": 16710
    },
    {
      "entropy": 5.580508518218994,
      "epoch": 1.658892417626042,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047300288420807463,
      "loss": 5.1744,
      "mean_token_accuracy": 0.1661687821149826,
      "num_tokens": 36635961.0,
      "step": 16715
    },
    {
      "entropy": 5.543060684204102,
      "epoch": 1.6593886462882095,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047298605356642387,
      "loss": 5.1289,
      "mean_token_accuracy": 0.17295888513326646,
      "num_tokens": 36646710.0,
      "step": 16720
    },
    {
      "entropy": 5.577163171768189,
      "epoch": 1.6598848749503772,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047296921801514565,
      "loss": 5.2162,
      "mean_token_accuracy": 0.16633112877607345,
      "num_tokens": 36657168.0,
      "step": 16725
    },
    {
      "entropy": 5.57464828491211,
      "epoch": 1.6603811036125447,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004729523775546574,
      "loss": 5.1967,
      "mean_token_accuracy": 0.17284595519304274,
      "num_tokens": 36668428.0,
      "step": 16730
    },
    {
      "entropy": 5.5711745262146,
      "epoch": 1.6608773322747123,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004729355321853767,
      "loss": 5.1361,
      "mean_token_accuracy": 0.17802993655204774,
      "num_tokens": 36679361.0,
      "step": 16735
    },
    {
      "entropy": 5.648694181442261,
      "epoch": 1.6613735609368798,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004729186819077214,
      "loss": 5.194,
      "mean_token_accuracy": 0.170845428109169,
      "num_tokens": 36689480.0,
      "step": 16740
    },
    {
      "entropy": 5.64420747756958,
      "epoch": 1.6618697895990473,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004729018267221092,
      "loss": 5.2333,
      "mean_token_accuracy": 0.17059906125068663,
      "num_tokens": 36700420.0,
      "step": 16745
    },
    {
      "entropy": 5.64308443069458,
      "epoch": 1.6623660182612148,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004728849666289581,
      "loss": 5.1824,
      "mean_token_accuracy": 0.17325215488672258,
      "num_tokens": 36710106.0,
      "step": 16750
    },
    {
      "entropy": 5.581542301177978,
      "epoch": 1.6628622469233822,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047286810162868625,
      "loss": 5.0579,
      "mean_token_accuracy": 0.18079546242952346,
      "num_tokens": 36719630.0,
      "step": 16755
    },
    {
      "entropy": 5.464664649963379,
      "epoch": 1.6633584755855497,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004728512317217119,
      "loss": 5.0399,
      "mean_token_accuracy": 0.17958485782146455,
      "num_tokens": 36729379.0,
      "step": 16760
    },
    {
      "entropy": 5.484744596481323,
      "epoch": 1.6638547042477172,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047283435690845336,
      "loss": 5.0921,
      "mean_token_accuracy": 0.1769113913178444,
      "num_tokens": 36740103.0,
      "step": 16765
    },
    {
      "entropy": 5.565602731704712,
      "epoch": 1.664350932909885,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047281747718932906,
      "loss": 5.135,
      "mean_token_accuracy": 0.17403532713651657,
      "num_tokens": 36750928.0,
      "step": 16770
    },
    {
      "entropy": 5.570236349105835,
      "epoch": 1.6648471615720524,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004728005925647575,
      "loss": 5.1433,
      "mean_token_accuracy": 0.17115781158208848,
      "num_tokens": 36761415.0,
      "step": 16775
    },
    {
      "entropy": 5.541550922393799,
      "epoch": 1.66534339023422,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047278370303515744,
      "loss": 5.1716,
      "mean_token_accuracy": 0.16619772613048553,
      "num_tokens": 36772560.0,
      "step": 16780
    },
    {
      "entropy": 5.570325660705566,
      "epoch": 1.6658396188963875,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004727668086009478,
      "loss": 5.1771,
      "mean_token_accuracy": 0.16833605915307998,
      "num_tokens": 36783597.0,
      "step": 16785
    },
    {
      "entropy": 5.623755311965942,
      "epoch": 1.666335847558555,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047274990926254753,
      "loss": 5.241,
      "mean_token_accuracy": 0.16613737046718596,
      "num_tokens": 36794219.0,
      "step": 16790
    },
    {
      "entropy": 5.531423044204712,
      "epoch": 1.6668320762207225,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004727330050203755,
      "loss": 5.0706,
      "mean_token_accuracy": 0.17751879394054412,
      "num_tokens": 36804156.0,
      "step": 16795
    },
    {
      "entropy": 5.526420593261719,
      "epoch": 1.66732830488289,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004727160958748511,
      "loss": 5.0992,
      "mean_token_accuracy": 0.17404341101646423,
      "num_tokens": 36814414.0,
      "step": 16800
    },
    {
      "entropy": 5.537476396560669,
      "epoch": 1.6678245335450574,
      "grad_norm": 0.9375,
      "learning_rate": 0.00047269918182639355,
      "loss": 5.0538,
      "mean_token_accuracy": 0.17120650112628938,
      "num_tokens": 36826173.0,
      "step": 16805
    },
    {
      "entropy": 5.599848461151123,
      "epoch": 1.6683207622072251,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004726822628754222,
      "loss": 5.1583,
      "mean_token_accuracy": 0.17181871086359024,
      "num_tokens": 36837726.0,
      "step": 16810
    },
    {
      "entropy": 5.598222827911377,
      "epoch": 1.6688169908693926,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047266533902235683,
      "loss": 5.1289,
      "mean_token_accuracy": 0.16565078496932983,
      "num_tokens": 36848605.0,
      "step": 16815
    },
    {
      "entropy": 5.582981300354004,
      "epoch": 1.66931321953156,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004726484102676169,
      "loss": 5.153,
      "mean_token_accuracy": 0.17065990567207337,
      "num_tokens": 36858730.0,
      "step": 16820
    },
    {
      "entropy": 5.526777505874634,
      "epoch": 1.6698094481937278,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004726314766116223,
      "loss": 5.0192,
      "mean_token_accuracy": 0.17582650780677794,
      "num_tokens": 36868239.0,
      "step": 16825
    },
    {
      "entropy": 5.5396997928619385,
      "epoch": 1.6703056768558953,
      "grad_norm": 1.15625,
      "learning_rate": 0.000472614538054793,
      "loss": 5.105,
      "mean_token_accuracy": 0.17341177314519882,
      "num_tokens": 36878077.0,
      "step": 16830
    },
    {
      "entropy": 5.55140323638916,
      "epoch": 1.6708019055180627,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00047259759459754893,
      "loss": 5.1533,
      "mean_token_accuracy": 0.16827925592660903,
      "num_tokens": 36888874.0,
      "step": 16835
    },
    {
      "entropy": 5.578630161285401,
      "epoch": 1.6712981341802302,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004725806462403103,
      "loss": 5.1435,
      "mean_token_accuracy": 0.17273767292499542,
      "num_tokens": 36898584.0,
      "step": 16840
    },
    {
      "entropy": 5.573261022567749,
      "epoch": 1.6717943628423977,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047256369298349736,
      "loss": 5.1368,
      "mean_token_accuracy": 0.17315253764390945,
      "num_tokens": 36908662.0,
      "step": 16845
    },
    {
      "entropy": 5.5238433361053465,
      "epoch": 1.6722905915045652,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047254673482753065,
      "loss": 5.0781,
      "mean_token_accuracy": 0.17288859486579894,
      "num_tokens": 36919482.0,
      "step": 16850
    },
    {
      "entropy": 5.538987350463867,
      "epoch": 1.6727868201667329,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00047252977177283045,
      "loss": 5.1121,
      "mean_token_accuracy": 0.16581951528787614,
      "num_tokens": 36930470.0,
      "step": 16855
    },
    {
      "entropy": 5.563366413116455,
      "epoch": 1.6732830488289003,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00047251280381981764,
      "loss": 5.1026,
      "mean_token_accuracy": 0.1766013354063034,
      "num_tokens": 36942837.0,
      "step": 16860
    },
    {
      "entropy": 5.6174417495727536,
      "epoch": 1.673779277491068,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004724958309689129,
      "loss": 5.1788,
      "mean_token_accuracy": 0.166561958193779,
      "num_tokens": 36952994.0,
      "step": 16865
    },
    {
      "entropy": 5.551107740402221,
      "epoch": 1.6742755061532355,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00047247885322053705,
      "loss": 5.1597,
      "mean_token_accuracy": 0.16894342005252838,
      "num_tokens": 36963973.0,
      "step": 16870
    },
    {
      "entropy": 5.503463888168335,
      "epoch": 1.674771734815403,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00047246187057511127,
      "loss": 5.0956,
      "mean_token_accuracy": 0.17601613700389862,
      "num_tokens": 36975897.0,
      "step": 16875
    },
    {
      "entropy": 5.553459167480469,
      "epoch": 1.6752679634775705,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004724448830330565,
      "loss": 5.1027,
      "mean_token_accuracy": 0.1754691183567047,
      "num_tokens": 36986481.0,
      "step": 16880
    },
    {
      "entropy": 5.567030382156372,
      "epoch": 1.675764192139738,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00047242789059479413,
      "loss": 5.192,
      "mean_token_accuracy": 0.1680533319711685,
      "num_tokens": 36997863.0,
      "step": 16885
    },
    {
      "entropy": 5.620091247558594,
      "epoch": 1.6762604208019054,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047241089326074547,
      "loss": 5.1383,
      "mean_token_accuracy": 0.1653734490275383,
      "num_tokens": 37009294.0,
      "step": 16890
    },
    {
      "entropy": 5.612838077545166,
      "epoch": 1.6767566494640729,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004723938910313321,
      "loss": 5.1261,
      "mean_token_accuracy": 0.17353014796972274,
      "num_tokens": 37019748.0,
      "step": 16895
    },
    {
      "entropy": 5.588705968856812,
      "epoch": 1.6772528781262406,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047237688390697555,
      "loss": 5.1778,
      "mean_token_accuracy": 0.17644532546401023,
      "num_tokens": 37030903.0,
      "step": 16900
    },
    {
      "entropy": 5.5499029636383055,
      "epoch": 1.677749106788408,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004723598718880975,
      "loss": 5.1043,
      "mean_token_accuracy": 0.1754912927746773,
      "num_tokens": 37042551.0,
      "step": 16905
    },
    {
      "entropy": 5.577929353713989,
      "epoch": 1.6782453354505757,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047234285497512005,
      "loss": 5.1709,
      "mean_token_accuracy": 0.16915378719568253,
      "num_tokens": 37053151.0,
      "step": 16910
    },
    {
      "entropy": 5.629922771453858,
      "epoch": 1.6787415641127432,
      "grad_norm": 0.875,
      "learning_rate": 0.00047232583316846494,
      "loss": 5.1829,
      "mean_token_accuracy": 0.166587033867836,
      "num_tokens": 37064421.0,
      "step": 16915
    },
    {
      "entropy": 5.578013801574707,
      "epoch": 1.6792377927749107,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004723088064685544,
      "loss": 5.1295,
      "mean_token_accuracy": 0.17486939132213591,
      "num_tokens": 37074758.0,
      "step": 16920
    },
    {
      "entropy": 5.577193069458008,
      "epoch": 1.6797340214370782,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047229177487581055,
      "loss": 5.0896,
      "mean_token_accuracy": 0.16904166638851165,
      "num_tokens": 37084898.0,
      "step": 16925
    },
    {
      "entropy": 5.608043193817139,
      "epoch": 1.6802302500992456,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047227473839065587,
      "loss": 5.1651,
      "mean_token_accuracy": 0.1672091618180275,
      "num_tokens": 37094366.0,
      "step": 16930
    },
    {
      "entropy": 5.676841306686401,
      "epoch": 1.6807264787614131,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047225769701351284,
      "loss": 5.2253,
      "mean_token_accuracy": 0.1685462027788162,
      "num_tokens": 37104965.0,
      "step": 16935
    },
    {
      "entropy": 5.537209129333496,
      "epoch": 1.6812227074235808,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004722406507448038,
      "loss": 5.0931,
      "mean_token_accuracy": 0.17549068480730057,
      "num_tokens": 37115622.0,
      "step": 16940
    },
    {
      "entropy": 5.609269428253174,
      "epoch": 1.6817189360857483,
      "grad_norm": 0.875,
      "learning_rate": 0.00047222359958495175,
      "loss": 5.1824,
      "mean_token_accuracy": 0.16754831671714782,
      "num_tokens": 37127234.0,
      "step": 16945
    },
    {
      "entropy": 5.570038747787476,
      "epoch": 1.682215164747916,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004722065435343794,
      "loss": 5.0474,
      "mean_token_accuracy": 0.17715921103954316,
      "num_tokens": 37136533.0,
      "step": 16950
    },
    {
      "entropy": 5.583553886413574,
      "epoch": 1.6827113934100835,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004721894825935097,
      "loss": 5.1544,
      "mean_token_accuracy": 0.1715320512652397,
      "num_tokens": 37147821.0,
      "step": 16955
    },
    {
      "entropy": 5.5517964363098145,
      "epoch": 1.683207622072251,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004721724167627657,
      "loss": 5.1583,
      "mean_token_accuracy": 0.17215365469455718,
      "num_tokens": 37158929.0,
      "step": 16960
    },
    {
      "entropy": 5.596142292022705,
      "epoch": 1.6837038507344184,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004721553460425708,
      "loss": 5.1548,
      "mean_token_accuracy": 0.17849822491407394,
      "num_tokens": 37169817.0,
      "step": 16965
    },
    {
      "entropy": 5.5814528465271,
      "epoch": 1.684200079396586,
      "grad_norm": 1.03125,
      "learning_rate": 0.000472138270433348,
      "loss": 5.2621,
      "mean_token_accuracy": 0.16471544280648232,
      "num_tokens": 37181009.0,
      "step": 16970
    },
    {
      "entropy": 5.552694463729859,
      "epoch": 1.6846963080587534,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047212118993552097,
      "loss": 5.089,
      "mean_token_accuracy": 0.1761760577559471,
      "num_tokens": 37192110.0,
      "step": 16975
    },
    {
      "entropy": 5.575069570541382,
      "epoch": 1.6851925367209208,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004721041045495131,
      "loss": 5.136,
      "mean_token_accuracy": 0.17346104234457016,
      "num_tokens": 37203426.0,
      "step": 16980
    },
    {
      "entropy": 5.619303369522095,
      "epoch": 1.6856887653830885,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047208701427574817,
      "loss": 5.2176,
      "mean_token_accuracy": 0.16681872457265853,
      "num_tokens": 37215025.0,
      "step": 16985
    },
    {
      "entropy": 5.6421293258667,
      "epoch": 1.686184994045256,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047206991911465,
      "loss": 5.1477,
      "mean_token_accuracy": 0.17911407649517058,
      "num_tokens": 37225653.0,
      "step": 16990
    },
    {
      "entropy": 5.562642669677734,
      "epoch": 1.6866812227074237,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004720528190666425,
      "loss": 5.1179,
      "mean_token_accuracy": 0.17100996524095535,
      "num_tokens": 37237282.0,
      "step": 16995
    },
    {
      "entropy": 5.496473217010498,
      "epoch": 1.6871774513695912,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047203571413214967,
      "loss": 5.0817,
      "mean_token_accuracy": 0.18428921550512314,
      "num_tokens": 37248476.0,
      "step": 17000
    },
    {
      "entropy": 5.570226001739502,
      "epoch": 1.6876736800317587,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004720186043115957,
      "loss": 5.0737,
      "mean_token_accuracy": 0.1730433762073517,
      "num_tokens": 37259354.0,
      "step": 17005
    },
    {
      "entropy": 5.569587993621826,
      "epoch": 1.6881699086939261,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004720014896054048,
      "loss": 5.1428,
      "mean_token_accuracy": 0.16817787587642669,
      "num_tokens": 37269895.0,
      "step": 17010
    },
    {
      "entropy": 5.536753702163696,
      "epoch": 1.6886661373560936,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004719843700140017,
      "loss": 5.1154,
      "mean_token_accuracy": 0.17986234724521638,
      "num_tokens": 37281201.0,
      "step": 17015
    },
    {
      "entropy": 5.558519506454468,
      "epoch": 1.689162366018261,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004719672455378105,
      "loss": 5.1291,
      "mean_token_accuracy": 0.1699021965265274,
      "num_tokens": 37293362.0,
      "step": 17020
    },
    {
      "entropy": 5.637897920608521,
      "epoch": 1.6896585946804288,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000471950116177256,
      "loss": 5.1611,
      "mean_token_accuracy": 0.1686394229531288,
      "num_tokens": 37304291.0,
      "step": 17025
    },
    {
      "entropy": 5.596599674224853,
      "epoch": 1.6901548233425963,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004719329819327631,
      "loss": 5.1438,
      "mean_token_accuracy": 0.1711001694202423,
      "num_tokens": 37315608.0,
      "step": 17030
    },
    {
      "entropy": 5.569362735748291,
      "epoch": 1.690651052004764,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004719158428047566,
      "loss": 5.0943,
      "mean_token_accuracy": 0.168352709710598,
      "num_tokens": 37327559.0,
      "step": 17035
    },
    {
      "entropy": 5.5401458740234375,
      "epoch": 1.6911472806669314,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004718986987936615,
      "loss": 5.1212,
      "mean_token_accuracy": 0.17786616533994676,
      "num_tokens": 37338101.0,
      "step": 17040
    },
    {
      "entropy": 5.581073331832886,
      "epoch": 1.691643509329099,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00047188154989990296,
      "loss": 5.2042,
      "mean_token_accuracy": 0.17016657441854477,
      "num_tokens": 37349124.0,
      "step": 17045
    },
    {
      "entropy": 5.540998935699463,
      "epoch": 1.6921397379912664,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004718643961239062,
      "loss": 5.0752,
      "mean_token_accuracy": 0.17614357471466063,
      "num_tokens": 37360061.0,
      "step": 17050
    },
    {
      "entropy": 5.579684495925903,
      "epoch": 1.6926359666534339,
      "grad_norm": 0.90625,
      "learning_rate": 0.00047184723746609666,
      "loss": 5.1343,
      "mean_token_accuracy": 0.17226943820714952,
      "num_tokens": 37371210.0,
      "step": 17055
    },
    {
      "entropy": 5.542012166976929,
      "epoch": 1.6931321953156013,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047183007392689977,
      "loss": 5.0946,
      "mean_token_accuracy": 0.17004089504480363,
      "num_tokens": 37382547.0,
      "step": 17060
    },
    {
      "entropy": 5.546593952178955,
      "epoch": 1.6936284239777688,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047181290550674115,
      "loss": 5.1562,
      "mean_token_accuracy": 0.17347490042448044,
      "num_tokens": 37392311.0,
      "step": 17065
    },
    {
      "entropy": 5.5597306251525875,
      "epoch": 1.6941246526399365,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004717957322060466,
      "loss": 5.0964,
      "mean_token_accuracy": 0.17364707738161086,
      "num_tokens": 37402626.0,
      "step": 17070
    },
    {
      "entropy": 5.561833381652832,
      "epoch": 1.694620881302104,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004717785540252419,
      "loss": 5.1036,
      "mean_token_accuracy": 0.17350115478038788,
      "num_tokens": 37412100.0,
      "step": 17075
    },
    {
      "entropy": 5.498662233352661,
      "epoch": 1.6951171099642717,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004717613709647532,
      "loss": 5.0116,
      "mean_token_accuracy": 0.17816671580076218,
      "num_tokens": 37421735.0,
      "step": 17080
    },
    {
      "entropy": 5.607385110855103,
      "epoch": 1.6956133386264391,
      "grad_norm": 1.0,
      "learning_rate": 0.00047174418302500635,
      "loss": 5.2093,
      "mean_token_accuracy": 0.16542862951755524,
      "num_tokens": 37432179.0,
      "step": 17085
    },
    {
      "entropy": 5.561559677124023,
      "epoch": 1.6961095672886066,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004717269902064278,
      "loss": 5.0763,
      "mean_token_accuracy": 0.18370807766914368,
      "num_tokens": 37443276.0,
      "step": 17090
    },
    {
      "entropy": 5.583922719955444,
      "epoch": 1.696605795950774,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004717097925094438,
      "loss": 5.1894,
      "mean_token_accuracy": 0.1680811569094658,
      "num_tokens": 37454537.0,
      "step": 17095
    },
    {
      "entropy": 5.591526508331299,
      "epoch": 1.6971020246129416,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004716925899344808,
      "loss": 5.1049,
      "mean_token_accuracy": 0.1662537544965744,
      "num_tokens": 37465493.0,
      "step": 17100
    },
    {
      "entropy": 5.507003307342529,
      "epoch": 1.697598253275109,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004716753824819654,
      "loss": 5.0996,
      "mean_token_accuracy": 0.17380036562681198,
      "num_tokens": 37475511.0,
      "step": 17105
    },
    {
      "entropy": 5.503319501876831,
      "epoch": 1.6980944819372767,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004716581701523243,
      "loss": 5.0102,
      "mean_token_accuracy": 0.17425770163536072,
      "num_tokens": 37487046.0,
      "step": 17110
    },
    {
      "entropy": 5.565830039978027,
      "epoch": 1.6985907105994442,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00047164095294598433,
      "loss": 5.1641,
      "mean_token_accuracy": 0.16879702657461165,
      "num_tokens": 37498398.0,
      "step": 17115
    },
    {
      "entropy": 5.57945761680603,
      "epoch": 1.699086939261612,
      "grad_norm": 0.90625,
      "learning_rate": 0.00047162373086337243,
      "loss": 5.1716,
      "mean_token_accuracy": 0.16734508275985718,
      "num_tokens": 37509397.0,
      "step": 17120
    },
    {
      "entropy": 5.572624397277832,
      "epoch": 1.6995831679237794,
      "grad_norm": 0.9375,
      "learning_rate": 0.00047160650390491574,
      "loss": 5.0499,
      "mean_token_accuracy": 0.18090201318264007,
      "num_tokens": 37519284.0,
      "step": 17125
    },
    {
      "entropy": 5.5761829853057865,
      "epoch": 1.7000793965859469,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004715892720710414,
      "loss": 5.1518,
      "mean_token_accuracy": 0.17086864560842513,
      "num_tokens": 37529699.0,
      "step": 17130
    },
    {
      "entropy": 5.636133861541748,
      "epoch": 1.7005756252481143,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004715720353621767,
      "loss": 5.2017,
      "mean_token_accuracy": 0.16297321021556854,
      "num_tokens": 37541756.0,
      "step": 17135
    },
    {
      "entropy": 5.502932071685791,
      "epoch": 1.7010718539102818,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047155479377874917,
      "loss": 5.0829,
      "mean_token_accuracy": 0.17542338222265244,
      "num_tokens": 37552693.0,
      "step": 17140
    },
    {
      "entropy": 5.546161079406739,
      "epoch": 1.7015680825724493,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047153754732118616,
      "loss": 5.062,
      "mean_token_accuracy": 0.1763912707567215,
      "num_tokens": 37563826.0,
      "step": 17145
    },
    {
      "entropy": 5.5860276222229,
      "epoch": 1.7020643112346168,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004715202959899156,
      "loss": 5.1042,
      "mean_token_accuracy": 0.17517814189195632,
      "num_tokens": 37576645.0,
      "step": 17150
    },
    {
      "entropy": 5.4545149326324465,
      "epoch": 1.7025605398967845,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004715030397853651,
      "loss": 4.9822,
      "mean_token_accuracy": 0.1835605651140213,
      "num_tokens": 37587467.0,
      "step": 17155
    },
    {
      "entropy": 5.592034387588501,
      "epoch": 1.703056768558952,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047148577870796264,
      "loss": 5.1635,
      "mean_token_accuracy": 0.1691695600748062,
      "num_tokens": 37598475.0,
      "step": 17160
    },
    {
      "entropy": 5.627919244766235,
      "epoch": 1.7035529972211196,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004714685127581363,
      "loss": 5.2541,
      "mean_token_accuracy": 0.16691816747188568,
      "num_tokens": 37610260.0,
      "step": 17165
    },
    {
      "entropy": 5.5737567901611325,
      "epoch": 1.704049225883287,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047145124193631415,
      "loss": 5.1827,
      "mean_token_accuracy": 0.16948528289794923,
      "num_tokens": 37621129.0,
      "step": 17170
    },
    {
      "entropy": 5.548666667938233,
      "epoch": 1.7045454545454546,
      "grad_norm": 0.921875,
      "learning_rate": 0.00047143396624292454,
      "loss": 5.0774,
      "mean_token_accuracy": 0.17280406951904298,
      "num_tokens": 37633130.0,
      "step": 17175
    },
    {
      "entropy": 5.599744749069214,
      "epoch": 1.705041683207622,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00047141668567839584,
      "loss": 5.2022,
      "mean_token_accuracy": 0.1700626716017723,
      "num_tokens": 37644247.0,
      "step": 17180
    },
    {
      "entropy": 5.616266393661499,
      "epoch": 1.7055379118697895,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004713994002431566,
      "loss": 5.1247,
      "mean_token_accuracy": 0.17535771131515504,
      "num_tokens": 37655602.0,
      "step": 17185
    },
    {
      "entropy": 5.5037469387054445,
      "epoch": 1.706034140531957,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00047138210993763534,
      "loss": 5.0519,
      "mean_token_accuracy": 0.18014779239892958,
      "num_tokens": 37667150.0,
      "step": 17190
    },
    {
      "entropy": 5.492997026443481,
      "epoch": 1.7065303691941247,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000471364814762261,
      "loss": 5.0593,
      "mean_token_accuracy": 0.1819021925330162,
      "num_tokens": 37677626.0,
      "step": 17195
    },
    {
      "entropy": 5.625694274902344,
      "epoch": 1.7070265978562922,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004713475147174624,
      "loss": 5.2086,
      "mean_token_accuracy": 0.16772229075431824,
      "num_tokens": 37689050.0,
      "step": 17200
    },
    {
      "entropy": 5.645072174072266,
      "epoch": 1.7075228265184597,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004713302098036685,
      "loss": 5.0943,
      "mean_token_accuracy": 0.18364367932081221,
      "num_tokens": 37699190.0,
      "step": 17205
    },
    {
      "entropy": 5.486385536193848,
      "epoch": 1.7080190551806274,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004713129000213085,
      "loss": 5.0834,
      "mean_token_accuracy": 0.17613888382911683,
      "num_tokens": 37710465.0,
      "step": 17210
    },
    {
      "entropy": 5.506699085235596,
      "epoch": 1.7085152838427948,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004712955853708115,
      "loss": 5.1841,
      "mean_token_accuracy": 0.17233196049928665,
      "num_tokens": 37722426.0,
      "step": 17215
    },
    {
      "entropy": 5.529746675491333,
      "epoch": 1.7090115125049623,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000471278265852607,
      "loss": 5.079,
      "mean_token_accuracy": 0.17865019738674165,
      "num_tokens": 37733104.0,
      "step": 17220
    },
    {
      "entropy": 5.527663230895996,
      "epoch": 1.7095077411671298,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004712609414671245,
      "loss": 5.0179,
      "mean_token_accuracy": 0.1805179536342621,
      "num_tokens": 37742747.0,
      "step": 17225
    },
    {
      "entropy": 5.536157608032227,
      "epoch": 1.7100039698292973,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004712436122147935,
      "loss": 5.0757,
      "mean_token_accuracy": 0.17503509670495987,
      "num_tokens": 37753338.0,
      "step": 17230
    },
    {
      "entropy": 5.518141984939575,
      "epoch": 1.7105001984914647,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004712262780960438,
      "loss": 5.0722,
      "mean_token_accuracy": 0.17581305652856827,
      "num_tokens": 37763122.0,
      "step": 17235
    },
    {
      "entropy": 5.539923191070557,
      "epoch": 1.7109964271536324,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047120893911130525,
      "loss": 5.0676,
      "mean_token_accuracy": 0.17038348764181138,
      "num_tokens": 37774677.0,
      "step": 17240
    },
    {
      "entropy": 5.534102201461792,
      "epoch": 1.7114926558158,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00047119159526100773,
      "loss": 5.0839,
      "mean_token_accuracy": 0.16821239292621612,
      "num_tokens": 37786186.0,
      "step": 17245
    },
    {
      "entropy": 5.517325782775879,
      "epoch": 1.7119888844779676,
      "grad_norm": 0.890625,
      "learning_rate": 0.00047117424654558146,
      "loss": 5.1005,
      "mean_token_accuracy": 0.1793801948428154,
      "num_tokens": 37796837.0,
      "step": 17250
    },
    {
      "entropy": 5.575711917877197,
      "epoch": 1.712485113140135,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004711568929654566,
      "loss": 5.1631,
      "mean_token_accuracy": 0.17081210911273956,
      "num_tokens": 37808813.0,
      "step": 17255
    },
    {
      "entropy": 5.549413204193115,
      "epoch": 1.7129813418023025,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004711395345210635,
      "loss": 5.026,
      "mean_token_accuracy": 0.17955968677997589,
      "num_tokens": 37819770.0,
      "step": 17260
    },
    {
      "entropy": 5.550781917572022,
      "epoch": 1.71347757046447,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004711221712128325,
      "loss": 5.1137,
      "mean_token_accuracy": 0.17487804740667343,
      "num_tokens": 37831831.0,
      "step": 17265
    },
    {
      "entropy": 5.555217266082764,
      "epoch": 1.7139737991266375,
      "grad_norm": 1.0,
      "learning_rate": 0.0004711048030411943,
      "loss": 5.1576,
      "mean_token_accuracy": 0.17047970294952391,
      "num_tokens": 37843092.0,
      "step": 17270
    },
    {
      "entropy": 5.58795166015625,
      "epoch": 1.714470027788805,
      "grad_norm": 0.921875,
      "learning_rate": 0.00047108743000657963,
      "loss": 5.1244,
      "mean_token_accuracy": 0.1759021446108818,
      "num_tokens": 37853192.0,
      "step": 17275
    },
    {
      "entropy": 5.647727727890015,
      "epoch": 1.7149662564509724,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004710700521094192,
      "loss": 5.1529,
      "mean_token_accuracy": 0.17165095061063768,
      "num_tokens": 37863510.0,
      "step": 17280
    },
    {
      "entropy": 5.532742357254028,
      "epoch": 1.7154624851131401,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00047105266935014385,
      "loss": 5.0631,
      "mean_token_accuracy": 0.18225870579481124,
      "num_tokens": 37876156.0,
      "step": 17285
    },
    {
      "entropy": 5.518691778182983,
      "epoch": 1.7159587137753076,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047103528172918487,
      "loss": 5.1169,
      "mean_token_accuracy": 0.17749424576759337,
      "num_tokens": 37887153.0,
      "step": 17290
    },
    {
      "entropy": 5.5618085861206055,
      "epoch": 1.7164549424374753,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004710178892469733,
      "loss": 5.0978,
      "mean_token_accuracy": 0.1794215679168701,
      "num_tokens": 37898820.0,
      "step": 17295
    },
    {
      "entropy": 5.578125381469727,
      "epoch": 1.7169511710996428,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00047100049190394044,
      "loss": 5.0888,
      "mean_token_accuracy": 0.17426283210515975,
      "num_tokens": 37910257.0,
      "step": 17300
    },
    {
      "entropy": 5.544584131240844,
      "epoch": 1.7174473997618103,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004709830897005178,
      "loss": 5.1771,
      "mean_token_accuracy": 0.17154236435890197,
      "num_tokens": 37921574.0,
      "step": 17305
    },
    {
      "entropy": 5.616960573196411,
      "epoch": 1.7179436284239777,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004709656826371367,
      "loss": 5.1606,
      "mean_token_accuracy": 0.17311425507068634,
      "num_tokens": 37932741.0,
      "step": 17310
    },
    {
      "entropy": 5.556202793121338,
      "epoch": 1.7184398570861452,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004709482707142291,
      "loss": 5.1318,
      "mean_token_accuracy": 0.17382349967956542,
      "num_tokens": 37944592.0,
      "step": 17315
    },
    {
      "entropy": 5.557074880599975,
      "epoch": 1.7189360857483127,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047093085393222646,
      "loss": 5.0809,
      "mean_token_accuracy": 0.17793885618448257,
      "num_tokens": 37954983.0,
      "step": 17320
    },
    {
      "entropy": 5.562156009674072,
      "epoch": 1.7194323144104804,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047091343229156094,
      "loss": 5.1546,
      "mean_token_accuracy": 0.17035369724035263,
      "num_tokens": 37964491.0,
      "step": 17325
    },
    {
      "entropy": 5.5098920345306395,
      "epoch": 1.7199285430726479,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004708960057926644,
      "loss": 5.0451,
      "mean_token_accuracy": 0.1774196907877922,
      "num_tokens": 37976140.0,
      "step": 17330
    },
    {
      "entropy": 5.549550008773804,
      "epoch": 1.7204247717348156,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047087857443596905,
      "loss": 5.0867,
      "mean_token_accuracy": 0.1775013253092766,
      "num_tokens": 37987776.0,
      "step": 17335
    },
    {
      "entropy": 5.560797166824341,
      "epoch": 1.720921000396983,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047086113822190716,
      "loss": 5.1484,
      "mean_token_accuracy": 0.16919077336788177,
      "num_tokens": 37998088.0,
      "step": 17340
    },
    {
      "entropy": 5.53006944656372,
      "epoch": 1.7214172290591505,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00047084369715091103,
      "loss": 5.1337,
      "mean_token_accuracy": 0.16951314210891724,
      "num_tokens": 38009802.0,
      "step": 17345
    },
    {
      "entropy": 5.5932230949401855,
      "epoch": 1.721913457721318,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004708262512234133,
      "loss": 5.1716,
      "mean_token_accuracy": 0.17263431549072267,
      "num_tokens": 38020797.0,
      "step": 17350
    },
    {
      "entropy": 5.6425275802612305,
      "epoch": 1.7224096863834855,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004708088004398464,
      "loss": 5.1624,
      "mean_token_accuracy": 0.17210831940174104,
      "num_tokens": 38031816.0,
      "step": 17355
    },
    {
      "entropy": 5.545508813858032,
      "epoch": 1.722905915045653,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004707913448006433,
      "loss": 5.1609,
      "mean_token_accuracy": 0.16241195946931838,
      "num_tokens": 38043953.0,
      "step": 17360
    },
    {
      "entropy": 5.514689874649048,
      "epoch": 1.7234021437078204,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004707738843062367,
      "loss": 5.0318,
      "mean_token_accuracy": 0.1831573724746704,
      "num_tokens": 38053929.0,
      "step": 17365
    },
    {
      "entropy": 5.626897382736206,
      "epoch": 1.723898372369988,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004707564189570596,
      "loss": 5.2587,
      "mean_token_accuracy": 0.17311705201864241,
      "num_tokens": 38067091.0,
      "step": 17370
    },
    {
      "entropy": 5.591850090026855,
      "epoch": 1.7243946010321556,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004707389487535451,
      "loss": 5.0969,
      "mean_token_accuracy": 0.17243184596300126,
      "num_tokens": 38079002.0,
      "step": 17375
    },
    {
      "entropy": 5.5826293468475345,
      "epoch": 1.7248908296943233,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004707214736961265,
      "loss": 5.1032,
      "mean_token_accuracy": 0.1832143947482109,
      "num_tokens": 38089526.0,
      "step": 17380
    },
    {
      "entropy": 5.559901714324951,
      "epoch": 1.7253870583564908,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004707039937852371,
      "loss": 5.1073,
      "mean_token_accuracy": 0.1793068215250969,
      "num_tokens": 38099902.0,
      "step": 17385
    },
    {
      "entropy": 5.523968744277954,
      "epoch": 1.7258832870186582,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00047068650902131034,
      "loss": 5.0738,
      "mean_token_accuracy": 0.17942894697189332,
      "num_tokens": 38111725.0,
      "step": 17390
    },
    {
      "entropy": 5.558438062667847,
      "epoch": 1.7263795156808257,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004706690194047798,
      "loss": 5.0404,
      "mean_token_accuracy": 0.17444614320993423,
      "num_tokens": 38123522.0,
      "step": 17395
    },
    {
      "entropy": 5.573927783966065,
      "epoch": 1.7268757443429932,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004706515249360792,
      "loss": 5.1705,
      "mean_token_accuracy": 0.17031463235616684,
      "num_tokens": 38133968.0,
      "step": 17400
    },
    {
      "entropy": 5.529282188415527,
      "epoch": 1.7273719730051607,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004706340256156425,
      "loss": 5.221,
      "mean_token_accuracy": 0.16289281845092773,
      "num_tokens": 38145288.0,
      "step": 17405
    },
    {
      "entropy": 5.58859338760376,
      "epoch": 1.7278682016673284,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004706165214439034,
      "loss": 5.1151,
      "mean_token_accuracy": 0.17147918045520782,
      "num_tokens": 38156141.0,
      "step": 17410
    },
    {
      "entropy": 5.618618202209473,
      "epoch": 1.7283644303294958,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004705990124212961,
      "loss": 5.0777,
      "mean_token_accuracy": 0.18603018969297408,
      "num_tokens": 38167211.0,
      "step": 17415
    },
    {
      "entropy": 5.545669031143189,
      "epoch": 1.7288606589916635,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004705814985482548,
      "loss": 5.1414,
      "mean_token_accuracy": 0.17378313094377518,
      "num_tokens": 38177360.0,
      "step": 17420
    },
    {
      "entropy": 5.508595895767212,
      "epoch": 1.729356887653831,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047056397982521366,
      "loss": 5.0437,
      "mean_token_accuracy": 0.17996789664030075,
      "num_tokens": 38187857.0,
      "step": 17425
    },
    {
      "entropy": 5.590577268600464,
      "epoch": 1.7298531163159985,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004705464562526074,
      "loss": 5.1715,
      "mean_token_accuracy": 0.16224278807640075,
      "num_tokens": 38198161.0,
      "step": 17430
    },
    {
      "entropy": 5.585971641540527,
      "epoch": 1.730349344978166,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004705289278308703,
      "loss": 5.1016,
      "mean_token_accuracy": 0.17599351853132247,
      "num_tokens": 38208612.0,
      "step": 17435
    },
    {
      "entropy": 5.5811385154724125,
      "epoch": 1.7308455736403334,
      "grad_norm": 0.90625,
      "learning_rate": 0.00047051139456043707,
      "loss": 5.135,
      "mean_token_accuracy": 0.17004906982183457,
      "num_tokens": 38220153.0,
      "step": 17440
    },
    {
      "entropy": 5.555471706390381,
      "epoch": 1.731341802302501,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004704938564417427,
      "loss": 5.1011,
      "mean_token_accuracy": 0.17772171795368194,
      "num_tokens": 38230790.0,
      "step": 17445
    },
    {
      "entropy": 5.55522928237915,
      "epoch": 1.7318380309646684,
      "grad_norm": 0.9375,
      "learning_rate": 0.00047047631347522175,
      "loss": 5.049,
      "mean_token_accuracy": 0.18061629682779312,
      "num_tokens": 38242585.0,
      "step": 17450
    },
    {
      "entropy": 5.5654490947723385,
      "epoch": 1.732334259626836,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00047045876566130957,
      "loss": 5.1042,
      "mean_token_accuracy": 0.1754191666841507,
      "num_tokens": 38253316.0,
      "step": 17455
    },
    {
      "entropy": 5.488184547424316,
      "epoch": 1.7328304882890035,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004704412130004411,
      "loss": 5.0802,
      "mean_token_accuracy": 0.17691100835800172,
      "num_tokens": 38263230.0,
      "step": 17460
    },
    {
      "entropy": 5.581308841705322,
      "epoch": 1.7333267169511712,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047042365549305174,
      "loss": 5.1468,
      "mean_token_accuracy": 0.17157044857740403,
      "num_tokens": 38273416.0,
      "step": 17465
    },
    {
      "entropy": 5.53314242362976,
      "epoch": 1.7338229456133387,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004704060931395768,
      "loss": 5.0769,
      "mean_token_accuracy": 0.18006590902805328,
      "num_tokens": 38285764.0,
      "step": 17470
    },
    {
      "entropy": 5.591202259063721,
      "epoch": 1.7343191742755062,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047038852594045186,
      "loss": 5.1833,
      "mean_token_accuracy": 0.17256945371627808,
      "num_tokens": 38296899.0,
      "step": 17475
    },
    {
      "entropy": 5.616226673126221,
      "epoch": 1.7348154029376737,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047037095389611245,
      "loss": 5.1743,
      "mean_token_accuracy": 0.17670390009880066,
      "num_tokens": 38309050.0,
      "step": 17480
    },
    {
      "entropy": 5.5264490127563475,
      "epoch": 1.7353116315998411,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004703533770069944,
      "loss": 5.1091,
      "mean_token_accuracy": 0.1777672663331032,
      "num_tokens": 38320831.0,
      "step": 17485
    },
    {
      "entropy": 5.567055511474609,
      "epoch": 1.7358078602620086,
      "grad_norm": 0.890625,
      "learning_rate": 0.00047033579527353355,
      "loss": 5.135,
      "mean_token_accuracy": 0.17228993624448777,
      "num_tokens": 38332096.0,
      "step": 17490
    },
    {
      "entropy": 5.595486259460449,
      "epoch": 1.7363040889241763,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004703182086961659,
      "loss": 5.1546,
      "mean_token_accuracy": 0.1682416930794716,
      "num_tokens": 38343110.0,
      "step": 17495
    },
    {
      "entropy": 5.6186693668365475,
      "epoch": 1.7368003175863438,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004703006172753274,
      "loss": 5.1382,
      "mean_token_accuracy": 0.17911583930253983,
      "num_tokens": 38353971.0,
      "step": 17500
    },
    {
      "entropy": 5.522997951507568,
      "epoch": 1.7372965462485113,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047028302101145467,
      "loss": 5.0693,
      "mean_token_accuracy": 0.178188294172287,
      "num_tokens": 38365114.0,
      "step": 17505
    },
    {
      "entropy": 5.573307800292969,
      "epoch": 1.737792774910679,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004702654199049836,
      "loss": 5.0953,
      "mean_token_accuracy": 0.17266631126403809,
      "num_tokens": 38375784.0,
      "step": 17510
    },
    {
      "entropy": 5.538006830215454,
      "epoch": 1.7382890035728464,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000470247813956351,
      "loss": 5.0821,
      "mean_token_accuracy": 0.17605000883340835,
      "num_tokens": 38386053.0,
      "step": 17515
    },
    {
      "entropy": 5.484797859191895,
      "epoch": 1.738785232235014,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047023020316599334,
      "loss": 5.0833,
      "mean_token_accuracy": 0.17451546639204024,
      "num_tokens": 38396545.0,
      "step": 17520
    },
    {
      "entropy": 5.551498031616211,
      "epoch": 1.7392814608971814,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00047021258753434725,
      "loss": 5.0629,
      "mean_token_accuracy": 0.17322328239679335,
      "num_tokens": 38406344.0,
      "step": 17525
    },
    {
      "entropy": 5.5825286388397215,
      "epoch": 1.7397776895593489,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047019496706184966,
      "loss": 5.0563,
      "mean_token_accuracy": 0.17969710230827332,
      "num_tokens": 38416388.0,
      "step": 17530
    },
    {
      "entropy": 5.5995628356933596,
      "epoch": 1.7402739182215163,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004701773417489375,
      "loss": 5.1858,
      "mean_token_accuracy": 0.16599820107221602,
      "num_tokens": 38427608.0,
      "step": 17535
    },
    {
      "entropy": 5.542110347747803,
      "epoch": 1.740770146883684,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00047015971159604787,
      "loss": 5.1031,
      "mean_token_accuracy": 0.17624692171812056,
      "num_tokens": 38437921.0,
      "step": 17540
    },
    {
      "entropy": 5.555644750595093,
      "epoch": 1.7412663755458515,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000470142076603618,
      "loss": 5.1605,
      "mean_token_accuracy": 0.16873769760131835,
      "num_tokens": 38448767.0,
      "step": 17545
    },
    {
      "entropy": 5.589807844161987,
      "epoch": 1.7417626042080192,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00047012443677208496,
      "loss": 5.2283,
      "mean_token_accuracy": 0.16700267493724824,
      "num_tokens": 38460052.0,
      "step": 17550
    },
    {
      "entropy": 5.576025199890137,
      "epoch": 1.7422588328701867,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004701067921018864,
      "loss": 5.0786,
      "mean_token_accuracy": 0.17620374113321305,
      "num_tokens": 38470177.0,
      "step": 17555
    },
    {
      "entropy": 5.590360116958618,
      "epoch": 1.7427550615323542,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047008914259345986,
      "loss": 5.0911,
      "mean_token_accuracy": 0.1702023595571518,
      "num_tokens": 38481608.0,
      "step": 17560
    },
    {
      "entropy": 5.55975661277771,
      "epoch": 1.7432512901945216,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00047007148824724295,
      "loss": 5.1581,
      "mean_token_accuracy": 0.16911829859018326,
      "num_tokens": 38492534.0,
      "step": 17565
    },
    {
      "entropy": 5.557606792449951,
      "epoch": 1.743747518856689,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004700538290636734,
      "loss": 5.1581,
      "mean_token_accuracy": 0.17308571338653564,
      "num_tokens": 38503629.0,
      "step": 17570
    },
    {
      "entropy": 5.496676015853882,
      "epoch": 1.7442437475188566,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047003616504318925,
      "loss": 5.0418,
      "mean_token_accuracy": 0.17432612031698227,
      "num_tokens": 38514428.0,
      "step": 17575
    },
    {
      "entropy": 5.581231307983399,
      "epoch": 1.744739976181024,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047001849618622844,
      "loss": 5.1344,
      "mean_token_accuracy": 0.1703056275844574,
      "num_tokens": 38525468.0,
      "step": 17580
    },
    {
      "entropy": 5.517770147323608,
      "epoch": 1.7452362048431918,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047000082249322916,
      "loss": 5.1007,
      "mean_token_accuracy": 0.17329324334859847,
      "num_tokens": 38535095.0,
      "step": 17585
    },
    {
      "entropy": 5.579011344909668,
      "epoch": 1.7457324335053592,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004699831439646297,
      "loss": 5.0906,
      "mean_token_accuracy": 0.17359646409749985,
      "num_tokens": 38546867.0,
      "step": 17590
    },
    {
      "entropy": 5.495063209533692,
      "epoch": 1.746228662167527,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004699654606008684,
      "loss": 5.0017,
      "mean_token_accuracy": 0.1814451649785042,
      "num_tokens": 38557231.0,
      "step": 17595
    },
    {
      "entropy": 5.537030839920044,
      "epoch": 1.7467248908296944,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004699477724023838,
      "loss": 5.1224,
      "mean_token_accuracy": 0.17315296828746796,
      "num_tokens": 38568857.0,
      "step": 17600
    },
    {
      "entropy": 5.599961471557617,
      "epoch": 1.7472211194918619,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004699300793696145,
      "loss": 5.0976,
      "mean_token_accuracy": 0.17201314270496368,
      "num_tokens": 38579177.0,
      "step": 17605
    },
    {
      "entropy": 5.570987796783447,
      "epoch": 1.7477173481540293,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004699123815029993,
      "loss": 5.0079,
      "mean_token_accuracy": 0.1800644278526306,
      "num_tokens": 38588683.0,
      "step": 17610
    },
    {
      "entropy": 5.527664422988892,
      "epoch": 1.7482135768161968,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046989467880297704,
      "loss": 5.0854,
      "mean_token_accuracy": 0.1740689605474472,
      "num_tokens": 38599636.0,
      "step": 17615
    },
    {
      "entropy": 5.626832914352417,
      "epoch": 1.7487098054783643,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046987697126998673,
      "loss": 5.1991,
      "mean_token_accuracy": 0.16754450649023056,
      "num_tokens": 38609360.0,
      "step": 17620
    },
    {
      "entropy": 5.58344874382019,
      "epoch": 1.749206034140532,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004698592589044674,
      "loss": 5.1239,
      "mean_token_accuracy": 0.17207718640565872,
      "num_tokens": 38620996.0,
      "step": 17625
    },
    {
      "entropy": 5.572943210601807,
      "epoch": 1.7497022628026995,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046984154170685833,
      "loss": 5.2215,
      "mean_token_accuracy": 0.1728046104311943,
      "num_tokens": 38632935.0,
      "step": 17630
    },
    {
      "entropy": 5.658014822006225,
      "epoch": 1.7501984914648672,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004698238196775989,
      "loss": 5.211,
      "mean_token_accuracy": 0.16946618705987931,
      "num_tokens": 38644299.0,
      "step": 17635
    },
    {
      "entropy": 5.589119291305542,
      "epoch": 1.7506947201270346,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00046980609281712857,
      "loss": 5.1507,
      "mean_token_accuracy": 0.1749320074915886,
      "num_tokens": 38654970.0,
      "step": 17640
    },
    {
      "entropy": 5.580189180374146,
      "epoch": 1.7511909487892021,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004697883611258868,
      "loss": 5.1893,
      "mean_token_accuracy": 0.1741221010684967,
      "num_tokens": 38666302.0,
      "step": 17645
    },
    {
      "entropy": 5.562500905990601,
      "epoch": 1.7516871774513696,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004697706246043136,
      "loss": 5.1081,
      "mean_token_accuracy": 0.17513372302055358,
      "num_tokens": 38678346.0,
      "step": 17650
    },
    {
      "entropy": 5.5413085460662845,
      "epoch": 1.752183406113537,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046975288325284845,
      "loss": 5.0493,
      "mean_token_accuracy": 0.17577243149280547,
      "num_tokens": 38688242.0,
      "step": 17655
    },
    {
      "entropy": 5.630189752578735,
      "epoch": 1.7526796347757045,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004697351370719315,
      "loss": 5.2204,
      "mean_token_accuracy": 0.16486080139875411,
      "num_tokens": 38698756.0,
      "step": 17660
    },
    {
      "entropy": 5.596133756637573,
      "epoch": 1.753175863437872,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004697173860620027,
      "loss": 5.19,
      "mean_token_accuracy": 0.1659785345196724,
      "num_tokens": 38710410.0,
      "step": 17665
    },
    {
      "entropy": 5.63321442604065,
      "epoch": 1.7536720921000397,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004696996302235024,
      "loss": 5.136,
      "mean_token_accuracy": 0.16877845376729966,
      "num_tokens": 38721917.0,
      "step": 17670
    },
    {
      "entropy": 5.595292472839356,
      "epoch": 1.7541683207622072,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00046968186955687074,
      "loss": 5.1582,
      "mean_token_accuracy": 0.17791620194911956,
      "num_tokens": 38733085.0,
      "step": 17675
    },
    {
      "entropy": 5.528179788589478,
      "epoch": 1.7546645494243749,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004696641040625483,
      "loss": 5.18,
      "mean_token_accuracy": 0.16956325024366378,
      "num_tokens": 38744785.0,
      "step": 17680
    },
    {
      "entropy": 5.569801473617554,
      "epoch": 1.7551607780865424,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00046964633374097544,
      "loss": 5.1533,
      "mean_token_accuracy": 0.1736989811062813,
      "num_tokens": 38755626.0,
      "step": 17685
    },
    {
      "entropy": 5.615735197067261,
      "epoch": 1.7556570067487098,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046962855859259306,
      "loss": 5.0364,
      "mean_token_accuracy": 0.1725809618830681,
      "num_tokens": 38766360.0,
      "step": 17690
    },
    {
      "entropy": 5.581758260726929,
      "epoch": 1.7561532354108773,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046961077861784167,
      "loss": 5.1084,
      "mean_token_accuracy": 0.1726755529642105,
      "num_tokens": 38776763.0,
      "step": 17695
    },
    {
      "entropy": 5.521144676208496,
      "epoch": 1.7566494640730448,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004695929938171624,
      "loss": 5.1758,
      "mean_token_accuracy": 0.17600510865449906,
      "num_tokens": 38787664.0,
      "step": 17700
    },
    {
      "entropy": 5.577087640762329,
      "epoch": 1.7571456927352123,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004695752041909962,
      "loss": 5.1017,
      "mean_token_accuracy": 0.1711129814386368,
      "num_tokens": 38798018.0,
      "step": 17705
    },
    {
      "entropy": 5.573191833496094,
      "epoch": 1.75764192139738,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004695574097397842,
      "loss": 5.0615,
      "mean_token_accuracy": 0.17492159307003022,
      "num_tokens": 38810381.0,
      "step": 17710
    },
    {
      "entropy": 5.499499940872193,
      "epoch": 1.7581381500595474,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046953961046396767,
      "loss": 5.0662,
      "mean_token_accuracy": 0.18041797429323198,
      "num_tokens": 38821042.0,
      "step": 17715
    },
    {
      "entropy": 5.508300542831421,
      "epoch": 1.7586343787217151,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046952180636398793,
      "loss": 5.0218,
      "mean_token_accuracy": 0.17752404361963273,
      "num_tokens": 38830890.0,
      "step": 17720
    },
    {
      "entropy": 5.5899986743927,
      "epoch": 1.7591306073838826,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004695039974402866,
      "loss": 5.066,
      "mean_token_accuracy": 0.1826263725757599,
      "num_tokens": 38842648.0,
      "step": 17725
    },
    {
      "entropy": 5.554860401153564,
      "epoch": 1.75962683604605,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00046948618369330533,
      "loss": 5.0192,
      "mean_token_accuracy": 0.1832275778055191,
      "num_tokens": 38852740.0,
      "step": 17730
    },
    {
      "entropy": 5.5347191333770756,
      "epoch": 1.7601230647082176,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046946836512348566,
      "loss": 5.144,
      "mean_token_accuracy": 0.1655426040291786,
      "num_tokens": 38863371.0,
      "step": 17735
    },
    {
      "entropy": 5.598482465744018,
      "epoch": 1.760619293370385,
      "grad_norm": 1.0,
      "learning_rate": 0.00046945054173126965,
      "loss": 5.1268,
      "mean_token_accuracy": 0.1708410531282425,
      "num_tokens": 38873893.0,
      "step": 17740
    },
    {
      "entropy": 5.578781270980835,
      "epoch": 1.7611155220325525,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046943271351709923,
      "loss": 5.1665,
      "mean_token_accuracy": 0.1677718788385391,
      "num_tokens": 38884778.0,
      "step": 17745
    },
    {
      "entropy": 5.510950374603271,
      "epoch": 1.76161175069472,
      "grad_norm": 1.0,
      "learning_rate": 0.0004694148804814164,
      "loss": 5.1125,
      "mean_token_accuracy": 0.16857096701860427,
      "num_tokens": 38896011.0,
      "step": 17750
    },
    {
      "entropy": 5.59821457862854,
      "epoch": 1.7621079793568877,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004693970426246635,
      "loss": 5.1468,
      "mean_token_accuracy": 0.16961099356412887,
      "num_tokens": 38907686.0,
      "step": 17755
    },
    {
      "entropy": 5.5256627082824705,
      "epoch": 1.7626042080190552,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00046937919994728287,
      "loss": 5.1085,
      "mean_token_accuracy": 0.1772584065794945,
      "num_tokens": 38919180.0,
      "step": 17760
    },
    {
      "entropy": 5.549013710021972,
      "epoch": 1.7631004366812228,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004693613524497168,
      "loss": 5.09,
      "mean_token_accuracy": 0.17301513105630875,
      "num_tokens": 38929600.0,
      "step": 17765
    },
    {
      "entropy": 5.5715728282928465,
      "epoch": 1.7635966653433903,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046934350013240814,
      "loss": 5.0393,
      "mean_token_accuracy": 0.18329585194587708,
      "num_tokens": 38940404.0,
      "step": 17770
    },
    {
      "entropy": 5.54627799987793,
      "epoch": 1.7640928940055578,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004693256429957994,
      "loss": 5.095,
      "mean_token_accuracy": 0.17642540484666824,
      "num_tokens": 38951929.0,
      "step": 17775
    },
    {
      "entropy": 5.516851282119751,
      "epoch": 1.7645891226677253,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004693077810403335,
      "loss": 5.0842,
      "mean_token_accuracy": 0.175152924656868,
      "num_tokens": 38961954.0,
      "step": 17780
    },
    {
      "entropy": 5.495822095870972,
      "epoch": 1.7650853513298927,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046928991426645325,
      "loss": 5.0658,
      "mean_token_accuracy": 0.17466057687997819,
      "num_tokens": 38970974.0,
      "step": 17785
    },
    {
      "entropy": 5.581322765350341,
      "epoch": 1.7655815799920602,
      "grad_norm": 0.953125,
      "learning_rate": 0.00046927204267460177,
      "loss": 5.1415,
      "mean_token_accuracy": 0.17215966135263444,
      "num_tokens": 38982481.0,
      "step": 17790
    },
    {
      "entropy": 5.634880399703979,
      "epoch": 1.766077808654228,
      "grad_norm": 1.0,
      "learning_rate": 0.00046925416626522235,
      "loss": 5.1025,
      "mean_token_accuracy": 0.17899129390716553,
      "num_tokens": 38993279.0,
      "step": 17795
    },
    {
      "entropy": 5.540242004394531,
      "epoch": 1.7665740373163954,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004692362850387581,
      "loss": 5.0784,
      "mean_token_accuracy": 0.18178649246692657,
      "num_tokens": 39003748.0,
      "step": 17800
    },
    {
      "entropy": 5.497012138366699,
      "epoch": 1.7670702659785629,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046921839899565263,
      "loss": 5.1051,
      "mean_token_accuracy": 0.16572619527578353,
      "num_tokens": 39014902.0,
      "step": 17805
    },
    {
      "entropy": 5.602819156646729,
      "epoch": 1.7675664946407306,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004692005081363493,
      "loss": 5.1662,
      "mean_token_accuracy": 0.1694572687149048,
      "num_tokens": 39026484.0,
      "step": 17810
    },
    {
      "entropy": 5.564551830291748,
      "epoch": 1.768062723302898,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00046918261246129176,
      "loss": 5.1538,
      "mean_token_accuracy": 0.17185938358306885,
      "num_tokens": 39036268.0,
      "step": 17815
    },
    {
      "entropy": 5.598098278045654,
      "epoch": 1.7685589519650655,
      "grad_norm": 0.96875,
      "learning_rate": 0.000469164711970924,
      "loss": 5.2003,
      "mean_token_accuracy": 0.17065991163253785,
      "num_tokens": 39049092.0,
      "step": 17820
    },
    {
      "entropy": 5.625868272781372,
      "epoch": 1.769055180627233,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004691468066656896,
      "loss": 5.1554,
      "mean_token_accuracy": 0.17015611827373506,
      "num_tokens": 39059255.0,
      "step": 17825
    },
    {
      "entropy": 5.516210508346558,
      "epoch": 1.7695514092894005,
      "grad_norm": 0.90625,
      "learning_rate": 0.00046912889654603286,
      "loss": 5.053,
      "mean_token_accuracy": 0.1774064302444458,
      "num_tokens": 39069913.0,
      "step": 17830
    },
    {
      "entropy": 5.541619348526001,
      "epoch": 1.770047637951568,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004691109816123977,
      "loss": 5.0727,
      "mean_token_accuracy": 0.17889440953731536,
      "num_tokens": 39080666.0,
      "step": 17835
    },
    {
      "entropy": 5.591982507705689,
      "epoch": 1.7705438666137356,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004690930618652286,
      "loss": 5.1424,
      "mean_token_accuracy": 0.17367461025714875,
      "num_tokens": 39091562.0,
      "step": 17840
    },
    {
      "entropy": 5.603212833404541,
      "epoch": 1.7710400952759031,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046907513730496975,
      "loss": 5.2189,
      "mean_token_accuracy": 0.16420638412237168,
      "num_tokens": 39102567.0,
      "step": 17845
    },
    {
      "entropy": 5.631757402420044,
      "epoch": 1.7715363239380708,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004690572079320657,
      "loss": 5.105,
      "mean_token_accuracy": 0.17596133500337602,
      "num_tokens": 39113260.0,
      "step": 17850
    },
    {
      "entropy": 5.5827446460723875,
      "epoch": 1.7720325526002383,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00046903927374696106,
      "loss": 5.1448,
      "mean_token_accuracy": 0.1742891326546669,
      "num_tokens": 39125790.0,
      "step": 17855
    },
    {
      "entropy": 5.5558558940887455,
      "epoch": 1.7725287812624058,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004690213347501005,
      "loss": 5.1849,
      "mean_token_accuracy": 0.1703210711479187,
      "num_tokens": 39137220.0,
      "step": 17860
    },
    {
      "entropy": 5.560394239425659,
      "epoch": 1.7730250099245732,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004690033909419289,
      "loss": 5.0677,
      "mean_token_accuracy": 0.17490962743759156,
      "num_tokens": 39146362.0,
      "step": 17865
    },
    {
      "entropy": 5.616075277328491,
      "epoch": 1.7735212385867407,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046898544232289126,
      "loss": 5.1466,
      "mean_token_accuracy": 0.16869162917137145,
      "num_tokens": 39156822.0,
      "step": 17870
    },
    {
      "entropy": 5.626028823852539,
      "epoch": 1.7740174672489082,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00046896748889343267,
      "loss": 5.1299,
      "mean_token_accuracy": 0.16527492702007293,
      "num_tokens": 39166693.0,
      "step": 17875
    },
    {
      "entropy": 5.496959972381592,
      "epoch": 1.7745136959110757,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046894953065399836,
      "loss": 5.0989,
      "mean_token_accuracy": 0.17684730589389802,
      "num_tokens": 39176889.0,
      "step": 17880
    },
    {
      "entropy": 5.560200452804565,
      "epoch": 1.7750099245732434,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046893156760503357,
      "loss": 5.099,
      "mean_token_accuracy": 0.18187216073274612,
      "num_tokens": 39188077.0,
      "step": 17885
    },
    {
      "entropy": 5.536591482162476,
      "epoch": 1.7755061532354108,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004689135997469838,
      "loss": 5.0894,
      "mean_token_accuracy": 0.16988267600536347,
      "num_tokens": 39198182.0,
      "step": 17890
    },
    {
      "entropy": 5.625663089752197,
      "epoch": 1.7760023818975785,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004688956270802945,
      "loss": 5.0667,
      "mean_token_accuracy": 0.1798384740948677,
      "num_tokens": 39208801.0,
      "step": 17895
    },
    {
      "entropy": 5.457943677902222,
      "epoch": 1.776498610559746,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004688776496054117,
      "loss": 4.9411,
      "mean_token_accuracy": 0.18785025030374528,
      "num_tokens": 39220658.0,
      "step": 17900
    },
    {
      "entropy": 5.540046167373657,
      "epoch": 1.7769948392219135,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046885966732278077,
      "loss": 5.1807,
      "mean_token_accuracy": 0.16862889379262924,
      "num_tokens": 39231987.0,
      "step": 17905
    },
    {
      "entropy": 5.576083469390869,
      "epoch": 1.777491067884081,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046884168023284787,
      "loss": 5.066,
      "mean_token_accuracy": 0.1771724745631218,
      "num_tokens": 39242797.0,
      "step": 17910
    },
    {
      "entropy": 5.576232814788819,
      "epoch": 1.7779872965462484,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046882368833605894,
      "loss": 5.1202,
      "mean_token_accuracy": 0.17469854950904845,
      "num_tokens": 39253621.0,
      "step": 17915
    },
    {
      "entropy": 5.52582483291626,
      "epoch": 1.778483525208416,
      "grad_norm": 1.0,
      "learning_rate": 0.00046880569163286023,
      "loss": 5.1398,
      "mean_token_accuracy": 0.16921603828668594,
      "num_tokens": 39264126.0,
      "step": 17920
    },
    {
      "entropy": 5.563271617889404,
      "epoch": 1.7789797538705836,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046878769012369807,
      "loss": 5.0743,
      "mean_token_accuracy": 0.180909962952137,
      "num_tokens": 39275118.0,
      "step": 17925
    },
    {
      "entropy": 5.562721061706543,
      "epoch": 1.779475982532751,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046876968380901864,
      "loss": 5.0942,
      "mean_token_accuracy": 0.17614645957946778,
      "num_tokens": 39287237.0,
      "step": 17930
    },
    {
      "entropy": 5.624432182312011,
      "epoch": 1.7799722111949188,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046875167268926865,
      "loss": 5.1524,
      "mean_token_accuracy": 0.17139471918344498,
      "num_tokens": 39296995.0,
      "step": 17935
    },
    {
      "entropy": 5.53447003364563,
      "epoch": 1.7804684398570862,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004687336567648947,
      "loss": 5.0538,
      "mean_token_accuracy": 0.17358405739068986,
      "num_tokens": 39305980.0,
      "step": 17940
    },
    {
      "entropy": 5.584849643707275,
      "epoch": 1.7809646685192537,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046871563603634347,
      "loss": 5.1741,
      "mean_token_accuracy": 0.16430944502353667,
      "num_tokens": 39316494.0,
      "step": 17945
    },
    {
      "entropy": 5.581762599945068,
      "epoch": 1.7814608971814212,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004686976105040619,
      "loss": 5.147,
      "mean_token_accuracy": 0.17480543851852418,
      "num_tokens": 39327699.0,
      "step": 17950
    },
    {
      "entropy": 5.584734869003296,
      "epoch": 1.7819571258435887,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004686795801684969,
      "loss": 5.1629,
      "mean_token_accuracy": 0.16990833282470702,
      "num_tokens": 39338749.0,
      "step": 17955
    },
    {
      "entropy": 5.569218683242798,
      "epoch": 1.7824533545057561,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004686615450300958,
      "loss": 5.1362,
      "mean_token_accuracy": 0.17690266221761702,
      "num_tokens": 39349747.0,
      "step": 17960
    },
    {
      "entropy": 5.508942604064941,
      "epoch": 1.7829495831679236,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004686435050893056,
      "loss": 4.9987,
      "mean_token_accuracy": 0.18178412765264512,
      "num_tokens": 39360068.0,
      "step": 17965
    },
    {
      "entropy": 5.502421951293945,
      "epoch": 1.7834458118300913,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046862546034657376,
      "loss": 5.0282,
      "mean_token_accuracy": 0.1788963109254837,
      "num_tokens": 39369651.0,
      "step": 17970
    },
    {
      "entropy": 5.586280727386475,
      "epoch": 1.7839420404922588,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004686074108023477,
      "loss": 5.2313,
      "mean_token_accuracy": 0.1694331258535385,
      "num_tokens": 39381354.0,
      "step": 17975
    },
    {
      "entropy": 5.585847520828247,
      "epoch": 1.7844382691544265,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046858935645707506,
      "loss": 5.083,
      "mean_token_accuracy": 0.17145538777112962,
      "num_tokens": 39391793.0,
      "step": 17980
    },
    {
      "entropy": 5.527411937713623,
      "epoch": 1.784934497816594,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004685712973112035,
      "loss": 5.0998,
      "mean_token_accuracy": 0.17774528712034227,
      "num_tokens": 39402194.0,
      "step": 17985
    },
    {
      "entropy": 5.532209873199463,
      "epoch": 1.7854307264787614,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004685532333651809,
      "loss": 5.1059,
      "mean_token_accuracy": 0.17310286313295364,
      "num_tokens": 39413193.0,
      "step": 17990
    },
    {
      "entropy": 5.549567317962646,
      "epoch": 1.785926955140929,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046853516461945523,
      "loss": 5.0774,
      "mean_token_accuracy": 0.18055932968854904,
      "num_tokens": 39423072.0,
      "step": 17995
    },
    {
      "entropy": 5.604977178573608,
      "epoch": 1.7864231838030964,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046851709107447446,
      "loss": 5.1968,
      "mean_token_accuracy": 0.1684126600623131,
      "num_tokens": 39435088.0,
      "step": 18000
    },
    {
      "epoch": 1.7864231838030964,
      "eval_entropy": 5.376621442555985,
      "eval_loss": 5.217236042022705,
      "eval_mean_token_accuracy": 0.17458677464106157,
      "eval_num_tokens": 39435088.0,
      "eval_runtime": 20.473,
      "eval_samples_per_second": 1644.557,
      "eval_steps_per_second": 205.588,
      "step": 18000
    },
    {
      "entropy": 5.516323137283325,
      "epoch": 1.7869194124652639,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004684990127306868,
      "loss": 5.0587,
      "mean_token_accuracy": 0.1767568454146385,
      "num_tokens": 39446129.0,
      "step": 18005
    },
    {
      "entropy": 5.520525789260864,
      "epoch": 1.7874156411274316,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046848092958854065,
      "loss": 5.011,
      "mean_token_accuracy": 0.189851376414299,
      "num_tokens": 39458127.0,
      "step": 18010
    },
    {
      "entropy": 5.557810926437378,
      "epoch": 1.787911869789599,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004684628416484843,
      "loss": 5.0693,
      "mean_token_accuracy": 0.17873586267232894,
      "num_tokens": 39468204.0,
      "step": 18015
    },
    {
      "entropy": 5.5416114807128904,
      "epoch": 1.7884080984517667,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004684447489109663,
      "loss": 5.1392,
      "mean_token_accuracy": 0.17283686697483064,
      "num_tokens": 39479114.0,
      "step": 18020
    },
    {
      "entropy": 5.547650241851807,
      "epoch": 1.7889043271139342,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046842665137643546,
      "loss": 5.1128,
      "mean_token_accuracy": 0.1774169459939003,
      "num_tokens": 39489691.0,
      "step": 18025
    },
    {
      "entropy": 5.612976789474487,
      "epoch": 1.7894005557761017,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046840854904534046,
      "loss": 5.2269,
      "mean_token_accuracy": 0.16386041939258575,
      "num_tokens": 39501879.0,
      "step": 18030
    },
    {
      "entropy": 5.608426809310913,
      "epoch": 1.7898967844382692,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004683904419181302,
      "loss": 5.1345,
      "mean_token_accuracy": 0.17699406296014786,
      "num_tokens": 39513413.0,
      "step": 18035
    },
    {
      "entropy": 5.554781579971314,
      "epoch": 1.7903930131004366,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046837232999525366,
      "loss": 5.0795,
      "mean_token_accuracy": 0.17090435475111007,
      "num_tokens": 39522719.0,
      "step": 18040
    },
    {
      "entropy": 5.556888055801392,
      "epoch": 1.7908892417626041,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046835421327716,
      "loss": 5.1405,
      "mean_token_accuracy": 0.1730068013072014,
      "num_tokens": 39532463.0,
      "step": 18045
    },
    {
      "entropy": 5.5412859439849855,
      "epoch": 1.7913854704247716,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004683360917642986,
      "loss": 5.0943,
      "mean_token_accuracy": 0.17171480357646943,
      "num_tokens": 39543005.0,
      "step": 18050
    },
    {
      "entropy": 5.601563215255737,
      "epoch": 1.7918816990869393,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046831796545711865,
      "loss": 5.164,
      "mean_token_accuracy": 0.16778888255357743,
      "num_tokens": 39553468.0,
      "step": 18055
    },
    {
      "entropy": 5.67397928237915,
      "epoch": 1.7923779277491068,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046829983435606973,
      "loss": 5.2231,
      "mean_token_accuracy": 0.17129427939653397,
      "num_tokens": 39563539.0,
      "step": 18060
    },
    {
      "entropy": 5.592680835723877,
      "epoch": 1.7928741564112745,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004682816984616014,
      "loss": 5.1576,
      "mean_token_accuracy": 0.17221810519695283,
      "num_tokens": 39574835.0,
      "step": 18065
    },
    {
      "entropy": 5.569009590148926,
      "epoch": 1.793370385073442,
      "grad_norm": 0.921875,
      "learning_rate": 0.00046826355777416346,
      "loss": 5.1046,
      "mean_token_accuracy": 0.17700168639421462,
      "num_tokens": 39586021.0,
      "step": 18070
    },
    {
      "entropy": 5.585173940658569,
      "epoch": 1.7938666137356094,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00046824541229420583,
      "loss": 5.044,
      "mean_token_accuracy": 0.17448382079601288,
      "num_tokens": 39596333.0,
      "step": 18075
    },
    {
      "entropy": 5.534476661682129,
      "epoch": 1.7943628423977769,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046822726202217825,
      "loss": 5.0682,
      "mean_token_accuracy": 0.17738160490989685,
      "num_tokens": 39607553.0,
      "step": 18080
    },
    {
      "entropy": 5.530224609375,
      "epoch": 1.7948590710599444,
      "grad_norm": 0.90625,
      "learning_rate": 0.000468209106958531,
      "loss": 5.1664,
      "mean_token_accuracy": 0.16989864856004716,
      "num_tokens": 39619700.0,
      "step": 18085
    },
    {
      "entropy": 5.542677688598633,
      "epoch": 1.7953552997221118,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004681909471037142,
      "loss": 5.0765,
      "mean_token_accuracy": 0.18002297282218932,
      "num_tokens": 39630623.0,
      "step": 18090
    },
    {
      "entropy": 5.554100227355957,
      "epoch": 1.7958515283842795,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046817278245817815,
      "loss": 5.0711,
      "mean_token_accuracy": 0.18067716807127,
      "num_tokens": 39641089.0,
      "step": 18095
    },
    {
      "entropy": 5.530361700057983,
      "epoch": 1.796347757046447,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004681546130223735,
      "loss": 5.1262,
      "mean_token_accuracy": 0.1715655505657196,
      "num_tokens": 39652381.0,
      "step": 18100
    },
    {
      "entropy": 5.532079219818115,
      "epoch": 1.7968439857086145,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004681364387967505,
      "loss": 5.0607,
      "mean_token_accuracy": 0.1748769223690033,
      "num_tokens": 39662536.0,
      "step": 18105
    },
    {
      "entropy": 5.543280553817749,
      "epoch": 1.7973402143707822,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004681182597817601,
      "loss": 5.1066,
      "mean_token_accuracy": 0.17961354702711105,
      "num_tokens": 39673140.0,
      "step": 18110
    },
    {
      "entropy": 5.520239210128784,
      "epoch": 1.7978364430329496,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004681000759778529,
      "loss": 5.0997,
      "mean_token_accuracy": 0.1677226424217224,
      "num_tokens": 39683843.0,
      "step": 18115
    },
    {
      "entropy": 5.520877456665039,
      "epoch": 1.7983326716951171,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004680818873854799,
      "loss": 5.0518,
      "mean_token_accuracy": 0.17436048984527588,
      "num_tokens": 39694178.0,
      "step": 18120
    },
    {
      "entropy": 5.472235441207886,
      "epoch": 1.7988289003572846,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004680636940050922,
      "loss": 5.0536,
      "mean_token_accuracy": 0.17695568650960922,
      "num_tokens": 39704818.0,
      "step": 18125
    },
    {
      "entropy": 5.556345510482788,
      "epoch": 1.799325129019452,
      "grad_norm": 0.984375,
      "learning_rate": 0.000468045495837141,
      "loss": 5.0663,
      "mean_token_accuracy": 0.1716505452990532,
      "num_tokens": 39715599.0,
      "step": 18130
    },
    {
      "entropy": 5.579806804656982,
      "epoch": 1.7998213576816195,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046802729288207735,
      "loss": 5.0814,
      "mean_token_accuracy": 0.17834774702787398,
      "num_tokens": 39725736.0,
      "step": 18135
    },
    {
      "entropy": 5.5522219181060795,
      "epoch": 1.8003175863437872,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046800908514035285,
      "loss": 5.1251,
      "mean_token_accuracy": 0.17335045337677002,
      "num_tokens": 39735736.0,
      "step": 18140
    },
    {
      "entropy": 5.561379528045654,
      "epoch": 1.8008138150059547,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046799087261241884,
      "loss": 5.1614,
      "mean_token_accuracy": 0.16988508850336076,
      "num_tokens": 39747443.0,
      "step": 18145
    },
    {
      "entropy": 5.578796148300171,
      "epoch": 1.8013100436681224,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004679726552987271,
      "loss": 5.0945,
      "mean_token_accuracy": 0.17009207457304001,
      "num_tokens": 39758128.0,
      "step": 18150
    },
    {
      "entropy": 5.557246732711792,
      "epoch": 1.80180627233029,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004679544331997294,
      "loss": 5.0297,
      "mean_token_accuracy": 0.18322060257196426,
      "num_tokens": 39769153.0,
      "step": 18155
    },
    {
      "entropy": 5.56773009300232,
      "epoch": 1.8023025009924574,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046793620631587743,
      "loss": 5.1124,
      "mean_token_accuracy": 0.16667832881212236,
      "num_tokens": 39779666.0,
      "step": 18160
    },
    {
      "entropy": 5.566453886032105,
      "epoch": 1.8027987296546248,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046791797464762334,
      "loss": 5.1462,
      "mean_token_accuracy": 0.16968514919281005,
      "num_tokens": 39790329.0,
      "step": 18165
    },
    {
      "entropy": 5.615738344192505,
      "epoch": 1.8032949583167923,
      "grad_norm": 0.921875,
      "learning_rate": 0.00046789973819541915,
      "loss": 5.1297,
      "mean_token_accuracy": 0.17371605485677719,
      "num_tokens": 39801996.0,
      "step": 18170
    },
    {
      "entropy": 5.625317525863648,
      "epoch": 1.8037911869789598,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004678814969597171,
      "loss": 5.1072,
      "mean_token_accuracy": 0.17198027819395065,
      "num_tokens": 39814064.0,
      "step": 18175
    },
    {
      "entropy": 5.553517389297485,
      "epoch": 1.8042874156411273,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004678632509409696,
      "loss": 5.0665,
      "mean_token_accuracy": 0.1776961162686348,
      "num_tokens": 39825369.0,
      "step": 18180
    },
    {
      "entropy": 5.510013246536255,
      "epoch": 1.804783644303295,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000467845000139629,
      "loss": 5.149,
      "mean_token_accuracy": 0.17642490863800048,
      "num_tokens": 39835218.0,
      "step": 18185
    },
    {
      "entropy": 5.5924444675445555,
      "epoch": 1.8052798729654624,
      "grad_norm": 1.03125,
      "learning_rate": 0.000467826744556148,
      "loss": 5.1726,
      "mean_token_accuracy": 0.16996136754751207,
      "num_tokens": 39844706.0,
      "step": 18190
    },
    {
      "entropy": 5.506793737411499,
      "epoch": 1.8057761016276301,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004678084841909792,
      "loss": 5.0975,
      "mean_token_accuracy": 0.17927393168210984,
      "num_tokens": 39856348.0,
      "step": 18195
    },
    {
      "entropy": 5.550699615478516,
      "epoch": 1.8062723302897976,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046779021904457555,
      "loss": 5.1446,
      "mean_token_accuracy": 0.1717533841729164,
      "num_tokens": 39866850.0,
      "step": 18200
    },
    {
      "entropy": 5.590773010253907,
      "epoch": 1.806768558951965,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004677719491173898,
      "loss": 5.1677,
      "mean_token_accuracy": 0.1759544476866722,
      "num_tokens": 39877647.0,
      "step": 18205
    },
    {
      "entropy": 5.584692096710205,
      "epoch": 1.8072647876141326,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00046775367440987506,
      "loss": 5.1894,
      "mean_token_accuracy": 0.17400931119918822,
      "num_tokens": 39889172.0,
      "step": 18210
    },
    {
      "entropy": 5.597168016433716,
      "epoch": 1.8077610162763,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004677353949224846,
      "loss": 5.1029,
      "mean_token_accuracy": 0.17396012246608733,
      "num_tokens": 39900445.0,
      "step": 18215
    },
    {
      "entropy": 5.571915483474731,
      "epoch": 1.8082572449384675,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00046771711065567167,
      "loss": 5.1331,
      "mean_token_accuracy": 0.17335934489965438,
      "num_tokens": 39912825.0,
      "step": 18220
    },
    {
      "entropy": 5.494789886474609,
      "epoch": 1.8087534736006352,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004676988216098896,
      "loss": 5.0799,
      "mean_token_accuracy": 0.17813094556331635,
      "num_tokens": 39923370.0,
      "step": 18225
    },
    {
      "entropy": 5.617679738998413,
      "epoch": 1.8092497022628027,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004676805277855921,
      "loss": 5.1446,
      "mean_token_accuracy": 0.17175042629241943,
      "num_tokens": 39934724.0,
      "step": 18230
    },
    {
      "entropy": 5.565017414093018,
      "epoch": 1.8097459309249704,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004676622291832326,
      "loss": 5.0718,
      "mean_token_accuracy": 0.16928407102823256,
      "num_tokens": 39947064.0,
      "step": 18235
    },
    {
      "entropy": 5.548071336746216,
      "epoch": 1.8102421595871379,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046764392580326506,
      "loss": 5.0727,
      "mean_token_accuracy": 0.17396903783082962,
      "num_tokens": 39957649.0,
      "step": 18240
    },
    {
      "entropy": 5.533809518814087,
      "epoch": 1.8107383882493053,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046762561764614316,
      "loss": 5.1187,
      "mean_token_accuracy": 0.1732972890138626,
      "num_tokens": 39968086.0,
      "step": 18245
    },
    {
      "entropy": 5.532074880599976,
      "epoch": 1.8112346169114728,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000467607304712321,
      "loss": 5.0719,
      "mean_token_accuracy": 0.1758315846323967,
      "num_tokens": 39978794.0,
      "step": 18250
    },
    {
      "entropy": 5.588777637481689,
      "epoch": 1.8117308455736403,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046758898700225283,
      "loss": 5.0938,
      "mean_token_accuracy": 0.1740478590130806,
      "num_tokens": 39990383.0,
      "step": 18255
    },
    {
      "entropy": 5.586848640441895,
      "epoch": 1.8122270742358078,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004675706645163927,
      "loss": 5.1662,
      "mean_token_accuracy": 0.17162053436040878,
      "num_tokens": 40001900.0,
      "step": 18260
    },
    {
      "entropy": 5.588620519638061,
      "epoch": 1.8127233028979752,
      "grad_norm": 0.953125,
      "learning_rate": 0.000467552337255195,
      "loss": 5.0829,
      "mean_token_accuracy": 0.17045395374298095,
      "num_tokens": 40012482.0,
      "step": 18265
    },
    {
      "entropy": 5.531330394744873,
      "epoch": 1.813219531560143,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004675340052191144,
      "loss": 5.1015,
      "mean_token_accuracy": 0.1747794896364212,
      "num_tokens": 40023264.0,
      "step": 18270
    },
    {
      "entropy": 5.606415128707885,
      "epoch": 1.8137157602223104,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004675156684086052,
      "loss": 5.1628,
      "mean_token_accuracy": 0.1697589173913002,
      "num_tokens": 40034754.0,
      "step": 18275
    },
    {
      "entropy": 5.5492490291595455,
      "epoch": 1.814211988884478,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004674973268241224,
      "loss": 5.0722,
      "mean_token_accuracy": 0.17935317307710646,
      "num_tokens": 40045781.0,
      "step": 18280
    },
    {
      "entropy": 5.591843318939209,
      "epoch": 1.8147082175466456,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004674789804661205,
      "loss": 5.1598,
      "mean_token_accuracy": 0.17444380968809128,
      "num_tokens": 40055700.0,
      "step": 18285
    },
    {
      "entropy": 5.605121040344239,
      "epoch": 1.815204446208813,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046746062933505477,
      "loss": 5.1735,
      "mean_token_accuracy": 0.17266058176755905,
      "num_tokens": 40066702.0,
      "step": 18290
    },
    {
      "entropy": 5.588819313049316,
      "epoch": 1.8157006748709805,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004674422734313801,
      "loss": 5.1766,
      "mean_token_accuracy": 0.17474929690361024,
      "num_tokens": 40077856.0,
      "step": 18295
    },
    {
      "entropy": 5.500308036804199,
      "epoch": 1.816196903533148,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046742391275555176,
      "loss": 5.064,
      "mean_token_accuracy": 0.17913249582052232,
      "num_tokens": 40089205.0,
      "step": 18300
    },
    {
      "entropy": 5.607101631164551,
      "epoch": 1.8166931321953155,
      "grad_norm": 0.875,
      "learning_rate": 0.000467405547308025,
      "loss": 5.1133,
      "mean_token_accuracy": 0.17143297642469407,
      "num_tokens": 40100028.0,
      "step": 18305
    },
    {
      "entropy": 5.647699689865112,
      "epoch": 1.8171893608574832,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004673871770892553,
      "loss": 5.1331,
      "mean_token_accuracy": 0.17165503948926925,
      "num_tokens": 40109635.0,
      "step": 18310
    },
    {
      "entropy": 5.551734590530396,
      "epoch": 1.8176855895196506,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00046736880209969814,
      "loss": 5.0925,
      "mean_token_accuracy": 0.17392676770687104,
      "num_tokens": 40120539.0,
      "step": 18315
    },
    {
      "entropy": 5.57048544883728,
      "epoch": 1.8181818181818183,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046735042233980923,
      "loss": 5.1512,
      "mean_token_accuracy": 0.16403615176677705,
      "num_tokens": 40131738.0,
      "step": 18320
    },
    {
      "entropy": 5.663658237457275,
      "epoch": 1.8186780468439858,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046733203781004426,
      "loss": 5.2196,
      "mean_token_accuracy": 0.17271289974451065,
      "num_tokens": 40142800.0,
      "step": 18325
    },
    {
      "entropy": 5.660841226577759,
      "epoch": 1.8191742755061533,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00046731364851085923,
      "loss": 5.1706,
      "mean_token_accuracy": 0.16808565855026245,
      "num_tokens": 40153992.0,
      "step": 18330
    },
    {
      "entropy": 5.589001703262329,
      "epoch": 1.8196705041683208,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004672952544427101,
      "loss": 5.1184,
      "mean_token_accuracy": 0.17254485189914703,
      "num_tokens": 40165213.0,
      "step": 18335
    },
    {
      "entropy": 5.560780811309814,
      "epoch": 1.8201667328304882,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046727685560605305,
      "loss": 5.1788,
      "mean_token_accuracy": 0.16840216815471648,
      "num_tokens": 40176763.0,
      "step": 18340
    },
    {
      "entropy": 5.56354169845581,
      "epoch": 1.8206629614926557,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004672584520013443,
      "loss": 5.0568,
      "mean_token_accuracy": 0.17163143754005433,
      "num_tokens": 40187958.0,
      "step": 18345
    },
    {
      "entropy": 5.533384513854981,
      "epoch": 1.8211591901548232,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004672400436290402,
      "loss": 5.0706,
      "mean_token_accuracy": 0.17672607004642488,
      "num_tokens": 40197881.0,
      "step": 18350
    },
    {
      "entropy": 5.583889198303223,
      "epoch": 1.821655418816991,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046722163048959725,
      "loss": 5.0538,
      "mean_token_accuracy": 0.18200553208589554,
      "num_tokens": 40209322.0,
      "step": 18355
    },
    {
      "entropy": 5.565239143371582,
      "epoch": 1.8221516474791584,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004672032125834721,
      "loss": 5.1173,
      "mean_token_accuracy": 0.17447513192892075,
      "num_tokens": 40222186.0,
      "step": 18360
    },
    {
      "entropy": 5.58490309715271,
      "epoch": 1.822647876141326,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004671847899111214,
      "loss": 5.1384,
      "mean_token_accuracy": 0.17238964587450029,
      "num_tokens": 40232136.0,
      "step": 18365
    },
    {
      "entropy": 5.6306617736816404,
      "epoch": 1.8231441048034935,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046716636247300206,
      "loss": 5.1782,
      "mean_token_accuracy": 0.16945978850126267,
      "num_tokens": 40244468.0,
      "step": 18370
    },
    {
      "entropy": 5.599199819564819,
      "epoch": 1.823640333465661,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046714793026957096,
      "loss": 5.1356,
      "mean_token_accuracy": 0.17083004713058472,
      "num_tokens": 40255441.0,
      "step": 18375
    },
    {
      "entropy": 5.516262674331665,
      "epoch": 1.8241365621278285,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004671294933012853,
      "loss": 5.045,
      "mean_token_accuracy": 0.1725158527493477,
      "num_tokens": 40266109.0,
      "step": 18380
    },
    {
      "entropy": 5.526769495010376,
      "epoch": 1.824632790789996,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046711105156860223,
      "loss": 5.0707,
      "mean_token_accuracy": 0.17202406227588654,
      "num_tokens": 40277635.0,
      "step": 18385
    },
    {
      "entropy": 5.570279455184936,
      "epoch": 1.8251290194521634,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046709260507197896,
      "loss": 5.2046,
      "mean_token_accuracy": 0.17353739440441132,
      "num_tokens": 40288668.0,
      "step": 18390
    },
    {
      "entropy": 5.53749303817749,
      "epoch": 1.8256252481143311,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046707415381187306,
      "loss": 5.0698,
      "mean_token_accuracy": 0.17616031616926192,
      "num_tokens": 40300262.0,
      "step": 18395
    },
    {
      "entropy": 5.5743152618408205,
      "epoch": 1.8261214767764986,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000467055697788742,
      "loss": 5.1363,
      "mean_token_accuracy": 0.1635638028383255,
      "num_tokens": 40310931.0,
      "step": 18400
    },
    {
      "entropy": 5.546328544616699,
      "epoch": 1.826617705438666,
      "grad_norm": 0.890625,
      "learning_rate": 0.00046703723700304345,
      "loss": 5.0857,
      "mean_token_accuracy": 0.17368315309286117,
      "num_tokens": 40321623.0,
      "step": 18405
    },
    {
      "entropy": 5.523750495910645,
      "epoch": 1.8271139341008338,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004670187714552353,
      "loss": 5.0479,
      "mean_token_accuracy": 0.175416561961174,
      "num_tokens": 40330951.0,
      "step": 18410
    },
    {
      "entropy": 5.586254405975342,
      "epoch": 1.8276101627630013,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00046700030114577533,
      "loss": 5.1344,
      "mean_token_accuracy": 0.17705303132534028,
      "num_tokens": 40342971.0,
      "step": 18415
    },
    {
      "entropy": 5.542151260375976,
      "epoch": 1.8281063914251687,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004669818260751217,
      "loss": 5.0877,
      "mean_token_accuracy": 0.18032497316598892,
      "num_tokens": 40353837.0,
      "step": 18420
    },
    {
      "entropy": 5.543359470367432,
      "epoch": 1.8286026200873362,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004669633462437324,
      "loss": 5.1785,
      "mean_token_accuracy": 0.17377234101295472,
      "num_tokens": 40367389.0,
      "step": 18425
    },
    {
      "entropy": 5.562901163101197,
      "epoch": 1.8290988487495037,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046694486165206573,
      "loss": 5.0014,
      "mean_token_accuracy": 0.1825328454375267,
      "num_tokens": 40376793.0,
      "step": 18430
    },
    {
      "entropy": 5.574186515808106,
      "epoch": 1.8295950774116712,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004669263723005801,
      "loss": 5.0984,
      "mean_token_accuracy": 0.17968219071626662,
      "num_tokens": 40388025.0,
      "step": 18435
    },
    {
      "entropy": 5.554730749130249,
      "epoch": 1.8300913060738389,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046690787818973406,
      "loss": 5.181,
      "mean_token_accuracy": 0.17453974038362502,
      "num_tokens": 40399731.0,
      "step": 18440
    },
    {
      "entropy": 5.575783920288086,
      "epoch": 1.8305875347360063,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00046688937931998613,
      "loss": 5.2303,
      "mean_token_accuracy": 0.1716172441840172,
      "num_tokens": 40411078.0,
      "step": 18445
    },
    {
      "entropy": 5.63482928276062,
      "epoch": 1.831083763398174,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004668708756917951,
      "loss": 5.1847,
      "mean_token_accuracy": 0.16972215175628663,
      "num_tokens": 40422806.0,
      "step": 18450
    },
    {
      "entropy": 5.704753303527832,
      "epoch": 1.8315799920603415,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046685236730561975,
      "loss": 5.2332,
      "mean_token_accuracy": 0.16531258299946786,
      "num_tokens": 40434041.0,
      "step": 18455
    },
    {
      "entropy": 5.601125764846802,
      "epoch": 1.832076220722509,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00046683385416191914,
      "loss": 5.0761,
      "mean_token_accuracy": 0.18101150542497635,
      "num_tokens": 40444658.0,
      "step": 18460
    },
    {
      "entropy": 5.5495442867279055,
      "epoch": 1.8325724493846764,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046681533626115226,
      "loss": 5.0834,
      "mean_token_accuracy": 0.17381613701581955,
      "num_tokens": 40455745.0,
      "step": 18465
    },
    {
      "entropy": 5.513105678558349,
      "epoch": 1.833068678046844,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004667968136037784,
      "loss": 5.1744,
      "mean_token_accuracy": 0.17271911799907685,
      "num_tokens": 40468829.0,
      "step": 18470
    },
    {
      "entropy": 5.526120615005493,
      "epoch": 1.8335649067090114,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046677828619025687,
      "loss": 5.0861,
      "mean_token_accuracy": 0.17547255456447602,
      "num_tokens": 40479817.0,
      "step": 18475
    },
    {
      "entropy": 5.669960689544678,
      "epoch": 1.8340611353711789,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046675975402104706,
      "loss": 5.191,
      "mean_token_accuracy": 0.1685073569417,
      "num_tokens": 40490398.0,
      "step": 18480
    },
    {
      "entropy": 5.539244556427002,
      "epoch": 1.8345573640333466,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046674121709660855,
      "loss": 5.0237,
      "mean_token_accuracy": 0.1787686377763748,
      "num_tokens": 40500483.0,
      "step": 18485
    },
    {
      "entropy": 5.5947342872619625,
      "epoch": 1.835053592695514,
      "grad_norm": 0.90625,
      "learning_rate": 0.000466722675417401,
      "loss": 5.1497,
      "mean_token_accuracy": 0.17010664641857148,
      "num_tokens": 40511494.0,
      "step": 18490
    },
    {
      "entropy": 5.575565528869629,
      "epoch": 1.8355498213576817,
      "grad_norm": 1.0,
      "learning_rate": 0.00046670412898388425,
      "loss": 5.0914,
      "mean_token_accuracy": 0.18553085327148439,
      "num_tokens": 40523573.0,
      "step": 18495
    },
    {
      "entropy": 5.555435085296631,
      "epoch": 1.8360460500198492,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00046668557779651825,
      "loss": 5.1492,
      "mean_token_accuracy": 0.17197633981704713,
      "num_tokens": 40534904.0,
      "step": 18500
    },
    {
      "entropy": 5.5364326477050785,
      "epoch": 1.8365422786820167,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046666702185576295,
      "loss": 5.1073,
      "mean_token_accuracy": 0.17526595890522004,
      "num_tokens": 40545053.0,
      "step": 18505
    },
    {
      "entropy": 5.5635459423065186,
      "epoch": 1.8370385073441842,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004666484611620785,
      "loss": 5.0389,
      "mean_token_accuracy": 0.18085704743862152,
      "num_tokens": 40555305.0,
      "step": 18510
    },
    {
      "entropy": 5.541917037963867,
      "epoch": 1.8375347360063516,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004666298957159251,
      "loss": 5.0736,
      "mean_token_accuracy": 0.17188790440559387,
      "num_tokens": 40565611.0,
      "step": 18515
    },
    {
      "entropy": 5.573732471466064,
      "epoch": 1.8380309646685191,
      "grad_norm": 1.0,
      "learning_rate": 0.00046661132551776327,
      "loss": 5.1574,
      "mean_token_accuracy": 0.17759806364774705,
      "num_tokens": 40578321.0,
      "step": 18520
    },
    {
      "entropy": 5.617476129531861,
      "epoch": 1.8385271933306868,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004665927505680535,
      "loss": 5.1566,
      "mean_token_accuracy": 0.1810738116502762,
      "num_tokens": 40588696.0,
      "step": 18525
    },
    {
      "entropy": 5.58264422416687,
      "epoch": 1.8390234219928543,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046657417086725625,
      "loss": 5.1116,
      "mean_token_accuracy": 0.1716594085097313,
      "num_tokens": 40599699.0,
      "step": 18530
    },
    {
      "entropy": 5.552099990844726,
      "epoch": 1.839519650655022,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004665555864158324,
      "loss": 5.1198,
      "mean_token_accuracy": 0.1748165175318718,
      "num_tokens": 40609706.0,
      "step": 18535
    },
    {
      "entropy": 5.559789562225342,
      "epoch": 1.8400158793171895,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004665369972142428,
      "loss": 5.1505,
      "mean_token_accuracy": 0.17779080122709273,
      "num_tokens": 40619431.0,
      "step": 18540
    },
    {
      "entropy": 5.611735916137695,
      "epoch": 1.840512107979357,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046651840326294845,
      "loss": 5.1182,
      "mean_token_accuracy": 0.17686475366353988,
      "num_tokens": 40629904.0,
      "step": 18545
    },
    {
      "entropy": 5.561684465408325,
      "epoch": 1.8410083366415244,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004664998045624104,
      "loss": 5.154,
      "mean_token_accuracy": 0.17088982909917833,
      "num_tokens": 40641266.0,
      "step": 18550
    },
    {
      "entropy": 5.518081474304199,
      "epoch": 1.8415045653036919,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004664812011130897,
      "loss": 5.0953,
      "mean_token_accuracy": 0.17238613218069077,
      "num_tokens": 40652454.0,
      "step": 18555
    },
    {
      "entropy": 5.577287769317627,
      "epoch": 1.8420007939658594,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046646259291544787,
      "loss": 5.0842,
      "mean_token_accuracy": 0.17364238500595092,
      "num_tokens": 40664311.0,
      "step": 18560
    },
    {
      "entropy": 5.50480432510376,
      "epoch": 1.8424970226280268,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004664439799699463,
      "loss": 5.012,
      "mean_token_accuracy": 0.17981128692626952,
      "num_tokens": 40675241.0,
      "step": 18565
    },
    {
      "entropy": 5.547020816802979,
      "epoch": 1.8429932512901945,
      "grad_norm": 0.921875,
      "learning_rate": 0.00046642536227704665,
      "loss": 5.0885,
      "mean_token_accuracy": 0.17679386138916015,
      "num_tokens": 40685972.0,
      "step": 18570
    },
    {
      "entropy": 5.533656358718872,
      "epoch": 1.843489479952362,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046640673983721037,
      "loss": 5.1091,
      "mean_token_accuracy": 0.17504760921001433,
      "num_tokens": 40696967.0,
      "step": 18575
    },
    {
      "entropy": 5.583533382415771,
      "epoch": 1.8439857086145297,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00046638811265089944,
      "loss": 5.1032,
      "mean_token_accuracy": 0.17084284424781798,
      "num_tokens": 40707710.0,
      "step": 18580
    },
    {
      "entropy": 5.6240739822387695,
      "epoch": 1.8444819372766972,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004663694807185758,
      "loss": 5.239,
      "mean_token_accuracy": 0.16960958391427994,
      "num_tokens": 40718586.0,
      "step": 18585
    },
    {
      "entropy": 5.6136102199554445,
      "epoch": 1.8449781659388647,
      "grad_norm": 1.0,
      "learning_rate": 0.00046635084404070135,
      "loss": 5.1142,
      "mean_token_accuracy": 0.1783537745475769,
      "num_tokens": 40729792.0,
      "step": 18590
    },
    {
      "entropy": 5.712875890731811,
      "epoch": 1.8454743946010321,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004663322026177382,
      "loss": 5.171,
      "mean_token_accuracy": 0.16736815646290779,
      "num_tokens": 40740974.0,
      "step": 18595
    },
    {
      "entropy": 5.526970338821411,
      "epoch": 1.8459706232631996,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046631355645014875,
      "loss": 4.9854,
      "mean_token_accuracy": 0.18436610251665114,
      "num_tokens": 40750208.0,
      "step": 18600
    },
    {
      "entropy": 5.544992113113404,
      "epoch": 1.846466851925367,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00046629490553839537,
      "loss": 5.0853,
      "mean_token_accuracy": 0.1767660841345787,
      "num_tokens": 40762163.0,
      "step": 18605
    },
    {
      "entropy": 5.58039116859436,
      "epoch": 1.8469630805875348,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004662762498829405,
      "loss": 5.1792,
      "mean_token_accuracy": 0.17419902086257935,
      "num_tokens": 40773028.0,
      "step": 18610
    },
    {
      "entropy": 5.637732410430909,
      "epoch": 1.8474593092497023,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004662575894842469,
      "loss": 5.1514,
      "mean_token_accuracy": 0.16426871418952943,
      "num_tokens": 40784939.0,
      "step": 18615
    },
    {
      "entropy": 5.591274976730347,
      "epoch": 1.84795553791187,
      "grad_norm": 0.84375,
      "learning_rate": 0.00046623892434277713,
      "loss": 5.1387,
      "mean_token_accuracy": 0.17336395382881165,
      "num_tokens": 40796170.0,
      "step": 18620
    },
    {
      "entropy": 5.578694629669189,
      "epoch": 1.8484517665740374,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004662202544589941,
      "loss": 5.0881,
      "mean_token_accuracy": 0.17862146049737931,
      "num_tokens": 40807184.0,
      "step": 18625
    },
    {
      "entropy": 5.542214488983154,
      "epoch": 1.848947995236205,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00046620157983336087,
      "loss": 5.0602,
      "mean_token_accuracy": 0.1744668588042259,
      "num_tokens": 40818474.0,
      "step": 18630
    },
    {
      "entropy": 5.55736722946167,
      "epoch": 1.8494442238983724,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004661829004663404,
      "loss": 5.1175,
      "mean_token_accuracy": 0.17546663284301758,
      "num_tokens": 40828789.0,
      "step": 18635
    },
    {
      "entropy": 5.568894910812378,
      "epoch": 1.8499404525605398,
      "grad_norm": 1.0,
      "learning_rate": 0.0004661642163583961,
      "loss": 5.0705,
      "mean_token_accuracy": 0.18279811441898347,
      "num_tokens": 40840564.0,
      "step": 18640
    },
    {
      "entropy": 5.534308624267578,
      "epoch": 1.8504366812227073,
      "grad_norm": 0.828125,
      "learning_rate": 0.00046614552750999093,
      "loss": 5.062,
      "mean_token_accuracy": 0.17633196860551834,
      "num_tokens": 40852528.0,
      "step": 18645
    },
    {
      "entropy": 5.551772880554199,
      "epoch": 1.8509329098848748,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004661268339215887,
      "loss": 5.1048,
      "mean_token_accuracy": 0.17363428324460983,
      "num_tokens": 40863242.0,
      "step": 18650
    },
    {
      "entropy": 5.53049521446228,
      "epoch": 1.8514291385470425,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004661081355936528,
      "loss": 5.0643,
      "mean_token_accuracy": 0.18281110972166062,
      "num_tokens": 40873721.0,
      "step": 18655
    },
    {
      "entropy": 5.599796867370605,
      "epoch": 1.85192536720921,
      "grad_norm": 1.015625,
      "learning_rate": 0.000466089432526647,
      "loss": 5.1475,
      "mean_token_accuracy": 0.17103291004896165,
      "num_tokens": 40884921.0,
      "step": 18660
    },
    {
      "entropy": 5.590263605117798,
      "epoch": 1.8524215958713777,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046607072472103493,
      "loss": 5.0436,
      "mean_token_accuracy": 0.17644645273685455,
      "num_tokens": 40896243.0,
      "step": 18665
    },
    {
      "entropy": 5.554067230224609,
      "epoch": 1.8529178245335451,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00046605201217728067,
      "loss": 5.1731,
      "mean_token_accuracy": 0.16817651987075805,
      "num_tokens": 40907191.0,
      "step": 18670
    },
    {
      "entropy": 5.596529293060303,
      "epoch": 1.8534140531957126,
      "grad_norm": 0.921875,
      "learning_rate": 0.00046603329489584817,
      "loss": 5.1277,
      "mean_token_accuracy": 0.18083939254283904,
      "num_tokens": 40918737.0,
      "step": 18675
    },
    {
      "entropy": 5.6284534454345705,
      "epoch": 1.85391028185788,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004660145728772016,
      "loss": 5.1377,
      "mean_token_accuracy": 0.17307621091604233,
      "num_tokens": 40930184.0,
      "step": 18680
    },
    {
      "entropy": 5.540746450424194,
      "epoch": 1.8544065105200476,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004659958461218052,
      "loss": 5.0406,
      "mean_token_accuracy": 0.18058212399482726,
      "num_tokens": 40941483.0,
      "step": 18685
    },
    {
      "entropy": 5.5579094886779785,
      "epoch": 1.854902739182215,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00046597711463012344,
      "loss": 5.1316,
      "mean_token_accuracy": 0.17433445304632186,
      "num_tokens": 40952392.0,
      "step": 18690
    },
    {
      "entropy": 5.528490924835205,
      "epoch": 1.8553989678443827,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004659583784026207,
      "loss": 5.0566,
      "mean_token_accuracy": 0.1776428610086441,
      "num_tokens": 40962539.0,
      "step": 18695
    },
    {
      "entropy": 5.543492603302002,
      "epoch": 1.8558951965065502,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046593963743976163,
      "loss": 4.9924,
      "mean_token_accuracy": 0.18435071557760238,
      "num_tokens": 40972594.0,
      "step": 18700
    },
    {
      "entropy": 5.5710138320922855,
      "epoch": 1.8563914251687177,
      "grad_norm": 0.9375,
      "learning_rate": 0.000465920891742011,
      "loss": 5.1198,
      "mean_token_accuracy": 0.17292593121528627,
      "num_tokens": 40983460.0,
      "step": 18705
    },
    {
      "entropy": 5.540741014480591,
      "epoch": 1.8568876538308854,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004659021413098337,
      "loss": 5.1035,
      "mean_token_accuracy": 0.1728368043899536,
      "num_tokens": 40993945.0,
      "step": 18710
    },
    {
      "entropy": 5.4902032852172855,
      "epoch": 1.8573838824930529,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046588338614369465,
      "loss": 5.0951,
      "mean_token_accuracy": 0.17472022622823716,
      "num_tokens": 41005781.0,
      "step": 18715
    },
    {
      "entropy": 5.621731233596802,
      "epoch": 1.8578801111552203,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00046586462624405897,
      "loss": 5.1968,
      "mean_token_accuracy": 0.16917262822389603,
      "num_tokens": 41017650.0,
      "step": 18720
    },
    {
      "entropy": 5.581646299362182,
      "epoch": 1.8583763398173878,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004658458616113918,
      "loss": 5.0719,
      "mean_token_accuracy": 0.17837297469377517,
      "num_tokens": 41029196.0,
      "step": 18725
    },
    {
      "entropy": 5.583410120010376,
      "epoch": 1.8588725684795553,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004658270922461585,
      "loss": 5.1414,
      "mean_token_accuracy": 0.1769915610551834,
      "num_tokens": 41040204.0,
      "step": 18730
    },
    {
      "entropy": 5.585629796981811,
      "epoch": 1.8593687971417228,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046580831814882446,
      "loss": 5.123,
      "mean_token_accuracy": 0.17069858312606812,
      "num_tokens": 41050507.0,
      "step": 18735
    },
    {
      "entropy": 5.527240896224976,
      "epoch": 1.8598650258038905,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046578953931985536,
      "loss": 5.092,
      "mean_token_accuracy": 0.17268568873405457,
      "num_tokens": 41062001.0,
      "step": 18740
    },
    {
      "entropy": 5.580925846099854,
      "epoch": 1.860361254466058,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004657707557597167,
      "loss": 5.0962,
      "mean_token_accuracy": 0.17323880940675734,
      "num_tokens": 41073097.0,
      "step": 18745
    },
    {
      "entropy": 5.583203363418579,
      "epoch": 1.8608574831282256,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046575196746887447,
      "loss": 5.1209,
      "mean_token_accuracy": 0.1713651418685913,
      "num_tokens": 41083883.0,
      "step": 18750
    },
    {
      "entropy": 5.489950323104859,
      "epoch": 1.861353711790393,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046573317444779445,
      "loss": 5.0949,
      "mean_token_accuracy": 0.18363923728466033,
      "num_tokens": 41095506.0,
      "step": 18755
    },
    {
      "entropy": 5.579948759078979,
      "epoch": 1.8618499404525606,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004657143766969427,
      "loss": 5.1232,
      "mean_token_accuracy": 0.17421770989894866,
      "num_tokens": 41106568.0,
      "step": 18760
    },
    {
      "entropy": 5.578218603134156,
      "epoch": 1.862346169114728,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004656955742167853,
      "loss": 5.015,
      "mean_token_accuracy": 0.17773817032575606,
      "num_tokens": 41117661.0,
      "step": 18765
    },
    {
      "entropy": 5.559988880157471,
      "epoch": 1.8628423977768955,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00046567676700778865,
      "loss": 5.1031,
      "mean_token_accuracy": 0.17744435220956803,
      "num_tokens": 41129209.0,
      "step": 18770
    },
    {
      "entropy": 5.564537715911865,
      "epoch": 1.863338626439063,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00046565795507041904,
      "loss": 5.1725,
      "mean_token_accuracy": 0.17351065427064896,
      "num_tokens": 41140715.0,
      "step": 18775
    },
    {
      "entropy": 5.572348117828369,
      "epoch": 1.8638348551012305,
      "grad_norm": 0.921875,
      "learning_rate": 0.000465639138405143,
      "loss": 5.1359,
      "mean_token_accuracy": 0.17433889359235763,
      "num_tokens": 41151080.0,
      "step": 18780
    },
    {
      "entropy": 5.595568132400513,
      "epoch": 1.8643310837633982,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046562031701242703,
      "loss": 5.1464,
      "mean_token_accuracy": 0.1780318945646286,
      "num_tokens": 41162816.0,
      "step": 18785
    },
    {
      "entropy": 5.568077230453492,
      "epoch": 1.8648273124255657,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046560149089273803,
      "loss": 5.0409,
      "mean_token_accuracy": 0.1847884625196457,
      "num_tokens": 41173456.0,
      "step": 18790
    },
    {
      "entropy": 5.5413374423980715,
      "epoch": 1.8653235410877333,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004655826600465427,
      "loss": 5.0037,
      "mean_token_accuracy": 0.1799089252948761,
      "num_tokens": 41184436.0,
      "step": 18795
    },
    {
      "entropy": 5.512774038314819,
      "epoch": 1.8658197697499008,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046556382447430816,
      "loss": 5.0765,
      "mean_token_accuracy": 0.17751057744026183,
      "num_tokens": 41196417.0,
      "step": 18800
    },
    {
      "entropy": 5.5446936130523685,
      "epoch": 1.8663159984120683,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046554498417650134,
      "loss": 5.0831,
      "mean_token_accuracy": 0.17722584307193756,
      "num_tokens": 41206054.0,
      "step": 18805
    },
    {
      "entropy": 5.537550544738769,
      "epoch": 1.8668122270742358,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004655261391535895,
      "loss": 5.0938,
      "mean_token_accuracy": 0.17636859863996507,
      "num_tokens": 41217002.0,
      "step": 18810
    },
    {
      "entropy": 5.706327629089356,
      "epoch": 1.8673084557364032,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004655072894060399,
      "loss": 5.2437,
      "mean_token_accuracy": 0.16572582125663757,
      "num_tokens": 41228485.0,
      "step": 18815
    },
    {
      "entropy": 5.609031391143799,
      "epoch": 1.8678046843985707,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004654884349343201,
      "loss": 5.1189,
      "mean_token_accuracy": 0.17708613574504853,
      "num_tokens": 41240600.0,
      "step": 18820
    },
    {
      "entropy": 5.643232965469361,
      "epoch": 1.8683009130607384,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046546957573889757,
      "loss": 5.211,
      "mean_token_accuracy": 0.1694723054766655,
      "num_tokens": 41251870.0,
      "step": 18825
    },
    {
      "entropy": 5.54521598815918,
      "epoch": 1.868797141722906,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046545071182024,
      "loss": 5.1097,
      "mean_token_accuracy": 0.16906451284885407,
      "num_tokens": 41264679.0,
      "step": 18830
    },
    {
      "entropy": 5.6929511547088625,
      "epoch": 1.8692933703850736,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046543184317881516,
      "loss": 5.1976,
      "mean_token_accuracy": 0.16911463141441346,
      "num_tokens": 41275690.0,
      "step": 18835
    },
    {
      "entropy": 5.554496812820434,
      "epoch": 1.869789599047241,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000465412969815091,
      "loss": 5.0817,
      "mean_token_accuracy": 0.17793331444263458,
      "num_tokens": 41285118.0,
      "step": 18840
    },
    {
      "entropy": 5.5289019584655765,
      "epoch": 1.8702858277094085,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004653940917295354,
      "loss": 5.0799,
      "mean_token_accuracy": 0.1762320175766945,
      "num_tokens": 41296330.0,
      "step": 18845
    },
    {
      "entropy": 5.521486616134643,
      "epoch": 1.870782056371576,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004653752089226166,
      "loss": 5.1071,
      "mean_token_accuracy": 0.17323946654796601,
      "num_tokens": 41307051.0,
      "step": 18850
    },
    {
      "entropy": 5.562676906585693,
      "epoch": 1.8712782850337435,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004653563213948029,
      "loss": 5.0459,
      "mean_token_accuracy": 0.18045729994773865,
      "num_tokens": 41317111.0,
      "step": 18855
    },
    {
      "entropy": 5.500308227539063,
      "epoch": 1.871774513695911,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046533742914656254,
      "loss": 5.0317,
      "mean_token_accuracy": 0.17823178917169571,
      "num_tokens": 41327509.0,
      "step": 18860
    },
    {
      "entropy": 5.558806753158569,
      "epoch": 1.8722707423580784,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046531853217836414,
      "loss": 5.1217,
      "mean_token_accuracy": 0.1714998885989189,
      "num_tokens": 41338721.0,
      "step": 18865
    },
    {
      "entropy": 5.633945941925049,
      "epoch": 1.8727669710202461,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046529963049067623,
      "loss": 5.1667,
      "mean_token_accuracy": 0.1708723098039627,
      "num_tokens": 41349767.0,
      "step": 18870
    },
    {
      "entropy": 5.686631917953491,
      "epoch": 1.8732631996824136,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046528072408396747,
      "loss": 5.2718,
      "mean_token_accuracy": 0.1667434275150299,
      "num_tokens": 41361151.0,
      "step": 18875
    },
    {
      "entropy": 5.5772655487060545,
      "epoch": 1.8737594283445813,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004652618129587069,
      "loss": 5.0664,
      "mean_token_accuracy": 0.17504362016916275,
      "num_tokens": 41372875.0,
      "step": 18880
    },
    {
      "entropy": 5.622480297088623,
      "epoch": 1.8742556570067488,
      "grad_norm": 1.0,
      "learning_rate": 0.00046524289711536327,
      "loss": 5.1375,
      "mean_token_accuracy": 0.17084690034389496,
      "num_tokens": 41383284.0,
      "step": 18885
    },
    {
      "entropy": 5.5345375537872314,
      "epoch": 1.8747518856689163,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004652239765544057,
      "loss": 5.0917,
      "mean_token_accuracy": 0.17910555750131607,
      "num_tokens": 41395389.0,
      "step": 18890
    },
    {
      "entropy": 5.506894922256469,
      "epoch": 1.8752481143310837,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046520505127630346,
      "loss": 5.0512,
      "mean_token_accuracy": 0.17614742517471313,
      "num_tokens": 41406740.0,
      "step": 18895
    },
    {
      "entropy": 5.541567993164063,
      "epoch": 1.8757443429932512,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046518612128152576,
      "loss": 5.0679,
      "mean_token_accuracy": 0.17749993801116942,
      "num_tokens": 41417279.0,
      "step": 18900
    },
    {
      "entropy": 5.512280082702636,
      "epoch": 1.8762405716554187,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004651671865705421,
      "loss": 5.0373,
      "mean_token_accuracy": 0.17643697261810304,
      "num_tokens": 41428574.0,
      "step": 18905
    },
    {
      "entropy": 5.560574626922607,
      "epoch": 1.8767368003175864,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00046514824714382204,
      "loss": 5.0315,
      "mean_token_accuracy": 0.17742726802825928,
      "num_tokens": 41438784.0,
      "step": 18910
    },
    {
      "entropy": 5.559384393692016,
      "epoch": 1.8772330289797539,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004651293030018351,
      "loss": 5.0711,
      "mean_token_accuracy": 0.1744977653026581,
      "num_tokens": 41448481.0,
      "step": 18915
    },
    {
      "entropy": 5.5868010997772215,
      "epoch": 1.8777292576419216,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046511035414505115,
      "loss": 5.1155,
      "mean_token_accuracy": 0.1788717344403267,
      "num_tokens": 41459097.0,
      "step": 18920
    },
    {
      "entropy": 5.60605640411377,
      "epoch": 1.878225486304089,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004650914005739401,
      "loss": 5.1579,
      "mean_token_accuracy": 0.17140321284532548,
      "num_tokens": 41469647.0,
      "step": 18925
    },
    {
      "entropy": 5.562855243682861,
      "epoch": 1.8787217149662565,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004650724422889719,
      "loss": 5.1008,
      "mean_token_accuracy": 0.16781231462955476,
      "num_tokens": 41479826.0,
      "step": 18930
    },
    {
      "entropy": 5.600466108322143,
      "epoch": 1.879217943628424,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004650534792906167,
      "loss": 5.0571,
      "mean_token_accuracy": 0.17631438374519348,
      "num_tokens": 41490551.0,
      "step": 18935
    },
    {
      "entropy": 5.556157064437866,
      "epoch": 1.8797141722905915,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004650345115793447,
      "loss": 5.0914,
      "mean_token_accuracy": 0.179885895550251,
      "num_tokens": 41500908.0,
      "step": 18940
    },
    {
      "entropy": 5.537628269195556,
      "epoch": 1.880210400952759,
      "grad_norm": 0.953125,
      "learning_rate": 0.00046501553915562644,
      "loss": 5.064,
      "mean_token_accuracy": 0.1736200124025345,
      "num_tokens": 41511311.0,
      "step": 18945
    },
    {
      "entropy": 5.530085325241089,
      "epoch": 1.8807066296149264,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046499656201993213,
      "loss": 5.093,
      "mean_token_accuracy": 0.17708706259727477,
      "num_tokens": 41522068.0,
      "step": 18950
    },
    {
      "entropy": 5.543504238128662,
      "epoch": 1.881202858277094,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046497758017273255,
      "loss": 5.0983,
      "mean_token_accuracy": 0.17089073359966278,
      "num_tokens": 41534099.0,
      "step": 18955
    },
    {
      "entropy": 5.606848621368409,
      "epoch": 1.8816990869392616,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00046495859361449834,
      "loss": 5.1634,
      "mean_token_accuracy": 0.16555947065353394,
      "num_tokens": 41545870.0,
      "step": 18960
    },
    {
      "entropy": 5.581093740463257,
      "epoch": 1.8821953156014293,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004649396023457003,
      "loss": 5.1551,
      "mean_token_accuracy": 0.17180070132017136,
      "num_tokens": 41558175.0,
      "step": 18965
    },
    {
      "entropy": 5.548663139343262,
      "epoch": 1.8826915442635968,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004649206063668094,
      "loss": 5.1078,
      "mean_token_accuracy": 0.18083097487688066,
      "num_tokens": 41569326.0,
      "step": 18970
    },
    {
      "entropy": 5.562871885299683,
      "epoch": 1.8831877729257642,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004649016056782968,
      "loss": 5.0559,
      "mean_token_accuracy": 0.18182073831558226,
      "num_tokens": 41580065.0,
      "step": 18975
    },
    {
      "entropy": 5.634759473800659,
      "epoch": 1.8836840015879317,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004648826002806334,
      "loss": 5.1979,
      "mean_token_accuracy": 0.1764739543199539,
      "num_tokens": 41590915.0,
      "step": 18980
    },
    {
      "entropy": 5.533243465423584,
      "epoch": 1.8841802302500992,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004648635901742909,
      "loss": 5.0467,
      "mean_token_accuracy": 0.18027180135250093,
      "num_tokens": 41602096.0,
      "step": 18985
    },
    {
      "entropy": 5.566839551925659,
      "epoch": 1.8846764589122667,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004648445753597403,
      "loss": 5.1614,
      "mean_token_accuracy": 0.17083170861005784,
      "num_tokens": 41612589.0,
      "step": 18990
    },
    {
      "entropy": 5.540729522705078,
      "epoch": 1.8851726875744343,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00046482555583745336,
      "loss": 5.0539,
      "mean_token_accuracy": 0.1739143282175064,
      "num_tokens": 41624163.0,
      "step": 18995
    },
    {
      "entropy": 5.574202299118042,
      "epoch": 1.8856689162366018,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004648065316079017,
      "loss": 5.055,
      "mean_token_accuracy": 0.17677053064107895,
      "num_tokens": 41635919.0,
      "step": 19000
    },
    {
      "entropy": 5.536738491058349,
      "epoch": 1.8861651448987695,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046478750267155703,
      "loss": 5.0786,
      "mean_token_accuracy": 0.17330556958913804,
      "num_tokens": 41644723.0,
      "step": 19005
    },
    {
      "entropy": 5.548032522201538,
      "epoch": 1.886661373560937,
      "grad_norm": 0.83984375,
      "learning_rate": 0.00046476846902889125,
      "loss": 5.1432,
      "mean_token_accuracy": 0.17674286663532257,
      "num_tokens": 41657137.0,
      "step": 19010
    },
    {
      "entropy": 5.599433040618896,
      "epoch": 1.8871576022231045,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004647494306803764,
      "loss": 5.1179,
      "mean_token_accuracy": 0.17483253180980682,
      "num_tokens": 41667773.0,
      "step": 19015
    },
    {
      "entropy": 5.624449348449707,
      "epoch": 1.887653830885272,
      "grad_norm": 0.890625,
      "learning_rate": 0.00046473038762648446,
      "loss": 5.1877,
      "mean_token_accuracy": 0.17361018806695938,
      "num_tokens": 41680549.0,
      "step": 19020
    },
    {
      "entropy": 5.569558334350586,
      "epoch": 1.8881500595474394,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046471133986768784,
      "loss": 5.0835,
      "mean_token_accuracy": 0.17772890329360963,
      "num_tokens": 41689700.0,
      "step": 19025
    },
    {
      "entropy": 5.63198037147522,
      "epoch": 1.888646288209607,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046469228740445874,
      "loss": 5.1248,
      "mean_token_accuracy": 0.17821198403835298,
      "num_tokens": 41699734.0,
      "step": 19030
    },
    {
      "entropy": 5.5688129425048825,
      "epoch": 1.8891425168717744,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046467323023726965,
      "loss": 5.1243,
      "mean_token_accuracy": 0.17509949058294297,
      "num_tokens": 41710644.0,
      "step": 19035
    },
    {
      "entropy": 5.499895334243774,
      "epoch": 1.889638745533942,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046465416836659313,
      "loss": 5.0394,
      "mean_token_accuracy": 0.1806145504117012,
      "num_tokens": 41721416.0,
      "step": 19040
    },
    {
      "entropy": 5.582535171508789,
      "epoch": 1.8901349741961095,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000464635101792902,
      "loss": 5.0517,
      "mean_token_accuracy": 0.17709997147321702,
      "num_tokens": 41732139.0,
      "step": 19045
    },
    {
      "entropy": 5.561202049255371,
      "epoch": 1.8906312028582772,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046461603051666883,
      "loss": 5.1488,
      "mean_token_accuracy": 0.17352749407291412,
      "num_tokens": 41744233.0,
      "step": 19050
    },
    {
      "entropy": 5.561340999603272,
      "epoch": 1.8911274315204447,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046459695453836676,
      "loss": 5.0699,
      "mean_token_accuracy": 0.1834057793021202,
      "num_tokens": 41755514.0,
      "step": 19055
    },
    {
      "entropy": 5.596374559402466,
      "epoch": 1.8916236601826122,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004645778738584688,
      "loss": 5.1034,
      "mean_token_accuracy": 0.17541026175022126,
      "num_tokens": 41765613.0,
      "step": 19060
    },
    {
      "entropy": 5.609683656692505,
      "epoch": 1.8921198888447797,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004645587884774481,
      "loss": 5.1388,
      "mean_token_accuracy": 0.1724128544330597,
      "num_tokens": 41777294.0,
      "step": 19065
    },
    {
      "entropy": 5.5225907325744625,
      "epoch": 1.8926161175069471,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004645396983957778,
      "loss": 5.0444,
      "mean_token_accuracy": 0.18220995217561722,
      "num_tokens": 41788278.0,
      "step": 19070
    },
    {
      "entropy": 5.578003025054931,
      "epoch": 1.8931123461691146,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046452060361393147,
      "loss": 5.1093,
      "mean_token_accuracy": 0.1702814817428589,
      "num_tokens": 41798663.0,
      "step": 19075
    },
    {
      "entropy": 5.584049367904663,
      "epoch": 1.8936085748312823,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004645015041323825,
      "loss": 5.0664,
      "mean_token_accuracy": 0.1794762507081032,
      "num_tokens": 41808711.0,
      "step": 19080
    },
    {
      "entropy": 5.580535745620727,
      "epoch": 1.8941048034934498,
      "grad_norm": 1.0,
      "learning_rate": 0.00046448239995160455,
      "loss": 5.1147,
      "mean_token_accuracy": 0.1716906785964966,
      "num_tokens": 41819679.0,
      "step": 19085
    },
    {
      "entropy": 5.611795473098755,
      "epoch": 1.8946010321556173,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004644632910720715,
      "loss": 5.0996,
      "mean_token_accuracy": 0.17616001665592193,
      "num_tokens": 41829649.0,
      "step": 19090
    },
    {
      "entropy": 5.5847875595092775,
      "epoch": 1.895097260817785,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000464444177494257,
      "loss": 5.1472,
      "mean_token_accuracy": 0.17002528309822082,
      "num_tokens": 41842523.0,
      "step": 19095
    },
    {
      "entropy": 5.566792058944702,
      "epoch": 1.8955934894799524,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004644250592186351,
      "loss": 5.1119,
      "mean_token_accuracy": 0.17423339039087296,
      "num_tokens": 41854514.0,
      "step": 19100
    },
    {
      "entropy": 5.598566293716431,
      "epoch": 1.89608971814212,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00046440593624568,
      "loss": 5.0832,
      "mean_token_accuracy": 0.17401032745838166,
      "num_tokens": 41865166.0,
      "step": 19105
    },
    {
      "entropy": 5.559074068069458,
      "epoch": 1.8965859468042874,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004643868085758657,
      "loss": 5.0405,
      "mean_token_accuracy": 0.18056625127792358,
      "num_tokens": 41877264.0,
      "step": 19110
    },
    {
      "entropy": 5.581889581680298,
      "epoch": 1.8970821754664549,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046436767620966675,
      "loss": 5.0646,
      "mean_token_accuracy": 0.17500241994857788,
      "num_tokens": 41887345.0,
      "step": 19115
    },
    {
      "entropy": 5.5845636367797855,
      "epoch": 1.8975784041286223,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046434853914755744,
      "loss": 5.1214,
      "mean_token_accuracy": 0.17222516536712645,
      "num_tokens": 41898877.0,
      "step": 19120
    },
    {
      "entropy": 5.510692548751831,
      "epoch": 1.89807463279079,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00046432939739001237,
      "loss": 5.0427,
      "mean_token_accuracy": 0.17791842073202133,
      "num_tokens": 41910593.0,
      "step": 19125
    },
    {
      "entropy": 5.540350389480591,
      "epoch": 1.8985708614529575,
      "grad_norm": 0.953125,
      "learning_rate": 0.00046431025093750613,
      "loss": 5.0833,
      "mean_token_accuracy": 0.18365292400121688,
      "num_tokens": 41920719.0,
      "step": 19130
    },
    {
      "entropy": 5.597641515731811,
      "epoch": 1.8990670901151252,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00046429109979051367,
      "loss": 5.1212,
      "mean_token_accuracy": 0.16393977999687195,
      "num_tokens": 41931630.0,
      "step": 19135
    },
    {
      "entropy": 5.574547672271729,
      "epoch": 1.8995633187772927,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046427194394950985,
      "loss": 5.1501,
      "mean_token_accuracy": 0.1732749342918396,
      "num_tokens": 41943076.0,
      "step": 19140
    },
    {
      "entropy": 5.592162847518921,
      "epoch": 1.9000595474394602,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004642527834149696,
      "loss": 5.0451,
      "mean_token_accuracy": 0.17585303485393525,
      "num_tokens": 41954294.0,
      "step": 19145
    },
    {
      "entropy": 5.541206645965576,
      "epoch": 1.9005557761016276,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046423361818736814,
      "loss": 5.0446,
      "mean_token_accuracy": 0.17360924184322357,
      "num_tokens": 41964943.0,
      "step": 19150
    },
    {
      "entropy": 5.504830980300904,
      "epoch": 1.901052004763795,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004642144482671807,
      "loss": 5.0784,
      "mean_token_accuracy": 0.17577015608549118,
      "num_tokens": 41975211.0,
      "step": 19155
    },
    {
      "entropy": 5.558339405059814,
      "epoch": 1.9015482334259626,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00046419527365488275,
      "loss": 5.0883,
      "mean_token_accuracy": 0.17465862333774568,
      "num_tokens": 41986492.0,
      "step": 19160
    },
    {
      "entropy": 5.530906438827515,
      "epoch": 1.90204446208813,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004641760943509496,
      "loss": 5.0697,
      "mean_token_accuracy": 0.17716875225305556,
      "num_tokens": 41997613.0,
      "step": 19165
    },
    {
      "entropy": 5.547230434417725,
      "epoch": 1.9025406907502977,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000464156910355857,
      "loss": 5.1404,
      "mean_token_accuracy": 0.16788488924503325,
      "num_tokens": 42008597.0,
      "step": 19170
    },
    {
      "entropy": 5.544430875778199,
      "epoch": 1.9030369194124652,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00046413772167008057,
      "loss": 5.1206,
      "mean_token_accuracy": 0.17881128191947937,
      "num_tokens": 42019698.0,
      "step": 19175
    },
    {
      "entropy": 5.563129472732544,
      "epoch": 1.903533148074633,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046411852829409626,
      "loss": 5.0122,
      "mean_token_accuracy": 0.17747138440608978,
      "num_tokens": 42029473.0,
      "step": 19180
    },
    {
      "entropy": 5.5570985794067385,
      "epoch": 1.9040293767368004,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046409933022837984,
      "loss": 5.1532,
      "mean_token_accuracy": 0.17229220420122146,
      "num_tokens": 42041494.0,
      "step": 19185
    },
    {
      "entropy": 5.57109317779541,
      "epoch": 1.9045256053989679,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004640801274734076,
      "loss": 5.1121,
      "mean_token_accuracy": 0.17497511059045792,
      "num_tokens": 42051798.0,
      "step": 19190
    },
    {
      "entropy": 5.569985723495483,
      "epoch": 1.9050218340611353,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004640609200296557,
      "loss": 5.0381,
      "mean_token_accuracy": 0.17674174457788466,
      "num_tokens": 42061802.0,
      "step": 19195
    },
    {
      "entropy": 5.5907776832580565,
      "epoch": 1.9055180627233028,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004640417078976003,
      "loss": 5.1602,
      "mean_token_accuracy": 0.16748594790697097,
      "num_tokens": 42071503.0,
      "step": 19200
    },
    {
      "entropy": 5.445274019241333,
      "epoch": 1.9060142913854703,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004640224910777179,
      "loss": 5.0066,
      "mean_token_accuracy": 0.1784880205988884,
      "num_tokens": 42082104.0,
      "step": 19205
    },
    {
      "entropy": 5.5312236785888675,
      "epoch": 1.906510520047638,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000464003269570485,
      "loss": 5.0701,
      "mean_token_accuracy": 0.17583508640527726,
      "num_tokens": 42092798.0,
      "step": 19210
    },
    {
      "entropy": 5.574150419235229,
      "epoch": 1.9070067487098055,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004639840433763783,
      "loss": 5.1179,
      "mean_token_accuracy": 0.1736740291118622,
      "num_tokens": 42103062.0,
      "step": 19215
    },
    {
      "entropy": 5.599778842926026,
      "epoch": 1.9075029773719732,
      "grad_norm": 1.0,
      "learning_rate": 0.00046396481249587454,
      "loss": 5.1575,
      "mean_token_accuracy": 0.17718523293733596,
      "num_tokens": 42115062.0,
      "step": 19220
    },
    {
      "entropy": 5.651457977294922,
      "epoch": 1.9079992060341406,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004639455769294506,
      "loss": 5.1511,
      "mean_token_accuracy": 0.17218709439039231,
      "num_tokens": 42125937.0,
      "step": 19225
    },
    {
      "entropy": 5.579854249954224,
      "epoch": 1.9084954346963081,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004639263366775835,
      "loss": 5.0857,
      "mean_token_accuracy": 0.17947132289409637,
      "num_tokens": 42136916.0,
      "step": 19230
    },
    {
      "entropy": 5.60452446937561,
      "epoch": 1.9089916633584756,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046390709174075036,
      "loss": 5.1483,
      "mean_token_accuracy": 0.16964043527841569,
      "num_tokens": 42148348.0,
      "step": 19235
    },
    {
      "entropy": 5.621536302566528,
      "epoch": 1.909487892020643,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004638878421194284,
      "loss": 5.0551,
      "mean_token_accuracy": 0.1832812622189522,
      "num_tokens": 42157586.0,
      "step": 19240
    },
    {
      "entropy": 5.566982412338257,
      "epoch": 1.9099841206828105,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000463868587814095,
      "loss": 5.1245,
      "mean_token_accuracy": 0.16906868666410446,
      "num_tokens": 42169536.0,
      "step": 19245
    },
    {
      "entropy": 5.4944923400878904,
      "epoch": 1.910480349344978,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004638493288252276,
      "loss": 5.0521,
      "mean_token_accuracy": 0.1763198718428612,
      "num_tokens": 42180024.0,
      "step": 19250
    },
    {
      "entropy": 5.565520906448365,
      "epoch": 1.9109765780071457,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046383006515330374,
      "loss": 5.0623,
      "mean_token_accuracy": 0.1832951694726944,
      "num_tokens": 42191070.0,
      "step": 19255
    },
    {
      "entropy": 5.570163774490356,
      "epoch": 1.9114728066693132,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004638107967988012,
      "loss": 5.115,
      "mean_token_accuracy": 0.17710933834314346,
      "num_tokens": 42203643.0,
      "step": 19260
    },
    {
      "entropy": 5.5978234767913815,
      "epoch": 1.9119690353314809,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046379152376219775,
      "loss": 5.1622,
      "mean_token_accuracy": 0.17120399177074433,
      "num_tokens": 42214979.0,
      "step": 19265
    },
    {
      "entropy": 5.677051115036011,
      "epoch": 1.9124652639936484,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046377224604397127,
      "loss": 5.1499,
      "mean_token_accuracy": 0.17099551260471343,
      "num_tokens": 42225152.0,
      "step": 19270
    },
    {
      "entropy": 5.624697017669678,
      "epoch": 1.9129614926558158,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004637529636445998,
      "loss": 5.0693,
      "mean_token_accuracy": 0.1809826523065567,
      "num_tokens": 42235151.0,
      "step": 19275
    },
    {
      "entropy": 5.546418952941894,
      "epoch": 1.9134577213179833,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004637336765645617,
      "loss": 5.0656,
      "mean_token_accuracy": 0.1829977810382843,
      "num_tokens": 42245419.0,
      "step": 19280
    },
    {
      "entropy": 5.508101892471314,
      "epoch": 1.9139539499801508,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046371438480433504,
      "loss": 5.0482,
      "mean_token_accuracy": 0.17987146526575087,
      "num_tokens": 42255836.0,
      "step": 19285
    },
    {
      "entropy": 5.5303544998168945,
      "epoch": 1.9144501786423183,
      "grad_norm": 0.875,
      "learning_rate": 0.0004636950883643983,
      "loss": 5.1416,
      "mean_token_accuracy": 0.17315745651721953,
      "num_tokens": 42267490.0,
      "step": 19290
    },
    {
      "entropy": 5.603589725494385,
      "epoch": 1.914946407304486,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046367578724523,
      "loss": 5.0997,
      "mean_token_accuracy": 0.17101542502641678,
      "num_tokens": 42276957.0,
      "step": 19295
    },
    {
      "entropy": 5.602015352249145,
      "epoch": 1.9154426359666534,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004636564814473087,
      "loss": 4.979,
      "mean_token_accuracy": 0.18534846901893615,
      "num_tokens": 42288469.0,
      "step": 19300
    },
    {
      "entropy": 5.541460561752319,
      "epoch": 1.9159388646288211,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004636371709711131,
      "loss": 5.0443,
      "mean_token_accuracy": 0.18952454775571823,
      "num_tokens": 42299141.0,
      "step": 19305
    },
    {
      "entropy": 5.529378271102905,
      "epoch": 1.9164350932909886,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00046361785581712224,
      "loss": 5.1296,
      "mean_token_accuracy": 0.17912334352731704,
      "num_tokens": 42310651.0,
      "step": 19310
    },
    {
      "entropy": 5.513909864425659,
      "epoch": 1.916931321953156,
      "grad_norm": 1.046875,
      "learning_rate": 0.000463598535985815,
      "loss": 4.9541,
      "mean_token_accuracy": 0.19018761366605758,
      "num_tokens": 42320529.0,
      "step": 19315
    },
    {
      "entropy": 5.556969070434571,
      "epoch": 1.9174275506153236,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004635792114776704,
      "loss": 5.0079,
      "mean_token_accuracy": 0.18231214135885238,
      "num_tokens": 42331216.0,
      "step": 19320
    },
    {
      "entropy": 5.490019845962524,
      "epoch": 1.917923779277491,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004635598822931677,
      "loss": 5.0202,
      "mean_token_accuracy": 0.18094714879989623,
      "num_tokens": 42341730.0,
      "step": 19325
    },
    {
      "entropy": 5.572424507141113,
      "epoch": 1.9184200079396585,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046354054843278624,
      "loss": 5.1173,
      "mean_token_accuracy": 0.1721300482749939,
      "num_tokens": 42352643.0,
      "step": 19330
    },
    {
      "entropy": 5.626467752456665,
      "epoch": 1.918916236601826,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046352120989700553,
      "loss": 5.1645,
      "mean_token_accuracy": 0.16845486909151078,
      "num_tokens": 42363979.0,
      "step": 19335
    },
    {
      "entropy": 5.629286670684815,
      "epoch": 1.9194124652639937,
      "grad_norm": 1.046875,
      "learning_rate": 0.000463501866686305,
      "loss": 5.105,
      "mean_token_accuracy": 0.1746789187192917,
      "num_tokens": 42374104.0,
      "step": 19340
    },
    {
      "entropy": 5.61504340171814,
      "epoch": 1.9199086939261611,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046348251880116433,
      "loss": 5.0956,
      "mean_token_accuracy": 0.17424288243055344,
      "num_tokens": 42383576.0,
      "step": 19345
    },
    {
      "entropy": 5.573254346847534,
      "epoch": 1.9204049225883288,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046346316624206325,
      "loss": 5.1033,
      "mean_token_accuracy": 0.1774061381816864,
      "num_tokens": 42394625.0,
      "step": 19350
    },
    {
      "entropy": 5.543413782119751,
      "epoch": 1.9209011512504963,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004634438090094819,
      "loss": 5.0496,
      "mean_token_accuracy": 0.17962605953216554,
      "num_tokens": 42404764.0,
      "step": 19355
    },
    {
      "entropy": 5.584619140625,
      "epoch": 1.9213973799126638,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046342444710390006,
      "loss": 5.0269,
      "mean_token_accuracy": 0.17842495441436768,
      "num_tokens": 42414386.0,
      "step": 19360
    },
    {
      "entropy": 5.570868921279907,
      "epoch": 1.9218936085748313,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000463405080525798,
      "loss": 5.1242,
      "mean_token_accuracy": 0.17460551410913466,
      "num_tokens": 42425671.0,
      "step": 19365
    },
    {
      "entropy": 5.583004426956177,
      "epoch": 1.9223898372369987,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004633857092756559,
      "loss": 5.1392,
      "mean_token_accuracy": 0.16956614702939987,
      "num_tokens": 42436365.0,
      "step": 19370
    },
    {
      "entropy": 5.547567653656006,
      "epoch": 1.9228860658991662,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004633663333539542,
      "loss": 5.0038,
      "mean_token_accuracy": 0.1752679243683815,
      "num_tokens": 42447624.0,
      "step": 19375
    },
    {
      "entropy": 5.629079103469849,
      "epoch": 1.923382294561334,
      "grad_norm": 0.9375,
      "learning_rate": 0.00046334695276117324,
      "loss": 5.1018,
      "mean_token_accuracy": 0.1740894228219986,
      "num_tokens": 42458013.0,
      "step": 19380
    },
    {
      "entropy": 5.59863395690918,
      "epoch": 1.9238785232235014,
      "grad_norm": 1.0,
      "learning_rate": 0.0004633275674977937,
      "loss": 5.1276,
      "mean_token_accuracy": 0.16924300342798232,
      "num_tokens": 42469807.0,
      "step": 19385
    },
    {
      "entropy": 5.503483772277832,
      "epoch": 1.9243747518856689,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004633081775642964,
      "loss": 5.0564,
      "mean_token_accuracy": 0.17505667060613633,
      "num_tokens": 42479240.0,
      "step": 19390
    },
    {
      "entropy": 5.5568524360656735,
      "epoch": 1.9248709805478366,
      "grad_norm": 1.0,
      "learning_rate": 0.00046328878296116196,
      "loss": 5.118,
      "mean_token_accuracy": 0.17725289314985276,
      "num_tokens": 42489558.0,
      "step": 19395
    },
    {
      "entropy": 5.608658218383789,
      "epoch": 1.925367209210004,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046326938368887146,
      "loss": 5.1321,
      "mean_token_accuracy": 0.17715611010789872,
      "num_tokens": 42501252.0,
      "step": 19400
    },
    {
      "entropy": 5.625342082977295,
      "epoch": 1.9258634378721715,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00046324997974790596,
      "loss": 5.1558,
      "mean_token_accuracy": 0.1770201161503792,
      "num_tokens": 42512330.0,
      "step": 19405
    },
    {
      "entropy": 5.582589769363404,
      "epoch": 1.926359666534339,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004632305711387466,
      "loss": 5.1488,
      "mean_token_accuracy": 0.17336562871932984,
      "num_tokens": 42525216.0,
      "step": 19410
    },
    {
      "entropy": 5.603672742843628,
      "epoch": 1.9268558951965065,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046321115786187466,
      "loss": 5.0941,
      "mean_token_accuracy": 0.17508437484502792,
      "num_tokens": 42537047.0,
      "step": 19415
    },
    {
      "entropy": 5.593495798110962,
      "epoch": 1.927352123858674,
      "grad_norm": 0.953125,
      "learning_rate": 0.00046319173991777153,
      "loss": 5.0749,
      "mean_token_accuracy": 0.17257878929376602,
      "num_tokens": 42549784.0,
      "step": 19420
    },
    {
      "entropy": 5.562648630142212,
      "epoch": 1.9278483525208416,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004631723173069188,
      "loss": 5.0714,
      "mean_token_accuracy": 0.17260258197784423,
      "num_tokens": 42561079.0,
      "step": 19425
    },
    {
      "entropy": 5.558985090255737,
      "epoch": 1.928344581183009,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004631528900297981,
      "loss": 5.1354,
      "mean_token_accuracy": 0.17199122309684753,
      "num_tokens": 42572531.0,
      "step": 19430
    },
    {
      "entropy": 5.562539577484131,
      "epoch": 1.9288408098451768,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046313345808689113,
      "loss": 5.0504,
      "mean_token_accuracy": 0.1781085506081581,
      "num_tokens": 42583836.0,
      "step": 19435
    },
    {
      "entropy": 5.622388315200806,
      "epoch": 1.9293370385073443,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004631140214786798,
      "loss": 5.1592,
      "mean_token_accuracy": 0.17296061366796495,
      "num_tokens": 42595338.0,
      "step": 19440
    },
    {
      "entropy": 5.5818993091583256,
      "epoch": 1.9298332671695118,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046309458020564606,
      "loss": 5.0962,
      "mean_token_accuracy": 0.17329725325107576,
      "num_tokens": 42606804.0,
      "step": 19445
    },
    {
      "entropy": 5.544586896896362,
      "epoch": 1.9303294958316792,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046307513426827214,
      "loss": 5.0295,
      "mean_token_accuracy": 0.18118116408586502,
      "num_tokens": 42616987.0,
      "step": 19450
    },
    {
      "entropy": 5.569060659408569,
      "epoch": 1.9308257244938467,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046305568366704006,
      "loss": 5.1048,
      "mean_token_accuracy": 0.1745738834142685,
      "num_tokens": 42627633.0,
      "step": 19455
    },
    {
      "entropy": 5.592783117294312,
      "epoch": 1.9313219531560142,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004630362284024322,
      "loss": 5.0653,
      "mean_token_accuracy": 0.1722562924027443,
      "num_tokens": 42638802.0,
      "step": 19460
    },
    {
      "entropy": 5.5772136688232425,
      "epoch": 1.9318181818181817,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046301676847493116,
      "loss": 5.061,
      "mean_token_accuracy": 0.1808040753006935,
      "num_tokens": 42650802.0,
      "step": 19465
    },
    {
      "entropy": 5.551653146743774,
      "epoch": 1.9323144104803494,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046299730388501935,
      "loss": 5.032,
      "mean_token_accuracy": 0.1813851073384285,
      "num_tokens": 42660290.0,
      "step": 19470
    },
    {
      "entropy": 5.590027475357056,
      "epoch": 1.9328106391425168,
      "grad_norm": 0.921875,
      "learning_rate": 0.00046297783463317957,
      "loss": 5.1203,
      "mean_token_accuracy": 0.17696295827627181,
      "num_tokens": 42671182.0,
      "step": 19475
    },
    {
      "entropy": 5.60194354057312,
      "epoch": 1.9333068678046845,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004629583607198945,
      "loss": 5.1239,
      "mean_token_accuracy": 0.17695219218730926,
      "num_tokens": 42682499.0,
      "step": 19480
    },
    {
      "entropy": 5.540981197357178,
      "epoch": 1.933803096466852,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046293888214564704,
      "loss": 5.1313,
      "mean_token_accuracy": 0.17261911928653717,
      "num_tokens": 42693728.0,
      "step": 19485
    },
    {
      "entropy": 5.573365783691406,
      "epoch": 1.9342993251290195,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046291939891092036,
      "loss": 5.1155,
      "mean_token_accuracy": 0.1752532184123993,
      "num_tokens": 42704877.0,
      "step": 19490
    },
    {
      "entropy": 5.617432928085327,
      "epoch": 1.934795553791187,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004628999110161975,
      "loss": 5.1425,
      "mean_token_accuracy": 0.16973139494657516,
      "num_tokens": 42715787.0,
      "step": 19495
    },
    {
      "entropy": 5.596118021011352,
      "epoch": 1.9352917824533544,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046288041846196174,
      "loss": 5.1041,
      "mean_token_accuracy": 0.1750999927520752,
      "num_tokens": 42727557.0,
      "step": 19500
    },
    {
      "entropy": 5.606228828430176,
      "epoch": 1.935788011115522,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046286092124869647,
      "loss": 5.1045,
      "mean_token_accuracy": 0.17468319982290267,
      "num_tokens": 42737540.0,
      "step": 19505
    },
    {
      "entropy": 5.637428569793701,
      "epoch": 1.9362842397776896,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046284141937688513,
      "loss": 5.1319,
      "mean_token_accuracy": 0.1786578565835953,
      "num_tokens": 42748716.0,
      "step": 19510
    },
    {
      "entropy": 5.673902893066407,
      "epoch": 1.936780468439857,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046282191284701123,
      "loss": 5.1662,
      "mean_token_accuracy": 0.17017856389284133,
      "num_tokens": 42758991.0,
      "step": 19515
    },
    {
      "entropy": 5.653537130355835,
      "epoch": 1.9372766971020248,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004628024016595587,
      "loss": 5.1422,
      "mean_token_accuracy": 0.1710558831691742,
      "num_tokens": 42771105.0,
      "step": 19520
    },
    {
      "entropy": 5.55591344833374,
      "epoch": 1.9377729257641922,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00046278288581501136,
      "loss": 5.103,
      "mean_token_accuracy": 0.17486774027347565,
      "num_tokens": 42783317.0,
      "step": 19525
    },
    {
      "entropy": 5.652013731002808,
      "epoch": 1.9382691544263597,
      "grad_norm": 0.921875,
      "learning_rate": 0.00046276336531385305,
      "loss": 5.2283,
      "mean_token_accuracy": 0.1676255762577057,
      "num_tokens": 42794998.0,
      "step": 19530
    },
    {
      "entropy": 5.6051544666290285,
      "epoch": 1.9387653830885272,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046274384015656785,
      "loss": 5.1016,
      "mean_token_accuracy": 0.18168456107378006,
      "num_tokens": 42806197.0,
      "step": 19535
    },
    {
      "entropy": 5.53239369392395,
      "epoch": 1.9392616117506947,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046272431034363997,
      "loss": 5.002,
      "mean_token_accuracy": 0.17452119886875153,
      "num_tokens": 42816522.0,
      "step": 19540
    },
    {
      "entropy": 5.553054761886597,
      "epoch": 1.9397578404128621,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004627047758755538,
      "loss": 5.0123,
      "mean_token_accuracy": 0.18434111177921295,
      "num_tokens": 42826003.0,
      "step": 19545
    },
    {
      "entropy": 5.59314866065979,
      "epoch": 1.9402540690750296,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004626852367527935,
      "loss": 5.1002,
      "mean_token_accuracy": 0.17019644677639006,
      "num_tokens": 42836346.0,
      "step": 19550
    },
    {
      "entropy": 5.5753350257873535,
      "epoch": 1.9407502977371973,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004626656929758438,
      "loss": 5.1586,
      "mean_token_accuracy": 0.17343145310878755,
      "num_tokens": 42846296.0,
      "step": 19555
    },
    {
      "entropy": 5.672791290283203,
      "epoch": 1.9412465263993648,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004626461445451894,
      "loss": 5.1485,
      "mean_token_accuracy": 0.17419494092464446,
      "num_tokens": 42856323.0,
      "step": 19560
    },
    {
      "entropy": 5.6711437702178955,
      "epoch": 1.9417427550615325,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004626265914613149,
      "loss": 5.1654,
      "mean_token_accuracy": 0.1695869415998459,
      "num_tokens": 42866937.0,
      "step": 19565
    },
    {
      "entropy": 5.593677425384522,
      "epoch": 1.9422389837237,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004626070337247052,
      "loss": 5.1287,
      "mean_token_accuracy": 0.1693687006831169,
      "num_tokens": 42877194.0,
      "step": 19570
    },
    {
      "entropy": 5.594156122207641,
      "epoch": 1.9427352123858674,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004625874713358453,
      "loss": 5.129,
      "mean_token_accuracy": 0.17129995226860045,
      "num_tokens": 42887741.0,
      "step": 19575
    },
    {
      "entropy": 5.572790956497192,
      "epoch": 1.943231441048035,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046256790429522035,
      "loss": 5.1079,
      "mean_token_accuracy": 0.18464816659688948,
      "num_tokens": 42898195.0,
      "step": 19580
    },
    {
      "entropy": 5.655848360061645,
      "epoch": 1.9437276697102024,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004625483326033155,
      "loss": 5.149,
      "mean_token_accuracy": 0.17260519415140152,
      "num_tokens": 42908852.0,
      "step": 19585
    },
    {
      "entropy": 5.556072950363159,
      "epoch": 1.9442238983723699,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004625287562606162,
      "loss": 5.0419,
      "mean_token_accuracy": 0.1763271987438202,
      "num_tokens": 42920166.0,
      "step": 19590
    },
    {
      "entropy": 5.607384920120239,
      "epoch": 1.9447201270345376,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004625091752676078,
      "loss": 5.1226,
      "mean_token_accuracy": 0.1745381087064743,
      "num_tokens": 42931677.0,
      "step": 19595
    },
    {
      "entropy": 5.584395074844361,
      "epoch": 1.945216355696705,
      "grad_norm": 1.0625,
      "learning_rate": 0.000462489589624776,
      "loss": 5.107,
      "mean_token_accuracy": 0.17392319291830063,
      "num_tokens": 42942346.0,
      "step": 19600
    },
    {
      "entropy": 5.475983476638794,
      "epoch": 1.9457125843588727,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004624699993326063,
      "loss": 5.0702,
      "mean_token_accuracy": 0.1802891030907631,
      "num_tokens": 42953821.0,
      "step": 19605
    },
    {
      "entropy": 5.543852710723877,
      "epoch": 1.9462088130210402,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046245040439158456,
      "loss": 5.0678,
      "mean_token_accuracy": 0.17612649202346803,
      "num_tokens": 42964358.0,
      "step": 19610
    },
    {
      "entropy": 5.573706245422363,
      "epoch": 1.9467050416832077,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00046243080480219673,
      "loss": 5.0572,
      "mean_token_accuracy": 0.17840345799922944,
      "num_tokens": 42976459.0,
      "step": 19615
    },
    {
      "entropy": 5.641165781021118,
      "epoch": 1.9472012703453752,
      "grad_norm": 1.0,
      "learning_rate": 0.0004624112005649288,
      "loss": 5.2039,
      "mean_token_accuracy": 0.16884401738643645,
      "num_tokens": 42988603.0,
      "step": 19620
    },
    {
      "entropy": 5.568244600296021,
      "epoch": 1.9476974990075426,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046239159168026694,
      "loss": 5.1193,
      "mean_token_accuracy": 0.1738400474190712,
      "num_tokens": 43000262.0,
      "step": 19625
    },
    {
      "entropy": 5.521918392181396,
      "epoch": 1.94819372766971,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004623719781486974,
      "loss": 5.0561,
      "mean_token_accuracy": 0.1762985348701477,
      "num_tokens": 43009363.0,
      "step": 19630
    },
    {
      "entropy": 5.628265619277954,
      "epoch": 1.9486899563318776,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046235235997070664,
      "loss": 5.1252,
      "mean_token_accuracy": 0.17919848263263702,
      "num_tokens": 43019893.0,
      "step": 19635
    },
    {
      "entropy": 5.5354039669036865,
      "epoch": 1.9491861849940453,
      "grad_norm": 0.984375,
      "learning_rate": 0.000462332737146781,
      "loss": 5.0287,
      "mean_token_accuracy": 0.18238697797060013,
      "num_tokens": 43031152.0,
      "step": 19640
    },
    {
      "entropy": 5.574510145187378,
      "epoch": 1.9496824136562128,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004623131096774071,
      "loss": 5.1669,
      "mean_token_accuracy": 0.170855288207531,
      "num_tokens": 43041247.0,
      "step": 19645
    },
    {
      "entropy": 5.628740358352661,
      "epoch": 1.9501786423183805,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004622934775630717,
      "loss": 5.0827,
      "mean_token_accuracy": 0.17058514654636384,
      "num_tokens": 43052065.0,
      "step": 19650
    },
    {
      "entropy": 5.5691609382629395,
      "epoch": 1.950674870980548,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00046227384080426163,
      "loss": 5.12,
      "mean_token_accuracy": 0.17122100442647933,
      "num_tokens": 43063217.0,
      "step": 19655
    },
    {
      "entropy": 5.553304052352905,
      "epoch": 1.9511710996427154,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004622541994014639,
      "loss": 5.1243,
      "mean_token_accuracy": 0.1771386981010437,
      "num_tokens": 43074463.0,
      "step": 19660
    },
    {
      "entropy": 5.691059684753418,
      "epoch": 1.9516673283048829,
      "grad_norm": 0.875,
      "learning_rate": 0.0004622345533551655,
      "loss": 5.1616,
      "mean_token_accuracy": 0.16897969990968703,
      "num_tokens": 43086004.0,
      "step": 19665
    },
    {
      "entropy": 5.620903253555298,
      "epoch": 1.9521635569670504,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004622149026658537,
      "loss": 5.0595,
      "mean_token_accuracy": 0.18073871582746506,
      "num_tokens": 43096973.0,
      "step": 19670
    },
    {
      "entropy": 5.6457103252410885,
      "epoch": 1.9526597856292178,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004621952473340156,
      "loss": 5.1584,
      "mean_token_accuracy": 0.1755485862493515,
      "num_tokens": 43107337.0,
      "step": 19675
    },
    {
      "entropy": 5.611739778518677,
      "epoch": 1.9531560142913855,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004621755873601388,
      "loss": 5.0806,
      "mean_token_accuracy": 0.18096780627965928,
      "num_tokens": 43118668.0,
      "step": 19680
    },
    {
      "entropy": 5.654886245727539,
      "epoch": 1.953652242953553,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004621559227447108,
      "loss": 5.1876,
      "mean_token_accuracy": 0.1683372363448143,
      "num_tokens": 43129030.0,
      "step": 19685
    },
    {
      "entropy": 5.520848894119263,
      "epoch": 1.9541484716157205,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004621362534882192,
      "loss": 5.0365,
      "mean_token_accuracy": 0.1843307673931122,
      "num_tokens": 43139120.0,
      "step": 19690
    },
    {
      "entropy": 5.572016954421997,
      "epoch": 1.9546447002778882,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004621165795911517,
      "loss": 5.122,
      "mean_token_accuracy": 0.17558340430259706,
      "num_tokens": 43152390.0,
      "step": 19695
    },
    {
      "entropy": 5.554745244979858,
      "epoch": 1.9551409289400556,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046209690105399623,
      "loss": 4.9344,
      "mean_token_accuracy": 0.18695660680532455,
      "num_tokens": 43162230.0,
      "step": 19700
    },
    {
      "entropy": 5.505560779571534,
      "epoch": 1.9556371576022231,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046207721787724085,
      "loss": 5.0673,
      "mean_token_accuracy": 0.17772364467382432,
      "num_tokens": 43172751.0,
      "step": 19705
    },
    {
      "entropy": 5.642329216003418,
      "epoch": 1.9561333862643906,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046205753006137347,
      "loss": 5.152,
      "mean_token_accuracy": 0.1744347706437111,
      "num_tokens": 43183835.0,
      "step": 19710
    },
    {
      "entropy": 5.65907917022705,
      "epoch": 1.956629614926558,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046203783760688245,
      "loss": 5.0485,
      "mean_token_accuracy": 0.1729263558983803,
      "num_tokens": 43193178.0,
      "step": 19715
    },
    {
      "entropy": 5.583413743972779,
      "epoch": 1.9571258435887255,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004620181405142561,
      "loss": 5.1988,
      "mean_token_accuracy": 0.1671207293868065,
      "num_tokens": 43204968.0,
      "step": 19720
    },
    {
      "entropy": 5.561521339416504,
      "epoch": 1.9576220722508932,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004619984387839829,
      "loss": 5.2202,
      "mean_token_accuracy": 0.16834264099597931,
      "num_tokens": 43214916.0,
      "step": 19725
    },
    {
      "entropy": 5.711319255828857,
      "epoch": 1.9581183009130607,
      "grad_norm": 0.90625,
      "learning_rate": 0.00046197873241655133,
      "loss": 5.2134,
      "mean_token_accuracy": 0.16977116018533706,
      "num_tokens": 43226374.0,
      "step": 19730
    },
    {
      "entropy": 5.58314266204834,
      "epoch": 1.9586145295752284,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004619590214124501,
      "loss": 5.0853,
      "mean_token_accuracy": 0.1738823115825653,
      "num_tokens": 43237296.0,
      "step": 19735
    },
    {
      "entropy": 5.66984281539917,
      "epoch": 1.9591107582373959,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046193930577216797,
      "loss": 5.2423,
      "mean_token_accuracy": 0.1598547577857971,
      "num_tokens": 43248220.0,
      "step": 19740
    },
    {
      "entropy": 5.560368251800537,
      "epoch": 1.9596069868995634,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004619195854961939,
      "loss": 5.0149,
      "mean_token_accuracy": 0.18297584503889083,
      "num_tokens": 43259913.0,
      "step": 19745
    },
    {
      "entropy": 5.622790813446045,
      "epoch": 1.9601032155617308,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00046189986058501687,
      "loss": 5.1485,
      "mean_token_accuracy": 0.17128514349460602,
      "num_tokens": 43272317.0,
      "step": 19750
    },
    {
      "entropy": 5.491859865188599,
      "epoch": 1.9605994442238983,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000461880131039126,
      "loss": 5.0153,
      "mean_token_accuracy": 0.18247572034597398,
      "num_tokens": 43283170.0,
      "step": 19755
    },
    {
      "entropy": 5.536263084411621,
      "epoch": 1.9610956728860658,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004618603968590107,
      "loss": 5.1072,
      "mean_token_accuracy": 0.17903425693511962,
      "num_tokens": 43293899.0,
      "step": 19760
    },
    {
      "entropy": 5.605640316009522,
      "epoch": 1.9615919015482333,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046184065804516007,
      "loss": 5.1354,
      "mean_token_accuracy": 0.17757341265678406,
      "num_tokens": 43305638.0,
      "step": 19765
    },
    {
      "entropy": 5.64197883605957,
      "epoch": 1.962088130210401,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004618209145980638,
      "loss": 5.2056,
      "mean_token_accuracy": 0.16710670292377472,
      "num_tokens": 43315900.0,
      "step": 19770
    },
    {
      "entropy": 5.67323489189148,
      "epoch": 1.9625843588725684,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004618011665182114,
      "loss": 5.272,
      "mean_token_accuracy": 0.1656080037355423,
      "num_tokens": 43328464.0,
      "step": 19775
    },
    {
      "entropy": 5.53949465751648,
      "epoch": 1.9630805875347361,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004617814138060926,
      "loss": 4.9582,
      "mean_token_accuracy": 0.18672350347042083,
      "num_tokens": 43339303.0,
      "step": 19780
    },
    {
      "entropy": 5.621024036407471,
      "epoch": 1.9635768161969036,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004617616564621972,
      "loss": 5.1992,
      "mean_token_accuracy": 0.17171578705310822,
      "num_tokens": 43349269.0,
      "step": 19785
    },
    {
      "entropy": 5.5267102241516115,
      "epoch": 1.964073044859071,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00046174189448701516,
      "loss": 5.0495,
      "mean_token_accuracy": 0.17921780347824096,
      "num_tokens": 43361090.0,
      "step": 19790
    },
    {
      "entropy": 5.556002187728882,
      "epoch": 1.9645692735212386,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004617221278810365,
      "loss": 5.0389,
      "mean_token_accuracy": 0.174630868434906,
      "num_tokens": 43373448.0,
      "step": 19795
    },
    {
      "entropy": 5.671497583389282,
      "epoch": 1.965065502183406,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046170235664475147,
      "loss": 5.1783,
      "mean_token_accuracy": 0.17180072665214538,
      "num_tokens": 43386202.0,
      "step": 19800
    },
    {
      "entropy": 5.580495452880859,
      "epoch": 1.9655617308455735,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046168258077865026,
      "loss": 5.09,
      "mean_token_accuracy": 0.1797773450613022,
      "num_tokens": 43396363.0,
      "step": 19805
    },
    {
      "entropy": 5.556813192367554,
      "epoch": 1.9660579595077412,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004616628002832233,
      "loss": 5.0049,
      "mean_token_accuracy": 0.18110644817352295,
      "num_tokens": 43406076.0,
      "step": 19810
    },
    {
      "entropy": 5.587572956085205,
      "epoch": 1.9665541881699087,
      "grad_norm": 1.0,
      "learning_rate": 0.00046164301515896124,
      "loss": 5.1448,
      "mean_token_accuracy": 0.17329322695732116,
      "num_tokens": 43416226.0,
      "step": 19815
    },
    {
      "entropy": 5.553575944900513,
      "epoch": 1.9670504168320764,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004616232254063544,
      "loss": 4.9986,
      "mean_token_accuracy": 0.18101046234369278,
      "num_tokens": 43427364.0,
      "step": 19820
    },
    {
      "entropy": 5.598277282714844,
      "epoch": 1.9675466454942439,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046160343102589384,
      "loss": 5.1203,
      "mean_token_accuracy": 0.17934240251779557,
      "num_tokens": 43437854.0,
      "step": 19825
    },
    {
      "entropy": 5.573452568054199,
      "epoch": 1.9680428741564113,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004615836320180702,
      "loss": 5.1511,
      "mean_token_accuracy": 0.17664049863815307,
      "num_tokens": 43449370.0,
      "step": 19830
    },
    {
      "entropy": 5.559548044204712,
      "epoch": 1.9685391028185788,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046156382838337455,
      "loss": 5.0941,
      "mean_token_accuracy": 0.1821209594607353,
      "num_tokens": 43461631.0,
      "step": 19835
    },
    {
      "entropy": 5.607298088073731,
      "epoch": 1.9690353314807463,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046154402012229807,
      "loss": 5.0927,
      "mean_token_accuracy": 0.17239862084388732,
      "num_tokens": 43472357.0,
      "step": 19840
    },
    {
      "entropy": 5.5414710521698,
      "epoch": 1.9695315601429138,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046152420723533175,
      "loss": 5.054,
      "mean_token_accuracy": 0.17707743346691132,
      "num_tokens": 43483642.0,
      "step": 19845
    },
    {
      "entropy": 5.6295366287231445,
      "epoch": 1.9700277888050812,
      "grad_norm": 1.125,
      "learning_rate": 0.00046150438972296706,
      "loss": 5.1349,
      "mean_token_accuracy": 0.17591917067766188,
      "num_tokens": 43495136.0,
      "step": 19850
    },
    {
      "entropy": 5.552084064483642,
      "epoch": 1.970524017467249,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046148456758569536,
      "loss": 5.1124,
      "mean_token_accuracy": 0.17267705798149108,
      "num_tokens": 43506376.0,
      "step": 19855
    },
    {
      "entropy": 5.558958959579468,
      "epoch": 1.9710202461294164,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046146474082400823,
      "loss": 5.0698,
      "mean_token_accuracy": 0.176774001121521,
      "num_tokens": 43516658.0,
      "step": 19860
    },
    {
      "entropy": 5.571771097183228,
      "epoch": 1.971516474791584,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046144490943839735,
      "loss": 5.0464,
      "mean_token_accuracy": 0.1817093849182129,
      "num_tokens": 43526907.0,
      "step": 19865
    },
    {
      "entropy": 5.655218315124512,
      "epoch": 1.9720127034537516,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004614250734293544,
      "loss": 5.1376,
      "mean_token_accuracy": 0.1728274032473564,
      "num_tokens": 43537739.0,
      "step": 19870
    },
    {
      "entropy": 5.522761249542237,
      "epoch": 1.972508932115919,
      "grad_norm": 1.0,
      "learning_rate": 0.00046140523279737146,
      "loss": 5.0943,
      "mean_token_accuracy": 0.17799045890569687,
      "num_tokens": 43549240.0,
      "step": 19875
    },
    {
      "entropy": 5.570582914352417,
      "epoch": 1.9730051607780865,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004613853875429403,
      "loss": 5.0538,
      "mean_token_accuracy": 0.17478798627853392,
      "num_tokens": 43559857.0,
      "step": 19880
    },
    {
      "entropy": 5.538764190673828,
      "epoch": 1.973501389440254,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004613655376665532,
      "loss": 5.042,
      "mean_token_accuracy": 0.1760262057185173,
      "num_tokens": 43570262.0,
      "step": 19885
    },
    {
      "entropy": 5.63384780883789,
      "epoch": 1.9739976181024215,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004613456831687024,
      "loss": 5.1151,
      "mean_token_accuracy": 0.17173005193471907,
      "num_tokens": 43581965.0,
      "step": 19890
    },
    {
      "entropy": 5.562789487838745,
      "epoch": 1.9744938467645892,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046132582404988004,
      "loss": 5.0769,
      "mean_token_accuracy": 0.17880486696958542,
      "num_tokens": 43592728.0,
      "step": 19895
    },
    {
      "entropy": 5.594052314758301,
      "epoch": 1.9749900754267566,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046130596031057885,
      "loss": 5.0441,
      "mean_token_accuracy": 0.1728651061654091,
      "num_tokens": 43602437.0,
      "step": 19900
    },
    {
      "entropy": 5.598533630371094,
      "epoch": 1.9754863040889243,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046128609195129135,
      "loss": 5.1628,
      "mean_token_accuracy": 0.17248115986585616,
      "num_tokens": 43613115.0,
      "step": 19905
    },
    {
      "entropy": 5.6184135437011715,
      "epoch": 1.9759825327510918,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004612662189725101,
      "loss": 5.1459,
      "mean_token_accuracy": 0.1716393396258354,
      "num_tokens": 43623700.0,
      "step": 19910
    },
    {
      "entropy": 5.6604790687561035,
      "epoch": 1.9764787614132593,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000461246341374728,
      "loss": 5.1604,
      "mean_token_accuracy": 0.1706886038184166,
      "num_tokens": 43633712.0,
      "step": 19915
    },
    {
      "entropy": 5.624731063842773,
      "epoch": 1.9769749900754268,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046122645915843796,
      "loss": 5.1894,
      "mean_token_accuracy": 0.17017436772584915,
      "num_tokens": 43643705.0,
      "step": 19920
    },
    {
      "entropy": 5.598883104324341,
      "epoch": 1.9774712187375942,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046120657232413305,
      "loss": 5.1417,
      "mean_token_accuracy": 0.1766065239906311,
      "num_tokens": 43655376.0,
      "step": 19925
    },
    {
      "entropy": 5.557476997375488,
      "epoch": 1.9779674473997617,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004611866808723064,
      "loss": 5.0388,
      "mean_token_accuracy": 0.1845523163676262,
      "num_tokens": 43666329.0,
      "step": 19930
    },
    {
      "entropy": 5.524384498596191,
      "epoch": 1.9784636760619292,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00046116678480345117,
      "loss": 5.0521,
      "mean_token_accuracy": 0.17806633710861205,
      "num_tokens": 43676415.0,
      "step": 19935
    },
    {
      "entropy": 5.523144102096557,
      "epoch": 1.9789599047240969,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00046114688411806083,
      "loss": 5.0676,
      "mean_token_accuracy": 0.18042845875024796,
      "num_tokens": 43686997.0,
      "step": 19940
    },
    {
      "entropy": 5.540007734298706,
      "epoch": 1.9794561333862644,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00046112697881662895,
      "loss": 5.0385,
      "mean_token_accuracy": 0.1834940642118454,
      "num_tokens": 43697342.0,
      "step": 19945
    },
    {
      "entropy": 5.652699279785156,
      "epoch": 1.979952362048432,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00046110706889964905,
      "loss": 5.1437,
      "mean_token_accuracy": 0.17826791256666183,
      "num_tokens": 43707876.0,
      "step": 19950
    },
    {
      "entropy": 5.57138090133667,
      "epoch": 1.9804485907105995,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004610871543676149,
      "loss": 4.9955,
      "mean_token_accuracy": 0.1790217474102974,
      "num_tokens": 43718268.0,
      "step": 19955
    },
    {
      "entropy": 5.569626045227051,
      "epoch": 1.980944819372767,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00046106723522102024,
      "loss": 5.1533,
      "mean_token_accuracy": 0.16895215064287186,
      "num_tokens": 43729853.0,
      "step": 19960
    },
    {
      "entropy": 5.580903387069702,
      "epoch": 1.9814410480349345,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046104731146035904,
      "loss": 5.1382,
      "mean_token_accuracy": 0.17228945195674897,
      "num_tokens": 43739886.0,
      "step": 19965
    },
    {
      "entropy": 5.628433465957642,
      "epoch": 1.981937276697102,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004610273830861255,
      "loss": 5.1118,
      "mean_token_accuracy": 0.17257425785064698,
      "num_tokens": 43750536.0,
      "step": 19970
    },
    {
      "entropy": 5.616104650497436,
      "epoch": 1.9824335053592694,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004610074500988137,
      "loss": 5.0988,
      "mean_token_accuracy": 0.17899229228496552,
      "num_tokens": 43763192.0,
      "step": 19975
    },
    {
      "entropy": 5.618440294265747,
      "epoch": 1.9829297340214371,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004609875124989179,
      "loss": 5.1365,
      "mean_token_accuracy": 0.1766981303691864,
      "num_tokens": 43774117.0,
      "step": 19980
    },
    {
      "entropy": 5.533320808410645,
      "epoch": 1.9834259626836046,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004609675702869326,
      "loss": 5.0666,
      "mean_token_accuracy": 0.1747085228562355,
      "num_tokens": 43784708.0,
      "step": 19985
    },
    {
      "entropy": 5.573682260513306,
      "epoch": 1.983922191345772,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004609476234633523,
      "loss": 5.1013,
      "mean_token_accuracy": 0.1803373023867607,
      "num_tokens": 43796767.0,
      "step": 19990
    },
    {
      "entropy": 5.636098623275757,
      "epoch": 1.9844184200079398,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046092767202867156,
      "loss": 5.2144,
      "mean_token_accuracy": 0.1675091952085495,
      "num_tokens": 43807203.0,
      "step": 19995
    },
    {
      "entropy": 5.572779273986816,
      "epoch": 1.9849146486701073,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046090771598338525,
      "loss": 5.1121,
      "mean_token_accuracy": 0.1792059138417244,
      "num_tokens": 43817123.0,
      "step": 20000
    },
    {
      "entropy": 5.598117160797119,
      "epoch": 1.9854108773322747,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004608877553279881,
      "loss": 5.089,
      "mean_token_accuracy": 0.18206135779619217,
      "num_tokens": 43828625.0,
      "step": 20005
    },
    {
      "entropy": 5.574577045440674,
      "epoch": 1.9859071059944422,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004608677900629753,
      "loss": 5.0506,
      "mean_token_accuracy": 0.17934031486511232,
      "num_tokens": 43840012.0,
      "step": 20010
    },
    {
      "entropy": 5.488186693191528,
      "epoch": 1.9864033346566097,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046084782018884177,
      "loss": 5.0489,
      "mean_token_accuracy": 0.17638782858848573,
      "num_tokens": 43851253.0,
      "step": 20015
    },
    {
      "entropy": 5.643233299255371,
      "epoch": 1.9868995633187772,
      "grad_norm": 0.90625,
      "learning_rate": 0.00046082784570608266,
      "loss": 5.1645,
      "mean_token_accuracy": 0.17455788254737853,
      "num_tokens": 43864389.0,
      "step": 20020
    },
    {
      "entropy": 5.639037036895752,
      "epoch": 1.9873957919809448,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046080786661519354,
      "loss": 5.1562,
      "mean_token_accuracy": 0.17175842076539993,
      "num_tokens": 43876446.0,
      "step": 20025
    },
    {
      "entropy": 5.603530979156494,
      "epoch": 1.9878920206431123,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004607878829166696,
      "loss": 5.1104,
      "mean_token_accuracy": 0.16865846365690232,
      "num_tokens": 43888318.0,
      "step": 20030
    },
    {
      "entropy": 5.520271158218383,
      "epoch": 1.98838824930528,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046076789461100653,
      "loss": 5.05,
      "mean_token_accuracy": 0.18364468663930894,
      "num_tokens": 43898358.0,
      "step": 20035
    },
    {
      "entropy": 5.591721105575561,
      "epoch": 1.9888844779674475,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004607479016987001,
      "loss": 5.0984,
      "mean_token_accuracy": 0.1757814645767212,
      "num_tokens": 43909831.0,
      "step": 20040
    },
    {
      "entropy": 5.597817754745483,
      "epoch": 1.989380706629615,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004607279041802458,
      "loss": 5.0673,
      "mean_token_accuracy": 0.17361391335725784,
      "num_tokens": 43919648.0,
      "step": 20045
    },
    {
      "entropy": 5.569124174118042,
      "epoch": 1.9898769352917824,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00046070790205613967,
      "loss": 5.12,
      "mean_token_accuracy": 0.17306884825229646,
      "num_tokens": 43930563.0,
      "step": 20050
    },
    {
      "entropy": 5.63607497215271,
      "epoch": 1.99037316395395,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00046068789532687784,
      "loss": 5.1932,
      "mean_token_accuracy": 0.17143940329551696,
      "num_tokens": 43942490.0,
      "step": 20055
    },
    {
      "entropy": 5.561767673492431,
      "epoch": 1.9908693926161174,
      "grad_norm": 1.0,
      "learning_rate": 0.0004606678839929563,
      "loss": 4.9763,
      "mean_token_accuracy": 0.18970065861940383,
      "num_tokens": 43952187.0,
      "step": 20060
    },
    {
      "entropy": 5.522125291824341,
      "epoch": 1.9913656212782849,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004606478680548713,
      "loss": 5.0381,
      "mean_token_accuracy": 0.1813553750514984,
      "num_tokens": 43964203.0,
      "step": 20065
    },
    {
      "entropy": 5.547319936752319,
      "epoch": 1.9918618499404526,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046062784751311917,
      "loss": 5.1192,
      "mean_token_accuracy": 0.17629277259111403,
      "num_tokens": 43974048.0,
      "step": 20070
    },
    {
      "entropy": 5.6360697746276855,
      "epoch": 1.99235807860262,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046060782236819647,
      "loss": 5.1016,
      "mean_token_accuracy": 0.17606257796287536,
      "num_tokens": 43984469.0,
      "step": 20075
    },
    {
      "entropy": 5.588944911956787,
      "epoch": 1.9928543072647877,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004605877926205996,
      "loss": 5.122,
      "mean_token_accuracy": 0.1758631095290184,
      "num_tokens": 43995494.0,
      "step": 20080
    },
    {
      "entropy": 5.58999457359314,
      "epoch": 1.9933505359269552,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004605677582708256,
      "loss": 5.0483,
      "mean_token_accuracy": 0.1770133838057518,
      "num_tokens": 44005869.0,
      "step": 20085
    },
    {
      "entropy": 5.629586315155029,
      "epoch": 1.9938467645891227,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004605477193193709,
      "loss": 5.1065,
      "mean_token_accuracy": 0.172397281229496,
      "num_tokens": 44016956.0,
      "step": 20090
    },
    {
      "entropy": 5.6551614761352536,
      "epoch": 1.9943429932512902,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004605276757667326,
      "loss": 5.1784,
      "mean_token_accuracy": 0.16540496796369553,
      "num_tokens": 44029524.0,
      "step": 20095
    },
    {
      "entropy": 5.574221801757813,
      "epoch": 1.9948392219134576,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004605076276134078,
      "loss": 5.0624,
      "mean_token_accuracy": 0.17908160239458085,
      "num_tokens": 44039523.0,
      "step": 20100
    },
    {
      "entropy": 5.659173631668091,
      "epoch": 1.9953354505756251,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046048757485989345,
      "loss": 5.1743,
      "mean_token_accuracy": 0.16826657056808472,
      "num_tokens": 44050538.0,
      "step": 20105
    },
    {
      "entropy": 5.641814517974853,
      "epoch": 1.9958316792377928,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000460467517506687,
      "loss": 5.1578,
      "mean_token_accuracy": 0.16938179433345796,
      "num_tokens": 44061108.0,
      "step": 20110
    },
    {
      "entropy": 5.551302719116211,
      "epoch": 1.9963279078999603,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004604474555542858,
      "loss": 5.0942,
      "mean_token_accuracy": 0.17868563234806062,
      "num_tokens": 44072706.0,
      "step": 20115
    },
    {
      "entropy": 5.610600852966309,
      "epoch": 1.996824136562128,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004604273890031872,
      "loss": 5.0865,
      "mean_token_accuracy": 0.17227104306221008,
      "num_tokens": 44082697.0,
      "step": 20120
    },
    {
      "entropy": 5.5398736000061035,
      "epoch": 1.9973203652242955,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000460407317853889,
      "loss": 5.0696,
      "mean_token_accuracy": 0.17819233387708663,
      "num_tokens": 44093265.0,
      "step": 20125
    },
    {
      "entropy": 5.562592697143555,
      "epoch": 1.997816593886463,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004603872421068888,
      "loss": 5.0619,
      "mean_token_accuracy": 0.1732928767800331,
      "num_tokens": 44103535.0,
      "step": 20130
    },
    {
      "entropy": 5.568638801574707,
      "epoch": 1.9983128225486304,
      "grad_norm": 1.0,
      "learning_rate": 0.0004603671617626845,
      "loss": 5.0523,
      "mean_token_accuracy": 0.18068394511938096,
      "num_tokens": 44113903.0,
      "step": 20135
    },
    {
      "entropy": 5.587209272384643,
      "epoch": 1.9988090512107979,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00046034707682177393,
      "loss": 5.088,
      "mean_token_accuracy": 0.1831910565495491,
      "num_tokens": 44125148.0,
      "step": 20140
    },
    {
      "entropy": 5.674484491348267,
      "epoch": 1.9993052798729654,
      "grad_norm": 1.0,
      "learning_rate": 0.0004603269872846554,
      "loss": 5.1869,
      "mean_token_accuracy": 0.16739544719457627,
      "num_tokens": 44136563.0,
      "step": 20145
    },
    {
      "entropy": 5.7027617454528805,
      "epoch": 1.9998015085351328,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00046030689315182676,
      "loss": 5.1505,
      "mean_token_accuracy": 0.17384083867073058,
      "num_tokens": 44147521.0,
      "step": 20150
    },
    {
      "entropy": 5.645961761474609,
      "epoch": 2.0002977371973003,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004602867944237866,
      "loss": 5.1377,
      "mean_token_accuracy": 0.1753172382712364,
      "num_tokens": 44158977.0,
      "step": 20155
    },
    {
      "entropy": 5.613375520706176,
      "epoch": 2.0007939658594682,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00046026669110103315,
      "loss": 5.099,
      "mean_token_accuracy": 0.17865435928106307,
      "num_tokens": 44170986.0,
      "step": 20160
    },
    {
      "entropy": 5.684283685684204,
      "epoch": 2.0012901945216357,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046024658318406493,
      "loss": 5.1495,
      "mean_token_accuracy": 0.17643476128578187,
      "num_tokens": 44180936.0,
      "step": 20165
    },
    {
      "entropy": 5.6822834491729735,
      "epoch": 2.001786423183803,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004602264706733807,
      "loss": 5.146,
      "mean_token_accuracy": 0.17399806380271912,
      "num_tokens": 44191741.0,
      "step": 20170
    },
    {
      "entropy": 5.602428531646728,
      "epoch": 2.0022826518459707,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004602063535694791,
      "loss": 5.0262,
      "mean_token_accuracy": 0.17787106484174728,
      "num_tokens": 44202310.0,
      "step": 20175
    },
    {
      "entropy": 5.586825752258301,
      "epoch": 2.002778880508138,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000460186231872859,
      "loss": 5.0745,
      "mean_token_accuracy": 0.18134640008211136,
      "num_tokens": 44213514.0,
      "step": 20180
    },
    {
      "entropy": 5.65569920539856,
      "epoch": 2.0032751091703056,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00046016610558401936,
      "loss": 5.1083,
      "mean_token_accuracy": 0.17185890674591064,
      "num_tokens": 44224718.0,
      "step": 20185
    },
    {
      "entropy": 5.617652320861817,
      "epoch": 2.003771337832473,
      "grad_norm": 0.9375,
      "learning_rate": 0.00046014597470345935,
      "loss": 5.1229,
      "mean_token_accuracy": 0.17804794311523436,
      "num_tokens": 44235999.0,
      "step": 20190
    },
    {
      "entropy": 5.578877592086792,
      "epoch": 2.0042675664946406,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004601258392316781,
      "loss": 4.98,
      "mean_token_accuracy": 0.18120295703411102,
      "num_tokens": 44246323.0,
      "step": 20195
    },
    {
      "entropy": 5.559261894226074,
      "epoch": 2.0047637951568085,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046010569916917496,
      "loss": 5.0216,
      "mean_token_accuracy": 0.17838912308216096,
      "num_tokens": 44256323.0,
      "step": 20200
    },
    {
      "entropy": 5.588315057754516,
      "epoch": 2.005260023818976,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004600855545164493,
      "loss": 5.0067,
      "mean_token_accuracy": 0.18068246096372603,
      "num_tokens": 44266845.0,
      "step": 20205
    },
    {
      "entropy": 5.624822759628296,
      "epoch": 2.0057562524811434,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004600654052740008,
      "loss": 5.1023,
      "mean_token_accuracy": 0.1778157964348793,
      "num_tokens": 44277861.0,
      "step": 20210
    },
    {
      "entropy": 5.593832492828369,
      "epoch": 2.006252481143311,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004600452514423289,
      "loss": 5.0396,
      "mean_token_accuracy": 0.18204380422830582,
      "num_tokens": 44288511.0,
      "step": 20215
    },
    {
      "entropy": 5.5884888648986815,
      "epoch": 2.0067487098054784,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004600250930219336,
      "loss": 5.1547,
      "mean_token_accuracy": 0.1687192514538765,
      "num_tokens": 44299752.0,
      "step": 20220
    },
    {
      "entropy": 5.593734645843506,
      "epoch": 2.007244938467646,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046000493001331465,
      "loss": 5.0004,
      "mean_token_accuracy": 0.18091526478528977,
      "num_tokens": 44309443.0,
      "step": 20225
    },
    {
      "entropy": 5.633876085281372,
      "epoch": 2.0077411671298133,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045998476241697214,
      "loss": 5.0247,
      "mean_token_accuracy": 0.18577445447444915,
      "num_tokens": 44320030.0,
      "step": 20230
    },
    {
      "entropy": 5.532174825668335,
      "epoch": 2.008237395791981,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004599645902334061,
      "loss": 4.935,
      "mean_token_accuracy": 0.18527044951915742,
      "num_tokens": 44330077.0,
      "step": 20235
    },
    {
      "entropy": 5.458415699005127,
      "epoch": 2.0087336244541483,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00045994441346311677,
      "loss": 5.0192,
      "mean_token_accuracy": 0.1848942831158638,
      "num_tokens": 44340336.0,
      "step": 20240
    },
    {
      "entropy": 5.585581874847412,
      "epoch": 2.009229853116316,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004599242321066045,
      "loss": 5.009,
      "mean_token_accuracy": 0.18270372301340104,
      "num_tokens": 44350704.0,
      "step": 20245
    },
    {
      "entropy": 5.635739755630493,
      "epoch": 2.0097260817784837,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045990404616436976,
      "loss": 5.0452,
      "mean_token_accuracy": 0.18373702913522721,
      "num_tokens": 44360541.0,
      "step": 20250
    },
    {
      "entropy": 5.4952953338623045,
      "epoch": 2.010222310440651,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004598838556369131,
      "loss": 4.9566,
      "mean_token_accuracy": 0.18585868179798126,
      "num_tokens": 44372132.0,
      "step": 20255
    },
    {
      "entropy": 5.505873441696167,
      "epoch": 2.0107185391028186,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00045986366052473525,
      "loss": 4.9553,
      "mean_token_accuracy": 0.1907545283436775,
      "num_tokens": 44383594.0,
      "step": 20260
    },
    {
      "entropy": 5.52165994644165,
      "epoch": 2.011214767764986,
      "grad_norm": 1.125,
      "learning_rate": 0.0004598434608283369,
      "loss": 4.98,
      "mean_token_accuracy": 0.18184978663921356,
      "num_tokens": 44394101.0,
      "step": 20265
    },
    {
      "entropy": 5.632853412628174,
      "epoch": 2.0117109964271536,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004598232565482191,
      "loss": 5.0892,
      "mean_token_accuracy": 0.1794629603624344,
      "num_tokens": 44405670.0,
      "step": 20270
    },
    {
      "entropy": 5.620148849487305,
      "epoch": 2.012207225089321,
      "grad_norm": 0.9375,
      "learning_rate": 0.00045980304768488274,
      "loss": 5.0422,
      "mean_token_accuracy": 0.1825747773051262,
      "num_tokens": 44417748.0,
      "step": 20275
    },
    {
      "entropy": 5.638302278518677,
      "epoch": 2.0127034537514885,
      "grad_norm": 0.984375,
      "learning_rate": 0.00045978283423882906,
      "loss": 5.0334,
      "mean_token_accuracy": 0.17779522985219956,
      "num_tokens": 44429648.0,
      "step": 20280
    },
    {
      "entropy": 5.541291809082031,
      "epoch": 2.0131996824136564,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045976261621055915,
      "loss": 5.0155,
      "mean_token_accuracy": 0.1786918967962265,
      "num_tokens": 44440867.0,
      "step": 20285
    },
    {
      "entropy": 5.561808490753174,
      "epoch": 2.013695911075824,
      "grad_norm": 1.0,
      "learning_rate": 0.00045974239360057457,
      "loss": 5.0633,
      "mean_token_accuracy": 0.1781747654080391,
      "num_tokens": 44452981.0,
      "step": 20290
    },
    {
      "entropy": 5.58002667427063,
      "epoch": 2.0141921397379914,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045972216640937676,
      "loss": 4.9991,
      "mean_token_accuracy": 0.1874930053949356,
      "num_tokens": 44464243.0,
      "step": 20295
    },
    {
      "entropy": 5.6033953666687015,
      "epoch": 2.014688368400159,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004597019346374672,
      "loss": 5.0875,
      "mean_token_accuracy": 0.1769336000084877,
      "num_tokens": 44474376.0,
      "step": 20300
    },
    {
      "entropy": 5.548331308364868,
      "epoch": 2.0151845970623263,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004596816982853477,
      "loss": 4.9697,
      "mean_token_accuracy": 0.1939048945903778,
      "num_tokens": 44485686.0,
      "step": 20305
    },
    {
      "entropy": 5.498635196685791,
      "epoch": 2.015680825724494,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045966145735352,
      "loss": 4.9881,
      "mean_token_accuracy": 0.18503205478191376,
      "num_tokens": 44496096.0,
      "step": 20310
    },
    {
      "entropy": 5.633474683761596,
      "epoch": 2.0161770543866613,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045964121184248604,
      "loss": 5.0861,
      "mean_token_accuracy": 0.17507048845291137,
      "num_tokens": 44507555.0,
      "step": 20315
    },
    {
      "entropy": 5.5817920684814455,
      "epoch": 2.0166732830488288,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004596209617527479,
      "loss": 5.0544,
      "mean_token_accuracy": 0.177122001349926,
      "num_tokens": 44519288.0,
      "step": 20320
    },
    {
      "entropy": 5.625044870376587,
      "epoch": 2.0171695117109962,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004596007070848078,
      "loss": 5.0913,
      "mean_token_accuracy": 0.1673508882522583,
      "num_tokens": 44530146.0,
      "step": 20325
    },
    {
      "entropy": 5.62387170791626,
      "epoch": 2.017665740373164,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045958044783916795,
      "loss": 5.0636,
      "mean_token_accuracy": 0.18019708693027497,
      "num_tokens": 44541411.0,
      "step": 20330
    },
    {
      "entropy": 5.5671141147613525,
      "epoch": 2.0181619690353316,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004595601840163306,
      "loss": 5.033,
      "mean_token_accuracy": 0.1766697272658348,
      "num_tokens": 44552556.0,
      "step": 20335
    },
    {
      "entropy": 5.537384510040283,
      "epoch": 2.018658197697499,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045953991561679855,
      "loss": 5.042,
      "mean_token_accuracy": 0.18193720579147338,
      "num_tokens": 44563052.0,
      "step": 20340
    },
    {
      "entropy": 5.534431600570679,
      "epoch": 2.0191544263596666,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004595196426410742,
      "loss": 5.0239,
      "mean_token_accuracy": 0.18166431784629822,
      "num_tokens": 44574020.0,
      "step": 20345
    },
    {
      "entropy": 5.5410583972930905,
      "epoch": 2.019650655021834,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045949936508966015,
      "loss": 4.9769,
      "mean_token_accuracy": 0.1821667656302452,
      "num_tokens": 44586261.0,
      "step": 20350
    },
    {
      "entropy": 5.594756269454956,
      "epoch": 2.0201468836840015,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004594790829630596,
      "loss": 5.0251,
      "mean_token_accuracy": 0.17324368953704833,
      "num_tokens": 44596391.0,
      "step": 20355
    },
    {
      "entropy": 5.5267126083374025,
      "epoch": 2.020643112346169,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004594587962617752,
      "loss": 4.9813,
      "mean_token_accuracy": 0.18209591507911682,
      "num_tokens": 44606263.0,
      "step": 20360
    },
    {
      "entropy": 5.571532487869263,
      "epoch": 2.0211393410083365,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004594385049863102,
      "loss": 5.0498,
      "mean_token_accuracy": 0.17598610371351242,
      "num_tokens": 44617332.0,
      "step": 20365
    },
    {
      "entropy": 5.571857976913452,
      "epoch": 2.021635569670504,
      "grad_norm": 1.5,
      "learning_rate": 0.00045941820913716766,
      "loss": 5.0423,
      "mean_token_accuracy": 0.19036736637353896,
      "num_tokens": 44629784.0,
      "step": 20370
    },
    {
      "entropy": 5.672733354568481,
      "epoch": 2.022131798332672,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045939790871485093,
      "loss": 5.1552,
      "mean_token_accuracy": 0.16935770064592362,
      "num_tokens": 44641398.0,
      "step": 20375
    },
    {
      "entropy": 5.507920789718628,
      "epoch": 2.0226280269948393,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00045937760371986334,
      "loss": 4.9055,
      "mean_token_accuracy": 0.1905221089720726,
      "num_tokens": 44652592.0,
      "step": 20380
    },
    {
      "entropy": 5.561168527603149,
      "epoch": 2.023124255657007,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045935729415270857,
      "loss": 4.9746,
      "mean_token_accuracy": 0.19756082445383072,
      "num_tokens": 44663896.0,
      "step": 20385
    },
    {
      "entropy": 5.667262077331543,
      "epoch": 2.0236204843191743,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004593369800138901,
      "loss": 5.1233,
      "mean_token_accuracy": 0.17189263701438903,
      "num_tokens": 44675004.0,
      "step": 20390
    },
    {
      "entropy": 5.621305751800537,
      "epoch": 2.0241167129813418,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004593166613039118,
      "loss": 5.0913,
      "mean_token_accuracy": 0.17793205380439758,
      "num_tokens": 44686778.0,
      "step": 20395
    },
    {
      "entropy": 5.559695434570313,
      "epoch": 2.0246129416435092,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045929633802327743,
      "loss": 5.0152,
      "mean_token_accuracy": 0.18322455286979675,
      "num_tokens": 44697600.0,
      "step": 20400
    },
    {
      "entropy": 5.547638893127441,
      "epoch": 2.0251091703056767,
      "grad_norm": 1.046875,
      "learning_rate": 0.000459276010172491,
      "loss": 5.0716,
      "mean_token_accuracy": 0.17049810588359832,
      "num_tokens": 44707453.0,
      "step": 20405
    },
    {
      "entropy": 5.590932893753052,
      "epoch": 2.025605398967844,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045925567775205656,
      "loss": 5.1391,
      "mean_token_accuracy": 0.1736005961894989,
      "num_tokens": 44718351.0,
      "step": 20410
    },
    {
      "entropy": 5.60587887763977,
      "epoch": 2.026101627630012,
      "grad_norm": 0.984375,
      "learning_rate": 0.00045923534076247846,
      "loss": 5.0128,
      "mean_token_accuracy": 0.18642785400152206,
      "num_tokens": 44728884.0,
      "step": 20415
    },
    {
      "entropy": 5.600027799606323,
      "epoch": 2.0265978562921796,
      "grad_norm": 0.90625,
      "learning_rate": 0.00045921499920426083,
      "loss": 5.0263,
      "mean_token_accuracy": 0.1786925658583641,
      "num_tokens": 44740769.0,
      "step": 20420
    },
    {
      "entropy": 5.599120426177978,
      "epoch": 2.027094084954347,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045919465307790817,
      "loss": 5.1447,
      "mean_token_accuracy": 0.17507598847150802,
      "num_tokens": 44753094.0,
      "step": 20425
    },
    {
      "entropy": 5.5000285625457765,
      "epoch": 2.0275903136165145,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045917430238392505,
      "loss": 4.8942,
      "mean_token_accuracy": 0.19176001399755477,
      "num_tokens": 44762996.0,
      "step": 20430
    },
    {
      "entropy": 5.528482675552368,
      "epoch": 2.028086542278682,
      "grad_norm": 0.828125,
      "learning_rate": 0.000459153947122816,
      "loss": 5.0858,
      "mean_token_accuracy": 0.18000393211841584,
      "num_tokens": 44775343.0,
      "step": 20435
    },
    {
      "entropy": 5.519376182556153,
      "epoch": 2.0285827709408495,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004591335872950859,
      "loss": 4.9389,
      "mean_token_accuracy": 0.18735790699720384,
      "num_tokens": 44785383.0,
      "step": 20440
    },
    {
      "entropy": 5.554380083084107,
      "epoch": 2.029078999603017,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004591132229012397,
      "loss": 5.0015,
      "mean_token_accuracy": 0.18096659630537032,
      "num_tokens": 44796805.0,
      "step": 20445
    },
    {
      "entropy": 5.569212436676025,
      "epoch": 2.0295752282651844,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045909285394178217,
      "loss": 5.0113,
      "mean_token_accuracy": 0.18607627898454665,
      "num_tokens": 44806800.0,
      "step": 20450
    },
    {
      "entropy": 5.605460548400879,
      "epoch": 2.030071456927352,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004590724804172187,
      "loss": 5.1266,
      "mean_token_accuracy": 0.16906161308288575,
      "num_tokens": 44818745.0,
      "step": 20455
    },
    {
      "entropy": 5.524948310852051,
      "epoch": 2.03056768558952,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004590521023280543,
      "loss": 4.9927,
      "mean_token_accuracy": 0.18356360048055648,
      "num_tokens": 44829825.0,
      "step": 20460
    },
    {
      "entropy": 5.536785507202149,
      "epoch": 2.0310639142516873,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00045903171967479425,
      "loss": 5.0033,
      "mean_token_accuracy": 0.18167675882577897,
      "num_tokens": 44841742.0,
      "step": 20465
    },
    {
      "entropy": 5.5943561553955075,
      "epoch": 2.031560142913855,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045901133245794426,
      "loss": 5.129,
      "mean_token_accuracy": 0.18166817724704742,
      "num_tokens": 44853019.0,
      "step": 20470
    },
    {
      "entropy": 5.555161666870117,
      "epoch": 2.0320563715760223,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004589909406780096,
      "loss": 4.9651,
      "mean_token_accuracy": 0.18661444634199142,
      "num_tokens": 44863147.0,
      "step": 20475
    },
    {
      "entropy": 5.630882787704468,
      "epoch": 2.0325526002381897,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045897054433549617,
      "loss": 5.0364,
      "mean_token_accuracy": 0.1834707573056221,
      "num_tokens": 44873638.0,
      "step": 20480
    },
    {
      "entropy": 5.5578453540802,
      "epoch": 2.033048828900357,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045895014343090963,
      "loss": 5.0343,
      "mean_token_accuracy": 0.18064282238483428,
      "num_tokens": 44885752.0,
      "step": 20485
    },
    {
      "entropy": 5.511915683746338,
      "epoch": 2.0335450575625247,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00045892973796475594,
      "loss": 4.9878,
      "mean_token_accuracy": 0.17926738709211348,
      "num_tokens": 44897471.0,
      "step": 20490
    },
    {
      "entropy": 5.641724109649658,
      "epoch": 2.034041286224692,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000458909327937541,
      "loss": 5.0801,
      "mean_token_accuracy": 0.17359787672758104,
      "num_tokens": 44907236.0,
      "step": 20495
    },
    {
      "entropy": 5.594384717941284,
      "epoch": 2.03453751488686,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00045888891334977105,
      "loss": 4.9902,
      "mean_token_accuracy": 0.18277155011892318,
      "num_tokens": 44918103.0,
      "step": 20500
    },
    {
      "entropy": 5.546513462066651,
      "epoch": 2.0350337435490276,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004588684942019524,
      "loss": 5.0475,
      "mean_token_accuracy": 0.17986328303813934,
      "num_tokens": 44928263.0,
      "step": 20505
    },
    {
      "entropy": 5.5183717727661135,
      "epoch": 2.035529972211195,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00045884807049459116,
      "loss": 5.0121,
      "mean_token_accuracy": 0.1820213943719864,
      "num_tokens": 44939487.0,
      "step": 20510
    },
    {
      "entropy": 5.574971866607666,
      "epoch": 2.0360262008733625,
      "grad_norm": 1.046875,
      "learning_rate": 0.000458827642228194,
      "loss": 5.1081,
      "mean_token_accuracy": 0.17568829357624055,
      "num_tokens": 44949749.0,
      "step": 20515
    },
    {
      "entropy": 5.58670449256897,
      "epoch": 2.03652242953553,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045880720940326744,
      "loss": 5.0346,
      "mean_token_accuracy": 0.17874034345149994,
      "num_tokens": 44960347.0,
      "step": 20520
    },
    {
      "entropy": 5.665294504165649,
      "epoch": 2.0370186581976975,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045878677202031804,
      "loss": 5.1019,
      "mean_token_accuracy": 0.17289525419473648,
      "num_tokens": 44970420.0,
      "step": 20525
    },
    {
      "entropy": 5.560429620742798,
      "epoch": 2.037514886859865,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004587663300798529,
      "loss": 5.0347,
      "mean_token_accuracy": 0.17720650285482406,
      "num_tokens": 44980888.0,
      "step": 20530
    },
    {
      "entropy": 5.5810709476470945,
      "epoch": 2.0380111155220324,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004587458835823786,
      "loss": 5.0337,
      "mean_token_accuracy": 0.17814827263355254,
      "num_tokens": 44992721.0,
      "step": 20535
    },
    {
      "entropy": 5.580349922180176,
      "epoch": 2.0385073441842,
      "grad_norm": 1.125,
      "learning_rate": 0.00045872543252840236,
      "loss": 5.0041,
      "mean_token_accuracy": 0.18439416140317916,
      "num_tokens": 45003603.0,
      "step": 20540
    },
    {
      "entropy": 5.557141971588135,
      "epoch": 2.039003572846368,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045870497691843134,
      "loss": 5.0339,
      "mean_token_accuracy": 0.18700573444366456,
      "num_tokens": 45014864.0,
      "step": 20545
    },
    {
      "entropy": 5.586525249481201,
      "epoch": 2.0394998015085353,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004586845167529727,
      "loss": 5.0793,
      "mean_token_accuracy": 0.1750664606690407,
      "num_tokens": 45027766.0,
      "step": 20550
    },
    {
      "entropy": 5.557331609725952,
      "epoch": 2.0399960301707027,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00045866405203253384,
      "loss": 5.0372,
      "mean_token_accuracy": 0.17988807559013367,
      "num_tokens": 45039524.0,
      "step": 20555
    },
    {
      "entropy": 5.582472419738769,
      "epoch": 2.04049225883287,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004586435827576223,
      "loss": 5.0617,
      "mean_token_accuracy": 0.17601825445890426,
      "num_tokens": 45050848.0,
      "step": 20560
    },
    {
      "entropy": 5.564962100982666,
      "epoch": 2.0409884874950377,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004586231089287456,
      "loss": 5.0089,
      "mean_token_accuracy": 0.178389011323452,
      "num_tokens": 45061229.0,
      "step": 20565
    },
    {
      "entropy": 5.525482606887818,
      "epoch": 2.041484716157205,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004586026305464115,
      "loss": 4.9572,
      "mean_token_accuracy": 0.18500153720378876,
      "num_tokens": 45070466.0,
      "step": 20570
    },
    {
      "entropy": 5.589316558837891,
      "epoch": 2.0419809448193726,
      "grad_norm": 1.0,
      "learning_rate": 0.00045858214761112777,
      "loss": 5.0467,
      "mean_token_accuracy": 0.17747531831264496,
      "num_tokens": 45081839.0,
      "step": 20575
    },
    {
      "entropy": 5.652382278442383,
      "epoch": 2.04247717348154,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045856166012340235,
      "loss": 5.1038,
      "mean_token_accuracy": 0.17503010034561156,
      "num_tokens": 45092458.0,
      "step": 20580
    },
    {
      "entropy": 5.643860960006714,
      "epoch": 2.0429734021437076,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004585411680837433,
      "loss": 5.0836,
      "mean_token_accuracy": 0.17511983811855317,
      "num_tokens": 45104629.0,
      "step": 20585
    },
    {
      "entropy": 5.581271648406982,
      "epoch": 2.0434696308058755,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00045852067149265875,
      "loss": 4.9434,
      "mean_token_accuracy": 0.1826970860362053,
      "num_tokens": 45115556.0,
      "step": 20590
    },
    {
      "entropy": 5.5476508140563965,
      "epoch": 2.043965859468043,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000458500170350657,
      "loss": 5.0175,
      "mean_token_accuracy": 0.1785544604063034,
      "num_tokens": 45126489.0,
      "step": 20595
    },
    {
      "entropy": 5.579299736022949,
      "epoch": 2.0444620881302105,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00045847966465824654,
      "loss": 5.1267,
      "mean_token_accuracy": 0.17368541061878204,
      "num_tokens": 45138882.0,
      "step": 20600
    },
    {
      "entropy": 5.636334228515625,
      "epoch": 2.044958316792378,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004584591544159356,
      "loss": 5.0553,
      "mean_token_accuracy": 0.18055800050497056,
      "num_tokens": 45149941.0,
      "step": 20605
    },
    {
      "entropy": 5.667939281463623,
      "epoch": 2.0454545454545454,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004584386396242331,
      "loss": 5.0874,
      "mean_token_accuracy": 0.17820067554712296,
      "num_tokens": 45162267.0,
      "step": 20610
    },
    {
      "entropy": 5.63246488571167,
      "epoch": 2.045950774116713,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045841812028364756,
      "loss": 5.1739,
      "mean_token_accuracy": 0.17613014876842498,
      "num_tokens": 45174444.0,
      "step": 20615
    },
    {
      "entropy": 5.559367942810058,
      "epoch": 2.0464470027788804,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045839759639468774,
      "loss": 5.0265,
      "mean_token_accuracy": 0.18383753448724746,
      "num_tokens": 45183957.0,
      "step": 20620
    },
    {
      "entropy": 5.605564212799072,
      "epoch": 2.046943231441048,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004583770679578628,
      "loss": 5.1133,
      "mean_token_accuracy": 0.1812033087015152,
      "num_tokens": 45195607.0,
      "step": 20625
    },
    {
      "entropy": 5.54872465133667,
      "epoch": 2.0474394601032158,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004583565349736818,
      "loss": 5.0028,
      "mean_token_accuracy": 0.1790957823395729,
      "num_tokens": 45205627.0,
      "step": 20630
    },
    {
      "entropy": 5.530519294738769,
      "epoch": 2.0479356887653832,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004583359974426538,
      "loss": 5.0661,
      "mean_token_accuracy": 0.1742791309952736,
      "num_tokens": 45216331.0,
      "step": 20635
    },
    {
      "entropy": 5.567686986923218,
      "epoch": 2.0484319174275507,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045831545536528806,
      "loss": 5.0862,
      "mean_token_accuracy": 0.16973341405391693,
      "num_tokens": 45227826.0,
      "step": 20640
    },
    {
      "entropy": 5.595687961578369,
      "epoch": 2.048928146089718,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045829490874209404,
      "loss": 5.0147,
      "mean_token_accuracy": 0.18066619485616683,
      "num_tokens": 45238093.0,
      "step": 20645
    },
    {
      "entropy": 5.571276521682739,
      "epoch": 2.0494243747518857,
      "grad_norm": 0.96875,
      "learning_rate": 0.00045827435757358136,
      "loss": 5.0443,
      "mean_token_accuracy": 0.18174657374620437,
      "num_tokens": 45248660.0,
      "step": 20650
    },
    {
      "entropy": 5.636414813995361,
      "epoch": 2.049920603414053,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004582538018602594,
      "loss": 5.1127,
      "mean_token_accuracy": 0.1737897887825966,
      "num_tokens": 45261059.0,
      "step": 20655
    },
    {
      "entropy": 5.557131910324097,
      "epoch": 2.0504168320762206,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004582332416026381,
      "loss": 5.0132,
      "mean_token_accuracy": 0.1771009922027588,
      "num_tokens": 45272210.0,
      "step": 20660
    },
    {
      "entropy": 5.6142806053161625,
      "epoch": 2.050913060738388,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00045821267680122725,
      "loss": 5.1071,
      "mean_token_accuracy": 0.1789221405982971,
      "num_tokens": 45283488.0,
      "step": 20665
    },
    {
      "entropy": 5.565041828155517,
      "epoch": 2.0514092894005556,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004581921074565368,
      "loss": 4.9556,
      "mean_token_accuracy": 0.186887463927269,
      "num_tokens": 45292824.0,
      "step": 20670
    },
    {
      "entropy": 5.557746458053589,
      "epoch": 2.0519055180627235,
      "grad_norm": 1.0,
      "learning_rate": 0.0004581715335690768,
      "loss": 5.0468,
      "mean_token_accuracy": 0.17812294661998748,
      "num_tokens": 45303102.0,
      "step": 20675
    },
    {
      "entropy": 5.580737829208374,
      "epoch": 2.052401746724891,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045815095513935747,
      "loss": 4.9565,
      "mean_token_accuracy": 0.17806898355484008,
      "num_tokens": 45313825.0,
      "step": 20680
    },
    {
      "entropy": 5.549282169342041,
      "epoch": 2.0528979753870584,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004581303721678891,
      "loss": 5.0033,
      "mean_token_accuracy": 0.1876699820160866,
      "num_tokens": 45325789.0,
      "step": 20685
    },
    {
      "entropy": 5.570593595504761,
      "epoch": 2.053394204049226,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004581097846551821,
      "loss": 5.0795,
      "mean_token_accuracy": 0.1743321478366852,
      "num_tokens": 45337630.0,
      "step": 20690
    },
    {
      "entropy": 5.640217065811157,
      "epoch": 2.0538904327113934,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000458089192601747,
      "loss": 5.0978,
      "mean_token_accuracy": 0.17524106204509735,
      "num_tokens": 45349815.0,
      "step": 20695
    },
    {
      "entropy": 5.586354780197143,
      "epoch": 2.054386661373561,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045806859600809455,
      "loss": 5.0368,
      "mean_token_accuracy": 0.17634499669075013,
      "num_tokens": 45360741.0,
      "step": 20700
    },
    {
      "entropy": 5.476849508285523,
      "epoch": 2.0548828900357283,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004580479948747353,
      "loss": 4.9155,
      "mean_token_accuracy": 0.18714760094881058,
      "num_tokens": 45371258.0,
      "step": 20705
    },
    {
      "entropy": 5.587560987472534,
      "epoch": 2.055379118697896,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045802738920218015,
      "loss": 5.045,
      "mean_token_accuracy": 0.17742789685726165,
      "num_tokens": 45381914.0,
      "step": 20710
    },
    {
      "entropy": 5.555529451370239,
      "epoch": 2.0558753473600637,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004580067789909403,
      "loss": 5.0516,
      "mean_token_accuracy": 0.1820403218269348,
      "num_tokens": 45393159.0,
      "step": 20715
    },
    {
      "entropy": 5.545459651947022,
      "epoch": 2.056371576022231,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045798616424152654,
      "loss": 4.9319,
      "mean_token_accuracy": 0.18739939481019974,
      "num_tokens": 45403076.0,
      "step": 20720
    },
    {
      "entropy": 5.561541271209717,
      "epoch": 2.0568678046843987,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00045796554495445023,
      "loss": 5.0797,
      "mean_token_accuracy": 0.1781062051653862,
      "num_tokens": 45412962.0,
      "step": 20725
    },
    {
      "entropy": 5.5583171367645265,
      "epoch": 2.057364033346566,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004579449211302228,
      "loss": 4.9967,
      "mean_token_accuracy": 0.1842828556895256,
      "num_tokens": 45423620.0,
      "step": 20730
    },
    {
      "entropy": 5.584582662582397,
      "epoch": 2.0578602620087336,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004579242927693554,
      "loss": 5.037,
      "mean_token_accuracy": 0.17163565903902053,
      "num_tokens": 45435901.0,
      "step": 20735
    },
    {
      "entropy": 5.574634218215943,
      "epoch": 2.058356490670901,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004579036598723598,
      "loss": 4.9683,
      "mean_token_accuracy": 0.19019347727298735,
      "num_tokens": 45445409.0,
      "step": 20740
    },
    {
      "entropy": 5.6521838188171385,
      "epoch": 2.0588527193330686,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045788302243974745,
      "loss": 5.1238,
      "mean_token_accuracy": 0.1788850173354149,
      "num_tokens": 45455794.0,
      "step": 20745
    },
    {
      "entropy": 5.595630216598511,
      "epoch": 2.059348947995236,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004578623804720302,
      "loss": 4.9991,
      "mean_token_accuracy": 0.18362150490283966,
      "num_tokens": 45465922.0,
      "step": 20750
    },
    {
      "entropy": 5.556346273422241,
      "epoch": 2.0598451766574035,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004578417339697201,
      "loss": 5.1042,
      "mean_token_accuracy": 0.1750722348690033,
      "num_tokens": 45476314.0,
      "step": 20755
    },
    {
      "entropy": 5.588579797744751,
      "epoch": 2.0603414053195714,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045782108293332884,
      "loss": 5.0197,
      "mean_token_accuracy": 0.1753836527466774,
      "num_tokens": 45487232.0,
      "step": 20760
    },
    {
      "entropy": 5.664383506774902,
      "epoch": 2.060837633981739,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004578004273633687,
      "loss": 5.0885,
      "mean_token_accuracy": 0.18001712560653688,
      "num_tokens": 45498368.0,
      "step": 20765
    },
    {
      "entropy": 5.57822003364563,
      "epoch": 2.0613338626439064,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045777976726035194,
      "loss": 5.0287,
      "mean_token_accuracy": 0.17257740795612336,
      "num_tokens": 45507600.0,
      "step": 20770
    },
    {
      "entropy": 5.590188694000244,
      "epoch": 2.061830091306074,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004577591026247907,
      "loss": 5.1225,
      "mean_token_accuracy": 0.16889785826206208,
      "num_tokens": 45518594.0,
      "step": 20775
    },
    {
      "entropy": 5.566431140899658,
      "epoch": 2.0623263199682413,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004577384334571976,
      "loss": 4.9925,
      "mean_token_accuracy": 0.17890307307243347,
      "num_tokens": 45529929.0,
      "step": 20780
    },
    {
      "entropy": 5.577402019500733,
      "epoch": 2.062822548630409,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004577177597580851,
      "loss": 5.0013,
      "mean_token_accuracy": 0.18637326806783677,
      "num_tokens": 45540228.0,
      "step": 20785
    },
    {
      "entropy": 5.521492528915405,
      "epoch": 2.0633187772925763,
      "grad_norm": 0.9375,
      "learning_rate": 0.00045769708152796593,
      "loss": 4.9699,
      "mean_token_accuracy": 0.18163398802280425,
      "num_tokens": 45552026.0,
      "step": 20790
    },
    {
      "entropy": 5.519711446762085,
      "epoch": 2.0638150059547438,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045767639876735276,
      "loss": 5.0472,
      "mean_token_accuracy": 0.1865169957280159,
      "num_tokens": 45563579.0,
      "step": 20795
    },
    {
      "entropy": 5.650913572311401,
      "epoch": 2.0643112346169117,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004576557114767585,
      "loss": 5.1299,
      "mean_token_accuracy": 0.17230444103479386,
      "num_tokens": 45573800.0,
      "step": 20800
    },
    {
      "entropy": 5.556518840789795,
      "epoch": 2.064807463279079,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045763501965669627,
      "loss": 5.0027,
      "mean_token_accuracy": 0.17949966043233873,
      "num_tokens": 45583758.0,
      "step": 20805
    },
    {
      "entropy": 5.580332851409912,
      "epoch": 2.0653036919412466,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000457614323307679,
      "loss": 5.0007,
      "mean_token_accuracy": 0.18121497631072997,
      "num_tokens": 45595154.0,
      "step": 20810
    },
    {
      "entropy": 5.621364593505859,
      "epoch": 2.065799920603414,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045759362243022,
      "loss": 5.1041,
      "mean_token_accuracy": 0.17656422704458236,
      "num_tokens": 45606557.0,
      "step": 20815
    },
    {
      "entropy": 5.539298009872437,
      "epoch": 2.0662961492655816,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004575729170248327,
      "loss": 4.9654,
      "mean_token_accuracy": 0.18942419141530992,
      "num_tokens": 45615531.0,
      "step": 20820
    },
    {
      "entropy": 5.498677349090576,
      "epoch": 2.066792377927749,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004575522070920305,
      "loss": 4.9728,
      "mean_token_accuracy": 0.1830257222056389,
      "num_tokens": 45626512.0,
      "step": 20825
    },
    {
      "entropy": 5.6015135765075685,
      "epoch": 2.0672886065899165,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045753149263232687,
      "loss": 5.0803,
      "mean_token_accuracy": 0.17098679393529892,
      "num_tokens": 45638005.0,
      "step": 20830
    },
    {
      "entropy": 5.598339986801148,
      "epoch": 2.067784835252084,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004575107736462355,
      "loss": 5.0812,
      "mean_token_accuracy": 0.17462422102689742,
      "num_tokens": 45647740.0,
      "step": 20835
    },
    {
      "entropy": 5.544892501831055,
      "epoch": 2.0682810639142515,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004574900501342703,
      "loss": 5.0056,
      "mean_token_accuracy": 0.18812091648578644,
      "num_tokens": 45659021.0,
      "step": 20840
    },
    {
      "entropy": 5.579664802551269,
      "epoch": 2.0687772925764194,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004574693220969451,
      "loss": 5.0811,
      "mean_token_accuracy": 0.17835566401481628,
      "num_tokens": 45669413.0,
      "step": 20845
    },
    {
      "entropy": 5.667920875549316,
      "epoch": 2.069273521238587,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004574485895347738,
      "loss": 5.1388,
      "mean_token_accuracy": 0.17369325011968612,
      "num_tokens": 45680719.0,
      "step": 20850
    },
    {
      "entropy": 5.518199396133423,
      "epoch": 2.0697697499007544,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045742785244827066,
      "loss": 4.9628,
      "mean_token_accuracy": 0.18290352523326875,
      "num_tokens": 45690393.0,
      "step": 20855
    },
    {
      "entropy": 5.548664474487305,
      "epoch": 2.070265978562922,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004574071108379499,
      "loss": 5.0783,
      "mean_token_accuracy": 0.17684251368045806,
      "num_tokens": 45701304.0,
      "step": 20860
    },
    {
      "entropy": 5.5731456756591795,
      "epoch": 2.0707622072250893,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004573863647043259,
      "loss": 4.9414,
      "mean_token_accuracy": 0.18446728438138962,
      "num_tokens": 45711892.0,
      "step": 20865
    },
    {
      "entropy": 5.581446313858033,
      "epoch": 2.0712584358872568,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004573656140479129,
      "loss": 5.0151,
      "mean_token_accuracy": 0.1787782147526741,
      "num_tokens": 45723693.0,
      "step": 20870
    },
    {
      "entropy": 5.5282111167907715,
      "epoch": 2.0717546645494243,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004573448588692258,
      "loss": 5.0033,
      "mean_token_accuracy": 0.17832859605550766,
      "num_tokens": 45734392.0,
      "step": 20875
    },
    {
      "entropy": 5.464937925338745,
      "epoch": 2.0722508932115917,
      "grad_norm": 1.03125,
      "learning_rate": 0.000457324099168779,
      "loss": 4.9329,
      "mean_token_accuracy": 0.19012268781661987,
      "num_tokens": 45745732.0,
      "step": 20880
    },
    {
      "entropy": 5.52878246307373,
      "epoch": 2.0727471218737596,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004573033349470874,
      "loss": 4.99,
      "mean_token_accuracy": 0.18597166687250138,
      "num_tokens": 45756671.0,
      "step": 20885
    },
    {
      "entropy": 5.636493492126465,
      "epoch": 2.073243350535927,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004572825662046661,
      "loss": 5.1048,
      "mean_token_accuracy": 0.17231640368700027,
      "num_tokens": 45767685.0,
      "step": 20890
    },
    {
      "entropy": 5.59394941329956,
      "epoch": 2.0737395791980946,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004572617929420298,
      "loss": 5.01,
      "mean_token_accuracy": 0.18423440754413606,
      "num_tokens": 45777251.0,
      "step": 20895
    },
    {
      "entropy": 5.6230322360992435,
      "epoch": 2.074235807860262,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045724101515969375,
      "loss": 5.0783,
      "mean_token_accuracy": 0.18082635700702668,
      "num_tokens": 45790037.0,
      "step": 20900
    },
    {
      "entropy": 5.601011610031128,
      "epoch": 2.0747320365224295,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045722023285817327,
      "loss": 5.1267,
      "mean_token_accuracy": 0.17161785960197448,
      "num_tokens": 45800803.0,
      "step": 20905
    },
    {
      "entropy": 5.622574949264527,
      "epoch": 2.075228265184597,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004571994460379837,
      "loss": 5.0605,
      "mean_token_accuracy": 0.17137459367513658,
      "num_tokens": 45812447.0,
      "step": 20910
    },
    {
      "entropy": 5.591508960723877,
      "epoch": 2.0757244938467645,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045717865469964045,
      "loss": 5.0135,
      "mean_token_accuracy": 0.18045831173658372,
      "num_tokens": 45822172.0,
      "step": 20915
    },
    {
      "entropy": 5.6481201171875,
      "epoch": 2.076220722508932,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045715785884365914,
      "loss": 5.1355,
      "mean_token_accuracy": 0.1753174141049385,
      "num_tokens": 45832583.0,
      "step": 20920
    },
    {
      "entropy": 5.630967044830323,
      "epoch": 2.0767169511710994,
      "grad_norm": 1.125,
      "learning_rate": 0.00045713705847055537,
      "loss": 5.1425,
      "mean_token_accuracy": 0.17449038326740265,
      "num_tokens": 45842640.0,
      "step": 20925
    },
    {
      "entropy": 5.637538766860962,
      "epoch": 2.0772131798332674,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004571162535808451,
      "loss": 5.1388,
      "mean_token_accuracy": 0.17166436910629274,
      "num_tokens": 45854618.0,
      "step": 20930
    },
    {
      "entropy": 5.60342173576355,
      "epoch": 2.077709408495435,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004570954441750441,
      "loss": 5.0317,
      "mean_token_accuracy": 0.18146111220121383,
      "num_tokens": 45866130.0,
      "step": 20935
    },
    {
      "entropy": 5.628126239776611,
      "epoch": 2.0782056371576023,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004570746302536685,
      "loss": 5.057,
      "mean_token_accuracy": 0.17616796046495437,
      "num_tokens": 45877187.0,
      "step": 20940
    },
    {
      "entropy": 5.549054765701294,
      "epoch": 2.07870186581977,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045705381181723444,
      "loss": 5.0679,
      "mean_token_accuracy": 0.17600455582141877,
      "num_tokens": 45889557.0,
      "step": 20945
    },
    {
      "entropy": 5.573263311386109,
      "epoch": 2.0791980944819373,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00045703298886625804,
      "loss": 5.0654,
      "mean_token_accuracy": 0.1764945477247238,
      "num_tokens": 45900860.0,
      "step": 20950
    },
    {
      "entropy": 5.593493032455444,
      "epoch": 2.0796943231441047,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004570121614012558,
      "loss": 5.0869,
      "mean_token_accuracy": 0.1741206780076027,
      "num_tokens": 45912231.0,
      "step": 20955
    },
    {
      "entropy": 5.642436122894287,
      "epoch": 2.080190551806272,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045699132942274425,
      "loss": 5.1233,
      "mean_token_accuracy": 0.17317160815000535,
      "num_tokens": 45922722.0,
      "step": 20960
    },
    {
      "entropy": 5.580073833465576,
      "epoch": 2.0806867804684397,
      "grad_norm": 1.0,
      "learning_rate": 0.00045697049293123984,
      "loss": 5.0352,
      "mean_token_accuracy": 0.18277282565832137,
      "num_tokens": 45932444.0,
      "step": 20965
    },
    {
      "entropy": 5.633468198776245,
      "epoch": 2.0811830091306076,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004569496519272593,
      "loss": 5.0605,
      "mean_token_accuracy": 0.18099953830242158,
      "num_tokens": 45942826.0,
      "step": 20970
    },
    {
      "entropy": 5.600572204589843,
      "epoch": 2.081679237792775,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045692880641131945,
      "loss": 5.055,
      "mean_token_accuracy": 0.18229408115148543,
      "num_tokens": 45953935.0,
      "step": 20975
    },
    {
      "entropy": 5.627800035476684,
      "epoch": 2.0821754664549426,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004569079563839372,
      "loss": 5.1464,
      "mean_token_accuracy": 0.17057626247406005,
      "num_tokens": 45964693.0,
      "step": 20980
    },
    {
      "entropy": 5.63632116317749,
      "epoch": 2.08267169511711,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045688710184562965,
      "loss": 5.0366,
      "mean_token_accuracy": 0.174713633954525,
      "num_tokens": 45976481.0,
      "step": 20985
    },
    {
      "entropy": 5.652490949630737,
      "epoch": 2.0831679237792775,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045686624279691387,
      "loss": 5.1148,
      "mean_token_accuracy": 0.18101923912763596,
      "num_tokens": 45988879.0,
      "step": 20990
    },
    {
      "entropy": 5.561849355697632,
      "epoch": 2.083664152441445,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004568453792383072,
      "loss": 5.0078,
      "mean_token_accuracy": 0.18511599600315093,
      "num_tokens": 45999332.0,
      "step": 20995
    },
    {
      "entropy": 5.597072505950928,
      "epoch": 2.0841603811036125,
      "grad_norm": 1.015625,
      "learning_rate": 0.000456824511170327,
      "loss": 5.046,
      "mean_token_accuracy": 0.17673124521970748,
      "num_tokens": 46009841.0,
      "step": 21000
    },
    {
      "epoch": 2.0841603811036125,
      "eval_entropy": 5.378207194037606,
      "eval_loss": 5.182725429534912,
      "eval_mean_token_accuracy": 0.17813740089008206,
      "eval_num_tokens": 46009841.0,
      "eval_runtime": 20.5344,
      "eval_samples_per_second": 1639.64,
      "eval_steps_per_second": 204.973,
      "step": 21000
    },
    {
      "entropy": 5.549616813659668,
      "epoch": 2.08465660976578,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004568036385934906,
      "loss": 4.9721,
      "mean_token_accuracy": 0.18178554326295854,
      "num_tokens": 46020678.0,
      "step": 21005
    },
    {
      "entropy": 5.596362352371216,
      "epoch": 2.0851528384279474,
      "grad_norm": 0.953125,
      "learning_rate": 0.00045678276150831586,
      "loss": 5.0583,
      "mean_token_accuracy": 0.176519213616848,
      "num_tokens": 46032036.0,
      "step": 21010
    },
    {
      "entropy": 5.606713151931762,
      "epoch": 2.0856490670901153,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004567618799153202,
      "loss": 5.0492,
      "mean_token_accuracy": 0.17937204688787461,
      "num_tokens": 46042675.0,
      "step": 21015
    },
    {
      "entropy": 5.5188436031341555,
      "epoch": 2.086145295752283,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045674099381502167,
      "loss": 5.064,
      "mean_token_accuracy": 0.1755911096930504,
      "num_tokens": 46053616.0,
      "step": 21020
    },
    {
      "entropy": 5.60694580078125,
      "epoch": 2.0866415244144503,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004567201032079381,
      "loss": 5.0804,
      "mean_token_accuracy": 0.1770681068301201,
      "num_tokens": 46063755.0,
      "step": 21025
    },
    {
      "entropy": 5.634601068496704,
      "epoch": 2.0871377530766178,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045669920809458754,
      "loss": 5.0601,
      "mean_token_accuracy": 0.18070498555898667,
      "num_tokens": 46074459.0,
      "step": 21030
    },
    {
      "entropy": 5.676368474960327,
      "epoch": 2.0876339817387852,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004566783084754881,
      "loss": 5.1244,
      "mean_token_accuracy": 0.17696245908737182,
      "num_tokens": 46085053.0,
      "step": 21035
    },
    {
      "entropy": 5.5432586669921875,
      "epoch": 2.0881302104009527,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00045665740435115815,
      "loss": 4.9672,
      "mean_token_accuracy": 0.18689364790916443,
      "num_tokens": 46094697.0,
      "step": 21040
    },
    {
      "entropy": 5.554885816574097,
      "epoch": 2.08862643906312,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000456636495722116,
      "loss": 5.0853,
      "mean_token_accuracy": 0.17936250269412995,
      "num_tokens": 46105617.0,
      "step": 21045
    },
    {
      "entropy": 5.668656253814698,
      "epoch": 2.0891226677252877,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004566155825888801,
      "loss": 5.0325,
      "mean_token_accuracy": 0.18122997283935546,
      "num_tokens": 46115450.0,
      "step": 21050
    },
    {
      "entropy": 5.590940666198731,
      "epoch": 2.089618896387455,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004565946649519691,
      "loss": 5.1039,
      "mean_token_accuracy": 0.17471627295017242,
      "num_tokens": 46126511.0,
      "step": 21055
    },
    {
      "entropy": 5.581959390640259,
      "epoch": 2.090115125049623,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045657374281190177,
      "loss": 5.0017,
      "mean_token_accuracy": 0.18319119811058043,
      "num_tokens": 46136294.0,
      "step": 21060
    },
    {
      "entropy": 5.576907157897949,
      "epoch": 2.0906113537117905,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004565528161691969,
      "loss": 5.0833,
      "mean_token_accuracy": 0.17087069451808928,
      "num_tokens": 46148926.0,
      "step": 21065
    },
    {
      "entropy": 5.626527452468872,
      "epoch": 2.091107582373958,
      "grad_norm": 1.0,
      "learning_rate": 0.0004565318850243733,
      "loss": 5.0837,
      "mean_token_accuracy": 0.18009043782949447,
      "num_tokens": 46160130.0,
      "step": 21070
    },
    {
      "entropy": 5.5552350044250485,
      "epoch": 2.0916038110361255,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004565109493779502,
      "loss": 4.9911,
      "mean_token_accuracy": 0.1845992088317871,
      "num_tokens": 46169919.0,
      "step": 21075
    },
    {
      "entropy": 5.603697395324707,
      "epoch": 2.092100039698293,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00045649000923044664,
      "loss": 5.0767,
      "mean_token_accuracy": 0.1763385906815529,
      "num_tokens": 46181186.0,
      "step": 21080
    },
    {
      "entropy": 5.597004985809326,
      "epoch": 2.0925962683604604,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045646906458238196,
      "loss": 5.077,
      "mean_token_accuracy": 0.17143227756023408,
      "num_tokens": 46192172.0,
      "step": 21085
    },
    {
      "entropy": 5.579356384277344,
      "epoch": 2.093092497022628,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004564481154342754,
      "loss": 5.0511,
      "mean_token_accuracy": 0.17318841218948364,
      "num_tokens": 46202678.0,
      "step": 21090
    },
    {
      "entropy": 5.604232311248779,
      "epoch": 2.0935887256847954,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004564271617866467,
      "loss": 5.0585,
      "mean_token_accuracy": 0.17549093961715698,
      "num_tokens": 46212213.0,
      "step": 21095
    },
    {
      "entropy": 5.658325958251953,
      "epoch": 2.0940849543469633,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045640620364001524,
      "loss": 5.0877,
      "mean_token_accuracy": 0.17705549746751786,
      "num_tokens": 46223044.0,
      "step": 21100
    },
    {
      "entropy": 5.64846773147583,
      "epoch": 2.0945811830091308,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045638524099490086,
      "loss": 5.0949,
      "mean_token_accuracy": 0.18088916093111038,
      "num_tokens": 46236008.0,
      "step": 21105
    },
    {
      "entropy": 5.559442663192749,
      "epoch": 2.0950774116712982,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045636427385182334,
      "loss": 5.0494,
      "mean_token_accuracy": 0.1757390931248665,
      "num_tokens": 46247631.0,
      "step": 21110
    },
    {
      "entropy": 5.583029460906983,
      "epoch": 2.0955736403334657,
      "grad_norm": 0.96875,
      "learning_rate": 0.00045634330221130265,
      "loss": 5.0186,
      "mean_token_accuracy": 0.17828223407268523,
      "num_tokens": 46258294.0,
      "step": 21115
    },
    {
      "entropy": 5.602357721328735,
      "epoch": 2.096069868995633,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004563223260738589,
      "loss": 5.0279,
      "mean_token_accuracy": 0.1764216259121895,
      "num_tokens": 46269611.0,
      "step": 21120
    },
    {
      "entropy": 5.579091024398804,
      "epoch": 2.0965660976578007,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000456301345440012,
      "loss": 5.1014,
      "mean_token_accuracy": 0.1849465474486351,
      "num_tokens": 46280731.0,
      "step": 21125
    },
    {
      "entropy": 5.563394832611084,
      "epoch": 2.097062326319968,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004562803603102825,
      "loss": 5.0249,
      "mean_token_accuracy": 0.1745460405945778,
      "num_tokens": 46290226.0,
      "step": 21130
    },
    {
      "entropy": 5.623651313781738,
      "epoch": 2.0975585549821356,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045625937068519075,
      "loss": 5.0362,
      "mean_token_accuracy": 0.17938980907201768,
      "num_tokens": 46300388.0,
      "step": 21135
    },
    {
      "entropy": 5.58364429473877,
      "epoch": 2.098054783644303,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004562383765652571,
      "loss": 5.0781,
      "mean_token_accuracy": 0.18040805459022521,
      "num_tokens": 46312214.0,
      "step": 21140
    },
    {
      "entropy": 5.605428457260132,
      "epoch": 2.098551012306471,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004562173779510023,
      "loss": 5.0853,
      "mean_token_accuracy": 0.16679899394512177,
      "num_tokens": 46322728.0,
      "step": 21145
    },
    {
      "entropy": 5.55753345489502,
      "epoch": 2.0990472409686385,
      "grad_norm": 0.984375,
      "learning_rate": 0.000456196374842947,
      "loss": 5.0137,
      "mean_token_accuracy": 0.18286693692207337,
      "num_tokens": 46334169.0,
      "step": 21150
    },
    {
      "entropy": 5.59223051071167,
      "epoch": 2.099543469630806,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045617536724161203,
      "loss": 4.9982,
      "mean_token_accuracy": 0.17731491923332215,
      "num_tokens": 46344601.0,
      "step": 21155
    },
    {
      "entropy": 5.576746559143066,
      "epoch": 2.1000396982929734,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045615435514751834,
      "loss": 5.0743,
      "mean_token_accuracy": 0.17338691800832748,
      "num_tokens": 46355633.0,
      "step": 21160
    },
    {
      "entropy": 5.609165668487549,
      "epoch": 2.100535926955141,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000456133338561187,
      "loss": 5.1392,
      "mean_token_accuracy": 0.16641507893800736,
      "num_tokens": 46365981.0,
      "step": 21165
    },
    {
      "entropy": 5.617049264907837,
      "epoch": 2.1010321556173084,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004561123174831391,
      "loss": 5.0966,
      "mean_token_accuracy": 0.18057404905557634,
      "num_tokens": 46376664.0,
      "step": 21170
    },
    {
      "entropy": 5.558592653274536,
      "epoch": 2.101528384279476,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004560912919138962,
      "loss": 5.005,
      "mean_token_accuracy": 0.1823573961853981,
      "num_tokens": 46387918.0,
      "step": 21175
    },
    {
      "entropy": 5.637909650802612,
      "epoch": 2.1020246129416433,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045607026185397924,
      "loss": 5.0274,
      "mean_token_accuracy": 0.1830432340502739,
      "num_tokens": 46398660.0,
      "step": 21180
    },
    {
      "entropy": 5.574792861938477,
      "epoch": 2.102520841603811,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004560492273039101,
      "loss": 4.9803,
      "mean_token_accuracy": 0.18094398230314254,
      "num_tokens": 46409365.0,
      "step": 21185
    },
    {
      "entropy": 5.567476844787597,
      "epoch": 2.1030170702659787,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045602818826421024,
      "loss": 4.9945,
      "mean_token_accuracy": 0.18630912601947786,
      "num_tokens": 46419457.0,
      "step": 21190
    },
    {
      "entropy": 5.569611120223999,
      "epoch": 2.103513298928146,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004560071447354013,
      "loss": 5.0868,
      "mean_token_accuracy": 0.17638052105903626,
      "num_tokens": 46430189.0,
      "step": 21195
    },
    {
      "entropy": 5.5185614109039305,
      "epoch": 2.1040095275903137,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004559860967180053,
      "loss": 5.0292,
      "mean_token_accuracy": 0.18278742134571074,
      "num_tokens": 46440555.0,
      "step": 21200
    },
    {
      "entropy": 5.65135498046875,
      "epoch": 2.104505756252481,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004559650442125441,
      "loss": 5.0524,
      "mean_token_accuracy": 0.1734270229935646,
      "num_tokens": 46451582.0,
      "step": 21205
    },
    {
      "entropy": 5.700384044647217,
      "epoch": 2.1050019849146486,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004559439872195397,
      "loss": 5.0824,
      "mean_token_accuracy": 0.17643293291330336,
      "num_tokens": 46463614.0,
      "step": 21210
    },
    {
      "entropy": 5.634904289245606,
      "epoch": 2.105498213576816,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004559229257395144,
      "loss": 5.0733,
      "mean_token_accuracy": 0.17539695501327515,
      "num_tokens": 46474279.0,
      "step": 21215
    },
    {
      "entropy": 5.630218744277954,
      "epoch": 2.1059944422389836,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045590185977299025,
      "loss": 5.1763,
      "mean_token_accuracy": 0.17436592876911164,
      "num_tokens": 46485531.0,
      "step": 21220
    },
    {
      "entropy": 5.5639465808868405,
      "epoch": 2.106490670901151,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004558807893204899,
      "loss": 4.9858,
      "mean_token_accuracy": 0.18145814687013626,
      "num_tokens": 46496197.0,
      "step": 21225
    },
    {
      "entropy": 5.614350986480713,
      "epoch": 2.106986899563319,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004558597143825357,
      "loss": 5.0644,
      "mean_token_accuracy": 0.17861939668655397,
      "num_tokens": 46506759.0,
      "step": 21230
    },
    {
      "entropy": 5.568059349060059,
      "epoch": 2.1074831282254864,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004558386349596503,
      "loss": 5.0681,
      "mean_token_accuracy": 0.17717475593090057,
      "num_tokens": 46518956.0,
      "step": 21235
    },
    {
      "entropy": 5.684637022018433,
      "epoch": 2.107979356887654,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004558175510523564,
      "loss": 5.1443,
      "mean_token_accuracy": 0.1723005011677742,
      "num_tokens": 46530109.0,
      "step": 21240
    },
    {
      "entropy": 5.635321807861328,
      "epoch": 2.1084755855498214,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00045579646266117696,
      "loss": 5.0007,
      "mean_token_accuracy": 0.17881372272968293,
      "num_tokens": 46542206.0,
      "step": 21245
    },
    {
      "entropy": 5.608251285552979,
      "epoch": 2.108971814211989,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045577536978663475,
      "loss": 5.1574,
      "mean_token_accuracy": 0.18266464322805404,
      "num_tokens": 46554143.0,
      "step": 21250
    },
    {
      "entropy": 5.552614355087281,
      "epoch": 2.1094680428741563,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004557542724292529,
      "loss": 4.9873,
      "mean_token_accuracy": 0.18234004825353622,
      "num_tokens": 46565697.0,
      "step": 21255
    },
    {
      "entropy": 5.617684650421142,
      "epoch": 2.109964271536324,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004557331705895545,
      "loss": 5.0454,
      "mean_token_accuracy": 0.17975567728281022,
      "num_tokens": 46576358.0,
      "step": 21260
    },
    {
      "entropy": 5.618617820739746,
      "epoch": 2.1104605001984913,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004557120642680631,
      "loss": 5.0854,
      "mean_token_accuracy": 0.17682051062583923,
      "num_tokens": 46589422.0,
      "step": 21265
    },
    {
      "entropy": 5.582330846786499,
      "epoch": 2.1109567288606588,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004556909534653018,
      "loss": 5.0578,
      "mean_token_accuracy": 0.18285891860723497,
      "num_tokens": 46599623.0,
      "step": 21270
    },
    {
      "entropy": 5.584320545196533,
      "epoch": 2.1114529575228267,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004556698381817942,
      "loss": 4.9968,
      "mean_token_accuracy": 0.1861279621720314,
      "num_tokens": 46609321.0,
      "step": 21275
    },
    {
      "entropy": 5.525045394897461,
      "epoch": 2.111949186184994,
      "grad_norm": 1.015625,
      "learning_rate": 0.000455648718418064,
      "loss": 5.0684,
      "mean_token_accuracy": 0.17370790094137192,
      "num_tokens": 46620265.0,
      "step": 21280
    },
    {
      "entropy": 5.595817279815674,
      "epoch": 2.1124454148471616,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004556275941746348,
      "loss": 5.0825,
      "mean_token_accuracy": 0.18267000168561937,
      "num_tokens": 46630776.0,
      "step": 21285
    },
    {
      "entropy": 5.549660873413086,
      "epoch": 2.112941643509329,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00045560646545203044,
      "loss": 4.9673,
      "mean_token_accuracy": 0.18396178781986236,
      "num_tokens": 46641382.0,
      "step": 21290
    },
    {
      "entropy": 5.523974895477295,
      "epoch": 2.1134378721714966,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00045558533225077496,
      "loss": 4.9798,
      "mean_token_accuracy": 0.18110697567462922,
      "num_tokens": 46653589.0,
      "step": 21295
    },
    {
      "entropy": 5.54365119934082,
      "epoch": 2.113934100833664,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00045556419457139235,
      "loss": 5.0228,
      "mean_token_accuracy": 0.1827416092157364,
      "num_tokens": 46664595.0,
      "step": 21300
    },
    {
      "entropy": 5.598300743103027,
      "epoch": 2.1144303294958315,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045554305241440684,
      "loss": 5.0444,
      "mean_token_accuracy": 0.1780799373984337,
      "num_tokens": 46675453.0,
      "step": 21305
    },
    {
      "entropy": 5.557474088668823,
      "epoch": 2.114926558157999,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004555219057803426,
      "loss": 4.9611,
      "mean_token_accuracy": 0.18098766803741456,
      "num_tokens": 46685668.0,
      "step": 21310
    },
    {
      "entropy": 5.576448583602906,
      "epoch": 2.115422786820167,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004555007546697242,
      "loss": 5.0626,
      "mean_token_accuracy": 0.1745098739862442,
      "num_tokens": 46696144.0,
      "step": 21315
    },
    {
      "entropy": 5.593879127502442,
      "epoch": 2.1159190154823344,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045547959908307585,
      "loss": 5.0453,
      "mean_token_accuracy": 0.17621316462755204,
      "num_tokens": 46707892.0,
      "step": 21320
    },
    {
      "entropy": 5.609086227416992,
      "epoch": 2.116415244144502,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004554584390209225,
      "loss": 5.0646,
      "mean_token_accuracy": 0.18206441402435303,
      "num_tokens": 46721099.0,
      "step": 21325
    },
    {
      "entropy": 5.61162543296814,
      "epoch": 2.1169114728066694,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004554372744837886,
      "loss": 5.0917,
      "mean_token_accuracy": 0.1766042962670326,
      "num_tokens": 46732974.0,
      "step": 21330
    },
    {
      "entropy": 5.6034228801727295,
      "epoch": 2.117407701468837,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004554161054721992,
      "loss": 5.1137,
      "mean_token_accuracy": 0.1791682094335556,
      "num_tokens": 46743148.0,
      "step": 21335
    },
    {
      "entropy": 5.563817501068115,
      "epoch": 2.1179039301310043,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004553949319866791,
      "loss": 5.0163,
      "mean_token_accuracy": 0.18600744605064393,
      "num_tokens": 46754214.0,
      "step": 21340
    },
    {
      "entropy": 5.593805646896362,
      "epoch": 2.118400158793172,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00045537375402775334,
      "loss": 5.0441,
      "mean_token_accuracy": 0.17925716787576676,
      "num_tokens": 46764112.0,
      "step": 21345
    },
    {
      "entropy": 5.554880094528198,
      "epoch": 2.1188963874553393,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00045535257159594724,
      "loss": 5.0067,
      "mean_token_accuracy": 0.18660915940999984,
      "num_tokens": 46775202.0,
      "step": 21350
    },
    {
      "entropy": 5.699235200881958,
      "epoch": 2.1193926161175067,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000455331384691786,
      "loss": 5.0569,
      "mean_token_accuracy": 0.17765737175941468,
      "num_tokens": 46786557.0,
      "step": 21355
    },
    {
      "entropy": 5.577398157119751,
      "epoch": 2.1198888447796747,
      "grad_norm": 1.125,
      "learning_rate": 0.0004553101933157949,
      "loss": 5.0301,
      "mean_token_accuracy": 0.16968678534030915,
      "num_tokens": 46797082.0,
      "step": 21360
    },
    {
      "entropy": 5.548353099822998,
      "epoch": 2.120385073441842,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045528899746849965,
      "loss": 5.0615,
      "mean_token_accuracy": 0.17988609224557878,
      "num_tokens": 46806679.0,
      "step": 21365
    },
    {
      "entropy": 5.666635942459107,
      "epoch": 2.1208813021040096,
      "grad_norm": 1.0,
      "learning_rate": 0.0004552677971504257,
      "loss": 5.1856,
      "mean_token_accuracy": 0.16439101248979568,
      "num_tokens": 46818169.0,
      "step": 21370
    },
    {
      "entropy": 5.750371599197388,
      "epoch": 2.121377530766177,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004552465923620988,
      "loss": 5.2345,
      "mean_token_accuracy": 0.1687205970287323,
      "num_tokens": 46829552.0,
      "step": 21375
    },
    {
      "entropy": 5.646593999862671,
      "epoch": 2.1218737594283446,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004552253831040448,
      "loss": 5.1341,
      "mean_token_accuracy": 0.1758819043636322,
      "num_tokens": 46840808.0,
      "step": 21380
    },
    {
      "entropy": 5.631982135772705,
      "epoch": 2.122369988090512,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045520416937678973,
      "loss": 5.0601,
      "mean_token_accuracy": 0.1720468893647194,
      "num_tokens": 46852066.0,
      "step": 21385
    },
    {
      "entropy": 5.577331781387329,
      "epoch": 2.1228662167526795,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045518295118085955,
      "loss": 4.9917,
      "mean_token_accuracy": 0.18183973878622056,
      "num_tokens": 46863163.0,
      "step": 21390
    },
    {
      "entropy": 5.616700220108032,
      "epoch": 2.123362445414847,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045516172851678036,
      "loss": 5.0539,
      "mean_token_accuracy": 0.17280440777540207,
      "num_tokens": 46873246.0,
      "step": 21395
    },
    {
      "entropy": 5.569489622116089,
      "epoch": 2.123858674077015,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045514050138507857,
      "loss": 5.0692,
      "mean_token_accuracy": 0.17490116208791734,
      "num_tokens": 46883881.0,
      "step": 21400
    },
    {
      "entropy": 5.6027342796325685,
      "epoch": 2.1243549027391824,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004551192697862805,
      "loss": 5.1019,
      "mean_token_accuracy": 0.1771396741271019,
      "num_tokens": 46894907.0,
      "step": 21405
    },
    {
      "entropy": 5.57938060760498,
      "epoch": 2.12485113140135,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004550980337209127,
      "loss": 5.0046,
      "mean_token_accuracy": 0.1740194231271744,
      "num_tokens": 46905220.0,
      "step": 21410
    },
    {
      "entropy": 5.564579582214355,
      "epoch": 2.1253473600635173,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045507679318950166,
      "loss": 5.0103,
      "mean_token_accuracy": 0.1850699871778488,
      "num_tokens": 46916328.0,
      "step": 21415
    },
    {
      "entropy": 5.766316890716553,
      "epoch": 2.125843588725685,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004550555481925743,
      "loss": 5.2208,
      "mean_token_accuracy": 0.16631329655647278,
      "num_tokens": 46927450.0,
      "step": 21420
    },
    {
      "entropy": 5.5532958984375,
      "epoch": 2.1263398173878523,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045503429873065723,
      "loss": 4.9692,
      "mean_token_accuracy": 0.18178583979606627,
      "num_tokens": 46936758.0,
      "step": 21425
    },
    {
      "entropy": 5.501639175415039,
      "epoch": 2.1268360460500197,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004550130448042775,
      "loss": 4.9285,
      "mean_token_accuracy": 0.18358418792486192,
      "num_tokens": 46945776.0,
      "step": 21430
    },
    {
      "entropy": 5.5447389602661135,
      "epoch": 2.127332274712187,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004549917864139622,
      "loss": 5.0362,
      "mean_token_accuracy": 0.18409579396247863,
      "num_tokens": 46955621.0,
      "step": 21435
    },
    {
      "entropy": 5.7072838306427,
      "epoch": 2.1278285033743547,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004549705235602384,
      "loss": 5.1466,
      "mean_token_accuracy": 0.17853861302137375,
      "num_tokens": 46966890.0,
      "step": 21440
    },
    {
      "entropy": 5.624867057800293,
      "epoch": 2.1283247320365226,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004549492562436334,
      "loss": 5.1556,
      "mean_token_accuracy": 0.17566467076539993,
      "num_tokens": 46978380.0,
      "step": 21445
    },
    {
      "entropy": 5.585629653930664,
      "epoch": 2.12882096069869,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045492798446467465,
      "loss": 4.971,
      "mean_token_accuracy": 0.18519706428050994,
      "num_tokens": 46988286.0,
      "step": 21450
    },
    {
      "entropy": 5.5865966796875,
      "epoch": 2.1293171893608576,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045490670822388957,
      "loss": 4.991,
      "mean_token_accuracy": 0.1812462627887726,
      "num_tokens": 46997795.0,
      "step": 21455
    },
    {
      "entropy": 5.516621398925781,
      "epoch": 2.129813418023025,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004548854275218059,
      "loss": 4.8902,
      "mean_token_accuracy": 0.19194891452789306,
      "num_tokens": 47007970.0,
      "step": 21460
    },
    {
      "entropy": 5.539976692199707,
      "epoch": 2.1303096466851925,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045486414235895115,
      "loss": 5.0138,
      "mean_token_accuracy": 0.18090228438377381,
      "num_tokens": 47019548.0,
      "step": 21465
    },
    {
      "entropy": 5.584608554840088,
      "epoch": 2.13080587534736,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004548428527358532,
      "loss": 5.1162,
      "mean_token_accuracy": 0.1752985894680023,
      "num_tokens": 47029568.0,
      "step": 21470
    },
    {
      "entropy": 5.660408687591553,
      "epoch": 2.1313021040095275,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004548215586530401,
      "loss": 5.1133,
      "mean_token_accuracy": 0.17004141807556153,
      "num_tokens": 47042048.0,
      "step": 21475
    },
    {
      "entropy": 5.582299661636353,
      "epoch": 2.131798332671695,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045480026011103975,
      "loss": 4.9804,
      "mean_token_accuracy": 0.17573142796754837,
      "num_tokens": 47052099.0,
      "step": 21480
    },
    {
      "entropy": 5.592738914489746,
      "epoch": 2.132294561333863,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00045477895711038045,
      "loss": 5.015,
      "mean_token_accuracy": 0.18177002668380737,
      "num_tokens": 47063310.0,
      "step": 21485
    },
    {
      "entropy": 5.58268518447876,
      "epoch": 2.1327907899960303,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004547576496515904,
      "loss": 5.044,
      "mean_token_accuracy": 0.17684397399425505,
      "num_tokens": 47073545.0,
      "step": 21490
    },
    {
      "entropy": 5.565859699249268,
      "epoch": 2.133287018658198,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00045473633773519795,
      "loss": 5.0895,
      "mean_token_accuracy": 0.17818187475204467,
      "num_tokens": 47085456.0,
      "step": 21495
    },
    {
      "entropy": 5.6535240650177006,
      "epoch": 2.1337832473203653,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045471502136173173,
      "loss": 5.0619,
      "mean_token_accuracy": 0.17919227927923204,
      "num_tokens": 47096590.0,
      "step": 21500
    },
    {
      "entropy": 5.678283214569092,
      "epoch": 2.1342794759825328,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004546937005317201,
      "loss": 5.0062,
      "mean_token_accuracy": 0.18891891837120056,
      "num_tokens": 47107876.0,
      "step": 21505
    },
    {
      "entropy": 5.577515172958374,
      "epoch": 2.1347757046447002,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045467237524569193,
      "loss": 5.0449,
      "mean_token_accuracy": 0.17991174459457399,
      "num_tokens": 47117662.0,
      "step": 21510
    },
    {
      "entropy": 5.562215518951416,
      "epoch": 2.1352719333068677,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045465104550417604,
      "loss": 5.0753,
      "mean_token_accuracy": 0.17737892866134644,
      "num_tokens": 47129283.0,
      "step": 21515
    },
    {
      "entropy": 5.664295816421509,
      "epoch": 2.135768161969035,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045462971130770123,
      "loss": 5.1532,
      "mean_token_accuracy": 0.17320260852575303,
      "num_tokens": 47140824.0,
      "step": 21520
    },
    {
      "entropy": 5.575448846817016,
      "epoch": 2.1362643906312027,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004546083726567968,
      "loss": 5.0058,
      "mean_token_accuracy": 0.18087561279535294,
      "num_tokens": 47151733.0,
      "step": 21525
    },
    {
      "entropy": 5.583763504028321,
      "epoch": 2.1367606192933706,
      "grad_norm": 1.0,
      "learning_rate": 0.0004545870295519917,
      "loss": 5.1205,
      "mean_token_accuracy": 0.16902925968170165,
      "num_tokens": 47162846.0,
      "step": 21530
    },
    {
      "entropy": 5.631829500198364,
      "epoch": 2.137256847955538,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045456568199381524,
      "loss": 5.0145,
      "mean_token_accuracy": 0.1806974560022354,
      "num_tokens": 47174969.0,
      "step": 21535
    },
    {
      "entropy": 5.666506385803222,
      "epoch": 2.1377530766177055,
      "grad_norm": 1.0,
      "learning_rate": 0.00045454432998279676,
      "loss": 5.1054,
      "mean_token_accuracy": 0.17508367747068404,
      "num_tokens": 47186903.0,
      "step": 21540
    },
    {
      "entropy": 5.617109107971191,
      "epoch": 2.138249305279873,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004545229735194658,
      "loss": 5.0834,
      "mean_token_accuracy": 0.1783437430858612,
      "num_tokens": 47197716.0,
      "step": 21545
    },
    {
      "entropy": 5.604199981689453,
      "epoch": 2.1387455339420405,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045450161260435196,
      "loss": 5.0092,
      "mean_token_accuracy": 0.1813771605491638,
      "num_tokens": 47208623.0,
      "step": 21550
    },
    {
      "entropy": 5.547342920303345,
      "epoch": 2.139241762604208,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004544802472379849,
      "loss": 5.0182,
      "mean_token_accuracy": 0.18133753836154937,
      "num_tokens": 47218248.0,
      "step": 21555
    },
    {
      "entropy": 5.6151481628417965,
      "epoch": 2.1397379912663754,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004544588774208945,
      "loss": 5.1487,
      "mean_token_accuracy": 0.17479496449232101,
      "num_tokens": 47232028.0,
      "step": 21560
    },
    {
      "entropy": 5.662732410430908,
      "epoch": 2.140234219928543,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045443750315361055,
      "loss": 5.015,
      "mean_token_accuracy": 0.18409377932548524,
      "num_tokens": 47241976.0,
      "step": 21565
    },
    {
      "entropy": 5.601552867889405,
      "epoch": 2.140730448590711,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004544161244366632,
      "loss": 5.0221,
      "mean_token_accuracy": 0.1823471352458,
      "num_tokens": 47253228.0,
      "step": 21570
    },
    {
      "entropy": 5.578529977798462,
      "epoch": 2.1412266772528783,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004543947412705826,
      "loss": 5.0402,
      "mean_token_accuracy": 0.1810141086578369,
      "num_tokens": 47265210.0,
      "step": 21575
    },
    {
      "entropy": 5.560053873062134,
      "epoch": 2.1417229059150458,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045437335365589905,
      "loss": 5.0548,
      "mean_token_accuracy": 0.17746611684560776,
      "num_tokens": 47275723.0,
      "step": 21580
    },
    {
      "entropy": 5.6039714336395265,
      "epoch": 2.1422191345772132,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045435196159314276,
      "loss": 5.0951,
      "mean_token_accuracy": 0.17851237952709198,
      "num_tokens": 47286469.0,
      "step": 21585
    },
    {
      "entropy": 5.668009424209595,
      "epoch": 2.1427153632393807,
      "grad_norm": 0.984375,
      "learning_rate": 0.00045433056508284425,
      "loss": 5.1128,
      "mean_token_accuracy": 0.17833200544118882,
      "num_tokens": 47296461.0,
      "step": 21590
    },
    {
      "entropy": 5.643115234375,
      "epoch": 2.143211591901548,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004543091641255342,
      "loss": 5.0124,
      "mean_token_accuracy": 0.18718140125274657,
      "num_tokens": 47307217.0,
      "step": 21595
    },
    {
      "entropy": 5.6398458003997805,
      "epoch": 2.1437078205637157,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004542877587217432,
      "loss": 5.1423,
      "mean_token_accuracy": 0.17120454907417298,
      "num_tokens": 47317041.0,
      "step": 21600
    },
    {
      "entropy": 5.5971814632415775,
      "epoch": 2.144204049225883,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045426634887200224,
      "loss": 5.05,
      "mean_token_accuracy": 0.17875636965036393,
      "num_tokens": 47328185.0,
      "step": 21605
    },
    {
      "entropy": 5.639533615112304,
      "epoch": 2.1447002778880506,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000454244934576842,
      "loss": 5.0975,
      "mean_token_accuracy": 0.17185043394565583,
      "num_tokens": 47339984.0,
      "step": 21610
    },
    {
      "entropy": 5.609605741500855,
      "epoch": 2.1451965065502185,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045422351583679365,
      "loss": 5.0474,
      "mean_token_accuracy": 0.18795174956321717,
      "num_tokens": 47350911.0,
      "step": 21615
    },
    {
      "entropy": 5.610928535461426,
      "epoch": 2.145692735212386,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004542020926523883,
      "loss": 5.0395,
      "mean_token_accuracy": 0.1735944226384163,
      "num_tokens": 47361703.0,
      "step": 21620
    },
    {
      "entropy": 5.568384313583374,
      "epoch": 2.1461889638745535,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004541806650241572,
      "loss": 5.0093,
      "mean_token_accuracy": 0.17709755897521973,
      "num_tokens": 47372878.0,
      "step": 21625
    },
    {
      "entropy": 5.575262784957886,
      "epoch": 2.146685192536721,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004541592329526317,
      "loss": 5.0408,
      "mean_token_accuracy": 0.17615410089492797,
      "num_tokens": 47383056.0,
      "step": 21630
    },
    {
      "entropy": 5.559898614883423,
      "epoch": 2.1471814211988884,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004541377964383434,
      "loss": 5.0717,
      "mean_token_accuracy": 0.1783818393945694,
      "num_tokens": 47394160.0,
      "step": 21635
    },
    {
      "entropy": 5.630073499679566,
      "epoch": 2.147677649861056,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004541163554818237,
      "loss": 5.0163,
      "mean_token_accuracy": 0.17626212686300277,
      "num_tokens": 47404861.0,
      "step": 21640
    },
    {
      "entropy": 5.636077785491944,
      "epoch": 2.1481738785232234,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045409491008360423,
      "loss": 5.0962,
      "mean_token_accuracy": 0.17747374922037124,
      "num_tokens": 47415236.0,
      "step": 21645
    },
    {
      "entropy": 5.627088594436645,
      "epoch": 2.148670107185391,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004540734602442169,
      "loss": 5.0511,
      "mean_token_accuracy": 0.17763217240571977,
      "num_tokens": 47425789.0,
      "step": 21650
    },
    {
      "entropy": 5.551941680908203,
      "epoch": 2.149166335847559,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004540520059641937,
      "loss": 4.9286,
      "mean_token_accuracy": 0.18573008775711058,
      "num_tokens": 47436236.0,
      "step": 21655
    },
    {
      "entropy": 5.515349864959717,
      "epoch": 2.1496625645097263,
      "grad_norm": 0.875,
      "learning_rate": 0.0004540305472440666,
      "loss": 4.9507,
      "mean_token_accuracy": 0.18640548586845399,
      "num_tokens": 47447041.0,
      "step": 21660
    },
    {
      "entropy": 5.571415376663208,
      "epoch": 2.1501587931718937,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004540090840843677,
      "loss": 5.0967,
      "mean_token_accuracy": 0.17228649109601973,
      "num_tokens": 47458001.0,
      "step": 21665
    },
    {
      "entropy": 5.560210418701172,
      "epoch": 2.150655021834061,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004539876164856292,
      "loss": 4.9613,
      "mean_token_accuracy": 0.18584735095500945,
      "num_tokens": 47468621.0,
      "step": 21670
    },
    {
      "entropy": 5.553755044937134,
      "epoch": 2.1511512504962287,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004539661444483835,
      "loss": 5.0842,
      "mean_token_accuracy": 0.17947480976581573,
      "num_tokens": 47478957.0,
      "step": 21675
    },
    {
      "entropy": 5.559771156311035,
      "epoch": 2.151647479158396,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004539446679731631,
      "loss": 5.0215,
      "mean_token_accuracy": 0.18373374491930008,
      "num_tokens": 47489640.0,
      "step": 21680
    },
    {
      "entropy": 5.651854991912842,
      "epoch": 2.1521437078205636,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004539231870605005,
      "loss": 5.1519,
      "mean_token_accuracy": 0.17401155084371567,
      "num_tokens": 47501581.0,
      "step": 21685
    },
    {
      "entropy": 5.6084897994995115,
      "epoch": 2.152639936482731,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00045390170171092845,
      "loss": 5.0834,
      "mean_token_accuracy": 0.17708443850278854,
      "num_tokens": 47513605.0,
      "step": 21690
    },
    {
      "entropy": 5.658263921737671,
      "epoch": 2.1531361651448986,
      "grad_norm": 0.953125,
      "learning_rate": 0.00045388021192497956,
      "loss": 5.0849,
      "mean_token_accuracy": 0.1809326931834221,
      "num_tokens": 47523784.0,
      "step": 21695
    },
    {
      "entropy": 5.575326347351075,
      "epoch": 2.1536323938070665,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000453858717703187,
      "loss": 4.9834,
      "mean_token_accuracy": 0.17962647080421448,
      "num_tokens": 47534643.0,
      "step": 21700
    },
    {
      "entropy": 5.583701038360596,
      "epoch": 2.154128622469234,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004538372190460836,
      "loss": 5.0903,
      "mean_token_accuracy": 0.17195199579000472,
      "num_tokens": 47545920.0,
      "step": 21705
    },
    {
      "entropy": 5.5628283500671385,
      "epoch": 2.1546248511314015,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004538157159542025,
      "loss": 5.0381,
      "mean_token_accuracy": 0.17441756576299666,
      "num_tokens": 47557024.0,
      "step": 21710
    },
    {
      "entropy": 5.595124387741089,
      "epoch": 2.155121079793569,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004537942084280771,
      "loss": 5.0093,
      "mean_token_accuracy": 0.18362217545509338,
      "num_tokens": 47567531.0,
      "step": 21715
    },
    {
      "entropy": 5.585353660583496,
      "epoch": 2.1556173084557364,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00045377269646824036,
      "loss": 5.1367,
      "mean_token_accuracy": 0.17484066337347032,
      "num_tokens": 47578642.0,
      "step": 21720
    },
    {
      "entropy": 5.6135828495025635,
      "epoch": 2.156113537117904,
      "grad_norm": 0.953125,
      "learning_rate": 0.00045375118007522607,
      "loss": 5.1837,
      "mean_token_accuracy": 0.1732584223151207,
      "num_tokens": 47589346.0,
      "step": 21725
    },
    {
      "entropy": 5.559467029571533,
      "epoch": 2.1566097657800714,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004537296592495678,
      "loss": 4.9829,
      "mean_token_accuracy": 0.18416567593812944,
      "num_tokens": 47599689.0,
      "step": 21730
    },
    {
      "entropy": 5.640780639648438,
      "epoch": 2.157105994442239,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000453708133991799,
      "loss": 5.0802,
      "mean_token_accuracy": 0.17734809070825577,
      "num_tokens": 47609214.0,
      "step": 21735
    },
    {
      "entropy": 5.633480453491211,
      "epoch": 2.1576022231044063,
      "grad_norm": 0.984375,
      "learning_rate": 0.00045368660430245364,
      "loss": 5.0252,
      "mean_token_accuracy": 0.17884812653064727,
      "num_tokens": 47620119.0,
      "step": 21740
    },
    {
      "entropy": 5.592954492568969,
      "epoch": 2.158098451766574,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004536650701820654,
      "loss": 4.9973,
      "mean_token_accuracy": 0.17908246219158172,
      "num_tokens": 47630085.0,
      "step": 21745
    },
    {
      "entropy": 5.571291160583496,
      "epoch": 2.1585946804287417,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004536435316311685,
      "loss": 4.9873,
      "mean_token_accuracy": 0.18062832802534104,
      "num_tokens": 47641457.0,
      "step": 21750
    },
    {
      "entropy": 5.591070032119751,
      "epoch": 2.159090909090909,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045362198865029696,
      "loss": 5.012,
      "mean_token_accuracy": 0.17439776957035064,
      "num_tokens": 47652269.0,
      "step": 21755
    },
    {
      "entropy": 5.578510570526123,
      "epoch": 2.1595871377530766,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045360044123998495,
      "loss": 5.0399,
      "mean_token_accuracy": 0.17895068377256393,
      "num_tokens": 47664601.0,
      "step": 21760
    },
    {
      "entropy": 5.627600049972534,
      "epoch": 2.160083366415244,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045357888940076685,
      "loss": 5.1446,
      "mean_token_accuracy": 0.1762605682015419,
      "num_tokens": 47676187.0,
      "step": 21765
    },
    {
      "entropy": 5.554150152206421,
      "epoch": 2.1605795950774116,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004535573331331771,
      "loss": 4.9491,
      "mean_token_accuracy": 0.1862216755747795,
      "num_tokens": 47687057.0,
      "step": 21770
    },
    {
      "entropy": 5.589615774154663,
      "epoch": 2.161075823739579,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00045353577243775035,
      "loss": 5.001,
      "mean_token_accuracy": 0.18201490342617035,
      "num_tokens": 47699074.0,
      "step": 21775
    },
    {
      "entropy": 5.601614141464234,
      "epoch": 2.1615720524017465,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004535142073150211,
      "loss": 5.0044,
      "mean_token_accuracy": 0.18161804229021072,
      "num_tokens": 47710400.0,
      "step": 21780
    },
    {
      "entropy": 5.571734857559204,
      "epoch": 2.162068281063914,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004534926377655241,
      "loss": 4.8857,
      "mean_token_accuracy": 0.19429706782102585,
      "num_tokens": 47720900.0,
      "step": 21785
    },
    {
      "entropy": 5.532523965835571,
      "epoch": 2.162564509726082,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045347106378979435,
      "loss": 4.9175,
      "mean_token_accuracy": 0.18466942608356476,
      "num_tokens": 47730702.0,
      "step": 21790
    },
    {
      "entropy": 5.5342902660369875,
      "epoch": 2.1630607383882494,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045344948538836674,
      "loss": 4.994,
      "mean_token_accuracy": 0.18384909480810166,
      "num_tokens": 47740451.0,
      "step": 21795
    },
    {
      "entropy": 5.598954153060913,
      "epoch": 2.163556967050417,
      "grad_norm": 0.96875,
      "learning_rate": 0.00045342790256177646,
      "loss": 5.0729,
      "mean_token_accuracy": 0.18085743933916093,
      "num_tokens": 47752285.0,
      "step": 21800
    },
    {
      "entropy": 5.642742872238159,
      "epoch": 2.1640531957125844,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004534063153105587,
      "loss": 5.0656,
      "mean_token_accuracy": 0.17590310126543046,
      "num_tokens": 47761895.0,
      "step": 21805
    },
    {
      "entropy": 5.5979901313781735,
      "epoch": 2.164549424374752,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004533847236352487,
      "loss": 4.9629,
      "mean_token_accuracy": 0.18740931749343873,
      "num_tokens": 47771718.0,
      "step": 21810
    },
    {
      "entropy": 5.511568021774292,
      "epoch": 2.1650456530369193,
      "grad_norm": 0.953125,
      "learning_rate": 0.000453363127536382,
      "loss": 5.0046,
      "mean_token_accuracy": 0.18217131346464158,
      "num_tokens": 47782183.0,
      "step": 21815
    },
    {
      "entropy": 5.48533787727356,
      "epoch": 2.165541881699087,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045334152701449407,
      "loss": 4.9516,
      "mean_token_accuracy": 0.1853344812989235,
      "num_tokens": 47792190.0,
      "step": 21820
    },
    {
      "entropy": 5.67718653678894,
      "epoch": 2.1660381103612543,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004533199220701205,
      "loss": 5.1564,
      "mean_token_accuracy": 0.17121699452400208,
      "num_tokens": 47803245.0,
      "step": 21825
    },
    {
      "entropy": 5.64677996635437,
      "epoch": 2.166534339023422,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045329831270379714,
      "loss": 5.0933,
      "mean_token_accuracy": 0.1739192485809326,
      "num_tokens": 47814605.0,
      "step": 21830
    },
    {
      "entropy": 5.582768774032592,
      "epoch": 2.1670305676855897,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00045327669891605975,
      "loss": 5.0682,
      "mean_token_accuracy": 0.18000198155641556,
      "num_tokens": 47825945.0,
      "step": 21835
    },
    {
      "entropy": 5.546384859085083,
      "epoch": 2.167526796347757,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004532550807074445,
      "loss": 5.066,
      "mean_token_accuracy": 0.1756667673587799,
      "num_tokens": 47838119.0,
      "step": 21840
    },
    {
      "entropy": 5.613620281219482,
      "epoch": 2.1680230250099246,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004532334580784873,
      "loss": 5.0584,
      "mean_token_accuracy": 0.18069804608821868,
      "num_tokens": 47849560.0,
      "step": 21845
    },
    {
      "entropy": 5.650173473358154,
      "epoch": 2.168519253672092,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004532118310297244,
      "loss": 5.0721,
      "mean_token_accuracy": 0.1677744470536709,
      "num_tokens": 47860036.0,
      "step": 21850
    },
    {
      "entropy": 5.518974924087525,
      "epoch": 2.1690154823342596,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004531901995616921,
      "loss": 4.9548,
      "mean_token_accuracy": 0.18143303841352462,
      "num_tokens": 47870826.0,
      "step": 21855
    },
    {
      "entropy": 5.575003433227539,
      "epoch": 2.169511710996427,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004531685636749268,
      "loss": 5.0753,
      "mean_token_accuracy": 0.1716070294380188,
      "num_tokens": 47881415.0,
      "step": 21860
    },
    {
      "entropy": 5.651249122619629,
      "epoch": 2.1700079396585945,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004531469233699651,
      "loss": 5.0786,
      "mean_token_accuracy": 0.17890810072422028,
      "num_tokens": 47893343.0,
      "step": 21865
    },
    {
      "entropy": 5.5685631275177006,
      "epoch": 2.170504168320762,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004531252786473436,
      "loss": 5.0631,
      "mean_token_accuracy": 0.1751348689198494,
      "num_tokens": 47904871.0,
      "step": 21870
    },
    {
      "entropy": 5.562549304962158,
      "epoch": 2.17100039698293,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004531036295075989,
      "loss": 4.9804,
      "mean_token_accuracy": 0.18189692348241807,
      "num_tokens": 47915367.0,
      "step": 21875
    },
    {
      "entropy": 5.5944239616394045,
      "epoch": 2.1714966256450974,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00045308197595126797,
      "loss": 4.9977,
      "mean_token_accuracy": 0.1846961185336113,
      "num_tokens": 47925987.0,
      "step": 21880
    },
    {
      "entropy": 5.604113674163818,
      "epoch": 2.171992854307265,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004530603179788879,
      "loss": 5.0887,
      "mean_token_accuracy": 0.18021042346954347,
      "num_tokens": 47937688.0,
      "step": 21885
    },
    {
      "entropy": 5.566831064224243,
      "epoch": 2.1724890829694323,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004530386555909955,
      "loss": 5.041,
      "mean_token_accuracy": 0.18573734015226365,
      "num_tokens": 47948550.0,
      "step": 21890
    },
    {
      "entropy": 5.604857778549194,
      "epoch": 2.1729853116316,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045301698878812813,
      "loss": 5.0379,
      "mean_token_accuracy": 0.1845029205083847,
      "num_tokens": 47957982.0,
      "step": 21895
    },
    {
      "entropy": 5.624063301086426,
      "epoch": 2.1734815402937673,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004529953175708231,
      "loss": 5.0848,
      "mean_token_accuracy": 0.17916077822446824,
      "num_tokens": 47969562.0,
      "step": 21900
    },
    {
      "entropy": 5.604766082763672,
      "epoch": 2.1739777689559348,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00045297364193961763,
      "loss": 5.1039,
      "mean_token_accuracy": 0.17346563190221786,
      "num_tokens": 47981323.0,
      "step": 21905
    },
    {
      "entropy": 5.609239768981934,
      "epoch": 2.1744739976181022,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045295196189504934,
      "loss": 5.0263,
      "mean_token_accuracy": 0.17552904635667801,
      "num_tokens": 47991454.0,
      "step": 21910
    },
    {
      "entropy": 5.564348745346069,
      "epoch": 2.17497022628027,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045293027743765595,
      "loss": 4.9135,
      "mean_token_accuracy": 0.1843430742621422,
      "num_tokens": 48001300.0,
      "step": 21915
    },
    {
      "entropy": 5.591091442108154,
      "epoch": 2.1754664549424376,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045290858856797503,
      "loss": 5.0179,
      "mean_token_accuracy": 0.18275899440050125,
      "num_tokens": 48012608.0,
      "step": 21920
    },
    {
      "entropy": 5.616801643371582,
      "epoch": 2.175962683604605,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045288689528654444,
      "loss": 5.0416,
      "mean_token_accuracy": 0.18189008086919783,
      "num_tokens": 48022120.0,
      "step": 21925
    },
    {
      "entropy": 5.600833606719971,
      "epoch": 2.1764589122667726,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045286519759390217,
      "loss": 5.0471,
      "mean_token_accuracy": 0.1793038547039032,
      "num_tokens": 48032629.0,
      "step": 21930
    },
    {
      "entropy": 5.564683151245117,
      "epoch": 2.17695514092894,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004528434954905863,
      "loss": 4.9657,
      "mean_token_accuracy": 0.18391641229391098,
      "num_tokens": 48042934.0,
      "step": 21935
    },
    {
      "entropy": 5.576721429824829,
      "epoch": 2.1774513695911075,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004528217889771349,
      "loss": 5.0524,
      "mean_token_accuracy": 0.1794678181409836,
      "num_tokens": 48052825.0,
      "step": 21940
    },
    {
      "entropy": 5.639779329299927,
      "epoch": 2.177947598253275,
      "grad_norm": 1.0,
      "learning_rate": 0.00045280007805408636,
      "loss": 5.1136,
      "mean_token_accuracy": 0.1706728979945183,
      "num_tokens": 48064565.0,
      "step": 21945
    },
    {
      "entropy": 5.648127508163452,
      "epoch": 2.1784438269154425,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00045277836272197895,
      "loss": 5.0776,
      "mean_token_accuracy": 0.18024928718805314,
      "num_tokens": 48076006.0,
      "step": 21950
    },
    {
      "entropy": 5.675761795043945,
      "epoch": 2.17894005557761,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045275664298135134,
      "loss": 5.182,
      "mean_token_accuracy": 0.17125353962183,
      "num_tokens": 48087110.0,
      "step": 21955
    },
    {
      "entropy": 5.618146514892578,
      "epoch": 2.179436284239778,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004527349188327419,
      "loss": 5.0635,
      "mean_token_accuracy": 0.1855880782008171,
      "num_tokens": 48098894.0,
      "step": 21960
    },
    {
      "entropy": 5.6330725193023685,
      "epoch": 2.1799325129019453,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004527131902766895,
      "loss": 5.0785,
      "mean_token_accuracy": 0.17861806303262712,
      "num_tokens": 48109729.0,
      "step": 21965
    },
    {
      "entropy": 5.625303792953491,
      "epoch": 2.180428741564113,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004526914573137329,
      "loss": 5.071,
      "mean_token_accuracy": 0.17511035948991777,
      "num_tokens": 48120345.0,
      "step": 21970
    },
    {
      "entropy": 5.609857416152954,
      "epoch": 2.1809249702262803,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045266971994441116,
      "loss": 5.0109,
      "mean_token_accuracy": 0.1806703433394432,
      "num_tokens": 48130185.0,
      "step": 21975
    },
    {
      "entropy": 5.555281162261963,
      "epoch": 2.1814211988884478,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000452647978169263,
      "loss": 4.9787,
      "mean_token_accuracy": 0.18263042122125625,
      "num_tokens": 48141020.0,
      "step": 21980
    },
    {
      "entropy": 5.572640323638916,
      "epoch": 2.1819174275506152,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045262623198882797,
      "loss": 5.064,
      "mean_token_accuracy": 0.18618489056825638,
      "num_tokens": 48153361.0,
      "step": 21985
    },
    {
      "entropy": 5.6059380054473875,
      "epoch": 2.1824136562127827,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004526044814036451,
      "loss": 5.0251,
      "mean_token_accuracy": 0.17799050956964493,
      "num_tokens": 48163604.0,
      "step": 21990
    },
    {
      "entropy": 5.58221960067749,
      "epoch": 2.18290988487495,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045258272641425377,
      "loss": 5.0414,
      "mean_token_accuracy": 0.17689698189496994,
      "num_tokens": 48174732.0,
      "step": 21995
    },
    {
      "entropy": 5.618593025207519,
      "epoch": 2.183406113537118,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004525609670211934,
      "loss": 5.0081,
      "mean_token_accuracy": 0.1845436543226242,
      "num_tokens": 48187082.0,
      "step": 22000
    },
    {
      "entropy": 5.605060577392578,
      "epoch": 2.1839023421992856,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045253920322500373,
      "loss": 5.0222,
      "mean_token_accuracy": 0.18025557845830917,
      "num_tokens": 48196547.0,
      "step": 22005
    },
    {
      "entropy": 5.580152559280395,
      "epoch": 2.184398570861453,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00045251743502622445,
      "loss": 5.0699,
      "mean_token_accuracy": 0.1801171272993088,
      "num_tokens": 48207404.0,
      "step": 22010
    },
    {
      "entropy": 5.547487783432007,
      "epoch": 2.1848947995236205,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004524956624253952,
      "loss": 5.0439,
      "mean_token_accuracy": 0.18376276940107344,
      "num_tokens": 48217093.0,
      "step": 22015
    },
    {
      "entropy": 5.603080797195434,
      "epoch": 2.185391028185788,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045247388542305595,
      "loss": 5.0678,
      "mean_token_accuracy": 0.17807838767766954,
      "num_tokens": 48228093.0,
      "step": 22020
    },
    {
      "entropy": 5.561899852752686,
      "epoch": 2.1858872568479555,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004524521040197469,
      "loss": 5.033,
      "mean_token_accuracy": 0.18567698299884797,
      "num_tokens": 48238834.0,
      "step": 22025
    },
    {
      "entropy": 5.645340538024902,
      "epoch": 2.186383485510123,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00045243031821600794,
      "loss": 5.0757,
      "mean_token_accuracy": 0.18889536261558532,
      "num_tokens": 48250782.0,
      "step": 22030
    },
    {
      "entropy": 5.500328874588012,
      "epoch": 2.1868797141722904,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004524085280123795,
      "loss": 4.9717,
      "mean_token_accuracy": 0.1805126041173935,
      "num_tokens": 48261711.0,
      "step": 22035
    },
    {
      "entropy": 5.574241590499878,
      "epoch": 2.187375942834458,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004523867334094018,
      "loss": 5.0873,
      "mean_token_accuracy": 0.17435535192489623,
      "num_tokens": 48273505.0,
      "step": 22040
    },
    {
      "entropy": 5.560463953018188,
      "epoch": 2.187872171496626,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004523649344076154,
      "loss": 4.9822,
      "mean_token_accuracy": 0.18139338344335557,
      "num_tokens": 48284639.0,
      "step": 22045
    },
    {
      "entropy": 5.596578931808471,
      "epoch": 2.1883684001587933,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00045234313100756077,
      "loss": 5.0647,
      "mean_token_accuracy": 0.1765772208571434,
      "num_tokens": 48296275.0,
      "step": 22050
    },
    {
      "entropy": 5.557593774795532,
      "epoch": 2.188864628820961,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045232132320977854,
      "loss": 4.9709,
      "mean_token_accuracy": 0.18602284938097,
      "num_tokens": 48306859.0,
      "step": 22055
    },
    {
      "entropy": 5.609091567993164,
      "epoch": 2.1893608574831283,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004522995110148097,
      "loss": 5.1127,
      "mean_token_accuracy": 0.18117768466472625,
      "num_tokens": 48317561.0,
      "step": 22060
    },
    {
      "entropy": 5.5942830562591555,
      "epoch": 2.1898570861452957,
      "grad_norm": 0.9375,
      "learning_rate": 0.000452277694423195,
      "loss": 5.1209,
      "mean_token_accuracy": 0.16993902176618575,
      "num_tokens": 48327814.0,
      "step": 22065
    },
    {
      "entropy": 5.586060476303101,
      "epoch": 2.190353314807463,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004522558734354755,
      "loss": 4.9752,
      "mean_token_accuracy": 0.18480440974235535,
      "num_tokens": 48339658.0,
      "step": 22070
    },
    {
      "entropy": 5.6134397983551025,
      "epoch": 2.1908495434696307,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045223404805219225,
      "loss": 5.0883,
      "mean_token_accuracy": 0.17792799770832063,
      "num_tokens": 48352115.0,
      "step": 22075
    },
    {
      "entropy": 5.6445409774780275,
      "epoch": 2.191345772131798,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045221221827388655,
      "loss": 5.0387,
      "mean_token_accuracy": 0.18001949936151504,
      "num_tokens": 48362378.0,
      "step": 22080
    },
    {
      "entropy": 5.578653192520141,
      "epoch": 2.191842000793966,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004521903841010997,
      "loss": 4.9759,
      "mean_token_accuracy": 0.1877150759100914,
      "num_tokens": 48373035.0,
      "step": 22085
    },
    {
      "entropy": 5.534437036514282,
      "epoch": 2.1923382294561335,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004521685455343731,
      "loss": 5.0309,
      "mean_token_accuracy": 0.18416130542755127,
      "num_tokens": 48383737.0,
      "step": 22090
    },
    {
      "entropy": 5.56206693649292,
      "epoch": 2.192834458118301,
      "grad_norm": 0.96875,
      "learning_rate": 0.00045214670257424835,
      "loss": 5.0225,
      "mean_token_accuracy": 0.17904094606637955,
      "num_tokens": 48394547.0,
      "step": 22095
    },
    {
      "entropy": 5.609930038452148,
      "epoch": 2.1933306867804685,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004521248552212671,
      "loss": 5.0893,
      "mean_token_accuracy": 0.17812312692403792,
      "num_tokens": 48404906.0,
      "step": 22100
    },
    {
      "entropy": 5.57509446144104,
      "epoch": 2.193826915442636,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004521030034759711,
      "loss": 5.0235,
      "mean_token_accuracy": 0.18149878233671188,
      "num_tokens": 48416425.0,
      "step": 22105
    },
    {
      "entropy": 5.598795938491821,
      "epoch": 2.1943231441048034,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004520811473389023,
      "loss": 4.9865,
      "mean_token_accuracy": 0.1817646488547325,
      "num_tokens": 48426967.0,
      "step": 22110
    },
    {
      "entropy": 5.612984704971313,
      "epoch": 2.194819372766971,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045205928681060255,
      "loss": 5.1058,
      "mean_token_accuracy": 0.18466767072677612,
      "num_tokens": 48439720.0,
      "step": 22115
    },
    {
      "entropy": 5.585451602935791,
      "epoch": 2.1953156014291384,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000452037421891614,
      "loss": 5.0926,
      "mean_token_accuracy": 0.18281929045915604,
      "num_tokens": 48449949.0,
      "step": 22120
    },
    {
      "entropy": 5.5391075134277346,
      "epoch": 2.195811830091306,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045201555258247896,
      "loss": 4.98,
      "mean_token_accuracy": 0.18449413776397705,
      "num_tokens": 48462515.0,
      "step": 22125
    },
    {
      "entropy": 5.56147837638855,
      "epoch": 2.196308058753474,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004519936788837396,
      "loss": 5.0111,
      "mean_token_accuracy": 0.18165423572063447,
      "num_tokens": 48471984.0,
      "step": 22130
    },
    {
      "entropy": 5.6573656558990475,
      "epoch": 2.1968042874156413,
      "grad_norm": 0.9375,
      "learning_rate": 0.00045197180079593836,
      "loss": 5.1004,
      "mean_token_accuracy": 0.17616873532533645,
      "num_tokens": 48482814.0,
      "step": 22135
    },
    {
      "entropy": 5.655974102020264,
      "epoch": 2.1973005160778087,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045194991831961777,
      "loss": 5.0457,
      "mean_token_accuracy": 0.18555772602558135,
      "num_tokens": 48493670.0,
      "step": 22140
    },
    {
      "entropy": 5.546384620666504,
      "epoch": 2.197796744739976,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004519280314553206,
      "loss": 5.0243,
      "mean_token_accuracy": 0.17614435255527497,
      "num_tokens": 48504758.0,
      "step": 22145
    },
    {
      "entropy": 5.6040102481842045,
      "epoch": 2.1982929734021437,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004519061402035894,
      "loss": 5.0543,
      "mean_token_accuracy": 0.1704815298318863,
      "num_tokens": 48515846.0,
      "step": 22150
    },
    {
      "entropy": 5.612509298324585,
      "epoch": 2.198789202064311,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045188424456496723,
      "loss": 5.0394,
      "mean_token_accuracy": 0.18775732070207596,
      "num_tokens": 48526136.0,
      "step": 22155
    },
    {
      "entropy": 5.641967296600342,
      "epoch": 2.1992854307264786,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045186234453999693,
      "loss": 5.1534,
      "mean_token_accuracy": 0.17862827479839324,
      "num_tokens": 48536031.0,
      "step": 22160
    },
    {
      "entropy": 5.487728214263916,
      "epoch": 2.199781659388646,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045184044012922154,
      "loss": 4.9683,
      "mean_token_accuracy": 0.18186060041189195,
      "num_tokens": 48546965.0,
      "step": 22165
    },
    {
      "entropy": 5.60263466835022,
      "epoch": 2.200277888050814,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045181853133318425,
      "loss": 5.0504,
      "mean_token_accuracy": 0.17720620930194855,
      "num_tokens": 48557637.0,
      "step": 22170
    },
    {
      "entropy": 5.631606817245483,
      "epoch": 2.2007741167129815,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004517966181524285,
      "loss": 5.0938,
      "mean_token_accuracy": 0.18279265016317367,
      "num_tokens": 48568578.0,
      "step": 22175
    },
    {
      "entropy": 5.659185266494751,
      "epoch": 2.201270345375149,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004517747005874975,
      "loss": 5.1313,
      "mean_token_accuracy": 0.17839040607213974,
      "num_tokens": 48581782.0,
      "step": 22180
    },
    {
      "entropy": 5.524752521514893,
      "epoch": 2.2017665740373165,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004517527786389349,
      "loss": 4.9938,
      "mean_token_accuracy": 0.1855344757437706,
      "num_tokens": 48592432.0,
      "step": 22185
    },
    {
      "entropy": 5.538179397583008,
      "epoch": 2.202262802699484,
      "grad_norm": 1.0,
      "learning_rate": 0.00045173085230728417,
      "loss": 5.0282,
      "mean_token_accuracy": 0.18091219216585158,
      "num_tokens": 48603251.0,
      "step": 22190
    },
    {
      "entropy": 5.633578586578369,
      "epoch": 2.2027590313616514,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004517089215930892,
      "loss": 5.1247,
      "mean_token_accuracy": 0.1747087761759758,
      "num_tokens": 48616118.0,
      "step": 22195
    },
    {
      "entropy": 5.640906906127929,
      "epoch": 2.203255260023819,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045168698649689375,
      "loss": 5.0896,
      "mean_token_accuracy": 0.17727308422327043,
      "num_tokens": 48626903.0,
      "step": 22200
    },
    {
      "entropy": 5.558699989318848,
      "epoch": 2.2037514886859864,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00045166504701924176,
      "loss": 4.9701,
      "mean_token_accuracy": 0.18535945266485215,
      "num_tokens": 48638186.0,
      "step": 22205
    },
    {
      "entropy": 5.613149642944336,
      "epoch": 2.204247717348154,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00045164310316067727,
      "loss": 5.0943,
      "mean_token_accuracy": 0.1711938574910164,
      "num_tokens": 48649948.0,
      "step": 22210
    },
    {
      "entropy": 5.587204265594482,
      "epoch": 2.2047439460103218,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004516211549217444,
      "loss": 4.9992,
      "mean_token_accuracy": 0.18400440067052842,
      "num_tokens": 48660869.0,
      "step": 22215
    },
    {
      "entropy": 5.526603412628174,
      "epoch": 2.2052401746724892,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004515992023029875,
      "loss": 5.0232,
      "mean_token_accuracy": 0.17694871872663498,
      "num_tokens": 48673383.0,
      "step": 22220
    },
    {
      "entropy": 5.513467025756836,
      "epoch": 2.2057364033346567,
      "grad_norm": 1.046875,
      "learning_rate": 0.000451577245304951,
      "loss": 4.9807,
      "mean_token_accuracy": 0.18683461397886275,
      "num_tokens": 48684071.0,
      "step": 22225
    },
    {
      "entropy": 5.580668544769287,
      "epoch": 2.206232631996824,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004515552839281792,
      "loss": 5.0145,
      "mean_token_accuracy": 0.17973570823669432,
      "num_tokens": 48695432.0,
      "step": 22230
    },
    {
      "entropy": 5.58124189376831,
      "epoch": 2.2067288606589917,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004515333181732169,
      "loss": 5.0237,
      "mean_token_accuracy": 0.18465565145015717,
      "num_tokens": 48706102.0,
      "step": 22235
    },
    {
      "entropy": 5.547732591629028,
      "epoch": 2.207225089321159,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045151134804060863,
      "loss": 5.077,
      "mean_token_accuracy": 0.17627473622560502,
      "num_tokens": 48717748.0,
      "step": 22240
    },
    {
      "entropy": 5.588041067123413,
      "epoch": 2.2077213179833266,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004514893735308993,
      "loss": 4.9801,
      "mean_token_accuracy": 0.1840432822704315,
      "num_tokens": 48728941.0,
      "step": 22245
    },
    {
      "entropy": 5.569143676757813,
      "epoch": 2.208217546645494,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004514673946446338,
      "loss": 5.0443,
      "mean_token_accuracy": 0.1831594243645668,
      "num_tokens": 48738565.0,
      "step": 22250
    },
    {
      "entropy": 5.562809848785401,
      "epoch": 2.208713775307662,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004514454113823572,
      "loss": 5.019,
      "mean_token_accuracy": 0.1765393540263176,
      "num_tokens": 48749537.0,
      "step": 22255
    },
    {
      "entropy": 5.545636415481567,
      "epoch": 2.2092100039698295,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004514234237446146,
      "loss": 5.0205,
      "mean_token_accuracy": 0.17443419694900514,
      "num_tokens": 48760388.0,
      "step": 22260
    },
    {
      "entropy": 5.5714531421661375,
      "epoch": 2.209706232631997,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045140143173195117,
      "loss": 5.0509,
      "mean_token_accuracy": 0.1768399953842163,
      "num_tokens": 48770551.0,
      "step": 22265
    },
    {
      "entropy": 5.548145580291748,
      "epoch": 2.2102024612941644,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004513794353449124,
      "loss": 4.9972,
      "mean_token_accuracy": 0.17944179028272628,
      "num_tokens": 48780739.0,
      "step": 22270
    },
    {
      "entropy": 5.593333292007446,
      "epoch": 2.210698689956332,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004513574345840437,
      "loss": 5.1188,
      "mean_token_accuracy": 0.17226570546627046,
      "num_tokens": 48791478.0,
      "step": 22275
    },
    {
      "entropy": 5.6152276515960695,
      "epoch": 2.2111949186184994,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004513354294498907,
      "loss": 5.0739,
      "mean_token_accuracy": 0.17670032978057862,
      "num_tokens": 48802202.0,
      "step": 22280
    },
    {
      "entropy": 5.589963054656982,
      "epoch": 2.211691147280667,
      "grad_norm": 0.96875,
      "learning_rate": 0.00045131341994299895,
      "loss": 5.0476,
      "mean_token_accuracy": 0.18130356073379517,
      "num_tokens": 48813619.0,
      "step": 22285
    },
    {
      "entropy": 5.607101535797119,
      "epoch": 2.2121873759428343,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004512914060639143,
      "loss": 5.0065,
      "mean_token_accuracy": 0.18936029970645904,
      "num_tokens": 48824951.0,
      "step": 22290
    },
    {
      "entropy": 5.5926981449127195,
      "epoch": 2.212683604605002,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004512693878131827,
      "loss": 5.0194,
      "mean_token_accuracy": 0.18406404107809066,
      "num_tokens": 48834328.0,
      "step": 22295
    },
    {
      "entropy": 5.57903528213501,
      "epoch": 2.2131798332671697,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045124736519135003,
      "loss": 5.0581,
      "mean_token_accuracy": 0.17980943322181703,
      "num_tokens": 48843880.0,
      "step": 22300
    },
    {
      "entropy": 5.5731042385101315,
      "epoch": 2.213676061929337,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045122533819896257,
      "loss": 5.0582,
      "mean_token_accuracy": 0.18477343320846557,
      "num_tokens": 48854893.0,
      "step": 22305
    },
    {
      "entropy": 5.580644655227661,
      "epoch": 2.2141722905915047,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004512033068365664,
      "loss": 5.1188,
      "mean_token_accuracy": 0.17615607529878616,
      "num_tokens": 48867308.0,
      "step": 22310
    },
    {
      "entropy": 5.627501344680786,
      "epoch": 2.214668519253672,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004511812711047079,
      "loss": 5.1042,
      "mean_token_accuracy": 0.17725855112075806,
      "num_tokens": 48880222.0,
      "step": 22315
    },
    {
      "entropy": 5.652177000045777,
      "epoch": 2.2151647479158396,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045115923100393356,
      "loss": 5.0476,
      "mean_token_accuracy": 0.1829438477754593,
      "num_tokens": 48890345.0,
      "step": 22320
    },
    {
      "entropy": 5.583768510818482,
      "epoch": 2.215660976578007,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045113718653478984,
      "loss": 4.9856,
      "mean_token_accuracy": 0.18363598585128785,
      "num_tokens": 48901267.0,
      "step": 22325
    },
    {
      "entropy": 5.604900121688843,
      "epoch": 2.2161572052401746,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004511151376978234,
      "loss": 5.0508,
      "mean_token_accuracy": 0.18134261071681976,
      "num_tokens": 48911739.0,
      "step": 22330
    },
    {
      "entropy": 5.607231092453003,
      "epoch": 2.216653433902342,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045109308449358114,
      "loss": 4.9937,
      "mean_token_accuracy": 0.1829239159822464,
      "num_tokens": 48922376.0,
      "step": 22335
    },
    {
      "entropy": 5.58563380241394,
      "epoch": 2.21714966256451,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004510710269226098,
      "loss": 4.933,
      "mean_token_accuracy": 0.18203675746917725,
      "num_tokens": 48934101.0,
      "step": 22340
    },
    {
      "entropy": 5.537929487228394,
      "epoch": 2.2176458912266774,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004510489649854564,
      "loss": 5.0186,
      "mean_token_accuracy": 0.18465807884931565,
      "num_tokens": 48944684.0,
      "step": 22345
    },
    {
      "entropy": 5.565070056915284,
      "epoch": 2.218142119888845,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000451026898682668,
      "loss": 5.0212,
      "mean_token_accuracy": 0.1741677850484848,
      "num_tokens": 48955255.0,
      "step": 22350
    },
    {
      "entropy": 5.5622382164001465,
      "epoch": 2.2186383485510124,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004510048280147918,
      "loss": 5.0449,
      "mean_token_accuracy": 0.18326972424983978,
      "num_tokens": 48967034.0,
      "step": 22355
    },
    {
      "entropy": 5.620797395706177,
      "epoch": 2.21913457721318,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004509827529823751,
      "loss": 5.1404,
      "mean_token_accuracy": 0.17184707522392273,
      "num_tokens": 48977903.0,
      "step": 22360
    },
    {
      "entropy": 5.613242053985596,
      "epoch": 2.2196308058753473,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00045096067358596536,
      "loss": 5.0165,
      "mean_token_accuracy": 0.18206143230199814,
      "num_tokens": 48988889.0,
      "step": 22365
    },
    {
      "entropy": 5.6383514404296875,
      "epoch": 2.220127034537515,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045093858982611006,
      "loss": 5.12,
      "mean_token_accuracy": 0.16795546412467957,
      "num_tokens": 48998946.0,
      "step": 22370
    },
    {
      "entropy": 5.592536306381225,
      "epoch": 2.2206232631996823,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004509165017033569,
      "loss": 5.0197,
      "mean_token_accuracy": 0.18399660438299179,
      "num_tokens": 49008141.0,
      "step": 22375
    },
    {
      "entropy": 5.547129583358765,
      "epoch": 2.2211194918618498,
      "grad_norm": 0.953125,
      "learning_rate": 0.00045089440921825354,
      "loss": 4.9411,
      "mean_token_accuracy": 0.19075675904750825,
      "num_tokens": 49019387.0,
      "step": 22380
    },
    {
      "entropy": 5.526688098907471,
      "epoch": 2.2216157205240172,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045087231237134787,
      "loss": 4.9956,
      "mean_token_accuracy": 0.18144616335630417,
      "num_tokens": 49030280.0,
      "step": 22385
    },
    {
      "entropy": 5.620566320419312,
      "epoch": 2.222111949186185,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045085021116318783,
      "loss": 4.9849,
      "mean_token_accuracy": 0.18748746067285538,
      "num_tokens": 49041766.0,
      "step": 22390
    },
    {
      "entropy": 5.714295768737793,
      "epoch": 2.2226081778483526,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004508281055943214,
      "loss": 5.221,
      "mean_token_accuracy": 0.1705726847052574,
      "num_tokens": 49052931.0,
      "step": 22395
    },
    {
      "entropy": 5.563606691360474,
      "epoch": 2.22310440651052,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004508059956652969,
      "loss": 5.0187,
      "mean_token_accuracy": 0.1782955199480057,
      "num_tokens": 49064153.0,
      "step": 22400
    },
    {
      "entropy": 5.556827735900879,
      "epoch": 2.2236006351726876,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004507838813766625,
      "loss": 5.0361,
      "mean_token_accuracy": 0.18064209669828415,
      "num_tokens": 49075857.0,
      "step": 22405
    },
    {
      "entropy": 5.678604650497436,
      "epoch": 2.224096863834855,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004507617627289666,
      "loss": 5.0593,
      "mean_token_accuracy": 0.1782527133822441,
      "num_tokens": 49086632.0,
      "step": 22410
    },
    {
      "entropy": 5.607068967819214,
      "epoch": 2.2245930924970225,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004507396397227577,
      "loss": 5.0279,
      "mean_token_accuracy": 0.1786979392170906,
      "num_tokens": 49098086.0,
      "step": 22415
    },
    {
      "entropy": 5.527521991729737,
      "epoch": 2.22508932115919,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00045071751235858454,
      "loss": 4.9404,
      "mean_token_accuracy": 0.18760869055986404,
      "num_tokens": 49108857.0,
      "step": 22420
    },
    {
      "entropy": 5.549077796936035,
      "epoch": 2.2255855498213575,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004506953806369956,
      "loss": 4.9754,
      "mean_token_accuracy": 0.1857747495174408,
      "num_tokens": 49119609.0,
      "step": 22425
    },
    {
      "entropy": 5.608169221878052,
      "epoch": 2.2260817784835254,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004506732445585399,
      "loss": 5.074,
      "mean_token_accuracy": 0.1825532153248787,
      "num_tokens": 49130047.0,
      "step": 22430
    },
    {
      "entropy": 5.572369861602783,
      "epoch": 2.226578007145693,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004506511041237662,
      "loss": 5.0245,
      "mean_token_accuracy": 0.18269540518522262,
      "num_tokens": 49140517.0,
      "step": 22435
    },
    {
      "entropy": 5.601465511322021,
      "epoch": 2.2270742358078603,
      "grad_norm": 1.125,
      "learning_rate": 0.00045062895933322366,
      "loss": 5.0356,
      "mean_token_accuracy": 0.1782595306634903,
      "num_tokens": 49151497.0,
      "step": 22440
    },
    {
      "entropy": 5.482938241958618,
      "epoch": 2.227570464470028,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004506068101874614,
      "loss": 4.9599,
      "mean_token_accuracy": 0.17888348251581193,
      "num_tokens": 49161568.0,
      "step": 22445
    },
    {
      "entropy": 5.6057229995727536,
      "epoch": 2.2280666931321953,
      "grad_norm": 1.0,
      "learning_rate": 0.00045058465668702856,
      "loss": 5.0732,
      "mean_token_accuracy": 0.1794983848929405,
      "num_tokens": 49173554.0,
      "step": 22450
    },
    {
      "entropy": 5.671862602233887,
      "epoch": 2.2285629217943628,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00045056249883247457,
      "loss": 5.0718,
      "mean_token_accuracy": 0.1773956760764122,
      "num_tokens": 49185399.0,
      "step": 22455
    },
    {
      "entropy": 5.608171844482422,
      "epoch": 2.2290591504565302,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045054033662434903,
      "loss": 4.9957,
      "mean_token_accuracy": 0.18290940374135972,
      "num_tokens": 49195489.0,
      "step": 22460
    },
    {
      "entropy": 5.496192836761475,
      "epoch": 2.2295553791186977,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045051817006320127,
      "loss": 4.9601,
      "mean_token_accuracy": 0.1836245521903038,
      "num_tokens": 49204652.0,
      "step": 22465
    },
    {
      "entropy": 5.54185004234314,
      "epoch": 2.230051607780865,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00045049599914958124,
      "loss": 5.0546,
      "mean_token_accuracy": 0.17424410134553908,
      "num_tokens": 49215420.0,
      "step": 22470
    },
    {
      "entropy": 5.594494771957398,
      "epoch": 2.230547836443033,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004504738238840385,
      "loss": 5.0139,
      "mean_token_accuracy": 0.18304983228445054,
      "num_tokens": 49225375.0,
      "step": 22475
    },
    {
      "entropy": 5.633879947662353,
      "epoch": 2.2310440651052006,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004504516442671231,
      "loss": 5.0182,
      "mean_token_accuracy": 0.17765786349773408,
      "num_tokens": 49238399.0,
      "step": 22480
    },
    {
      "entropy": 5.5827446460723875,
      "epoch": 2.231540293767368,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00045042946029938495,
      "loss": 4.9979,
      "mean_token_accuracy": 0.17983777970075607,
      "num_tokens": 49251069.0,
      "step": 22485
    },
    {
      "entropy": 5.5566212177276615,
      "epoch": 2.2320365224295355,
      "grad_norm": 1.0,
      "learning_rate": 0.00045040727198137416,
      "loss": 5.0305,
      "mean_token_accuracy": 0.1777949184179306,
      "num_tokens": 49263859.0,
      "step": 22490
    },
    {
      "entropy": 5.604022073745727,
      "epoch": 2.232532751091703,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00045038507931364103,
      "loss": 5.0295,
      "mean_token_accuracy": 0.1811242073774338,
      "num_tokens": 49274846.0,
      "step": 22495
    },
    {
      "entropy": 5.56999454498291,
      "epoch": 2.2330289797538705,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045036288229673583,
      "loss": 4.9791,
      "mean_token_accuracy": 0.18691035509109497,
      "num_tokens": 49285197.0,
      "step": 22500
    },
    {
      "entropy": 5.534236335754395,
      "epoch": 2.233525208416038,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045034068093120907,
      "loss": 4.9766,
      "mean_token_accuracy": 0.17641551047563553,
      "num_tokens": 49296238.0,
      "step": 22505
    },
    {
      "entropy": 5.677579641342163,
      "epoch": 2.2340214370782054,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004503184752176112,
      "loss": 5.2035,
      "mean_token_accuracy": 0.1695760115981102,
      "num_tokens": 49308262.0,
      "step": 22510
    },
    {
      "entropy": 5.586322879791259,
      "epoch": 2.2345176657403734,
      "grad_norm": 1.0,
      "learning_rate": 0.000450296265156493,
      "loss": 4.9835,
      "mean_token_accuracy": 0.18736137598752975,
      "num_tokens": 49318881.0,
      "step": 22515
    },
    {
      "entropy": 5.651030826568603,
      "epoch": 2.235013894402541,
      "grad_norm": 1.0,
      "learning_rate": 0.00045027405074840506,
      "loss": 5.1322,
      "mean_token_accuracy": 0.17456569522619247,
      "num_tokens": 49329470.0,
      "step": 22520
    },
    {
      "entropy": 5.595328664779663,
      "epoch": 2.2355101230647083,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004502518319938984,
      "loss": 5.0198,
      "mean_token_accuracy": 0.18555116206407546,
      "num_tokens": 49339806.0,
      "step": 22525
    },
    {
      "entropy": 5.701306343078613,
      "epoch": 2.236006351726876,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045022960889352394,
      "loss": 5.1294,
      "mean_token_accuracy": 0.17799793034791947,
      "num_tokens": 49350295.0,
      "step": 22530
    },
    {
      "entropy": 5.607735109329224,
      "epoch": 2.2365025803890433,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045020738144783273,
      "loss": 5.0372,
      "mean_token_accuracy": 0.17820966094732285,
      "num_tokens": 49360735.0,
      "step": 22535
    },
    {
      "entropy": 5.548648738861084,
      "epoch": 2.2369988090512107,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004501851496573761,
      "loss": 4.9877,
      "mean_token_accuracy": 0.18256026655435562,
      "num_tokens": 49372278.0,
      "step": 22540
    },
    {
      "entropy": 5.572845602035523,
      "epoch": 2.237495037713378,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004501629135227051,
      "loss": 5.0928,
      "mean_token_accuracy": 0.17433278560638427,
      "num_tokens": 49382842.0,
      "step": 22545
    },
    {
      "entropy": 5.617898464202881,
      "epoch": 2.2379912663755457,
      "grad_norm": 1.0,
      "learning_rate": 0.0004501406730443714,
      "loss": 5.0912,
      "mean_token_accuracy": 0.17576729357242585,
      "num_tokens": 49394633.0,
      "step": 22550
    },
    {
      "entropy": 5.668472146987915,
      "epoch": 2.238487495037713,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045011842822292636,
      "loss": 5.0675,
      "mean_token_accuracy": 0.17675844132900237,
      "num_tokens": 49404731.0,
      "step": 22555
    },
    {
      "entropy": 5.641077184677124,
      "epoch": 2.238983723699881,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004500961790589216,
      "loss": 5.0544,
      "mean_token_accuracy": 0.17910027354955674,
      "num_tokens": 49415918.0,
      "step": 22560
    },
    {
      "entropy": 5.553595685958863,
      "epoch": 2.2394799523620486,
      "grad_norm": 1.046875,
      "learning_rate": 0.000450073925552909,
      "loss": 4.9969,
      "mean_token_accuracy": 0.17839334011077881,
      "num_tokens": 49427458.0,
      "step": 22565
    },
    {
      "entropy": 5.584511375427246,
      "epoch": 2.239976181024216,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004500516677054403,
      "loss": 5.0178,
      "mean_token_accuracy": 0.18439576029777527,
      "num_tokens": 49437696.0,
      "step": 22570
    },
    {
      "entropy": 5.65137825012207,
      "epoch": 2.2404724096863835,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045002940551706726,
      "loss": 5.2217,
      "mean_token_accuracy": 0.1747758075594902,
      "num_tokens": 49449220.0,
      "step": 22575
    },
    {
      "entropy": 5.620813655853271,
      "epoch": 2.240968638348551,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045000713898834225,
      "loss": 5.0493,
      "mean_token_accuracy": 0.17836696058511733,
      "num_tokens": 49459624.0,
      "step": 22580
    },
    {
      "entropy": 5.617305517196655,
      "epoch": 2.2414648670107185,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004499848681198173,
      "loss": 4.9687,
      "mean_token_accuracy": 0.18799397051334382,
      "num_tokens": 49471813.0,
      "step": 22585
    },
    {
      "entropy": 5.5926868438720705,
      "epoch": 2.241961095672886,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004499625929120447,
      "loss": 5.0383,
      "mean_token_accuracy": 0.18021085262298583,
      "num_tokens": 49482624.0,
      "step": 22590
    },
    {
      "entropy": 5.596084833145142,
      "epoch": 2.2424573243350534,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004499403133655768,
      "loss": 5.0643,
      "mean_token_accuracy": 0.1739892914891243,
      "num_tokens": 49493791.0,
      "step": 22595
    },
    {
      "entropy": 5.6493675231933596,
      "epoch": 2.2429535529972213,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004499180294809661,
      "loss": 5.096,
      "mean_token_accuracy": 0.17527344971895217,
      "num_tokens": 49504060.0,
      "step": 22600
    },
    {
      "entropy": 5.548713541030883,
      "epoch": 2.243449781659389,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044989574125876516,
      "loss": 4.9347,
      "mean_token_accuracy": 0.18660158216953276,
      "num_tokens": 49515962.0,
      "step": 22605
    },
    {
      "entropy": 5.6037561893463135,
      "epoch": 2.2439460103215563,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004498734486995267,
      "loss": 5.0652,
      "mean_token_accuracy": 0.17424511015415192,
      "num_tokens": 49526676.0,
      "step": 22610
    },
    {
      "entropy": 5.607731294631958,
      "epoch": 2.2444422389837237,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044985115180380355,
      "loss": 5.0577,
      "mean_token_accuracy": 0.17544593513011933,
      "num_tokens": 49537347.0,
      "step": 22615
    },
    {
      "entropy": 5.624332332611084,
      "epoch": 2.2449384676458912,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00044982885057214857,
      "loss": 5.0639,
      "mean_token_accuracy": 0.17598291337490082,
      "num_tokens": 49547422.0,
      "step": 22620
    },
    {
      "entropy": 5.583463382720947,
      "epoch": 2.2454346963080587,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00044980654500511485,
      "loss": 4.9204,
      "mean_token_accuracy": 0.18898916244506836,
      "num_tokens": 49557250.0,
      "step": 22625
    },
    {
      "entropy": 5.654331064224243,
      "epoch": 2.245930924970226,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044978423510325544,
      "loss": 5.117,
      "mean_token_accuracy": 0.17775555253028869,
      "num_tokens": 49567106.0,
      "step": 22630
    },
    {
      "entropy": 5.620014476776123,
      "epoch": 2.2464271536323936,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004497619208671237,
      "loss": 5.0672,
      "mean_token_accuracy": 0.17509078681468965,
      "num_tokens": 49578978.0,
      "step": 22635
    },
    {
      "entropy": 5.686333179473877,
      "epoch": 2.246923382294561,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004497396022972728,
      "loss": 5.1064,
      "mean_token_accuracy": 0.1785639524459839,
      "num_tokens": 49589730.0,
      "step": 22640
    },
    {
      "entropy": 5.603769016265869,
      "epoch": 2.247419610956729,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004497172793942564,
      "loss": 5.0191,
      "mean_token_accuracy": 0.18040618747472764,
      "num_tokens": 49600881.0,
      "step": 22645
    },
    {
      "entropy": 5.599695825576783,
      "epoch": 2.2479158396188965,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004496949521586279,
      "loss": 5.0659,
      "mean_token_accuracy": 0.1783287286758423,
      "num_tokens": 49611120.0,
      "step": 22650
    },
    {
      "entropy": 5.618778085708618,
      "epoch": 2.248412068281064,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044967262059094097,
      "loss": 5.0855,
      "mean_token_accuracy": 0.17207848131656647,
      "num_tokens": 49622978.0,
      "step": 22655
    },
    {
      "entropy": 5.601924800872803,
      "epoch": 2.2489082969432315,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044965028469174956,
      "loss": 5.0726,
      "mean_token_accuracy": 0.17792664468288422,
      "num_tokens": 49633678.0,
      "step": 22660
    },
    {
      "entropy": 5.610531854629516,
      "epoch": 2.249404525605399,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00044962794446160736,
      "loss": 5.0619,
      "mean_token_accuracy": 0.17468793988227843,
      "num_tokens": 49644886.0,
      "step": 22665
    },
    {
      "entropy": 5.597045993804931,
      "epoch": 2.2499007542675664,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044960559990106844,
      "loss": 4.9987,
      "mean_token_accuracy": 0.1867280676960945,
      "num_tokens": 49656790.0,
      "step": 22670
    },
    {
      "entropy": 5.632965612411499,
      "epoch": 2.250396982929734,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004495832510106869,
      "loss": 5.0479,
      "mean_token_accuracy": 0.17537580132484437,
      "num_tokens": 49667943.0,
      "step": 22675
    },
    {
      "entropy": 5.569033288955689,
      "epoch": 2.2508932115919014,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004495608977910169,
      "loss": 5.0299,
      "mean_token_accuracy": 0.17802347838878632,
      "num_tokens": 49678652.0,
      "step": 22680
    },
    {
      "entropy": 5.57305941581726,
      "epoch": 2.2513894402540693,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004495385402426128,
      "loss": 4.9958,
      "mean_token_accuracy": 0.18683501034975053,
      "num_tokens": 49688951.0,
      "step": 22685
    },
    {
      "entropy": 5.588244962692261,
      "epoch": 2.2518856689162368,
      "grad_norm": 1.03125,
      "learning_rate": 0.000449516178366029,
      "loss": 5.0565,
      "mean_token_accuracy": 0.1780592381954193,
      "num_tokens": 49698750.0,
      "step": 22690
    },
    {
      "entropy": 5.625252962112427,
      "epoch": 2.2523818975784042,
      "grad_norm": 1.0,
      "learning_rate": 0.00044949381216182,
      "loss": 5.1349,
      "mean_token_accuracy": 0.171563021838665,
      "num_tokens": 49709442.0,
      "step": 22695
    },
    {
      "entropy": 5.6081953048706055,
      "epoch": 2.2528781262405717,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044947144163054036,
      "loss": 5.029,
      "mean_token_accuracy": 0.18069500029087066,
      "num_tokens": 49720517.0,
      "step": 22700
    },
    {
      "entropy": 5.608110952377319,
      "epoch": 2.253374354902739,
      "grad_norm": 1.125,
      "learning_rate": 0.0004494490667727451,
      "loss": 5.0241,
      "mean_token_accuracy": 0.17928366959095002,
      "num_tokens": 49731413.0,
      "step": 22705
    },
    {
      "entropy": 5.653209590911866,
      "epoch": 2.2538705835649067,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00044942668758898873,
      "loss": 5.1929,
      "mean_token_accuracy": 0.16980450451374055,
      "num_tokens": 49742492.0,
      "step": 22710
    },
    {
      "entropy": 5.644893836975098,
      "epoch": 2.254366812227074,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004494043040798265,
      "loss": 5.0306,
      "mean_token_accuracy": 0.1753566488623619,
      "num_tokens": 49753981.0,
      "step": 22715
    },
    {
      "entropy": 5.615680742263794,
      "epoch": 2.2548630408892416,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004493819162458133,
      "loss": 5.0769,
      "mean_token_accuracy": 0.17856322079896927,
      "num_tokens": 49764718.0,
      "step": 22720
    },
    {
      "entropy": 5.599643993377685,
      "epoch": 2.255359269551409,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004493595240875043,
      "loss": 5.0248,
      "mean_token_accuracy": 0.18065377324819565,
      "num_tokens": 49775361.0,
      "step": 22725
    },
    {
      "entropy": 5.544538784027099,
      "epoch": 2.255855498213577,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044933712760545486,
      "loss": 5.0098,
      "mean_token_accuracy": 0.18104665130376815,
      "num_tokens": 49786670.0,
      "step": 22730
    },
    {
      "entropy": 5.645787715911865,
      "epoch": 2.2563517268757445,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004493147268002203,
      "loss": 5.1408,
      "mean_token_accuracy": 0.17564889937639236,
      "num_tokens": 49796514.0,
      "step": 22735
    },
    {
      "entropy": 5.540385770797729,
      "epoch": 2.256847955537912,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044929232167235613,
      "loss": 4.9953,
      "mean_token_accuracy": 0.1855500653386116,
      "num_tokens": 49807042.0,
      "step": 22740
    },
    {
      "entropy": 5.565149736404419,
      "epoch": 2.2573441842000794,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004492699122224179,
      "loss": 5.0739,
      "mean_token_accuracy": 0.17502318024635316,
      "num_tokens": 49818779.0,
      "step": 22745
    },
    {
      "entropy": 5.616581821441651,
      "epoch": 2.257840412862247,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044924749845096144,
      "loss": 5.0671,
      "mean_token_accuracy": 0.17724036574363708,
      "num_tokens": 49829901.0,
      "step": 22750
    },
    {
      "entropy": 5.604426765441895,
      "epoch": 2.2583366415244144,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004492250803585425,
      "loss": 5.0507,
      "mean_token_accuracy": 0.18493872433900832,
      "num_tokens": 49840837.0,
      "step": 22755
    },
    {
      "entropy": 5.614662170410156,
      "epoch": 2.258832870186582,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004492026579457169,
      "loss": 5.0105,
      "mean_token_accuracy": 0.1774394094944,
      "num_tokens": 49853052.0,
      "step": 22760
    },
    {
      "entropy": 5.581817293167115,
      "epoch": 2.2593290988487493,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004491802312130408,
      "loss": 5.0243,
      "mean_token_accuracy": 0.17834030240774154,
      "num_tokens": 49864210.0,
      "step": 22765
    },
    {
      "entropy": 5.563812208175659,
      "epoch": 2.2598253275109172,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004491578001610703,
      "loss": 4.9886,
      "mean_token_accuracy": 0.186432908475399,
      "num_tokens": 49874995.0,
      "step": 22770
    },
    {
      "entropy": 5.50725531578064,
      "epoch": 2.2603215561730847,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044913536479036163,
      "loss": 5.0049,
      "mean_token_accuracy": 0.1801043227314949,
      "num_tokens": 49886240.0,
      "step": 22775
    },
    {
      "entropy": 5.56751651763916,
      "epoch": 2.260817784835252,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00044911292510147117,
      "loss": 5.0153,
      "mean_token_accuracy": 0.18055537790060044,
      "num_tokens": 49897580.0,
      "step": 22780
    },
    {
      "entropy": 5.583517980575562,
      "epoch": 2.2613140134974197,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004490904810949553,
      "loss": 5.0583,
      "mean_token_accuracy": 0.1789379447698593,
      "num_tokens": 49907280.0,
      "step": 22785
    },
    {
      "entropy": 5.611277866363525,
      "epoch": 2.261810242159587,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044906803277137064,
      "loss": 5.0761,
      "mean_token_accuracy": 0.1817319795489311,
      "num_tokens": 49917552.0,
      "step": 22790
    },
    {
      "entropy": 5.586445045471192,
      "epoch": 2.2623064708217546,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004490455801312738,
      "loss": 5.0481,
      "mean_token_accuracy": 0.1772630751132965,
      "num_tokens": 49930610.0,
      "step": 22795
    },
    {
      "entropy": 5.599416589736938,
      "epoch": 2.262802699483922,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004490231231752217,
      "loss": 5.0794,
      "mean_token_accuracy": 0.17749821841716767,
      "num_tokens": 49941054.0,
      "step": 22800
    },
    {
      "entropy": 5.500024700164795,
      "epoch": 2.2632989281460896,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004490006619037711,
      "loss": 4.9263,
      "mean_token_accuracy": 0.19124432057142257,
      "num_tokens": 49950885.0,
      "step": 22805
    },
    {
      "entropy": 5.5474159717559814,
      "epoch": 2.263795156808257,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00044897819631747903,
      "loss": 5.0357,
      "mean_token_accuracy": 0.17892571538686752,
      "num_tokens": 49961723.0,
      "step": 22810
    },
    {
      "entropy": 5.54910945892334,
      "epoch": 2.2642913854704245,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00044895572641690256,
      "loss": 5.0085,
      "mean_token_accuracy": 0.18188804090023042,
      "num_tokens": 49974124.0,
      "step": 22815
    },
    {
      "entropy": 5.572559213638305,
      "epoch": 2.2647876141325924,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000448933252202599,
      "loss": 5.0346,
      "mean_token_accuracy": 0.18043317794799804,
      "num_tokens": 49985792.0,
      "step": 22820
    },
    {
      "entropy": 5.541806745529175,
      "epoch": 2.26528384279476,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044891077367512544,
      "loss": 4.9758,
      "mean_token_accuracy": 0.1844248056411743,
      "num_tokens": 49996935.0,
      "step": 22825
    },
    {
      "entropy": 5.653744173049927,
      "epoch": 2.2657800714569274,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004488882908350396,
      "loss": 5.1389,
      "mean_token_accuracy": 0.17303285300731658,
      "num_tokens": 50007545.0,
      "step": 22830
    },
    {
      "entropy": 5.632570362091064,
      "epoch": 2.266276300119095,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004488658036828987,
      "loss": 5.0448,
      "mean_token_accuracy": 0.18197034299373627,
      "num_tokens": 50018797.0,
      "step": 22835
    },
    {
      "entropy": 5.581188726425171,
      "epoch": 2.2667725287812623,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004488433122192606,
      "loss": 5.0106,
      "mean_token_accuracy": 0.18376223891973495,
      "num_tokens": 50030884.0,
      "step": 22840
    },
    {
      "entropy": 5.642255640029907,
      "epoch": 2.26726875744343,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044882081644468286,
      "loss": 5.1477,
      "mean_token_accuracy": 0.17101453244686127,
      "num_tokens": 50042327.0,
      "step": 22845
    },
    {
      "entropy": 5.608316659927368,
      "epoch": 2.2677649861055973,
      "grad_norm": 1.0,
      "learning_rate": 0.00044879831635972345,
      "loss": 5.0854,
      "mean_token_accuracy": 0.18410026282072067,
      "num_tokens": 50053733.0,
      "step": 22850
    },
    {
      "entropy": 5.595193719863891,
      "epoch": 2.268261214767765,
      "grad_norm": 1.125,
      "learning_rate": 0.0004487758119649403,
      "loss": 5.0191,
      "mean_token_accuracy": 0.18390018343925477,
      "num_tokens": 50064153.0,
      "step": 22855
    },
    {
      "entropy": 5.519360589981079,
      "epoch": 2.2687574434299327,
      "grad_norm": 1.0,
      "learning_rate": 0.00044875330326089153,
      "loss": 4.9702,
      "mean_token_accuracy": 0.19200130850076674,
      "num_tokens": 50074397.0,
      "step": 22860
    },
    {
      "entropy": 5.520435428619384,
      "epoch": 2.2692536720921,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004487307902481352,
      "loss": 4.9746,
      "mean_token_accuracy": 0.18198251873254775,
      "num_tokens": 50084074.0,
      "step": 22865
    },
    {
      "entropy": 5.581186103820801,
      "epoch": 2.2697499007542676,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044870827292722967,
      "loss": 5.0013,
      "mean_token_accuracy": 0.17841627299785615,
      "num_tokens": 50094535.0,
      "step": 22870
    },
    {
      "entropy": 5.589366722106933,
      "epoch": 2.270246129416435,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004486857512987333,
      "loss": 5.0679,
      "mean_token_accuracy": 0.17727411836385726,
      "num_tokens": 50105907.0,
      "step": 22875
    },
    {
      "entropy": 5.544551610946655,
      "epoch": 2.2707423580786026,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044866322536320447,
      "loss": 5.0399,
      "mean_token_accuracy": 0.18140538930892944,
      "num_tokens": 50117361.0,
      "step": 22880
    },
    {
      "entropy": 5.611779260635376,
      "epoch": 2.27123858674077,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00044864069512120185,
      "loss": 5.0599,
      "mean_token_accuracy": 0.18520863503217697,
      "num_tokens": 50128598.0,
      "step": 22885
    },
    {
      "entropy": 5.591342782974243,
      "epoch": 2.2717348154029375,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044861816057328424,
      "loss": 5.0369,
      "mean_token_accuracy": 0.17884714901447296,
      "num_tokens": 50140470.0,
      "step": 22890
    },
    {
      "entropy": 5.624648380279541,
      "epoch": 2.272231044065105,
      "grad_norm": 0.984375,
      "learning_rate": 0.00044859562172001036,
      "loss": 5.0997,
      "mean_token_accuracy": 0.17759020030498504,
      "num_tokens": 50150950.0,
      "step": 22895
    },
    {
      "entropy": 5.535177564620971,
      "epoch": 2.2727272727272725,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004485730785619391,
      "loss": 4.9409,
      "mean_token_accuracy": 0.18485622704029084,
      "num_tokens": 50161173.0,
      "step": 22900
    },
    {
      "entropy": 5.609437751770019,
      "epoch": 2.2732235013894404,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004485505310996295,
      "loss": 5.1723,
      "mean_token_accuracy": 0.17111800909042357,
      "num_tokens": 50173874.0,
      "step": 22905
    },
    {
      "entropy": 5.715093755722046,
      "epoch": 2.273719730051608,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004485279793336406,
      "loss": 5.1213,
      "mean_token_accuracy": 0.17077192962169646,
      "num_tokens": 50185123.0,
      "step": 22910
    },
    {
      "entropy": 5.612595701217652,
      "epoch": 2.2742159587137754,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004485054232645319,
      "loss": 5.0728,
      "mean_token_accuracy": 0.18398701399564743,
      "num_tokens": 50195699.0,
      "step": 22915
    },
    {
      "entropy": 5.651986026763916,
      "epoch": 2.274712187375943,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004484828628928625,
      "loss": 5.0851,
      "mean_token_accuracy": 0.18250927031040193,
      "num_tokens": 50206623.0,
      "step": 22920
    },
    {
      "entropy": 5.673154497146607,
      "epoch": 2.2752084160381103,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00044846029821919193,
      "loss": 5.149,
      "mean_token_accuracy": 0.17489539235830306,
      "num_tokens": 50216086.0,
      "step": 22925
    },
    {
      "entropy": 5.536364316940308,
      "epoch": 2.275704644700278,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004484377292440797,
      "loss": 5.0339,
      "mean_token_accuracy": 0.1776548519730568,
      "num_tokens": 50228472.0,
      "step": 22930
    },
    {
      "entropy": 5.597187137603759,
      "epoch": 2.2762008733624453,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044841515596808553,
      "loss": 5.0654,
      "mean_token_accuracy": 0.1734164297580719,
      "num_tokens": 50238607.0,
      "step": 22935
    },
    {
      "entropy": 5.606688213348389,
      "epoch": 2.276697102024613,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00044839257839176914,
      "loss": 5.0168,
      "mean_token_accuracy": 0.18222397267818452,
      "num_tokens": 50249890.0,
      "step": 22940
    },
    {
      "entropy": 5.586983013153076,
      "epoch": 2.2771933306867806,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044836999651569055,
      "loss": 5.013,
      "mean_token_accuracy": 0.18418598771095276,
      "num_tokens": 50260234.0,
      "step": 22945
    },
    {
      "entropy": 5.57249608039856,
      "epoch": 2.277689559348948,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004483474103404095,
      "loss": 5.0647,
      "mean_token_accuracy": 0.18492246121168138,
      "num_tokens": 50271605.0,
      "step": 22950
    },
    {
      "entropy": 5.596893405914306,
      "epoch": 2.2781857880111156,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044832481986648633,
      "loss": 4.9435,
      "mean_token_accuracy": 0.18492139875888824,
      "num_tokens": 50281837.0,
      "step": 22955
    },
    {
      "entropy": 5.549652242660523,
      "epoch": 2.278682016673283,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000448302225094481,
      "loss": 4.9834,
      "mean_token_accuracy": 0.18546690791845322,
      "num_tokens": 50292524.0,
      "step": 22960
    },
    {
      "entropy": 5.5931768894195555,
      "epoch": 2.2791782453354505,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044827962602495393,
      "loss": 5.0043,
      "mean_token_accuracy": 0.17987931817770003,
      "num_tokens": 50304009.0,
      "step": 22965
    },
    {
      "entropy": 5.624735116958618,
      "epoch": 2.279674473997618,
      "grad_norm": 1.0,
      "learning_rate": 0.00044825702265846557,
      "loss": 5.0764,
      "mean_token_accuracy": 0.17724401950836183,
      "num_tokens": 50316103.0,
      "step": 22970
    },
    {
      "entropy": 5.602565097808838,
      "epoch": 2.2801707026597855,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004482344149955763,
      "loss": 5.0832,
      "mean_token_accuracy": 0.18215854465961456,
      "num_tokens": 50326127.0,
      "step": 22975
    },
    {
      "entropy": 5.533067321777343,
      "epoch": 2.280666931321953,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004482118030368469,
      "loss": 4.9872,
      "mean_token_accuracy": 0.18493663221597673,
      "num_tokens": 50337044.0,
      "step": 22980
    },
    {
      "entropy": 5.619342613220215,
      "epoch": 2.2811631599841204,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000448189186782838,
      "loss": 5.1865,
      "mean_token_accuracy": 0.17090394496917724,
      "num_tokens": 50347796.0,
      "step": 22985
    },
    {
      "entropy": 5.5643086433410645,
      "epoch": 2.2816593886462884,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00044816656623411044,
      "loss": 5.0543,
      "mean_token_accuracy": 0.17792009562253952,
      "num_tokens": 50358454.0,
      "step": 22990
    },
    {
      "entropy": 5.565145683288574,
      "epoch": 2.282155617308456,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004481439413912252,
      "loss": 4.9573,
      "mean_token_accuracy": 0.1811808779835701,
      "num_tokens": 50369415.0,
      "step": 22995
    },
    {
      "entropy": 5.566468095779419,
      "epoch": 2.2826518459706233,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00044812131225474333,
      "loss": 5.0594,
      "mean_token_accuracy": 0.17492768466472625,
      "num_tokens": 50379977.0,
      "step": 23000
    },
    {
      "entropy": 5.638420152664184,
      "epoch": 2.283148074632791,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004480986788252259,
      "loss": 5.0548,
      "mean_token_accuracy": 0.18053642213344573,
      "num_tokens": 50390465.0,
      "step": 23005
    },
    {
      "entropy": 5.644487953186035,
      "epoch": 2.2836443032949583,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004480760411032343,
      "loss": 5.0818,
      "mean_token_accuracy": 0.1787111684679985,
      "num_tokens": 50402230.0,
      "step": 23010
    },
    {
      "entropy": 5.537339544296264,
      "epoch": 2.2841405319571257,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004480533990893298,
      "loss": 4.9694,
      "mean_token_accuracy": 0.18669238835573196,
      "num_tokens": 50413819.0,
      "step": 23015
    },
    {
      "entropy": 5.743681001663208,
      "epoch": 2.284636760619293,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004480307527840739,
      "loss": 5.2497,
      "mean_token_accuracy": 0.1648907855153084,
      "num_tokens": 50424977.0,
      "step": 23020
    },
    {
      "entropy": 5.656344699859619,
      "epoch": 2.285132989281461,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004480081021880282,
      "loss": 5.0544,
      "mean_token_accuracy": 0.17721066027879714,
      "num_tokens": 50438357.0,
      "step": 23025
    },
    {
      "entropy": 5.6191794872283936,
      "epoch": 2.2856292179436286,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004479854473017543,
      "loss": 5.0492,
      "mean_token_accuracy": 0.1796061411499977,
      "num_tokens": 50449081.0,
      "step": 23030
    },
    {
      "entropy": 5.515747547149658,
      "epoch": 2.286125446605796,
      "grad_norm": 0.96875,
      "learning_rate": 0.00044796278812581417,
      "loss": 4.9475,
      "mean_token_accuracy": 0.18713658303022385,
      "num_tokens": 50460358.0,
      "step": 23035
    },
    {
      "entropy": 5.592691373825073,
      "epoch": 2.2866216752679636,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004479401246607695,
      "loss": 5.0327,
      "mean_token_accuracy": 0.1791094422340393,
      "num_tokens": 50470955.0,
      "step": 23040
    },
    {
      "entropy": 5.631878757476807,
      "epoch": 2.287117903930131,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004479174569071823,
      "loss": 5.0057,
      "mean_token_accuracy": 0.1853179767727852,
      "num_tokens": 50482250.0,
      "step": 23045
    },
    {
      "entropy": 5.584647178649902,
      "epoch": 2.2876141325922985,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004478947848656149,
      "loss": 5.0378,
      "mean_token_accuracy": 0.18050197064876555,
      "num_tokens": 50493868.0,
      "step": 23050
    },
    {
      "entropy": 5.5868284702301025,
      "epoch": 2.288110361254466,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044787210853662943,
      "loss": 5.0855,
      "mean_token_accuracy": 0.18069671839475632,
      "num_tokens": 50504012.0,
      "step": 23055
    },
    {
      "entropy": 5.620962047576905,
      "epoch": 2.2886065899166335,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004478494279207881,
      "loss": 5.0049,
      "mean_token_accuracy": 0.18260658979415895,
      "num_tokens": 50515503.0,
      "step": 23060
    },
    {
      "entropy": 5.589963817596436,
      "epoch": 2.289102818578801,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044782674301865346,
      "loss": 5.0503,
      "mean_token_accuracy": 0.1753444865345955,
      "num_tokens": 50526180.0,
      "step": 23065
    },
    {
      "entropy": 5.607716512680054,
      "epoch": 2.2895990472409684,
      "grad_norm": 1.0,
      "learning_rate": 0.0004478040538307879,
      "loss": 5.0687,
      "mean_token_accuracy": 0.17163751274347305,
      "num_tokens": 50536711.0,
      "step": 23070
    },
    {
      "entropy": 5.6650806903839115,
      "epoch": 2.2900952759031363,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004477813603577543,
      "loss": 5.1253,
      "mean_token_accuracy": 0.1744353875517845,
      "num_tokens": 50547736.0,
      "step": 23075
    },
    {
      "entropy": 5.620262527465821,
      "epoch": 2.290591504565304,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00044775866260011516,
      "loss": 5.0212,
      "mean_token_accuracy": 0.18726903200149536,
      "num_tokens": 50558173.0,
      "step": 23080
    },
    {
      "entropy": 5.599652814865112,
      "epoch": 2.2910877332274713,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004477359605584335,
      "loss": 5.0489,
      "mean_token_accuracy": 0.18064558058977126,
      "num_tokens": 50568421.0,
      "step": 23085
    },
    {
      "entropy": 5.595942878723145,
      "epoch": 2.2915839618896388,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044771325423327227,
      "loss": 5.0119,
      "mean_token_accuracy": 0.18216795921325685,
      "num_tokens": 50579182.0,
      "step": 23090
    },
    {
      "entropy": 5.6086602210998535,
      "epoch": 2.2920801905518062,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004476905436251945,
      "loss": 5.0363,
      "mean_token_accuracy": 0.1848099038004875,
      "num_tokens": 50590916.0,
      "step": 23095
    },
    {
      "entropy": 5.505478191375732,
      "epoch": 2.2925764192139737,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044766782873476333,
      "loss": 4.92,
      "mean_token_accuracy": 0.19027017056941986,
      "num_tokens": 50600682.0,
      "step": 23100
    },
    {
      "entropy": 5.664739608764648,
      "epoch": 2.293072647876141,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000447645109562542,
      "loss": 5.1874,
      "mean_token_accuracy": 0.16733266711235045,
      "num_tokens": 50612103.0,
      "step": 23105
    },
    {
      "entropy": 5.674100589752197,
      "epoch": 2.293568876538309,
      "grad_norm": 1.0,
      "learning_rate": 0.00044762238610909414,
      "loss": 5.1153,
      "mean_token_accuracy": 0.17316448390483857,
      "num_tokens": 50623894.0,
      "step": 23110
    },
    {
      "entropy": 5.6742846965789795,
      "epoch": 2.2940651052004766,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004475996583749829,
      "loss": 5.0839,
      "mean_token_accuracy": 0.17228189408779143,
      "num_tokens": 50634495.0,
      "step": 23115
    },
    {
      "entropy": 5.66628680229187,
      "epoch": 2.294561333862644,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044757692636077215,
      "loss": 5.215,
      "mean_token_accuracy": 0.17369554191827774,
      "num_tokens": 50645921.0,
      "step": 23120
    },
    {
      "entropy": 5.66722297668457,
      "epoch": 2.2950575625248115,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044755419006702545,
      "loss": 5.0582,
      "mean_token_accuracy": 0.17647605389356613,
      "num_tokens": 50657047.0,
      "step": 23125
    },
    {
      "entropy": 5.639464855194092,
      "epoch": 2.295553791186979,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044753144949430667,
      "loss": 5.0339,
      "mean_token_accuracy": 0.17725845724344252,
      "num_tokens": 50668271.0,
      "step": 23130
    },
    {
      "entropy": 5.5653282642364506,
      "epoch": 2.2960500198491465,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004475087046431798,
      "loss": 4.9852,
      "mean_token_accuracy": 0.1874163419008255,
      "num_tokens": 50678870.0,
      "step": 23135
    },
    {
      "entropy": 5.573322582244873,
      "epoch": 2.296546248511314,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004474859555142087,
      "loss": 5.0394,
      "mean_token_accuracy": 0.18124370872974396,
      "num_tokens": 50690322.0,
      "step": 23140
    },
    {
      "entropy": 5.5733558177948,
      "epoch": 2.2970424771734814,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044746320210795753,
      "loss": 4.9793,
      "mean_token_accuracy": 0.1834281638264656,
      "num_tokens": 50701958.0,
      "step": 23145
    },
    {
      "entropy": 5.583808469772339,
      "epoch": 2.297538705835649,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004474404444249906,
      "loss": 5.0474,
      "mean_token_accuracy": 0.18413593173027037,
      "num_tokens": 50712962.0,
      "step": 23150
    },
    {
      "entropy": 5.574769163131714,
      "epoch": 2.2980349344978164,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044741768246587226,
      "loss": 5.079,
      "mean_token_accuracy": 0.18358581215143205,
      "num_tokens": 50723442.0,
      "step": 23155
    },
    {
      "entropy": 5.599833822250366,
      "epoch": 2.2985311631599843,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004473949162311669,
      "loss": 4.957,
      "mean_token_accuracy": 0.181428362429142,
      "num_tokens": 50733188.0,
      "step": 23160
    },
    {
      "entropy": 5.619607734680176,
      "epoch": 2.2990273918221518,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004473721457214391,
      "loss": 5.0967,
      "mean_token_accuracy": 0.1729112982749939,
      "num_tokens": 50744023.0,
      "step": 23165
    },
    {
      "entropy": 5.695945358276367,
      "epoch": 2.2995236204843192,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004473493709372535,
      "loss": 5.2631,
      "mean_token_accuracy": 0.17567385882139205,
      "num_tokens": 50756807.0,
      "step": 23170
    },
    {
      "entropy": 5.663753366470337,
      "epoch": 2.3000198491464867,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004473265918791749,
      "loss": 5.1487,
      "mean_token_accuracy": 0.17493315190076827,
      "num_tokens": 50768981.0,
      "step": 23175
    },
    {
      "entropy": 5.654820537567138,
      "epoch": 2.300516077808654,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004473038085477682,
      "loss": 5.0518,
      "mean_token_accuracy": 0.1841997131705284,
      "num_tokens": 50780111.0,
      "step": 23180
    },
    {
      "entropy": 5.6414525508880615,
      "epoch": 2.3010123064708217,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044728102094359826,
      "loss": 5.0876,
      "mean_token_accuracy": 0.1773157000541687,
      "num_tokens": 50790318.0,
      "step": 23185
    },
    {
      "entropy": 5.627245235443115,
      "epoch": 2.301508535132989,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004472582290672303,
      "loss": 5.1141,
      "mean_token_accuracy": 0.1752907156944275,
      "num_tokens": 50800734.0,
      "step": 23190
    },
    {
      "entropy": 5.528088760375977,
      "epoch": 2.3020047637951566,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044723543291922936,
      "loss": 5.0149,
      "mean_token_accuracy": 0.18594680428504945,
      "num_tokens": 50811236.0,
      "step": 23195
    },
    {
      "entropy": 5.569911146163941,
      "epoch": 2.3025009924573245,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044721263250016086,
      "loss": 5.0118,
      "mean_token_accuracy": 0.18104765713214874,
      "num_tokens": 50821776.0,
      "step": 23200
    },
    {
      "entropy": 5.65333776473999,
      "epoch": 2.302997221119492,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044718982781059017,
      "loss": 5.1511,
      "mean_token_accuracy": 0.17451186925172807,
      "num_tokens": 50834156.0,
      "step": 23205
    },
    {
      "entropy": 5.58366847038269,
      "epoch": 2.3034934497816595,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044716701885108275,
      "loss": 5.0241,
      "mean_token_accuracy": 0.1829358920454979,
      "num_tokens": 50845086.0,
      "step": 23210
    },
    {
      "entropy": 5.618365049362183,
      "epoch": 2.303989678443827,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044714420562220423,
      "loss": 5.0307,
      "mean_token_accuracy": 0.18179574459791184,
      "num_tokens": 50858637.0,
      "step": 23215
    },
    {
      "entropy": 5.644404697418213,
      "epoch": 2.3044859071059944,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004471213881245204,
      "loss": 5.1558,
      "mean_token_accuracy": 0.1749633491039276,
      "num_tokens": 50871159.0,
      "step": 23220
    },
    {
      "entropy": 5.5893453598022464,
      "epoch": 2.304982135768162,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000447098566358597,
      "loss": 5.0408,
      "mean_token_accuracy": 0.17604662775993346,
      "num_tokens": 50884072.0,
      "step": 23225
    },
    {
      "entropy": 5.557044172286988,
      "epoch": 2.3054783644303294,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00044707574032500005,
      "loss": 5.0443,
      "mean_token_accuracy": 0.18032577335834504,
      "num_tokens": 50896627.0,
      "step": 23230
    },
    {
      "entropy": 5.555281066894532,
      "epoch": 2.305974593092497,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004470529100242955,
      "loss": 4.9643,
      "mean_token_accuracy": 0.1855953097343445,
      "num_tokens": 50907476.0,
      "step": 23235
    },
    {
      "entropy": 5.589204359054565,
      "epoch": 2.3064708217546643,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004470300754570495,
      "loss": 5.0463,
      "mean_token_accuracy": 0.17860807180404664,
      "num_tokens": 50918104.0,
      "step": 23240
    },
    {
      "entropy": 5.560319232940674,
      "epoch": 2.3069670504168323,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004470072366238284,
      "loss": 5.0121,
      "mean_token_accuracy": 0.17882056087255477,
      "num_tokens": 50930575.0,
      "step": 23245
    },
    {
      "entropy": 5.568819761276245,
      "epoch": 2.3074632790789997,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044698439352519836,
      "loss": 4.97,
      "mean_token_accuracy": 0.18190665692090988,
      "num_tokens": 50940592.0,
      "step": 23250
    },
    {
      "entropy": 5.57731990814209,
      "epoch": 2.307959507741167,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044696154616172604,
      "loss": 5.0411,
      "mean_token_accuracy": 0.17612005770206451,
      "num_tokens": 50952134.0,
      "step": 23255
    },
    {
      "entropy": 5.5701545715332035,
      "epoch": 2.3084557364033347,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004469386945339779,
      "loss": 5.0846,
      "mean_token_accuracy": 0.1784474566578865,
      "num_tokens": 50962721.0,
      "step": 23260
    },
    {
      "entropy": 5.57421989440918,
      "epoch": 2.308951965065502,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004469158386425206,
      "loss": 5.0229,
      "mean_token_accuracy": 0.18003182262182235,
      "num_tokens": 50974893.0,
      "step": 23265
    },
    {
      "entropy": 5.594786310195923,
      "epoch": 2.3094481937276696,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000446892978487921,
      "loss": 4.9969,
      "mean_token_accuracy": 0.1797194093465805,
      "num_tokens": 50986225.0,
      "step": 23270
    },
    {
      "entropy": 5.5979804515838625,
      "epoch": 2.309944422389837,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00044687011407074583,
      "loss": 5.1009,
      "mean_token_accuracy": 0.1689686268568039,
      "num_tokens": 50997924.0,
      "step": 23275
    },
    {
      "entropy": 5.584272670745849,
      "epoch": 2.3104406510520046,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004468472453915623,
      "loss": 4.9806,
      "mean_token_accuracy": 0.18198306262493133,
      "num_tokens": 51008179.0,
      "step": 23280
    },
    {
      "entropy": 5.544312620162964,
      "epoch": 2.3109368797141725,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004468243724509373,
      "loss": 4.9953,
      "mean_token_accuracy": 0.18630722910165787,
      "num_tokens": 51018462.0,
      "step": 23285
    },
    {
      "entropy": 5.618963098526001,
      "epoch": 2.31143310837634,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004468014952494382,
      "loss": 5.0724,
      "mean_token_accuracy": 0.17896728962659836,
      "num_tokens": 51029769.0,
      "step": 23290
    },
    {
      "entropy": 5.616843557357788,
      "epoch": 2.3119293370385074,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044677861378763205,
      "loss": 5.0759,
      "mean_token_accuracy": 0.1816832974553108,
      "num_tokens": 51041058.0,
      "step": 23295
    },
    {
      "entropy": 5.587506484985352,
      "epoch": 2.312425565700675,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044675572806608655,
      "loss": 5.0129,
      "mean_token_accuracy": 0.18079742789268494,
      "num_tokens": 51050958.0,
      "step": 23300
    },
    {
      "entropy": 5.618343257904053,
      "epoch": 2.3129217943628424,
      "grad_norm": 0.984375,
      "learning_rate": 0.00044673283808536915,
      "loss": 5.0231,
      "mean_token_accuracy": 0.18827854543924333,
      "num_tokens": 51062459.0,
      "step": 23305
    },
    {
      "entropy": 5.688998031616211,
      "epoch": 2.31341802302501,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004467099438460474,
      "loss": 5.211,
      "mean_token_accuracy": 0.1744321808218956,
      "num_tokens": 51074596.0,
      "step": 23310
    },
    {
      "entropy": 5.66031699180603,
      "epoch": 2.3139142516871773,
      "grad_norm": 1.125,
      "learning_rate": 0.00044668704534868886,
      "loss": 5.0712,
      "mean_token_accuracy": 0.17126382291316986,
      "num_tokens": 51085265.0,
      "step": 23315
    },
    {
      "entropy": 5.60770001411438,
      "epoch": 2.314410480349345,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00044666414259386173,
      "loss": 5.0168,
      "mean_token_accuracy": 0.17705831974744796,
      "num_tokens": 51095986.0,
      "step": 23320
    },
    {
      "entropy": 5.623257780075074,
      "epoch": 2.3149067090115123,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044664123558213367,
      "loss": 5.016,
      "mean_token_accuracy": 0.17910957038402558,
      "num_tokens": 51107518.0,
      "step": 23325
    },
    {
      "entropy": 5.569105386734009,
      "epoch": 2.31540293767368,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004466183243140729,
      "loss": 4.9738,
      "mean_token_accuracy": 0.18646175414323807,
      "num_tokens": 51119715.0,
      "step": 23330
    },
    {
      "entropy": 5.548243427276612,
      "epoch": 2.3158991663358477,
      "grad_norm": 0.9375,
      "learning_rate": 0.00044659540879024746,
      "loss": 4.9943,
      "mean_token_accuracy": 0.18227538615465164,
      "num_tokens": 51130801.0,
      "step": 23335
    },
    {
      "entropy": 5.555860805511474,
      "epoch": 2.316395394998015,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004465724890112256,
      "loss": 5.042,
      "mean_token_accuracy": 0.17628263980150222,
      "num_tokens": 51142509.0,
      "step": 23340
    },
    {
      "entropy": 5.6402171611785885,
      "epoch": 2.3168916236601826,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044654956497757576,
      "loss": 5.0763,
      "mean_token_accuracy": 0.17936841696500777,
      "num_tokens": 51154839.0,
      "step": 23345
    },
    {
      "entropy": 5.57779016494751,
      "epoch": 2.31738785232235,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044652663668986634,
      "loss": 5.0225,
      "mean_token_accuracy": 0.18539883345365524,
      "num_tokens": 51165293.0,
      "step": 23350
    },
    {
      "entropy": 5.556305551528931,
      "epoch": 2.3178840809845176,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000446503704148666,
      "loss": 5.0857,
      "mean_token_accuracy": 0.1782493069767952,
      "num_tokens": 51176371.0,
      "step": 23355
    },
    {
      "entropy": 5.564503955841064,
      "epoch": 2.318380309646685,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004464807673545433,
      "loss": 5.0106,
      "mean_token_accuracy": 0.18717833757400512,
      "num_tokens": 51187240.0,
      "step": 23360
    },
    {
      "entropy": 5.670996856689453,
      "epoch": 2.3188765383088525,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004464578263080671,
      "loss": 5.0991,
      "mean_token_accuracy": 0.17741835564374925,
      "num_tokens": 51199084.0,
      "step": 23365
    },
    {
      "entropy": 5.589241361618042,
      "epoch": 2.3193727669710205,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044643488100980623,
      "loss": 4.9714,
      "mean_token_accuracy": 0.1834746092557907,
      "num_tokens": 51209885.0,
      "step": 23370
    },
    {
      "entropy": 5.594709682464599,
      "epoch": 2.319868995633188,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004464119314603298,
      "loss": 5.0283,
      "mean_token_accuracy": 0.18390836268663407,
      "num_tokens": 51220644.0,
      "step": 23375
    },
    {
      "entropy": 5.58029088973999,
      "epoch": 2.3203652242953554,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00044638897766020673,
      "loss": 5.0103,
      "mean_token_accuracy": 0.1815899059176445,
      "num_tokens": 51231342.0,
      "step": 23380
    },
    {
      "entropy": 5.5645177364349365,
      "epoch": 2.320861452957523,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004463660196100064,
      "loss": 4.9299,
      "mean_token_accuracy": 0.1838017001748085,
      "num_tokens": 51241368.0,
      "step": 23385
    },
    {
      "entropy": 5.547626113891601,
      "epoch": 2.3213576816196904,
      "grad_norm": 0.984375,
      "learning_rate": 0.00044634305731029803,
      "loss": 5.1229,
      "mean_token_accuracy": 0.1810004472732544,
      "num_tokens": 51252784.0,
      "step": 23390
    },
    {
      "entropy": 5.5737354278564455,
      "epoch": 2.321853910281858,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044632009076165096,
      "loss": 5.0087,
      "mean_token_accuracy": 0.1803363710641861,
      "num_tokens": 51264246.0,
      "step": 23395
    },
    {
      "entropy": 5.581436586380005,
      "epoch": 2.3223501389440253,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000446297119964635,
      "loss": 4.9831,
      "mean_token_accuracy": 0.18832659125328063,
      "num_tokens": 51274575.0,
      "step": 23400
    },
    {
      "entropy": 5.611540079116821,
      "epoch": 2.322846367606193,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004462741449198194,
      "loss": 4.9903,
      "mean_token_accuracy": 0.18128154873847963,
      "num_tokens": 51285384.0,
      "step": 23405
    },
    {
      "entropy": 5.653060007095337,
      "epoch": 2.3233425962683603,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044625116562777413,
      "loss": 5.08,
      "mean_token_accuracy": 0.18511968553066255,
      "num_tokens": 51297983.0,
      "step": 23410
    },
    {
      "entropy": 5.511483383178711,
      "epoch": 2.323838824930528,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004462281820890689,
      "loss": 4.9822,
      "mean_token_accuracy": 0.18765173256397247,
      "num_tokens": 51308795.0,
      "step": 23415
    },
    {
      "entropy": 5.547490835189819,
      "epoch": 2.3243350535926957,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044620519430427376,
      "loss": 5.0025,
      "mean_token_accuracy": 0.18458894938230513,
      "num_tokens": 51318597.0,
      "step": 23420
    },
    {
      "entropy": 5.550917911529541,
      "epoch": 2.324831282254863,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004461822022739588,
      "loss": 4.9494,
      "mean_token_accuracy": 0.18288534581661225,
      "num_tokens": 51329238.0,
      "step": 23425
    },
    {
      "entropy": 5.615041065216064,
      "epoch": 2.3253275109170306,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000446159205998694,
      "loss": 5.0423,
      "mean_token_accuracy": 0.18075837790966034,
      "num_tokens": 51340295.0,
      "step": 23430
    },
    {
      "entropy": 5.564624977111817,
      "epoch": 2.325823739579198,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004461362054790497,
      "loss": 5.015,
      "mean_token_accuracy": 0.1820700079202652,
      "num_tokens": 51350702.0,
      "step": 23435
    },
    {
      "entropy": 5.612828254699707,
      "epoch": 2.3263199682413656,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044611320071559637,
      "loss": 5.0318,
      "mean_token_accuracy": 0.18055972903966905,
      "num_tokens": 51359794.0,
      "step": 23440
    },
    {
      "entropy": 5.606128787994384,
      "epoch": 2.326816196903533,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044609019170890424,
      "loss": 5.0004,
      "mean_token_accuracy": 0.18530600368976594,
      "num_tokens": 51368932.0,
      "step": 23445
    },
    {
      "entropy": 5.617540693283081,
      "epoch": 2.3273124255657005,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00044606717845954415,
      "loss": 5.0662,
      "mean_token_accuracy": 0.18132403939962388,
      "num_tokens": 51381111.0,
      "step": 23450
    },
    {
      "entropy": 5.534998655319214,
      "epoch": 2.3278086542278684,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044604416096808656,
      "loss": 4.9757,
      "mean_token_accuracy": 0.19520024508237838,
      "num_tokens": 51392976.0,
      "step": 23455
    },
    {
      "entropy": 5.614948272705078,
      "epoch": 2.328304882890036,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044602113923510235,
      "loss": 5.0058,
      "mean_token_accuracy": 0.17893098592758178,
      "num_tokens": 51403586.0,
      "step": 23460
    },
    {
      "entropy": 5.6019322872161865,
      "epoch": 2.3288011115522034,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044599811326116246,
      "loss": 5.0636,
      "mean_token_accuracy": 0.17843217104673387,
      "num_tokens": 51414035.0,
      "step": 23465
    },
    {
      "entropy": 5.590606069564819,
      "epoch": 2.329297340214371,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044597508304683784,
      "loss": 5.0662,
      "mean_token_accuracy": 0.17720458209514617,
      "num_tokens": 51424111.0,
      "step": 23470
    },
    {
      "entropy": 5.5653258800506595,
      "epoch": 2.3297935688765383,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004459520485926995,
      "loss": 4.9697,
      "mean_token_accuracy": 0.1882004588842392,
      "num_tokens": 51434566.0,
      "step": 23475
    },
    {
      "entropy": 5.644761228561402,
      "epoch": 2.330289797538706,
      "grad_norm": 1.0,
      "learning_rate": 0.0004459290098993188,
      "loss": 5.0334,
      "mean_token_accuracy": 0.18696822673082353,
      "num_tokens": 51445568.0,
      "step": 23480
    },
    {
      "entropy": 5.636776876449585,
      "epoch": 2.3307860262008733,
      "grad_norm": 0.953125,
      "learning_rate": 0.00044590596696726686,
      "loss": 5.0246,
      "mean_token_accuracy": 0.18721263408660888,
      "num_tokens": 51456482.0,
      "step": 23485
    },
    {
      "entropy": 5.620083570480347,
      "epoch": 2.3312822548630407,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004458829197971153,
      "loss": 5.1081,
      "mean_token_accuracy": 0.17495281994342804,
      "num_tokens": 51467809.0,
      "step": 23490
    },
    {
      "entropy": 5.63603720664978,
      "epoch": 2.3317784835252082,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044585986838943556,
      "loss": 5.0175,
      "mean_token_accuracy": 0.17824462950229644,
      "num_tokens": 51477927.0,
      "step": 23495
    },
    {
      "entropy": 5.593960237503052,
      "epoch": 2.3322747121873757,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044583681274479927,
      "loss": 5.071,
      "mean_token_accuracy": 0.18142592459917067,
      "num_tokens": 51488432.0,
      "step": 23500
    },
    {
      "entropy": 5.617846202850342,
      "epoch": 2.3327709408495436,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004458137528637781,
      "loss": 5.0394,
      "mean_token_accuracy": 0.18071080446243287,
      "num_tokens": 51499149.0,
      "step": 23505
    },
    {
      "entropy": 5.666125011444092,
      "epoch": 2.333267169511711,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044579068874694386,
      "loss": 5.071,
      "mean_token_accuracy": 0.17561647146940232,
      "num_tokens": 51510672.0,
      "step": 23510
    },
    {
      "entropy": 5.641594696044922,
      "epoch": 2.3337633981738786,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004457676203948687,
      "loss": 5.0617,
      "mean_token_accuracy": 0.17588749378919602,
      "num_tokens": 51521855.0,
      "step": 23515
    },
    {
      "entropy": 5.645274972915649,
      "epoch": 2.334259626836046,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004457445478081244,
      "loss": 5.0383,
      "mean_token_accuracy": 0.18006182760000228,
      "num_tokens": 51531970.0,
      "step": 23520
    },
    {
      "entropy": 5.60600733757019,
      "epoch": 2.3347558554982135,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044572147098728335,
      "loss": 5.1153,
      "mean_token_accuracy": 0.17813000679016114,
      "num_tokens": 51543110.0,
      "step": 23525
    },
    {
      "entropy": 5.5759031772613525,
      "epoch": 2.335252084160381,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044569838993291757,
      "loss": 5.0453,
      "mean_token_accuracy": 0.17850546538829803,
      "num_tokens": 51552991.0,
      "step": 23530
    },
    {
      "entropy": 5.639141035079956,
      "epoch": 2.3357483128225485,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044567530464559963,
      "loss": 5.0638,
      "mean_token_accuracy": 0.18162497878074646,
      "num_tokens": 51564472.0,
      "step": 23535
    },
    {
      "entropy": 5.589152526855469,
      "epoch": 2.3362445414847164,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004456522151259019,
      "loss": 5.0814,
      "mean_token_accuracy": 0.18080969154834747,
      "num_tokens": 51576051.0,
      "step": 23540
    },
    {
      "entropy": 5.670563268661499,
      "epoch": 2.336740770146884,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000445629121374397,
      "loss": 5.0722,
      "mean_token_accuracy": 0.17418465614318848,
      "num_tokens": 51586658.0,
      "step": 23545
    },
    {
      "entropy": 5.614601182937622,
      "epoch": 2.3372369988090513,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044560602339165746,
      "loss": 5.0389,
      "mean_token_accuracy": 0.1779758632183075,
      "num_tokens": 51596614.0,
      "step": 23550
    },
    {
      "entropy": 5.590232896804809,
      "epoch": 2.337733227471219,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004455829211782562,
      "loss": 5.0717,
      "mean_token_accuracy": 0.17915811985731125,
      "num_tokens": 51607102.0,
      "step": 23555
    },
    {
      "entropy": 5.675381803512574,
      "epoch": 2.3382294561333863,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004455598147347661,
      "loss": 5.1025,
      "mean_token_accuracy": 0.1766846612095833,
      "num_tokens": 51619079.0,
      "step": 23560
    },
    {
      "entropy": 5.6101583480834964,
      "epoch": 2.3387256847955538,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004455367040617601,
      "loss": 4.9859,
      "mean_token_accuracy": 0.18753135502338408,
      "num_tokens": 51630273.0,
      "step": 23565
    },
    {
      "entropy": 5.56528606414795,
      "epoch": 2.3392219134577212,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00044551358915981135,
      "loss": 5.0409,
      "mean_token_accuracy": 0.1820089712738991,
      "num_tokens": 51640480.0,
      "step": 23570
    },
    {
      "entropy": 5.593130826950073,
      "epoch": 2.3397181421198887,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004454904700294929,
      "loss": 5.0766,
      "mean_token_accuracy": 0.1786801502108574,
      "num_tokens": 51651831.0,
      "step": 23575
    },
    {
      "entropy": 5.60858006477356,
      "epoch": 2.340214370782056,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004454673466713783,
      "loss": 5.0751,
      "mean_token_accuracy": 0.17778751105070115,
      "num_tokens": 51664061.0,
      "step": 23580
    },
    {
      "entropy": 5.590447902679443,
      "epoch": 2.3407105994442237,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044544421908604073,
      "loss": 5.026,
      "mean_token_accuracy": 0.1818692833185196,
      "num_tokens": 51675081.0,
      "step": 23585
    },
    {
      "entropy": 5.575238418579102,
      "epoch": 2.3412068281063916,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044542108727405386,
      "loss": 4.9373,
      "mean_token_accuracy": 0.1871587172150612,
      "num_tokens": 51684477.0,
      "step": 23590
    },
    {
      "entropy": 5.702288722991943,
      "epoch": 2.341703056768559,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044539795123599117,
      "loss": 5.0957,
      "mean_token_accuracy": 0.18402961492538453,
      "num_tokens": 51694420.0,
      "step": 23595
    },
    {
      "entropy": 5.474022626876831,
      "epoch": 2.3421992854307265,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00044537481097242654,
      "loss": 4.9272,
      "mean_token_accuracy": 0.1897786557674408,
      "num_tokens": 51704635.0,
      "step": 23600
    },
    {
      "entropy": 5.610105323791504,
      "epoch": 2.342695514092894,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044535166648393377,
      "loss": 5.1388,
      "mean_token_accuracy": 0.17504112869501115,
      "num_tokens": 51716149.0,
      "step": 23605
    },
    {
      "entropy": 5.548909664154053,
      "epoch": 2.3431917427550615,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00044532851777108663,
      "loss": 4.9788,
      "mean_token_accuracy": 0.18450577110052108,
      "num_tokens": 51726861.0,
      "step": 23610
    },
    {
      "entropy": 5.648274898529053,
      "epoch": 2.343687971417229,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004453053648344593,
      "loss": 5.2094,
      "mean_token_accuracy": 0.17432949244976043,
      "num_tokens": 51738111.0,
      "step": 23615
    },
    {
      "entropy": 5.565129804611206,
      "epoch": 2.3441842000793964,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044528220767462595,
      "loss": 5.0422,
      "mean_token_accuracy": 0.18279070258140565,
      "num_tokens": 51749004.0,
      "step": 23620
    },
    {
      "entropy": 5.580155849456787,
      "epoch": 2.3446804287415643,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004452590462921607,
      "loss": 5.022,
      "mean_token_accuracy": 0.17733656764030456,
      "num_tokens": 51759557.0,
      "step": 23625
    },
    {
      "entropy": 5.626314878463745,
      "epoch": 2.345176657403732,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044523588068763804,
      "loss": 5.0883,
      "mean_token_accuracy": 0.17816792726516723,
      "num_tokens": 51769921.0,
      "step": 23630
    },
    {
      "entropy": 5.572683095932007,
      "epoch": 2.3456728860658993,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004452127108616324,
      "loss": 4.9548,
      "mean_token_accuracy": 0.19134241491556167,
      "num_tokens": 51780158.0,
      "step": 23635
    },
    {
      "entropy": 5.638480758666992,
      "epoch": 2.3461691147280668,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044518953681471824,
      "loss": 5.0588,
      "mean_token_accuracy": 0.17901730984449388,
      "num_tokens": 51790502.0,
      "step": 23640
    },
    {
      "entropy": 5.5948234558105465,
      "epoch": 2.3466653433902342,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004451663585474703,
      "loss": 5.0759,
      "mean_token_accuracy": 0.17232829928398133,
      "num_tokens": 51801420.0,
      "step": 23645
    },
    {
      "entropy": 5.60740098953247,
      "epoch": 2.3471615720524017,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044514317606046336,
      "loss": 5.0742,
      "mean_token_accuracy": 0.18954353034496307,
      "num_tokens": 51812399.0,
      "step": 23650
    },
    {
      "entropy": 5.660820198059082,
      "epoch": 2.347657800714569,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004451199893542723,
      "loss": 5.1183,
      "mean_token_accuracy": 0.17228747755289078,
      "num_tokens": 51824268.0,
      "step": 23655
    },
    {
      "entropy": 5.561337614059449,
      "epoch": 2.3481540293767367,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004450967984294721,
      "loss": 5.0448,
      "mean_token_accuracy": 0.17698400169610978,
      "num_tokens": 51835842.0,
      "step": 23660
    },
    {
      "entropy": 5.583753490447998,
      "epoch": 2.348650258038904,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004450736032866377,
      "loss": 4.9846,
      "mean_token_accuracy": 0.1826761394739151,
      "num_tokens": 51847501.0,
      "step": 23665
    },
    {
      "entropy": 5.595281982421875,
      "epoch": 2.3491464867010716,
      "grad_norm": 1.0,
      "learning_rate": 0.00044505040392634447,
      "loss": 5.0305,
      "mean_token_accuracy": 0.1872783422470093,
      "num_tokens": 51858056.0,
      "step": 23670
    },
    {
      "entropy": 5.636955642700196,
      "epoch": 2.3496427153632395,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004450272003491677,
      "loss": 4.9967,
      "mean_token_accuracy": 0.1887053519487381,
      "num_tokens": 51869248.0,
      "step": 23675
    },
    {
      "entropy": 5.593786430358887,
      "epoch": 2.350138944025407,
      "grad_norm": 0.953125,
      "learning_rate": 0.00044500399255568274,
      "loss": 5.0904,
      "mean_token_accuracy": 0.1746561273932457,
      "num_tokens": 51881496.0,
      "step": 23680
    },
    {
      "entropy": 5.569082164764405,
      "epoch": 2.3506351726875745,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00044498078054646507,
      "loss": 5.0407,
      "mean_token_accuracy": 0.17979726344347,
      "num_tokens": 51891851.0,
      "step": 23685
    },
    {
      "entropy": 5.6511420726776125,
      "epoch": 2.351131401349742,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044495756432209025,
      "loss": 5.1273,
      "mean_token_accuracy": 0.1777437835931778,
      "num_tokens": 51903581.0,
      "step": 23690
    },
    {
      "entropy": 5.58378381729126,
      "epoch": 2.3516276300119094,
      "grad_norm": 0.96875,
      "learning_rate": 0.00044493434388313423,
      "loss": 4.9651,
      "mean_token_accuracy": 0.18419304937124253,
      "num_tokens": 51913650.0,
      "step": 23695
    },
    {
      "entropy": 5.526974248886108,
      "epoch": 2.352123858674077,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044491111923017254,
      "loss": 4.9804,
      "mean_token_accuracy": 0.18517538905143738,
      "num_tokens": 51924704.0,
      "step": 23700
    },
    {
      "entropy": 5.654710197448731,
      "epoch": 2.3526200873362444,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00044488789036378117,
      "loss": 5.1115,
      "mean_token_accuracy": 0.177008980512619,
      "num_tokens": 51937576.0,
      "step": 23705
    },
    {
      "entropy": 5.613334655761719,
      "epoch": 2.3531163159984123,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004448646572845363,
      "loss": 4.9575,
      "mean_token_accuracy": 0.1870013654232025,
      "num_tokens": 51947359.0,
      "step": 23710
    },
    {
      "entropy": 5.598943471908569,
      "epoch": 2.35361254466058,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004448414199930139,
      "loss": 5.1005,
      "mean_token_accuracy": 0.1712775155901909,
      "num_tokens": 51958016.0,
      "step": 23715
    },
    {
      "entropy": 5.67420072555542,
      "epoch": 2.3541087733227473,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044481817848979026,
      "loss": 5.0988,
      "mean_token_accuracy": 0.1787247657775879,
      "num_tokens": 51968770.0,
      "step": 23720
    },
    {
      "entropy": 5.59364686012268,
      "epoch": 2.3546050019849147,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004447949327754418,
      "loss": 4.9672,
      "mean_token_accuracy": 0.1912851944565773,
      "num_tokens": 51980052.0,
      "step": 23725
    },
    {
      "entropy": 5.6142831325531,
      "epoch": 2.355101230647082,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004447716828505447,
      "loss": 5.1244,
      "mean_token_accuracy": 0.17506493330001832,
      "num_tokens": 51992080.0,
      "step": 23730
    },
    {
      "entropy": 5.582031965255737,
      "epoch": 2.3555974593092497,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004447484287156758,
      "loss": 5.0468,
      "mean_token_accuracy": 0.18303615152835845,
      "num_tokens": 52004313.0,
      "step": 23735
    },
    {
      "entropy": 5.650063705444336,
      "epoch": 2.356093687971417,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004447251703714117,
      "loss": 5.0334,
      "mean_token_accuracy": 0.18002406358718873,
      "num_tokens": 52015207.0,
      "step": 23740
    },
    {
      "entropy": 5.6015260219573975,
      "epoch": 2.3565899166335846,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000444701907818329,
      "loss": 5.0347,
      "mean_token_accuracy": 0.17736193239688874,
      "num_tokens": 52026565.0,
      "step": 23745
    },
    {
      "entropy": 5.635419511795044,
      "epoch": 2.357086145295752,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004446786410570048,
      "loss": 5.0646,
      "mean_token_accuracy": 0.17616048753261565,
      "num_tokens": 52037654.0,
      "step": 23750
    },
    {
      "entropy": 5.558535528182984,
      "epoch": 2.3575823739579196,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044465537008801575,
      "loss": 4.961,
      "mean_token_accuracy": 0.1841454967856407,
      "num_tokens": 52049772.0,
      "step": 23755
    },
    {
      "entropy": 5.580855417251587,
      "epoch": 2.3580786026200875,
      "grad_norm": 1.125,
      "learning_rate": 0.0004446320949119392,
      "loss": 5.105,
      "mean_token_accuracy": 0.17487892657518386,
      "num_tokens": 52061363.0,
      "step": 23760
    },
    {
      "entropy": 5.607885313034058,
      "epoch": 2.358574831282255,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044460881552935215,
      "loss": 4.9709,
      "mean_token_accuracy": 0.18495057225227357,
      "num_tokens": 52073153.0,
      "step": 23765
    },
    {
      "entropy": 5.561751937866211,
      "epoch": 2.3590710599444225,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044458553194083204,
      "loss": 5.0096,
      "mean_token_accuracy": 0.18456631302833557,
      "num_tokens": 52083369.0,
      "step": 23770
    },
    {
      "entropy": 5.523240756988526,
      "epoch": 2.35956728860659,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004445622441469562,
      "loss": 4.9656,
      "mean_token_accuracy": 0.18823008239269257,
      "num_tokens": 52093701.0,
      "step": 23775
    },
    {
      "entropy": 5.569080257415772,
      "epoch": 2.3600635172687574,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00044453895214830195,
      "loss": 5.0255,
      "mean_token_accuracy": 0.1830786868929863,
      "num_tokens": 52105863.0,
      "step": 23780
    },
    {
      "entropy": 5.543252277374267,
      "epoch": 2.360559745930925,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004445156559454471,
      "loss": 5.017,
      "mean_token_accuracy": 0.1818300813436508,
      "num_tokens": 52116476.0,
      "step": 23785
    },
    {
      "entropy": 5.666142463684082,
      "epoch": 2.3610559745930924,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044449235553896914,
      "loss": 5.1165,
      "mean_token_accuracy": 0.1691933959722519,
      "num_tokens": 52127262.0,
      "step": 23790
    },
    {
      "entropy": 5.686756420135498,
      "epoch": 2.36155220325526,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004444690509294461,
      "loss": 5.0729,
      "mean_token_accuracy": 0.17669105380773545,
      "num_tokens": 52138575.0,
      "step": 23795
    },
    {
      "entropy": 5.585786199569702,
      "epoch": 2.3620484319174277,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004444457421174558,
      "loss": 5.0278,
      "mean_token_accuracy": 0.18147281259298326,
      "num_tokens": 52149620.0,
      "step": 23800
    },
    {
      "entropy": 5.639217281341553,
      "epoch": 2.3625446605795952,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004444224291035761,
      "loss": 5.0612,
      "mean_token_accuracy": 0.1762116014957428,
      "num_tokens": 52161929.0,
      "step": 23805
    },
    {
      "entropy": 5.512679529190064,
      "epoch": 2.3630408892417627,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044439911188838533,
      "loss": 4.9882,
      "mean_token_accuracy": 0.18888108134269715,
      "num_tokens": 52172283.0,
      "step": 23810
    },
    {
      "entropy": 5.584146881103516,
      "epoch": 2.36353711790393,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004443757904724615,
      "loss": 5.0783,
      "mean_token_accuracy": 0.1781835824251175,
      "num_tokens": 52184130.0,
      "step": 23815
    },
    {
      "entropy": 5.588577032089233,
      "epoch": 2.3640333465660976,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00044435246485638317,
      "loss": 5.0846,
      "mean_token_accuracy": 0.18076565861701965,
      "num_tokens": 52195358.0,
      "step": 23820
    },
    {
      "entropy": 5.568464565277099,
      "epoch": 2.364529575228265,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004443291350407285,
      "loss": 4.979,
      "mean_token_accuracy": 0.18197174072265626,
      "num_tokens": 52207070.0,
      "step": 23825
    },
    {
      "entropy": 5.605605506896973,
      "epoch": 2.3650258038904326,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004443058010260762,
      "loss": 5.1006,
      "mean_token_accuracy": 0.17477748095989226,
      "num_tokens": 52217658.0,
      "step": 23830
    },
    {
      "entropy": 5.577688264846802,
      "epoch": 2.3655220325526,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044428246281300493,
      "loss": 5.0051,
      "mean_token_accuracy": 0.18622718900442123,
      "num_tokens": 52229350.0,
      "step": 23835
    },
    {
      "entropy": 5.656297874450684,
      "epoch": 2.3660182612147675,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00044425912040209326,
      "loss": 5.1592,
      "mean_token_accuracy": 0.1753564551472664,
      "num_tokens": 52241260.0,
      "step": 23840
    },
    {
      "entropy": 5.675198030471802,
      "epoch": 2.3665144898769355,
      "grad_norm": 0.953125,
      "learning_rate": 0.00044423577379392023,
      "loss": 5.061,
      "mean_token_accuracy": 0.18123331516981125,
      "num_tokens": 52253049.0,
      "step": 23845
    },
    {
      "entropy": 5.585350799560547,
      "epoch": 2.367010718539103,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004442124229890646,
      "loss": 5.0358,
      "mean_token_accuracy": 0.18232847154140472,
      "num_tokens": 52263405.0,
      "step": 23850
    },
    {
      "entropy": 5.5854637145996096,
      "epoch": 2.3675069472012704,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044418906798810554,
      "loss": 5.0412,
      "mean_token_accuracy": 0.18200973719358443,
      "num_tokens": 52273075.0,
      "step": 23855
    },
    {
      "entropy": 5.7175764560699465,
      "epoch": 2.368003175863438,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00044416570879162213,
      "loss": 5.2322,
      "mean_token_accuracy": 0.17419291734695436,
      "num_tokens": 52284275.0,
      "step": 23860
    },
    {
      "entropy": 5.59305853843689,
      "epoch": 2.3684994045256054,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004441423454001936,
      "loss": 5.034,
      "mean_token_accuracy": 0.17180906534194945,
      "num_tokens": 52295084.0,
      "step": 23865
    },
    {
      "entropy": 5.586873531341553,
      "epoch": 2.368995633187773,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044411897781439946,
      "loss": 5.0266,
      "mean_token_accuracy": 0.18246820122003554,
      "num_tokens": 52304172.0,
      "step": 23870
    },
    {
      "entropy": 5.498567056655884,
      "epoch": 2.3694918618499403,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004440956060348191,
      "loss": 4.9748,
      "mean_token_accuracy": 0.18676338642835616,
      "num_tokens": 52315370.0,
      "step": 23875
    },
    {
      "entropy": 5.597283029556275,
      "epoch": 2.369988090512108,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004440722300620321,
      "loss": 5.0375,
      "mean_token_accuracy": 0.18130150586366653,
      "num_tokens": 52326517.0,
      "step": 23880
    },
    {
      "entropy": 5.662319087982178,
      "epoch": 2.3704843191742757,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044404884989661804,
      "loss": 5.0865,
      "mean_token_accuracy": 0.18310338705778123,
      "num_tokens": 52337110.0,
      "step": 23885
    },
    {
      "entropy": 5.592140817642212,
      "epoch": 2.370980547836443,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004440254655391568,
      "loss": 5.0039,
      "mean_token_accuracy": 0.1850810244679451,
      "num_tokens": 52347509.0,
      "step": 23890
    },
    {
      "entropy": 5.670315742492676,
      "epoch": 2.3714767764986107,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004440020769902282,
      "loss": 5.1361,
      "mean_token_accuracy": 0.17586909085512162,
      "num_tokens": 52357920.0,
      "step": 23895
    },
    {
      "entropy": 5.63978009223938,
      "epoch": 2.371973005160778,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00044397868425041234,
      "loss": 5.0995,
      "mean_token_accuracy": 0.1759955793619156,
      "num_tokens": 52368138.0,
      "step": 23900
    },
    {
      "entropy": 5.667237234115601,
      "epoch": 2.3724692338229456,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004439552873202891,
      "loss": 5.0594,
      "mean_token_accuracy": 0.1795011579990387,
      "num_tokens": 52380503.0,
      "step": 23905
    },
    {
      "entropy": 5.6044487953186035,
      "epoch": 2.372965462485113,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044393188620043893,
      "loss": 5.0472,
      "mean_token_accuracy": 0.18059685826301575,
      "num_tokens": 52390850.0,
      "step": 23910
    },
    {
      "entropy": 5.542678165435791,
      "epoch": 2.3734616911472806,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004439084808914419,
      "loss": 4.985,
      "mean_token_accuracy": 0.18484220504760743,
      "num_tokens": 52400829.0,
      "step": 23915
    },
    {
      "entropy": 5.641800737380981,
      "epoch": 2.373957919809448,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044388507139387864,
      "loss": 5.0681,
      "mean_token_accuracy": 0.18003148436546326,
      "num_tokens": 52412892.0,
      "step": 23920
    },
    {
      "entropy": 5.581310033798218,
      "epoch": 2.3744541484716155,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004438616577083293,
      "loss": 5.0116,
      "mean_token_accuracy": 0.18533659875392913,
      "num_tokens": 52423662.0,
      "step": 23925
    },
    {
      "entropy": 5.600772953033447,
      "epoch": 2.3749503771337834,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000443838239835375,
      "loss": 5.0575,
      "mean_token_accuracy": 0.18391835391521455,
      "num_tokens": 52433541.0,
      "step": 23930
    },
    {
      "entropy": 5.622206354141236,
      "epoch": 2.375446605795951,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004438148177755959,
      "loss": 5.0949,
      "mean_token_accuracy": 0.17893042415380478,
      "num_tokens": 52445504.0,
      "step": 23935
    },
    {
      "entropy": 5.650042963027954,
      "epoch": 2.3759428344581184,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044379139152957327,
      "loss": 5.057,
      "mean_token_accuracy": 0.1753670021891594,
      "num_tokens": 52456116.0,
      "step": 23940
    },
    {
      "entropy": 5.59142107963562,
      "epoch": 2.376439063120286,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044376796109788777,
      "loss": 4.9383,
      "mean_token_accuracy": 0.18848557472229005,
      "num_tokens": 52465462.0,
      "step": 23945
    },
    {
      "entropy": 5.570306777954102,
      "epoch": 2.3769352917824533,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044374452648112046,
      "loss": 5.0982,
      "mean_token_accuracy": 0.17667741924524308,
      "num_tokens": 52477213.0,
      "step": 23950
    },
    {
      "entropy": 5.641315078735351,
      "epoch": 2.377431520444621,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044372108767985253,
      "loss": 5.1004,
      "mean_token_accuracy": 0.17777794152498244,
      "num_tokens": 52488095.0,
      "step": 23955
    },
    {
      "entropy": 5.674362277984619,
      "epoch": 2.3779277491067883,
      "grad_norm": 1.0,
      "learning_rate": 0.0004436976446946652,
      "loss": 5.1435,
      "mean_token_accuracy": 0.17102427035570145,
      "num_tokens": 52499078.0,
      "step": 23960
    },
    {
      "entropy": 5.617254734039307,
      "epoch": 2.3784239777689558,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044367419752613983,
      "loss": 5.0295,
      "mean_token_accuracy": 0.17971991002559662,
      "num_tokens": 52510829.0,
      "step": 23965
    },
    {
      "entropy": 5.6331230163574215,
      "epoch": 2.3789202064311237,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004436507461748577,
      "loss": 5.0403,
      "mean_token_accuracy": 0.17940446585416794,
      "num_tokens": 52521466.0,
      "step": 23970
    },
    {
      "entropy": 5.6306994438171385,
      "epoch": 2.379416435093291,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004436272906414007,
      "loss": 4.961,
      "mean_token_accuracy": 0.1843634158372879,
      "num_tokens": 52532617.0,
      "step": 23975
    },
    {
      "entropy": 5.515544366836548,
      "epoch": 2.3799126637554586,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004436038309263501,
      "loss": 5.0202,
      "mean_token_accuracy": 0.17911292910575866,
      "num_tokens": 52543826.0,
      "step": 23980
    },
    {
      "entropy": 5.560767602920532,
      "epoch": 2.380408892417626,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044358036703028773,
      "loss": 5.0595,
      "mean_token_accuracy": 0.18059613108634948,
      "num_tokens": 52556752.0,
      "step": 23985
    },
    {
      "entropy": 5.628003120422363,
      "epoch": 2.3809051210797936,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00044355689895379563,
      "loss": 5.1201,
      "mean_token_accuracy": 0.17333704084157944,
      "num_tokens": 52567268.0,
      "step": 23990
    },
    {
      "entropy": 5.613879251480102,
      "epoch": 2.381401349741961,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00044353342669745566,
      "loss": 5.0479,
      "mean_token_accuracy": 0.18038588464260102,
      "num_tokens": 52576894.0,
      "step": 23995
    },
    {
      "entropy": 5.578713655471802,
      "epoch": 2.3818975784041285,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00044350995026184976,
      "loss": 5.0341,
      "mean_token_accuracy": 0.18142016530036925,
      "num_tokens": 52587434.0,
      "step": 24000
    },
    {
      "epoch": 2.3818975784041285,
      "eval_entropy": 5.412793871738418,
      "eval_loss": 5.156675815582275,
      "eval_mean_token_accuracy": 0.18144856775336607,
      "eval_num_tokens": 52587434.0,
      "eval_runtime": 20.6019,
      "eval_samples_per_second": 1634.269,
      "eval_steps_per_second": 204.302,
      "step": 24000
    },
    {
      "entropy": 5.662291526794434,
      "epoch": 2.382393807066296,
      "grad_norm": 0.984375,
      "learning_rate": 0.00044348646964756036,
      "loss": 5.1029,
      "mean_token_accuracy": 0.18088118731975555,
      "num_tokens": 52599455.0,
      "step": 24005
    },
    {
      "entropy": 5.65318751335144,
      "epoch": 2.3828900357284635,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004434629848551694,
      "loss": 5.0804,
      "mean_token_accuracy": 0.1775888442993164,
      "num_tokens": 52611020.0,
      "step": 24010
    },
    {
      "entropy": 5.594597339630127,
      "epoch": 2.3833862643906314,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004434394958852595,
      "loss": 5.0312,
      "mean_token_accuracy": 0.1794052690267563,
      "num_tokens": 52621743.0,
      "step": 24015
    },
    {
      "entropy": 5.623204946517944,
      "epoch": 2.383882493052799,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044341600273841304,
      "loss": 5.0771,
      "mean_token_accuracy": 0.18326006978750228,
      "num_tokens": 52633437.0,
      "step": 24020
    },
    {
      "entropy": 5.562138605117798,
      "epoch": 2.3843787217149663,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044339250541521264,
      "loss": 4.9976,
      "mean_token_accuracy": 0.1838457018136978,
      "num_tokens": 52643849.0,
      "step": 24025
    },
    {
      "entropy": 5.693951940536499,
      "epoch": 2.384874950377134,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00044336900391624096,
      "loss": 5.1679,
      "mean_token_accuracy": 0.16831765770912172,
      "num_tokens": 52656417.0,
      "step": 24030
    },
    {
      "entropy": 5.594033718109131,
      "epoch": 2.3853711790393013,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004433454982420808,
      "loss": 4.9486,
      "mean_token_accuracy": 0.18690738677978516,
      "num_tokens": 52668491.0,
      "step": 24035
    },
    {
      "entropy": 5.609128856658936,
      "epoch": 2.3858674077014688,
      "grad_norm": 1.09375,
      "learning_rate": 0.000443321988393315,
      "loss": 5.0869,
      "mean_token_accuracy": 0.17601223289966583,
      "num_tokens": 52679532.0,
      "step": 24040
    },
    {
      "entropy": 5.572027969360351,
      "epoch": 2.3863636363636362,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004432984743705265,
      "loss": 4.9712,
      "mean_token_accuracy": 0.18051217794418334,
      "num_tokens": 52689850.0,
      "step": 24045
    },
    {
      "entropy": 5.529065370559692,
      "epoch": 2.3868598650258037,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044327495617429866,
      "loss": 4.9536,
      "mean_token_accuracy": 0.188804292678833,
      "num_tokens": 52699529.0,
      "step": 24050
    },
    {
      "entropy": 5.5093008518219,
      "epoch": 2.3873560936879716,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004432514338052144,
      "loss": 4.9808,
      "mean_token_accuracy": 0.1861027628183365,
      "num_tokens": 52711449.0,
      "step": 24055
    },
    {
      "entropy": 5.662768077850342,
      "epoch": 2.387852322350139,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004432279072638571,
      "loss": 5.1067,
      "mean_token_accuracy": 0.16947699338197708,
      "num_tokens": 52721528.0,
      "step": 24060
    },
    {
      "entropy": 5.628865575790405,
      "epoch": 2.3883485510123066,
      "grad_norm": 1.671875,
      "learning_rate": 0.00044320437655081033,
      "loss": 5.0066,
      "mean_token_accuracy": 0.18313945084810257,
      "num_tokens": 52731367.0,
      "step": 24065
    },
    {
      "entropy": 5.538984918594361,
      "epoch": 2.388844779674474,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044318084166665737,
      "loss": 4.9281,
      "mean_token_accuracy": 0.19015485644340516,
      "num_tokens": 52740822.0,
      "step": 24070
    },
    {
      "entropy": 5.526667928695678,
      "epoch": 2.3893410083366415,
      "grad_norm": 1.015625,
      "learning_rate": 0.000443157302611982,
      "loss": 5.025,
      "mean_token_accuracy": 0.17424940764904023,
      "num_tokens": 52751609.0,
      "step": 24075
    },
    {
      "entropy": 5.6400188446044925,
      "epoch": 2.389837236998809,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004431337593873678,
      "loss": 5.0647,
      "mean_token_accuracy": 0.17497017085552216,
      "num_tokens": 52762914.0,
      "step": 24080
    },
    {
      "entropy": 5.730286550521851,
      "epoch": 2.3903334656609765,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004431102119933987,
      "loss": 5.105,
      "mean_token_accuracy": 0.1768881320953369,
      "num_tokens": 52773894.0,
      "step": 24085
    },
    {
      "entropy": 5.555190801620483,
      "epoch": 2.390829694323144,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004430866604306586,
      "loss": 5.0362,
      "mean_token_accuracy": 0.1773328274488449,
      "num_tokens": 52784606.0,
      "step": 24090
    },
    {
      "entropy": 5.5716712951660154,
      "epoch": 2.3913259229853114,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00044306310469973147,
      "loss": 5.0161,
      "mean_token_accuracy": 0.18024498373270034,
      "num_tokens": 52795275.0,
      "step": 24095
    },
    {
      "entropy": 5.6666923522949215,
      "epoch": 2.391822151647479,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044303954480120155,
      "loss": 5.0972,
      "mean_token_accuracy": 0.17821034938097,
      "num_tokens": 52805058.0,
      "step": 24100
    },
    {
      "entropy": 5.616100931167603,
      "epoch": 2.392318380309647,
      "grad_norm": 1.125,
      "learning_rate": 0.00044301598073565295,
      "loss": 5.0191,
      "mean_token_accuracy": 0.17934695184230803,
      "num_tokens": 52815486.0,
      "step": 24105
    },
    {
      "entropy": 5.579297876358032,
      "epoch": 2.3928146089718143,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004429924125036702,
      "loss": 5.0377,
      "mean_token_accuracy": 0.18106404095888137,
      "num_tokens": 52827039.0,
      "step": 24110
    },
    {
      "entropy": 5.57388596534729,
      "epoch": 2.393310837633982,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004429688401058375,
      "loss": 5.0553,
      "mean_token_accuracy": 0.1833604693412781,
      "num_tokens": 52837712.0,
      "step": 24115
    },
    {
      "entropy": 5.575838041305542,
      "epoch": 2.3938070662961493,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004429452635427394,
      "loss": 4.9405,
      "mean_token_accuracy": 0.18876053243875504,
      "num_tokens": 52846321.0,
      "step": 24120
    },
    {
      "entropy": 5.598903369903565,
      "epoch": 2.3943032949583167,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044292168281496086,
      "loss": 4.9939,
      "mean_token_accuracy": 0.18464927971363068,
      "num_tokens": 52856342.0,
      "step": 24125
    },
    {
      "entropy": 5.623614835739136,
      "epoch": 2.394799523620484,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004428980979230863,
      "loss": 5.1225,
      "mean_token_accuracy": 0.17436445951461793,
      "num_tokens": 52866602.0,
      "step": 24130
    },
    {
      "entropy": 5.581123876571655,
      "epoch": 2.3952957522826517,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044287450886770075,
      "loss": 4.9366,
      "mean_token_accuracy": 0.1866896256804466,
      "num_tokens": 52877066.0,
      "step": 24135
    },
    {
      "entropy": 5.555600833892822,
      "epoch": 2.3957919809448196,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004428509156493891,
      "loss": 4.9713,
      "mean_token_accuracy": 0.1866391584277153,
      "num_tokens": 52887907.0,
      "step": 24140
    },
    {
      "entropy": 5.663693141937256,
      "epoch": 2.396288209606987,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004428273182687364,
      "loss": 5.1319,
      "mean_token_accuracy": 0.1766095757484436,
      "num_tokens": 52899308.0,
      "step": 24145
    },
    {
      "entropy": 5.583540725708008,
      "epoch": 2.3967844382691545,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004428037167263279,
      "loss": 4.9408,
      "mean_token_accuracy": 0.18997844457626342,
      "num_tokens": 52909127.0,
      "step": 24150
    },
    {
      "entropy": 5.575205898284912,
      "epoch": 2.397280666931322,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044278011102274887,
      "loss": 4.9786,
      "mean_token_accuracy": 0.1858937993645668,
      "num_tokens": 52920503.0,
      "step": 24155
    },
    {
      "entropy": 5.574775123596192,
      "epoch": 2.3977768955934895,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044275650115858445,
      "loss": 5.0014,
      "mean_token_accuracy": 0.17913050651550294,
      "num_tokens": 52931904.0,
      "step": 24160
    },
    {
      "entropy": 5.635742139816284,
      "epoch": 2.398273124255657,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004427328871344204,
      "loss": 5.1495,
      "mean_token_accuracy": 0.16960639953613282,
      "num_tokens": 52945101.0,
      "step": 24165
    },
    {
      "entropy": 5.601398515701294,
      "epoch": 2.3987693529178244,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004427092689508422,
      "loss": 5.0319,
      "mean_token_accuracy": 0.1814688816666603,
      "num_tokens": 52955999.0,
      "step": 24170
    },
    {
      "entropy": 5.568216037750244,
      "epoch": 2.399265581579992,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004426856466084354,
      "loss": 5.0532,
      "mean_token_accuracy": 0.18089765161275864,
      "num_tokens": 52967273.0,
      "step": 24175
    },
    {
      "entropy": 5.63888463973999,
      "epoch": 2.3997618102421594,
      "grad_norm": 1.078125,
      "learning_rate": 0.000442662020107786,
      "loss": 5.1002,
      "mean_token_accuracy": 0.18492758572101592,
      "num_tokens": 52977415.0,
      "step": 24180
    },
    {
      "entropy": 5.5796044826507565,
      "epoch": 2.400258038904327,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004426383894494797,
      "loss": 4.9934,
      "mean_token_accuracy": 0.18083257228136063,
      "num_tokens": 52990071.0,
      "step": 24185
    },
    {
      "entropy": 5.530057191848755,
      "epoch": 2.400754267566495,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004426147546341027,
      "loss": 4.9811,
      "mean_token_accuracy": 0.1810405433177948,
      "num_tokens": 52999874.0,
      "step": 24190
    },
    {
      "entropy": 5.600704383850098,
      "epoch": 2.4012504962286623,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044259111566224084,
      "loss": 5.016,
      "mean_token_accuracy": 0.18385784775018693,
      "num_tokens": 53011349.0,
      "step": 24195
    },
    {
      "entropy": 5.540091514587402,
      "epoch": 2.4017467248908297,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00044256747253448045,
      "loss": 4.9814,
      "mean_token_accuracy": 0.18911761343479155,
      "num_tokens": 53024256.0,
      "step": 24200
    },
    {
      "entropy": 5.582849502563477,
      "epoch": 2.402242953552997,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044254382525140784,
      "loss": 4.9572,
      "mean_token_accuracy": 0.19450874179601668,
      "num_tokens": 53035617.0,
      "step": 24205
    },
    {
      "entropy": 5.579328107833862,
      "epoch": 2.4027391822151647,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044252017381360935,
      "loss": 4.9911,
      "mean_token_accuracy": 0.1905856505036354,
      "num_tokens": 53046219.0,
      "step": 24210
    },
    {
      "entropy": 5.679724884033203,
      "epoch": 2.403235410877332,
      "grad_norm": 1.125,
      "learning_rate": 0.0004424965182216716,
      "loss": 5.11,
      "mean_token_accuracy": 0.1763088271021843,
      "num_tokens": 53057114.0,
      "step": 24215
    },
    {
      "entropy": 5.633517122268676,
      "epoch": 2.4037316395394996,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044247285847618105,
      "loss": 5.0349,
      "mean_token_accuracy": 0.1805433988571167,
      "num_tokens": 53067481.0,
      "step": 24220
    },
    {
      "entropy": 5.665051794052124,
      "epoch": 2.4042278682016676,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004424491945777245,
      "loss": 5.1296,
      "mean_token_accuracy": 0.18244045823812485,
      "num_tokens": 53078129.0,
      "step": 24225
    },
    {
      "entropy": 5.594828033447266,
      "epoch": 2.404724096863835,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044242552652688885,
      "loss": 4.9728,
      "mean_token_accuracy": 0.1837316021323204,
      "num_tokens": 53088919.0,
      "step": 24230
    },
    {
      "entropy": 5.598760986328125,
      "epoch": 2.4052203255260025,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004424018543242608,
      "loss": 5.0457,
      "mean_token_accuracy": 0.1878911539912224,
      "num_tokens": 53099716.0,
      "step": 24235
    },
    {
      "entropy": 5.551165580749512,
      "epoch": 2.40571655418817,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004423781779704275,
      "loss": 4.9623,
      "mean_token_accuracy": 0.1876218721270561,
      "num_tokens": 53108798.0,
      "step": 24240
    },
    {
      "entropy": 5.585519742965698,
      "epoch": 2.4062127828503375,
      "grad_norm": 1.0,
      "learning_rate": 0.0004423544974659761,
      "loss": 5.0721,
      "mean_token_accuracy": 0.17753576040267943,
      "num_tokens": 53120655.0,
      "step": 24245
    },
    {
      "entropy": 5.580986881256104,
      "epoch": 2.406709011512505,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044233081281149373,
      "loss": 4.9227,
      "mean_token_accuracy": 0.1847422182559967,
      "num_tokens": 53131760.0,
      "step": 24250
    },
    {
      "entropy": 5.61481237411499,
      "epoch": 2.4072052401746724,
      "grad_norm": 1.125,
      "learning_rate": 0.00044230712400756774,
      "loss": 4.9811,
      "mean_token_accuracy": 0.18387256264686586,
      "num_tokens": 53142347.0,
      "step": 24255
    },
    {
      "entropy": 5.578934526443481,
      "epoch": 2.40770146883684,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044228343105478564,
      "loss": 5.0327,
      "mean_token_accuracy": 0.18056386858224868,
      "num_tokens": 53153678.0,
      "step": 24260
    },
    {
      "entropy": 5.563760137557983,
      "epoch": 2.4081976974990074,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004422597339537349,
      "loss": 5.0587,
      "mean_token_accuracy": 0.18140977025032043,
      "num_tokens": 53165089.0,
      "step": 24265
    },
    {
      "entropy": 5.621790647506714,
      "epoch": 2.408693926161175,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004422360327050032,
      "loss": 5.0118,
      "mean_token_accuracy": 0.17450100630521775,
      "num_tokens": 53175614.0,
      "step": 24270
    },
    {
      "entropy": 5.638982915878296,
      "epoch": 2.4091901548233428,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00044221232730917823,
      "loss": 4.9984,
      "mean_token_accuracy": 0.1831998720765114,
      "num_tokens": 53186306.0,
      "step": 24275
    },
    {
      "entropy": 5.6010290622711185,
      "epoch": 2.4096863834855102,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00044218861776684776,
      "loss": 5.0553,
      "mean_token_accuracy": 0.18409262001514434,
      "num_tokens": 53197730.0,
      "step": 24280
    },
    {
      "entropy": 5.555887603759766,
      "epoch": 2.4101826121476777,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044216490407859995,
      "loss": 4.9832,
      "mean_token_accuracy": 0.18099249750375748,
      "num_tokens": 53208486.0,
      "step": 24285
    },
    {
      "entropy": 5.627083492279053,
      "epoch": 2.410678840809845,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044214118624502267,
      "loss": 5.0407,
      "mean_token_accuracy": 0.18060191869735717,
      "num_tokens": 53220045.0,
      "step": 24290
    },
    {
      "entropy": 5.6182936668396,
      "epoch": 2.4111750694720127,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004421174642667042,
      "loss": 5.123,
      "mean_token_accuracy": 0.17749010175466537,
      "num_tokens": 53231385.0,
      "step": 24295
    },
    {
      "entropy": 5.660434150695801,
      "epoch": 2.41167129813418,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044209373814423265,
      "loss": 5.098,
      "mean_token_accuracy": 0.17906394004821777,
      "num_tokens": 53243038.0,
      "step": 24300
    },
    {
      "entropy": 5.642292213439942,
      "epoch": 2.4121675267963476,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004420700078781965,
      "loss": 5.0675,
      "mean_token_accuracy": 0.18085843771696092,
      "num_tokens": 53252959.0,
      "step": 24305
    },
    {
      "entropy": 5.526022338867188,
      "epoch": 2.4126637554585155,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00044204627346918407,
      "loss": 5.0002,
      "mean_token_accuracy": 0.18352188020944596,
      "num_tokens": 53263542.0,
      "step": 24310
    },
    {
      "entropy": 5.573980045318604,
      "epoch": 2.413159984120683,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044202253491778404,
      "loss": 4.9914,
      "mean_token_accuracy": 0.18750749677419662,
      "num_tokens": 53273511.0,
      "step": 24315
    },
    {
      "entropy": 5.600523662567139,
      "epoch": 2.4136562127828505,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004419987922245852,
      "loss": 5.1105,
      "mean_token_accuracy": 0.18141085803508758,
      "num_tokens": 53286173.0,
      "step": 24320
    },
    {
      "entropy": 5.596400547027588,
      "epoch": 2.414152441445018,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000441975045390176,
      "loss": 5.0013,
      "mean_token_accuracy": 0.18684254139661788,
      "num_tokens": 53296784.0,
      "step": 24325
    },
    {
      "entropy": 5.644485902786255,
      "epoch": 2.4146486701071854,
      "grad_norm": 0.96875,
      "learning_rate": 0.00044195129441514547,
      "loss": 5.0283,
      "mean_token_accuracy": 0.18385814875364304,
      "num_tokens": 53308044.0,
      "step": 24330
    },
    {
      "entropy": 5.564671277999878,
      "epoch": 2.415144898769353,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044192753930008267,
      "loss": 4.992,
      "mean_token_accuracy": 0.19118549823760986,
      "num_tokens": 53319161.0,
      "step": 24335
    },
    {
      "entropy": 5.554568147659301,
      "epoch": 2.4156411274315204,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004419037800455765,
      "loss": 4.964,
      "mean_token_accuracy": 0.1920511931180954,
      "num_tokens": 53329766.0,
      "step": 24340
    },
    {
      "entropy": 5.615224599838257,
      "epoch": 2.416137356093688,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004418800166522163,
      "loss": 5.0686,
      "mean_token_accuracy": 0.18233283758163452,
      "num_tokens": 53340329.0,
      "step": 24345
    },
    {
      "entropy": 5.650993394851684,
      "epoch": 2.4166335847558553,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044185624912059125,
      "loss": 5.061,
      "mean_token_accuracy": 0.1823356881737709,
      "num_tokens": 53352592.0,
      "step": 24350
    },
    {
      "entropy": 5.6580921649932865,
      "epoch": 2.417129813418023,
      "grad_norm": 1.0,
      "learning_rate": 0.0004418324774512908,
      "loss": 5.0138,
      "mean_token_accuracy": 0.17795349806547164,
      "num_tokens": 53362877.0,
      "step": 24355
    },
    {
      "entropy": 5.590397644042969,
      "epoch": 2.4176260420801907,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004418087016449043,
      "loss": 5.0322,
      "mean_token_accuracy": 0.18421638309955596,
      "num_tokens": 53374612.0,
      "step": 24360
    },
    {
      "entropy": 5.622679901123047,
      "epoch": 2.418122270742358,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044178492170202153,
      "loss": 5.0696,
      "mean_token_accuracy": 0.17929739356040955,
      "num_tokens": 53385129.0,
      "step": 24365
    },
    {
      "entropy": 5.678403663635254,
      "epoch": 2.4186184994045257,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044176113762323213,
      "loss": 5.1357,
      "mean_token_accuracy": 0.1765660509467125,
      "num_tokens": 53395403.0,
      "step": 24370
    },
    {
      "entropy": 5.653208684921265,
      "epoch": 2.419114728066693,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044173734940912586,
      "loss": 5.1065,
      "mean_token_accuracy": 0.17346630990505219,
      "num_tokens": 53406684.0,
      "step": 24375
    },
    {
      "entropy": 5.58662428855896,
      "epoch": 2.4196109567288606,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044171355706029247,
      "loss": 4.9754,
      "mean_token_accuracy": 0.18618477433919906,
      "num_tokens": 53417256.0,
      "step": 24380
    },
    {
      "entropy": 5.5342942237854,
      "epoch": 2.420107185391028,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044168976057732225,
      "loss": 4.9856,
      "mean_token_accuracy": 0.17997187227010727,
      "num_tokens": 53428305.0,
      "step": 24385
    },
    {
      "entropy": 5.534274864196777,
      "epoch": 2.4206034140531956,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004416659599608051,
      "loss": 5.014,
      "mean_token_accuracy": 0.17817320972681044,
      "num_tokens": 53439301.0,
      "step": 24390
    },
    {
      "entropy": 5.663099193572998,
      "epoch": 2.4210996427153635,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004416421552113312,
      "loss": 5.1344,
      "mean_token_accuracy": 0.17113237082958221,
      "num_tokens": 53450478.0,
      "step": 24395
    },
    {
      "entropy": 5.662753009796143,
      "epoch": 2.421595871377531,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000441618346329491,
      "loss": 5.0564,
      "mean_token_accuracy": 0.17686022371053695,
      "num_tokens": 53461293.0,
      "step": 24400
    },
    {
      "entropy": 5.5635889053344725,
      "epoch": 2.4220921000396984,
      "grad_norm": 1.0,
      "learning_rate": 0.00044159453331587493,
      "loss": 5.0235,
      "mean_token_accuracy": 0.18409996032714843,
      "num_tokens": 53472014.0,
      "step": 24405
    },
    {
      "entropy": 5.596433210372925,
      "epoch": 2.422588328701866,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044157071617107325,
      "loss": 5.0718,
      "mean_token_accuracy": 0.1815323367714882,
      "num_tokens": 53483010.0,
      "step": 24410
    },
    {
      "entropy": 5.636553716659546,
      "epoch": 2.4230845573640334,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00044154689489567684,
      "loss": 5.0602,
      "mean_token_accuracy": 0.17905300259590148,
      "num_tokens": 53494151.0,
      "step": 24415
    },
    {
      "entropy": 5.606340599060059,
      "epoch": 2.423580786026201,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044152306949027625,
      "loss": 5.0449,
      "mean_token_accuracy": 0.1821301445364952,
      "num_tokens": 53504462.0,
      "step": 24420
    },
    {
      "entropy": 5.6185462474823,
      "epoch": 2.4240770146883683,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044149923995546244,
      "loss": 5.0238,
      "mean_token_accuracy": 0.18389852344989777,
      "num_tokens": 53514896.0,
      "step": 24425
    },
    {
      "entropy": 5.57132921218872,
      "epoch": 2.424573243350536,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004414754062918262,
      "loss": 4.9888,
      "mean_token_accuracy": 0.1845644474029541,
      "num_tokens": 53524656.0,
      "step": 24430
    },
    {
      "entropy": 5.6260094165802,
      "epoch": 2.4250694720127033,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004414515684999587,
      "loss": 5.0488,
      "mean_token_accuracy": 0.18247200548648834,
      "num_tokens": 53534896.0,
      "step": 24435
    },
    {
      "entropy": 5.573774957656861,
      "epoch": 2.4255657006748708,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004414277265804508,
      "loss": 4.9952,
      "mean_token_accuracy": 0.18591091483831407,
      "num_tokens": 53546108.0,
      "step": 24440
    },
    {
      "entropy": 5.553739356994629,
      "epoch": 2.4260619293370387,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000441403880533894,
      "loss": 5.0153,
      "mean_token_accuracy": 0.17458850592374803,
      "num_tokens": 53557325.0,
      "step": 24445
    },
    {
      "entropy": 5.503540706634522,
      "epoch": 2.426558157999206,
      "grad_norm": 0.953125,
      "learning_rate": 0.00044138003036087956,
      "loss": 4.9385,
      "mean_token_accuracy": 0.18516125231981279,
      "num_tokens": 53568737.0,
      "step": 24450
    },
    {
      "entropy": 5.624419641494751,
      "epoch": 2.4270543866613736,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004413561760619988,
      "loss": 5.0455,
      "mean_token_accuracy": 0.17774686962366104,
      "num_tokens": 53579312.0,
      "step": 24455
    },
    {
      "entropy": 5.572875213623047,
      "epoch": 2.427550615323541,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004413323176378434,
      "loss": 4.9585,
      "mean_token_accuracy": 0.18648573458194734,
      "num_tokens": 53590476.0,
      "step": 24460
    },
    {
      "entropy": 5.618983364105224,
      "epoch": 2.4280468439857086,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004413084550890049,
      "loss": 5.088,
      "mean_token_accuracy": 0.17718201279640197,
      "num_tokens": 53601284.0,
      "step": 24465
    },
    {
      "entropy": 5.591473340988159,
      "epoch": 2.428543072647876,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004412845884160751,
      "loss": 5.068,
      "mean_token_accuracy": 0.17332033962011337,
      "num_tokens": 53612590.0,
      "step": 24470
    },
    {
      "entropy": 5.622741746902466,
      "epoch": 2.4290393013100435,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004412607176196458,
      "loss": 5.0811,
      "mean_token_accuracy": 0.17733932435512542,
      "num_tokens": 53623049.0,
      "step": 24475
    },
    {
      "entropy": 5.588508176803589,
      "epoch": 2.429535529972211,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044123684270030907,
      "loss": 5.0055,
      "mean_token_accuracy": 0.18623065799474717,
      "num_tokens": 53633961.0,
      "step": 24480
    },
    {
      "entropy": 5.584644412994384,
      "epoch": 2.430031758634379,
      "grad_norm": 0.96875,
      "learning_rate": 0.00044121296365865674,
      "loss": 5.0508,
      "mean_token_accuracy": 0.17703040540218354,
      "num_tokens": 53644634.0,
      "step": 24485
    },
    {
      "entropy": 5.588081645965576,
      "epoch": 2.4305279872965464,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044118908049528105,
      "loss": 4.9873,
      "mean_token_accuracy": 0.1846154287457466,
      "num_tokens": 53655982.0,
      "step": 24490
    },
    {
      "entropy": 5.669098424911499,
      "epoch": 2.431024215958714,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00044116519321077435,
      "loss": 5.0983,
      "mean_token_accuracy": 0.172780904173851,
      "num_tokens": 53667367.0,
      "step": 24495
    },
    {
      "entropy": 5.576200532913208,
      "epoch": 2.4315204446208813,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044114130180572883,
      "loss": 4.984,
      "mean_token_accuracy": 0.18253543078899384,
      "num_tokens": 53677191.0,
      "step": 24500
    },
    {
      "entropy": 5.590221881866455,
      "epoch": 2.432016673283049,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004411174062807371,
      "loss": 5.0635,
      "mean_token_accuracy": 0.17831269949674605,
      "num_tokens": 53688636.0,
      "step": 24505
    },
    {
      "entropy": 5.632370233535767,
      "epoch": 2.4325129019452163,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004410935066363916,
      "loss": 5.0011,
      "mean_token_accuracy": 0.18055623322725295,
      "num_tokens": 53699588.0,
      "step": 24510
    },
    {
      "entropy": 5.622808980941772,
      "epoch": 2.4330091306073838,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044106960287328503,
      "loss": 5.126,
      "mean_token_accuracy": 0.17195386290550232,
      "num_tokens": 53710427.0,
      "step": 24515
    },
    {
      "entropy": 5.636325168609619,
      "epoch": 2.4335053592695512,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004410456949920102,
      "loss": 5.0867,
      "mean_token_accuracy": 0.17121079713106155,
      "num_tokens": 53722319.0,
      "step": 24520
    },
    {
      "entropy": 5.610949611663818,
      "epoch": 2.4340015879317187,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044102178299315994,
      "loss": 5.0859,
      "mean_token_accuracy": 0.17673904001712798,
      "num_tokens": 53734059.0,
      "step": 24525
    },
    {
      "entropy": 5.61175389289856,
      "epoch": 2.4344978165938866,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044099786687732713,
      "loss": 5.0573,
      "mean_token_accuracy": 0.18032858818769454,
      "num_tokens": 53745795.0,
      "step": 24530
    },
    {
      "entropy": 5.626543760299683,
      "epoch": 2.434994045256054,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004409739466451049,
      "loss": 5.0742,
      "mean_token_accuracy": 0.17744001895189285,
      "num_tokens": 53756781.0,
      "step": 24535
    },
    {
      "entropy": 5.672981262207031,
      "epoch": 2.4354902739182216,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00044095002229708655,
      "loss": 5.1188,
      "mean_token_accuracy": 0.17729998379945755,
      "num_tokens": 53769725.0,
      "step": 24540
    },
    {
      "entropy": 5.58683180809021,
      "epoch": 2.435986502580389,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044092609383386517,
      "loss": 4.9952,
      "mean_token_accuracy": 0.19077104032039643,
      "num_tokens": 53781217.0,
      "step": 24545
    },
    {
      "entropy": 5.645798301696777,
      "epoch": 2.4364827312425565,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044090216125603426,
      "loss": 5.1015,
      "mean_token_accuracy": 0.18322383761405944,
      "num_tokens": 53791478.0,
      "step": 24550
    },
    {
      "entropy": 5.569611215591431,
      "epoch": 2.436978959904724,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004408782245641872,
      "loss": 4.9879,
      "mean_token_accuracy": 0.19129831045866014,
      "num_tokens": 53802844.0,
      "step": 24555
    },
    {
      "entropy": 5.624389886856079,
      "epoch": 2.4374751885668915,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004408542837589176,
      "loss": 5.0493,
      "mean_token_accuracy": 0.1850482165813446,
      "num_tokens": 53812938.0,
      "step": 24560
    },
    {
      "entropy": 5.636423254013062,
      "epoch": 2.437971417229059,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044083033884081914,
      "loss": 5.001,
      "mean_token_accuracy": 0.18272792994976045,
      "num_tokens": 53824105.0,
      "step": 24565
    },
    {
      "entropy": 5.683406114578247,
      "epoch": 2.438467645891227,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004408063898104856,
      "loss": 5.1537,
      "mean_token_accuracy": 0.17532306164503098,
      "num_tokens": 53834956.0,
      "step": 24570
    },
    {
      "entropy": 5.61568512916565,
      "epoch": 2.4389638745533944,
      "grad_norm": 0.9375,
      "learning_rate": 0.00044078243666851094,
      "loss": 5.0556,
      "mean_token_accuracy": 0.18020149767398835,
      "num_tokens": 53846257.0,
      "step": 24575
    },
    {
      "entropy": 5.6047505855560305,
      "epoch": 2.439460103215562,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044075847941548905,
      "loss": 4.9688,
      "mean_token_accuracy": 0.1835252344608307,
      "num_tokens": 53857107.0,
      "step": 24580
    },
    {
      "entropy": 5.604504489898682,
      "epoch": 2.4399563318777293,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00044073451805201396,
      "loss": 5.0381,
      "mean_token_accuracy": 0.18071888983249665,
      "num_tokens": 53867779.0,
      "step": 24585
    },
    {
      "entropy": 5.6059540748596195,
      "epoch": 2.440452560539897,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044071055257868,
      "loss": 5.016,
      "mean_token_accuracy": 0.17710161805152894,
      "num_tokens": 53878661.0,
      "step": 24590
    },
    {
      "entropy": 5.692973041534424,
      "epoch": 2.4409487892020643,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00044068658299608145,
      "loss": 5.1116,
      "mean_token_accuracy": 0.17986399978399276,
      "num_tokens": 53891300.0,
      "step": 24595
    },
    {
      "entropy": 5.633247566223145,
      "epoch": 2.4414450178642317,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044066260930481264,
      "loss": 4.9763,
      "mean_token_accuracy": 0.18791133910417557,
      "num_tokens": 53901086.0,
      "step": 24600
    },
    {
      "entropy": 5.65188627243042,
      "epoch": 2.441941246526399,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004406386315054681,
      "loss": 5.0846,
      "mean_token_accuracy": 0.17226369231939315,
      "num_tokens": 53912599.0,
      "step": 24605
    },
    {
      "entropy": 5.553772592544556,
      "epoch": 2.4424374751885667,
      "grad_norm": 1.125,
      "learning_rate": 0.00044061464959864243,
      "loss": 5.0548,
      "mean_token_accuracy": 0.18020074665546418,
      "num_tokens": 53924293.0,
      "step": 24610
    },
    {
      "entropy": 5.613297367095948,
      "epoch": 2.4429337038507346,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004405906635849302,
      "loss": 5.1079,
      "mean_token_accuracy": 0.17676304280757904,
      "num_tokens": 53935074.0,
      "step": 24615
    },
    {
      "entropy": 5.6033872127532955,
      "epoch": 2.443429932512902,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004405666734649265,
      "loss": 5.0481,
      "mean_token_accuracy": 0.18174927085638046,
      "num_tokens": 53946770.0,
      "step": 24620
    },
    {
      "entropy": 5.6157489776611325,
      "epoch": 2.4439261611750696,
      "grad_norm": 1.125,
      "learning_rate": 0.00044054267923922594,
      "loss": 5.0199,
      "mean_token_accuracy": 0.17674887627363206,
      "num_tokens": 53957077.0,
      "step": 24625
    },
    {
      "entropy": 5.619725179672241,
      "epoch": 2.444422389837237,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044051868090842376,
      "loss": 5.0498,
      "mean_token_accuracy": 0.1832478180527687,
      "num_tokens": 53968455.0,
      "step": 24630
    },
    {
      "entropy": 5.523649454116821,
      "epoch": 2.4449186184994045,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044049467847311486,
      "loss": 5.0045,
      "mean_token_accuracy": 0.18154477179050446,
      "num_tokens": 53980429.0,
      "step": 24635
    },
    {
      "entropy": 5.585433149337769,
      "epoch": 2.445414847161572,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004404706719338946,
      "loss": 5.0162,
      "mean_token_accuracy": 0.18648228347301482,
      "num_tokens": 53990937.0,
      "step": 24640
    },
    {
      "entropy": 5.591693687438965,
      "epoch": 2.4459110758237395,
      "grad_norm": 0.90625,
      "learning_rate": 0.00044044666129135825,
      "loss": 5.0024,
      "mean_token_accuracy": 0.18123873621225356,
      "num_tokens": 54003526.0,
      "step": 24645
    },
    {
      "entropy": 5.572509241104126,
      "epoch": 2.446407304485907,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044042264654610115,
      "loss": 4.9739,
      "mean_token_accuracy": 0.19311117082834245,
      "num_tokens": 54013875.0,
      "step": 24650
    },
    {
      "entropy": 5.643759059906006,
      "epoch": 2.446903533148075,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044039862769871884,
      "loss": 5.1454,
      "mean_token_accuracy": 0.1739455610513687,
      "num_tokens": 54024103.0,
      "step": 24655
    },
    {
      "entropy": 5.626368999481201,
      "epoch": 2.4473997618102423,
      "grad_norm": 0.96875,
      "learning_rate": 0.00044037460474980703,
      "loss": 5.1224,
      "mean_token_accuracy": 0.17327839136123657,
      "num_tokens": 54035417.0,
      "step": 24660
    },
    {
      "entropy": 5.6699607372283936,
      "epoch": 2.44789599047241,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004403505776999614,
      "loss": 5.1116,
      "mean_token_accuracy": 0.17129187136888505,
      "num_tokens": 54045180.0,
      "step": 24665
    },
    {
      "entropy": 5.630414628982544,
      "epoch": 2.4483922191345773,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044032654654977766,
      "loss": 4.9993,
      "mean_token_accuracy": 0.1907757818698883,
      "num_tokens": 54056493.0,
      "step": 24670
    },
    {
      "entropy": 5.567400932312012,
      "epoch": 2.4488884477967447,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004403025112998519,
      "loss": 5.0273,
      "mean_token_accuracy": 0.18152531385421752,
      "num_tokens": 54067445.0,
      "step": 24675
    },
    {
      "entropy": 5.601558494567871,
      "epoch": 2.4493846764589122,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004402784719507799,
      "loss": 5.0159,
      "mean_token_accuracy": 0.18429240882396697,
      "num_tokens": 54076937.0,
      "step": 24680
    },
    {
      "entropy": 5.649069356918335,
      "epoch": 2.4498809051210797,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004402544285031581,
      "loss": 5.0588,
      "mean_token_accuracy": 0.1875091478228569,
      "num_tokens": 54088014.0,
      "step": 24685
    },
    {
      "entropy": 5.655137014389038,
      "epoch": 2.450377133783247,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004402303809575825,
      "loss": 5.0047,
      "mean_token_accuracy": 0.18605092465877532,
      "num_tokens": 54098984.0,
      "step": 24690
    },
    {
      "entropy": 5.651613235473633,
      "epoch": 2.4508733624454146,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004402063293146495,
      "loss": 5.1655,
      "mean_token_accuracy": 0.17314006835222245,
      "num_tokens": 54110490.0,
      "step": 24695
    },
    {
      "entropy": 5.62675724029541,
      "epoch": 2.451369591107582,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004401822735749555,
      "loss": 5.0524,
      "mean_token_accuracy": 0.17702329605817796,
      "num_tokens": 54120785.0,
      "step": 24700
    },
    {
      "entropy": 5.6131676197052,
      "epoch": 2.45186581976975,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044015821373909707,
      "loss": 4.9891,
      "mean_token_accuracy": 0.1817745327949524,
      "num_tokens": 54131339.0,
      "step": 24705
    },
    {
      "entropy": 5.608229112625122,
      "epoch": 2.4523620484319175,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044013414980767084,
      "loss": 5.072,
      "mean_token_accuracy": 0.17239224463701247,
      "num_tokens": 54142270.0,
      "step": 24710
    },
    {
      "entropy": 5.603488540649414,
      "epoch": 2.452858277094085,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004401100817812735,
      "loss": 5.0313,
      "mean_token_accuracy": 0.1825444683432579,
      "num_tokens": 54153045.0,
      "step": 24715
    },
    {
      "entropy": 5.593999004364013,
      "epoch": 2.4533545057562525,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044008600966050196,
      "loss": 4.966,
      "mean_token_accuracy": 0.18523489087820053,
      "num_tokens": 54164069.0,
      "step": 24720
    },
    {
      "entropy": 5.632596349716186,
      "epoch": 2.45385073441842,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004400619334459531,
      "loss": 5.0592,
      "mean_token_accuracy": 0.17960473895072937,
      "num_tokens": 54176902.0,
      "step": 24725
    },
    {
      "entropy": 5.560891342163086,
      "epoch": 2.4543469630805874,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044003785313822386,
      "loss": 5.013,
      "mean_token_accuracy": 0.18618395328521728,
      "num_tokens": 54188515.0,
      "step": 24730
    },
    {
      "entropy": 5.633498668670654,
      "epoch": 2.454843191742755,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044001376873791166,
      "loss": 5.0737,
      "mean_token_accuracy": 0.1818646028637886,
      "num_tokens": 54199077.0,
      "step": 24735
    },
    {
      "entropy": 5.698347854614258,
      "epoch": 2.455339420404923,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043998968024561337,
      "loss": 5.1031,
      "mean_token_accuracy": 0.18825859427452088,
      "num_tokens": 54210216.0,
      "step": 24740
    },
    {
      "entropy": 5.601244640350342,
      "epoch": 2.4558356490670903,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004399655876619267,
      "loss": 5.0269,
      "mean_token_accuracy": 0.18205155730247496,
      "num_tokens": 54221143.0,
      "step": 24745
    },
    {
      "entropy": 5.625874137878418,
      "epoch": 2.4563318777292578,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043994149098744884,
      "loss": 5.0814,
      "mean_token_accuracy": 0.18046271055936813,
      "num_tokens": 54233070.0,
      "step": 24750
    },
    {
      "entropy": 5.607621955871582,
      "epoch": 2.4568281063914252,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004399173902227775,
      "loss": 5.0268,
      "mean_token_accuracy": 0.18263722360134124,
      "num_tokens": 54243793.0,
      "step": 24755
    },
    {
      "entropy": 5.54004168510437,
      "epoch": 2.4573243350535927,
      "grad_norm": 0.9375,
      "learning_rate": 0.00043989328536851014,
      "loss": 5.02,
      "mean_token_accuracy": 0.18297266513109206,
      "num_tokens": 54256983.0,
      "step": 24760
    },
    {
      "entropy": 5.592242097854614,
      "epoch": 2.45782056371576,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043986917642524465,
      "loss": 5.0157,
      "mean_token_accuracy": 0.18150011152029039,
      "num_tokens": 54268789.0,
      "step": 24765
    },
    {
      "entropy": 5.7046884059906,
      "epoch": 2.4583167923779277,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00043984506339357887,
      "loss": 5.0451,
      "mean_token_accuracy": 0.17431846261024475,
      "num_tokens": 54279517.0,
      "step": 24770
    },
    {
      "entropy": 5.6320991039276125,
      "epoch": 2.458813021040095,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004398209462741107,
      "loss": 5.0334,
      "mean_token_accuracy": 0.178621344268322,
      "num_tokens": 54290865.0,
      "step": 24775
    },
    {
      "entropy": 5.72394962310791,
      "epoch": 2.4593092497022626,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004397968250674383,
      "loss": 5.0949,
      "mean_token_accuracy": 0.1876548558473587,
      "num_tokens": 54302658.0,
      "step": 24780
    },
    {
      "entropy": 5.61916298866272,
      "epoch": 2.45980547836443,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004397726997741597,
      "loss": 5.0614,
      "mean_token_accuracy": 0.18128456622362138,
      "num_tokens": 54314000.0,
      "step": 24785
    },
    {
      "entropy": 5.566324043273926,
      "epoch": 2.460301707026598,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004397485703948732,
      "loss": 5.0267,
      "mean_token_accuracy": 0.18582546412944795,
      "num_tokens": 54325952.0,
      "step": 24790
    },
    {
      "entropy": 5.653996706008911,
      "epoch": 2.4607979356887655,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004397244369301771,
      "loss": 5.1178,
      "mean_token_accuracy": 0.17647729218006133,
      "num_tokens": 54338582.0,
      "step": 24795
    },
    {
      "entropy": 5.629953670501709,
      "epoch": 2.461294164350933,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043970029938066995,
      "loss": 5.0882,
      "mean_token_accuracy": 0.17371853590011596,
      "num_tokens": 54349513.0,
      "step": 24800
    },
    {
      "entropy": 5.631373262405395,
      "epoch": 2.4617903930131004,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004396761577469503,
      "loss": 5.0113,
      "mean_token_accuracy": 0.1783672273159027,
      "num_tokens": 54360264.0,
      "step": 24805
    },
    {
      "entropy": 5.618650150299072,
      "epoch": 2.462286621675268,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00043965201202961673,
      "loss": 5.0739,
      "mean_token_accuracy": 0.17500193864107133,
      "num_tokens": 54371379.0,
      "step": 24810
    },
    {
      "entropy": 5.616022348403931,
      "epoch": 2.4627828503374354,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043962786222926807,
      "loss": 5.0535,
      "mean_token_accuracy": 0.18545264452695848,
      "num_tokens": 54382844.0,
      "step": 24815
    },
    {
      "entropy": 5.604498243331909,
      "epoch": 2.463279078999603,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043960370834650325,
      "loss": 5.0212,
      "mean_token_accuracy": 0.18380396217107772,
      "num_tokens": 54392807.0,
      "step": 24820
    },
    {
      "entropy": 5.571725273132325,
      "epoch": 2.4637753076617708,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004395795503819211,
      "loss": 5.0399,
      "mean_token_accuracy": 0.1810228317975998,
      "num_tokens": 54404060.0,
      "step": 24825
    },
    {
      "entropy": 5.583786392211914,
      "epoch": 2.4642715363239382,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004395553883361207,
      "loss": 4.9584,
      "mean_token_accuracy": 0.1844467431306839,
      "num_tokens": 54414636.0,
      "step": 24830
    },
    {
      "entropy": 5.559342384338379,
      "epoch": 2.4647677649861057,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043953122220970127,
      "loss": 5.0021,
      "mean_token_accuracy": 0.18629788905382155,
      "num_tokens": 54426185.0,
      "step": 24835
    },
    {
      "entropy": 5.579334449768067,
      "epoch": 2.465263993648273,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043950705200326205,
      "loss": 5.0075,
      "mean_token_accuracy": 0.1769101932644844,
      "num_tokens": 54436468.0,
      "step": 24840
    },
    {
      "entropy": 5.624745607376099,
      "epoch": 2.4657602223104407,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004394828777174025,
      "loss": 5.0233,
      "mean_token_accuracy": 0.18556296825408936,
      "num_tokens": 54446228.0,
      "step": 24845
    },
    {
      "entropy": 5.634768056869507,
      "epoch": 2.466256450972608,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043945869935272194,
      "loss": 5.0857,
      "mean_token_accuracy": 0.17250876128673553,
      "num_tokens": 54456704.0,
      "step": 24850
    },
    {
      "entropy": 5.658349275588989,
      "epoch": 2.4667526796347756,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043943451690982006,
      "loss": 5.03,
      "mean_token_accuracy": 0.18911212980747222,
      "num_tokens": 54467780.0,
      "step": 24855
    },
    {
      "entropy": 5.7213929176330565,
      "epoch": 2.467248908296943,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004394103303892964,
      "loss": 5.1369,
      "mean_token_accuracy": 0.17702567130327224,
      "num_tokens": 54478428.0,
      "step": 24860
    },
    {
      "entropy": 5.554982995986938,
      "epoch": 2.4677451369591106,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043938613979175087,
      "loss": 4.9494,
      "mean_token_accuracy": 0.18643166720867158,
      "num_tokens": 54489544.0,
      "step": 24865
    },
    {
      "entropy": 5.582428026199341,
      "epoch": 2.468241365621278,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004393619451177834,
      "loss": 5.0202,
      "mean_token_accuracy": 0.18617129921913148,
      "num_tokens": 54501438.0,
      "step": 24870
    },
    {
      "entropy": 5.552873754501343,
      "epoch": 2.468737594283446,
      "grad_norm": 1.125,
      "learning_rate": 0.00043933774636799365,
      "loss": 4.9519,
      "mean_token_accuracy": 0.183061483502388,
      "num_tokens": 54511746.0,
      "step": 24875
    },
    {
      "entropy": 5.610507678985596,
      "epoch": 2.4692338229456134,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043931354354298204,
      "loss": 5.0572,
      "mean_token_accuracy": 0.1831992819905281,
      "num_tokens": 54523141.0,
      "step": 24880
    },
    {
      "entropy": 5.742952346801758,
      "epoch": 2.469730051607781,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004392893366433486,
      "loss": 5.1854,
      "mean_token_accuracy": 0.1755501225590706,
      "num_tokens": 54533275.0,
      "step": 24885
    },
    {
      "entropy": 5.697040557861328,
      "epoch": 2.4702262802699484,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00043926512566969366,
      "loss": 5.0237,
      "mean_token_accuracy": 0.17898010313510895,
      "num_tokens": 54545309.0,
      "step": 24890
    },
    {
      "entropy": 5.581230354309082,
      "epoch": 2.470722508932116,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043924091062261743,
      "loss": 4.9986,
      "mean_token_accuracy": 0.18815975040197372,
      "num_tokens": 54554905.0,
      "step": 24895
    },
    {
      "entropy": 5.599094438552856,
      "epoch": 2.4712187375942833,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004392166915027206,
      "loss": 5.0385,
      "mean_token_accuracy": 0.1779889464378357,
      "num_tokens": 54566050.0,
      "step": 24900
    },
    {
      "entropy": 5.634348678588867,
      "epoch": 2.471714966256451,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004391924683106037,
      "loss": 5.1019,
      "mean_token_accuracy": 0.17665312737226485,
      "num_tokens": 54577791.0,
      "step": 24905
    },
    {
      "entropy": 5.65377140045166,
      "epoch": 2.4722111949186187,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004391682410468674,
      "loss": 5.0348,
      "mean_token_accuracy": 0.1845640704035759,
      "num_tokens": 54589183.0,
      "step": 24910
    },
    {
      "entropy": 5.599700546264648,
      "epoch": 2.472707423580786,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00043914400971211255,
      "loss": 5.0331,
      "mean_token_accuracy": 0.17670703828334808,
      "num_tokens": 54600776.0,
      "step": 24915
    },
    {
      "entropy": 5.576289081573487,
      "epoch": 2.4732036522429537,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004391197743069398,
      "loss": 4.9946,
      "mean_token_accuracy": 0.18819257318973542,
      "num_tokens": 54611167.0,
      "step": 24920
    },
    {
      "entropy": 5.60039267539978,
      "epoch": 2.473699880905121,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004390955348319503,
      "loss": 5.0356,
      "mean_token_accuracy": 0.18310957551002502,
      "num_tokens": 54622811.0,
      "step": 24925
    },
    {
      "entropy": 5.596719884872437,
      "epoch": 2.4741961095672886,
      "grad_norm": 1.0,
      "learning_rate": 0.00043907129128774525,
      "loss": 5.0371,
      "mean_token_accuracy": 0.18296483606100084,
      "num_tokens": 54635109.0,
      "step": 24930
    },
    {
      "entropy": 5.664292430877685,
      "epoch": 2.474692338229456,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043904704367492563,
      "loss": 5.0759,
      "mean_token_accuracy": 0.18220090121030807,
      "num_tokens": 54644260.0,
      "step": 24935
    },
    {
      "entropy": 5.602024221420288,
      "epoch": 2.4751885668916236,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004390227919940928,
      "loss": 4.976,
      "mean_token_accuracy": 0.18378848880529403,
      "num_tokens": 54654433.0,
      "step": 24940
    },
    {
      "entropy": 5.619705772399902,
      "epoch": 2.475684795553791,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043899853624584825,
      "loss": 5.0958,
      "mean_token_accuracy": 0.175523242354393,
      "num_tokens": 54666582.0,
      "step": 24945
    },
    {
      "entropy": 5.587623548507691,
      "epoch": 2.4761810242159585,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043897427643079336,
      "loss": 4.9914,
      "mean_token_accuracy": 0.1809183731675148,
      "num_tokens": 54677247.0,
      "step": 24950
    },
    {
      "entropy": 5.611222267150879,
      "epoch": 2.476677252878126,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00043895001254952963,
      "loss": 5.062,
      "mean_token_accuracy": 0.1783335193991661,
      "num_tokens": 54688423.0,
      "step": 24955
    },
    {
      "entropy": 5.637845087051391,
      "epoch": 2.477173481540294,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000438925744602659,
      "loss": 5.0397,
      "mean_token_accuracy": 0.18561952710151672,
      "num_tokens": 54698517.0,
      "step": 24960
    },
    {
      "entropy": 5.673588705062866,
      "epoch": 2.4776697102024614,
      "grad_norm": 0.984375,
      "learning_rate": 0.00043890147259078314,
      "loss": 5.0913,
      "mean_token_accuracy": 0.17386962324380875,
      "num_tokens": 54711635.0,
      "step": 24965
    },
    {
      "entropy": 5.605737209320068,
      "epoch": 2.478165938864629,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043887719651450384,
      "loss": 5.0428,
      "mean_token_accuracy": 0.1861738681793213,
      "num_tokens": 54723051.0,
      "step": 24970
    },
    {
      "entropy": 5.711588430404663,
      "epoch": 2.4786621675267964,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004388529163744233,
      "loss": 5.0436,
      "mean_token_accuracy": 0.18400874882936477,
      "num_tokens": 54734191.0,
      "step": 24975
    },
    {
      "entropy": 5.625216341018676,
      "epoch": 2.479158396188964,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004388286321711435,
      "loss": 5.0172,
      "mean_token_accuracy": 0.18316898196935655,
      "num_tokens": 54745632.0,
      "step": 24980
    },
    {
      "entropy": 5.612901115417481,
      "epoch": 2.4796546248511313,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004388043439052665,
      "loss": 5.0624,
      "mean_token_accuracy": 0.17178695201873778,
      "num_tokens": 54755871.0,
      "step": 24985
    },
    {
      "entropy": 5.545900201797485,
      "epoch": 2.480150853513299,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043878005157739486,
      "loss": 4.9684,
      "mean_token_accuracy": 0.1879187449812889,
      "num_tokens": 54768036.0,
      "step": 24990
    },
    {
      "entropy": 5.596815347671509,
      "epoch": 2.4806470821754667,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004387557551881308,
      "loss": 4.9975,
      "mean_token_accuracy": 0.181893190741539,
      "num_tokens": 54778518.0,
      "step": 24995
    },
    {
      "entropy": 5.592529916763306,
      "epoch": 2.481143310837634,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000438731454738077,
      "loss": 4.9884,
      "mean_token_accuracy": 0.18764616847038268,
      "num_tokens": 54790786.0,
      "step": 25000
    },
    {
      "entropy": 5.576348972320557,
      "epoch": 2.4816395394998016,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043870715022783575,
      "loss": 4.9967,
      "mean_token_accuracy": 0.18101539760828017,
      "num_tokens": 54800063.0,
      "step": 25005
    },
    {
      "entropy": 5.618507671356201,
      "epoch": 2.482135768161969,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004386828416580101,
      "loss": 5.0691,
      "mean_token_accuracy": 0.17662140727043152,
      "num_tokens": 54810605.0,
      "step": 25010
    },
    {
      "entropy": 5.630087041854859,
      "epoch": 2.4826319968241366,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004386585290292026,
      "loss": 4.992,
      "mean_token_accuracy": 0.183344529569149,
      "num_tokens": 54820626.0,
      "step": 25015
    },
    {
      "entropy": 5.587936687469482,
      "epoch": 2.483128225486304,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004386342123420161,
      "loss": 4.9907,
      "mean_token_accuracy": 0.1797950029373169,
      "num_tokens": 54831395.0,
      "step": 25020
    },
    {
      "entropy": 5.537699890136719,
      "epoch": 2.4836244541484715,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043860989159705396,
      "loss": 4.9992,
      "mean_token_accuracy": 0.17836591452360154,
      "num_tokens": 54841238.0,
      "step": 25025
    },
    {
      "entropy": 5.646393775939941,
      "epoch": 2.484120682810639,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043858556679491896,
      "loss": 5.0635,
      "mean_token_accuracy": 0.18250280767679214,
      "num_tokens": 54851301.0,
      "step": 25030
    },
    {
      "entropy": 5.6216959953308105,
      "epoch": 2.4846169114728065,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043856123793621437,
      "loss": 5.0094,
      "mean_token_accuracy": 0.18658150136470794,
      "num_tokens": 54863135.0,
      "step": 25035
    },
    {
      "entropy": 5.613266944885254,
      "epoch": 2.485113140134974,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043853690502154355,
      "loss": 5.0685,
      "mean_token_accuracy": 0.17698876857757567,
      "num_tokens": 54874780.0,
      "step": 25040
    },
    {
      "entropy": 5.658929920196533,
      "epoch": 2.485609368797142,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004385125680515099,
      "loss": 5.0706,
      "mean_token_accuracy": 0.18166109770536423,
      "num_tokens": 54885237.0,
      "step": 25045
    },
    {
      "entropy": 5.65393557548523,
      "epoch": 2.4861055974593094,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004384882270267168,
      "loss": 5.0631,
      "mean_token_accuracy": 0.1794355422258377,
      "num_tokens": 54896801.0,
      "step": 25050
    },
    {
      "entropy": 5.5936578750610355,
      "epoch": 2.486601826121477,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004384638819477681,
      "loss": 5.0048,
      "mean_token_accuracy": 0.17853158563375474,
      "num_tokens": 54908427.0,
      "step": 25055
    },
    {
      "entropy": 5.5150984764099125,
      "epoch": 2.4870980547836443,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043843953281526727,
      "loss": 4.9884,
      "mean_token_accuracy": 0.18781101554632187,
      "num_tokens": 54918394.0,
      "step": 25060
    },
    {
      "entropy": 5.6067102432250975,
      "epoch": 2.487594283445812,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00043841517962981823,
      "loss": 5.0171,
      "mean_token_accuracy": 0.18458457440137863,
      "num_tokens": 54928613.0,
      "step": 25065
    },
    {
      "entropy": 5.617119359970093,
      "epoch": 2.4880905121079793,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043839082239202496,
      "loss": 4.9688,
      "mean_token_accuracy": 0.1821914792060852,
      "num_tokens": 54938938.0,
      "step": 25070
    },
    {
      "entropy": 5.688495206832886,
      "epoch": 2.4885867407701467,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004383664611024912,
      "loss": 5.0962,
      "mean_token_accuracy": 0.1826512783765793,
      "num_tokens": 54950152.0,
      "step": 25075
    },
    {
      "entropy": 5.658412313461303,
      "epoch": 2.489082969432314,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004383420957618213,
      "loss": 5.1355,
      "mean_token_accuracy": 0.17032338827848434,
      "num_tokens": 54962191.0,
      "step": 25080
    },
    {
      "entropy": 5.5589070320129395,
      "epoch": 2.489579198094482,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00043831772637061944,
      "loss": 4.9478,
      "mean_token_accuracy": 0.1838442638516426,
      "num_tokens": 54972532.0,
      "step": 25085
    },
    {
      "entropy": 5.6510967254638675,
      "epoch": 2.4900754267566496,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00043829335292948986,
      "loss": 5.0796,
      "mean_token_accuracy": 0.1764182984828949,
      "num_tokens": 54983212.0,
      "step": 25090
    },
    {
      "entropy": 5.546608924865723,
      "epoch": 2.490571655418817,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043826897543903696,
      "loss": 4.9853,
      "mean_token_accuracy": 0.18756668120622635,
      "num_tokens": 54994094.0,
      "step": 25095
    },
    {
      "entropy": 5.616532182693481,
      "epoch": 2.4910678840809846,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043824459389986534,
      "loss": 5.0637,
      "mean_token_accuracy": 0.18052673637866973,
      "num_tokens": 55004098.0,
      "step": 25100
    },
    {
      "entropy": 5.576532363891602,
      "epoch": 2.491564112743152,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004382202083125795,
      "loss": 4.953,
      "mean_token_accuracy": 0.18839882016181947,
      "num_tokens": 55013680.0,
      "step": 25105
    },
    {
      "entropy": 5.593909358978271,
      "epoch": 2.4920603414053195,
      "grad_norm": 1.125,
      "learning_rate": 0.0004381958186777842,
      "loss": 4.9418,
      "mean_token_accuracy": 0.18684090822935104,
      "num_tokens": 55024311.0,
      "step": 25110
    },
    {
      "entropy": 5.610959577560425,
      "epoch": 2.492556570067487,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004381714249960843,
      "loss": 5.19,
      "mean_token_accuracy": 0.1729530870914459,
      "num_tokens": 55036215.0,
      "step": 25115
    },
    {
      "entropy": 5.683563184738159,
      "epoch": 2.4930527987296545,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004381470272680846,
      "loss": 5.04,
      "mean_token_accuracy": 0.1786735847592354,
      "num_tokens": 55046402.0,
      "step": 25120
    },
    {
      "entropy": 5.629692077636719,
      "epoch": 2.493549027391822,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004381226254943902,
      "loss": 4.9851,
      "mean_token_accuracy": 0.18200834542512895,
      "num_tokens": 55056307.0,
      "step": 25125
    },
    {
      "entropy": 5.594213438034058,
      "epoch": 2.49404525605399,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004380982196756062,
      "loss": 5.0085,
      "mean_token_accuracy": 0.1827903851866722,
      "num_tokens": 55066918.0,
      "step": 25130
    },
    {
      "entropy": 5.575310373306275,
      "epoch": 2.4945414847161573,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00043807380981233777,
      "loss": 5.0266,
      "mean_token_accuracy": 0.18188861906528472,
      "num_tokens": 55078166.0,
      "step": 25135
    },
    {
      "entropy": 5.580396938323974,
      "epoch": 2.495037713378325,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004380493959051903,
      "loss": 5.0502,
      "mean_token_accuracy": 0.1841185823082924,
      "num_tokens": 55090919.0,
      "step": 25140
    },
    {
      "entropy": 5.681678295135498,
      "epoch": 2.4955339420404923,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043802497795476916,
      "loss": 5.0665,
      "mean_token_accuracy": 0.1872013971209526,
      "num_tokens": 55100512.0,
      "step": 25145
    },
    {
      "entropy": 5.592142248153687,
      "epoch": 2.4960301707026598,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004380005559616798,
      "loss": 5.0133,
      "mean_token_accuracy": 0.18684851229190827,
      "num_tokens": 55111845.0,
      "step": 25150
    },
    {
      "entropy": 5.6454627990722654,
      "epoch": 2.4965263993648272,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004379761299265279,
      "loss": 5.1425,
      "mean_token_accuracy": 0.17119362503290175,
      "num_tokens": 55123744.0,
      "step": 25155
    },
    {
      "entropy": 5.688706541061402,
      "epoch": 2.4970226280269947,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004379516998499192,
      "loss": 5.0407,
      "mean_token_accuracy": 0.18875586837530137,
      "num_tokens": 55133789.0,
      "step": 25160
    },
    {
      "entropy": 5.629477643966675,
      "epoch": 2.497518856689162,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004379272657324594,
      "loss": 5.0399,
      "mean_token_accuracy": 0.18446365296840667,
      "num_tokens": 55145325.0,
      "step": 25165
    },
    {
      "entropy": 5.609633874893189,
      "epoch": 2.49801508535133,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004379028275747545,
      "loss": 5.0044,
      "mean_token_accuracy": 0.17975860834121704,
      "num_tokens": 55157639.0,
      "step": 25170
    },
    {
      "entropy": 5.584935569763184,
      "epoch": 2.4985113140134976,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043787838537741054,
      "loss": 4.9674,
      "mean_token_accuracy": 0.18287350982427597,
      "num_tokens": 55167796.0,
      "step": 25175
    },
    {
      "entropy": 5.587255239486694,
      "epoch": 2.499007542675665,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004378539391410336,
      "loss": 5.0707,
      "mean_token_accuracy": 0.18034869134426118,
      "num_tokens": 55178384.0,
      "step": 25180
    },
    {
      "entropy": 5.681662511825562,
      "epoch": 2.4995037713378325,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043782948886622987,
      "loss": 5.0983,
      "mean_token_accuracy": 0.17667626440525055,
      "num_tokens": 55188363.0,
      "step": 25185
    },
    {
      "entropy": 5.654328393936157,
      "epoch": 2.5,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004378050345536057,
      "loss": 5.0747,
      "mean_token_accuracy": 0.17885033786296844,
      "num_tokens": 55198293.0,
      "step": 25190
    },
    {
      "entropy": 5.612078380584717,
      "epoch": 2.5004962286621675,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043778057620376745,
      "loss": 5.0439,
      "mean_token_accuracy": 0.18344273120164872,
      "num_tokens": 55208368.0,
      "step": 25195
    },
    {
      "entropy": 5.62855896949768,
      "epoch": 2.500992457324335,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043775611381732164,
      "loss": 5.0716,
      "mean_token_accuracy": 0.18054731637239457,
      "num_tokens": 55220075.0,
      "step": 25200
    },
    {
      "entropy": 5.621571111679077,
      "epoch": 2.5014886859865024,
      "grad_norm": 1.0625,
      "learning_rate": 0.000437731647394875,
      "loss": 5.0291,
      "mean_token_accuracy": 0.17928869128227234,
      "num_tokens": 55231060.0,
      "step": 25205
    },
    {
      "entropy": 5.659281873703003,
      "epoch": 2.50198491464867,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004377071769370341,
      "loss": 5.0562,
      "mean_token_accuracy": 0.17779871970415115,
      "num_tokens": 55242443.0,
      "step": 25210
    },
    {
      "entropy": 5.4976722240448,
      "epoch": 2.5024811433108374,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043768270244440587,
      "loss": 4.9843,
      "mean_token_accuracy": 0.1917301669716835,
      "num_tokens": 55253285.0,
      "step": 25215
    },
    {
      "entropy": 5.53802638053894,
      "epoch": 2.5029773719730053,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00043765822391759714,
      "loss": 4.991,
      "mean_token_accuracy": 0.1817561134696007,
      "num_tokens": 55264694.0,
      "step": 25220
    },
    {
      "entropy": 5.6204822063446045,
      "epoch": 2.5034736006351728,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004376337413572149,
      "loss": 5.0219,
      "mean_token_accuracy": 0.1823040157556534,
      "num_tokens": 55275429.0,
      "step": 25225
    },
    {
      "entropy": 5.654348421096802,
      "epoch": 2.5039698292973402,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043760925476386635,
      "loss": 5.0788,
      "mean_token_accuracy": 0.1801487222313881,
      "num_tokens": 55286764.0,
      "step": 25230
    },
    {
      "entropy": 5.503701114654541,
      "epoch": 2.5044660579595077,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004375847641381587,
      "loss": 4.9434,
      "mean_token_accuracy": 0.18740667700767516,
      "num_tokens": 55296284.0,
      "step": 25235
    },
    {
      "entropy": 5.606362295150757,
      "epoch": 2.504962286621675,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004375602694806993,
      "loss": 5.0272,
      "mean_token_accuracy": 0.18572374433279037,
      "num_tokens": 55307684.0,
      "step": 25240
    },
    {
      "entropy": 5.606105613708496,
      "epoch": 2.5054585152838427,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043753577079209537,
      "loss": 5.0067,
      "mean_token_accuracy": 0.1818462133407593,
      "num_tokens": 55318202.0,
      "step": 25245
    },
    {
      "entropy": 5.645253705978393,
      "epoch": 2.5059547439460106,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043751126807295464,
      "loss": 5.0937,
      "mean_token_accuracy": 0.17801566123962403,
      "num_tokens": 55328827.0,
      "step": 25250
    },
    {
      "entropy": 5.6482831001281735,
      "epoch": 2.506450972608178,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043748676132388464,
      "loss": 5.0782,
      "mean_token_accuracy": 0.1764460876584053,
      "num_tokens": 55338171.0,
      "step": 25255
    },
    {
      "entropy": 5.601218843460083,
      "epoch": 2.5069472012703455,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043746225054549316,
      "loss": 5.0349,
      "mean_token_accuracy": 0.17681822329759597,
      "num_tokens": 55349081.0,
      "step": 25260
    },
    {
      "entropy": 5.590871810913086,
      "epoch": 2.507443429932513,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043743773573838785,
      "loss": 5.0006,
      "mean_token_accuracy": 0.18336262553930283,
      "num_tokens": 55359344.0,
      "step": 25265
    },
    {
      "entropy": 5.567030525207519,
      "epoch": 2.5079396585946805,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043741321690317673,
      "loss": 4.9294,
      "mean_token_accuracy": 0.1927078902721405,
      "num_tokens": 55369643.0,
      "step": 25270
    },
    {
      "entropy": 5.56604380607605,
      "epoch": 2.508435887256848,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004373886940404677,
      "loss": 4.9433,
      "mean_token_accuracy": 0.19568266570568085,
      "num_tokens": 55380812.0,
      "step": 25275
    },
    {
      "entropy": 5.5357897758483885,
      "epoch": 2.5089321159190154,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004373641671508691,
      "loss": 5.0097,
      "mean_token_accuracy": 0.19161544740200043,
      "num_tokens": 55393136.0,
      "step": 25280
    },
    {
      "entropy": 5.5399254322052,
      "epoch": 2.509428344581183,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000437339636234989,
      "loss": 4.9442,
      "mean_token_accuracy": 0.18168800175189972,
      "num_tokens": 55404317.0,
      "step": 25285
    },
    {
      "entropy": 5.601834917068482,
      "epoch": 2.5099245732433504,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043731510129343574,
      "loss": 5.0885,
      "mean_token_accuracy": 0.17560949176549911,
      "num_tokens": 55415812.0,
      "step": 25290
    },
    {
      "entropy": 5.664914083480835,
      "epoch": 2.510420801905518,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043729056232681766,
      "loss": 5.0732,
      "mean_token_accuracy": 0.17686787843704224,
      "num_tokens": 55426402.0,
      "step": 25295
    },
    {
      "entropy": 5.625409412384033,
      "epoch": 2.5109170305676853,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004372660193357434,
      "loss": 5.0777,
      "mean_token_accuracy": 0.18300699293613434,
      "num_tokens": 55437474.0,
      "step": 25300
    },
    {
      "entropy": 5.605809354782105,
      "epoch": 2.5114132592298533,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004372414723208215,
      "loss": 5.0205,
      "mean_token_accuracy": 0.18146351873874664,
      "num_tokens": 55447587.0,
      "step": 25305
    },
    {
      "entropy": 5.731449794769287,
      "epoch": 2.5119094878920207,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004372169212826606,
      "loss": 5.1636,
      "mean_token_accuracy": 0.17572268843650818,
      "num_tokens": 55460587.0,
      "step": 25310
    },
    {
      "entropy": 5.631799697875977,
      "epoch": 2.512405716554188,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00043719236622186976,
      "loss": 5.0168,
      "mean_token_accuracy": 0.1847177892923355,
      "num_tokens": 55471150.0,
      "step": 25315
    },
    {
      "entropy": 5.5658527374267575,
      "epoch": 2.5129019452163557,
      "grad_norm": 1.25,
      "learning_rate": 0.0004371678071390576,
      "loss": 5.0161,
      "mean_token_accuracy": 0.18901062607765198,
      "num_tokens": 55480368.0,
      "step": 25320
    },
    {
      "entropy": 5.642173910140992,
      "epoch": 2.513398173878523,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043714324403483333,
      "loss": 5.0778,
      "mean_token_accuracy": 0.17868364751338958,
      "num_tokens": 55491954.0,
      "step": 25325
    },
    {
      "entropy": 5.655166435241699,
      "epoch": 2.5138944025406906,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043711867690980605,
      "loss": 5.0368,
      "mean_token_accuracy": 0.1766806274652481,
      "num_tokens": 55502560.0,
      "step": 25330
    },
    {
      "entropy": 5.5948121547698975,
      "epoch": 2.5143906312028586,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00043709410576458477,
      "loss": 5.014,
      "mean_token_accuracy": 0.1829821601510048,
      "num_tokens": 55514571.0,
      "step": 25335
    },
    {
      "entropy": 5.612494421005249,
      "epoch": 2.514886859865026,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043706953059977907,
      "loss": 5.0264,
      "mean_token_accuracy": 0.18312541097402574,
      "num_tokens": 55525206.0,
      "step": 25340
    },
    {
      "entropy": 5.607606220245361,
      "epoch": 2.5153830885271935,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043704495141599816,
      "loss": 5.0529,
      "mean_token_accuracy": 0.17500218898057937,
      "num_tokens": 55536551.0,
      "step": 25345
    },
    {
      "entropy": 5.630746078491211,
      "epoch": 2.515879317189361,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004370203682138517,
      "loss": 5.0268,
      "mean_token_accuracy": 0.17296613454818727,
      "num_tokens": 55547842.0,
      "step": 25350
    },
    {
      "entropy": 5.705642175674439,
      "epoch": 2.5163755458515285,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004369957809939492,
      "loss": 4.9982,
      "mean_token_accuracy": 0.19703430235385894,
      "num_tokens": 55558248.0,
      "step": 25355
    },
    {
      "entropy": 5.591320657730103,
      "epoch": 2.516871774513696,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004369711897569004,
      "loss": 4.987,
      "mean_token_accuracy": 0.17640934884548187,
      "num_tokens": 55569195.0,
      "step": 25360
    },
    {
      "entropy": 5.584616088867188,
      "epoch": 2.5173680031758634,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004369465945033151,
      "loss": 4.9675,
      "mean_token_accuracy": 0.18890831917524337,
      "num_tokens": 55579649.0,
      "step": 25365
    },
    {
      "entropy": 5.546411228179932,
      "epoch": 2.517864231838031,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043692199523380325,
      "loss": 4.8655,
      "mean_token_accuracy": 0.18947683721780778,
      "num_tokens": 55589550.0,
      "step": 25370
    },
    {
      "entropy": 5.601106977462768,
      "epoch": 2.5183604605001984,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004368973919489748,
      "loss": 5.0284,
      "mean_token_accuracy": 0.18239061236381532,
      "num_tokens": 55600673.0,
      "step": 25375
    },
    {
      "entropy": 5.490999555587768,
      "epoch": 2.518856689162366,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00043687278464943986,
      "loss": 4.9899,
      "mean_token_accuracy": 0.18310610353946685,
      "num_tokens": 55612045.0,
      "step": 25380
    },
    {
      "entropy": 5.5714569091796875,
      "epoch": 2.5193529178245333,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004368481733358088,
      "loss": 5.0227,
      "mean_token_accuracy": 0.1809934437274933,
      "num_tokens": 55623666.0,
      "step": 25385
    },
    {
      "entropy": 5.619052505493164,
      "epoch": 2.519849146486701,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004368235580086917,
      "loss": 5.0114,
      "mean_token_accuracy": 0.18551373183727266,
      "num_tokens": 55634236.0,
      "step": 25390
    },
    {
      "entropy": 5.54365644454956,
      "epoch": 2.5203453751488687,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043679893866869904,
      "loss": 4.9544,
      "mean_token_accuracy": 0.18328907042741777,
      "num_tokens": 55645412.0,
      "step": 25395
    },
    {
      "entropy": 5.578490734100342,
      "epoch": 2.520841603811036,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004367743153164414,
      "loss": 4.9793,
      "mean_token_accuracy": 0.18199385702610016,
      "num_tokens": 55655861.0,
      "step": 25400
    },
    {
      "entropy": 5.579327058792114,
      "epoch": 2.5213378324732036,
      "grad_norm": 1.125,
      "learning_rate": 0.00043674968795252926,
      "loss": 4.9636,
      "mean_token_accuracy": 0.18757071048021318,
      "num_tokens": 55665227.0,
      "step": 25405
    },
    {
      "entropy": 5.570992803573608,
      "epoch": 2.521834061135371,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004367250565775734,
      "loss": 4.9995,
      "mean_token_accuracy": 0.18337539285421373,
      "num_tokens": 55676104.0,
      "step": 25410
    },
    {
      "entropy": 5.604399251937866,
      "epoch": 2.5223302897975386,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004367004211921847,
      "loss": 5.1001,
      "mean_token_accuracy": 0.17614561319351196,
      "num_tokens": 55688222.0,
      "step": 25415
    },
    {
      "entropy": 5.636390542984008,
      "epoch": 2.522826518459706,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00043667578179697387,
      "loss": 5.0259,
      "mean_token_accuracy": 0.18236683756113053,
      "num_tokens": 55700138.0,
      "step": 25420
    },
    {
      "entropy": 5.624015522003174,
      "epoch": 2.523322747121874,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004366511383925521,
      "loss": 4.9874,
      "mean_token_accuracy": 0.18538923114538192,
      "num_tokens": 55709444.0,
      "step": 25425
    },
    {
      "entropy": 5.697638416290284,
      "epoch": 2.5238189757840415,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004366264909795305,
      "loss": 5.111,
      "mean_token_accuracy": 0.1787201538681984,
      "num_tokens": 55720327.0,
      "step": 25430
    },
    {
      "entropy": 5.540494823455811,
      "epoch": 2.524315204446209,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004366018395585201,
      "loss": 4.9004,
      "mean_token_accuracy": 0.1879577487707138,
      "num_tokens": 55730727.0,
      "step": 25435
    },
    {
      "entropy": 5.619707822799683,
      "epoch": 2.5248114331083764,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004365771841301323,
      "loss": 5.0603,
      "mean_token_accuracy": 0.1808421105146408,
      "num_tokens": 55741568.0,
      "step": 25440
    },
    {
      "entropy": 5.561898946762085,
      "epoch": 2.525307661770544,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004365525246949786,
      "loss": 5.0016,
      "mean_token_accuracy": 0.1854325130581856,
      "num_tokens": 55752511.0,
      "step": 25445
    },
    {
      "entropy": 5.621946525573731,
      "epoch": 2.5258038904327114,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043652786125367035,
      "loss": 4.9873,
      "mean_token_accuracy": 0.18922401517629622,
      "num_tokens": 55763004.0,
      "step": 25450
    },
    {
      "entropy": 5.625278615951538,
      "epoch": 2.526300119094879,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043650319380681915,
      "loss": 5.0671,
      "mean_token_accuracy": 0.17767291367053986,
      "num_tokens": 55773276.0,
      "step": 25455
    },
    {
      "entropy": 5.619245719909668,
      "epoch": 2.5267963477570463,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004364785223550368,
      "loss": 5.0548,
      "mean_token_accuracy": 0.18142981380224227,
      "num_tokens": 55784607.0,
      "step": 25460
    },
    {
      "entropy": 5.65968165397644,
      "epoch": 2.527292576419214,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00043645384689893504,
      "loss": 5.0713,
      "mean_token_accuracy": 0.17207679897546768,
      "num_tokens": 55796730.0,
      "step": 25465
    },
    {
      "entropy": 5.6546252250671385,
      "epoch": 2.5277888050813813,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004364291674391258,
      "loss": 5.0666,
      "mean_token_accuracy": 0.17559448927640914,
      "num_tokens": 55806940.0,
      "step": 25470
    },
    {
      "entropy": 5.613480806350708,
      "epoch": 2.528285033743549,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000436404483976221,
      "loss": 5.0039,
      "mean_token_accuracy": 0.17970138788223267,
      "num_tokens": 55818229.0,
      "step": 25475
    },
    {
      "entropy": 5.618879175186157,
      "epoch": 2.5287812624057167,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043637979651083285,
      "loss": 5.085,
      "mean_token_accuracy": 0.1783071279525757,
      "num_tokens": 55831100.0,
      "step": 25480
    },
    {
      "entropy": 5.622269821166992,
      "epoch": 2.529277491067884,
      "grad_norm": 1.125,
      "learning_rate": 0.00043635510504357343,
      "loss": 5.0112,
      "mean_token_accuracy": 0.18103124648332597,
      "num_tokens": 55841713.0,
      "step": 25485
    },
    {
      "entropy": 5.599541568756104,
      "epoch": 2.5297737197300516,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043633040957505515,
      "loss": 4.9942,
      "mean_token_accuracy": 0.17404750734567642,
      "num_tokens": 55852737.0,
      "step": 25490
    },
    {
      "entropy": 5.667076778411865,
      "epoch": 2.530269948392219,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004363057101058903,
      "loss": 5.0954,
      "mean_token_accuracy": 0.18068393766880037,
      "num_tokens": 55863471.0,
      "step": 25495
    },
    {
      "entropy": 5.56984395980835,
      "epoch": 2.5307661770543866,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004362810066366915,
      "loss": 4.9566,
      "mean_token_accuracy": 0.18414071798324586,
      "num_tokens": 55874305.0,
      "step": 25500
    },
    {
      "entropy": 5.636424398422241,
      "epoch": 2.531262405716554,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00043625629916807117,
      "loss": 5.0989,
      "mean_token_accuracy": 0.17998090535402297,
      "num_tokens": 55887400.0,
      "step": 25505
    },
    {
      "entropy": 5.567593669891357,
      "epoch": 2.531758634378722,
      "grad_norm": 0.890625,
      "learning_rate": 0.00043623158770064214,
      "loss": 5.0421,
      "mean_token_accuracy": 0.18298677802085878,
      "num_tokens": 55900052.0,
      "step": 25510
    },
    {
      "entropy": 5.704504346847534,
      "epoch": 2.5322548630408894,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004362068722350172,
      "loss": 5.1664,
      "mean_token_accuracy": 0.17730359435081483,
      "num_tokens": 55911609.0,
      "step": 25515
    },
    {
      "entropy": 5.638509225845337,
      "epoch": 2.532751091703057,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043618215277180917,
      "loss": 4.9833,
      "mean_token_accuracy": 0.18710464835166932,
      "num_tokens": 55922873.0,
      "step": 25520
    },
    {
      "entropy": 5.642351150512695,
      "epoch": 2.5332473203652244,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043615742931163106,
      "loss": 5.0383,
      "mean_token_accuracy": 0.17975827902555466,
      "num_tokens": 55932161.0,
      "step": 25525
    },
    {
      "entropy": 5.657623720169068,
      "epoch": 2.533743549027392,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000436132701855096,
      "loss": 5.1483,
      "mean_token_accuracy": 0.1738502100110054,
      "num_tokens": 55943601.0,
      "step": 25530
    },
    {
      "entropy": 5.542093229293823,
      "epoch": 2.5342397776895593,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043610797040281705,
      "loss": 5.0199,
      "mean_token_accuracy": 0.180764339864254,
      "num_tokens": 55954981.0,
      "step": 25535
    },
    {
      "entropy": 5.6258495330810545,
      "epoch": 2.534736006351727,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004360832349554077,
      "loss": 5.0795,
      "mean_token_accuracy": 0.1805799275636673,
      "num_tokens": 55964407.0,
      "step": 25540
    },
    {
      "entropy": 5.588602542877197,
      "epoch": 2.5352322350138943,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004360584955134811,
      "loss": 4.9846,
      "mean_token_accuracy": 0.17978089153766633,
      "num_tokens": 55974634.0,
      "step": 25545
    },
    {
      "entropy": 5.614993572235107,
      "epoch": 2.5357284636760618,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043603375207765103,
      "loss": 5.0451,
      "mean_token_accuracy": 0.17299122959375382,
      "num_tokens": 55986753.0,
      "step": 25550
    },
    {
      "entropy": 5.604440498352051,
      "epoch": 2.5362246923382292,
      "grad_norm": 1.0,
      "learning_rate": 0.0004360090046485308,
      "loss": 5.0073,
      "mean_token_accuracy": 0.18535103052854537,
      "num_tokens": 55997839.0,
      "step": 25555
    },
    {
      "entropy": 5.63296594619751,
      "epoch": 2.536720921000397,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004359842532267342,
      "loss": 5.0369,
      "mean_token_accuracy": 0.18951851576566697,
      "num_tokens": 56008765.0,
      "step": 25560
    },
    {
      "entropy": 5.595081186294555,
      "epoch": 2.5372171496625646,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043595949781287505,
      "loss": 5.0227,
      "mean_token_accuracy": 0.18444891273975372,
      "num_tokens": 56020077.0,
      "step": 25565
    },
    {
      "entropy": 5.5984735012054445,
      "epoch": 2.537713378324732,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043593473840756715,
      "loss": 5.0154,
      "mean_token_accuracy": 0.18994038701057434,
      "num_tokens": 56031298.0,
      "step": 25570
    },
    {
      "entropy": 5.652325963973999,
      "epoch": 2.5382096069868996,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004359099750114245,
      "loss": 5.0573,
      "mean_token_accuracy": 0.1790929466485977,
      "num_tokens": 56043955.0,
      "step": 25575
    },
    {
      "entropy": 5.594660043716431,
      "epoch": 2.538705835649067,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004358852076250613,
      "loss": 5.0083,
      "mean_token_accuracy": 0.18631981015205384,
      "num_tokens": 56055734.0,
      "step": 25580
    },
    {
      "entropy": 5.665326976776123,
      "epoch": 2.5392020643112345,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043586043624909153,
      "loss": 5.0478,
      "mean_token_accuracy": 0.1827930748462677,
      "num_tokens": 56065472.0,
      "step": 25585
    },
    {
      "entropy": 5.70714955329895,
      "epoch": 2.539698292973402,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004358356608841296,
      "loss": 5.0502,
      "mean_token_accuracy": 0.1836530700325966,
      "num_tokens": 56075691.0,
      "step": 25590
    },
    {
      "entropy": 5.5869465351104735,
      "epoch": 2.54019452163557,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043581088153078984,
      "loss": 5.0324,
      "mean_token_accuracy": 0.1850854590535164,
      "num_tokens": 56085900.0,
      "step": 25595
    },
    {
      "entropy": 5.609371328353882,
      "epoch": 2.5406907502977374,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00043578609818968676,
      "loss": 5.0181,
      "mean_token_accuracy": 0.18428054004907607,
      "num_tokens": 56097675.0,
      "step": 25600
    },
    {
      "entropy": 5.701621723175049,
      "epoch": 2.541186978959905,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043576131086143496,
      "loss": 5.0972,
      "mean_token_accuracy": 0.1665532946586609,
      "num_tokens": 56108102.0,
      "step": 25605
    },
    {
      "entropy": 5.542317676544189,
      "epoch": 2.5416832076220723,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043573651954664897,
      "loss": 4.9304,
      "mean_token_accuracy": 0.18984250128269195,
      "num_tokens": 56118489.0,
      "step": 25610
    },
    {
      "entropy": 5.5337389469146725,
      "epoch": 2.54217943628424,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043571172424594363,
      "loss": 4.9443,
      "mean_token_accuracy": 0.18903415650129318,
      "num_tokens": 56127949.0,
      "step": 25615
    },
    {
      "entropy": 5.58496241569519,
      "epoch": 2.5426756649464073,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004356869249599339,
      "loss": 4.9813,
      "mean_token_accuracy": 0.18364671170711516,
      "num_tokens": 56138209.0,
      "step": 25620
    },
    {
      "entropy": 5.599924182891845,
      "epoch": 2.5431718936085748,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004356621216892347,
      "loss": 5.051,
      "mean_token_accuracy": 0.1853262022137642,
      "num_tokens": 56149005.0,
      "step": 25625
    },
    {
      "entropy": 5.6181371212005615,
      "epoch": 2.5436681222707422,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043563731443446095,
      "loss": 4.9751,
      "mean_token_accuracy": 0.1808355137705803,
      "num_tokens": 56159406.0,
      "step": 25630
    },
    {
      "entropy": 5.591839790344238,
      "epoch": 2.5441643509329097,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043561250319622806,
      "loss": 5.0372,
      "mean_token_accuracy": 0.18283226042985917,
      "num_tokens": 56169998.0,
      "step": 25635
    },
    {
      "entropy": 5.695139312744141,
      "epoch": 2.544660579595077,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00043558768797515116,
      "loss": 5.1021,
      "mean_token_accuracy": 0.1771041199564934,
      "num_tokens": 56181096.0,
      "step": 25640
    },
    {
      "entropy": 5.636304616928101,
      "epoch": 2.5451568082572447,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043556286877184566,
      "loss": 5.0237,
      "mean_token_accuracy": 0.18300354927778245,
      "num_tokens": 56191909.0,
      "step": 25645
    },
    {
      "entropy": 5.63016905784607,
      "epoch": 2.5456530369194126,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004355380455869269,
      "loss": 5.0525,
      "mean_token_accuracy": 0.17828502207994462,
      "num_tokens": 56202725.0,
      "step": 25650
    },
    {
      "entropy": 5.624461269378662,
      "epoch": 2.54614926558158,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004355132184210106,
      "loss": 5.0189,
      "mean_token_accuracy": 0.18688997626304626,
      "num_tokens": 56213596.0,
      "step": 25655
    },
    {
      "entropy": 5.594005680084228,
      "epoch": 2.5466454942437475,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004354883872747124,
      "loss": 5.0259,
      "mean_token_accuracy": 0.18167044818401337,
      "num_tokens": 56224416.0,
      "step": 25660
    },
    {
      "entropy": 5.58353180885315,
      "epoch": 2.547141722905915,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043546355214864796,
      "loss": 5.0288,
      "mean_token_accuracy": 0.1826405018568039,
      "num_tokens": 56234325.0,
      "step": 25665
    },
    {
      "entropy": 5.607661771774292,
      "epoch": 2.5476379515680825,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004354387130434332,
      "loss": 5.0155,
      "mean_token_accuracy": 0.18114465475082397,
      "num_tokens": 56244178.0,
      "step": 25670
    },
    {
      "entropy": 5.583818006515503,
      "epoch": 2.54813418023025,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004354138699596841,
      "loss": 5.0495,
      "mean_token_accuracy": 0.18329914808273315,
      "num_tokens": 56256036.0,
      "step": 25675
    },
    {
      "entropy": 5.579014730453491,
      "epoch": 2.548630408892418,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004353890228980167,
      "loss": 5.0151,
      "mean_token_accuracy": 0.17689924389123918,
      "num_tokens": 56266203.0,
      "step": 25680
    },
    {
      "entropy": 5.595081329345703,
      "epoch": 2.5491266375545854,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004353641718590471,
      "loss": 5.0387,
      "mean_token_accuracy": 0.1842021882534027,
      "num_tokens": 56276958.0,
      "step": 25685
    },
    {
      "entropy": 5.625451707839966,
      "epoch": 2.549622866216753,
      "grad_norm": 0.96875,
      "learning_rate": 0.00043533931684339156,
      "loss": 5.0602,
      "mean_token_accuracy": 0.18240056931972504,
      "num_tokens": 56287941.0,
      "step": 25690
    },
    {
      "entropy": 5.553532648086548,
      "epoch": 2.5501190948789203,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004353144578516665,
      "loss": 4.9501,
      "mean_token_accuracy": 0.18917568475008012,
      "num_tokens": 56299390.0,
      "step": 25695
    },
    {
      "entropy": 5.57025294303894,
      "epoch": 2.5506153235410878,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043528959488448835,
      "loss": 5.009,
      "mean_token_accuracy": 0.18140523582696916,
      "num_tokens": 56310762.0,
      "step": 25700
    },
    {
      "entropy": 5.642427635192871,
      "epoch": 2.5511115522032553,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00043526472794247365,
      "loss": 4.9911,
      "mean_token_accuracy": 0.18194476813077926,
      "num_tokens": 56322522.0,
      "step": 25705
    },
    {
      "entropy": 5.637915468215942,
      "epoch": 2.5516077808654227,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043523985702623896,
      "loss": 5.0624,
      "mean_token_accuracy": 0.17940168529748918,
      "num_tokens": 56334124.0,
      "step": 25710
    },
    {
      "entropy": 5.578952693939209,
      "epoch": 2.55210400952759,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043521498213640113,
      "loss": 5.0023,
      "mean_token_accuracy": 0.18222629725933076,
      "num_tokens": 56344761.0,
      "step": 25715
    },
    {
      "entropy": 5.55111517906189,
      "epoch": 2.5526002381897577,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00043519010327357704,
      "loss": 4.9821,
      "mean_token_accuracy": 0.18635161519050597,
      "num_tokens": 56355518.0,
      "step": 25720
    },
    {
      "entropy": 5.603583765029907,
      "epoch": 2.553096466851925,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004351652204383836,
      "loss": 5.0722,
      "mean_token_accuracy": 0.18054696023464203,
      "num_tokens": 56366307.0,
      "step": 25725
    },
    {
      "entropy": 5.633018493652344,
      "epoch": 2.5535926955140926,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004351403336314377,
      "loss": 4.97,
      "mean_token_accuracy": 0.1901824116706848,
      "num_tokens": 56377065.0,
      "step": 25730
    },
    {
      "entropy": 5.653081655502319,
      "epoch": 2.5540889241762605,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004351154428533566,
      "loss": 5.0366,
      "mean_token_accuracy": 0.17846063077449797,
      "num_tokens": 56387741.0,
      "step": 25735
    },
    {
      "entropy": 5.616034984588623,
      "epoch": 2.554585152838428,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004350905481047576,
      "loss": 5.2021,
      "mean_token_accuracy": 0.17490432113409043,
      "num_tokens": 56398927.0,
      "step": 25740
    },
    {
      "entropy": 5.702428245544434,
      "epoch": 2.5550813815005955,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004350656493862579,
      "loss": 5.1581,
      "mean_token_accuracy": 0.172688065469265,
      "num_tokens": 56408791.0,
      "step": 25745
    },
    {
      "entropy": 5.6360736846923825,
      "epoch": 2.555577610162763,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00043504074669847517,
      "loss": 4.9708,
      "mean_token_accuracy": 0.1880552962422371,
      "num_tokens": 56419206.0,
      "step": 25750
    },
    {
      "entropy": 5.71256947517395,
      "epoch": 2.5560738388249304,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004350158400420266,
      "loss": 5.1602,
      "mean_token_accuracy": 0.17241645753383636,
      "num_tokens": 56431774.0,
      "step": 25755
    },
    {
      "entropy": 5.683459997177124,
      "epoch": 2.556570067487098,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043499092941753003,
      "loss": 5.0505,
      "mean_token_accuracy": 0.1784671202301979,
      "num_tokens": 56442048.0,
      "step": 25760
    },
    {
      "entropy": 5.5669252395629885,
      "epoch": 2.557066296149266,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043496601482560327,
      "loss": 5.0362,
      "mean_token_accuracy": 0.17566805928945542,
      "num_tokens": 56453347.0,
      "step": 25765
    },
    {
      "entropy": 5.543325281143188,
      "epoch": 2.5575625248114333,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043494109626686394,
      "loss": 4.9943,
      "mean_token_accuracy": 0.18804705291986465,
      "num_tokens": 56465323.0,
      "step": 25770
    },
    {
      "entropy": 5.641653251647949,
      "epoch": 2.558058753473601,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043491617374193017,
      "loss": 5.0431,
      "mean_token_accuracy": 0.18054264783859253,
      "num_tokens": 56474704.0,
      "step": 25775
    },
    {
      "entropy": 5.620493412017822,
      "epoch": 2.5585549821357683,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004348912472514197,
      "loss": 5.0519,
      "mean_token_accuracy": 0.1843789577484131,
      "num_tokens": 56485028.0,
      "step": 25780
    },
    {
      "entropy": 5.615398836135864,
      "epoch": 2.5590512107979357,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043486631679595097,
      "loss": 4.9495,
      "mean_token_accuracy": 0.19075653851032257,
      "num_tokens": 56496169.0,
      "step": 25785
    },
    {
      "entropy": 5.5327721118927,
      "epoch": 2.559547439460103,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004348413823761421,
      "loss": 4.9637,
      "mean_token_accuracy": 0.18663237988948822,
      "num_tokens": 56506324.0,
      "step": 25790
    },
    {
      "entropy": 5.560867691040039,
      "epoch": 2.5600436681222707,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004348164439926112,
      "loss": 4.9234,
      "mean_token_accuracy": 0.18896648287773132,
      "num_tokens": 56516901.0,
      "step": 25795
    },
    {
      "entropy": 5.559589147567749,
      "epoch": 2.560539896784438,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043479150164597695,
      "loss": 4.9321,
      "mean_token_accuracy": 0.18817751705646515,
      "num_tokens": 56527241.0,
      "step": 25800
    },
    {
      "entropy": 5.6210329055786135,
      "epoch": 2.5610361254466056,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043476655533685783,
      "loss": 5.1247,
      "mean_token_accuracy": 0.18013921082019807,
      "num_tokens": 56539078.0,
      "step": 25805
    },
    {
      "entropy": 5.569391441345215,
      "epoch": 2.561532354108773,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043474160506587234,
      "loss": 4.9706,
      "mean_token_accuracy": 0.18349270075559615,
      "num_tokens": 56551141.0,
      "step": 25810
    },
    {
      "entropy": 5.637751293182373,
      "epoch": 2.5620285827709406,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004347166508336393,
      "loss": 5.0953,
      "mean_token_accuracy": 0.17926084697246553,
      "num_tokens": 56561836.0,
      "step": 25815
    },
    {
      "entropy": 5.673055934906006,
      "epoch": 2.5625248114331085,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043469169264077754,
      "loss": 5.0033,
      "mean_token_accuracy": 0.18479280322790145,
      "num_tokens": 56572699.0,
      "step": 25820
    },
    {
      "entropy": 5.649527168273925,
      "epoch": 2.563021040095276,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004346667304879058,
      "loss": 5.0997,
      "mean_token_accuracy": 0.17993267625570297,
      "num_tokens": 56582296.0,
      "step": 25825
    },
    {
      "entropy": 5.601381015777588,
      "epoch": 2.5635172687574435,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00043464176437564317,
      "loss": 5.0201,
      "mean_token_accuracy": 0.1793950691819191,
      "num_tokens": 56592907.0,
      "step": 25830
    },
    {
      "entropy": 5.586054897308349,
      "epoch": 2.564013497419611,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004346167943046089,
      "loss": 5.0046,
      "mean_token_accuracy": 0.19203923791646957,
      "num_tokens": 56604995.0,
      "step": 25835
    },
    {
      "entropy": 5.597304391860962,
      "epoch": 2.5645097260817784,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000434591820275422,
      "loss": 4.9859,
      "mean_token_accuracy": 0.18328161984682084,
      "num_tokens": 56614809.0,
      "step": 25840
    },
    {
      "entropy": 5.672670078277588,
      "epoch": 2.565005954743946,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004345668422887018,
      "loss": 5.0696,
      "mean_token_accuracy": 0.1739903137087822,
      "num_tokens": 56624409.0,
      "step": 25845
    },
    {
      "entropy": 5.624929475784302,
      "epoch": 2.565502183406114,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004345418603450678,
      "loss": 5.0533,
      "mean_token_accuracy": 0.172790789604187,
      "num_tokens": 56634596.0,
      "step": 25850
    },
    {
      "entropy": 5.598063564300537,
      "epoch": 2.5659984120682813,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004345168744451395,
      "loss": 5.0264,
      "mean_token_accuracy": 0.18219223916530608,
      "num_tokens": 56646710.0,
      "step": 25855
    },
    {
      "entropy": 5.596146726608277,
      "epoch": 2.5664946407304488,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004344918845895364,
      "loss": 5.0146,
      "mean_token_accuracy": 0.18157119154930115,
      "num_tokens": 56658427.0,
      "step": 25860
    },
    {
      "entropy": 5.639008045196533,
      "epoch": 2.5669908693926162,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004344668907788782,
      "loss": 5.0933,
      "mean_token_accuracy": 0.17930882126092912,
      "num_tokens": 56669039.0,
      "step": 25865
    },
    {
      "entropy": 5.611382150650025,
      "epoch": 2.5674870980547837,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004344418930137848,
      "loss": 4.9864,
      "mean_token_accuracy": 0.18810024857521057,
      "num_tokens": 56679063.0,
      "step": 25870
    },
    {
      "entropy": 5.6468727588653564,
      "epoch": 2.567983326716951,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004344168912948759,
      "loss": 5.0289,
      "mean_token_accuracy": 0.18402439802885057,
      "num_tokens": 56690142.0,
      "step": 25875
    },
    {
      "entropy": 5.594340038299561,
      "epoch": 2.5684795553791187,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043439188562277174,
      "loss": 5.0446,
      "mean_token_accuracy": 0.18425420373678209,
      "num_tokens": 56700918.0,
      "step": 25880
    },
    {
      "entropy": 5.711795949935913,
      "epoch": 2.568975784041286,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043436687599809226,
      "loss": 5.1709,
      "mean_token_accuracy": 0.17287722378969192,
      "num_tokens": 56713520.0,
      "step": 25885
    },
    {
      "entropy": 5.617305278778076,
      "epoch": 2.5694720127034536,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043434186242145763,
      "loss": 5.0698,
      "mean_token_accuracy": 0.17742855846881866,
      "num_tokens": 56723428.0,
      "step": 25890
    },
    {
      "entropy": 5.622738790512085,
      "epoch": 2.569968241365621,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004343168448934881,
      "loss": 5.0014,
      "mean_token_accuracy": 0.18396189510822297,
      "num_tokens": 56734167.0,
      "step": 25895
    },
    {
      "entropy": 5.618954706192016,
      "epoch": 2.5704644700277886,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043429182341480423,
      "loss": 4.9686,
      "mean_token_accuracy": 0.18685033023357392,
      "num_tokens": 56745196.0,
      "step": 25900
    },
    {
      "entropy": 5.583302021026611,
      "epoch": 2.5709606986899565,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004342667979860263,
      "loss": 4.9706,
      "mean_token_accuracy": 0.1803652301430702,
      "num_tokens": 56756166.0,
      "step": 25905
    },
    {
      "entropy": 5.5895223140716555,
      "epoch": 2.571456927352124,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043424176860777495,
      "loss": 4.9985,
      "mean_token_accuracy": 0.1835214763879776,
      "num_tokens": 56766690.0,
      "step": 25910
    },
    {
      "entropy": 5.6034010887146,
      "epoch": 2.5719531560142914,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043421673528067095,
      "loss": 5.0514,
      "mean_token_accuracy": 0.18161885291337967,
      "num_tokens": 56776995.0,
      "step": 25915
    },
    {
      "entropy": 5.685805225372315,
      "epoch": 2.572449384676459,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043419169800533496,
      "loss": 5.11,
      "mean_token_accuracy": 0.18113733530044557,
      "num_tokens": 56788509.0,
      "step": 25920
    },
    {
      "entropy": 5.626034116744995,
      "epoch": 2.5729456133386264,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004341666567823879,
      "loss": 5.0866,
      "mean_token_accuracy": 0.17504209280014038,
      "num_tokens": 56799027.0,
      "step": 25925
    },
    {
      "entropy": 5.644856643676758,
      "epoch": 2.573441842000794,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004341416116124508,
      "loss": 5.0629,
      "mean_token_accuracy": 0.1799465537071228,
      "num_tokens": 56809796.0,
      "step": 25930
    },
    {
      "entropy": 5.641553449630737,
      "epoch": 2.5739380706629618,
      "grad_norm": 1.0,
      "learning_rate": 0.0004341165624961445,
      "loss": 5.0698,
      "mean_token_accuracy": 0.17619462460279464,
      "num_tokens": 56820523.0,
      "step": 25935
    },
    {
      "entropy": 5.610048913955689,
      "epoch": 2.5744342993251292,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043409150943409035,
      "loss": 5.0382,
      "mean_token_accuracy": 0.1770606130361557,
      "num_tokens": 56831668.0,
      "step": 25940
    },
    {
      "entropy": 5.544254350662231,
      "epoch": 2.5749305279872967,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043406645242690967,
      "loss": 4.9439,
      "mean_token_accuracy": 0.18510528057813644,
      "num_tokens": 56841950.0,
      "step": 25945
    },
    {
      "entropy": 5.604423761367798,
      "epoch": 2.575426756649464,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004340413914752237,
      "loss": 5.0561,
      "mean_token_accuracy": 0.17846496552228927,
      "num_tokens": 56852190.0,
      "step": 25950
    },
    {
      "entropy": 5.6655614376068115,
      "epoch": 2.5759229853116317,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004340163265796539,
      "loss": 5.0726,
      "mean_token_accuracy": 0.17876274436712264,
      "num_tokens": 56864255.0,
      "step": 25955
    },
    {
      "entropy": 5.680318927764892,
      "epoch": 2.576419213973799,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004339912577408218,
      "loss": 4.9919,
      "mean_token_accuracy": 0.18346475958824157,
      "num_tokens": 56874711.0,
      "step": 25960
    },
    {
      "entropy": 5.533146858215332,
      "epoch": 2.5769154426359666,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004339661849593492,
      "loss": 4.9846,
      "mean_token_accuracy": 0.17937853336334228,
      "num_tokens": 56885790.0,
      "step": 25965
    },
    {
      "entropy": 5.607720184326172,
      "epoch": 2.577411671298134,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004339411082358577,
      "loss": 5.0272,
      "mean_token_accuracy": 0.18358489125967026,
      "num_tokens": 56896452.0,
      "step": 25970
    },
    {
      "entropy": 5.619990682601928,
      "epoch": 2.5779078999603016,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00043391602757096923,
      "loss": 5.1435,
      "mean_token_accuracy": 0.18090844601392747,
      "num_tokens": 56908216.0,
      "step": 25975
    },
    {
      "entropy": 5.659113121032715,
      "epoch": 2.578404128622469,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043389094296530565,
      "loss": 5.0306,
      "mean_token_accuracy": 0.1734323963522911,
      "num_tokens": 56918346.0,
      "step": 25980
    },
    {
      "entropy": 5.6343835353851315,
      "epoch": 2.5789003572846365,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004338658544194891,
      "loss": 5.1414,
      "mean_token_accuracy": 0.1708892196416855,
      "num_tokens": 56930950.0,
      "step": 25985
    },
    {
      "entropy": 5.6456057071685795,
      "epoch": 2.5793965859468044,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043384076193414176,
      "loss": 5.0749,
      "mean_token_accuracy": 0.18096836805343627,
      "num_tokens": 56940622.0,
      "step": 25990
    },
    {
      "entropy": 5.624816226959228,
      "epoch": 2.579892814608972,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043381566550988564,
      "loss": 5.014,
      "mean_token_accuracy": 0.18652373403310776,
      "num_tokens": 56952463.0,
      "step": 25995
    },
    {
      "entropy": 5.612960243225098,
      "epoch": 2.5803890432711394,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043379056514734334,
      "loss": 4.9784,
      "mean_token_accuracy": 0.18591365963220596,
      "num_tokens": 56961705.0,
      "step": 26000
    },
    {
      "entropy": 5.611596632003784,
      "epoch": 2.580885271933307,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004337654608471372,
      "loss": 5.0542,
      "mean_token_accuracy": 0.17360775023698807,
      "num_tokens": 56972359.0,
      "step": 26005
    },
    {
      "entropy": 5.606084966659546,
      "epoch": 2.5813815005954743,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043374035260988964,
      "loss": 4.9931,
      "mean_token_accuracy": 0.18574919998645784,
      "num_tokens": 56983342.0,
      "step": 26010
    },
    {
      "entropy": 5.597082996368409,
      "epoch": 2.581877729257642,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043371524043622344,
      "loss": 4.9908,
      "mean_token_accuracy": 0.188584566116333,
      "num_tokens": 56993859.0,
      "step": 26015
    },
    {
      "entropy": 5.649916696548462,
      "epoch": 2.5823739579198093,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004336901243267614,
      "loss": 5.1152,
      "mean_token_accuracy": 0.17783222496509551,
      "num_tokens": 57006631.0,
      "step": 26020
    },
    {
      "entropy": 5.674210071563721,
      "epoch": 2.582870186581977,
      "grad_norm": 1.015625,
      "learning_rate": 0.000433665004282126,
      "loss": 5.0624,
      "mean_token_accuracy": 0.17885658890008926,
      "num_tokens": 57018121.0,
      "step": 26025
    },
    {
      "entropy": 5.582159376144409,
      "epoch": 2.5833664152441447,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00043363988030294055,
      "loss": 5.0048,
      "mean_token_accuracy": 0.18431920260190965,
      "num_tokens": 57028127.0,
      "step": 26030
    },
    {
      "entropy": 5.59366397857666,
      "epoch": 2.583862643906312,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004336147523898279,
      "loss": 4.9832,
      "mean_token_accuracy": 0.18686074167490005,
      "num_tokens": 57038523.0,
      "step": 26035
    },
    {
      "entropy": 5.607472229003906,
      "epoch": 2.5843588725684796,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004335896205434111,
      "loss": 4.9923,
      "mean_token_accuracy": 0.18289553225040436,
      "num_tokens": 57048641.0,
      "step": 26040
    },
    {
      "entropy": 5.538111734390259,
      "epoch": 2.584855101230647,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043356448476431354,
      "loss": 4.9145,
      "mean_token_accuracy": 0.1833857998251915,
      "num_tokens": 57058794.0,
      "step": 26045
    },
    {
      "entropy": 5.614986419677734,
      "epoch": 2.5853513298928146,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004335393450531583,
      "loss": 5.0321,
      "mean_token_accuracy": 0.1798558607697487,
      "num_tokens": 57069814.0,
      "step": 26050
    },
    {
      "entropy": 5.708199977874756,
      "epoch": 2.585847558554982,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000433514201410569,
      "loss": 5.0984,
      "mean_token_accuracy": 0.1848915234208107,
      "num_tokens": 57081053.0,
      "step": 26055
    },
    {
      "entropy": 5.604388999938965,
      "epoch": 2.5863437872171495,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043348905383716905,
      "loss": 4.996,
      "mean_token_accuracy": 0.1817118600010872,
      "num_tokens": 57091386.0,
      "step": 26060
    },
    {
      "entropy": 5.585550308227539,
      "epoch": 2.586840015879317,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043346390233358213,
      "loss": 4.9661,
      "mean_token_accuracy": 0.18073562979698182,
      "num_tokens": 57102659.0,
      "step": 26065
    },
    {
      "entropy": 5.632572412490845,
      "epoch": 2.5873362445414845,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004334387469004319,
      "loss": 5.0523,
      "mean_token_accuracy": 0.18094513267278672,
      "num_tokens": 57114257.0,
      "step": 26070
    },
    {
      "entropy": 5.60391845703125,
      "epoch": 2.5878324732036524,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004334135875383421,
      "loss": 5.0533,
      "mean_token_accuracy": 0.18569535613059998,
      "num_tokens": 57125037.0,
      "step": 26075
    },
    {
      "entropy": 5.585127210617065,
      "epoch": 2.58832870186582,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004333884242479367,
      "loss": 5.0328,
      "mean_token_accuracy": 0.17596432715654373,
      "num_tokens": 57137581.0,
      "step": 26080
    },
    {
      "entropy": 5.605485153198242,
      "epoch": 2.5888249305279873,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00043336325702983966,
      "loss": 4.9858,
      "mean_token_accuracy": 0.185415019094944,
      "num_tokens": 57148507.0,
      "step": 26085
    },
    {
      "entropy": 5.599481868743896,
      "epoch": 2.589321159190155,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043333808588467504,
      "loss": 4.9459,
      "mean_token_accuracy": 0.19427523016929626,
      "num_tokens": 57159518.0,
      "step": 26090
    },
    {
      "entropy": 5.581620931625366,
      "epoch": 2.5898173878523223,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004333129108130671,
      "loss": 4.9962,
      "mean_token_accuracy": 0.1890954077243805,
      "num_tokens": 57170600.0,
      "step": 26095
    },
    {
      "entropy": 5.613749027252197,
      "epoch": 2.5903136165144898,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004332877318156402,
      "loss": 5.0166,
      "mean_token_accuracy": 0.18247200846672057,
      "num_tokens": 57181137.0,
      "step": 26100
    },
    {
      "entropy": 5.619913244247437,
      "epoch": 2.5908098451766572,
      "grad_norm": 0.9375,
      "learning_rate": 0.00043326254889301856,
      "loss": 5.0284,
      "mean_token_accuracy": 0.18105290830135345,
      "num_tokens": 57191975.0,
      "step": 26105
    },
    {
      "entropy": 5.698441410064698,
      "epoch": 2.591306073838825,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004332373620458268,
      "loss": 5.0591,
      "mean_token_accuracy": 0.1784238561987877,
      "num_tokens": 57203139.0,
      "step": 26110
    },
    {
      "entropy": 5.624229574203492,
      "epoch": 2.5918023025009926,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00043321217127468935,
      "loss": 5.0919,
      "mean_token_accuracy": 0.17494309693574905,
      "num_tokens": 57215542.0,
      "step": 26115
    },
    {
      "entropy": 5.623013067245483,
      "epoch": 2.59229853116316,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043318697658023097,
      "loss": 5.011,
      "mean_token_accuracy": 0.18505025655031204,
      "num_tokens": 57226525.0,
      "step": 26120
    },
    {
      "entropy": 5.597219514846802,
      "epoch": 2.5927947598253276,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004331617779630765,
      "loss": 4.9773,
      "mean_token_accuracy": 0.18829383254051207,
      "num_tokens": 57236120.0,
      "step": 26125
    },
    {
      "entropy": 5.508963298797608,
      "epoch": 2.593290988487495,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004331365754238507,
      "loss": 4.9412,
      "mean_token_accuracy": 0.18292835503816604,
      "num_tokens": 57246947.0,
      "step": 26130
    },
    {
      "entropy": 5.583683967590332,
      "epoch": 2.5937872171496625,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043311136896317855,
      "loss": 4.9681,
      "mean_token_accuracy": 0.1818282961845398,
      "num_tokens": 57258885.0,
      "step": 26135
    },
    {
      "entropy": 5.669933700561524,
      "epoch": 2.59428344581183,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043308615858168517,
      "loss": 5.0775,
      "mean_token_accuracy": 0.18671441078186035,
      "num_tokens": 57270047.0,
      "step": 26140
    },
    {
      "entropy": 5.671764230728149,
      "epoch": 2.5947796744739975,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004330609442799957,
      "loss": 5.1113,
      "mean_token_accuracy": 0.18158612996339799,
      "num_tokens": 57281405.0,
      "step": 26145
    },
    {
      "entropy": 5.634307050704956,
      "epoch": 2.595275903136165,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004330357260587355,
      "loss": 5.0308,
      "mean_token_accuracy": 0.1848935678601265,
      "num_tokens": 57292880.0,
      "step": 26150
    },
    {
      "entropy": 5.606462907791138,
      "epoch": 2.5957721317983324,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004330105039185297,
      "loss": 5.0018,
      "mean_token_accuracy": 0.18563878089189528,
      "num_tokens": 57304544.0,
      "step": 26155
    },
    {
      "entropy": 5.603926372528076,
      "epoch": 2.5962683604605004,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043298527786000393,
      "loss": 4.9894,
      "mean_token_accuracy": 0.1847218915820122,
      "num_tokens": 57315869.0,
      "step": 26160
    },
    {
      "entropy": 5.605328130722046,
      "epoch": 2.596764589122668,
      "grad_norm": 1.0,
      "learning_rate": 0.0004329600478837837,
      "loss": 5.0275,
      "mean_token_accuracy": 0.18461506366729735,
      "num_tokens": 57327933.0,
      "step": 26165
    },
    {
      "entropy": 5.632087135314942,
      "epoch": 2.5972608177848353,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043293481399049474,
      "loss": 5.0204,
      "mean_token_accuracy": 0.17754058986902238,
      "num_tokens": 57338941.0,
      "step": 26170
    },
    {
      "entropy": 5.629198932647705,
      "epoch": 2.597757046447003,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004329095761807626,
      "loss": 5.0171,
      "mean_token_accuracy": 0.18668123334646225,
      "num_tokens": 57350128.0,
      "step": 26175
    },
    {
      "entropy": 5.557136106491089,
      "epoch": 2.5982532751091703,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004328843344552133,
      "loss": 4.9977,
      "mean_token_accuracy": 0.18262481540441514,
      "num_tokens": 57361602.0,
      "step": 26180
    },
    {
      "entropy": 5.6945250034332275,
      "epoch": 2.5987495037713377,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004328590888144726,
      "loss": 5.0879,
      "mean_token_accuracy": 0.17418928444385529,
      "num_tokens": 57373093.0,
      "step": 26185
    },
    {
      "entropy": 5.7002311706542965,
      "epoch": 2.599245732433505,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004328338392591668,
      "loss": 5.1235,
      "mean_token_accuracy": 0.1730456084012985,
      "num_tokens": 57384799.0,
      "step": 26190
    },
    {
      "entropy": 5.555413389205933,
      "epoch": 2.599741961095673,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043280858578992185,
      "loss": 4.9551,
      "mean_token_accuracy": 0.18845021426677705,
      "num_tokens": 57395703.0,
      "step": 26195
    },
    {
      "entropy": 5.6574475288391115,
      "epoch": 2.6002381897578406,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043278332840736397,
      "loss": 5.1328,
      "mean_token_accuracy": 0.18020251542329788,
      "num_tokens": 57406725.0,
      "step": 26200
    },
    {
      "entropy": 5.619461917877198,
      "epoch": 2.600734418420008,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043275806711211957,
      "loss": 5.0012,
      "mean_token_accuracy": 0.18628361076116562,
      "num_tokens": 57417362.0,
      "step": 26205
    },
    {
      "entropy": 5.6103064060211185,
      "epoch": 2.6012306470821756,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00043273280190481503,
      "loss": 5.0125,
      "mean_token_accuracy": 0.18652498871088027,
      "num_tokens": 57429178.0,
      "step": 26210
    },
    {
      "entropy": 5.59174952507019,
      "epoch": 2.601726875744343,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043270753278607687,
      "loss": 5.0153,
      "mean_token_accuracy": 0.18329422324895858,
      "num_tokens": 57439741.0,
      "step": 26215
    },
    {
      "entropy": 5.575345087051391,
      "epoch": 2.6022231044065105,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043268225975653176,
      "loss": 5.0764,
      "mean_token_accuracy": 0.18172555714845656,
      "num_tokens": 57451754.0,
      "step": 26220
    },
    {
      "entropy": 5.661819744110107,
      "epoch": 2.602719333068678,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004326569828168064,
      "loss": 5.0317,
      "mean_token_accuracy": 0.18366955369710922,
      "num_tokens": 57463396.0,
      "step": 26225
    },
    {
      "entropy": 5.579652643203735,
      "epoch": 2.6032155617308455,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004326317019675275,
      "loss": 5.0104,
      "mean_token_accuracy": 0.18594853579998016,
      "num_tokens": 57474023.0,
      "step": 26230
    },
    {
      "entropy": 5.600180006027221,
      "epoch": 2.603711790393013,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043260641720932214,
      "loss": 5.0048,
      "mean_token_accuracy": 0.1833518475294113,
      "num_tokens": 57484751.0,
      "step": 26235
    },
    {
      "entropy": 5.602564191818237,
      "epoch": 2.6042080190551804,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043258112854281726,
      "loss": 4.9848,
      "mean_token_accuracy": 0.181386736035347,
      "num_tokens": 57496502.0,
      "step": 26240
    },
    {
      "entropy": 5.607271432876587,
      "epoch": 2.6047042477173483,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004325558359686399,
      "loss": 5.0192,
      "mean_token_accuracy": 0.1874239668250084,
      "num_tokens": 57507111.0,
      "step": 26245
    },
    {
      "entropy": 5.620730304718018,
      "epoch": 2.605200476379516,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043253053948741734,
      "loss": 5.0998,
      "mean_token_accuracy": 0.17302487641572953,
      "num_tokens": 57517844.0,
      "step": 26250
    },
    {
      "entropy": 5.6552510261535645,
      "epoch": 2.6056967050416833,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043250523909977684,
      "loss": 5.0547,
      "mean_token_accuracy": 0.18682085126638412,
      "num_tokens": 57528047.0,
      "step": 26255
    },
    {
      "entropy": 5.646855640411377,
      "epoch": 2.6061929337038507,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004324799348063457,
      "loss": 5.0293,
      "mean_token_accuracy": 0.1818633571267128,
      "num_tokens": 57538489.0,
      "step": 26260
    },
    {
      "entropy": 5.5869231700897215,
      "epoch": 2.606689162366018,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004324546266077518,
      "loss": 5.0271,
      "mean_token_accuracy": 0.1835519254207611,
      "num_tokens": 57549741.0,
      "step": 26265
    },
    {
      "entropy": 5.553083229064941,
      "epoch": 2.6071853910281857,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004324293145046222,
      "loss": 4.9531,
      "mean_token_accuracy": 0.18884621560573578,
      "num_tokens": 57559803.0,
      "step": 26270
    },
    {
      "entropy": 5.6010825634002686,
      "epoch": 2.607681619690353,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043240399849758484,
      "loss": 5.0248,
      "mean_token_accuracy": 0.18445219993591308,
      "num_tokens": 57570609.0,
      "step": 26275
    },
    {
      "entropy": 5.712609481811524,
      "epoch": 2.608177848352521,
      "grad_norm": 1.125,
      "learning_rate": 0.0004323786785872675,
      "loss": 5.0496,
      "mean_token_accuracy": 0.1852916419506073,
      "num_tokens": 57580934.0,
      "step": 26280
    },
    {
      "entropy": 5.665864419937134,
      "epoch": 2.6086740770146886,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004323533547742981,
      "loss": 5.1558,
      "mean_token_accuracy": 0.17857598066329955,
      "num_tokens": 57594671.0,
      "step": 26285
    },
    {
      "entropy": 5.672724056243896,
      "epoch": 2.609170305676856,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004323280270593046,
      "loss": 5.0803,
      "mean_token_accuracy": 0.17851771265268326,
      "num_tokens": 57605582.0,
      "step": 26290
    },
    {
      "entropy": 5.611728620529175,
      "epoch": 2.6096665343390235,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043230269544291503,
      "loss": 5.0111,
      "mean_token_accuracy": 0.17926060855388642,
      "num_tokens": 57617185.0,
      "step": 26295
    },
    {
      "entropy": 5.621560287475586,
      "epoch": 2.610162763001191,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004322773599257575,
      "loss": 4.9634,
      "mean_token_accuracy": 0.1846672773361206,
      "num_tokens": 57627205.0,
      "step": 26300
    },
    {
      "entropy": 5.55421290397644,
      "epoch": 2.6106589916633585,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043225202050846043,
      "loss": 4.9743,
      "mean_token_accuracy": 0.1863032728433609,
      "num_tokens": 57639496.0,
      "step": 26305
    },
    {
      "entropy": 5.645316314697266,
      "epoch": 2.611155220325526,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000432226677191652,
      "loss": 5.0723,
      "mean_token_accuracy": 0.17763450294733046,
      "num_tokens": 57650345.0,
      "step": 26310
    },
    {
      "entropy": 5.597909736633301,
      "epoch": 2.6116514489876934,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043220132997596084,
      "loss": 4.9686,
      "mean_token_accuracy": 0.18912501037120819,
      "num_tokens": 57661409.0,
      "step": 26315
    },
    {
      "entropy": 5.718633031845092,
      "epoch": 2.612147677649861,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004321759788620155,
      "loss": 5.1314,
      "mean_token_accuracy": 0.175704488158226,
      "num_tokens": 57672639.0,
      "step": 26320
    },
    {
      "entropy": 5.637755537033081,
      "epoch": 2.6126439063120284,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00043215062385044446,
      "loss": 5.1089,
      "mean_token_accuracy": 0.17685452997684478,
      "num_tokens": 57682249.0,
      "step": 26325
    },
    {
      "entropy": 5.5384767055511475,
      "epoch": 2.613140134974196,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043212526494187665,
      "loss": 4.9395,
      "mean_token_accuracy": 0.19167935699224473,
      "num_tokens": 57692319.0,
      "step": 26330
    },
    {
      "entropy": 5.660011100769043,
      "epoch": 2.6136363636363638,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004320999021369408,
      "loss": 5.1108,
      "mean_token_accuracy": 0.1801717147231102,
      "num_tokens": 57702334.0,
      "step": 26335
    },
    {
      "entropy": 5.622695589065552,
      "epoch": 2.6141325922985312,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043207453543626587,
      "loss": 5.0039,
      "mean_token_accuracy": 0.17977065742015838,
      "num_tokens": 57712407.0,
      "step": 26340
    },
    {
      "entropy": 5.588605070114136,
      "epoch": 2.6146288209606987,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043204916484048094,
      "loss": 4.9812,
      "mean_token_accuracy": 0.1872472271323204,
      "num_tokens": 57724252.0,
      "step": 26345
    },
    {
      "entropy": 5.594109535217285,
      "epoch": 2.615125049622866,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043202379035021515,
      "loss": 5.0458,
      "mean_token_accuracy": 0.1821434751152992,
      "num_tokens": 57735033.0,
      "step": 26350
    },
    {
      "entropy": 5.682437944412231,
      "epoch": 2.6156212782850337,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004319984119660977,
      "loss": 5.0245,
      "mean_token_accuracy": 0.18159531950950622,
      "num_tokens": 57746182.0,
      "step": 26355
    },
    {
      "entropy": 5.6042522430419925,
      "epoch": 2.616117506947201,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043197302968875795,
      "loss": 4.9712,
      "mean_token_accuracy": 0.18382258266210555,
      "num_tokens": 57756479.0,
      "step": 26360
    },
    {
      "entropy": 5.662573671340942,
      "epoch": 2.616613735609369,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004319476435188254,
      "loss": 5.0874,
      "mean_token_accuracy": 0.17670504301786422,
      "num_tokens": 57767773.0,
      "step": 26365
    },
    {
      "entropy": 5.615801191329956,
      "epoch": 2.6171099642715365,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043192225345692933,
      "loss": 4.9798,
      "mean_token_accuracy": 0.18966851085424424,
      "num_tokens": 57777331.0,
      "step": 26370
    },
    {
      "entropy": 5.558163166046143,
      "epoch": 2.617606192933704,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043189685950369963,
      "loss": 4.9937,
      "mean_token_accuracy": 0.1853550523519516,
      "num_tokens": 57788538.0,
      "step": 26375
    },
    {
      "entropy": 5.653491401672364,
      "epoch": 2.6181024215958715,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043187146165976577,
      "loss": 5.1501,
      "mean_token_accuracy": 0.17445281893014908,
      "num_tokens": 57799983.0,
      "step": 26380
    },
    {
      "entropy": 5.6188922882080075,
      "epoch": 2.618598650258039,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004318460599257578,
      "loss": 4.9752,
      "mean_token_accuracy": 0.18129622638225557,
      "num_tokens": 57810195.0,
      "step": 26385
    },
    {
      "entropy": 5.565737867355347,
      "epoch": 2.6190948789202064,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00043182065430230545,
      "loss": 4.9185,
      "mean_token_accuracy": 0.1923673763871193,
      "num_tokens": 57821930.0,
      "step": 26390
    },
    {
      "entropy": 5.611873817443848,
      "epoch": 2.619591107582374,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004317952447900388,
      "loss": 4.9722,
      "mean_token_accuracy": 0.1843975231051445,
      "num_tokens": 57834272.0,
      "step": 26395
    },
    {
      "entropy": 5.608261394500732,
      "epoch": 2.6200873362445414,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000431769831389588,
      "loss": 5.0055,
      "mean_token_accuracy": 0.18197612315416337,
      "num_tokens": 57845491.0,
      "step": 26400
    },
    {
      "entropy": 5.59079327583313,
      "epoch": 2.620583564906709,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004317444141015831,
      "loss": 5.0796,
      "mean_token_accuracy": 0.18275501132011412,
      "num_tokens": 57857816.0,
      "step": 26405
    },
    {
      "entropy": 5.637110471725464,
      "epoch": 2.6210797935688763,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043171899292665457,
      "loss": 5.0465,
      "mean_token_accuracy": 0.1781070977449417,
      "num_tokens": 57867769.0,
      "step": 26410
    },
    {
      "entropy": 5.6772431373596195,
      "epoch": 2.621576022231044,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004316935678654327,
      "loss": 5.1314,
      "mean_token_accuracy": 0.17217508852481841,
      "num_tokens": 57878027.0,
      "step": 26415
    },
    {
      "entropy": 5.612048149108887,
      "epoch": 2.6220722508932117,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043166813891854796,
      "loss": 5.0017,
      "mean_token_accuracy": 0.18342747241258622,
      "num_tokens": 57887938.0,
      "step": 26420
    },
    {
      "entropy": 5.653272390365601,
      "epoch": 2.622568479555379,
      "grad_norm": 1.0,
      "learning_rate": 0.000431642706086631,
      "loss": 4.9996,
      "mean_token_accuracy": 0.1937391921877861,
      "num_tokens": 57899375.0,
      "step": 26425
    },
    {
      "entropy": 5.651577138900757,
      "epoch": 2.6230647082175467,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043161726937031243,
      "loss": 5.0701,
      "mean_token_accuracy": 0.17937386631965638,
      "num_tokens": 57910393.0,
      "step": 26430
    },
    {
      "entropy": 5.598186922073364,
      "epoch": 2.623560936879714,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000431591828770223,
      "loss": 5.0578,
      "mean_token_accuracy": 0.18571165651082994,
      "num_tokens": 57922236.0,
      "step": 26435
    },
    {
      "entropy": 5.619221639633179,
      "epoch": 2.6240571655418816,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004315663842869938,
      "loss": 5.0342,
      "mean_token_accuracy": 0.18691803365945817,
      "num_tokens": 57931804.0,
      "step": 26440
    },
    {
      "entropy": 5.633609008789063,
      "epoch": 2.624553394204049,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043154093592125554,
      "loss": 5.0121,
      "mean_token_accuracy": 0.18181462287902833,
      "num_tokens": 57943259.0,
      "step": 26445
    },
    {
      "entropy": 5.570058584213257,
      "epoch": 2.625049622866217,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043151548367363943,
      "loss": 4.9246,
      "mean_token_accuracy": 0.1852260947227478,
      "num_tokens": 57952945.0,
      "step": 26450
    },
    {
      "entropy": 5.578832626342773,
      "epoch": 2.6255458515283845,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043149002754477655,
      "loss": 5.0077,
      "mean_token_accuracy": 0.18275612443685532,
      "num_tokens": 57964030.0,
      "step": 26455
    },
    {
      "entropy": 5.568461322784424,
      "epoch": 2.626042080190552,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004314645675352982,
      "loss": 4.9821,
      "mean_token_accuracy": 0.19326643496751786,
      "num_tokens": 57974888.0,
      "step": 26460
    },
    {
      "entropy": 5.676108455657959,
      "epoch": 2.6265383088527194,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004314391036458357,
      "loss": 5.0155,
      "mean_token_accuracy": 0.18703674226999284,
      "num_tokens": 57984922.0,
      "step": 26465
    },
    {
      "entropy": 5.566687774658203,
      "epoch": 2.627034537514887,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043141363587702054,
      "loss": 4.995,
      "mean_token_accuracy": 0.1846248686313629,
      "num_tokens": 57994881.0,
      "step": 26470
    },
    {
      "entropy": 5.610242938995361,
      "epoch": 2.6275307661770544,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004313881642294843,
      "loss": 5.0748,
      "mean_token_accuracy": 0.1823548823595047,
      "num_tokens": 58005335.0,
      "step": 26475
    },
    {
      "entropy": 5.620892143249511,
      "epoch": 2.628026994839222,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004313626887038585,
      "loss": 4.9543,
      "mean_token_accuracy": 0.19060834348201752,
      "num_tokens": 58016692.0,
      "step": 26480
    },
    {
      "entropy": 5.601221084594727,
      "epoch": 2.6285232235013893,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000431337209300775,
      "loss": 5.0548,
      "mean_token_accuracy": 0.18559518307447434,
      "num_tokens": 58028282.0,
      "step": 26485
    },
    {
      "entropy": 5.564561128616333,
      "epoch": 2.629019452163557,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004313117260208656,
      "loss": 4.9828,
      "mean_token_accuracy": 0.18707109242677689,
      "num_tokens": 58040108.0,
      "step": 26490
    },
    {
      "entropy": 5.581929492950439,
      "epoch": 2.6295156808257243,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004312862388647622,
      "loss": 4.9954,
      "mean_token_accuracy": 0.18484814167022706,
      "num_tokens": 58050156.0,
      "step": 26495
    },
    {
      "entropy": 5.584506702423096,
      "epoch": 2.6300119094878918,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043126074783309687,
      "loss": 4.9902,
      "mean_token_accuracy": 0.18150296360254287,
      "num_tokens": 58060929.0,
      "step": 26500
    },
    {
      "entropy": 5.6159087181091305,
      "epoch": 2.6305081381500597,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004312352529265016,
      "loss": 5.0439,
      "mean_token_accuracy": 0.176796555519104,
      "num_tokens": 58071684.0,
      "step": 26505
    },
    {
      "entropy": 5.633403873443603,
      "epoch": 2.631004366812227,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043120975414560886,
      "loss": 5.0692,
      "mean_token_accuracy": 0.1796170577406883,
      "num_tokens": 58082730.0,
      "step": 26510
    },
    {
      "entropy": 5.6776539325714115,
      "epoch": 2.6315005954743946,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043118425149105073,
      "loss": 5.1251,
      "mean_token_accuracy": 0.1798018380999565,
      "num_tokens": 58095278.0,
      "step": 26515
    },
    {
      "entropy": 5.724976682662964,
      "epoch": 2.631996824136562,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004311587449634597,
      "loss": 5.1849,
      "mean_token_accuracy": 0.17691968753933907,
      "num_tokens": 58107816.0,
      "step": 26520
    },
    {
      "entropy": 5.67445216178894,
      "epoch": 2.6324930527987296,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004311332345634683,
      "loss": 5.0996,
      "mean_token_accuracy": 0.17680224776268005,
      "num_tokens": 58117779.0,
      "step": 26525
    },
    {
      "entropy": 5.646727657318115,
      "epoch": 2.632989281460897,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043110772029170916,
      "loss": 5.0037,
      "mean_token_accuracy": 0.18940244168043135,
      "num_tokens": 58128665.0,
      "step": 26530
    },
    {
      "entropy": 5.674915790557861,
      "epoch": 2.633485510123065,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043108220214881494,
      "loss": 5.0759,
      "mean_token_accuracy": 0.1778171956539154,
      "num_tokens": 58139658.0,
      "step": 26535
    },
    {
      "entropy": 5.712341976165772,
      "epoch": 2.6339817387852325,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004310566801354184,
      "loss": 5.0627,
      "mean_token_accuracy": 0.1794685035943985,
      "num_tokens": 58149980.0,
      "step": 26540
    },
    {
      "entropy": 5.593758153915405,
      "epoch": 2.6344779674474,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043103115425215246,
      "loss": 4.9496,
      "mean_token_accuracy": 0.18666605204343795,
      "num_tokens": 58159809.0,
      "step": 26545
    },
    {
      "entropy": 5.579456567764282,
      "epoch": 2.6349741961095674,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043100562449965016,
      "loss": 5.022,
      "mean_token_accuracy": 0.18227325528860092,
      "num_tokens": 58169912.0,
      "step": 26550
    },
    {
      "entropy": 5.644398546218872,
      "epoch": 2.635470424771735,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043098009087854446,
      "loss": 5.0718,
      "mean_token_accuracy": 0.18029511719942093,
      "num_tokens": 58179949.0,
      "step": 26555
    },
    {
      "entropy": 5.599224615097046,
      "epoch": 2.6359666534339024,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043095455338946863,
      "loss": 5.0172,
      "mean_token_accuracy": 0.18796958327293395,
      "num_tokens": 58189847.0,
      "step": 26560
    },
    {
      "entropy": 5.579065370559692,
      "epoch": 2.63646288209607,
      "grad_norm": 1.03125,
      "learning_rate": 0.000430929012033056,
      "loss": 4.9324,
      "mean_token_accuracy": 0.18802070766687393,
      "num_tokens": 58201514.0,
      "step": 26565
    },
    {
      "entropy": 5.61249270439148,
      "epoch": 2.6369591107582373,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043090346680993983,
      "loss": 5.0211,
      "mean_token_accuracy": 0.18926084488630296,
      "num_tokens": 58211940.0,
      "step": 26570
    },
    {
      "entropy": 5.671613502502441,
      "epoch": 2.6374553394204048,
      "grad_norm": 1.125,
      "learning_rate": 0.0004308779177207536,
      "loss": 5.0396,
      "mean_token_accuracy": 0.17877403050661086,
      "num_tokens": 58222260.0,
      "step": 26575
    },
    {
      "entropy": 5.574507522583008,
      "epoch": 2.6379515680825723,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004308523647661309,
      "loss": 4.9921,
      "mean_token_accuracy": 0.18372125625610353,
      "num_tokens": 58232649.0,
      "step": 26580
    },
    {
      "entropy": 5.569385623931884,
      "epoch": 2.6384477967447397,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043082680794670535,
      "loss": 4.9335,
      "mean_token_accuracy": 0.1892600804567337,
      "num_tokens": 58243373.0,
      "step": 26585
    },
    {
      "entropy": 5.551293420791626,
      "epoch": 2.6389440254069076,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043080124726311084,
      "loss": 5.0225,
      "mean_token_accuracy": 0.18626249879598616,
      "num_tokens": 58253863.0,
      "step": 26590
    },
    {
      "entropy": 5.587895107269287,
      "epoch": 2.639440254069075,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004307756827159811,
      "loss": 5.0941,
      "mean_token_accuracy": 0.1889141544699669,
      "num_tokens": 58265326.0,
      "step": 26595
    },
    {
      "entropy": 5.645430040359497,
      "epoch": 2.6399364827312426,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043075011430594994,
      "loss": 5.0581,
      "mean_token_accuracy": 0.1866261199116707,
      "num_tokens": 58276006.0,
      "step": 26600
    },
    {
      "entropy": 5.605237579345703,
      "epoch": 2.64043271139341,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043072454203365175,
      "loss": 4.9132,
      "mean_token_accuracy": 0.1879487082362175,
      "num_tokens": 58285932.0,
      "step": 26605
    },
    {
      "entropy": 5.574843835830689,
      "epoch": 2.6409289400555775,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004306989658997204,
      "loss": 5.0109,
      "mean_token_accuracy": 0.18562981933355333,
      "num_tokens": 58297174.0,
      "step": 26610
    },
    {
      "entropy": 5.6590807914733885,
      "epoch": 2.641425168717745,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043067338590479023,
      "loss": 5.1338,
      "mean_token_accuracy": 0.17455951422452926,
      "num_tokens": 58308509.0,
      "step": 26615
    },
    {
      "entropy": 5.62342643737793,
      "epoch": 2.641921397379913,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043064780204949543,
      "loss": 5.0393,
      "mean_token_accuracy": 0.17511515319347382,
      "num_tokens": 58318449.0,
      "step": 26620
    },
    {
      "entropy": 5.663921642303467,
      "epoch": 2.6424176260420804,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004306222143344706,
      "loss": 5.0039,
      "mean_token_accuracy": 0.18436934798955917,
      "num_tokens": 58329759.0,
      "step": 26625
    },
    {
      "entropy": 5.598606967926026,
      "epoch": 2.642913854704248,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004305966227603503,
      "loss": 4.9833,
      "mean_token_accuracy": 0.1842211365699768,
      "num_tokens": 58339363.0,
      "step": 26630
    },
    {
      "entropy": 5.5661520004272464,
      "epoch": 2.6434100833664154,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004305710273277689,
      "loss": 5.0267,
      "mean_token_accuracy": 0.17713566720485688,
      "num_tokens": 58349482.0,
      "step": 26635
    },
    {
      "entropy": 5.652234220504761,
      "epoch": 2.643906312028583,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043054542803736126,
      "loss": 5.043,
      "mean_token_accuracy": 0.18166089057922363,
      "num_tokens": 58359879.0,
      "step": 26640
    },
    {
      "entropy": 5.681516695022583,
      "epoch": 2.6444025406907503,
      "grad_norm": 1.125,
      "learning_rate": 0.00043051982488976216,
      "loss": 5.0924,
      "mean_token_accuracy": 0.17325122952461242,
      "num_tokens": 58370475.0,
      "step": 26645
    },
    {
      "entropy": 5.638138771057129,
      "epoch": 2.644898769352918,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004304942178856066,
      "loss": 5.0317,
      "mean_token_accuracy": 0.18268633484840394,
      "num_tokens": 58381992.0,
      "step": 26650
    },
    {
      "entropy": 5.6303197860717775,
      "epoch": 2.6453949980150853,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043046860702552944,
      "loss": 4.9633,
      "mean_token_accuracy": 0.1925205945968628,
      "num_tokens": 58391616.0,
      "step": 26655
    },
    {
      "entropy": 5.663478088378906,
      "epoch": 2.6458912266772527,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004304429923101658,
      "loss": 5.0046,
      "mean_token_accuracy": 0.18007027208805085,
      "num_tokens": 58401800.0,
      "step": 26660
    },
    {
      "entropy": 5.604512023925781,
      "epoch": 2.64638745533942,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000430417373740151,
      "loss": 4.9775,
      "mean_token_accuracy": 0.18565159887075425,
      "num_tokens": 58413376.0,
      "step": 26665
    },
    {
      "entropy": 5.492952919006347,
      "epoch": 2.6468836840015877,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004303917513161201,
      "loss": 4.9508,
      "mean_token_accuracy": 0.18811583518981934,
      "num_tokens": 58423772.0,
      "step": 26670
    },
    {
      "entropy": 5.613302516937256,
      "epoch": 2.6473799126637556,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004303661250387088,
      "loss": 5.0679,
      "mean_token_accuracy": 0.1746920019388199,
      "num_tokens": 58434838.0,
      "step": 26675
    },
    {
      "entropy": 5.671370792388916,
      "epoch": 2.647876141325923,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004303404949085522,
      "loss": 5.084,
      "mean_token_accuracy": 0.1804327294230461,
      "num_tokens": 58445036.0,
      "step": 26680
    },
    {
      "entropy": 5.632583951950073,
      "epoch": 2.6483723699880906,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004303148609262861,
      "loss": 5.0032,
      "mean_token_accuracy": 0.18581858426332473,
      "num_tokens": 58455558.0,
      "step": 26685
    },
    {
      "entropy": 5.621822166442871,
      "epoch": 2.648868598650258,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004302892230925462,
      "loss": 5.0474,
      "mean_token_accuracy": 0.17653827667236327,
      "num_tokens": 58467716.0,
      "step": 26690
    },
    {
      "entropy": 5.579214954376221,
      "epoch": 2.6493648273124255,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00043026358140796806,
      "loss": 5.0163,
      "mean_token_accuracy": 0.17938157320022582,
      "num_tokens": 58479421.0,
      "step": 26695
    },
    {
      "entropy": 5.530521440505981,
      "epoch": 2.649861055974593,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004302379358731878,
      "loss": 5.0158,
      "mean_token_accuracy": 0.18639944940805436,
      "num_tokens": 58490115.0,
      "step": 26700
    },
    {
      "entropy": 5.603570413589478,
      "epoch": 2.6503572846367605,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004302122864888411,
      "loss": 4.9754,
      "mean_token_accuracy": 0.18610296696424483,
      "num_tokens": 58501711.0,
      "step": 26705
    },
    {
      "entropy": 5.658417272567749,
      "epoch": 2.6508535132989284,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004301866332555643,
      "loss": 5.0003,
      "mean_token_accuracy": 0.19129825830459596,
      "num_tokens": 58513441.0,
      "step": 26710
    },
    {
      "entropy": 5.606355857849121,
      "epoch": 2.651349741961096,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043016097617399327,
      "loss": 4.9943,
      "mean_token_accuracy": 0.18708724528551102,
      "num_tokens": 58523940.0,
      "step": 26715
    },
    {
      "entropy": 5.58193941116333,
      "epoch": 2.6518459706232633,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004301353152447644,
      "loss": 5.0524,
      "mean_token_accuracy": 0.18123987466096877,
      "num_tokens": 58536354.0,
      "step": 26720
    },
    {
      "entropy": 5.651720428466797,
      "epoch": 2.652342199285431,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043010965046851404,
      "loss": 5.053,
      "mean_token_accuracy": 0.1820437118411064,
      "num_tokens": 58547096.0,
      "step": 26725
    },
    {
      "entropy": 5.624227809906006,
      "epoch": 2.6528384279475983,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043008398184587847,
      "loss": 5.063,
      "mean_token_accuracy": 0.18470564633607864,
      "num_tokens": 58557951.0,
      "step": 26730
    },
    {
      "entropy": 5.568663024902344,
      "epoch": 2.6533346566097658,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043005830937749447,
      "loss": 4.9746,
      "mean_token_accuracy": 0.1884079873561859,
      "num_tokens": 58569384.0,
      "step": 26735
    },
    {
      "entropy": 5.631934547424317,
      "epoch": 2.6538308852719332,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004300326330639984,
      "loss": 5.0234,
      "mean_token_accuracy": 0.18148189187049865,
      "num_tokens": 58580308.0,
      "step": 26740
    },
    {
      "entropy": 5.607954788208008,
      "epoch": 2.6543271139341007,
      "grad_norm": 1.125,
      "learning_rate": 0.0004300069529060272,
      "loss": 5.0391,
      "mean_token_accuracy": 0.17917954474687575,
      "num_tokens": 58591780.0,
      "step": 26745
    },
    {
      "entropy": 5.57782621383667,
      "epoch": 2.654823342596268,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042998126890421754,
      "loss": 5.0496,
      "mean_token_accuracy": 0.18205637782812117,
      "num_tokens": 58601477.0,
      "step": 26750
    },
    {
      "entropy": 5.615290784835816,
      "epoch": 2.6553195712584357,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042995558105920627,
      "loss": 5.0179,
      "mean_token_accuracy": 0.1856437772512436,
      "num_tokens": 58612531.0,
      "step": 26755
    },
    {
      "entropy": 5.625716066360473,
      "epoch": 2.6558157999206036,
      "grad_norm": 0.984375,
      "learning_rate": 0.00042992988937163066,
      "loss": 5.0656,
      "mean_token_accuracy": 0.17846055775880815,
      "num_tokens": 58624951.0,
      "step": 26760
    },
    {
      "entropy": 5.623323392868042,
      "epoch": 2.656312028582771,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004299041938421275,
      "loss": 5.1064,
      "mean_token_accuracy": 0.17972584664821625,
      "num_tokens": 58636484.0,
      "step": 26765
    },
    {
      "entropy": 5.635343742370606,
      "epoch": 2.6568082572449385,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042987849447133423,
      "loss": 5.0176,
      "mean_token_accuracy": 0.17897830456495284,
      "num_tokens": 58647386.0,
      "step": 26770
    },
    {
      "entropy": 5.648524522781372,
      "epoch": 2.657304485907106,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004298527912598879,
      "loss": 5.096,
      "mean_token_accuracy": 0.1812959209084511,
      "num_tokens": 58658656.0,
      "step": 26775
    },
    {
      "entropy": 5.615984201431274,
      "epoch": 2.6578007145692735,
      "grad_norm": 1.0,
      "learning_rate": 0.0004298270842084261,
      "loss": 4.9749,
      "mean_token_accuracy": 0.18822856694459916,
      "num_tokens": 58669901.0,
      "step": 26780
    },
    {
      "entropy": 5.572210264205933,
      "epoch": 2.658296943231441,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042980137331758626,
      "loss": 5.0716,
      "mean_token_accuracy": 0.17581339478492736,
      "num_tokens": 58680849.0,
      "step": 26785
    },
    {
      "entropy": 5.588628149032592,
      "epoch": 2.6587931718936084,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042977565858800593,
      "loss": 4.9954,
      "mean_token_accuracy": 0.18232445865869523,
      "num_tokens": 58690642.0,
      "step": 26790
    },
    {
      "entropy": 5.599898099899292,
      "epoch": 2.6592894005557763,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042974994002032275,
      "loss": 4.9469,
      "mean_token_accuracy": 0.188630910217762,
      "num_tokens": 58700480.0,
      "step": 26795
    },
    {
      "entropy": 5.642024898529053,
      "epoch": 2.659785629217944,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042972421761517456,
      "loss": 5.0829,
      "mean_token_accuracy": 0.1757476046681404,
      "num_tokens": 58712443.0,
      "step": 26800
    },
    {
      "entropy": 5.561808061599732,
      "epoch": 2.6602818578801113,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004296984913731991,
      "loss": 4.9797,
      "mean_token_accuracy": 0.18213419914245604,
      "num_tokens": 58721172.0,
      "step": 26805
    },
    {
      "entropy": 5.612880516052246,
      "epoch": 2.6607780865422788,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004296727612950344,
      "loss": 5.0142,
      "mean_token_accuracy": 0.18662223517894744,
      "num_tokens": 58731060.0,
      "step": 26810
    },
    {
      "entropy": 5.607250452041626,
      "epoch": 2.6612743152044462,
      "grad_norm": 1.125,
      "learning_rate": 0.0004296470273813186,
      "loss": 5.0227,
      "mean_token_accuracy": 0.18771505802869798,
      "num_tokens": 58742189.0,
      "step": 26815
    },
    {
      "entropy": 5.673286056518554,
      "epoch": 2.6617705438666137,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004296212896326896,
      "loss": 5.05,
      "mean_token_accuracy": 0.18135914504528045,
      "num_tokens": 58753617.0,
      "step": 26820
    },
    {
      "entropy": 5.608369636535644,
      "epoch": 2.662266772528781,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042959554804978596,
      "loss": 5.0354,
      "mean_token_accuracy": 0.18591851890087127,
      "num_tokens": 58765299.0,
      "step": 26825
    },
    {
      "entropy": 5.649770450592041,
      "epoch": 2.6627630011909487,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042956980263324576,
      "loss": 4.9736,
      "mean_token_accuracy": 0.185293510556221,
      "num_tokens": 58775280.0,
      "step": 26830
    },
    {
      "entropy": 5.635244703292846,
      "epoch": 2.663259229853116,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004295440533837075,
      "loss": 4.9719,
      "mean_token_accuracy": 0.1882303163409233,
      "num_tokens": 58786078.0,
      "step": 26835
    },
    {
      "entropy": 5.610131359100341,
      "epoch": 2.6637554585152836,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00042951830030180983,
      "loss": 5.0664,
      "mean_token_accuracy": 0.18155399411916734,
      "num_tokens": 58796665.0,
      "step": 26840
    },
    {
      "entropy": 5.654162216186523,
      "epoch": 2.6642516871774515,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042949254338819123,
      "loss": 5.0688,
      "mean_token_accuracy": 0.18265877962112426,
      "num_tokens": 58808150.0,
      "step": 26845
    },
    {
      "entropy": 5.679914951324463,
      "epoch": 2.664747915839619,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042946678264349036,
      "loss": 4.9986,
      "mean_token_accuracy": 0.19301116317510605,
      "num_tokens": 58818278.0,
      "step": 26850
    },
    {
      "entropy": 5.6015888214111325,
      "epoch": 2.6652441445017865,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042944101806834627,
      "loss": 5.0481,
      "mean_token_accuracy": 0.1781671479344368,
      "num_tokens": 58828847.0,
      "step": 26855
    },
    {
      "entropy": 5.576524877548218,
      "epoch": 2.665740373163954,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004294152496633977,
      "loss": 5.008,
      "mean_token_accuracy": 0.18218873292207718,
      "num_tokens": 58839698.0,
      "step": 26860
    },
    {
      "entropy": 5.576607656478882,
      "epoch": 2.6662366018261214,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042938947742928357,
      "loss": 4.979,
      "mean_token_accuracy": 0.18223538398742675,
      "num_tokens": 58850794.0,
      "step": 26865
    },
    {
      "entropy": 5.641123867034912,
      "epoch": 2.666732830488289,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00042936370136664314,
      "loss": 5.0582,
      "mean_token_accuracy": 0.17693158984184265,
      "num_tokens": 58861590.0,
      "step": 26870
    },
    {
      "entropy": 5.656180715560913,
      "epoch": 2.6672290591504564,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004293379214761156,
      "loss": 5.0996,
      "mean_token_accuracy": 0.17577488869428634,
      "num_tokens": 58873796.0,
      "step": 26875
    },
    {
      "entropy": 5.641378164291382,
      "epoch": 2.6677252878126243,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004293121377583401,
      "loss": 5.0512,
      "mean_token_accuracy": 0.1827706977725029,
      "num_tokens": 58885296.0,
      "step": 26880
    },
    {
      "entropy": 5.597051382064819,
      "epoch": 2.6682215164747918,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004292863502139561,
      "loss": 5.0903,
      "mean_token_accuracy": 0.1800532177090645,
      "num_tokens": 58897764.0,
      "step": 26885
    },
    {
      "entropy": 5.643507242202759,
      "epoch": 2.6687177451369593,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004292605588436031,
      "loss": 5.0498,
      "mean_token_accuracy": 0.17635983675718309,
      "num_tokens": 58908964.0,
      "step": 26890
    },
    {
      "entropy": 5.531443166732788,
      "epoch": 2.6692139737991267,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042923476364792063,
      "loss": 4.9151,
      "mean_token_accuracy": 0.18714308440685273,
      "num_tokens": 58919152.0,
      "step": 26895
    },
    {
      "entropy": 5.524743366241455,
      "epoch": 2.669710202461294,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004292089646275484,
      "loss": 5.0475,
      "mean_token_accuracy": 0.18242392987012862,
      "num_tokens": 58928849.0,
      "step": 26900
    },
    {
      "entropy": 5.67999963760376,
      "epoch": 2.6702064311234617,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004291831617831261,
      "loss": 5.1602,
      "mean_token_accuracy": 0.17605929374694823,
      "num_tokens": 58941560.0,
      "step": 26905
    },
    {
      "entropy": 5.704592227935791,
      "epoch": 2.670702659785629,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042915735511529356,
      "loss": 5.1034,
      "mean_token_accuracy": 0.1740887388586998,
      "num_tokens": 58953365.0,
      "step": 26910
    },
    {
      "entropy": 5.6304315567016605,
      "epoch": 2.6711988884477966,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004291315446246909,
      "loss": 5.0141,
      "mean_token_accuracy": 0.1865653708577156,
      "num_tokens": 58964602.0,
      "step": 26915
    },
    {
      "entropy": 5.649563455581665,
      "epoch": 2.671695117109964,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042910573031195804,
      "loss": 5.0358,
      "mean_token_accuracy": 0.18009912967681885,
      "num_tokens": 58976154.0,
      "step": 26920
    },
    {
      "entropy": 5.635859441757202,
      "epoch": 2.6721913457721316,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004290799121777351,
      "loss": 5.0638,
      "mean_token_accuracy": 0.18599973917007445,
      "num_tokens": 58985989.0,
      "step": 26925
    },
    {
      "entropy": 5.686270236968994,
      "epoch": 2.672687574434299,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004290540902226624,
      "loss": 5.1111,
      "mean_token_accuracy": 0.1743547648191452,
      "num_tokens": 58997080.0,
      "step": 26930
    },
    {
      "entropy": 5.625657463073731,
      "epoch": 2.673183803096467,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004290282644473802,
      "loss": 5.0355,
      "mean_token_accuracy": 0.1843112215399742,
      "num_tokens": 59008683.0,
      "step": 26935
    },
    {
      "entropy": 5.6258203983306885,
      "epoch": 2.6736800317586344,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004290024348525289,
      "loss": 4.9635,
      "mean_token_accuracy": 0.1928812548518181,
      "num_tokens": 59019019.0,
      "step": 26940
    },
    {
      "entropy": 5.604146194458008,
      "epoch": 2.674176260420802,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004289766014387492,
      "loss": 5.0221,
      "mean_token_accuracy": 0.18435828238725663,
      "num_tokens": 59029706.0,
      "step": 26945
    },
    {
      "entropy": 5.616050863265992,
      "epoch": 2.6746724890829694,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004289507642066813,
      "loss": 4.95,
      "mean_token_accuracy": 0.18558347076177598,
      "num_tokens": 59040208.0,
      "step": 26950
    },
    {
      "entropy": 5.556113386154175,
      "epoch": 2.675168717745137,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004289249231569664,
      "loss": 5.0021,
      "mean_token_accuracy": 0.19311386495828628,
      "num_tokens": 59050909.0,
      "step": 26955
    },
    {
      "entropy": 5.543575191497803,
      "epoch": 2.6756649464073043,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042889907829024506,
      "loss": 4.9723,
      "mean_token_accuracy": 0.1874529853463173,
      "num_tokens": 59059834.0,
      "step": 26960
    },
    {
      "entropy": 5.653904676437378,
      "epoch": 2.6761611750694723,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042887322960715813,
      "loss": 5.0555,
      "mean_token_accuracy": 0.1772822320461273,
      "num_tokens": 59071348.0,
      "step": 26965
    },
    {
      "entropy": 5.613383674621582,
      "epoch": 2.6766574037316397,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004288473771083468,
      "loss": 5.0148,
      "mean_token_accuracy": 0.1794414058327675,
      "num_tokens": 59083053.0,
      "step": 26970
    },
    {
      "entropy": 5.65954556465149,
      "epoch": 2.677153632393807,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004288215207944519,
      "loss": 5.0953,
      "mean_token_accuracy": 0.1733824282884598,
      "num_tokens": 59093400.0,
      "step": 26975
    },
    {
      "entropy": 5.615987014770508,
      "epoch": 2.6776498610559747,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042879566066611474,
      "loss": 5.0603,
      "mean_token_accuracy": 0.1792568862438202,
      "num_tokens": 59103653.0,
      "step": 26980
    },
    {
      "entropy": 5.636954975128174,
      "epoch": 2.678146089718142,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004287697967239767,
      "loss": 5.0558,
      "mean_token_accuracy": 0.17804358452558516,
      "num_tokens": 59116660.0,
      "step": 26985
    },
    {
      "entropy": 5.537959718704224,
      "epoch": 2.6786423183803096,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000428743928968679,
      "loss": 4.9469,
      "mean_token_accuracy": 0.18944600522518157,
      "num_tokens": 59127424.0,
      "step": 26990
    },
    {
      "entropy": 5.6092428207397464,
      "epoch": 2.679138547042477,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042871805740086317,
      "loss": 5.0625,
      "mean_token_accuracy": 0.1845138520002365,
      "num_tokens": 59137566.0,
      "step": 26995
    },
    {
      "entropy": 5.586385011672974,
      "epoch": 2.6796347757046446,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004286921820211707,
      "loss": 4.9715,
      "mean_token_accuracy": 0.1848578780889511,
      "num_tokens": 59148404.0,
      "step": 27000
    },
    {
      "epoch": 2.6796347757046446,
      "eval_entropy": 5.3754649493098965,
      "eval_loss": 5.142398357391357,
      "eval_mean_token_accuracy": 0.18315990927681614,
      "eval_num_tokens": 59148404.0,
      "eval_runtime": 20.3045,
      "eval_samples_per_second": 1658.205,
      "eval_steps_per_second": 207.294,
      "step": 27000
    },
    {
      "entropy": 5.5433656692504885,
      "epoch": 2.680131004366812,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042866630283024327,
      "loss": 5.0287,
      "mean_token_accuracy": 0.17688723802566528,
      "num_tokens": 59158097.0,
      "step": 27005
    },
    {
      "entropy": 5.506463861465454,
      "epoch": 2.6806272330289795,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004286404198287227,
      "loss": 4.8893,
      "mean_token_accuracy": 0.19748805165290834,
      "num_tokens": 59168940.0,
      "step": 27010
    },
    {
      "entropy": 5.629644584655762,
      "epoch": 2.681123461691147,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004286145330172508,
      "loss": 5.03,
      "mean_token_accuracy": 0.18165774047374725,
      "num_tokens": 59180231.0,
      "step": 27015
    },
    {
      "entropy": 5.698243618011475,
      "epoch": 2.681619690353315,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004285886423964695,
      "loss": 5.1166,
      "mean_token_accuracy": 0.17195165753364564,
      "num_tokens": 59191565.0,
      "step": 27020
    },
    {
      "entropy": 5.607916259765625,
      "epoch": 2.6821159190154824,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004285627479670208,
      "loss": 5.0953,
      "mean_token_accuracy": 0.17935949862003325,
      "num_tokens": 59203394.0,
      "step": 27025
    },
    {
      "entropy": 5.528347539901733,
      "epoch": 2.68261214767765,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004285368497295468,
      "loss": 4.9492,
      "mean_token_accuracy": 0.18047820925712585,
      "num_tokens": 59213987.0,
      "step": 27030
    },
    {
      "entropy": 5.58594913482666,
      "epoch": 2.6831083763398174,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004285109476846898,
      "loss": 5.0003,
      "mean_token_accuracy": 0.18787861317396165,
      "num_tokens": 59225825.0,
      "step": 27035
    },
    {
      "entropy": 5.546665334701538,
      "epoch": 2.683604605001985,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004284850418330921,
      "loss": 4.9741,
      "mean_token_accuracy": 0.19124819934368134,
      "num_tokens": 59235978.0,
      "step": 27040
    },
    {
      "entropy": 5.678858661651612,
      "epoch": 2.6841008336641523,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042845913217539613,
      "loss": 5.1489,
      "mean_token_accuracy": 0.17519418895244598,
      "num_tokens": 59247505.0,
      "step": 27045
    },
    {
      "entropy": 5.59184079170227,
      "epoch": 2.6845970623263202,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042843321871224426,
      "loss": 4.9547,
      "mean_token_accuracy": 0.18520616888999938,
      "num_tokens": 59258401.0,
      "step": 27050
    },
    {
      "entropy": 5.6120024681091305,
      "epoch": 2.6850932909884877,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042840730144427927,
      "loss": 5.006,
      "mean_token_accuracy": 0.18425827324390412,
      "num_tokens": 59268068.0,
      "step": 27055
    },
    {
      "entropy": 5.592672300338745,
      "epoch": 2.685589519650655,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042838138037214375,
      "loss": 5.0195,
      "mean_token_accuracy": 0.18247674256563187,
      "num_tokens": 59279567.0,
      "step": 27060
    },
    {
      "entropy": 5.611498880386352,
      "epoch": 2.6860857483128227,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042835545549648054,
      "loss": 4.9503,
      "mean_token_accuracy": 0.18466290980577468,
      "num_tokens": 59289862.0,
      "step": 27065
    },
    {
      "entropy": 5.609827327728271,
      "epoch": 2.68658197697499,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042832952681793235,
      "loss": 4.9888,
      "mean_token_accuracy": 0.1900678649544716,
      "num_tokens": 59301304.0,
      "step": 27070
    },
    {
      "entropy": 5.589920616149902,
      "epoch": 2.6870782056371576,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042830359433714243,
      "loss": 5.0292,
      "mean_token_accuracy": 0.18565578013658524,
      "num_tokens": 59312184.0,
      "step": 27075
    },
    {
      "entropy": 5.562552976608276,
      "epoch": 2.687574434299325,
      "grad_norm": 1.015625,
      "learning_rate": 0.00042827765805475367,
      "loss": 4.9124,
      "mean_token_accuracy": 0.1858239069581032,
      "num_tokens": 59322688.0,
      "step": 27080
    },
    {
      "entropy": 5.58199520111084,
      "epoch": 2.6880706629614926,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042825171797140926,
      "loss": 4.9684,
      "mean_token_accuracy": 0.19157496392726897,
      "num_tokens": 59333349.0,
      "step": 27085
    },
    {
      "entropy": 5.555906295776367,
      "epoch": 2.68856689162366,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004282257740877525,
      "loss": 4.9743,
      "mean_token_accuracy": 0.19036263823509217,
      "num_tokens": 59344473.0,
      "step": 27090
    },
    {
      "entropy": 5.602223873138428,
      "epoch": 2.6890631202858275,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004281998264044267,
      "loss": 5.0507,
      "mean_token_accuracy": 0.18245213925838472,
      "num_tokens": 59355993.0,
      "step": 27095
    },
    {
      "entropy": 5.681280660629272,
      "epoch": 2.689559348947995,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004281738749220753,
      "loss": 5.0985,
      "mean_token_accuracy": 0.17890757620334624,
      "num_tokens": 59367182.0,
      "step": 27100
    },
    {
      "entropy": 5.646443843841553,
      "epoch": 2.690055577610163,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042814791964134184,
      "loss": 5.0679,
      "mean_token_accuracy": 0.17794458270072938,
      "num_tokens": 59379235.0,
      "step": 27105
    },
    {
      "entropy": 5.6045591831207275,
      "epoch": 2.6905518062723304,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042812196056287,
      "loss": 5.0134,
      "mean_token_accuracy": 0.1907332420349121,
      "num_tokens": 59388514.0,
      "step": 27110
    },
    {
      "entropy": 5.5873809337615965,
      "epoch": 2.691048034934498,
      "grad_norm": 1.015625,
      "learning_rate": 0.00042809599768730357,
      "loss": 5.0363,
      "mean_token_accuracy": 0.18245162665843964,
      "num_tokens": 59399496.0,
      "step": 27115
    },
    {
      "entropy": 5.579776239395142,
      "epoch": 2.6915442635966653,
      "grad_norm": 1.0,
      "learning_rate": 0.0004280700310152861,
      "loss": 5.0161,
      "mean_token_accuracy": 0.18616604655981064,
      "num_tokens": 59411986.0,
      "step": 27120
    },
    {
      "entropy": 5.695464372634888,
      "epoch": 2.692040492258833,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004280440605474619,
      "loss": 5.2084,
      "mean_token_accuracy": 0.17147329896688462,
      "num_tokens": 59424232.0,
      "step": 27125
    },
    {
      "entropy": 5.670924186706543,
      "epoch": 2.6925367209210003,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004280180862844747,
      "loss": 5.059,
      "mean_token_accuracy": 0.17807400971651077,
      "num_tokens": 59435201.0,
      "step": 27130
    },
    {
      "entropy": 5.6323775291442875,
      "epoch": 2.693032949583168,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042799210822696865,
      "loss": 5.0985,
      "mean_token_accuracy": 0.1749937802553177,
      "num_tokens": 59445552.0,
      "step": 27135
    },
    {
      "entropy": 5.699649381637573,
      "epoch": 2.6935291782453357,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000427966126375588,
      "loss": 5.0829,
      "mean_token_accuracy": 0.178082075715065,
      "num_tokens": 59455634.0,
      "step": 27140
    },
    {
      "entropy": 5.6921515464782715,
      "epoch": 2.694025406907503,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004279401407309771,
      "loss": 5.0591,
      "mean_token_accuracy": 0.18345436155796052,
      "num_tokens": 59467955.0,
      "step": 27145
    },
    {
      "entropy": 5.6287407875061035,
      "epoch": 2.6945216355696706,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004279141512937802,
      "loss": 5.0479,
      "mean_token_accuracy": 0.1822266936302185,
      "num_tokens": 59479566.0,
      "step": 27150
    },
    {
      "entropy": 5.518826198577881,
      "epoch": 2.695017864231838,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004278881580646418,
      "loss": 4.992,
      "mean_token_accuracy": 0.18783901780843734,
      "num_tokens": 59489967.0,
      "step": 27155
    },
    {
      "entropy": 5.609582901000977,
      "epoch": 2.6955140928940056,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004278621610442066,
      "loss": 5.0494,
      "mean_token_accuracy": 0.1788359686732292,
      "num_tokens": 59501640.0,
      "step": 27160
    },
    {
      "entropy": 5.604799461364746,
      "epoch": 2.696010321556173,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004278361602331192,
      "loss": 4.9836,
      "mean_token_accuracy": 0.1942935422062874,
      "num_tokens": 59512767.0,
      "step": 27165
    },
    {
      "entropy": 5.6147315979003904,
      "epoch": 2.6965065502183405,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042781015563202434,
      "loss": 4.9597,
      "mean_token_accuracy": 0.18769053965806962,
      "num_tokens": 59522188.0,
      "step": 27170
    },
    {
      "entropy": 5.6126384258270265,
      "epoch": 2.697002778880508,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004277841472415669,
      "loss": 4.9868,
      "mean_token_accuracy": 0.18905743807554246,
      "num_tokens": 59534871.0,
      "step": 27175
    },
    {
      "entropy": 5.692520523071289,
      "epoch": 2.6974990075426755,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004277581350623918,
      "loss": 5.1076,
      "mean_token_accuracy": 0.16885697692632676,
      "num_tokens": 59546858.0,
      "step": 27180
    },
    {
      "entropy": 5.5897492408752445,
      "epoch": 2.697995236204843,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004277321190951441,
      "loss": 4.965,
      "mean_token_accuracy": 0.18461227118968965,
      "num_tokens": 59557022.0,
      "step": 27185
    },
    {
      "entropy": 5.569014024734497,
      "epoch": 2.698491464867011,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042770609934046904,
      "loss": 4.9898,
      "mean_token_accuracy": 0.19153722524642944,
      "num_tokens": 59567564.0,
      "step": 27190
    },
    {
      "entropy": 5.706486177444458,
      "epoch": 2.6989876935291783,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042768007579901176,
      "loss": 5.1308,
      "mean_token_accuracy": 0.1716386318206787,
      "num_tokens": 59579122.0,
      "step": 27195
    },
    {
      "entropy": 5.7112874507904055,
      "epoch": 2.699483922191346,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042765404847141755,
      "loss": 5.0838,
      "mean_token_accuracy": 0.1865792065858841,
      "num_tokens": 59590477.0,
      "step": 27200
    },
    {
      "entropy": 5.630859851837158,
      "epoch": 2.6999801508535133,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004276280173583319,
      "loss": 4.9791,
      "mean_token_accuracy": 0.18462797701358796,
      "num_tokens": 59601217.0,
      "step": 27205
    },
    {
      "entropy": 5.572267055511475,
      "epoch": 2.7004763795156808,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042760198246040025,
      "loss": 4.9985,
      "mean_token_accuracy": 0.18656110763549805,
      "num_tokens": 59612271.0,
      "step": 27210
    },
    {
      "entropy": 5.657285594940186,
      "epoch": 2.7009726081778482,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004275759437782683,
      "loss": 5.0798,
      "mean_token_accuracy": 0.17912353873252868,
      "num_tokens": 59623948.0,
      "step": 27215
    },
    {
      "entropy": 5.582306528091431,
      "epoch": 2.701468836840016,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00042754990131258175,
      "loss": 4.9455,
      "mean_token_accuracy": 0.18733389377593995,
      "num_tokens": 59635720.0,
      "step": 27220
    },
    {
      "entropy": 5.6553467273712155,
      "epoch": 2.7019650655021836,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004275238550639863,
      "loss": 5.052,
      "mean_token_accuracy": 0.18440381139516832,
      "num_tokens": 59645171.0,
      "step": 27225
    },
    {
      "entropy": 5.675833082199096,
      "epoch": 2.702461294164351,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000427497805033128,
      "loss": 5.139,
      "mean_token_accuracy": 0.17492150515317917,
      "num_tokens": 59655266.0,
      "step": 27230
    },
    {
      "entropy": 5.59214415550232,
      "epoch": 2.7029575228265186,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004274717512206527,
      "loss": 5.0236,
      "mean_token_accuracy": 0.18714299649000168,
      "num_tokens": 59667055.0,
      "step": 27235
    },
    {
      "entropy": 5.576248216629028,
      "epoch": 2.703453751488686,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004274456936272065,
      "loss": 4.9873,
      "mean_token_accuracy": 0.18679561614990234,
      "num_tokens": 59678621.0,
      "step": 27240
    },
    {
      "entropy": 5.591213703155518,
      "epoch": 2.7039499801508535,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042741963225343554,
      "loss": 4.9734,
      "mean_token_accuracy": 0.19097980707883835,
      "num_tokens": 59690514.0,
      "step": 27245
    },
    {
      "entropy": 5.679116487503052,
      "epoch": 2.704446208813021,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004273935670999861,
      "loss": 5.0512,
      "mean_token_accuracy": 0.17929594218730927,
      "num_tokens": 59701271.0,
      "step": 27250
    },
    {
      "entropy": 5.631002521514892,
      "epoch": 2.7049424374751885,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042736749816750476,
      "loss": 5.1238,
      "mean_token_accuracy": 0.17323877960443496,
      "num_tokens": 59712787.0,
      "step": 27255
    },
    {
      "entropy": 5.543184423446656,
      "epoch": 2.705438666137356,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004273414254566376,
      "loss": 4.9462,
      "mean_token_accuracy": 0.1875184714794159,
      "num_tokens": 59722787.0,
      "step": 27260
    },
    {
      "entropy": 5.625221681594849,
      "epoch": 2.7059348947995234,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004273153489680315,
      "loss": 5.0213,
      "mean_token_accuracy": 0.18471357673406602,
      "num_tokens": 59733302.0,
      "step": 27265
    },
    {
      "entropy": 5.58648042678833,
      "epoch": 2.706431123461691,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042728926870233287,
      "loss": 5.0334,
      "mean_token_accuracy": 0.17918039560317994,
      "num_tokens": 59743290.0,
      "step": 27270
    },
    {
      "entropy": 5.667661714553833,
      "epoch": 2.706927352123859,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042726318466018855,
      "loss": 5.098,
      "mean_token_accuracy": 0.18062619864940643,
      "num_tokens": 59754145.0,
      "step": 27275
    },
    {
      "entropy": 5.666721296310425,
      "epoch": 2.7074235807860263,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004272370968422454,
      "loss": 5.05,
      "mean_token_accuracy": 0.18542413115501405,
      "num_tokens": 59764075.0,
      "step": 27280
    },
    {
      "entropy": 5.605688810348511,
      "epoch": 2.7079198094481938,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00042721100524915017,
      "loss": 5.064,
      "mean_token_accuracy": 0.18161149173974991,
      "num_tokens": 59775495.0,
      "step": 27285
    },
    {
      "entropy": 5.546535444259644,
      "epoch": 2.7084160381103612,
      "grad_norm": 1.015625,
      "learning_rate": 0.00042718490988155014,
      "loss": 4.9032,
      "mean_token_accuracy": 0.20499961972236633,
      "num_tokens": 59787006.0,
      "step": 27290
    },
    {
      "entropy": 5.614944362640381,
      "epoch": 2.7089122667725287,
      "grad_norm": 1.125,
      "learning_rate": 0.00042715881074009216,
      "loss": 5.0329,
      "mean_token_accuracy": 0.17987522184848787,
      "num_tokens": 59797680.0,
      "step": 27295
    },
    {
      "entropy": 5.579052925109863,
      "epoch": 2.709408495434696,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042713270782542364,
      "loss": 5.0145,
      "mean_token_accuracy": 0.18076301962137223,
      "num_tokens": 59809666.0,
      "step": 27300
    },
    {
      "entropy": 5.578750467300415,
      "epoch": 2.7099047240968637,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004271066011381917,
      "loss": 4.9807,
      "mean_token_accuracy": 0.18569900542497636,
      "num_tokens": 59819889.0,
      "step": 27305
    },
    {
      "entropy": 5.569624185562134,
      "epoch": 2.7104009527590316,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004270804906790438,
      "loss": 4.9429,
      "mean_token_accuracy": 0.19161884784698485,
      "num_tokens": 59830344.0,
      "step": 27310
    },
    {
      "entropy": 5.572251415252685,
      "epoch": 2.710897181421199,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004270543764486275,
      "loss": 4.9737,
      "mean_token_accuracy": 0.19120494574308394,
      "num_tokens": 59843498.0,
      "step": 27315
    },
    {
      "entropy": 5.660264396667481,
      "epoch": 2.7113934100833665,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004270282584475903,
      "loss": 5.039,
      "mean_token_accuracy": 0.1856732502579689,
      "num_tokens": 59854633.0,
      "step": 27320
    },
    {
      "entropy": 5.598152780532837,
      "epoch": 2.711889638745534,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042700213667657983,
      "loss": 5.1084,
      "mean_token_accuracy": 0.17954568862915038,
      "num_tokens": 59865925.0,
      "step": 27325
    },
    {
      "entropy": 5.574401044845581,
      "epoch": 2.7123858674077015,
      "grad_norm": 1.234375,
      "learning_rate": 0.000426976011136244,
      "loss": 4.9932,
      "mean_token_accuracy": 0.1855478048324585,
      "num_tokens": 59877196.0,
      "step": 27330
    },
    {
      "entropy": 5.663136911392212,
      "epoch": 2.712882096069869,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042694988182723047,
      "loss": 5.0789,
      "mean_token_accuracy": 0.17733621895313262,
      "num_tokens": 59887411.0,
      "step": 27335
    },
    {
      "entropy": 5.591849946975708,
      "epoch": 2.7133783247320364,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00042692374875018735,
      "loss": 5.02,
      "mean_token_accuracy": 0.19179949760437012,
      "num_tokens": 59899123.0,
      "step": 27340
    },
    {
      "entropy": 5.558926439285278,
      "epoch": 2.713874553394204,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042689761190576263,
      "loss": 4.9647,
      "mean_token_accuracy": 0.19419369846582413,
      "num_tokens": 59910326.0,
      "step": 27345
    },
    {
      "entropy": 5.6282310962677,
      "epoch": 2.7143707820563714,
      "grad_norm": 1.125,
      "learning_rate": 0.00042687147129460447,
      "loss": 5.0709,
      "mean_token_accuracy": 0.1801816463470459,
      "num_tokens": 59920722.0,
      "step": 27350
    },
    {
      "entropy": 5.646857643127442,
      "epoch": 2.714867010718539,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042684532691736104,
      "loss": 5.0364,
      "mean_token_accuracy": 0.18036623150110245,
      "num_tokens": 59932568.0,
      "step": 27355
    },
    {
      "entropy": 5.614872455596924,
      "epoch": 2.715363239380707,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042681917877468076,
      "loss": 5.0925,
      "mean_token_accuracy": 0.17709190547466278,
      "num_tokens": 59943805.0,
      "step": 27360
    },
    {
      "entropy": 5.609921264648437,
      "epoch": 2.7158594680428743,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042679302686721187,
      "loss": 4.9745,
      "mean_token_accuracy": 0.19219402819871903,
      "num_tokens": 59954455.0,
      "step": 27365
    },
    {
      "entropy": 5.627721786499023,
      "epoch": 2.7163556967050417,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004267668711956031,
      "loss": 4.9958,
      "mean_token_accuracy": 0.18588606417179107,
      "num_tokens": 59966722.0,
      "step": 27370
    },
    {
      "entropy": 5.601606607437134,
      "epoch": 2.716851925367209,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042674071176050296,
      "loss": 5.0038,
      "mean_token_accuracy": 0.18512384593486786,
      "num_tokens": 59976702.0,
      "step": 27375
    },
    {
      "entropy": 5.586561584472657,
      "epoch": 2.7173481540293767,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042671454856256,
      "loss": 4.9877,
      "mean_token_accuracy": 0.18496938198804855,
      "num_tokens": 59987038.0,
      "step": 27380
    },
    {
      "entropy": 5.641171932220459,
      "epoch": 2.717844382691544,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004266883816024233,
      "loss": 5.0462,
      "mean_token_accuracy": 0.18524792045354843,
      "num_tokens": 59996755.0,
      "step": 27385
    },
    {
      "entropy": 5.638524103164673,
      "epoch": 2.7183406113537116,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004266622108807415,
      "loss": 4.9952,
      "mean_token_accuracy": 0.1793353483080864,
      "num_tokens": 60006419.0,
      "step": 27390
    },
    {
      "entropy": 5.6336061477661135,
      "epoch": 2.7188368400158796,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004266360363981637,
      "loss": 5.0759,
      "mean_token_accuracy": 0.1853100448846817,
      "num_tokens": 60017793.0,
      "step": 27395
    },
    {
      "entropy": 5.571135139465332,
      "epoch": 2.719333068678047,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042660985815533905,
      "loss": 4.947,
      "mean_token_accuracy": 0.18834673762321472,
      "num_tokens": 60029392.0,
      "step": 27400
    },
    {
      "entropy": 5.612802267074585,
      "epoch": 2.7198292973402145,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042658367615291644,
      "loss": 5.0885,
      "mean_token_accuracy": 0.18273095041513443,
      "num_tokens": 60040467.0,
      "step": 27405
    },
    {
      "entropy": 5.6115881443023685,
      "epoch": 2.720325526002382,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004265574903915454,
      "loss": 5.0187,
      "mean_token_accuracy": 0.18095260262489318,
      "num_tokens": 60051583.0,
      "step": 27410
    },
    {
      "entropy": 5.629722929000854,
      "epoch": 2.7208217546645495,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004265313008718751,
      "loss": 5.0219,
      "mean_token_accuracy": 0.18260036408901215,
      "num_tokens": 60062218.0,
      "step": 27415
    },
    {
      "entropy": 5.593433809280396,
      "epoch": 2.721317983326717,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042650510759455516,
      "loss": 5.0057,
      "mean_token_accuracy": 0.18775643408298492,
      "num_tokens": 60074455.0,
      "step": 27420
    },
    {
      "entropy": 5.61510853767395,
      "epoch": 2.7218142119888844,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042647891056023495,
      "loss": 4.9949,
      "mean_token_accuracy": 0.17925068289041518,
      "num_tokens": 60085442.0,
      "step": 27425
    },
    {
      "entropy": 5.657591009140015,
      "epoch": 2.722310440651052,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042645270976956426,
      "loss": 5.0691,
      "mean_token_accuracy": 0.1836300879716873,
      "num_tokens": 60096441.0,
      "step": 27430
    },
    {
      "entropy": 5.68871660232544,
      "epoch": 2.7228066693132194,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042642650522319253,
      "loss": 5.0293,
      "mean_token_accuracy": 0.18722456246614455,
      "num_tokens": 60107399.0,
      "step": 27435
    },
    {
      "entropy": 5.629233741760254,
      "epoch": 2.723302897975387,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004264002969217699,
      "loss": 4.9949,
      "mean_token_accuracy": 0.18900584429502487,
      "num_tokens": 60117568.0,
      "step": 27440
    },
    {
      "entropy": 5.651252222061157,
      "epoch": 2.7237991266375547,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004263740848659461,
      "loss": 5.0634,
      "mean_token_accuracy": 0.18344887495040893,
      "num_tokens": 60127214.0,
      "step": 27445
    },
    {
      "entropy": 5.599195528030395,
      "epoch": 2.724295355299722,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00042634786905637123,
      "loss": 4.9935,
      "mean_token_accuracy": 0.18340423405170442,
      "num_tokens": 60138167.0,
      "step": 27450
    },
    {
      "entropy": 5.595355415344239,
      "epoch": 2.7247915839618897,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004263216494936953,
      "loss": 4.9817,
      "mean_token_accuracy": 0.18522557467222214,
      "num_tokens": 60148841.0,
      "step": 27455
    },
    {
      "entropy": 5.585482597351074,
      "epoch": 2.725287812624057,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042629542617856855,
      "loss": 5.0542,
      "mean_token_accuracy": 0.18300525844097137,
      "num_tokens": 60159512.0,
      "step": 27460
    },
    {
      "entropy": 5.573078918457031,
      "epoch": 2.7257840412862246,
      "grad_norm": 1.125,
      "learning_rate": 0.0004262691991116412,
      "loss": 5.0092,
      "mean_token_accuracy": 0.178063702583313,
      "num_tokens": 60170463.0,
      "step": 27465
    },
    {
      "entropy": 5.51702094078064,
      "epoch": 2.726280269948392,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004262429682935637,
      "loss": 4.8751,
      "mean_token_accuracy": 0.1908646732568741,
      "num_tokens": 60180431.0,
      "step": 27470
    },
    {
      "entropy": 5.6064173698425295,
      "epoch": 2.7267764986105596,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004262167337249865,
      "loss": 5.039,
      "mean_token_accuracy": 0.18532251417636872,
      "num_tokens": 60190410.0,
      "step": 27475
    },
    {
      "entropy": 5.625122117996216,
      "epoch": 2.7272727272727275,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004261904954065601,
      "loss": 4.9744,
      "mean_token_accuracy": 0.1867181956768036,
      "num_tokens": 60201909.0,
      "step": 27480
    },
    {
      "entropy": 5.671002817153931,
      "epoch": 2.727768955934895,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004261642533389352,
      "loss": 4.987,
      "mean_token_accuracy": 0.18742512464523314,
      "num_tokens": 60212577.0,
      "step": 27485
    },
    {
      "entropy": 5.592304277420044,
      "epoch": 2.7282651845970625,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042613800752276264,
      "loss": 5.0734,
      "mean_token_accuracy": 0.18112415373325347,
      "num_tokens": 60222615.0,
      "step": 27490
    },
    {
      "entropy": 5.606505918502807,
      "epoch": 2.72876141325923,
      "grad_norm": 0.984375,
      "learning_rate": 0.00042611175795869306,
      "loss": 5.0953,
      "mean_token_accuracy": 0.18306862264871598,
      "num_tokens": 60234338.0,
      "step": 27495
    },
    {
      "entropy": 5.6730384349823,
      "epoch": 2.7292576419213974,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004260855046473775,
      "loss": 4.977,
      "mean_token_accuracy": 0.18461020290851593,
      "num_tokens": 60244781.0,
      "step": 27500
    },
    {
      "entropy": 5.636649799346924,
      "epoch": 2.729753870583565,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042605924758946706,
      "loss": 5.0439,
      "mean_token_accuracy": 0.18205252587795256,
      "num_tokens": 60255150.0,
      "step": 27505
    },
    {
      "entropy": 5.570816612243652,
      "epoch": 2.7302500992457324,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042603298678561277,
      "loss": 5.0182,
      "mean_token_accuracy": 0.18281850665807725,
      "num_tokens": 60266073.0,
      "step": 27510
    },
    {
      "entropy": 5.642455339431763,
      "epoch": 2.7307463279079,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004260067222364659,
      "loss": 5.0565,
      "mean_token_accuracy": 0.17956717759370805,
      "num_tokens": 60277520.0,
      "step": 27515
    },
    {
      "entropy": 5.62664155960083,
      "epoch": 2.7312425565700673,
      "grad_norm": 0.984375,
      "learning_rate": 0.00042598045394267757,
      "loss": 5.037,
      "mean_token_accuracy": 0.18240980058908463,
      "num_tokens": 60288530.0,
      "step": 27520
    },
    {
      "entropy": 5.6168910503387455,
      "epoch": 2.731738785232235,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00042595418190489944,
      "loss": 4.9726,
      "mean_token_accuracy": 0.18819703310728073,
      "num_tokens": 60299926.0,
      "step": 27525
    },
    {
      "entropy": 5.660991239547729,
      "epoch": 2.7322350138944023,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00042592790612378277,
      "loss": 5.1476,
      "mean_token_accuracy": 0.17642896622419357,
      "num_tokens": 60312129.0,
      "step": 27530
    },
    {
      "entropy": 5.648805284500122,
      "epoch": 2.73273124255657,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042590162659997933,
      "loss": 5.0462,
      "mean_token_accuracy": 0.1765955999493599,
      "num_tokens": 60322405.0,
      "step": 27535
    },
    {
      "entropy": 5.675673341751098,
      "epoch": 2.7332274712187377,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004258753433341407,
      "loss": 5.0395,
      "mean_token_accuracy": 0.18028332740068437,
      "num_tokens": 60333112.0,
      "step": 27540
    },
    {
      "entropy": 5.644305944442749,
      "epoch": 2.733723699880905,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042584905632691867,
      "loss": 5.0054,
      "mean_token_accuracy": 0.18678477853536607,
      "num_tokens": 60343500.0,
      "step": 27545
    },
    {
      "entropy": 5.579323816299438,
      "epoch": 2.7342199285430726,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004258227655789651,
      "loss": 5.0103,
      "mean_token_accuracy": 0.18044755905866622,
      "num_tokens": 60354711.0,
      "step": 27550
    },
    {
      "entropy": 5.582811737060547,
      "epoch": 2.73471615720524,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004257964710909318,
      "loss": 5.0315,
      "mean_token_accuracy": 0.17886197119951247,
      "num_tokens": 60365064.0,
      "step": 27555
    },
    {
      "entropy": 5.6294948101043705,
      "epoch": 2.7352123858674076,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042577017286347116,
      "loss": 5.0532,
      "mean_token_accuracy": 0.1791486144065857,
      "num_tokens": 60376425.0,
      "step": 27560
    },
    {
      "entropy": 5.640968751907349,
      "epoch": 2.7357086145295755,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042574387089723504,
      "loss": 4.943,
      "mean_token_accuracy": 0.18689461052417755,
      "num_tokens": 60386745.0,
      "step": 27565
    },
    {
      "entropy": 5.600793266296387,
      "epoch": 2.736204843191743,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042571756519287567,
      "loss": 5.0,
      "mean_token_accuracy": 0.18125646114349364,
      "num_tokens": 60397038.0,
      "step": 27570
    },
    {
      "entropy": 5.552832794189453,
      "epoch": 2.7367010718539104,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042569125575104557,
      "loss": 5.0159,
      "mean_token_accuracy": 0.18392987102270125,
      "num_tokens": 60406955.0,
      "step": 27575
    },
    {
      "entropy": 5.6023496150970455,
      "epoch": 2.737197300516078,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042566494257239695,
      "loss": 5.0093,
      "mean_token_accuracy": 0.18417284190654754,
      "num_tokens": 60418245.0,
      "step": 27580
    },
    {
      "entropy": 5.643684244155883,
      "epoch": 2.7376935291782454,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004256386256575824,
      "loss": 5.0361,
      "mean_token_accuracy": 0.17585655599832534,
      "num_tokens": 60429310.0,
      "step": 27585
    },
    {
      "entropy": 5.628964614868164,
      "epoch": 2.738189757840413,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004256123050072546,
      "loss": 5.0635,
      "mean_token_accuracy": 0.18344567716121674,
      "num_tokens": 60441355.0,
      "step": 27590
    },
    {
      "entropy": 5.715586423873901,
      "epoch": 2.7386859865025803,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00042558598062206616,
      "loss": 5.2005,
      "mean_token_accuracy": 0.17644235938787461,
      "num_tokens": 60452853.0,
      "step": 27595
    },
    {
      "entropy": 5.633006477355957,
      "epoch": 2.739182215164748,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042555965250266984,
      "loss": 5.0106,
      "mean_token_accuracy": 0.18371304720640183,
      "num_tokens": 60462934.0,
      "step": 27600
    },
    {
      "entropy": 5.662494134902954,
      "epoch": 2.7396784438269153,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042553332064971866,
      "loss": 5.0933,
      "mean_token_accuracy": 0.1754643827676773,
      "num_tokens": 60475575.0,
      "step": 27605
    },
    {
      "entropy": 5.538043546676636,
      "epoch": 2.7401746724890828,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042550698506386536,
      "loss": 4.8997,
      "mean_token_accuracy": 0.1989493489265442,
      "num_tokens": 60486495.0,
      "step": 27610
    },
    {
      "entropy": 5.593028545379639,
      "epoch": 2.7406709011512502,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042548064574576323,
      "loss": 4.995,
      "mean_token_accuracy": 0.19008367210626603,
      "num_tokens": 60496806.0,
      "step": 27615
    },
    {
      "entropy": 5.587985324859619,
      "epoch": 2.741167129813418,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004254543026960654,
      "loss": 4.9868,
      "mean_token_accuracy": 0.1856980100274086,
      "num_tokens": 60508905.0,
      "step": 27620
    },
    {
      "entropy": 5.56434907913208,
      "epoch": 2.7416633584755856,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004254279559154249,
      "loss": 4.972,
      "mean_token_accuracy": 0.1840238556265831,
      "num_tokens": 60519603.0,
      "step": 27625
    },
    {
      "entropy": 5.618446254730225,
      "epoch": 2.742159587137753,
      "grad_norm": 1.0,
      "learning_rate": 0.00042540160540449536,
      "loss": 5.0479,
      "mean_token_accuracy": 0.17903169691562654,
      "num_tokens": 60530977.0,
      "step": 27630
    },
    {
      "entropy": 5.624520301818848,
      "epoch": 2.7426558157999206,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004253752511639301,
      "loss": 5.0575,
      "mean_token_accuracy": 0.1766205310821533,
      "num_tokens": 60541354.0,
      "step": 27635
    },
    {
      "entropy": 5.593533849716186,
      "epoch": 2.743152044462088,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004253488931943826,
      "loss": 5.0316,
      "mean_token_accuracy": 0.18703942745923996,
      "num_tokens": 60552531.0,
      "step": 27640
    },
    {
      "entropy": 5.5686863422393795,
      "epoch": 2.7436482731242555,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042532253149650653,
      "loss": 5.0032,
      "mean_token_accuracy": 0.18979061990976334,
      "num_tokens": 60562552.0,
      "step": 27645
    },
    {
      "entropy": 5.634743785858154,
      "epoch": 2.7441445017864234,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004252961660709556,
      "loss": 5.0279,
      "mean_token_accuracy": 0.18048362731933593,
      "num_tokens": 60572849.0,
      "step": 27650
    },
    {
      "entropy": 5.611437177658081,
      "epoch": 2.744640730448591,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042526979691838353,
      "loss": 4.9808,
      "mean_token_accuracy": 0.18515777140855788,
      "num_tokens": 60584095.0,
      "step": 27655
    },
    {
      "entropy": 5.636131668090821,
      "epoch": 2.7451369591107584,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042524342403944446,
      "loss": 5.062,
      "mean_token_accuracy": 0.18051713705062866,
      "num_tokens": 60594353.0,
      "step": 27660
    },
    {
      "entropy": 5.645056819915771,
      "epoch": 2.745633187772926,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00042521704743479213,
      "loss": 5.0569,
      "mean_token_accuracy": 0.17917411625385285,
      "num_tokens": 60604558.0,
      "step": 27665
    },
    {
      "entropy": 5.590495300292969,
      "epoch": 2.7461294164350933,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042519066710508064,
      "loss": 5.0042,
      "mean_token_accuracy": 0.1807366818189621,
      "num_tokens": 60615670.0,
      "step": 27670
    },
    {
      "entropy": 5.549590921401977,
      "epoch": 2.746625645097261,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042516428305096435,
      "loss": 4.9484,
      "mean_token_accuracy": 0.18771874606609346,
      "num_tokens": 60626452.0,
      "step": 27675
    },
    {
      "entropy": 5.607533407211304,
      "epoch": 2.7471218737594283,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004251378952730973,
      "loss": 5.0246,
      "mean_token_accuracy": 0.1833631217479706,
      "num_tokens": 60637122.0,
      "step": 27680
    },
    {
      "entropy": 5.623420143127442,
      "epoch": 2.7476181024215958,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004251115037721341,
      "loss": 5.0078,
      "mean_token_accuracy": 0.1911241590976715,
      "num_tokens": 60648835.0,
      "step": 27685
    },
    {
      "entropy": 5.644138669967651,
      "epoch": 2.7481143310837632,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042508510854872905,
      "loss": 5.0484,
      "mean_token_accuracy": 0.17923367619514466,
      "num_tokens": 60658537.0,
      "step": 27690
    },
    {
      "entropy": 5.641906452178955,
      "epoch": 2.7486105597459307,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042505870960353665,
      "loss": 5.0265,
      "mean_token_accuracy": 0.18923610001802443,
      "num_tokens": 60670450.0,
      "step": 27695
    },
    {
      "entropy": 5.685038757324219,
      "epoch": 2.749106788408098,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004250323069372116,
      "loss": 5.0395,
      "mean_token_accuracy": 0.18870363086462022,
      "num_tokens": 60680539.0,
      "step": 27700
    },
    {
      "entropy": 5.643326330184936,
      "epoch": 2.749603017070266,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042500590055040866,
      "loss": 5.1097,
      "mean_token_accuracy": 0.179111111164093,
      "num_tokens": 60693106.0,
      "step": 27705
    },
    {
      "entropy": 5.70630202293396,
      "epoch": 2.7500992457324336,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042497949044378267,
      "loss": 5.0814,
      "mean_token_accuracy": 0.17979989498853682,
      "num_tokens": 60703509.0,
      "step": 27710
    },
    {
      "entropy": 5.657323122024536,
      "epoch": 2.750595474394601,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042495307661798837,
      "loss": 5.0699,
      "mean_token_accuracy": 0.1769983649253845,
      "num_tokens": 60714113.0,
      "step": 27715
    },
    {
      "entropy": 5.6666731357574465,
      "epoch": 2.7510917030567685,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042492665907368094,
      "loss": 5.1148,
      "mean_token_accuracy": 0.1762701079249382,
      "num_tokens": 60726826.0,
      "step": 27720
    },
    {
      "entropy": 5.6214845180511475,
      "epoch": 2.751587931718936,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042490023781151546,
      "loss": 5.0466,
      "mean_token_accuracy": 0.18183017075061797,
      "num_tokens": 60738443.0,
      "step": 27725
    },
    {
      "entropy": 5.606932544708252,
      "epoch": 2.7520841603811035,
      "grad_norm": 1.125,
      "learning_rate": 0.00042487381283214706,
      "loss": 4.996,
      "mean_token_accuracy": 0.18725059777498246,
      "num_tokens": 60748865.0,
      "step": 27730
    },
    {
      "entropy": 5.644655799865722,
      "epoch": 2.7525803890432714,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004248473841362311,
      "loss": 5.0389,
      "mean_token_accuracy": 0.1857530802488327,
      "num_tokens": 60760001.0,
      "step": 27735
    },
    {
      "entropy": 5.680937147140503,
      "epoch": 2.753076617705439,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042482095172442286,
      "loss": 5.0727,
      "mean_token_accuracy": 0.18307537883520125,
      "num_tokens": 60772003.0,
      "step": 27740
    },
    {
      "entropy": 5.57878270149231,
      "epoch": 2.7535728463676064,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004247945155973778,
      "loss": 4.9546,
      "mean_token_accuracy": 0.1890112802386284,
      "num_tokens": 60783167.0,
      "step": 27745
    },
    {
      "entropy": 5.60573935508728,
      "epoch": 2.754069075029774,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004247680757557516,
      "loss": 5.0539,
      "mean_token_accuracy": 0.18707671761512756,
      "num_tokens": 60793917.0,
      "step": 27750
    },
    {
      "entropy": 5.699059915542603,
      "epoch": 2.7545653036919413,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042474163220019987,
      "loss": 5.1501,
      "mean_token_accuracy": 0.17729664891958236,
      "num_tokens": 60805166.0,
      "step": 27755
    },
    {
      "entropy": 5.721861505508423,
      "epoch": 2.7550615323541088,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042471518493137836,
      "loss": 5.0542,
      "mean_token_accuracy": 0.18474994450807572,
      "num_tokens": 60816984.0,
      "step": 27760
    },
    {
      "entropy": 5.577034187316895,
      "epoch": 2.7555577610162763,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004246887339499427,
      "loss": 4.9936,
      "mean_token_accuracy": 0.18638442754745482,
      "num_tokens": 60829481.0,
      "step": 27765
    },
    {
      "entropy": 5.6162824630737305,
      "epoch": 2.7560539896784437,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042466227925654917,
      "loss": 4.9923,
      "mean_token_accuracy": 0.18728916496038436,
      "num_tokens": 60839376.0,
      "step": 27770
    },
    {
      "entropy": 5.566398239135742,
      "epoch": 2.756550218340611,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004246358208518535,
      "loss": 5.0269,
      "mean_token_accuracy": 0.17810164839029313,
      "num_tokens": 60850383.0,
      "step": 27775
    },
    {
      "entropy": 5.584314680099487,
      "epoch": 2.7570464470027787,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042460935873651196,
      "loss": 4.9963,
      "mean_token_accuracy": 0.18513453602790833,
      "num_tokens": 60861249.0,
      "step": 27780
    },
    {
      "entropy": 5.62355785369873,
      "epoch": 2.757542675664946,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042458289291118066,
      "loss": 5.0291,
      "mean_token_accuracy": 0.18342422991991042,
      "num_tokens": 60872220.0,
      "step": 27785
    },
    {
      "entropy": 5.604116773605346,
      "epoch": 2.758038904327114,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042455642337651595,
      "loss": 5.0319,
      "mean_token_accuracy": 0.18161050230264664,
      "num_tokens": 60882006.0,
      "step": 27790
    },
    {
      "entropy": 5.6076277732849125,
      "epoch": 2.7585351329892815,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004245299501331743,
      "loss": 4.944,
      "mean_token_accuracy": 0.19380569010972976,
      "num_tokens": 60891732.0,
      "step": 27795
    },
    {
      "entropy": 5.643096113204956,
      "epoch": 2.759031361651449,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042450347318181194,
      "loss": 5.0716,
      "mean_token_accuracy": 0.18056030571460724,
      "num_tokens": 60902990.0,
      "step": 27800
    },
    {
      "entropy": 5.634044075012207,
      "epoch": 2.7595275903136165,
      "grad_norm": 1.015625,
      "learning_rate": 0.00042447699252308566,
      "loss": 5.0064,
      "mean_token_accuracy": 0.18053687065839769,
      "num_tokens": 60914673.0,
      "step": 27805
    },
    {
      "entropy": 5.588279771804809,
      "epoch": 2.760023818975784,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004244505081576521,
      "loss": 4.977,
      "mean_token_accuracy": 0.18592079728841782,
      "num_tokens": 60925872.0,
      "step": 27810
    },
    {
      "entropy": 5.619391918182373,
      "epoch": 2.7605200476379514,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042442402008616793,
      "loss": 5.0741,
      "mean_token_accuracy": 0.18028918355703355,
      "num_tokens": 60936589.0,
      "step": 27815
    },
    {
      "entropy": 5.613119649887085,
      "epoch": 2.7610162763001194,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042439752830929,
      "loss": 5.0289,
      "mean_token_accuracy": 0.178982475399971,
      "num_tokens": 60947213.0,
      "step": 27820
    },
    {
      "entropy": 5.69819393157959,
      "epoch": 2.761512504962287,
      "grad_norm": 1.125,
      "learning_rate": 0.0004243710328276754,
      "loss": 5.021,
      "mean_token_accuracy": 0.1834500849246979,
      "num_tokens": 60958615.0,
      "step": 27825
    },
    {
      "entropy": 5.618701839447022,
      "epoch": 2.7620087336244543,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000424344533641981,
      "loss": 4.9738,
      "mean_token_accuracy": 0.18677895814180373,
      "num_tokens": 60968830.0,
      "step": 27830
    },
    {
      "entropy": 5.581523370742798,
      "epoch": 2.762504962286622,
      "grad_norm": 1.109375,
      "learning_rate": 0.000424318030752864,
      "loss": 4.9526,
      "mean_token_accuracy": 0.1864494800567627,
      "num_tokens": 60979744.0,
      "step": 27835
    },
    {
      "entropy": 5.654686784744262,
      "epoch": 2.7630011909487893,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004242915241609815,
      "loss": 5.0551,
      "mean_token_accuracy": 0.17591994255781174,
      "num_tokens": 60990291.0,
      "step": 27840
    },
    {
      "entropy": 5.627245187759399,
      "epoch": 2.7634974196109567,
      "grad_norm": 1.078125,
      "learning_rate": 0.000424265013866991,
      "loss": 5.0066,
      "mean_token_accuracy": 0.18664557188749314,
      "num_tokens": 61002170.0,
      "step": 27845
    },
    {
      "entropy": 5.587162256240845,
      "epoch": 2.763993648273124,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004242384998715497,
      "loss": 5.0577,
      "mean_token_accuracy": 0.18185107856988908,
      "num_tokens": 61012639.0,
      "step": 27850
    },
    {
      "entropy": 5.656422853469849,
      "epoch": 2.7644898769352917,
      "grad_norm": 1.25,
      "learning_rate": 0.0004242119821753153,
      "loss": 5.0195,
      "mean_token_accuracy": 0.1852187305688858,
      "num_tokens": 61022657.0,
      "step": 27855
    },
    {
      "entropy": 5.613352680206299,
      "epoch": 2.764986105597459,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004241854607789452,
      "loss": 4.9768,
      "mean_token_accuracy": 0.18789129704236984,
      "num_tokens": 61033108.0,
      "step": 27860
    },
    {
      "entropy": 5.561254501342773,
      "epoch": 2.7654823342596266,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042415893568309714,
      "loss": 4.993,
      "mean_token_accuracy": 0.18533117622137069,
      "num_tokens": 61043651.0,
      "step": 27865
    },
    {
      "entropy": 5.565281915664673,
      "epoch": 2.765978562921794,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042413240688842884,
      "loss": 4.9462,
      "mean_token_accuracy": 0.1905208945274353,
      "num_tokens": 61053234.0,
      "step": 27870
    },
    {
      "entropy": 5.563858890533448,
      "epoch": 2.766474791583962,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004241058743955982,
      "loss": 4.9961,
      "mean_token_accuracy": 0.18286824077367783,
      "num_tokens": 61065178.0,
      "step": 27875
    },
    {
      "entropy": 5.596913814544678,
      "epoch": 2.7669710202461295,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004240793382052633,
      "loss": 4.9694,
      "mean_token_accuracy": 0.19241756200790405,
      "num_tokens": 61075709.0,
      "step": 27880
    },
    {
      "entropy": 5.649411916732788,
      "epoch": 2.767467248908297,
      "grad_norm": 1.046875,
      "learning_rate": 0.000424052798318082,
      "loss": 5.0807,
      "mean_token_accuracy": 0.17231455296278,
      "num_tokens": 61086880.0,
      "step": 27885
    },
    {
      "entropy": 5.589518690109253,
      "epoch": 2.7679634775704645,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004240262547347124,
      "loss": 5.017,
      "mean_token_accuracy": 0.18682194799184798,
      "num_tokens": 61097008.0,
      "step": 27890
    },
    {
      "entropy": 5.626911592483521,
      "epoch": 2.768459706232632,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042399970745581283,
      "loss": 4.9981,
      "mean_token_accuracy": 0.18668035864830018,
      "num_tokens": 61107844.0,
      "step": 27895
    },
    {
      "entropy": 5.562051916122437,
      "epoch": 2.7689559348947994,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042397315648204164,
      "loss": 4.9509,
      "mean_token_accuracy": 0.19240849912166597,
      "num_tokens": 61118652.0,
      "step": 27900
    },
    {
      "entropy": 5.668383646011352,
      "epoch": 2.7694521635569673,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042394660181405716,
      "loss": 5.0967,
      "mean_token_accuracy": 0.18060849308967591,
      "num_tokens": 61129143.0,
      "step": 27905
    },
    {
      "entropy": 5.63944263458252,
      "epoch": 2.769948392219135,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004239200434525179,
      "loss": 5.0373,
      "mean_token_accuracy": 0.1837241992354393,
      "num_tokens": 61140623.0,
      "step": 27910
    },
    {
      "entropy": 5.592364645004272,
      "epoch": 2.7704446208813023,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004238934813980824,
      "loss": 4.9982,
      "mean_token_accuracy": 0.18419128954410552,
      "num_tokens": 61150998.0,
      "step": 27915
    },
    {
      "entropy": 5.634286785125733,
      "epoch": 2.7709408495434698,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042386691565140947,
      "loss": 5.0223,
      "mean_token_accuracy": 0.18412060737609864,
      "num_tokens": 61162086.0,
      "step": 27920
    },
    {
      "entropy": 5.64199161529541,
      "epoch": 2.7714370782056372,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042384034621315773,
      "loss": 4.9959,
      "mean_token_accuracy": 0.18181274980306625,
      "num_tokens": 61171806.0,
      "step": 27925
    },
    {
      "entropy": 5.60626449584961,
      "epoch": 2.7719333068678047,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004238137730839862,
      "loss": 4.9847,
      "mean_token_accuracy": 0.18453231900930406,
      "num_tokens": 61182056.0,
      "step": 27930
    },
    {
      "entropy": 5.615599155426025,
      "epoch": 2.772429535529972,
      "grad_norm": 1.125,
      "learning_rate": 0.0004237871962645538,
      "loss": 5.0186,
      "mean_token_accuracy": 0.1858121782541275,
      "num_tokens": 61193231.0,
      "step": 27935
    },
    {
      "entropy": 5.572538805007935,
      "epoch": 2.7729257641921397,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00042376061575551946,
      "loss": 5.0015,
      "mean_token_accuracy": 0.18742585629224778,
      "num_tokens": 61203785.0,
      "step": 27940
    },
    {
      "entropy": 5.568484783172607,
      "epoch": 2.773421992854307,
      "grad_norm": 1.0,
      "learning_rate": 0.0004237340315575425,
      "loss": 5.0162,
      "mean_token_accuracy": 0.1888026550412178,
      "num_tokens": 61215244.0,
      "step": 27945
    },
    {
      "entropy": 5.639634609222412,
      "epoch": 2.7739182215164746,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004237074436712819,
      "loss": 4.986,
      "mean_token_accuracy": 0.18426254093647004,
      "num_tokens": 61227147.0,
      "step": 27950
    },
    {
      "entropy": 5.680387210845947,
      "epoch": 2.774414450178642,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042368085209739715,
      "loss": 5.0523,
      "mean_token_accuracy": 0.18678874373435975,
      "num_tokens": 61237687.0,
      "step": 27955
    },
    {
      "entropy": 5.532999467849732,
      "epoch": 2.77491067884081,
      "grad_norm": 1.125,
      "learning_rate": 0.0004236542568365478,
      "loss": 4.9542,
      "mean_token_accuracy": 0.18932979851961135,
      "num_tokens": 61247700.0,
      "step": 27960
    },
    {
      "entropy": 5.607886075973511,
      "epoch": 2.7754069075029775,
      "grad_norm": 1.25,
      "learning_rate": 0.00042362765788939307,
      "loss": 4.9733,
      "mean_token_accuracy": 0.18144610822200774,
      "num_tokens": 61258407.0,
      "step": 27965
    },
    {
      "entropy": 5.680282402038574,
      "epoch": 2.775903136165145,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004236010552565927,
      "loss": 5.1319,
      "mean_token_accuracy": 0.17552007734775543,
      "num_tokens": 61269065.0,
      "step": 27970
    },
    {
      "entropy": 5.652399635314941,
      "epoch": 2.7763993648273124,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042357444893880645,
      "loss": 5.0082,
      "mean_token_accuracy": 0.18172777444124222,
      "num_tokens": 61280767.0,
      "step": 27975
    },
    {
      "entropy": 5.558382034301758,
      "epoch": 2.77689559348948,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042354783893669385,
      "loss": 4.9593,
      "mean_token_accuracy": 0.18396374136209487,
      "num_tokens": 61291967.0,
      "step": 27980
    },
    {
      "entropy": 5.6664040088653564,
      "epoch": 2.7773918221516474,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004235212252509151,
      "loss": 5.1378,
      "mean_token_accuracy": 0.17554586231708527,
      "num_tokens": 61303572.0,
      "step": 27985
    },
    {
      "entropy": 5.636144065856934,
      "epoch": 2.777888050813815,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00042349460788212995,
      "loss": 5.0353,
      "mean_token_accuracy": 0.18538753390312196,
      "num_tokens": 61315674.0,
      "step": 27990
    },
    {
      "entropy": 5.627111339569092,
      "epoch": 2.7783842794759828,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042346798683099844,
      "loss": 4.9858,
      "mean_token_accuracy": 0.18407958298921584,
      "num_tokens": 61326334.0,
      "step": 27995
    },
    {
      "entropy": 5.6239160060882565,
      "epoch": 2.7788805081381502,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004234413620981809,
      "loss": 5.1125,
      "mean_token_accuracy": 0.17616498917341233,
      "num_tokens": 61337505.0,
      "step": 28000
    },
    {
      "entropy": 5.638072538375854,
      "epoch": 2.7793767368003177,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042341473368433735,
      "loss": 5.0627,
      "mean_token_accuracy": 0.18311095535755156,
      "num_tokens": 61348451.0,
      "step": 28005
    },
    {
      "entropy": 5.713073873519898,
      "epoch": 2.779872965462485,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004233881015901282,
      "loss": 5.2012,
      "mean_token_accuracy": 0.1703106075525284,
      "num_tokens": 61358356.0,
      "step": 28010
    },
    {
      "entropy": 5.713620138168335,
      "epoch": 2.7803691941246527,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004233614658162139,
      "loss": 5.0444,
      "mean_token_accuracy": 0.1876247152686119,
      "num_tokens": 61370996.0,
      "step": 28015
    },
    {
      "entropy": 5.689835929870606,
      "epoch": 2.78086542278682,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042333482636325504,
      "loss": 5.079,
      "mean_token_accuracy": 0.17999690175056457,
      "num_tokens": 61383928.0,
      "step": 28020
    },
    {
      "entropy": 5.6359117984771725,
      "epoch": 2.7813616514489876,
      "grad_norm": 1.09375,
      "learning_rate": 0.000423308183231912,
      "loss": 5.0541,
      "mean_token_accuracy": 0.18272197395563125,
      "num_tokens": 61394277.0,
      "step": 28025
    },
    {
      "entropy": 5.577258920669555,
      "epoch": 2.781857880111155,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00042328153642284567,
      "loss": 5.0127,
      "mean_token_accuracy": 0.18975501656532287,
      "num_tokens": 61405928.0,
      "step": 28030
    },
    {
      "entropy": 5.684264230728149,
      "epoch": 2.7823541087733226,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004232548859367167,
      "loss": 5.0909,
      "mean_token_accuracy": 0.17982769757509232,
      "num_tokens": 61417893.0,
      "step": 28035
    },
    {
      "entropy": 5.640292644500732,
      "epoch": 2.78285033743549,
      "grad_norm": 1.09375,
      "learning_rate": 0.000423228231774186,
      "loss": 5.0395,
      "mean_token_accuracy": 0.18183884918689727,
      "num_tokens": 61428099.0,
      "step": 28040
    },
    {
      "entropy": 5.620064640045166,
      "epoch": 2.783346566097658,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004232015739359146,
      "loss": 4.9745,
      "mean_token_accuracy": 0.18957398235797882,
      "num_tokens": 61437636.0,
      "step": 28045
    },
    {
      "entropy": 5.576571035385132,
      "epoch": 2.7838427947598254,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042317491242256346,
      "loss": 4.9567,
      "mean_token_accuracy": 0.1862746074795723,
      "num_tokens": 61448038.0,
      "step": 28050
    },
    {
      "entropy": 5.603193759918213,
      "epoch": 2.784339023421993,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004231482472347938,
      "loss": 5.0041,
      "mean_token_accuracy": 0.1831810414791107,
      "num_tokens": 61460258.0,
      "step": 28055
    },
    {
      "entropy": 5.616006851196289,
      "epoch": 2.7848352520841604,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004231215783732668,
      "loss": 5.0346,
      "mean_token_accuracy": 0.18467175513505935,
      "num_tokens": 61472274.0,
      "step": 28060
    },
    {
      "entropy": 5.6303221702575685,
      "epoch": 2.785331480746328,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042309490583864395,
      "loss": 5.0195,
      "mean_token_accuracy": 0.18563609570264816,
      "num_tokens": 61482975.0,
      "step": 28065
    },
    {
      "entropy": 5.660983943939209,
      "epoch": 2.7858277094084953,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004230682296315865,
      "loss": 5.0452,
      "mean_token_accuracy": 0.17922421395778657,
      "num_tokens": 61494223.0,
      "step": 28070
    },
    {
      "entropy": 5.720233058929443,
      "epoch": 2.786323938070663,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000423041549752756,
      "loss": 5.1021,
      "mean_token_accuracy": 0.18339602053165435,
      "num_tokens": 61505981.0,
      "step": 28075
    },
    {
      "entropy": 5.547557592391968,
      "epoch": 2.7868201667328307,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004230148662028141,
      "loss": 4.9702,
      "mean_token_accuracy": 0.18611379712820053,
      "num_tokens": 61516410.0,
      "step": 28080
    },
    {
      "entropy": 5.567597055435181,
      "epoch": 2.787316395394998,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042298817898242237,
      "loss": 4.9868,
      "mean_token_accuracy": 0.18130161762237548,
      "num_tokens": 61526551.0,
      "step": 28085
    },
    {
      "entropy": 5.677392625808716,
      "epoch": 2.7878126240571657,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042296148809224275,
      "loss": 5.1194,
      "mean_token_accuracy": 0.17953856736421586,
      "num_tokens": 61538214.0,
      "step": 28090
    },
    {
      "entropy": 5.6429298400878904,
      "epoch": 2.788308852719333,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004229347935329371,
      "loss": 5.0856,
      "mean_token_accuracy": 0.17560193836688995,
      "num_tokens": 61549726.0,
      "step": 28095
    },
    {
      "entropy": 5.70915207862854,
      "epoch": 2.7888050813815006,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042290809530516724,
      "loss": 5.1535,
      "mean_token_accuracy": 0.17094107568264008,
      "num_tokens": 61560070.0,
      "step": 28100
    },
    {
      "entropy": 5.706589031219482,
      "epoch": 2.789301310043668,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004228813934095954,
      "loss": 5.0795,
      "mean_token_accuracy": 0.1798699364066124,
      "num_tokens": 61571348.0,
      "step": 28105
    },
    {
      "entropy": 5.593964958190918,
      "epoch": 2.7897975387058356,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042285468784688363,
      "loss": 4.9574,
      "mean_token_accuracy": 0.1879449561238289,
      "num_tokens": 61581981.0,
      "step": 28110
    },
    {
      "entropy": 5.602452087402344,
      "epoch": 2.790293767368003,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042282797861769423,
      "loss": 5.057,
      "mean_token_accuracy": 0.18301763385534286,
      "num_tokens": 61591248.0,
      "step": 28115
    },
    {
      "entropy": 5.591927337646484,
      "epoch": 2.7907899960301705,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004228012657226895,
      "loss": 5.0613,
      "mean_token_accuracy": 0.18074851036071776,
      "num_tokens": 61602432.0,
      "step": 28120
    },
    {
      "entropy": 5.574327039718628,
      "epoch": 2.791286224692338,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00042277454916253186,
      "loss": 5.0298,
      "mean_token_accuracy": 0.17830269932746887,
      "num_tokens": 61613418.0,
      "step": 28125
    },
    {
      "entropy": 5.619820070266724,
      "epoch": 2.791782453354506,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004227478289378838,
      "loss": 5.0109,
      "mean_token_accuracy": 0.17809389382600785,
      "num_tokens": 61623783.0,
      "step": 28130
    },
    {
      "entropy": 5.6881509780883786,
      "epoch": 2.7922786820166734,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042272110504940795,
      "loss": 5.0967,
      "mean_token_accuracy": 0.1784144163131714,
      "num_tokens": 61633838.0,
      "step": 28135
    },
    {
      "entropy": 5.612725687026978,
      "epoch": 2.792774910678841,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042269437749776703,
      "loss": 4.9828,
      "mean_token_accuracy": 0.1877562716603279,
      "num_tokens": 61645982.0,
      "step": 28140
    },
    {
      "entropy": 5.587034034729004,
      "epoch": 2.7932711393410083,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004226676462836238,
      "loss": 4.9578,
      "mean_token_accuracy": 0.18453058302402497,
      "num_tokens": 61656030.0,
      "step": 28145
    },
    {
      "entropy": 5.54879903793335,
      "epoch": 2.793767368003176,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004226409114076411,
      "loss": 4.99,
      "mean_token_accuracy": 0.18540441542863845,
      "num_tokens": 61666746.0,
      "step": 28150
    },
    {
      "entropy": 5.66160888671875,
      "epoch": 2.7942635966653433,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042261417287048196,
      "loss": 5.1417,
      "mean_token_accuracy": 0.18062290102243422,
      "num_tokens": 61677681.0,
      "step": 28155
    },
    {
      "entropy": 5.728194904327393,
      "epoch": 2.7947598253275108,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004225874306728094,
      "loss": 5.0954,
      "mean_token_accuracy": 0.17671594619750977,
      "num_tokens": 61689564.0,
      "step": 28160
    },
    {
      "entropy": 5.632253551483155,
      "epoch": 2.7952560539896787,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042256068481528666,
      "loss": 5.0013,
      "mean_token_accuracy": 0.18339141309261323,
      "num_tokens": 61701074.0,
      "step": 28165
    },
    {
      "entropy": 5.5685595035552975,
      "epoch": 2.795752282651846,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004225339352985768,
      "loss": 4.9919,
      "mean_token_accuracy": 0.18690716624259948,
      "num_tokens": 61712203.0,
      "step": 28170
    },
    {
      "entropy": 5.6141517162323,
      "epoch": 2.7962485113140136,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042250718212334334,
      "loss": 4.9562,
      "mean_token_accuracy": 0.1851118490099907,
      "num_tokens": 61723362.0,
      "step": 28175
    },
    {
      "entropy": 5.6111420631408695,
      "epoch": 2.796744739976181,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042248042529024956,
      "loss": 4.9731,
      "mean_token_accuracy": 0.1900700569152832,
      "num_tokens": 61733454.0,
      "step": 28180
    },
    {
      "entropy": 5.6471720218658445,
      "epoch": 2.7972409686383486,
      "grad_norm": 1.015625,
      "learning_rate": 0.00042245366479995894,
      "loss": 5.0821,
      "mean_token_accuracy": 0.1788816437125206,
      "num_tokens": 61745014.0,
      "step": 28185
    },
    {
      "entropy": 5.598909521102906,
      "epoch": 2.797737197300516,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042242690065313535,
      "loss": 5.0302,
      "mean_token_accuracy": 0.18383824229240417,
      "num_tokens": 61755334.0,
      "step": 28190
    },
    {
      "entropy": 5.565058994293213,
      "epoch": 2.7982334259626835,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042240013285044215,
      "loss": 4.972,
      "mean_token_accuracy": 0.19374998062849044,
      "num_tokens": 61766024.0,
      "step": 28195
    },
    {
      "entropy": 5.6556683540344235,
      "epoch": 2.798729654624851,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00042237336139254336,
      "loss": 5.0189,
      "mean_token_accuracy": 0.1894935041666031,
      "num_tokens": 61778228.0,
      "step": 28200
    },
    {
      "entropy": 5.660628318786621,
      "epoch": 2.7992258832870185,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042234658628010274,
      "loss": 5.0663,
      "mean_token_accuracy": 0.17693185955286025,
      "num_tokens": 61789072.0,
      "step": 28205
    },
    {
      "entropy": 5.6409400463104244,
      "epoch": 2.799722111949186,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042231980751378435,
      "loss": 5.0092,
      "mean_token_accuracy": 0.1921718567609787,
      "num_tokens": 61799503.0,
      "step": 28210
    },
    {
      "entropy": 5.577624654769897,
      "epoch": 2.8002183406113534,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042229302509425205,
      "loss": 4.9705,
      "mean_token_accuracy": 0.18936125487089156,
      "num_tokens": 61811063.0,
      "step": 28215
    },
    {
      "entropy": 5.6281623363494875,
      "epoch": 2.8007145692735214,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004222662390221702,
      "loss": 5.0758,
      "mean_token_accuracy": 0.17948224693536757,
      "num_tokens": 61822832.0,
      "step": 28220
    },
    {
      "entropy": 5.665958738327026,
      "epoch": 2.801210797935689,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004222394492982029,
      "loss": 5.0518,
      "mean_token_accuracy": 0.1759596809744835,
      "num_tokens": 61832456.0,
      "step": 28225
    },
    {
      "entropy": 5.626626586914062,
      "epoch": 2.8017070265978563,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004222126559230146,
      "loss": 4.9891,
      "mean_token_accuracy": 0.18906847089529039,
      "num_tokens": 61843043.0,
      "step": 28230
    },
    {
      "entropy": 5.5559485912322994,
      "epoch": 2.802203255260024,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004221858588972696,
      "loss": 5.0306,
      "mean_token_accuracy": 0.17656009048223495,
      "num_tokens": 61853249.0,
      "step": 28235
    },
    {
      "entropy": 5.523712015151977,
      "epoch": 2.8026994839221913,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042215905822163245,
      "loss": 4.9599,
      "mean_token_accuracy": 0.19112093597650529,
      "num_tokens": 61863842.0,
      "step": 28240
    },
    {
      "entropy": 5.6194929599761965,
      "epoch": 2.8031957125843587,
      "grad_norm": 1.0,
      "learning_rate": 0.0004221322538967678,
      "loss": 5.0415,
      "mean_token_accuracy": 0.18196653574705124,
      "num_tokens": 61876924.0,
      "step": 28245
    },
    {
      "entropy": 5.715323257446289,
      "epoch": 2.8036919412465267,
      "grad_norm": 1.125,
      "learning_rate": 0.0004221054459233403,
      "loss": 5.0843,
      "mean_token_accuracy": 0.176398603618145,
      "num_tokens": 61886766.0,
      "step": 28250
    },
    {
      "entropy": 5.6948450088500975,
      "epoch": 2.804188169908694,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042207863430201466,
      "loss": 5.0362,
      "mean_token_accuracy": 0.18937778174877168,
      "num_tokens": 61896538.0,
      "step": 28255
    },
    {
      "entropy": 5.590991687774658,
      "epoch": 2.8046843985708616,
      "grad_norm": 0.953125,
      "learning_rate": 0.00042205181903345586,
      "loss": 5.0162,
      "mean_token_accuracy": 0.1758688673377037,
      "num_tokens": 61907328.0,
      "step": 28260
    },
    {
      "entropy": 5.611815643310547,
      "epoch": 2.805180627233029,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004220250001183288,
      "loss": 5.0467,
      "mean_token_accuracy": 0.1801968738436699,
      "num_tokens": 61918797.0,
      "step": 28265
    },
    {
      "entropy": 5.648912763595581,
      "epoch": 2.8056768558951966,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004219981775572986,
      "loss": 5.0691,
      "mean_token_accuracy": 0.1833320215344429,
      "num_tokens": 61928755.0,
      "step": 28270
    },
    {
      "entropy": 5.633308982849121,
      "epoch": 2.806173084557364,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004219713513510303,
      "loss": 5.0404,
      "mean_token_accuracy": 0.17843065708875655,
      "num_tokens": 61939993.0,
      "step": 28275
    },
    {
      "entropy": 5.5458070755004885,
      "epoch": 2.8066693132195315,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042194452150018917,
      "loss": 4.907,
      "mean_token_accuracy": 0.19532775580883027,
      "num_tokens": 61950982.0,
      "step": 28280
    },
    {
      "entropy": 5.618481731414795,
      "epoch": 2.807165541881699,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042191768800544055,
      "loss": 5.0038,
      "mean_token_accuracy": 0.17913904637098313,
      "num_tokens": 61961786.0,
      "step": 28285
    },
    {
      "entropy": 5.621355819702148,
      "epoch": 2.8076617705438665,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042189085086744986,
      "loss": 4.9977,
      "mean_token_accuracy": 0.19184398055076599,
      "num_tokens": 61972548.0,
      "step": 28290
    },
    {
      "entropy": 5.607481336593628,
      "epoch": 2.808157999206034,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004218640100868826,
      "loss": 4.9898,
      "mean_token_accuracy": 0.18982910811901094,
      "num_tokens": 61981822.0,
      "step": 28295
    },
    {
      "entropy": 5.635645723342895,
      "epoch": 2.8086542278682014,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042183716566440433,
      "loss": 5.1248,
      "mean_token_accuracy": 0.1728345274925232,
      "num_tokens": 61992997.0,
      "step": 28300
    },
    {
      "entropy": 5.625591897964478,
      "epoch": 2.8091504565303693,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004218103176006808,
      "loss": 5.0558,
      "mean_token_accuracy": 0.17887795716524124,
      "num_tokens": 62004391.0,
      "step": 28305
    },
    {
      "entropy": 5.629903268814087,
      "epoch": 2.809646685192537,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004217834658963777,
      "loss": 5.0159,
      "mean_token_accuracy": 0.1876690536737442,
      "num_tokens": 62014099.0,
      "step": 28310
    },
    {
      "entropy": 5.641525602340698,
      "epoch": 2.8101429138547043,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042175661055216093,
      "loss": 4.9894,
      "mean_token_accuracy": 0.18265165537595748,
      "num_tokens": 62023872.0,
      "step": 28315
    },
    {
      "entropy": 5.639899587631225,
      "epoch": 2.8106391425168717,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004217297515686965,
      "loss": 5.0005,
      "mean_token_accuracy": 0.18363700211048126,
      "num_tokens": 62035566.0,
      "step": 28320
    },
    {
      "entropy": 5.5944239616394045,
      "epoch": 2.811135371179039,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004217028889466503,
      "loss": 5.0151,
      "mean_token_accuracy": 0.18010144978761672,
      "num_tokens": 62046742.0,
      "step": 28325
    },
    {
      "entropy": 5.612712955474853,
      "epoch": 2.8116315998412067,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004216760226866887,
      "loss": 4.9764,
      "mean_token_accuracy": 0.1827520802617073,
      "num_tokens": 62056545.0,
      "step": 28330
    },
    {
      "entropy": 5.658690929412842,
      "epoch": 2.8121278285033746,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042164915278947775,
      "loss": 5.0738,
      "mean_token_accuracy": 0.18333644270896912,
      "num_tokens": 62066414.0,
      "step": 28335
    },
    {
      "entropy": 5.5768228530883786,
      "epoch": 2.812624057165542,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004216222792556838,
      "loss": 4.9758,
      "mean_token_accuracy": 0.18293156921863557,
      "num_tokens": 62078406.0,
      "step": 28340
    },
    {
      "entropy": 5.602375078201294,
      "epoch": 2.8131202858277096,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004215954020859732,
      "loss": 5.031,
      "mean_token_accuracy": 0.18085455745458603,
      "num_tokens": 62089516.0,
      "step": 28345
    },
    {
      "entropy": 5.620419979095459,
      "epoch": 2.813616514489877,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042156852128101264,
      "loss": 5.0152,
      "mean_token_accuracy": 0.18792759478092194,
      "num_tokens": 62099287.0,
      "step": 28350
    },
    {
      "entropy": 5.5905815124511715,
      "epoch": 2.8141127431520445,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042154163684146855,
      "loss": 4.98,
      "mean_token_accuracy": 0.18546606302261354,
      "num_tokens": 62111110.0,
      "step": 28355
    },
    {
      "entropy": 5.7087033748626705,
      "epoch": 2.814608971814212,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004215147487680076,
      "loss": 5.1444,
      "mean_token_accuracy": 0.17313922941684723,
      "num_tokens": 62123614.0,
      "step": 28360
    },
    {
      "entropy": 5.58901801109314,
      "epoch": 2.8151052004763795,
      "grad_norm": 1.125,
      "learning_rate": 0.00042148785706129657,
      "loss": 5.0773,
      "mean_token_accuracy": 0.18270352780818938,
      "num_tokens": 62133682.0,
      "step": 28365
    },
    {
      "entropy": 5.603968000411987,
      "epoch": 2.815601429138547,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004214609617220024,
      "loss": 5.0063,
      "mean_token_accuracy": 0.1815941661596298,
      "num_tokens": 62144802.0,
      "step": 28370
    },
    {
      "entropy": 5.654648351669311,
      "epoch": 2.8160976578007144,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042143406275079186,
      "loss": 4.9961,
      "mean_token_accuracy": 0.18042581677436828,
      "num_tokens": 62154205.0,
      "step": 28375
    },
    {
      "entropy": 5.595628547668457,
      "epoch": 2.816593886462882,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004214071601483322,
      "loss": 4.9113,
      "mean_token_accuracy": 0.19724029749631883,
      "num_tokens": 62164648.0,
      "step": 28380
    },
    {
      "entropy": 5.5516520023345945,
      "epoch": 2.8170901151250494,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004213802539152905,
      "loss": 4.999,
      "mean_token_accuracy": 0.18764600902795792,
      "num_tokens": 62176073.0,
      "step": 28385
    },
    {
      "entropy": 5.6369517803192135,
      "epoch": 2.8175863437872173,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00042135334405233384,
      "loss": 5.0388,
      "mean_token_accuracy": 0.17848848402500153,
      "num_tokens": 62187350.0,
      "step": 28390
    },
    {
      "entropy": 5.596481037139893,
      "epoch": 2.8180825724493848,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004213264305601296,
      "loss": 4.9739,
      "mean_token_accuracy": 0.1829273745417595,
      "num_tokens": 62197729.0,
      "step": 28395
    },
    {
      "entropy": 5.566349172592163,
      "epoch": 2.8185788011115522,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042129951343934527,
      "loss": 4.9545,
      "mean_token_accuracy": 0.1935669809579849,
      "num_tokens": 62209047.0,
      "step": 28400
    },
    {
      "entropy": 5.609026622772217,
      "epoch": 2.8190750297737197,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004212725926906482,
      "loss": 4.997,
      "mean_token_accuracy": 0.18274599611759185,
      "num_tokens": 62218859.0,
      "step": 28405
    },
    {
      "entropy": 5.585991525650025,
      "epoch": 2.819571258435887,
      "grad_norm": 1.125,
      "learning_rate": 0.00042124566831470604,
      "loss": 5.0028,
      "mean_token_accuracy": 0.18516783863306047,
      "num_tokens": 62230396.0,
      "step": 28410
    },
    {
      "entropy": 5.588314580917358,
      "epoch": 2.8200674870980547,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004212187403121864,
      "loss": 4.9785,
      "mean_token_accuracy": 0.1901271805167198,
      "num_tokens": 62241790.0,
      "step": 28415
    },
    {
      "entropy": 5.593439388275146,
      "epoch": 2.8205637157602226,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042119180868375707,
      "loss": 4.9509,
      "mean_token_accuracy": 0.1891007974743843,
      "num_tokens": 62253820.0,
      "step": 28420
    },
    {
      "entropy": 5.579290199279785,
      "epoch": 2.82105994442239,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004211648734300859,
      "loss": 5.0143,
      "mean_token_accuracy": 0.18295426219701766,
      "num_tokens": 62263875.0,
      "step": 28425
    },
    {
      "entropy": 5.592715644836426,
      "epoch": 2.8215561730845575,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042113793455184084,
      "loss": 5.0394,
      "mean_token_accuracy": 0.1810181573033333,
      "num_tokens": 62275390.0,
      "step": 28430
    },
    {
      "entropy": 5.6007843017578125,
      "epoch": 2.822052401746725,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004211109920496899,
      "loss": 5.0772,
      "mean_token_accuracy": 0.17341843247413635,
      "num_tokens": 62286969.0,
      "step": 28435
    },
    {
      "entropy": 5.613134574890137,
      "epoch": 2.8225486304088925,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042108404592430126,
      "loss": 5.0137,
      "mean_token_accuracy": 0.18292375802993774,
      "num_tokens": 62297810.0,
      "step": 28440
    },
    {
      "entropy": 5.619211721420288,
      "epoch": 2.82304485907106,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000421057096176343,
      "loss": 5.0255,
      "mean_token_accuracy": 0.18558906018733978,
      "num_tokens": 62309474.0,
      "step": 28445
    },
    {
      "entropy": 5.5670496940612795,
      "epoch": 2.8235410877332274,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004210301428064835,
      "loss": 5.0533,
      "mean_token_accuracy": 0.18186789602041245,
      "num_tokens": 62320022.0,
      "step": 28450
    },
    {
      "entropy": 5.664820575714112,
      "epoch": 2.824037316395395,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004210031858153911,
      "loss": 5.0206,
      "mean_token_accuracy": 0.18838034123182296,
      "num_tokens": 62330450.0,
      "step": 28455
    },
    {
      "entropy": 5.651589393615723,
      "epoch": 2.8245335450575624,
      "grad_norm": 1.0,
      "learning_rate": 0.00042097622520373434,
      "loss": 5.0476,
      "mean_token_accuracy": 0.1809871554374695,
      "num_tokens": 62341419.0,
      "step": 28460
    },
    {
      "entropy": 5.660876655578614,
      "epoch": 2.82502977371973,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004209492609721817,
      "loss": 5.0777,
      "mean_token_accuracy": 0.18043782114982604,
      "num_tokens": 62353312.0,
      "step": 28465
    },
    {
      "entropy": 5.659527730941773,
      "epoch": 2.8255260023818973,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004209222931214019,
      "loss": 5.0422,
      "mean_token_accuracy": 0.18412943482398986,
      "num_tokens": 62364838.0,
      "step": 28470
    },
    {
      "entropy": 5.599405145645141,
      "epoch": 2.8260222310440652,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042089532165206367,
      "loss": 4.9517,
      "mean_token_accuracy": 0.19128181636333466,
      "num_tokens": 62376443.0,
      "step": 28475
    },
    {
      "entropy": 5.677672672271728,
      "epoch": 2.8265184597062327,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004208683465648358,
      "loss": 5.1328,
      "mean_token_accuracy": 0.17443612664937974,
      "num_tokens": 62387263.0,
      "step": 28480
    },
    {
      "entropy": 5.595383071899414,
      "epoch": 2.8270146883684,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042084136786038737,
      "loss": 4.9482,
      "mean_token_accuracy": 0.18665664196014403,
      "num_tokens": 62398929.0,
      "step": 28485
    },
    {
      "entropy": 5.616779756546021,
      "epoch": 2.8275109170305677,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004208143855393872,
      "loss": 5.0587,
      "mean_token_accuracy": 0.17782157361507417,
      "num_tokens": 62409516.0,
      "step": 28490
    },
    {
      "entropy": 5.660712671279907,
      "epoch": 2.828007145692735,
      "grad_norm": 1.0,
      "learning_rate": 0.00042078739960250446,
      "loss": 5.0194,
      "mean_token_accuracy": 0.18727981448173522,
      "num_tokens": 62421856.0,
      "step": 28495
    },
    {
      "entropy": 5.666030168533325,
      "epoch": 2.8285033743549026,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042076041005040835,
      "loss": 4.9704,
      "mean_token_accuracy": 0.18266840428113937,
      "num_tokens": 62432340.0,
      "step": 28500
    },
    {
      "entropy": 5.605877017974853,
      "epoch": 2.8289996030170705,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004207334168837682,
      "loss": 5.0523,
      "mean_token_accuracy": 0.1754041776061058,
      "num_tokens": 62442652.0,
      "step": 28505
    },
    {
      "entropy": 5.639624547958374,
      "epoch": 2.829495831679238,
      "grad_norm": 1.0,
      "learning_rate": 0.0004207064201032533,
      "loss": 4.9776,
      "mean_token_accuracy": 0.19018022567033768,
      "num_tokens": 62454503.0,
      "step": 28510
    },
    {
      "entropy": 5.5767427444458,
      "epoch": 2.8299920603414055,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042067941970953316,
      "loss": 4.9471,
      "mean_token_accuracy": 0.18985274136066438,
      "num_tokens": 62464858.0,
      "step": 28515
    },
    {
      "entropy": 5.662875080108643,
      "epoch": 2.830488289003573,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042065241570327734,
      "loss": 5.0387,
      "mean_token_accuracy": 0.18194727450609208,
      "num_tokens": 62475153.0,
      "step": 28520
    },
    {
      "entropy": 5.630006742477417,
      "epoch": 2.8309845176657404,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004206254080851555,
      "loss": 5.0792,
      "mean_token_accuracy": 0.1763299897313118,
      "num_tokens": 62487883.0,
      "step": 28525
    },
    {
      "entropy": 5.588761997222901,
      "epoch": 2.831480746327908,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042059839685583726,
      "loss": 4.9871,
      "mean_token_accuracy": 0.18475978523492814,
      "num_tokens": 62499905.0,
      "step": 28530
    },
    {
      "entropy": 5.649273538589478,
      "epoch": 2.8319769749900754,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042057138201599247,
      "loss": 5.0588,
      "mean_token_accuracy": 0.17754292935132981,
      "num_tokens": 62511403.0,
      "step": 28535
    },
    {
      "entropy": 5.644236087799072,
      "epoch": 2.832473203652243,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042054436356629113,
      "loss": 5.0077,
      "mean_token_accuracy": 0.19396273642778397,
      "num_tokens": 62522543.0,
      "step": 28540
    },
    {
      "entropy": 5.591735172271728,
      "epoch": 2.8329694323144103,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004205173415074032,
      "loss": 4.9785,
      "mean_token_accuracy": 0.18382260650396348,
      "num_tokens": 62533053.0,
      "step": 28545
    },
    {
      "entropy": 5.634284591674804,
      "epoch": 2.833465660976578,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004204903158399989,
      "loss": 5.0964,
      "mean_token_accuracy": 0.17840754389762878,
      "num_tokens": 62543699.0,
      "step": 28550
    },
    {
      "entropy": 5.615767240524292,
      "epoch": 2.8339618896387453,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004204632865647481,
      "loss": 5.0396,
      "mean_token_accuracy": 0.18708068430423735,
      "num_tokens": 62554121.0,
      "step": 28555
    },
    {
      "entropy": 5.657007503509521,
      "epoch": 2.834458118300913,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004204362536823213,
      "loss": 4.9898,
      "mean_token_accuracy": 0.18404309898614885,
      "num_tokens": 62564010.0,
      "step": 28560
    },
    {
      "entropy": 5.594596672058105,
      "epoch": 2.8349543469630807,
      "grad_norm": 1.125,
      "learning_rate": 0.00042040921719338877,
      "loss": 4.9512,
      "mean_token_accuracy": 0.18907319903373718,
      "num_tokens": 62574451.0,
      "step": 28565
    },
    {
      "entropy": 5.601953458786011,
      "epoch": 2.835450575625248,
      "grad_norm": 1.109375,
      "learning_rate": 0.000420382177098621,
      "loss": 5.016,
      "mean_token_accuracy": 0.18277707397937776,
      "num_tokens": 62585298.0,
      "step": 28570
    },
    {
      "entropy": 5.641217660903931,
      "epoch": 2.8359468042874156,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042035513339868855,
      "loss": 5.0611,
      "mean_token_accuracy": 0.18451229333877564,
      "num_tokens": 62596248.0,
      "step": 28575
    },
    {
      "entropy": 5.654402446746826,
      "epoch": 2.836443032949583,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004203280860942619,
      "loss": 5.0408,
      "mean_token_accuracy": 0.17942994385957717,
      "num_tokens": 62607233.0,
      "step": 28580
    },
    {
      "entropy": 5.6108317375183105,
      "epoch": 2.8369392616117506,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004203010351860119,
      "loss": 4.9535,
      "mean_token_accuracy": 0.1916379824280739,
      "num_tokens": 62618564.0,
      "step": 28585
    },
    {
      "entropy": 5.574761152267456,
      "epoch": 2.837435490273918,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004202739806746094,
      "loss": 5.0387,
      "mean_token_accuracy": 0.18043111115694047,
      "num_tokens": 62630049.0,
      "step": 28590
    },
    {
      "entropy": 5.626397705078125,
      "epoch": 2.837931718936086,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004202469225607252,
      "loss": 5.0846,
      "mean_token_accuracy": 0.17508404701948166,
      "num_tokens": 62640887.0,
      "step": 28595
    },
    {
      "entropy": 5.638262414932251,
      "epoch": 2.8384279475982535,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004202198608450302,
      "loss": 5.0442,
      "mean_token_accuracy": 0.1808321788907051,
      "num_tokens": 62651816.0,
      "step": 28600
    },
    {
      "entropy": 5.606627464294434,
      "epoch": 2.838924176260421,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004201927955281957,
      "loss": 4.9895,
      "mean_token_accuracy": 0.18631562292575837,
      "num_tokens": 62662601.0,
      "step": 28605
    },
    {
      "entropy": 5.590560674667358,
      "epoch": 2.8394204049225884,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004201657266108926,
      "loss": 5.0136,
      "mean_token_accuracy": 0.1903885245323181,
      "num_tokens": 62673661.0,
      "step": 28610
    },
    {
      "entropy": 5.574826240539551,
      "epoch": 2.839916633584756,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042013865409379244,
      "loss": 4.9749,
      "mean_token_accuracy": 0.19163139015436173,
      "num_tokens": 62683984.0,
      "step": 28615
    },
    {
      "entropy": 5.6371497631073,
      "epoch": 2.8404128622469234,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004201115779775663,
      "loss": 5.0356,
      "mean_token_accuracy": 0.1839341938495636,
      "num_tokens": 62695338.0,
      "step": 28620
    },
    {
      "entropy": 5.664647054672241,
      "epoch": 2.840909090909091,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004200844982628857,
      "loss": 5.0896,
      "mean_token_accuracy": 0.17782351225614548,
      "num_tokens": 62706341.0,
      "step": 28625
    },
    {
      "entropy": 5.659587478637695,
      "epoch": 2.8414053195712583,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004200574149504222,
      "loss": 5.0633,
      "mean_token_accuracy": 0.18760404139757156,
      "num_tokens": 62717895.0,
      "step": 28630
    },
    {
      "entropy": 5.637781047821045,
      "epoch": 2.841901548233426,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004200303280408474,
      "loss": 5.0314,
      "mean_token_accuracy": 0.18040051460266113,
      "num_tokens": 62729745.0,
      "step": 28635
    },
    {
      "entropy": 5.612774229049682,
      "epoch": 2.8423977768955933,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042000323753483296,
      "loss": 5.0159,
      "mean_token_accuracy": 0.18837504237890243,
      "num_tokens": 62740298.0,
      "step": 28640
    },
    {
      "entropy": 5.564932537078858,
      "epoch": 2.842894005557761,
      "grad_norm": 1.0,
      "learning_rate": 0.00041997614343305067,
      "loss": 4.9958,
      "mean_token_accuracy": 0.1893626183271408,
      "num_tokens": 62751292.0,
      "step": 28645
    },
    {
      "entropy": 5.670356225967407,
      "epoch": 2.8433902342199286,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041994904573617247,
      "loss": 5.1163,
      "mean_token_accuracy": 0.1769591227173805,
      "num_tokens": 62761789.0,
      "step": 28650
    },
    {
      "entropy": 5.603038263320923,
      "epoch": 2.843886462882096,
      "grad_norm": 1.0,
      "learning_rate": 0.00041992194444487026,
      "loss": 4.9946,
      "mean_token_accuracy": 0.19008565396070481,
      "num_tokens": 62773317.0,
      "step": 28655
    },
    {
      "entropy": 5.596129083633423,
      "epoch": 2.8443826915442636,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041989483955981607,
      "loss": 4.9673,
      "mean_token_accuracy": 0.1889987051486969,
      "num_tokens": 62783780.0,
      "step": 28660
    },
    {
      "entropy": 5.629600572586059,
      "epoch": 2.844878920206431,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041986773108168216,
      "loss": 5.0173,
      "mean_token_accuracy": 0.1837546706199646,
      "num_tokens": 62794535.0,
      "step": 28665
    },
    {
      "entropy": 5.595682954788208,
      "epoch": 2.8453751488685985,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041984061901114057,
      "loss": 4.9994,
      "mean_token_accuracy": 0.18505898416042327,
      "num_tokens": 62804410.0,
      "step": 28670
    },
    {
      "entropy": 5.628465795516968,
      "epoch": 2.845871377530766,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004198135033488638,
      "loss": 4.9929,
      "mean_token_accuracy": 0.19691797494888305,
      "num_tokens": 62814693.0,
      "step": 28675
    },
    {
      "entropy": 5.615814304351806,
      "epoch": 2.846367606192934,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00041978638409552417,
      "loss": 4.9916,
      "mean_token_accuracy": 0.18218564242124557,
      "num_tokens": 62826114.0,
      "step": 28680
    },
    {
      "entropy": 5.584632778167725,
      "epoch": 2.8468638348551014,
      "grad_norm": 1.0,
      "learning_rate": 0.0004197592612517942,
      "loss": 5.0205,
      "mean_token_accuracy": 0.1851498991250992,
      "num_tokens": 62838130.0,
      "step": 28685
    },
    {
      "entropy": 5.716651391983032,
      "epoch": 2.847360063517269,
      "grad_norm": 1.125,
      "learning_rate": 0.0004197321348183465,
      "loss": 5.0897,
      "mean_token_accuracy": 0.178537118434906,
      "num_tokens": 62849290.0,
      "step": 28690
    },
    {
      "entropy": 5.6848667621612545,
      "epoch": 2.8478562921794364,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041970500479585364,
      "loss": 5.1497,
      "mean_token_accuracy": 0.1788558304309845,
      "num_tokens": 62860888.0,
      "step": 28695
    },
    {
      "entropy": 5.667258596420288,
      "epoch": 2.848352520841604,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041967787118498854,
      "loss": 5.0619,
      "mean_token_accuracy": 0.188059900701046,
      "num_tokens": 62870231.0,
      "step": 28700
    },
    {
      "entropy": 5.60745267868042,
      "epoch": 2.8488487495037713,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004196507339864241,
      "loss": 4.9697,
      "mean_token_accuracy": 0.18564898371696473,
      "num_tokens": 62881147.0,
      "step": 28705
    },
    {
      "entropy": 5.643290328979492,
      "epoch": 2.849344978165939,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000419623593200833,
      "loss": 4.994,
      "mean_token_accuracy": 0.1884832039475441,
      "num_tokens": 62891272.0,
      "step": 28710
    },
    {
      "entropy": 5.584864664077759,
      "epoch": 2.8498412068281063,
      "grad_norm": 1.015625,
      "learning_rate": 0.00041959644882888844,
      "loss": 4.9942,
      "mean_token_accuracy": 0.18157054036855697,
      "num_tokens": 62902376.0,
      "step": 28715
    },
    {
      "entropy": 5.565850257873535,
      "epoch": 2.8503374354902737,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00041956930087126357,
      "loss": 4.985,
      "mean_token_accuracy": 0.1942770227789879,
      "num_tokens": 62913811.0,
      "step": 28720
    },
    {
      "entropy": 5.601673221588134,
      "epoch": 2.850833664152441,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00041954214932863156,
      "loss": 5.0383,
      "mean_token_accuracy": 0.17784739285707474,
      "num_tokens": 62926001.0,
      "step": 28725
    },
    {
      "entropy": 5.609562587738037,
      "epoch": 2.851329892814609,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004195149942016657,
      "loss": 5.009,
      "mean_token_accuracy": 0.18117399513721466,
      "num_tokens": 62936183.0,
      "step": 28730
    },
    {
      "entropy": 5.609285116195679,
      "epoch": 2.8518261214767766,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004194878354910394,
      "loss": 4.9785,
      "mean_token_accuracy": 0.19009705483913422,
      "num_tokens": 62947427.0,
      "step": 28735
    },
    {
      "entropy": 5.640847492218017,
      "epoch": 2.852322350138944,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041946067319742605,
      "loss": 5.0539,
      "mean_token_accuracy": 0.18539511859416963,
      "num_tokens": 62958630.0,
      "step": 28740
    },
    {
      "entropy": 5.581965112686158,
      "epoch": 2.8528185788011116,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041943350732149935,
      "loss": 4.9825,
      "mean_token_accuracy": 0.18762819468975067,
      "num_tokens": 62968856.0,
      "step": 28745
    },
    {
      "entropy": 5.602175045013428,
      "epoch": 2.853314807463279,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041940633786393295,
      "loss": 4.9107,
      "mean_token_accuracy": 0.1923521414399147,
      "num_tokens": 62978393.0,
      "step": 28750
    },
    {
      "entropy": 5.607123327255249,
      "epoch": 2.8538110361254465,
      "grad_norm": 1.3125,
      "learning_rate": 0.00041937916482540045,
      "loss": 5.0128,
      "mean_token_accuracy": 0.18885039836168288,
      "num_tokens": 62989786.0,
      "step": 28755
    },
    {
      "entropy": 5.62238917350769,
      "epoch": 2.854307264787614,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041935198820657575,
      "loss": 5.0035,
      "mean_token_accuracy": 0.19164441227912904,
      "num_tokens": 63000021.0,
      "step": 28760
    },
    {
      "entropy": 5.621928930282593,
      "epoch": 2.854803493449782,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004193248080081329,
      "loss": 5.0153,
      "mean_token_accuracy": 0.1795743703842163,
      "num_tokens": 63011398.0,
      "step": 28765
    },
    {
      "entropy": 5.633696889877319,
      "epoch": 2.8552997221119494,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004192976242307457,
      "loss": 5.0007,
      "mean_token_accuracy": 0.18380239456892014,
      "num_tokens": 63021155.0,
      "step": 28770
    },
    {
      "entropy": 5.650878143310547,
      "epoch": 2.855795950774117,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004192704368750884,
      "loss": 5.0399,
      "mean_token_accuracy": 0.1785004436969757,
      "num_tokens": 63031510.0,
      "step": 28775
    },
    {
      "entropy": 5.693984317779541,
      "epoch": 2.8562921794362843,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004192432459418351,
      "loss": 5.0478,
      "mean_token_accuracy": 0.1821679025888443,
      "num_tokens": 63041978.0,
      "step": 28780
    },
    {
      "entropy": 5.675891351699829,
      "epoch": 2.856788408098452,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004192160514316602,
      "loss": 5.067,
      "mean_token_accuracy": 0.17787381261587143,
      "num_tokens": 63052194.0,
      "step": 28785
    },
    {
      "entropy": 5.567963409423828,
      "epoch": 2.8572846367606193,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004191888533452379,
      "loss": 4.9361,
      "mean_token_accuracy": 0.19006202965974808,
      "num_tokens": 63062144.0,
      "step": 28790
    },
    {
      "entropy": 5.568313503265381,
      "epoch": 2.8577808654227868,
      "grad_norm": 1.125,
      "learning_rate": 0.0004191616516832428,
      "loss": 5.0527,
      "mean_token_accuracy": 0.17921362519264222,
      "num_tokens": 63072403.0,
      "step": 28795
    },
    {
      "entropy": 5.5934816837310795,
      "epoch": 2.8582770940849542,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004191344464463494,
      "loss": 4.9779,
      "mean_token_accuracy": 0.1897140622138977,
      "num_tokens": 63084799.0,
      "step": 28800
    },
    {
      "entropy": 5.638767290115356,
      "epoch": 2.8587733227471217,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041910723763523224,
      "loss": 5.0654,
      "mean_token_accuracy": 0.18921996653079987,
      "num_tokens": 63095437.0,
      "step": 28805
    },
    {
      "entropy": 5.6170419216156,
      "epoch": 2.859269551409289,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004190800252505661,
      "loss": 5.0788,
      "mean_token_accuracy": 0.18285082578659057,
      "num_tokens": 63106743.0,
      "step": 28810
    },
    {
      "entropy": 5.559257650375367,
      "epoch": 2.8597657800714567,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004190528092930259,
      "loss": 4.9877,
      "mean_token_accuracy": 0.18583423495292664,
      "num_tokens": 63120525.0,
      "step": 28815
    },
    {
      "entropy": 5.625993871688843,
      "epoch": 2.8602620087336246,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004190255897632863,
      "loss": 5.0165,
      "mean_token_accuracy": 0.1808594211935997,
      "num_tokens": 63131510.0,
      "step": 28820
    },
    {
      "entropy": 5.674367952346802,
      "epoch": 2.860758237395792,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041899836666202254,
      "loss": 5.0667,
      "mean_token_accuracy": 0.18552121967077256,
      "num_tokens": 63143094.0,
      "step": 28825
    },
    {
      "entropy": 5.604450464248657,
      "epoch": 2.8612544660579595,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004189711399899095,
      "loss": 4.9782,
      "mean_token_accuracy": 0.18978180438280107,
      "num_tokens": 63155159.0,
      "step": 28830
    },
    {
      "entropy": 5.555520486831665,
      "epoch": 2.861750694720127,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004189439097476224,
      "loss": 5.0161,
      "mean_token_accuracy": 0.19096735268831252,
      "num_tokens": 63167132.0,
      "step": 28835
    },
    {
      "entropy": 5.60821042060852,
      "epoch": 2.8622469233822945,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004189166759358366,
      "loss": 5.0202,
      "mean_token_accuracy": 0.18507445901632308,
      "num_tokens": 63176864.0,
      "step": 28840
    },
    {
      "entropy": 5.609105730056763,
      "epoch": 2.862743152044462,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041888943855522727,
      "loss": 5.0076,
      "mean_token_accuracy": 0.18278055787086486,
      "num_tokens": 63188003.0,
      "step": 28845
    },
    {
      "entropy": 5.578560304641724,
      "epoch": 2.86323938070663,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041886219760646997,
      "loss": 4.95,
      "mean_token_accuracy": 0.17981291711330413,
      "num_tokens": 63197634.0,
      "step": 28850
    },
    {
      "entropy": 5.581772232055664,
      "epoch": 2.8637356093687973,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004188349530902402,
      "loss": 4.9388,
      "mean_token_accuracy": 0.18693489581346512,
      "num_tokens": 63207441.0,
      "step": 28855
    },
    {
      "entropy": 5.568036079406738,
      "epoch": 2.864231838030965,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041880770500721346,
      "loss": 4.9777,
      "mean_token_accuracy": 0.18755854070186614,
      "num_tokens": 63217541.0,
      "step": 28860
    },
    {
      "entropy": 5.552620935440063,
      "epoch": 2.8647280666931323,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004187804533580655,
      "loss": 4.9479,
      "mean_token_accuracy": 0.19071064591407777,
      "num_tokens": 63228391.0,
      "step": 28865
    },
    {
      "entropy": 5.601167011260986,
      "epoch": 2.8652242953552998,
      "grad_norm": 1.0,
      "learning_rate": 0.0004187531981434721,
      "loss": 4.9855,
      "mean_token_accuracy": 0.18833220452070237,
      "num_tokens": 63238462.0,
      "step": 28870
    },
    {
      "entropy": 5.613495397567749,
      "epoch": 2.8657205240174672,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041872593936410923,
      "loss": 5.0443,
      "mean_token_accuracy": 0.18311222195625304,
      "num_tokens": 63249305.0,
      "step": 28875
    },
    {
      "entropy": 5.656917428970337,
      "epoch": 2.8662167526796347,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041869867702065266,
      "loss": 5.0414,
      "mean_token_accuracy": 0.18493347018957138,
      "num_tokens": 63260832.0,
      "step": 28880
    },
    {
      "entropy": 5.581306266784668,
      "epoch": 2.866712981341802,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004186714111137787,
      "loss": 5.0238,
      "mean_token_accuracy": 0.18204674273729324,
      "num_tokens": 63273689.0,
      "step": 28885
    },
    {
      "entropy": 5.680056142807007,
      "epoch": 2.8672092100039697,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004186441416441632,
      "loss": 5.1617,
      "mean_token_accuracy": 0.17709024101495743,
      "num_tokens": 63284912.0,
      "step": 28890
    },
    {
      "entropy": 5.689363670349121,
      "epoch": 2.867705438666137,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004186168686124825,
      "loss": 5.0619,
      "mean_token_accuracy": 0.18060151487588882,
      "num_tokens": 63296428.0,
      "step": 28895
    },
    {
      "entropy": 5.598284101486206,
      "epoch": 2.8682016673283046,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041858959201941307,
      "loss": 4.9776,
      "mean_token_accuracy": 0.19212748110294342,
      "num_tokens": 63307509.0,
      "step": 28900
    },
    {
      "entropy": 5.647677183151245,
      "epoch": 2.8686978959904725,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041856231186563107,
      "loss": 5.092,
      "mean_token_accuracy": 0.18012225180864333,
      "num_tokens": 63319194.0,
      "step": 28905
    },
    {
      "entropy": 5.592281293869019,
      "epoch": 2.86919412465264,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041853502815181306,
      "loss": 4.9446,
      "mean_token_accuracy": 0.1850243628025055,
      "num_tokens": 63329122.0,
      "step": 28910
    },
    {
      "entropy": 5.619206953048706,
      "epoch": 2.8696903533148075,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004185077408786357,
      "loss": 5.0264,
      "mean_token_accuracy": 0.18307760059833528,
      "num_tokens": 63340304.0,
      "step": 28915
    },
    {
      "entropy": 5.631560802459717,
      "epoch": 2.870186581976975,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004184804500467756,
      "loss": 5.0262,
      "mean_token_accuracy": 0.1818143308162689,
      "num_tokens": 63351029.0,
      "step": 28920
    },
    {
      "entropy": 5.571732378005981,
      "epoch": 2.8706828106391424,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041845315565690945,
      "loss": 4.9529,
      "mean_token_accuracy": 0.1871739462018013,
      "num_tokens": 63361339.0,
      "step": 28925
    },
    {
      "entropy": 5.654687452316284,
      "epoch": 2.87117903930131,
      "grad_norm": 1.015625,
      "learning_rate": 0.00041842585770971417,
      "loss": 5.024,
      "mean_token_accuracy": 0.18697432577610015,
      "num_tokens": 63373431.0,
      "step": 28930
    },
    {
      "entropy": 5.589043807983399,
      "epoch": 2.871675267963478,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004183985562058667,
      "loss": 4.9843,
      "mean_token_accuracy": 0.1876607894897461,
      "num_tokens": 63384096.0,
      "step": 28935
    },
    {
      "entropy": 5.681875371932984,
      "epoch": 2.8721714966256453,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000418371251146044,
      "loss": 5.0509,
      "mean_token_accuracy": 0.18171531111001968,
      "num_tokens": 63394852.0,
      "step": 28940
    },
    {
      "entropy": 5.626764917373658,
      "epoch": 2.872667725287813,
      "grad_norm": 1.125,
      "learning_rate": 0.0004183439425309232,
      "loss": 5.0146,
      "mean_token_accuracy": 0.1938600391149521,
      "num_tokens": 63405575.0,
      "step": 28945
    },
    {
      "entropy": 5.5865175247192385,
      "epoch": 2.8731639539499803,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004183166303611816,
      "loss": 5.0439,
      "mean_token_accuracy": 0.18719707578420638,
      "num_tokens": 63416677.0,
      "step": 28950
    },
    {
      "entropy": 5.619169902801514,
      "epoch": 2.8736601826121477,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004182893146374963,
      "loss": 5.109,
      "mean_token_accuracy": 0.17997855991125106,
      "num_tokens": 63428613.0,
      "step": 28955
    },
    {
      "entropy": 5.679248285293579,
      "epoch": 2.874156411274315,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041826199536054476,
      "loss": 5.0572,
      "mean_token_accuracy": 0.17742540687322617,
      "num_tokens": 63439321.0,
      "step": 28960
    },
    {
      "entropy": 5.6768683910369875,
      "epoch": 2.8746526399364827,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004182346725310045,
      "loss": 5.06,
      "mean_token_accuracy": 0.18394505232572556,
      "num_tokens": 63451279.0,
      "step": 28965
    },
    {
      "entropy": 5.688011360168457,
      "epoch": 2.87514886859865,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041820734614955295,
      "loss": 5.0169,
      "mean_token_accuracy": 0.18736265301704408,
      "num_tokens": 63460801.0,
      "step": 28970
    },
    {
      "entropy": 5.642605638504028,
      "epoch": 2.8756450972608176,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004181800162168678,
      "loss": 4.9942,
      "mean_token_accuracy": 0.18576611578464508,
      "num_tokens": 63470232.0,
      "step": 28975
    },
    {
      "entropy": 5.581067943572998,
      "epoch": 2.876141325922985,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004181526827336268,
      "loss": 5.0296,
      "mean_token_accuracy": 0.18538580387830733,
      "num_tokens": 63480247.0,
      "step": 28980
    },
    {
      "entropy": 5.598515033721924,
      "epoch": 2.8766375545851526,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004181253457005077,
      "loss": 5.0048,
      "mean_token_accuracy": 0.17925434857606887,
      "num_tokens": 63492104.0,
      "step": 28985
    },
    {
      "entropy": 5.637013244628906,
      "epoch": 2.8771337832473205,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041809800511818853,
      "loss": 5.0422,
      "mean_token_accuracy": 0.18365007191896437,
      "num_tokens": 63502784.0,
      "step": 28990
    },
    {
      "entropy": 5.654964447021484,
      "epoch": 2.877630011909488,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041807066098734704,
      "loss": 5.1,
      "mean_token_accuracy": 0.18766946643590926,
      "num_tokens": 63514150.0,
      "step": 28995
    },
    {
      "entropy": 5.643490028381348,
      "epoch": 2.8781262405716554,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004180433133086615,
      "loss": 5.0432,
      "mean_token_accuracy": 0.18745111674070358,
      "num_tokens": 63526682.0,
      "step": 29000
    },
    {
      "entropy": 5.639063692092895,
      "epoch": 2.878622469233823,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004180159620828101,
      "loss": 5.063,
      "mean_token_accuracy": 0.1799523115158081,
      "num_tokens": 63537448.0,
      "step": 29005
    },
    {
      "entropy": 5.596715593338013,
      "epoch": 2.8791186978959904,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004179886073104709,
      "loss": 4.9925,
      "mean_token_accuracy": 0.18141426891088486,
      "num_tokens": 63547489.0,
      "step": 29010
    },
    {
      "entropy": 5.584376287460327,
      "epoch": 2.879614926558158,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004179612489923224,
      "loss": 4.9714,
      "mean_token_accuracy": 0.19245668351650239,
      "num_tokens": 63558324.0,
      "step": 29015
    },
    {
      "entropy": 5.5944126605987545,
      "epoch": 2.880111155220326,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000417933887129043,
      "loss": 5.0213,
      "mean_token_accuracy": 0.18617270439863204,
      "num_tokens": 63568630.0,
      "step": 29020
    },
    {
      "entropy": 5.578744268417358,
      "epoch": 2.8806073838824933,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004179065217213111,
      "loss": 4.961,
      "mean_token_accuracy": 0.18285456746816636,
      "num_tokens": 63578068.0,
      "step": 29025
    },
    {
      "entropy": 5.629943561553955,
      "epoch": 2.8811036125446607,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041787915276980556,
      "loss": 5.0523,
      "mean_token_accuracy": 0.18180556297302247,
      "num_tokens": 63589103.0,
      "step": 29030
    },
    {
      "entropy": 5.608882856369019,
      "epoch": 2.881599841206828,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041785178027520486,
      "loss": 5.0099,
      "mean_token_accuracy": 0.18732813447713853,
      "num_tokens": 63600305.0,
      "step": 29035
    },
    {
      "entropy": 5.703657436370849,
      "epoch": 2.8820960698689957,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00041782440423818775,
      "loss": 5.0755,
      "mean_token_accuracy": 0.18212238550186158,
      "num_tokens": 63611910.0,
      "step": 29040
    },
    {
      "entropy": 5.635914850234985,
      "epoch": 2.882592298531163,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004177970246594332,
      "loss": 4.9841,
      "mean_token_accuracy": 0.18674272447824478,
      "num_tokens": 63622674.0,
      "step": 29045
    },
    {
      "entropy": 5.632898664474487,
      "epoch": 2.8830885271933306,
      "grad_norm": 0.984375,
      "learning_rate": 0.00041776964153962017,
      "loss": 5.0778,
      "mean_token_accuracy": 0.17722410410642625,
      "num_tokens": 63633274.0,
      "step": 29050
    },
    {
      "entropy": 5.621895217895508,
      "epoch": 2.883584755855498,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041774225487942765,
      "loss": 5.0189,
      "mean_token_accuracy": 0.18822333514690398,
      "num_tokens": 63644738.0,
      "step": 29055
    },
    {
      "entropy": 5.648908281326294,
      "epoch": 2.8840809845176656,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004177148646795348,
      "loss": 5.0698,
      "mean_token_accuracy": 0.17872775644063948,
      "num_tokens": 63655468.0,
      "step": 29060
    },
    {
      "entropy": 5.567702960968018,
      "epoch": 2.884577213179833,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041768747094062084,
      "loss": 4.9268,
      "mean_token_accuracy": 0.18942926526069642,
      "num_tokens": 63667675.0,
      "step": 29065
    },
    {
      "entropy": 5.60233507156372,
      "epoch": 2.8850734418420005,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00041766007366336494,
      "loss": 5.0512,
      "mean_token_accuracy": 0.18126652240753174,
      "num_tokens": 63679251.0,
      "step": 29070
    },
    {
      "entropy": 5.590129089355469,
      "epoch": 2.8855696705041685,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041763267284844675,
      "loss": 4.996,
      "mean_token_accuracy": 0.18964481204748154,
      "num_tokens": 63690809.0,
      "step": 29075
    },
    {
      "entropy": 5.607123947143554,
      "epoch": 2.886065899166336,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004176052684965456,
      "loss": 4.9807,
      "mean_token_accuracy": 0.17863457202911376,
      "num_tokens": 63702434.0,
      "step": 29080
    },
    {
      "entropy": 5.697435283660889,
      "epoch": 2.8865621278285034,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004175778606083411,
      "loss": 5.0853,
      "mean_token_accuracy": 0.1846032738685608,
      "num_tokens": 63713173.0,
      "step": 29085
    },
    {
      "entropy": 5.685956621170044,
      "epoch": 2.887058356490671,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004175504491845128,
      "loss": 5.0499,
      "mean_token_accuracy": 0.1832095965743065,
      "num_tokens": 63723956.0,
      "step": 29090
    },
    {
      "entropy": 5.614120292663574,
      "epoch": 2.8875545851528384,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041752303422574067,
      "loss": 5.0859,
      "mean_token_accuracy": 0.1773693010210991,
      "num_tokens": 63735280.0,
      "step": 29095
    },
    {
      "entropy": 5.683985424041748,
      "epoch": 2.888050813815006,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004174956157327043,
      "loss": 5.0385,
      "mean_token_accuracy": 0.17639011591672898,
      "num_tokens": 63746075.0,
      "step": 29100
    },
    {
      "entropy": 5.638455963134765,
      "epoch": 2.8885470424771738,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004174681937060837,
      "loss": 5.0332,
      "mean_token_accuracy": 0.18563612252473832,
      "num_tokens": 63758152.0,
      "step": 29105
    },
    {
      "entropy": 5.582948493957519,
      "epoch": 2.8890432711393412,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041744076814655893,
      "loss": 5.0206,
      "mean_token_accuracy": 0.18459441959857942,
      "num_tokens": 63768855.0,
      "step": 29110
    },
    {
      "entropy": 5.64895658493042,
      "epoch": 2.8895394998015087,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041741333905481,
      "loss": 5.0553,
      "mean_token_accuracy": 0.17719324827194213,
      "num_tokens": 63780258.0,
      "step": 29115
    },
    {
      "entropy": 5.684383583068848,
      "epoch": 2.890035728463676,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004173859064315172,
      "loss": 5.0897,
      "mean_token_accuracy": 0.1815087988972664,
      "num_tokens": 63791959.0,
      "step": 29120
    },
    {
      "entropy": 5.658141565322876,
      "epoch": 2.8905319571258437,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041735847027736067,
      "loss": 5.0607,
      "mean_token_accuracy": 0.18203603774309157,
      "num_tokens": 63802859.0,
      "step": 29125
    },
    {
      "entropy": 5.612216997146606,
      "epoch": 2.891028185788011,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004173310305930209,
      "loss": 4.9785,
      "mean_token_accuracy": 0.18276638984680177,
      "num_tokens": 63814275.0,
      "step": 29130
    },
    {
      "entropy": 5.690074253082275,
      "epoch": 2.8915244144501786,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041730358737917823,
      "loss": 5.0326,
      "mean_token_accuracy": 0.18657053112983704,
      "num_tokens": 63824633.0,
      "step": 29135
    },
    {
      "entropy": 5.662819290161133,
      "epoch": 2.892020643112346,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041727614063651313,
      "loss": 5.0263,
      "mean_token_accuracy": 0.18296183347702027,
      "num_tokens": 63836478.0,
      "step": 29140
    },
    {
      "entropy": 5.683667421340942,
      "epoch": 2.8925168717745136,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004172486903657065,
      "loss": 5.0425,
      "mean_token_accuracy": 0.17953486293554305,
      "num_tokens": 63847917.0,
      "step": 29145
    },
    {
      "entropy": 5.60505518913269,
      "epoch": 2.893013100436681,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004172212365674388,
      "loss": 5.0485,
      "mean_token_accuracy": 0.1891883984208107,
      "num_tokens": 63857258.0,
      "step": 29150
    },
    {
      "entropy": 5.604137372970581,
      "epoch": 2.8935093290988485,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041719377924239083,
      "loss": 5.1013,
      "mean_token_accuracy": 0.1799338534474373,
      "num_tokens": 63868627.0,
      "step": 29155
    },
    {
      "entropy": 5.698397541046143,
      "epoch": 2.8940055577610164,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041716631839124356,
      "loss": 5.0789,
      "mean_token_accuracy": 0.1814470738172531,
      "num_tokens": 63879954.0,
      "step": 29160
    },
    {
      "entropy": 5.761444759368897,
      "epoch": 2.894501786423184,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000417138854014678,
      "loss": 5.1364,
      "mean_token_accuracy": 0.18111622631549834,
      "num_tokens": 63891147.0,
      "step": 29165
    },
    {
      "entropy": 5.607488489151001,
      "epoch": 2.8949980150853514,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004171113861133751,
      "loss": 4.9949,
      "mean_token_accuracy": 0.1858047977089882,
      "num_tokens": 63902380.0,
      "step": 29170
    },
    {
      "entropy": 5.6213630676269535,
      "epoch": 2.895494243747519,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004170839146880161,
      "loss": 5.0575,
      "mean_token_accuracy": 0.18845338821411134,
      "num_tokens": 63912557.0,
      "step": 29175
    },
    {
      "entropy": 5.648359251022339,
      "epoch": 2.8959904724096863,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004170564397392821,
      "loss": 5.0789,
      "mean_token_accuracy": 0.17962325513362884,
      "num_tokens": 63923959.0,
      "step": 29180
    },
    {
      "entropy": 5.663371944427491,
      "epoch": 2.896486701071854,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004170289612678545,
      "loss": 5.0001,
      "mean_token_accuracy": 0.1880306124687195,
      "num_tokens": 63935600.0,
      "step": 29185
    },
    {
      "entropy": 5.661167621612549,
      "epoch": 2.8969829297340213,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041700147927441477,
      "loss": 5.154,
      "mean_token_accuracy": 0.17193630188703538,
      "num_tokens": 63946371.0,
      "step": 29190
    },
    {
      "entropy": 5.670502662658691,
      "epoch": 2.897479158396189,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041697399375964435,
      "loss": 5.0905,
      "mean_token_accuracy": 0.17679031640291215,
      "num_tokens": 63956019.0,
      "step": 29195
    },
    {
      "entropy": 5.6590677261352536,
      "epoch": 2.8979753870583567,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041694650472422467,
      "loss": 5.0346,
      "mean_token_accuracy": 0.17920539379119874,
      "num_tokens": 63967330.0,
      "step": 29200
    },
    {
      "entropy": 5.626459980010987,
      "epoch": 2.898471615720524,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041691901216883763,
      "loss": 5.0261,
      "mean_token_accuracy": 0.1844634085893631,
      "num_tokens": 63978952.0,
      "step": 29205
    },
    {
      "entropy": 5.628128337860107,
      "epoch": 2.8989678443826916,
      "grad_norm": 1.15625,
      "learning_rate": 0.000416891516094165,
      "loss": 5.0426,
      "mean_token_accuracy": 0.17726188153028488,
      "num_tokens": 63989913.0,
      "step": 29210
    },
    {
      "entropy": 5.552118825912475,
      "epoch": 2.899464073044859,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004168640165008884,
      "loss": 4.9619,
      "mean_token_accuracy": 0.1916338548064232,
      "num_tokens": 64000033.0,
      "step": 29215
    },
    {
      "entropy": 5.598502779006958,
      "epoch": 2.8999603017070266,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041683651338968994,
      "loss": 5.0274,
      "mean_token_accuracy": 0.18182701915502547,
      "num_tokens": 64010746.0,
      "step": 29220
    },
    {
      "entropy": 5.620121765136719,
      "epoch": 2.900456530369194,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041680900676125147,
      "loss": 5.0057,
      "mean_token_accuracy": 0.18350052386522292,
      "num_tokens": 64021179.0,
      "step": 29225
    },
    {
      "entropy": 5.676047134399414,
      "epoch": 2.9009527590313615,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041678149661625525,
      "loss": 5.0566,
      "mean_token_accuracy": 0.18043424636125566,
      "num_tokens": 64032953.0,
      "step": 29230
    },
    {
      "entropy": 5.606753540039063,
      "epoch": 2.901448987693529,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004167539829553835,
      "loss": 4.9084,
      "mean_token_accuracy": 0.20246005207300186,
      "num_tokens": 64042548.0,
      "step": 29235
    },
    {
      "entropy": 5.607900381088257,
      "epoch": 2.9019452163556965,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041672646577931835,
      "loss": 4.9623,
      "mean_token_accuracy": 0.18515463620424272,
      "num_tokens": 64053254.0,
      "step": 29240
    },
    {
      "entropy": 5.592686700820923,
      "epoch": 2.9024414450178644,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041669894508874225,
      "loss": 5.0137,
      "mean_token_accuracy": 0.17645054906606675,
      "num_tokens": 64063307.0,
      "step": 29245
    },
    {
      "entropy": 5.609941816329956,
      "epoch": 2.902937673680032,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00041667142088433766,
      "loss": 5.0138,
      "mean_token_accuracy": 0.18753453642129897,
      "num_tokens": 64074274.0,
      "step": 29250
    },
    {
      "entropy": 5.467594718933105,
      "epoch": 2.9034339023421993,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041664389316678713,
      "loss": 4.8845,
      "mean_token_accuracy": 0.2002396434545517,
      "num_tokens": 64084243.0,
      "step": 29255
    },
    {
      "entropy": 5.599240827560425,
      "epoch": 2.903930131004367,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004166163619367733,
      "loss": 4.9875,
      "mean_token_accuracy": 0.1867930844426155,
      "num_tokens": 64095899.0,
      "step": 29260
    },
    {
      "entropy": 5.59771842956543,
      "epoch": 2.9044263596665343,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00041658882719497876,
      "loss": 4.9031,
      "mean_token_accuracy": 0.19081873297691346,
      "num_tokens": 64107408.0,
      "step": 29265
    },
    {
      "entropy": 5.677357912063599,
      "epoch": 2.9049225883287018,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004165612889420864,
      "loss": 5.0829,
      "mean_token_accuracy": 0.1779724270105362,
      "num_tokens": 64118935.0,
      "step": 29270
    },
    {
      "entropy": 5.648597860336304,
      "epoch": 2.9054188169908692,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004165337471787792,
      "loss": 5.0202,
      "mean_token_accuracy": 0.18675286918878556,
      "num_tokens": 64129085.0,
      "step": 29275
    },
    {
      "entropy": 5.557279300689697,
      "epoch": 2.905915045653037,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041650620190574004,
      "loss": 4.9348,
      "mean_token_accuracy": 0.19268961399793624,
      "num_tokens": 64139605.0,
      "step": 29280
    },
    {
      "entropy": 5.667985105514527,
      "epoch": 2.9064112743152046,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041647865312365186,
      "loss": 5.0692,
      "mean_token_accuracy": 0.18155786395072937,
      "num_tokens": 64151171.0,
      "step": 29285
    },
    {
      "entropy": 5.674433183670044,
      "epoch": 2.906907502977372,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041645110083319805,
      "loss": 5.0423,
      "mean_token_accuracy": 0.17669578939676284,
      "num_tokens": 64162417.0,
      "step": 29290
    },
    {
      "entropy": 5.643589019775391,
      "epoch": 2.9074037316395396,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041642354503506175,
      "loss": 5.0239,
      "mean_token_accuracy": 0.17653870433568955,
      "num_tokens": 64172755.0,
      "step": 29295
    },
    {
      "entropy": 5.635941505432129,
      "epoch": 2.907899960301707,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041639598572992625,
      "loss": 5.017,
      "mean_token_accuracy": 0.18304876685142518,
      "num_tokens": 64184270.0,
      "step": 29300
    },
    {
      "entropy": 5.626755857467652,
      "epoch": 2.9083961889638745,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041636842291847495,
      "loss": 5.0775,
      "mean_token_accuracy": 0.1765637680888176,
      "num_tokens": 64195400.0,
      "step": 29305
    },
    {
      "entropy": 5.673308181762695,
      "epoch": 2.908892417626042,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041634085660139145,
      "loss": 5.0587,
      "mean_token_accuracy": 0.17848499864339828,
      "num_tokens": 64205101.0,
      "step": 29310
    },
    {
      "entropy": 5.585682487487793,
      "epoch": 2.9093886462882095,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004163132867793592,
      "loss": 4.9236,
      "mean_token_accuracy": 0.18674274384975434,
      "num_tokens": 64216047.0,
      "step": 29315
    },
    {
      "entropy": 5.592884111404419,
      "epoch": 2.909884874950377,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000416285713453062,
      "loss": 5.0227,
      "mean_token_accuracy": 0.18670656830072402,
      "num_tokens": 64226195.0,
      "step": 29320
    },
    {
      "entropy": 5.624770164489746,
      "epoch": 2.9103811036125444,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004162581366231835,
      "loss": 5.001,
      "mean_token_accuracy": 0.19118838161230087,
      "num_tokens": 64236405.0,
      "step": 29325
    },
    {
      "entropy": 5.611018896102905,
      "epoch": 2.9108773322747123,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041623055629040764,
      "loss": 4.9645,
      "mean_token_accuracy": 0.18769015073776246,
      "num_tokens": 64246989.0,
      "step": 29330
    },
    {
      "entropy": 5.572210121154785,
      "epoch": 2.91137356093688,
      "grad_norm": 1.0,
      "learning_rate": 0.0004162029724554183,
      "loss": 5.0051,
      "mean_token_accuracy": 0.18124044239521026,
      "num_tokens": 64257928.0,
      "step": 29335
    },
    {
      "entropy": 5.656641578674316,
      "epoch": 2.9118697895990473,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041617538511889946,
      "loss": 5.0241,
      "mean_token_accuracy": 0.18298175185918808,
      "num_tokens": 64268985.0,
      "step": 29340
    },
    {
      "entropy": 5.743127346038818,
      "epoch": 2.9123660182612148,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004161477942815353,
      "loss": 5.151,
      "mean_token_accuracy": 0.1774548664689064,
      "num_tokens": 64279907.0,
      "step": 29345
    },
    {
      "entropy": 5.692344427108765,
      "epoch": 2.9128622469233822,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041612019994401,
      "loss": 5.0009,
      "mean_token_accuracy": 0.18899226039648057,
      "num_tokens": 64291496.0,
      "step": 29350
    },
    {
      "entropy": 5.678136587142944,
      "epoch": 2.9133584755855497,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004160926021070078,
      "loss": 5.0209,
      "mean_token_accuracy": 0.18401512801647185,
      "num_tokens": 64302160.0,
      "step": 29355
    },
    {
      "entropy": 5.65514497756958,
      "epoch": 2.913854704247717,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004160650007712132,
      "loss": 5.0773,
      "mean_token_accuracy": 0.17563798129558564,
      "num_tokens": 64313671.0,
      "step": 29360
    },
    {
      "entropy": 5.609063673019409,
      "epoch": 2.914350932909885,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041603739593731046,
      "loss": 5.0575,
      "mean_token_accuracy": 0.18001186549663545,
      "num_tokens": 64325376.0,
      "step": 29365
    },
    {
      "entropy": 5.67064847946167,
      "epoch": 2.9148471615720526,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004160097876059842,
      "loss": 5.0615,
      "mean_token_accuracy": 0.1829256996512413,
      "num_tokens": 64336048.0,
      "step": 29370
    },
    {
      "entropy": 5.73473858833313,
      "epoch": 2.91534339023422,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004159821757779191,
      "loss": 5.0825,
      "mean_token_accuracy": 0.17518284171819687,
      "num_tokens": 64346833.0,
      "step": 29375
    },
    {
      "entropy": 5.612217950820923,
      "epoch": 2.9158396188963875,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00041595456045379975,
      "loss": 5.0071,
      "mean_token_accuracy": 0.17883428037166596,
      "num_tokens": 64359656.0,
      "step": 29380
    },
    {
      "entropy": 5.585663223266602,
      "epoch": 2.916335847558555,
      "grad_norm": 1.0,
      "learning_rate": 0.0004159269416343111,
      "loss": 4.9608,
      "mean_token_accuracy": 0.18689645528793336,
      "num_tokens": 64371130.0,
      "step": 29385
    },
    {
      "entropy": 5.590322256088257,
      "epoch": 2.9168320762207225,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000415899319320138,
      "loss": 5.0073,
      "mean_token_accuracy": 0.18723549395799638,
      "num_tokens": 64381826.0,
      "step": 29390
    },
    {
      "entropy": 5.669737100601196,
      "epoch": 2.91732830488289,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041587169351196536,
      "loss": 5.1313,
      "mean_token_accuracy": 0.1724426344037056,
      "num_tokens": 64392725.0,
      "step": 29395
    },
    {
      "entropy": 5.713866758346557,
      "epoch": 2.9178245335450574,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041584406421047825,
      "loss": 5.0126,
      "mean_token_accuracy": 0.1844126045703888,
      "num_tokens": 64402587.0,
      "step": 29400
    },
    {
      "entropy": 5.642941474914551,
      "epoch": 2.918320762207225,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041581643141636184,
      "loss": 5.0617,
      "mean_token_accuracy": 0.18560199588537216,
      "num_tokens": 64414692.0,
      "step": 29405
    },
    {
      "entropy": 5.584514570236206,
      "epoch": 2.9188169908693924,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004157887951303014,
      "loss": 5.0808,
      "mean_token_accuracy": 0.18129976391792296,
      "num_tokens": 64427278.0,
      "step": 29410
    },
    {
      "entropy": 5.6327262878417965,
      "epoch": 2.91931321953156,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004157611553529822,
      "loss": 4.9689,
      "mean_token_accuracy": 0.185126493871212,
      "num_tokens": 64438221.0,
      "step": 29415
    },
    {
      "entropy": 5.657610177993774,
      "epoch": 2.919809448193728,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041573351208508966,
      "loss": 5.1281,
      "mean_token_accuracy": 0.17385951727628707,
      "num_tokens": 64450139.0,
      "step": 29420
    },
    {
      "entropy": 5.633916902542114,
      "epoch": 2.9203056768558953,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00041570586532730926,
      "loss": 5.0326,
      "mean_token_accuracy": 0.17937820553779601,
      "num_tokens": 64461523.0,
      "step": 29425
    },
    {
      "entropy": 5.629384613037109,
      "epoch": 2.9208019055180627,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004156782150803266,
      "loss": 5.098,
      "mean_token_accuracy": 0.17801516652107238,
      "num_tokens": 64474206.0,
      "step": 29430
    },
    {
      "entropy": 5.634992456436157,
      "epoch": 2.92129813418023,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041565056134482733,
      "loss": 5.0485,
      "mean_token_accuracy": 0.18397723287343978,
      "num_tokens": 64485585.0,
      "step": 29435
    },
    {
      "entropy": 5.713293743133545,
      "epoch": 2.9217943628423977,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004156229041214972,
      "loss": 5.0725,
      "mean_token_accuracy": 0.18324072957038878,
      "num_tokens": 64497237.0,
      "step": 29440
    },
    {
      "entropy": 5.646782064437867,
      "epoch": 2.922290591504565,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004155952434110221,
      "loss": 5.0176,
      "mean_token_accuracy": 0.18667127341032028,
      "num_tokens": 64508235.0,
      "step": 29445
    },
    {
      "entropy": 5.686424112319946,
      "epoch": 2.922786820166733,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004155675792140879,
      "loss": 5.0331,
      "mean_token_accuracy": 0.1851751208305359,
      "num_tokens": 64520345.0,
      "step": 29450
    },
    {
      "entropy": 5.68010721206665,
      "epoch": 2.9232830488289006,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004155399115313806,
      "loss": 5.0571,
      "mean_token_accuracy": 0.18286439776420593,
      "num_tokens": 64531428.0,
      "step": 29455
    },
    {
      "entropy": 5.681728267669678,
      "epoch": 2.923779277491068,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041551224036358644,
      "loss": 5.0669,
      "mean_token_accuracy": 0.18079429715871811,
      "num_tokens": 64541489.0,
      "step": 29460
    },
    {
      "entropy": 5.568420171737671,
      "epoch": 2.9242755061532355,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041548456571139145,
      "loss": 4.954,
      "mean_token_accuracy": 0.190568745136261,
      "num_tokens": 64553026.0,
      "step": 29465
    },
    {
      "entropy": 5.665519142150879,
      "epoch": 2.924771734815403,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041545688757548184,
      "loss": 5.128,
      "mean_token_accuracy": 0.18119923174381256,
      "num_tokens": 64565988.0,
      "step": 29470
    },
    {
      "entropy": 5.656896114349365,
      "epoch": 2.9252679634775705,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041542920595654423,
      "loss": 5.0091,
      "mean_token_accuracy": 0.18274008631706237,
      "num_tokens": 64578098.0,
      "step": 29475
    },
    {
      "entropy": 5.591707420349121,
      "epoch": 2.925764192139738,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00041540152085526476,
      "loss": 4.9294,
      "mean_token_accuracy": 0.18278551399707793,
      "num_tokens": 64588931.0,
      "step": 29480
    },
    {
      "entropy": 5.564826488494873,
      "epoch": 2.9262604208019054,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041537383227233023,
      "loss": 4.9462,
      "mean_token_accuracy": 0.1940802216529846,
      "num_tokens": 64599636.0,
      "step": 29485
    },
    {
      "entropy": 5.6110118389129635,
      "epoch": 2.926756649464073,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004153461402084271,
      "loss": 5.0621,
      "mean_token_accuracy": 0.18470342308282853,
      "num_tokens": 64609355.0,
      "step": 29490
    },
    {
      "entropy": 5.610093832015991,
      "epoch": 2.9272528781262404,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004153184446642421,
      "loss": 5.0057,
      "mean_token_accuracy": 0.18913337886333464,
      "num_tokens": 64619969.0,
      "step": 29495
    },
    {
      "entropy": 5.631777143478393,
      "epoch": 2.927749106788408,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041529074564046204,
      "loss": 4.9825,
      "mean_token_accuracy": 0.18742512315511703,
      "num_tokens": 64629538.0,
      "step": 29500
    },
    {
      "entropy": 5.645714664459229,
      "epoch": 2.9282453354505757,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004152630431377738,
      "loss": 5.0187,
      "mean_token_accuracy": 0.18873370438814163,
      "num_tokens": 64640255.0,
      "step": 29505
    },
    {
      "entropy": 5.560592412948608,
      "epoch": 2.9287415641127432,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004152353371568643,
      "loss": 4.9412,
      "mean_token_accuracy": 0.1883455768227577,
      "num_tokens": 64650051.0,
      "step": 29510
    },
    {
      "entropy": 5.577302932739258,
      "epoch": 2.9292377927749107,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041520762769842065,
      "loss": 4.9888,
      "mean_token_accuracy": 0.19090960919857025,
      "num_tokens": 64660575.0,
      "step": 29515
    },
    {
      "entropy": 5.670549011230468,
      "epoch": 2.929734021437078,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041517991476312994,
      "loss": 5.0524,
      "mean_token_accuracy": 0.18627830892801284,
      "num_tokens": 64671486.0,
      "step": 29520
    },
    {
      "entropy": 5.64156494140625,
      "epoch": 2.9302302500992456,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004151521983516794,
      "loss": 4.9632,
      "mean_token_accuracy": 0.18571656048297883,
      "num_tokens": 64682315.0,
      "step": 29525
    },
    {
      "entropy": 5.636375045776367,
      "epoch": 2.930726478761413,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004151244784647563,
      "loss": 5.0876,
      "mean_token_accuracy": 0.18856232315301896,
      "num_tokens": 64693940.0,
      "step": 29530
    },
    {
      "entropy": 5.597050666809082,
      "epoch": 2.931222707423581,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004150967551030482,
      "loss": 4.9929,
      "mean_token_accuracy": 0.1878893107175827,
      "num_tokens": 64703347.0,
      "step": 29535
    },
    {
      "entropy": 5.6149821281433105,
      "epoch": 2.9317189360857485,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041506902826724234,
      "loss": 5.061,
      "mean_token_accuracy": 0.17858110517263412,
      "num_tokens": 64714115.0,
      "step": 29540
    },
    {
      "entropy": 5.598868131637573,
      "epoch": 2.932215164747916,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004150412979580264,
      "loss": 4.9598,
      "mean_token_accuracy": 0.18797558695077896,
      "num_tokens": 64724888.0,
      "step": 29545
    },
    {
      "entropy": 5.694367980957031,
      "epoch": 2.9327113934100835,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041501356417608816,
      "loss": 5.0635,
      "mean_token_accuracy": 0.18335746228694916,
      "num_tokens": 64735626.0,
      "step": 29550
    },
    {
      "entropy": 5.6479228973388675,
      "epoch": 2.933207622072251,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00041498582692211513,
      "loss": 4.9875,
      "mean_token_accuracy": 0.18442681133747102,
      "num_tokens": 64747807.0,
      "step": 29555
    },
    {
      "entropy": 5.571166610717773,
      "epoch": 2.9337038507344184,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041495808619679525,
      "loss": 4.9692,
      "mean_token_accuracy": 0.1885893911123276,
      "num_tokens": 64756845.0,
      "step": 29560
    },
    {
      "entropy": 5.574899578094483,
      "epoch": 2.934200079396586,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041493034200081643,
      "loss": 5.0601,
      "mean_token_accuracy": 0.18705544173717498,
      "num_tokens": 64767316.0,
      "step": 29565
    },
    {
      "entropy": 5.618352890014648,
      "epoch": 2.9346963080587534,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004149025943348666,
      "loss": 4.9889,
      "mean_token_accuracy": 0.1922389194369316,
      "num_tokens": 64778329.0,
      "step": 29570
    },
    {
      "entropy": 5.6199562549591064,
      "epoch": 2.935192536720921,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041487484319963395,
      "loss": 5.0389,
      "mean_token_accuracy": 0.18682349920272828,
      "num_tokens": 64789215.0,
      "step": 29575
    },
    {
      "entropy": 5.616003704071045,
      "epoch": 2.9356887653830883,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004148470885958066,
      "loss": 4.9895,
      "mean_token_accuracy": 0.18633807003498076,
      "num_tokens": 64801878.0,
      "step": 29580
    },
    {
      "entropy": 5.5476514339447025,
      "epoch": 2.936184994045256,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004148193305240727,
      "loss": 4.9175,
      "mean_token_accuracy": 0.18343037813901902,
      "num_tokens": 64813442.0,
      "step": 29585
    },
    {
      "entropy": 5.620064306259155,
      "epoch": 2.9366812227074237,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00041479156898512076,
      "loss": 4.9875,
      "mean_token_accuracy": 0.18604617267847062,
      "num_tokens": 64824095.0,
      "step": 29590
    },
    {
      "entropy": 5.634258270263672,
      "epoch": 2.937177451369591,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004147638039796391,
      "loss": 5.0151,
      "mean_token_accuracy": 0.18663536608219147,
      "num_tokens": 64833101.0,
      "step": 29595
    },
    {
      "entropy": 5.5368232250213625,
      "epoch": 2.9376736800317587,
      "grad_norm": 1.015625,
      "learning_rate": 0.00041473603550831636,
      "loss": 4.9443,
      "mean_token_accuracy": 0.18840275555849076,
      "num_tokens": 64844315.0,
      "step": 29600
    },
    {
      "entropy": 5.630424737930298,
      "epoch": 2.938169908693926,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041470826357184096,
      "loss": 5.0311,
      "mean_token_accuracy": 0.1833059310913086,
      "num_tokens": 64855090.0,
      "step": 29605
    },
    {
      "entropy": 5.558199214935303,
      "epoch": 2.9386661373560936,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004146804881709017,
      "loss": 4.9848,
      "mean_token_accuracy": 0.18666452616453172,
      "num_tokens": 64865093.0,
      "step": 29610
    },
    {
      "entropy": 5.715952444076538,
      "epoch": 2.939162366018261,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004146527093061873,
      "loss": 5.0909,
      "mean_token_accuracy": 0.1765726238489151,
      "num_tokens": 64876250.0,
      "step": 29615
    },
    {
      "entropy": 5.672224235534668,
      "epoch": 2.939658594680429,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004146249269783867,
      "loss": 5.0166,
      "mean_token_accuracy": 0.18439213037490845,
      "num_tokens": 64886339.0,
      "step": 29620
    },
    {
      "entropy": 5.66644697189331,
      "epoch": 2.9401548233425965,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004145971411881886,
      "loss": 5.1025,
      "mean_token_accuracy": 0.18037427514791488,
      "num_tokens": 64899473.0,
      "step": 29625
    },
    {
      "entropy": 5.659638977050781,
      "epoch": 2.940651052004764,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041456935193628236,
      "loss": 5.0485,
      "mean_token_accuracy": 0.1827987626194954,
      "num_tokens": 64909919.0,
      "step": 29630
    },
    {
      "entropy": 5.6489335060119625,
      "epoch": 2.9411472806669314,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004145415592233569,
      "loss": 5.0512,
      "mean_token_accuracy": 0.17928548604249955,
      "num_tokens": 64921069.0,
      "step": 29635
    },
    {
      "entropy": 5.551986169815064,
      "epoch": 2.941643509329099,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041451376305010144,
      "loss": 4.9699,
      "mean_token_accuracy": 0.1946537584066391,
      "num_tokens": 64932173.0,
      "step": 29640
    },
    {
      "entropy": 5.6054114818573,
      "epoch": 2.9421397379912664,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004144859634172053,
      "loss": 4.9295,
      "mean_token_accuracy": 0.19296609461307526,
      "num_tokens": 64941650.0,
      "step": 29645
    },
    {
      "entropy": 5.562364912033081,
      "epoch": 2.942635966653434,
      "grad_norm": 1.0,
      "learning_rate": 0.00041445816032535784,
      "loss": 4.9798,
      "mean_token_accuracy": 0.19262303709983825,
      "num_tokens": 64952520.0,
      "step": 29650
    },
    {
      "entropy": 5.589870452880859,
      "epoch": 2.9431321953156013,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041443035377524844,
      "loss": 5.1096,
      "mean_token_accuracy": 0.17440439462661744,
      "num_tokens": 64965030.0,
      "step": 29655
    },
    {
      "entropy": 5.587843370437622,
      "epoch": 2.943628423977769,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004144025437675668,
      "loss": 4.9618,
      "mean_token_accuracy": 0.18726728707551957,
      "num_tokens": 64976260.0,
      "step": 29660
    },
    {
      "entropy": 5.693049335479737,
      "epoch": 2.9441246526399363,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004143747303030025,
      "loss": 5.1138,
      "mean_token_accuracy": 0.1846769705414772,
      "num_tokens": 64988049.0,
      "step": 29665
    },
    {
      "entropy": 5.632381248474121,
      "epoch": 2.9446208813021038,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041434691338224506,
      "loss": 4.9754,
      "mean_token_accuracy": 0.18913608342409133,
      "num_tokens": 64998506.0,
      "step": 29670
    },
    {
      "entropy": 5.608209943771362,
      "epoch": 2.9451171099642717,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004143190930059846,
      "loss": 4.9853,
      "mean_token_accuracy": 0.19040031731128693,
      "num_tokens": 65010533.0,
      "step": 29675
    },
    {
      "entropy": 5.665458393096924,
      "epoch": 2.945613338626439,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004142912691749107,
      "loss": 5.0553,
      "mean_token_accuracy": 0.18126529455184937,
      "num_tokens": 65021238.0,
      "step": 29680
    },
    {
      "entropy": 5.691421556472778,
      "epoch": 2.9461095672886066,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004142634418897135,
      "loss": 5.097,
      "mean_token_accuracy": 0.17697432190179824,
      "num_tokens": 65033251.0,
      "step": 29685
    },
    {
      "entropy": 5.661848640441894,
      "epoch": 2.946605795950774,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004142356111510831,
      "loss": 5.0182,
      "mean_token_accuracy": 0.18054499328136445,
      "num_tokens": 65043845.0,
      "step": 29690
    },
    {
      "entropy": 5.579871892929077,
      "epoch": 2.9471020246129416,
      "grad_norm": 1.0,
      "learning_rate": 0.00041420777695970945,
      "loss": 4.997,
      "mean_token_accuracy": 0.18567252457141875,
      "num_tokens": 65055395.0,
      "step": 29695
    },
    {
      "entropy": 5.646155071258545,
      "epoch": 2.947598253275109,
      "grad_norm": 1.078125,
      "learning_rate": 0.000414179939316283,
      "loss": 5.0429,
      "mean_token_accuracy": 0.18337540030479432,
      "num_tokens": 65064931.0,
      "step": 29700
    },
    {
      "entropy": 5.656322288513183,
      "epoch": 2.948094481937277,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041415209822149397,
      "loss": 5.0102,
      "mean_token_accuracy": 0.18660636395215988,
      "num_tokens": 65074678.0,
      "step": 29705
    },
    {
      "entropy": 5.64708890914917,
      "epoch": 2.9485907105994444,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004141242536760327,
      "loss": 5.0161,
      "mean_token_accuracy": 0.1873234912753105,
      "num_tokens": 65084562.0,
      "step": 29710
    },
    {
      "entropy": 5.6484424591064455,
      "epoch": 2.949086939261612,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004140964056805897,
      "loss": 5.0187,
      "mean_token_accuracy": 0.18499169647693633,
      "num_tokens": 65095924.0,
      "step": 29715
    },
    {
      "entropy": 5.647116708755493,
      "epoch": 2.9495831679237794,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004140685542358556,
      "loss": 5.0511,
      "mean_token_accuracy": 0.18970248103141785,
      "num_tokens": 65106802.0,
      "step": 29720
    },
    {
      "entropy": 5.624762439727784,
      "epoch": 2.950079396585947,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004140406993425211,
      "loss": 5.0936,
      "mean_token_accuracy": 0.1778579905629158,
      "num_tokens": 65118196.0,
      "step": 29725
    },
    {
      "entropy": 5.603924036026001,
      "epoch": 2.9505756252481143,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004140128410012767,
      "loss": 5.0024,
      "mean_token_accuracy": 0.19395073056221007,
      "num_tokens": 65129924.0,
      "step": 29730
    },
    {
      "entropy": 5.5188250064849855,
      "epoch": 2.951071853910282,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041398497921281354,
      "loss": 4.8749,
      "mean_token_accuracy": 0.19657872915267943,
      "num_tokens": 65140994.0,
      "step": 29735
    },
    {
      "entropy": 5.6372123718261715,
      "epoch": 2.9515680825724493,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041395711397782236,
      "loss": 5.094,
      "mean_token_accuracy": 0.18524870723485948,
      "num_tokens": 65153261.0,
      "step": 29740
    },
    {
      "entropy": 5.724872255325318,
      "epoch": 2.9520643112346168,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004139292452969941,
      "loss": 5.1507,
      "mean_token_accuracy": 0.1777268722653389,
      "num_tokens": 65164127.0,
      "step": 29745
    },
    {
      "entropy": 5.6130931854248045,
      "epoch": 2.9525605398967842,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00041390137317102,
      "loss": 4.9712,
      "mean_token_accuracy": 0.184186090528965,
      "num_tokens": 65175603.0,
      "step": 29750
    },
    {
      "entropy": 5.609373140335083,
      "epoch": 2.9530567685589517,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004138734976005911,
      "loss": 4.9555,
      "mean_token_accuracy": 0.18469937592744828,
      "num_tokens": 65186780.0,
      "step": 29755
    },
    {
      "entropy": 5.574729681015015,
      "epoch": 2.9535529972211196,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004138456185863986,
      "loss": 4.9223,
      "mean_token_accuracy": 0.19470570534467696,
      "num_tokens": 65198883.0,
      "step": 29760
    },
    {
      "entropy": 5.604518413543701,
      "epoch": 2.954049225883287,
      "grad_norm": 1.125,
      "learning_rate": 0.00041381773612913413,
      "loss": 5.0205,
      "mean_token_accuracy": 0.18472900837659836,
      "num_tokens": 65209088.0,
      "step": 29765
    },
    {
      "entropy": 5.581466388702393,
      "epoch": 2.9545454545454546,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004137898502294889,
      "loss": 4.9926,
      "mean_token_accuracy": 0.18780494183301927,
      "num_tokens": 65220816.0,
      "step": 29770
    },
    {
      "entropy": 5.629365062713623,
      "epoch": 2.955041683207622,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004137619608881544,
      "loss": 4.9892,
      "mean_token_accuracy": 0.1852090209722519,
      "num_tokens": 65232549.0,
      "step": 29775
    },
    {
      "entropy": 5.640566730499268,
      "epoch": 2.9555379118697895,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004137340681058223,
      "loss": 5.0243,
      "mean_token_accuracy": 0.18084525018930436,
      "num_tokens": 65243691.0,
      "step": 29780
    },
    {
      "entropy": 5.6202342987060545,
      "epoch": 2.956034140531957,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041370617188318424,
      "loss": 5.0419,
      "mean_token_accuracy": 0.18914755135774614,
      "num_tokens": 65254651.0,
      "step": 29785
    },
    {
      "entropy": 5.608315086364746,
      "epoch": 2.956530369194125,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041367827222093195,
      "loss": 4.9933,
      "mean_token_accuracy": 0.1917033612728119,
      "num_tokens": 65264669.0,
      "step": 29790
    },
    {
      "entropy": 5.610342025756836,
      "epoch": 2.9570265978562924,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004136503691197574,
      "loss": 5.0166,
      "mean_token_accuracy": 0.1837298735976219,
      "num_tokens": 65274663.0,
      "step": 29795
    },
    {
      "entropy": 5.62825837135315,
      "epoch": 2.95752282651846,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041362246258035254,
      "loss": 5.0301,
      "mean_token_accuracy": 0.1875256985425949,
      "num_tokens": 65286383.0,
      "step": 29800
    },
    {
      "entropy": 5.576198625564575,
      "epoch": 2.9580190551806274,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004135945526034091,
      "loss": 4.87,
      "mean_token_accuracy": 0.19135010093450547,
      "num_tokens": 65297156.0,
      "step": 29805
    },
    {
      "entropy": 5.669165086746216,
      "epoch": 2.958515283842795,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041356663918961956,
      "loss": 5.0663,
      "mean_token_accuracy": 0.18182418942451478,
      "num_tokens": 65306836.0,
      "step": 29810
    },
    {
      "entropy": 5.679946994781494,
      "epoch": 2.9590115125049623,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041353872233967595,
      "loss": 5.0835,
      "mean_token_accuracy": 0.178165502846241,
      "num_tokens": 65318355.0,
      "step": 29815
    },
    {
      "entropy": 5.633596897125244,
      "epoch": 2.95950774116713,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004135108020542705,
      "loss": 5.0622,
      "mean_token_accuracy": 0.18110385835170745,
      "num_tokens": 65330039.0,
      "step": 29820
    },
    {
      "entropy": 5.598168992996216,
      "epoch": 2.9600039698292973,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041348287833409557,
      "loss": 4.9628,
      "mean_token_accuracy": 0.18908904492855072,
      "num_tokens": 65339771.0,
      "step": 29825
    },
    {
      "entropy": 5.62923035621643,
      "epoch": 2.9605001984914647,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004134549511798438,
      "loss": 5.0332,
      "mean_token_accuracy": 0.18572789579629898,
      "num_tokens": 65352108.0,
      "step": 29830
    },
    {
      "entropy": 5.598220777511597,
      "epoch": 2.960996427153632,
      "grad_norm": 1.125,
      "learning_rate": 0.0004134270205922075,
      "loss": 4.977,
      "mean_token_accuracy": 0.18084535747766495,
      "num_tokens": 65364117.0,
      "step": 29835
    },
    {
      "entropy": 5.577403354644775,
      "epoch": 2.9614926558157997,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041339908657187933,
      "loss": 4.9575,
      "mean_token_accuracy": 0.1881212204694748,
      "num_tokens": 65374168.0,
      "step": 29840
    },
    {
      "entropy": 5.708590745925903,
      "epoch": 2.9619888844779676,
      "grad_norm": 1.25,
      "learning_rate": 0.0004133711491195522,
      "loss": 5.1473,
      "mean_token_accuracy": 0.1768685609102249,
      "num_tokens": 65385243.0,
      "step": 29845
    },
    {
      "entropy": 5.654151678085327,
      "epoch": 2.962485113140135,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00041334320823591857,
      "loss": 5.033,
      "mean_token_accuracy": 0.18130577504634857,
      "num_tokens": 65396674.0,
      "step": 29850
    },
    {
      "entropy": 5.663484859466553,
      "epoch": 2.9629813418023025,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041331526392167155,
      "loss": 5.0863,
      "mean_token_accuracy": 0.17268675565719604,
      "num_tokens": 65408121.0,
      "step": 29855
    },
    {
      "entropy": 5.651930141448974,
      "epoch": 2.96347757046447,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000413287316177504,
      "loss": 5.0462,
      "mean_token_accuracy": 0.18362269699573516,
      "num_tokens": 65420093.0,
      "step": 29860
    },
    {
      "entropy": 5.602506637573242,
      "epoch": 2.9639737991266375,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041325936500410905,
      "loss": 4.9693,
      "mean_token_accuracy": 0.18311800211668014,
      "num_tokens": 65431338.0,
      "step": 29865
    },
    {
      "entropy": 5.5937995433807375,
      "epoch": 2.964470027788805,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041323141040217976,
      "loss": 4.9917,
      "mean_token_accuracy": 0.18663060814142227,
      "num_tokens": 65442305.0,
      "step": 29870
    },
    {
      "entropy": 5.540653371810913,
      "epoch": 2.9649662564509724,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041320345237240923,
      "loss": 4.9619,
      "mean_token_accuracy": 0.1883355885744095,
      "num_tokens": 65453028.0,
      "step": 29875
    },
    {
      "entropy": 5.577840805053711,
      "epoch": 2.9654624851131404,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000413175490915491,
      "loss": 5.0635,
      "mean_token_accuracy": 0.18705219477415086,
      "num_tokens": 65463374.0,
      "step": 29880
    },
    {
      "entropy": 5.624992227554321,
      "epoch": 2.965958713775308,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004131475260321183,
      "loss": 4.9725,
      "mean_token_accuracy": 0.1825333371758461,
      "num_tokens": 65474144.0,
      "step": 29885
    },
    {
      "entropy": 5.57177996635437,
      "epoch": 2.9664549424374753,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041311955772298464,
      "loss": 4.9649,
      "mean_token_accuracy": 0.204286590218544,
      "num_tokens": 65485453.0,
      "step": 29890
    },
    {
      "entropy": 5.619845247268676,
      "epoch": 2.966951171099643,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004130915859887837,
      "loss": 4.9936,
      "mean_token_accuracy": 0.184697787463665,
      "num_tokens": 65496400.0,
      "step": 29895
    },
    {
      "entropy": 5.622707033157349,
      "epoch": 2.9674473997618103,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004130636108302088,
      "loss": 5.0593,
      "mean_token_accuracy": 0.1794567123055458,
      "num_tokens": 65508524.0,
      "step": 29900
    },
    {
      "entropy": 5.48760838508606,
      "epoch": 2.9679436284239777,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041303563224795393,
      "loss": 4.8959,
      "mean_token_accuracy": 0.1939314603805542,
      "num_tokens": 65519264.0,
      "step": 29905
    },
    {
      "entropy": 5.553054285049439,
      "epoch": 2.968439857086145,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004130076502427128,
      "loss": 4.9157,
      "mean_token_accuracy": 0.18867608159780502,
      "num_tokens": 65531049.0,
      "step": 29910
    },
    {
      "entropy": 5.603642559051513,
      "epoch": 2.9689360857483127,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041297966481517937,
      "loss": 5.048,
      "mean_token_accuracy": 0.18297618329524995,
      "num_tokens": 65543126.0,
      "step": 29915
    },
    {
      "entropy": 5.674145889282227,
      "epoch": 2.96943231441048,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004129516759660475,
      "loss": 5.1009,
      "mean_token_accuracy": 0.1718280479311943,
      "num_tokens": 65555050.0,
      "step": 29920
    },
    {
      "entropy": 5.677469635009766,
      "epoch": 2.9699285430726476,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041292368369601136,
      "loss": 5.0713,
      "mean_token_accuracy": 0.1916345998644829,
      "num_tokens": 65564156.0,
      "step": 29925
    },
    {
      "entropy": 5.672990131378174,
      "epoch": 2.9704247717348156,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000412895688005765,
      "loss": 5.1016,
      "mean_token_accuracy": 0.18110866844654083,
      "num_tokens": 65574639.0,
      "step": 29930
    },
    {
      "entropy": 5.6546159267425535,
      "epoch": 2.970921000396983,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004128676888960028,
      "loss": 4.9726,
      "mean_token_accuracy": 0.18655480295419694,
      "num_tokens": 65585628.0,
      "step": 29935
    },
    {
      "entropy": 5.599910163879395,
      "epoch": 2.9714172290591505,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041283968636741903,
      "loss": 4.986,
      "mean_token_accuracy": 0.1884813502430916,
      "num_tokens": 65595544.0,
      "step": 29940
    },
    {
      "entropy": 5.592312002182007,
      "epoch": 2.971913457721318,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004128116804207081,
      "loss": 5.0703,
      "mean_token_accuracy": 0.18596239537000656,
      "num_tokens": 65607963.0,
      "step": 29945
    },
    {
      "entropy": 5.623130893707275,
      "epoch": 2.9724096863834855,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004127836710565642,
      "loss": 4.9729,
      "mean_token_accuracy": 0.19936721175909042,
      "num_tokens": 65619881.0,
      "step": 29950
    },
    {
      "entropy": 5.663489818572998,
      "epoch": 2.972905915045653,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041275565827568234,
      "loss": 5.0849,
      "mean_token_accuracy": 0.18182734847068788,
      "num_tokens": 65629967.0,
      "step": 29955
    },
    {
      "entropy": 5.625498104095459,
      "epoch": 2.9734021437078204,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000412727642078757,
      "loss": 5.0084,
      "mean_token_accuracy": 0.1830184355378151,
      "num_tokens": 65640804.0,
      "step": 29960
    },
    {
      "entropy": 5.635215902328492,
      "epoch": 2.9738983723699883,
      "grad_norm": 1.0,
      "learning_rate": 0.0004126996224664829,
      "loss": 5.0979,
      "mean_token_accuracy": 0.17760427594184874,
      "num_tokens": 65652355.0,
      "step": 29965
    },
    {
      "entropy": 5.602761745452881,
      "epoch": 2.974394601032156,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041267159943955483,
      "loss": 5.0452,
      "mean_token_accuracy": 0.17833778262138367,
      "num_tokens": 65664047.0,
      "step": 29970
    },
    {
      "entropy": 5.641275835037232,
      "epoch": 2.9748908296943233,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004126435729986678,
      "loss": 5.0161,
      "mean_token_accuracy": 0.18311435729265213,
      "num_tokens": 65674942.0,
      "step": 29975
    },
    {
      "entropy": 5.637739372253418,
      "epoch": 2.9753870583564908,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041261554314451674,
      "loss": 4.9688,
      "mean_token_accuracy": 0.18891677558422088,
      "num_tokens": 65685285.0,
      "step": 29980
    },
    {
      "entropy": 5.566508436203003,
      "epoch": 2.9758832870186582,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004125875098777967,
      "loss": 4.956,
      "mean_token_accuracy": 0.19560834169387817,
      "num_tokens": 65695967.0,
      "step": 29985
    },
    {
      "entropy": 5.5857398986816404,
      "epoch": 2.9763795156808257,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041255947319920297,
      "loss": 5.0375,
      "mean_token_accuracy": 0.1842626839876175,
      "num_tokens": 65707434.0,
      "step": 29990
    },
    {
      "entropy": 5.552202415466309,
      "epoch": 2.976875744342993,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041253143310943064,
      "loss": 4.9788,
      "mean_token_accuracy": 0.19348690509796143,
      "num_tokens": 65717836.0,
      "step": 29995
    },
    {
      "entropy": 5.625916576385498,
      "epoch": 2.9773719730051607,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041250338960917515,
      "loss": 5.0217,
      "mean_token_accuracy": 0.18776267915964126,
      "num_tokens": 65728527.0,
      "step": 30000
    },
    {
      "epoch": 2.9773719730051607,
      "eval_entropy": 5.409579987366201,
      "eval_loss": 5.126099109649658,
      "eval_mean_token_accuracy": 0.18560175063966092,
      "eval_num_tokens": 65728527.0,
      "eval_runtime": 20.5435,
      "eval_samples_per_second": 1638.909,
      "eval_steps_per_second": 204.882,
      "step": 30000
    },
    {
      "entropy": 5.584886503219605,
      "epoch": 2.977868201667328,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00041247534269913197,
      "loss": 4.9784,
      "mean_token_accuracy": 0.1897513285279274,
      "num_tokens": 65739724.0,
      "step": 30005
    },
    {
      "entropy": 5.634951400756836,
      "epoch": 2.9783644303294956,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004124472923799964,
      "loss": 5.0566,
      "mean_token_accuracy": 0.18495203405618668,
      "num_tokens": 65750057.0,
      "step": 30010
    },
    {
      "entropy": 5.639446353912353,
      "epoch": 2.9788606589916635,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004124192386524642,
      "loss": 5.0397,
      "mean_token_accuracy": 0.18790996670722962,
      "num_tokens": 65760534.0,
      "step": 30015
    },
    {
      "entropy": 5.641406631469726,
      "epoch": 2.979356887653831,
      "grad_norm": 1.0625,
      "learning_rate": 0.000412391181517231,
      "loss": 5.0894,
      "mean_token_accuracy": 0.18363550156354905,
      "num_tokens": 65772524.0,
      "step": 30020
    },
    {
      "entropy": 5.655934286117554,
      "epoch": 2.9798531163159985,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041236312097499257,
      "loss": 5.0155,
      "mean_token_accuracy": 0.17715261429548262,
      "num_tokens": 65783003.0,
      "step": 30025
    },
    {
      "entropy": 5.577913665771485,
      "epoch": 2.980349344978166,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004123350570264447,
      "loss": 5.0107,
      "mean_token_accuracy": 0.18348318785429002,
      "num_tokens": 65793219.0,
      "step": 30030
    },
    {
      "entropy": 5.651893377304077,
      "epoch": 2.9808455736403334,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041230698967228334,
      "loss": 5.105,
      "mean_token_accuracy": 0.18621619194746017,
      "num_tokens": 65804105.0,
      "step": 30035
    },
    {
      "entropy": 5.773953819274903,
      "epoch": 2.981341802302501,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004122789189132045,
      "loss": 5.1046,
      "mean_token_accuracy": 0.1863139882683754,
      "num_tokens": 65814211.0,
      "step": 30040
    },
    {
      "entropy": 5.726456785202027,
      "epoch": 2.9818380309646684,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004122508447499043,
      "loss": 5.1185,
      "mean_token_accuracy": 0.17680369913578034,
      "num_tokens": 65825304.0,
      "step": 30045
    },
    {
      "entropy": 5.610083198547363,
      "epoch": 2.9823342596268363,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041222276718307896,
      "loss": 4.9546,
      "mean_token_accuracy": 0.19109073430299758,
      "num_tokens": 65835933.0,
      "step": 30050
    },
    {
      "entropy": 5.57972297668457,
      "epoch": 2.9828304882890038,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004121946862134246,
      "loss": 5.0433,
      "mean_token_accuracy": 0.18642766028642654,
      "num_tokens": 65846479.0,
      "step": 30055
    },
    {
      "entropy": 5.589495372772217,
      "epoch": 2.9833267169511712,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041216660184163767,
      "loss": 4.9402,
      "mean_token_accuracy": 0.19205747693777084,
      "num_tokens": 65857033.0,
      "step": 30060
    },
    {
      "entropy": 5.687922430038452,
      "epoch": 2.9838229456133387,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041213851406841456,
      "loss": 5.1116,
      "mean_token_accuracy": 0.17847366333007814,
      "num_tokens": 65867983.0,
      "step": 30065
    },
    {
      "entropy": 5.694590377807617,
      "epoch": 2.984319174275506,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004121104228944518,
      "loss": 5.0493,
      "mean_token_accuracy": 0.17700741887092591,
      "num_tokens": 65878535.0,
      "step": 30070
    },
    {
      "entropy": 5.684807872772216,
      "epoch": 2.9848154029376737,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004120823283204461,
      "loss": 5.018,
      "mean_token_accuracy": 0.1896379515528679,
      "num_tokens": 65890484.0,
      "step": 30075
    },
    {
      "entropy": 5.576067781448364,
      "epoch": 2.985311631599841,
      "grad_norm": 1.09375,
      "learning_rate": 0.000412054230347094,
      "loss": 4.9861,
      "mean_token_accuracy": 0.18700068593025207,
      "num_tokens": 65901220.0,
      "step": 30080
    },
    {
      "entropy": 5.672197580337524,
      "epoch": 2.9858078602620086,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004120261289750923,
      "loss": 5.0331,
      "mean_token_accuracy": 0.18340465873479844,
      "num_tokens": 65911875.0,
      "step": 30085
    },
    {
      "entropy": 5.609361076354981,
      "epoch": 2.986304088924176,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041199802420513785,
      "loss": 5.0124,
      "mean_token_accuracy": 0.18898148983716964,
      "num_tokens": 65921915.0,
      "step": 30090
    },
    {
      "entropy": 5.56067681312561,
      "epoch": 2.9868003175863436,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004119699160379276,
      "loss": 4.9735,
      "mean_token_accuracy": 0.18820373713970184,
      "num_tokens": 65932061.0,
      "step": 30095
    },
    {
      "entropy": 5.587253999710083,
      "epoch": 2.987296546248511,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041194180447415855,
      "loss": 4.9712,
      "mean_token_accuracy": 0.1893375799059868,
      "num_tokens": 65942818.0,
      "step": 30100
    },
    {
      "entropy": 5.626484155654907,
      "epoch": 2.987792774910679,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004119136895145279,
      "loss": 5.0122,
      "mean_token_accuracy": 0.18548336178064345,
      "num_tokens": 65953715.0,
      "step": 30105
    },
    {
      "entropy": 5.699669265747071,
      "epoch": 2.9882890035728464,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004118855711597327,
      "loss": 5.067,
      "mean_token_accuracy": 0.17751636505126953,
      "num_tokens": 65964235.0,
      "step": 30110
    },
    {
      "entropy": 5.587323617935181,
      "epoch": 2.988785232235014,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041185744941047033,
      "loss": 5.0291,
      "mean_token_accuracy": 0.18020953088998795,
      "num_tokens": 65976457.0,
      "step": 30115
    },
    {
      "entropy": 5.599376487731933,
      "epoch": 2.9892814608971814,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004118293242674382,
      "loss": 5.013,
      "mean_token_accuracy": 0.1833759754896164,
      "num_tokens": 65986810.0,
      "step": 30120
    },
    {
      "entropy": 5.648756504058838,
      "epoch": 2.989777689559349,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004118011957313336,
      "loss": 5.114,
      "mean_token_accuracy": 0.17542362958192825,
      "num_tokens": 65998019.0,
      "step": 30125
    },
    {
      "entropy": 5.656954526901245,
      "epoch": 2.9902739182215163,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041177306380285405,
      "loss": 5.0337,
      "mean_token_accuracy": 0.18912093937397004,
      "num_tokens": 66009404.0,
      "step": 30130
    },
    {
      "entropy": 5.587200498580932,
      "epoch": 2.9907701468836843,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004117449284826974,
      "loss": 4.9548,
      "mean_token_accuracy": 0.18638341277837753,
      "num_tokens": 66020610.0,
      "step": 30135
    },
    {
      "entropy": 5.648459625244141,
      "epoch": 2.9912663755458517,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041171678977156104,
      "loss": 5.0535,
      "mean_token_accuracy": 0.17769998908042908,
      "num_tokens": 66032576.0,
      "step": 30140
    },
    {
      "entropy": 5.633008527755737,
      "epoch": 2.991762604208019,
      "grad_norm": 1.125,
      "learning_rate": 0.00041168864767014295,
      "loss": 5.0235,
      "mean_token_accuracy": 0.19214165657758714,
      "num_tokens": 66043378.0,
      "step": 30145
    },
    {
      "entropy": 5.634005737304688,
      "epoch": 2.9922588328701867,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000411660502179141,
      "loss": 4.9563,
      "mean_token_accuracy": 0.18751318156719207,
      "num_tokens": 66053960.0,
      "step": 30150
    },
    {
      "entropy": 5.62046251296997,
      "epoch": 2.992755061532354,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004116323532992529,
      "loss": 5.0328,
      "mean_token_accuracy": 0.182614766061306,
      "num_tokens": 66066069.0,
      "step": 30155
    },
    {
      "entropy": 5.595282745361328,
      "epoch": 2.9932512901945216,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041160420103117704,
      "loss": 4.9865,
      "mean_token_accuracy": 0.18743032962083817,
      "num_tokens": 66076789.0,
      "step": 30160
    },
    {
      "entropy": 5.604577255249024,
      "epoch": 2.993747518856689,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041157604537561123,
      "loss": 4.971,
      "mean_token_accuracy": 0.1845577985048294,
      "num_tokens": 66088267.0,
      "step": 30165
    },
    {
      "entropy": 5.563841390609741,
      "epoch": 2.9942437475188566,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041154788633325387,
      "loss": 4.9537,
      "mean_token_accuracy": 0.1862846404314041,
      "num_tokens": 66098789.0,
      "step": 30170
    },
    {
      "entropy": 5.613836240768433,
      "epoch": 2.994739976181024,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004115197239048031,
      "loss": 5.002,
      "mean_token_accuracy": 0.1824750557541847,
      "num_tokens": 66110881.0,
      "step": 30175
    },
    {
      "entropy": 5.645247459411621,
      "epoch": 2.9952362048431915,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041149155809095734,
      "loss": 5.0051,
      "mean_token_accuracy": 0.17692070007324218,
      "num_tokens": 66121728.0,
      "step": 30180
    },
    {
      "entropy": 5.574751567840576,
      "epoch": 2.995732433505359,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041146338889241506,
      "loss": 4.9191,
      "mean_token_accuracy": 0.19036875814199447,
      "num_tokens": 66131386.0,
      "step": 30185
    },
    {
      "entropy": 5.637083911895752,
      "epoch": 2.996228662167527,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00041143521630987483,
      "loss": 5.0963,
      "mean_token_accuracy": 0.1778554365038872,
      "num_tokens": 66142905.0,
      "step": 30190
    },
    {
      "entropy": 5.6475873470306395,
      "epoch": 2.9967248908296944,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041140704034403527,
      "loss": 5.1434,
      "mean_token_accuracy": 0.17549051642417907,
      "num_tokens": 66152708.0,
      "step": 30195
    },
    {
      "entropy": 5.723256015777588,
      "epoch": 2.997221119491862,
      "grad_norm": 1.09375,
      "learning_rate": 0.000411378860995595,
      "loss": 5.1505,
      "mean_token_accuracy": 0.18080793619155883,
      "num_tokens": 66164351.0,
      "step": 30200
    },
    {
      "entropy": 5.700329065322876,
      "epoch": 2.9977173481540293,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004113506782652528,
      "loss": 5.0624,
      "mean_token_accuracy": 0.17873395681381227,
      "num_tokens": 66176641.0,
      "step": 30205
    },
    {
      "entropy": 5.739531183242798,
      "epoch": 2.998213576816197,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004113224921537076,
      "loss": 5.105,
      "mean_token_accuracy": 0.17775907814502717,
      "num_tokens": 66187822.0,
      "step": 30210
    },
    {
      "entropy": 5.575352478027344,
      "epoch": 2.9987098054783643,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041129430266165836,
      "loss": 5.012,
      "mean_token_accuracy": 0.1848052814602852,
      "num_tokens": 66199636.0,
      "step": 30215
    },
    {
      "entropy": 5.641059827804566,
      "epoch": 2.999206034140532,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004112661097898041,
      "loss": 5.0195,
      "mean_token_accuracy": 0.18138792365789413,
      "num_tokens": 66210649.0,
      "step": 30220
    },
    {
      "entropy": 5.631885147094726,
      "epoch": 2.9997022628026997,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041123791353884397,
      "loss": 5.0095,
      "mean_token_accuracy": 0.18125305622816085,
      "num_tokens": 66221806.0,
      "step": 30225
    },
    {
      "entropy": 5.598717975616455,
      "epoch": 3.000198491464867,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004112097139094771,
      "loss": 4.9716,
      "mean_token_accuracy": 0.18510637432336807,
      "num_tokens": 66232016.0,
      "step": 30230
    },
    {
      "entropy": 5.581386470794678,
      "epoch": 3.0006947201270346,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041118151090240284,
      "loss": 4.9498,
      "mean_token_accuracy": 0.1873907506465912,
      "num_tokens": 66242852.0,
      "step": 30235
    },
    {
      "entropy": 5.6308540344238285,
      "epoch": 3.001190948789202,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041115330451832065,
      "loss": 4.9939,
      "mean_token_accuracy": 0.1832643210887909,
      "num_tokens": 66253254.0,
      "step": 30240
    },
    {
      "entropy": 5.613007688522339,
      "epoch": 3.0016871774513696,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041112509475792983,
      "loss": 4.9592,
      "mean_token_accuracy": 0.18239415436983109,
      "num_tokens": 66262721.0,
      "step": 30245
    },
    {
      "entropy": 5.594451665878296,
      "epoch": 3.002183406113537,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041109688162193,
      "loss": 4.9208,
      "mean_token_accuracy": 0.1901039570569992,
      "num_tokens": 66273718.0,
      "step": 30250
    },
    {
      "entropy": 5.613263988494873,
      "epoch": 3.0026796347757045,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004110686651110207,
      "loss": 5.0013,
      "mean_token_accuracy": 0.18694380670785904,
      "num_tokens": 66285527.0,
      "step": 30255
    },
    {
      "entropy": 5.6330756664276125,
      "epoch": 3.003175863437872,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004110404452259017,
      "loss": 4.9744,
      "mean_token_accuracy": 0.18594790995121002,
      "num_tokens": 66295564.0,
      "step": 30260
    },
    {
      "entropy": 5.590390062332153,
      "epoch": 3.0036720921000395,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004110122219672729,
      "loss": 4.9361,
      "mean_token_accuracy": 0.19214808344841003,
      "num_tokens": 66306721.0,
      "step": 30265
    },
    {
      "entropy": 5.577317714691162,
      "epoch": 3.0041683207622074,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000410983995335834,
      "loss": 4.942,
      "mean_token_accuracy": 0.19656314104795455,
      "num_tokens": 66316990.0,
      "step": 30270
    },
    {
      "entropy": 5.584804725646973,
      "epoch": 3.004664549424375,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000410955765332285,
      "loss": 4.9772,
      "mean_token_accuracy": 0.18709995746612548,
      "num_tokens": 66327814.0,
      "step": 30275
    },
    {
      "entropy": 5.590845727920533,
      "epoch": 3.0051607780865424,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041092753195732607,
      "loss": 4.9213,
      "mean_token_accuracy": 0.19091932326555253,
      "num_tokens": 66338373.0,
      "step": 30280
    },
    {
      "entropy": 5.583441829681396,
      "epoch": 3.00565700674871,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00041089929521165725,
      "loss": 5.0007,
      "mean_token_accuracy": 0.18617116510868073,
      "num_tokens": 66349311.0,
      "step": 30285
    },
    {
      "entropy": 5.702427864074707,
      "epoch": 3.0061532354108773,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004108710550959787,
      "loss": 4.976,
      "mean_token_accuracy": 0.19142845124006272,
      "num_tokens": 66360771.0,
      "step": 30290
    },
    {
      "entropy": 5.646543836593628,
      "epoch": 3.006649464073045,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041084281161099073,
      "loss": 4.9791,
      "mean_token_accuracy": 0.18476541638374328,
      "num_tokens": 66371348.0,
      "step": 30295
    },
    {
      "entropy": 5.658541202545166,
      "epoch": 3.0071456927352123,
      "grad_norm": 1.125,
      "learning_rate": 0.0004108145647573938,
      "loss": 5.0378,
      "mean_token_accuracy": 0.18679469972848892,
      "num_tokens": 66382489.0,
      "step": 30300
    },
    {
      "entropy": 5.578683614730835,
      "epoch": 3.0076419213973797,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041078631453588834,
      "loss": 4.9155,
      "mean_token_accuracy": 0.1937975063920021,
      "num_tokens": 66392825.0,
      "step": 30305
    },
    {
      "entropy": 5.585844469070435,
      "epoch": 3.0081381500595477,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004107580609471748,
      "loss": 4.9853,
      "mean_token_accuracy": 0.18524574786424636,
      "num_tokens": 66403646.0,
      "step": 30310
    },
    {
      "entropy": 5.591515827178955,
      "epoch": 3.008634378721715,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041072980399195394,
      "loss": 4.8946,
      "mean_token_accuracy": 0.19687081426382064,
      "num_tokens": 66414199.0,
      "step": 30315
    },
    {
      "entropy": 5.607180500030518,
      "epoch": 3.0091306073838826,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004107015436709264,
      "loss": 4.8794,
      "mean_token_accuracy": 0.1927315726876259,
      "num_tokens": 66426331.0,
      "step": 30320
    },
    {
      "entropy": 5.592084646224976,
      "epoch": 3.00962683604605,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041067327998479296,
      "loss": 4.92,
      "mean_token_accuracy": 0.18736550956964493,
      "num_tokens": 66436854.0,
      "step": 30325
    },
    {
      "entropy": 5.594204092025757,
      "epoch": 3.0101230647082176,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041064501293425453,
      "loss": 4.9902,
      "mean_token_accuracy": 0.18795809000730515,
      "num_tokens": 66447757.0,
      "step": 30330
    },
    {
      "entropy": 5.625861406326294,
      "epoch": 3.010619293370385,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004106167425200121,
      "loss": 4.9678,
      "mean_token_accuracy": 0.18738853931427002,
      "num_tokens": 66458395.0,
      "step": 30335
    },
    {
      "entropy": 5.611331462860107,
      "epoch": 3.0111155220325525,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004105884687427667,
      "loss": 4.8505,
      "mean_token_accuracy": 0.1924513980746269,
      "num_tokens": 66467728.0,
      "step": 30340
    },
    {
      "entropy": 5.609290504455567,
      "epoch": 3.01161175069472,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004105601916032194,
      "loss": 4.9475,
      "mean_token_accuracy": 0.19594927430152892,
      "num_tokens": 66478113.0,
      "step": 30345
    },
    {
      "entropy": 5.6400596618652346,
      "epoch": 3.0121079793568875,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004105319111020714,
      "loss": 4.9355,
      "mean_token_accuracy": 0.1888850748538971,
      "num_tokens": 66489055.0,
      "step": 30350
    },
    {
      "entropy": 5.574275875091553,
      "epoch": 3.0126042080190554,
      "grad_norm": 1.125,
      "learning_rate": 0.00041050362724002413,
      "loss": 4.9244,
      "mean_token_accuracy": 0.19353559613227844,
      "num_tokens": 66499516.0,
      "step": 30355
    },
    {
      "entropy": 5.584369230270386,
      "epoch": 3.013100436681223,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041047534001777884,
      "loss": 4.9836,
      "mean_token_accuracy": 0.19290254563093184,
      "num_tokens": 66510832.0,
      "step": 30360
    },
    {
      "entropy": 5.693460893630982,
      "epoch": 3.0135966653433903,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004104470494360371,
      "loss": 5.0487,
      "mean_token_accuracy": 0.18019575029611587,
      "num_tokens": 66522457.0,
      "step": 30365
    },
    {
      "entropy": 5.651749753952027,
      "epoch": 3.014092894005558,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041041875549550043,
      "loss": 5.0006,
      "mean_token_accuracy": 0.1881660297513008,
      "num_tokens": 66532527.0,
      "step": 30370
    },
    {
      "entropy": 5.650031709671021,
      "epoch": 3.0145891226677253,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004103904581968703,
      "loss": 4.9774,
      "mean_token_accuracy": 0.19138551205396653,
      "num_tokens": 66542437.0,
      "step": 30375
    },
    {
      "entropy": 5.553804636001587,
      "epoch": 3.0150853513298927,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041036215754084864,
      "loss": 4.9406,
      "mean_token_accuracy": 0.18945419043302536,
      "num_tokens": 66552927.0,
      "step": 30380
    },
    {
      "entropy": 5.642523670196534,
      "epoch": 3.0155815799920602,
      "grad_norm": 1.125,
      "learning_rate": 0.00041033385352813714,
      "loss": 5.051,
      "mean_token_accuracy": 0.18118196129798889,
      "num_tokens": 66563664.0,
      "step": 30385
    },
    {
      "entropy": 5.616431665420532,
      "epoch": 3.0160778086542277,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004103055461594377,
      "loss": 4.9446,
      "mean_token_accuracy": 0.1943461999297142,
      "num_tokens": 66573968.0,
      "step": 30390
    },
    {
      "entropy": 5.599416923522949,
      "epoch": 3.0165740373163956,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004102772354354522,
      "loss": 5.0354,
      "mean_token_accuracy": 0.18301819562911986,
      "num_tokens": 66585048.0,
      "step": 30395
    },
    {
      "entropy": 5.634533166885376,
      "epoch": 3.017070265978563,
      "grad_norm": 1.359375,
      "learning_rate": 0.00041024892135688287,
      "loss": 5.0192,
      "mean_token_accuracy": 0.19041632860898972,
      "num_tokens": 66594813.0,
      "step": 30400
    },
    {
      "entropy": 5.665782260894775,
      "epoch": 3.0175664946407306,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004102206039244316,
      "loss": 4.9642,
      "mean_token_accuracy": 0.18979667127132416,
      "num_tokens": 66607096.0,
      "step": 30405
    },
    {
      "entropy": 5.66130313873291,
      "epoch": 3.018062723302898,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004101922831388008,
      "loss": 4.9949,
      "mean_token_accuracy": 0.1770506903529167,
      "num_tokens": 66617001.0,
      "step": 30410
    },
    {
      "entropy": 5.619752407073975,
      "epoch": 3.0185589519650655,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041016395900069267,
      "loss": 4.9733,
      "mean_token_accuracy": 0.184030419588089,
      "num_tokens": 66627036.0,
      "step": 30415
    },
    {
      "entropy": 5.6345385074615475,
      "epoch": 3.019055180627233,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004101356315108097,
      "loss": 4.991,
      "mean_token_accuracy": 0.18374181687831878,
      "num_tokens": 66637526.0,
      "step": 30420
    },
    {
      "entropy": 5.591116905212402,
      "epoch": 3.0195514092894005,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041010730066985414,
      "loss": 4.9338,
      "mean_token_accuracy": 0.1945575147867203,
      "num_tokens": 66650046.0,
      "step": 30425
    },
    {
      "entropy": 5.565946912765503,
      "epoch": 3.020047637951568,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004100789664785287,
      "loss": 4.8766,
      "mean_token_accuracy": 0.19755872189998627,
      "num_tokens": 66660922.0,
      "step": 30430
    },
    {
      "entropy": 5.62859263420105,
      "epoch": 3.0205438666137354,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004100506289375359,
      "loss": 5.03,
      "mean_token_accuracy": 0.1893402948975563,
      "num_tokens": 66671861.0,
      "step": 30435
    },
    {
      "entropy": 5.615510034561157,
      "epoch": 3.0210400952759033,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004100222880475786,
      "loss": 4.9942,
      "mean_token_accuracy": 0.1881416007876396,
      "num_tokens": 66682834.0,
      "step": 30440
    },
    {
      "entropy": 5.583151769638062,
      "epoch": 3.021536323938071,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040999394380935943,
      "loss": 4.9852,
      "mean_token_accuracy": 0.18673600405454635,
      "num_tokens": 66693692.0,
      "step": 30445
    },
    {
      "entropy": 5.619572877883911,
      "epoch": 3.0220325526002383,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040996559622358145,
      "loss": 4.9906,
      "mean_token_accuracy": 0.1890046328306198,
      "num_tokens": 66705035.0,
      "step": 30450
    },
    {
      "entropy": 5.551941871643066,
      "epoch": 3.0225287812624058,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004099372452909474,
      "loss": 4.8993,
      "mean_token_accuracy": 0.1936531648039818,
      "num_tokens": 66714713.0,
      "step": 30455
    },
    {
      "entropy": 5.585730791091919,
      "epoch": 3.0230250099245732,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004099088910121605,
      "loss": 5.0136,
      "mean_token_accuracy": 0.1835516706109047,
      "num_tokens": 66726454.0,
      "step": 30460
    },
    {
      "entropy": 5.590458106994629,
      "epoch": 3.0235212385867407,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040988053338792367,
      "loss": 4.9567,
      "mean_token_accuracy": 0.18618073612451552,
      "num_tokens": 66736103.0,
      "step": 30465
    },
    {
      "entropy": 5.6250709056854244,
      "epoch": 3.024017467248908,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004098521724189404,
      "loss": 5.0087,
      "mean_token_accuracy": 0.18078238815069197,
      "num_tokens": 66746554.0,
      "step": 30470
    },
    {
      "entropy": 5.618751430511475,
      "epoch": 3.0245136959110757,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040982380810591386,
      "loss": 4.9351,
      "mean_token_accuracy": 0.19151628166437148,
      "num_tokens": 66756568.0,
      "step": 30475
    },
    {
      "entropy": 5.6058601379394535,
      "epoch": 3.025009924573243,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004097954404495473,
      "loss": 4.9478,
      "mean_token_accuracy": 0.1897781029343605,
      "num_tokens": 66769036.0,
      "step": 30480
    },
    {
      "entropy": 5.540247201919556,
      "epoch": 3.025506153235411,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040976706945054434,
      "loss": 4.8522,
      "mean_token_accuracy": 0.20961830168962478,
      "num_tokens": 66779868.0,
      "step": 30485
    },
    {
      "entropy": 5.62428650856018,
      "epoch": 3.0260023818975785,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040973869510960837,
      "loss": 4.9805,
      "mean_token_accuracy": 0.1882914587855339,
      "num_tokens": 66791903.0,
      "step": 30490
    },
    {
      "entropy": 5.668175220489502,
      "epoch": 3.026498610559746,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040971031742744314,
      "loss": 4.9917,
      "mean_token_accuracy": 0.18591705411672593,
      "num_tokens": 66802118.0,
      "step": 30495
    },
    {
      "entropy": 5.672676467895508,
      "epoch": 3.0269948392219135,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040968193640475236,
      "loss": 5.0146,
      "mean_token_accuracy": 0.18685044944286347,
      "num_tokens": 66813653.0,
      "step": 30500
    },
    {
      "entropy": 5.6721399307250975,
      "epoch": 3.027491067884081,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004096535520422397,
      "loss": 5.0474,
      "mean_token_accuracy": 0.18411852568387985,
      "num_tokens": 66824752.0,
      "step": 30505
    },
    {
      "entropy": 5.604950761795044,
      "epoch": 3.0279872965462484,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040962516434060915,
      "loss": 4.9281,
      "mean_token_accuracy": 0.19168652445077897,
      "num_tokens": 66835483.0,
      "step": 30510
    },
    {
      "entropy": 5.673874568939209,
      "epoch": 3.028483525208416,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004095967733005646,
      "loss": 4.9874,
      "mean_token_accuracy": 0.18507133722305297,
      "num_tokens": 66846890.0,
      "step": 30515
    },
    {
      "entropy": 5.550619077682495,
      "epoch": 3.0289797538705834,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040956837892281014,
      "loss": 4.8812,
      "mean_token_accuracy": 0.1893573671579361,
      "num_tokens": 66857326.0,
      "step": 30520
    },
    {
      "entropy": 5.6155815601348875,
      "epoch": 3.0294759825327513,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040953998120804976,
      "loss": 4.958,
      "mean_token_accuracy": 0.18633511811494827,
      "num_tokens": 66868266.0,
      "step": 30525
    },
    {
      "entropy": 5.648244762420655,
      "epoch": 3.0299722111949188,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004095115801569878,
      "loss": 5.0147,
      "mean_token_accuracy": 0.17849545329809188,
      "num_tokens": 66879121.0,
      "step": 30530
    },
    {
      "entropy": 5.632192373275757,
      "epoch": 3.0304684398570862,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004094831757703285,
      "loss": 4.9322,
      "mean_token_accuracy": 0.1922020748257637,
      "num_tokens": 66889804.0,
      "step": 30535
    },
    {
      "entropy": 5.607153367996216,
      "epoch": 3.0309646685192537,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040945476804877623,
      "loss": 4.9577,
      "mean_token_accuracy": 0.18824877440929413,
      "num_tokens": 66900421.0,
      "step": 30540
    },
    {
      "entropy": 5.622883367538452,
      "epoch": 3.031460897181421,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004094263569930355,
      "loss": 5.0503,
      "mean_token_accuracy": 0.1818084254860878,
      "num_tokens": 66910754.0,
      "step": 30545
    },
    {
      "entropy": 5.674953460693359,
      "epoch": 3.0319571258435887,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040939794260381074,
      "loss": 5.0193,
      "mean_token_accuracy": 0.17803046852350235,
      "num_tokens": 66921107.0,
      "step": 30550
    },
    {
      "entropy": 5.5957482814788815,
      "epoch": 3.032453354505756,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004093695248818065,
      "loss": 4.9301,
      "mean_token_accuracy": 0.18645533174276352,
      "num_tokens": 66932851.0,
      "step": 30555
    },
    {
      "entropy": 5.64589900970459,
      "epoch": 3.0329495831679236,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004093411038277277,
      "loss": 4.9829,
      "mean_token_accuracy": 0.18938347548246384,
      "num_tokens": 66943403.0,
      "step": 30560
    },
    {
      "entropy": 5.694613885879517,
      "epoch": 3.033445811830091,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000409312679442279,
      "loss": 5.1132,
      "mean_token_accuracy": 0.18458633422851561,
      "num_tokens": 66955021.0,
      "step": 30565
    },
    {
      "entropy": 5.657150506973267,
      "epoch": 3.033942040492259,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040928425172616524,
      "loss": 4.9956,
      "mean_token_accuracy": 0.18762361854314805,
      "num_tokens": 66966147.0,
      "step": 30570
    },
    {
      "entropy": 5.655306482315064,
      "epoch": 3.0344382691544265,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040925582068009145,
      "loss": 4.943,
      "mean_token_accuracy": 0.1832023948431015,
      "num_tokens": 66977946.0,
      "step": 30575
    },
    {
      "entropy": 5.614497947692871,
      "epoch": 3.034934497816594,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004092273863047625,
      "loss": 5.0594,
      "mean_token_accuracy": 0.18674104362726213,
      "num_tokens": 66988014.0,
      "step": 30580
    },
    {
      "entropy": 5.680592107772827,
      "epoch": 3.0354307264787614,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040919894860088366,
      "loss": 5.0002,
      "mean_token_accuracy": 0.19153071492910384,
      "num_tokens": 66999142.0,
      "step": 30585
    },
    {
      "entropy": 5.604717111587524,
      "epoch": 3.035926955140929,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004091705075691601,
      "loss": 5.0294,
      "mean_token_accuracy": 0.18264827132225037,
      "num_tokens": 67010849.0,
      "step": 30590
    },
    {
      "entropy": 5.583193111419678,
      "epoch": 3.0364231838030964,
      "grad_norm": 1.15625,
      "learning_rate": 0.000409142063210297,
      "loss": 4.9389,
      "mean_token_accuracy": 0.19434720724821092,
      "num_tokens": 67021681.0,
      "step": 30595
    },
    {
      "entropy": 5.633980131149292,
      "epoch": 3.036919412465264,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040911361552499974,
      "loss": 5.021,
      "mean_token_accuracy": 0.1790219008922577,
      "num_tokens": 67034080.0,
      "step": 30600
    },
    {
      "entropy": 5.628406143188476,
      "epoch": 3.0374156411274313,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040908516451397395,
      "loss": 4.8936,
      "mean_token_accuracy": 0.1925090342760086,
      "num_tokens": 67045528.0,
      "step": 30605
    },
    {
      "entropy": 5.615469217300415,
      "epoch": 3.0379118697895993,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004090567101779249,
      "loss": 4.9306,
      "mean_token_accuracy": 0.19328001737594605,
      "num_tokens": 67057611.0,
      "step": 30610
    },
    {
      "entropy": 5.596962547302246,
      "epoch": 3.0384080984517667,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004090282525175583,
      "loss": 4.9613,
      "mean_token_accuracy": 0.18763742446899415,
      "num_tokens": 67067843.0,
      "step": 30615
    },
    {
      "entropy": 5.672213220596314,
      "epoch": 3.038904327113934,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004089997915335799,
      "loss": 5.0539,
      "mean_token_accuracy": 0.1902668818831444,
      "num_tokens": 67079125.0,
      "step": 30620
    },
    {
      "entropy": 5.6237907886505125,
      "epoch": 3.0394005557761017,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004089713272266954,
      "loss": 4.9828,
      "mean_token_accuracy": 0.18119250386953353,
      "num_tokens": 67089750.0,
      "step": 30625
    },
    {
      "entropy": 5.636394453048706,
      "epoch": 3.039896784438269,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004089428595976106,
      "loss": 5.0206,
      "mean_token_accuracy": 0.18942246884107589,
      "num_tokens": 67101738.0,
      "step": 30630
    },
    {
      "entropy": 5.632771015167236,
      "epoch": 3.0403930131004366,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004089143886470316,
      "loss": 4.9263,
      "mean_token_accuracy": 0.19426962435245515,
      "num_tokens": 67111941.0,
      "step": 30635
    },
    {
      "entropy": 5.638123655319214,
      "epoch": 3.040889241762604,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004088859143756643,
      "loss": 5.0665,
      "mean_token_accuracy": 0.17989249229431153,
      "num_tokens": 67124755.0,
      "step": 30640
    },
    {
      "entropy": 5.714755535125732,
      "epoch": 3.0413854704247716,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004088574367842148,
      "loss": 5.143,
      "mean_token_accuracy": 0.18082985430955886,
      "num_tokens": 67136597.0,
      "step": 30645
    },
    {
      "entropy": 5.649950885772705,
      "epoch": 3.041881699086939,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004088289558733893,
      "loss": 5.0326,
      "mean_token_accuracy": 0.19192678183317186,
      "num_tokens": 67146790.0,
      "step": 30650
    },
    {
      "entropy": 5.590872573852539,
      "epoch": 3.042377927749107,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004088004716438941,
      "loss": 4.9797,
      "mean_token_accuracy": 0.1938429668545723,
      "num_tokens": 67158054.0,
      "step": 30655
    },
    {
      "entropy": 5.615215826034546,
      "epoch": 3.0428741564112745,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004087719840964355,
      "loss": 4.9253,
      "mean_token_accuracy": 0.19147589206695556,
      "num_tokens": 67168244.0,
      "step": 30660
    },
    {
      "entropy": 5.644600534439087,
      "epoch": 3.043370385073442,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040874349323171994,
      "loss": 4.9527,
      "mean_token_accuracy": 0.1907864511013031,
      "num_tokens": 67179165.0,
      "step": 30665
    },
    {
      "entropy": 5.600915813446045,
      "epoch": 3.0438666137356094,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004087149990504539,
      "loss": 4.939,
      "mean_token_accuracy": 0.19038206934928895,
      "num_tokens": 67189653.0,
      "step": 30670
    },
    {
      "entropy": 5.597408246994019,
      "epoch": 3.044362842397777,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040868650155334406,
      "loss": 4.9893,
      "mean_token_accuracy": 0.19073462337255478,
      "num_tokens": 67203933.0,
      "step": 30675
    },
    {
      "entropy": 5.570241832733155,
      "epoch": 3.0448590710599444,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040865800074109713,
      "loss": 4.9357,
      "mean_token_accuracy": 0.1918495550751686,
      "num_tokens": 67215590.0,
      "step": 30680
    },
    {
      "entropy": 5.594831037521362,
      "epoch": 3.045355299722112,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004086294966144196,
      "loss": 4.93,
      "mean_token_accuracy": 0.19196883738040924,
      "num_tokens": 67225416.0,
      "step": 30685
    },
    {
      "entropy": 5.598183536529541,
      "epoch": 3.0458515283842793,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004086009891740187,
      "loss": 4.912,
      "mean_token_accuracy": 0.18535275757312775,
      "num_tokens": 67234492.0,
      "step": 30690
    },
    {
      "entropy": 5.596599531173706,
      "epoch": 3.046347757046447,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040857247842060105,
      "loss": 4.9736,
      "mean_token_accuracy": 0.18232698887586593,
      "num_tokens": 67245630.0,
      "step": 30695
    },
    {
      "entropy": 5.660459184646607,
      "epoch": 3.0468439857086147,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004085439643548737,
      "loss": 5.0493,
      "mean_token_accuracy": 0.18831884264945983,
      "num_tokens": 67257014.0,
      "step": 30700
    },
    {
      "entropy": 5.523923778533936,
      "epoch": 3.047340214370782,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004085154469775439,
      "loss": 4.8801,
      "mean_token_accuracy": 0.1925245627760887,
      "num_tokens": 67267165.0,
      "step": 30705
    },
    {
      "entropy": 5.609407997131347,
      "epoch": 3.0478364430329496,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004084869262893187,
      "loss": 4.9694,
      "mean_token_accuracy": 0.1899210825562477,
      "num_tokens": 67277077.0,
      "step": 30710
    },
    {
      "entropy": 5.622097063064575,
      "epoch": 3.048332671695117,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004084584022909053,
      "loss": 4.9751,
      "mean_token_accuracy": 0.19168352335691452,
      "num_tokens": 67287941.0,
      "step": 30715
    },
    {
      "entropy": 5.51892352104187,
      "epoch": 3.0488289003572846,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040842987498301115,
      "loss": 4.8718,
      "mean_token_accuracy": 0.1875634416937828,
      "num_tokens": 67299764.0,
      "step": 30720
    },
    {
      "entropy": 5.616883897781372,
      "epoch": 3.049325129019452,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004084013443663437,
      "loss": 4.9544,
      "mean_token_accuracy": 0.18413940370082854,
      "num_tokens": 67310527.0,
      "step": 30725
    },
    {
      "entropy": 5.626830673217773,
      "epoch": 3.0498213576816195,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004083728104416102,
      "loss": 4.9756,
      "mean_token_accuracy": 0.18897880464792252,
      "num_tokens": 67321369.0,
      "step": 30730
    },
    {
      "entropy": 5.568759536743164,
      "epoch": 3.050317586343787,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004083442732095185,
      "loss": 4.9084,
      "mean_token_accuracy": 0.20176333785057068,
      "num_tokens": 67332215.0,
      "step": 30735
    },
    {
      "entropy": 5.609928894042969,
      "epoch": 3.050813815005955,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004083157326707762,
      "loss": 5.0849,
      "mean_token_accuracy": 0.17823947072029114,
      "num_tokens": 67344407.0,
      "step": 30740
    },
    {
      "entropy": 5.646574354171753,
      "epoch": 3.0513100436681224,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040828718882609085,
      "loss": 4.9581,
      "mean_token_accuracy": 0.18673366010189058,
      "num_tokens": 67355513.0,
      "step": 30745
    },
    {
      "entropy": 5.618381547927856,
      "epoch": 3.05180627233029,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040825864167617047,
      "loss": 4.9913,
      "mean_token_accuracy": 0.1834542915225029,
      "num_tokens": 67366515.0,
      "step": 30750
    },
    {
      "entropy": 5.644259119033814,
      "epoch": 3.0523025009924574,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000408230091221723,
      "loss": 4.9577,
      "mean_token_accuracy": 0.19507146179676055,
      "num_tokens": 67376568.0,
      "step": 30755
    },
    {
      "entropy": 5.656472682952881,
      "epoch": 3.052798729654625,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004082015374634563,
      "loss": 5.0601,
      "mean_token_accuracy": 0.17866806834936141,
      "num_tokens": 67388639.0,
      "step": 30760
    },
    {
      "entropy": 5.626897668838501,
      "epoch": 3.0532949583167923,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004081729804020785,
      "loss": 4.9717,
      "mean_token_accuracy": 0.18269844949245453,
      "num_tokens": 67400330.0,
      "step": 30765
    },
    {
      "entropy": 5.579427003860474,
      "epoch": 3.05379118697896,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004081444200382978,
      "loss": 4.9518,
      "mean_token_accuracy": 0.1867471382021904,
      "num_tokens": 67411067.0,
      "step": 30770
    },
    {
      "entropy": 5.582539749145508,
      "epoch": 3.0542874156411273,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040811585637282234,
      "loss": 4.918,
      "mean_token_accuracy": 0.19195748567581178,
      "num_tokens": 67421142.0,
      "step": 30775
    },
    {
      "entropy": 5.586840581893921,
      "epoch": 3.0547836443032947,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004080872894063605,
      "loss": 4.9298,
      "mean_token_accuracy": 0.19341164231300353,
      "num_tokens": 67433609.0,
      "step": 30780
    },
    {
      "entropy": 5.702679109573364,
      "epoch": 3.0552798729654627,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040805871913962056,
      "loss": 5.1203,
      "mean_token_accuracy": 0.17652339935302735,
      "num_tokens": 67445281.0,
      "step": 30785
    },
    {
      "entropy": 5.575446748733521,
      "epoch": 3.05577610162763,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004080301455733112,
      "loss": 4.9253,
      "mean_token_accuracy": 0.18903401345014573,
      "num_tokens": 67455831.0,
      "step": 30790
    },
    {
      "entropy": 5.64455018043518,
      "epoch": 3.0562723302897976,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00040800156870814087,
      "loss": 5.0506,
      "mean_token_accuracy": 0.18833058923482895,
      "num_tokens": 67467195.0,
      "step": 30795
    },
    {
      "entropy": 5.566956377029419,
      "epoch": 3.056768558951965,
      "grad_norm": 1.125,
      "learning_rate": 0.00040797298854481826,
      "loss": 4.9253,
      "mean_token_accuracy": 0.19092370122671126,
      "num_tokens": 67478332.0,
      "step": 30800
    },
    {
      "entropy": 5.617889022827148,
      "epoch": 3.0572647876141326,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040794440508405195,
      "loss": 4.9282,
      "mean_token_accuracy": 0.19849851131439208,
      "num_tokens": 67489600.0,
      "step": 30805
    },
    {
      "entropy": 5.665069484710694,
      "epoch": 3.0577610162763,
      "grad_norm": 1.125,
      "learning_rate": 0.000407915818326551,
      "loss": 5.0623,
      "mean_token_accuracy": 0.18136073648929596,
      "num_tokens": 67498837.0,
      "step": 30810
    },
    {
      "entropy": 5.688076639175415,
      "epoch": 3.0582572449384675,
      "grad_norm": 1.09375,
      "learning_rate": 0.000407887228273024,
      "loss": 4.9944,
      "mean_token_accuracy": 0.19091964066028594,
      "num_tokens": 67509118.0,
      "step": 30815
    },
    {
      "entropy": 5.542440414428711,
      "epoch": 3.058753473600635,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004078586349241802,
      "loss": 4.9186,
      "mean_token_accuracy": 0.1857878625392914,
      "num_tokens": 67520577.0,
      "step": 30820
    },
    {
      "entropy": 5.596519470214844,
      "epoch": 3.059249702262803,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040783003828072855,
      "loss": 4.9111,
      "mean_token_accuracy": 0.19055615812540055,
      "num_tokens": 67532020.0,
      "step": 30825
    },
    {
      "entropy": 5.576579523086548,
      "epoch": 3.0597459309249704,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004078014383433782,
      "loss": 4.9424,
      "mean_token_accuracy": 0.18883258998394012,
      "num_tokens": 67540845.0,
      "step": 30830
    },
    {
      "entropy": 5.5755688667297365,
      "epoch": 3.060242159587138,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004077728351128383,
      "loss": 4.9725,
      "mean_token_accuracy": 0.1898255839943886,
      "num_tokens": 67552698.0,
      "step": 30835
    },
    {
      "entropy": 5.549128103256225,
      "epoch": 3.0607383882493053,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004077442285898181,
      "loss": 4.9538,
      "mean_token_accuracy": 0.18298746198415755,
      "num_tokens": 67563772.0,
      "step": 30840
    },
    {
      "entropy": 5.622981357574463,
      "epoch": 3.061234616911473,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040771561877502727,
      "loss": 4.9886,
      "mean_token_accuracy": 0.18567167967557907,
      "num_tokens": 67574955.0,
      "step": 30845
    },
    {
      "entropy": 5.58887209892273,
      "epoch": 3.0617308455736403,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004076870056691749,
      "loss": 4.9305,
      "mean_token_accuracy": 0.1925195962190628,
      "num_tokens": 67586086.0,
      "step": 30850
    },
    {
      "entropy": 5.670666122436524,
      "epoch": 3.0622270742358078,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040765838927297077,
      "loss": 5.026,
      "mean_token_accuracy": 0.18855949938297273,
      "num_tokens": 67597036.0,
      "step": 30855
    },
    {
      "entropy": 5.6248945713043215,
      "epoch": 3.0627233028979752,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004076297695871244,
      "loss": 4.8957,
      "mean_token_accuracy": 0.20713433027267455,
      "num_tokens": 67606499.0,
      "step": 30860
    },
    {
      "entropy": 5.633873844146729,
      "epoch": 3.0632195315601427,
      "grad_norm": 1.125,
      "learning_rate": 0.00040760114661234553,
      "loss": 5.0394,
      "mean_token_accuracy": 0.18878856301307678,
      "num_tokens": 67617820.0,
      "step": 30865
    },
    {
      "entropy": 5.5941245555877686,
      "epoch": 3.0637157602223106,
      "grad_norm": 1.125,
      "learning_rate": 0.000407572520349344,
      "loss": 4.8962,
      "mean_token_accuracy": 0.19868271797895432,
      "num_tokens": 67628122.0,
      "step": 30870
    },
    {
      "entropy": 5.596986246109009,
      "epoch": 3.064211988884478,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004075438907988296,
      "loss": 4.9943,
      "mean_token_accuracy": 0.18604467809200287,
      "num_tokens": 67639127.0,
      "step": 30875
    },
    {
      "entropy": 5.59669041633606,
      "epoch": 3.0647082175466456,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004075152579615124,
      "loss": 4.8945,
      "mean_token_accuracy": 0.1968715161085129,
      "num_tokens": 67650122.0,
      "step": 30880
    },
    {
      "entropy": 5.640858507156372,
      "epoch": 3.065204446208813,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004074866218381023,
      "loss": 5.0228,
      "mean_token_accuracy": 0.18087900578975677,
      "num_tokens": 67660427.0,
      "step": 30885
    },
    {
      "entropy": 5.589366722106933,
      "epoch": 3.0657006748709805,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040745798242930947,
      "loss": 5.0551,
      "mean_token_accuracy": 0.19178384393453599,
      "num_tokens": 67671151.0,
      "step": 30890
    },
    {
      "entropy": 5.6579938411712645,
      "epoch": 3.066196903533148,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000407429339735844,
      "loss": 5.0097,
      "mean_token_accuracy": 0.18353338837623595,
      "num_tokens": 67681338.0,
      "step": 30895
    },
    {
      "entropy": 5.605926370620727,
      "epoch": 3.0666931321953155,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040740069375841637,
      "loss": 4.9202,
      "mean_token_accuracy": 0.19130409955978395,
      "num_tokens": 67691884.0,
      "step": 30900
    },
    {
      "entropy": 5.576122856140136,
      "epoch": 3.067189360857483,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004073720444977368,
      "loss": 4.9816,
      "mean_token_accuracy": 0.18863677829504014,
      "num_tokens": 67703011.0,
      "step": 30905
    },
    {
      "entropy": 5.686889362335205,
      "epoch": 3.067685589519651,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004073433919545157,
      "loss": 5.1468,
      "mean_token_accuracy": 0.17638159543275833,
      "num_tokens": 67714042.0,
      "step": 30910
    },
    {
      "entropy": 5.627335929870606,
      "epoch": 3.0681818181818183,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004073147361294638,
      "loss": 5.0048,
      "mean_token_accuracy": 0.18647966533899307,
      "num_tokens": 67725449.0,
      "step": 30915
    },
    {
      "entropy": 5.640450954437256,
      "epoch": 3.068678046843986,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004072860770232915,
      "loss": 5.0693,
      "mean_token_accuracy": 0.1828252136707306,
      "num_tokens": 67737334.0,
      "step": 30920
    },
    {
      "entropy": 5.595211505889893,
      "epoch": 3.0691742755061533,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004072574146367095,
      "loss": 4.9698,
      "mean_token_accuracy": 0.1867480307817459,
      "num_tokens": 67747777.0,
      "step": 30925
    },
    {
      "entropy": 5.631493186950683,
      "epoch": 3.0696705041683208,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040722874897042863,
      "loss": 4.9251,
      "mean_token_accuracy": 0.19055636078119279,
      "num_tokens": 67758584.0,
      "step": 30930
    },
    {
      "entropy": 5.5891176700592045,
      "epoch": 3.0701667328304882,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004072000800251597,
      "loss": 4.9557,
      "mean_token_accuracy": 0.19063813835382462,
      "num_tokens": 67769168.0,
      "step": 30935
    },
    {
      "entropy": 5.614707279205322,
      "epoch": 3.0706629614926557,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040717140780161367,
      "loss": 4.9648,
      "mean_token_accuracy": 0.19147779047489166,
      "num_tokens": 67779680.0,
      "step": 30940
    },
    {
      "entropy": 5.593968152999878,
      "epoch": 3.071159190154823,
      "grad_norm": 1.125,
      "learning_rate": 0.00040714273230050153,
      "loss": 4.9826,
      "mean_token_accuracy": 0.1841825783252716,
      "num_tokens": 67789678.0,
      "step": 30945
    },
    {
      "entropy": 5.629078722000122,
      "epoch": 3.0716554188169907,
      "grad_norm": 1.125,
      "learning_rate": 0.0004071140535225343,
      "loss": 5.0217,
      "mean_token_accuracy": 0.19178111255168914,
      "num_tokens": 67800991.0,
      "step": 30950
    },
    {
      "entropy": 5.713760662078857,
      "epoch": 3.0721516474791586,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004070853714684233,
      "loss": 5.082,
      "mean_token_accuracy": 0.18138853460550308,
      "num_tokens": 67811319.0,
      "step": 30955
    },
    {
      "entropy": 5.678158807754516,
      "epoch": 3.072647876141326,
      "grad_norm": 1.125,
      "learning_rate": 0.0004070566861388797,
      "loss": 5.0848,
      "mean_token_accuracy": 0.18286885619163512,
      "num_tokens": 67823704.0,
      "step": 30960
    },
    {
      "entropy": 5.677220010757447,
      "epoch": 3.0731441048034935,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040702799753461477,
      "loss": 5.0223,
      "mean_token_accuracy": 0.1872129961848259,
      "num_tokens": 67835460.0,
      "step": 30965
    },
    {
      "entropy": 5.598894786834717,
      "epoch": 3.073640333465661,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004069993056563401,
      "loss": 4.9392,
      "mean_token_accuracy": 0.19336795955896377,
      "num_tokens": 67845464.0,
      "step": 30970
    },
    {
      "entropy": 5.706277656555176,
      "epoch": 3.0741365621278285,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000406970610504767,
      "loss": 5.0586,
      "mean_token_accuracy": 0.17798045724630357,
      "num_tokens": 67856718.0,
      "step": 30975
    },
    {
      "entropy": 5.648024559020996,
      "epoch": 3.074632790789996,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004069419120806072,
      "loss": 4.9601,
      "mean_token_accuracy": 0.19318547546863557,
      "num_tokens": 67867066.0,
      "step": 30980
    },
    {
      "entropy": 5.617588186264038,
      "epoch": 3.0751290194521634,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040691321038457224,
      "loss": 4.994,
      "mean_token_accuracy": 0.19054658114910125,
      "num_tokens": 67878213.0,
      "step": 30985
    },
    {
      "entropy": 5.626625728607178,
      "epoch": 3.075625248114331,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004068845054173739,
      "loss": 5.0082,
      "mean_token_accuracy": 0.1844394564628601,
      "num_tokens": 67888503.0,
      "step": 30990
    },
    {
      "entropy": 5.687537765502929,
      "epoch": 3.076121476776499,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004068557971797241,
      "loss": 4.9999,
      "mean_token_accuracy": 0.187854041159153,
      "num_tokens": 67899035.0,
      "step": 30995
    },
    {
      "entropy": 5.562118482589722,
      "epoch": 3.0766177054386663,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040682708567233464,
      "loss": 4.9752,
      "mean_token_accuracy": 0.19245949238538743,
      "num_tokens": 67909137.0,
      "step": 31000
    },
    {
      "entropy": 5.624712753295898,
      "epoch": 3.077113934100834,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040679837089591743,
      "loss": 4.9847,
      "mean_token_accuracy": 0.18359821736812593,
      "num_tokens": 67919026.0,
      "step": 31005
    },
    {
      "entropy": 5.675165414810181,
      "epoch": 3.0776101627630013,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040676965285118465,
      "loss": 5.0631,
      "mean_token_accuracy": 0.18130294382572174,
      "num_tokens": 67930333.0,
      "step": 31010
    },
    {
      "entropy": 5.624616956710815,
      "epoch": 3.0781063914251687,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004067409315388484,
      "loss": 4.919,
      "mean_token_accuracy": 0.19203022122383118,
      "num_tokens": 67940448.0,
      "step": 31015
    },
    {
      "entropy": 5.621329689025879,
      "epoch": 3.078602620087336,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040671220695962104,
      "loss": 4.9705,
      "mean_token_accuracy": 0.19193697571754456,
      "num_tokens": 67951398.0,
      "step": 31020
    },
    {
      "entropy": 5.654483604431152,
      "epoch": 3.0790988487495037,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004066834791142146,
      "loss": 4.9758,
      "mean_token_accuracy": 0.18562017232179642,
      "num_tokens": 67962741.0,
      "step": 31025
    },
    {
      "entropy": 5.637455081939697,
      "epoch": 3.079595077411671,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004066547480033417,
      "loss": 4.9589,
      "mean_token_accuracy": 0.19296547323465346,
      "num_tokens": 67975015.0,
      "step": 31030
    },
    {
      "entropy": 5.685034942626953,
      "epoch": 3.0800913060738386,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004066260136277148,
      "loss": 5.0406,
      "mean_token_accuracy": 0.18376687467098235,
      "num_tokens": 67985918.0,
      "step": 31035
    },
    {
      "entropy": 5.589951467514038,
      "epoch": 3.0805875347360065,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040659727598804635,
      "loss": 4.999,
      "mean_token_accuracy": 0.18790315538644792,
      "num_tokens": 67997419.0,
      "step": 31040
    },
    {
      "entropy": 5.611190938949585,
      "epoch": 3.081083763398174,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040656853508504897,
      "loss": 4.9503,
      "mean_token_accuracy": 0.19122907966375352,
      "num_tokens": 68008739.0,
      "step": 31045
    },
    {
      "entropy": 5.623060846328736,
      "epoch": 3.0815799920603415,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040653979091943555,
      "loss": 5.0086,
      "mean_token_accuracy": 0.18350944370031358,
      "num_tokens": 68020707.0,
      "step": 31050
    },
    {
      "entropy": 5.651588296890258,
      "epoch": 3.082076220722509,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040651104349191864,
      "loss": 4.9321,
      "mean_token_accuracy": 0.19023654013872146,
      "num_tokens": 68030825.0,
      "step": 31055
    },
    {
      "entropy": 5.606630849838257,
      "epoch": 3.0825724493846764,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040648229280321126,
      "loss": 4.9787,
      "mean_token_accuracy": 0.18300965279340745,
      "num_tokens": 68041385.0,
      "step": 31060
    },
    {
      "entropy": 5.6733451843261715,
      "epoch": 3.083068678046844,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004064535388540264,
      "loss": 5.0668,
      "mean_token_accuracy": 0.18191265016794206,
      "num_tokens": 68053123.0,
      "step": 31065
    },
    {
      "entropy": 5.6447728157043455,
      "epoch": 3.0835649067090114,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004064247816450769,
      "loss": 4.9607,
      "mean_token_accuracy": 0.1873933345079422,
      "num_tokens": 68064529.0,
      "step": 31070
    },
    {
      "entropy": 5.720857906341553,
      "epoch": 3.084061135371179,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004063960211770761,
      "loss": 5.0677,
      "mean_token_accuracy": 0.18538120836019517,
      "num_tokens": 68074091.0,
      "step": 31075
    },
    {
      "entropy": 5.64154806137085,
      "epoch": 3.084557364033347,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040636725745073714,
      "loss": 4.9692,
      "mean_token_accuracy": 0.19447302669286728,
      "num_tokens": 68085185.0,
      "step": 31080
    },
    {
      "entropy": 5.59161148071289,
      "epoch": 3.0850535926955143,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004063384904667732,
      "loss": 4.874,
      "mean_token_accuracy": 0.19459616839885713,
      "num_tokens": 68095977.0,
      "step": 31085
    },
    {
      "entropy": 5.605450582504273,
      "epoch": 3.0855498213576817,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040630972022589774,
      "loss": 5.0025,
      "mean_token_accuracy": 0.18767485320568084,
      "num_tokens": 68107469.0,
      "step": 31090
    },
    {
      "entropy": 5.721282815933227,
      "epoch": 3.086046050019849,
      "grad_norm": 1.125,
      "learning_rate": 0.00040628094672882415,
      "loss": 5.1444,
      "mean_token_accuracy": 0.1716878741979599,
      "num_tokens": 68119153.0,
      "step": 31095
    },
    {
      "entropy": 5.623871803283691,
      "epoch": 3.0865422786820167,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000406252169976266,
      "loss": 4.9282,
      "mean_token_accuracy": 0.18868185728788375,
      "num_tokens": 68130336.0,
      "step": 31100
    },
    {
      "entropy": 5.686665153503418,
      "epoch": 3.087038507344184,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004062233899689368,
      "loss": 4.9843,
      "mean_token_accuracy": 0.1883894369006157,
      "num_tokens": 68141231.0,
      "step": 31105
    },
    {
      "entropy": 5.6127887725830075,
      "epoch": 3.0875347360063516,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004061946067075503,
      "loss": 4.9642,
      "mean_token_accuracy": 0.1941882610321045,
      "num_tokens": 68151891.0,
      "step": 31110
    },
    {
      "entropy": 5.6143604755401615,
      "epoch": 3.088030964668519,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004061658201928203,
      "loss": 4.9551,
      "mean_token_accuracy": 0.19583060592412949,
      "num_tokens": 68162537.0,
      "step": 31115
    },
    {
      "entropy": 5.679628992080689,
      "epoch": 3.0885271933306866,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004061370304254606,
      "loss": 5.1319,
      "mean_token_accuracy": 0.1827751338481903,
      "num_tokens": 68173393.0,
      "step": 31120
    },
    {
      "entropy": 5.626796817779541,
      "epoch": 3.0890234219928545,
      "grad_norm": 1.140625,
      "learning_rate": 0.000406108237406185,
      "loss": 5.0191,
      "mean_token_accuracy": 0.18605466783046723,
      "num_tokens": 68184937.0,
      "step": 31125
    },
    {
      "entropy": 5.638377046585083,
      "epoch": 3.089519650655022,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040607944113570777,
      "loss": 5.0346,
      "mean_token_accuracy": 0.18998197168111802,
      "num_tokens": 68196962.0,
      "step": 31130
    },
    {
      "entropy": 5.725616836547852,
      "epoch": 3.0900158793171895,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004060506416147428,
      "loss": 5.0263,
      "mean_token_accuracy": 0.18621935397386552,
      "num_tokens": 68208603.0,
      "step": 31135
    },
    {
      "entropy": 5.625511837005615,
      "epoch": 3.090512107979357,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004060218388440043,
      "loss": 4.989,
      "mean_token_accuracy": 0.18789745718240738,
      "num_tokens": 68219424.0,
      "step": 31140
    },
    {
      "entropy": 5.594231748580933,
      "epoch": 3.0910083366415244,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004059930328242065,
      "loss": 4.9352,
      "mean_token_accuracy": 0.19473981708288193,
      "num_tokens": 68229758.0,
      "step": 31145
    },
    {
      "entropy": 5.548039817810059,
      "epoch": 3.091504565303692,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040596422355606383,
      "loss": 4.931,
      "mean_token_accuracy": 0.18891681879758834,
      "num_tokens": 68239146.0,
      "step": 31150
    },
    {
      "entropy": 5.618942737579346,
      "epoch": 3.0920007939658594,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004059354110402905,
      "loss": 5.0029,
      "mean_token_accuracy": 0.18471025228500365,
      "num_tokens": 68249339.0,
      "step": 31155
    },
    {
      "entropy": 5.643098592758179,
      "epoch": 3.092497022628027,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040590659527760123,
      "loss": 4.9698,
      "mean_token_accuracy": 0.19339598417282106,
      "num_tokens": 68260136.0,
      "step": 31160
    },
    {
      "entropy": 5.674410581588745,
      "epoch": 3.0929932512901943,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040587777626871046,
      "loss": 4.9975,
      "mean_token_accuracy": 0.1864364594221115,
      "num_tokens": 68270033.0,
      "step": 31165
    },
    {
      "entropy": 5.659496879577636,
      "epoch": 3.0934894799523622,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004058489540143328,
      "loss": 5.046,
      "mean_token_accuracy": 0.18138502687215804,
      "num_tokens": 68283344.0,
      "step": 31170
    },
    {
      "entropy": 5.637324619293213,
      "epoch": 3.0939857086145297,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004058201285151831,
      "loss": 4.9601,
      "mean_token_accuracy": 0.1891259327530861,
      "num_tokens": 68294604.0,
      "step": 31175
    },
    {
      "entropy": 5.634547472000122,
      "epoch": 3.094481937276697,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004057912997719761,
      "loss": 4.9435,
      "mean_token_accuracy": 0.19683267176151276,
      "num_tokens": 68304436.0,
      "step": 31180
    },
    {
      "entropy": 5.574007320404053,
      "epoch": 3.0949781659388647,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004057624677854266,
      "loss": 4.9102,
      "mean_token_accuracy": 0.19377656280994415,
      "num_tokens": 68315593.0,
      "step": 31185
    },
    {
      "entropy": 5.614682054519653,
      "epoch": 3.095474394601032,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00040573363255624976,
      "loss": 5.0078,
      "mean_token_accuracy": 0.1918174982070923,
      "num_tokens": 68327540.0,
      "step": 31190
    },
    {
      "entropy": 5.742231464385986,
      "epoch": 3.0959706232631996,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004057047940851605,
      "loss": 5.0747,
      "mean_token_accuracy": 0.1799856334924698,
      "num_tokens": 68339784.0,
      "step": 31195
    },
    {
      "entropy": 5.64006462097168,
      "epoch": 3.096466851925367,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004056759523728739,
      "loss": 4.9901,
      "mean_token_accuracy": 0.1858578160405159,
      "num_tokens": 68351182.0,
      "step": 31200
    },
    {
      "entropy": 5.634047603607177,
      "epoch": 3.0969630805875346,
      "grad_norm": 1.125,
      "learning_rate": 0.00040564710742010545,
      "loss": 5.0332,
      "mean_token_accuracy": 0.18002903014421462,
      "num_tokens": 68361726.0,
      "step": 31205
    },
    {
      "entropy": 5.633721590042114,
      "epoch": 3.0974593092497025,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004056182592275702,
      "loss": 5.0155,
      "mean_token_accuracy": 0.19055891633033753,
      "num_tokens": 68373167.0,
      "step": 31210
    },
    {
      "entropy": 5.685722208023071,
      "epoch": 3.09795553791187,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004055894077959835,
      "loss": 5.0566,
      "mean_token_accuracy": 0.1842707246541977,
      "num_tokens": 68384646.0,
      "step": 31215
    },
    {
      "entropy": 5.618046522140503,
      "epoch": 3.0984517665740374,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004055605531260609,
      "loss": 4.9794,
      "mean_token_accuracy": 0.19083010405302048,
      "num_tokens": 68395953.0,
      "step": 31220
    },
    {
      "entropy": 5.692149877548218,
      "epoch": 3.098947995236205,
      "grad_norm": 1.125,
      "learning_rate": 0.00040553169521851794,
      "loss": 5.0438,
      "mean_token_accuracy": 0.18652877509593963,
      "num_tokens": 68407167.0,
      "step": 31225
    },
    {
      "entropy": 5.659059524536133,
      "epoch": 3.0994442238983724,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040550283407407025,
      "loss": 5.0449,
      "mean_token_accuracy": 0.1863121509552002,
      "num_tokens": 68419530.0,
      "step": 31230
    },
    {
      "entropy": 5.665782117843628,
      "epoch": 3.09994045256054,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004054739696934334,
      "loss": 5.0087,
      "mean_token_accuracy": 0.18602059334516524,
      "num_tokens": 68430970.0,
      "step": 31235
    },
    {
      "entropy": 5.613737964630127,
      "epoch": 3.1004366812227073,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040544510207732337,
      "loss": 4.9488,
      "mean_token_accuracy": 0.19471430778503418,
      "num_tokens": 68441257.0,
      "step": 31240
    },
    {
      "entropy": 5.567920684814453,
      "epoch": 3.100932909884875,
      "grad_norm": 1.125,
      "learning_rate": 0.00040541623122645576,
      "loss": 4.9696,
      "mean_token_accuracy": 0.18740631341934205,
      "num_tokens": 68452493.0,
      "step": 31245
    },
    {
      "entropy": 5.561654090881348,
      "epoch": 3.1014291385470423,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040538735714154675,
      "loss": 4.9199,
      "mean_token_accuracy": 0.19658785462379455,
      "num_tokens": 68464023.0,
      "step": 31250
    },
    {
      "entropy": 5.62626166343689,
      "epoch": 3.10192536720921,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004053584798233122,
      "loss": 5.0036,
      "mean_token_accuracy": 0.19383625239133834,
      "num_tokens": 68474519.0,
      "step": 31255
    },
    {
      "entropy": 5.667145156860352,
      "epoch": 3.1024215958713777,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040532959927246814,
      "loss": 4.9779,
      "mean_token_accuracy": 0.19276228100061416,
      "num_tokens": 68486239.0,
      "step": 31260
    },
    {
      "entropy": 5.666027069091797,
      "epoch": 3.102917824533545,
      "grad_norm": 1.125,
      "learning_rate": 0.000405300715489731,
      "loss": 5.0352,
      "mean_token_accuracy": 0.18721785992383957,
      "num_tokens": 68495921.0,
      "step": 31265
    },
    {
      "entropy": 5.649004220962524,
      "epoch": 3.1034140531957126,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004052718284758169,
      "loss": 4.9424,
      "mean_token_accuracy": 0.19188570678234101,
      "num_tokens": 68505773.0,
      "step": 31270
    },
    {
      "entropy": 5.6930783748626705,
      "epoch": 3.10391028185788,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000405242938231442,
      "loss": 5.0662,
      "mean_token_accuracy": 0.18246551007032394,
      "num_tokens": 68516563.0,
      "step": 31275
    },
    {
      "entropy": 5.660870885848999,
      "epoch": 3.1044065105200476,
      "grad_norm": 1.171875,
      "learning_rate": 0.000405214044757323,
      "loss": 5.0088,
      "mean_token_accuracy": 0.19283432513475418,
      "num_tokens": 68527952.0,
      "step": 31280
    },
    {
      "entropy": 5.6522010326385494,
      "epoch": 3.104902739182215,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004051851480541763,
      "loss": 4.9912,
      "mean_token_accuracy": 0.18620000779628754,
      "num_tokens": 68538450.0,
      "step": 31285
    },
    {
      "entropy": 5.598599624633789,
      "epoch": 3.1053989678443825,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004051562481227184,
      "loss": 4.9487,
      "mean_token_accuracy": 0.18577575981616973,
      "num_tokens": 68547941.0,
      "step": 31290
    },
    {
      "entropy": 5.599919319152832,
      "epoch": 3.10589519650655,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000405127344963666,
      "loss": 4.9486,
      "mean_token_accuracy": 0.19299937933683395,
      "num_tokens": 68558525.0,
      "step": 31295
    },
    {
      "entropy": 5.6179687023162845,
      "epoch": 3.106391425168718,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004050984385777359,
      "loss": 4.9422,
      "mean_token_accuracy": 0.18543791025877,
      "num_tokens": 68569487.0,
      "step": 31300
    },
    {
      "entropy": 5.629094552993775,
      "epoch": 3.1068876538308854,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004050695289656448,
      "loss": 4.9911,
      "mean_token_accuracy": 0.18663630187511443,
      "num_tokens": 68580069.0,
      "step": 31305
    },
    {
      "entropy": 5.624674558639526,
      "epoch": 3.107383882493053,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040504061612810964,
      "loss": 4.9594,
      "mean_token_accuracy": 0.18712171018123627,
      "num_tokens": 68591511.0,
      "step": 31310
    },
    {
      "entropy": 5.5355064868927,
      "epoch": 3.1078801111552203,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004050117000658475,
      "loss": 4.9779,
      "mean_token_accuracy": 0.19108108580112457,
      "num_tokens": 68602464.0,
      "step": 31315
    },
    {
      "entropy": 5.635071516036987,
      "epoch": 3.108376339817388,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040498278077957534,
      "loss": 4.9528,
      "mean_token_accuracy": 0.18842452764511108,
      "num_tokens": 68613492.0,
      "step": 31320
    },
    {
      "entropy": 5.603519439697266,
      "epoch": 3.1088725684795553,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004049538582700102,
      "loss": 4.9166,
      "mean_token_accuracy": 0.19536224603652955,
      "num_tokens": 68625259.0,
      "step": 31325
    },
    {
      "entropy": 5.639528226852417,
      "epoch": 3.1093687971417228,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004049249325378695,
      "loss": 5.051,
      "mean_token_accuracy": 0.1806972250342369,
      "num_tokens": 68635455.0,
      "step": 31330
    },
    {
      "entropy": 5.650330352783203,
      "epoch": 3.1098650258038902,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040489600358387036,
      "loss": 5.0456,
      "mean_token_accuracy": 0.1790520206093788,
      "num_tokens": 68648102.0,
      "step": 31335
    },
    {
      "entropy": 5.676381349563599,
      "epoch": 3.110361254466058,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004048670714087303,
      "loss": 5.0137,
      "mean_token_accuracy": 0.1891563504934311,
      "num_tokens": 68660854.0,
      "step": 31340
    },
    {
      "entropy": 5.5275005340576175,
      "epoch": 3.1108574831282256,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004048381360131667,
      "loss": 4.8628,
      "mean_token_accuracy": 0.20227722972631454,
      "num_tokens": 68671839.0,
      "step": 31345
    },
    {
      "entropy": 5.655065298080444,
      "epoch": 3.111353711790393,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004048091973978971,
      "loss": 5.0045,
      "mean_token_accuracy": 0.19239673018455505,
      "num_tokens": 68682535.0,
      "step": 31350
    },
    {
      "entropy": 5.624544858932495,
      "epoch": 3.1118499404525606,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004047802555636391,
      "loss": 4.9617,
      "mean_token_accuracy": 0.19382238388061523,
      "num_tokens": 68694283.0,
      "step": 31355
    },
    {
      "entropy": 5.657701730728149,
      "epoch": 3.112346169114728,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004047513105111104,
      "loss": 4.9963,
      "mean_token_accuracy": 0.1855426549911499,
      "num_tokens": 68704268.0,
      "step": 31360
    },
    {
      "entropy": 5.615449714660644,
      "epoch": 3.1128423977768955,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040472236224102883,
      "loss": 4.9086,
      "mean_token_accuracy": 0.18791035413742066,
      "num_tokens": 68714762.0,
      "step": 31365
    },
    {
      "entropy": 5.585416173934936,
      "epoch": 3.113338626439063,
      "grad_norm": 1.125,
      "learning_rate": 0.00040469341075411217,
      "loss": 4.927,
      "mean_token_accuracy": 0.18320893496274948,
      "num_tokens": 68725302.0,
      "step": 31370
    },
    {
      "entropy": 5.6393999576568605,
      "epoch": 3.1138348551012305,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004046644560510783,
      "loss": 5.0398,
      "mean_token_accuracy": 0.17813299149274825,
      "num_tokens": 68736893.0,
      "step": 31375
    },
    {
      "entropy": 5.587645387649536,
      "epoch": 3.114331083763398,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040463549813264557,
      "loss": 4.9538,
      "mean_token_accuracy": 0.18947384953498841,
      "num_tokens": 68748202.0,
      "step": 31380
    },
    {
      "entropy": 5.667027807235717,
      "epoch": 3.114827312425566,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040460653699953166,
      "loss": 4.9441,
      "mean_token_accuracy": 0.1981617331504822,
      "num_tokens": 68759033.0,
      "step": 31385
    },
    {
      "entropy": 5.645000696182251,
      "epoch": 3.1153235410877333,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040457757265245485,
      "loss": 4.9497,
      "mean_token_accuracy": 0.19468911737203598,
      "num_tokens": 68770879.0,
      "step": 31390
    },
    {
      "entropy": 5.598604297637939,
      "epoch": 3.115819769749901,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004045486050921336,
      "loss": 4.9365,
      "mean_token_accuracy": 0.19535958021879196,
      "num_tokens": 68782087.0,
      "step": 31395
    },
    {
      "entropy": 5.605326747894287,
      "epoch": 3.1163159984120683,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000404519634319286,
      "loss": 4.9712,
      "mean_token_accuracy": 0.19178367406129837,
      "num_tokens": 68792355.0,
      "step": 31400
    },
    {
      "entropy": 5.611901473999024,
      "epoch": 3.1168122270742358,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040449066033463053,
      "loss": 4.9902,
      "mean_token_accuracy": 0.1868669033050537,
      "num_tokens": 68804255.0,
      "step": 31405
    },
    {
      "entropy": 5.6020527362823485,
      "epoch": 3.1173084557364032,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004044616831388858,
      "loss": 4.9477,
      "mean_token_accuracy": 0.1908596247434616,
      "num_tokens": 68816102.0,
      "step": 31410
    },
    {
      "entropy": 5.627028179168701,
      "epoch": 3.1178046843985707,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004044327027327703,
      "loss": 4.9921,
      "mean_token_accuracy": 0.18673000037670134,
      "num_tokens": 68827303.0,
      "step": 31415
    },
    {
      "entropy": 5.637691164016724,
      "epoch": 3.118300913060738,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004044037191170026,
      "loss": 4.9713,
      "mean_token_accuracy": 0.18843668699264526,
      "num_tokens": 68838418.0,
      "step": 31420
    },
    {
      "entropy": 5.712678384780884,
      "epoch": 3.118797141722906,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040437473229230155,
      "loss": 5.0094,
      "mean_token_accuracy": 0.18359093964099885,
      "num_tokens": 68848431.0,
      "step": 31425
    },
    {
      "entropy": 5.6165141582489015,
      "epoch": 3.1192933703850736,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004043457422593859,
      "loss": 4.9276,
      "mean_token_accuracy": 0.18596412986516953,
      "num_tokens": 68858983.0,
      "step": 31430
    },
    {
      "entropy": 5.628681898117065,
      "epoch": 3.119789599047241,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004043167490189746,
      "loss": 4.9798,
      "mean_token_accuracy": 0.19337759613990785,
      "num_tokens": 68870170.0,
      "step": 31435
    },
    {
      "entropy": 5.606418752670288,
      "epoch": 3.1202858277094085,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040428775257178655,
      "loss": 5.0155,
      "mean_token_accuracy": 0.19140416383743286,
      "num_tokens": 68880242.0,
      "step": 31440
    },
    {
      "entropy": 5.681541585922242,
      "epoch": 3.120782056371576,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040425875291854083,
      "loss": 5.0759,
      "mean_token_accuracy": 0.1782371073961258,
      "num_tokens": 68892150.0,
      "step": 31445
    },
    {
      "entropy": 5.710230875015259,
      "epoch": 3.1212782850337435,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040422975005995656,
      "loss": 5.0581,
      "mean_token_accuracy": 0.1810663804411888,
      "num_tokens": 68903164.0,
      "step": 31450
    },
    {
      "entropy": 5.606203508377075,
      "epoch": 3.121774513695911,
      "grad_norm": 1.125,
      "learning_rate": 0.0004042007439967529,
      "loss": 4.9593,
      "mean_token_accuracy": 0.18514100164175035,
      "num_tokens": 68914514.0,
      "step": 31455
    },
    {
      "entropy": 5.62603120803833,
      "epoch": 3.1222707423580784,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004041717347296493,
      "loss": 4.9686,
      "mean_token_accuracy": 0.19086654782295226,
      "num_tokens": 68925366.0,
      "step": 31460
    },
    {
      "entropy": 5.6191527366638185,
      "epoch": 3.122766971020246,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004041427222593649,
      "loss": 4.9282,
      "mean_token_accuracy": 0.1862796351313591,
      "num_tokens": 68936019.0,
      "step": 31465
    },
    {
      "entropy": 5.528175258636475,
      "epoch": 3.123263199682414,
      "grad_norm": 1.125,
      "learning_rate": 0.00040411370658661935,
      "loss": 4.8889,
      "mean_token_accuracy": 0.19416557550430297,
      "num_tokens": 68947405.0,
      "step": 31470
    },
    {
      "entropy": 5.602405786514282,
      "epoch": 3.1237594283445813,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040408468771213206,
      "loss": 4.9819,
      "mean_token_accuracy": 0.18933648616075516,
      "num_tokens": 68958822.0,
      "step": 31475
    },
    {
      "entropy": 5.633082580566406,
      "epoch": 3.124255657006749,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040405566563662265,
      "loss": 5.0068,
      "mean_token_accuracy": 0.18882711678743364,
      "num_tokens": 68969992.0,
      "step": 31480
    },
    {
      "entropy": 5.6115195751190186,
      "epoch": 3.1247518856689163,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040402664036081086,
      "loss": 4.9628,
      "mean_token_accuracy": 0.1841059908270836,
      "num_tokens": 68982081.0,
      "step": 31485
    },
    {
      "entropy": 5.6508087635040285,
      "epoch": 3.1252481143310837,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004039976118854164,
      "loss": 5.0009,
      "mean_token_accuracy": 0.18401008695364,
      "num_tokens": 68993430.0,
      "step": 31490
    },
    {
      "entropy": 5.692102432250977,
      "epoch": 3.125744342993251,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004039685802111591,
      "loss": 5.0503,
      "mean_token_accuracy": 0.1878976047039032,
      "num_tokens": 69004939.0,
      "step": 31495
    },
    {
      "entropy": 5.607555198669433,
      "epoch": 3.1262405716554187,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040393954533875904,
      "loss": 4.9807,
      "mean_token_accuracy": 0.19158879071474075,
      "num_tokens": 69016293.0,
      "step": 31500
    },
    {
      "entropy": 5.688988208770752,
      "epoch": 3.126736800317586,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040391050726893595,
      "loss": 5.0795,
      "mean_token_accuracy": 0.18065052330493928,
      "num_tokens": 69029087.0,
      "step": 31505
    },
    {
      "entropy": 5.650511837005615,
      "epoch": 3.127233028979754,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004038814660024101,
      "loss": 4.9836,
      "mean_token_accuracy": 0.18312305063009263,
      "num_tokens": 69040937.0,
      "step": 31510
    },
    {
      "entropy": 5.671603965759277,
      "epoch": 3.1277292576419216,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004038524215399016,
      "loss": 5.0589,
      "mean_token_accuracy": 0.18778420090675355,
      "num_tokens": 69052212.0,
      "step": 31515
    },
    {
      "entropy": 5.651321840286255,
      "epoch": 3.128225486304089,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004038233738821308,
      "loss": 4.9884,
      "mean_token_accuracy": 0.19498463869094848,
      "num_tokens": 69061951.0,
      "step": 31520
    },
    {
      "entropy": 5.67041392326355,
      "epoch": 3.1287217149662565,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004037943230298178,
      "loss": 4.9105,
      "mean_token_accuracy": 0.20215070098638535,
      "num_tokens": 69072156.0,
      "step": 31525
    },
    {
      "entropy": 5.640856552124023,
      "epoch": 3.129217943628424,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040376526898368314,
      "loss": 4.9808,
      "mean_token_accuracy": 0.18387744426727295,
      "num_tokens": 69084032.0,
      "step": 31530
    },
    {
      "entropy": 5.658621740341187,
      "epoch": 3.1297141722905915,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00040373621174444736,
      "loss": 5.027,
      "mean_token_accuracy": 0.18640570640563964,
      "num_tokens": 69096436.0,
      "step": 31535
    },
    {
      "entropy": 5.624767589569092,
      "epoch": 3.130210400952759,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040370715131283076,
      "loss": 4.9347,
      "mean_token_accuracy": 0.18998827785253525,
      "num_tokens": 69107090.0,
      "step": 31540
    },
    {
      "entropy": 5.607313871383667,
      "epoch": 3.1307066296149264,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004036780876895543,
      "loss": 4.9111,
      "mean_token_accuracy": 0.19557999521493913,
      "num_tokens": 69117644.0,
      "step": 31545
    },
    {
      "entropy": 5.58624677658081,
      "epoch": 3.131202858277094,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004036490208753385,
      "loss": 4.9838,
      "mean_token_accuracy": 0.18959536850452424,
      "num_tokens": 69129639.0,
      "step": 31550
    },
    {
      "entropy": 5.5926048278808596,
      "epoch": 3.131699086939262,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040361995087090407,
      "loss": 4.9284,
      "mean_token_accuracy": 0.1908608376979828,
      "num_tokens": 69139429.0,
      "step": 31555
    },
    {
      "entropy": 5.504916715621948,
      "epoch": 3.1321953156014293,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040359087767697223,
      "loss": 4.8852,
      "mean_token_accuracy": 0.19930580407381057,
      "num_tokens": 69150777.0,
      "step": 31560
    },
    {
      "entropy": 5.625134801864624,
      "epoch": 3.1326915442635968,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040356180129426356,
      "loss": 4.9519,
      "mean_token_accuracy": 0.18425585925579072,
      "num_tokens": 69161970.0,
      "step": 31565
    },
    {
      "entropy": 5.592458772659302,
      "epoch": 3.1331877729257642,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004035327217234993,
      "loss": 4.9342,
      "mean_token_accuracy": 0.19094129651784897,
      "num_tokens": 69172990.0,
      "step": 31570
    },
    {
      "entropy": 5.600011157989502,
      "epoch": 3.1336840015879317,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040350363896540054,
      "loss": 4.9736,
      "mean_token_accuracy": 0.19243034422397615,
      "num_tokens": 69184044.0,
      "step": 31575
    },
    {
      "entropy": 5.562719631195068,
      "epoch": 3.134180230250099,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040347455302068835,
      "loss": 4.9343,
      "mean_token_accuracy": 0.1842356353998184,
      "num_tokens": 69194841.0,
      "step": 31580
    },
    {
      "entropy": 5.636740732192993,
      "epoch": 3.1346764589122667,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040344546389008406,
      "loss": 5.0081,
      "mean_token_accuracy": 0.18695726841688157,
      "num_tokens": 69205465.0,
      "step": 31585
    },
    {
      "entropy": 5.6680398941040036,
      "epoch": 3.135172687574434,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040341637157430916,
      "loss": 5.0301,
      "mean_token_accuracy": 0.1845214009284973,
      "num_tokens": 69215831.0,
      "step": 31590
    },
    {
      "entropy": 5.634030961990357,
      "epoch": 3.135668916236602,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004033872760740849,
      "loss": 4.9737,
      "mean_token_accuracy": 0.19118852615356446,
      "num_tokens": 69225156.0,
      "step": 31595
    },
    {
      "entropy": 5.607449626922607,
      "epoch": 3.1361651448987695,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004033581773901328,
      "loss": 5.0156,
      "mean_token_accuracy": 0.18128662109375,
      "num_tokens": 69237590.0,
      "step": 31600
    },
    {
      "entropy": 5.6489661693572994,
      "epoch": 3.136661373560937,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040332907552317453,
      "loss": 4.9714,
      "mean_token_accuracy": 0.19201709032058717,
      "num_tokens": 69246904.0,
      "step": 31605
    },
    {
      "entropy": 5.7488209247589115,
      "epoch": 3.1371576022231045,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004032999704739317,
      "loss": 5.0849,
      "mean_token_accuracy": 0.17763118594884872,
      "num_tokens": 69258289.0,
      "step": 31610
    },
    {
      "entropy": 5.6261731624603275,
      "epoch": 3.137653830885272,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004032708622431261,
      "loss": 4.9855,
      "mean_token_accuracy": 0.18433631509542464,
      "num_tokens": 69268232.0,
      "step": 31615
    },
    {
      "entropy": 5.6252909183502195,
      "epoch": 3.1381500595474394,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004032417508314795,
      "loss": 4.9603,
      "mean_token_accuracy": 0.18173281699419022,
      "num_tokens": 69278596.0,
      "step": 31620
    },
    {
      "entropy": 5.646913814544678,
      "epoch": 3.138646288209607,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040321263623971376,
      "loss": 5.0537,
      "mean_token_accuracy": 0.18504312187433242,
      "num_tokens": 69290675.0,
      "step": 31625
    },
    {
      "entropy": 5.6617724895477295,
      "epoch": 3.1391425168717744,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000403183518468551,
      "loss": 5.0455,
      "mean_token_accuracy": 0.18834568411111832,
      "num_tokens": 69303600.0,
      "step": 31630
    },
    {
      "entropy": 5.67111234664917,
      "epoch": 3.139638745533942,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004031543975187132,
      "loss": 5.075,
      "mean_token_accuracy": 0.1843728870153427,
      "num_tokens": 69314672.0,
      "step": 31635
    },
    {
      "entropy": 5.592279291152954,
      "epoch": 3.1401349741961098,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040312527339092237,
      "loss": 4.9277,
      "mean_token_accuracy": 0.19602699279785157,
      "num_tokens": 69325574.0,
      "step": 31640
    },
    {
      "entropy": 5.601968240737915,
      "epoch": 3.1406312028582772,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040309614608590095,
      "loss": 4.916,
      "mean_token_accuracy": 0.18815799355506896,
      "num_tokens": 69336430.0,
      "step": 31645
    },
    {
      "entropy": 5.561579799652099,
      "epoch": 3.1411274315204447,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004030670156043711,
      "loss": 4.8516,
      "mean_token_accuracy": 0.2012013837695122,
      "num_tokens": 69347007.0,
      "step": 31650
    },
    {
      "entropy": 5.621777677536011,
      "epoch": 3.141623660182612,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040303788194705523,
      "loss": 4.9899,
      "mean_token_accuracy": 0.18767525553703307,
      "num_tokens": 69357796.0,
      "step": 31655
    },
    {
      "entropy": 5.653431081771851,
      "epoch": 3.1421198888447797,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040300874511467574,
      "loss": 5.0131,
      "mean_token_accuracy": 0.19130476713180541,
      "num_tokens": 69370098.0,
      "step": 31660
    },
    {
      "entropy": 5.6423180103302,
      "epoch": 3.142616117506947,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040297960510795526,
      "loss": 4.9863,
      "mean_token_accuracy": 0.17858834713697433,
      "num_tokens": 69380991.0,
      "step": 31665
    },
    {
      "entropy": 5.650356006622315,
      "epoch": 3.1431123461691146,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004029504619276163,
      "loss": 4.9696,
      "mean_token_accuracy": 0.18673238307237625,
      "num_tokens": 69391748.0,
      "step": 31670
    },
    {
      "entropy": 5.6788105964660645,
      "epoch": 3.143608574831282,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040292131557438166,
      "loss": 5.0215,
      "mean_token_accuracy": 0.17990265786647797,
      "num_tokens": 69404964.0,
      "step": 31675
    },
    {
      "entropy": 5.59475040435791,
      "epoch": 3.14410480349345,
      "grad_norm": 1.171875,
      "learning_rate": 0.000402892166048974,
      "loss": 4.9558,
      "mean_token_accuracy": 0.1934192031621933,
      "num_tokens": 69416045.0,
      "step": 31680
    },
    {
      "entropy": 5.597433423995971,
      "epoch": 3.1446010321556175,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004028630133521162,
      "loss": 4.9448,
      "mean_token_accuracy": 0.18341091722249986,
      "num_tokens": 69427489.0,
      "step": 31685
    },
    {
      "entropy": 5.583539867401123,
      "epoch": 3.145097260817785,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004028338574845312,
      "loss": 4.9506,
      "mean_token_accuracy": 0.1881953626871109,
      "num_tokens": 69437929.0,
      "step": 31690
    },
    {
      "entropy": 5.615532064437867,
      "epoch": 3.1455934894799524,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040280469844694203,
      "loss": 5.0172,
      "mean_token_accuracy": 0.18491577208042145,
      "num_tokens": 69449449.0,
      "step": 31695
    },
    {
      "entropy": 5.583782768249511,
      "epoch": 3.14608971814212,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004027755362400716,
      "loss": 5.0241,
      "mean_token_accuracy": 0.18237591832876204,
      "num_tokens": 69459451.0,
      "step": 31700
    },
    {
      "entropy": 5.680127096176148,
      "epoch": 3.1465859468042874,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004027463708646434,
      "loss": 5.0298,
      "mean_token_accuracy": 0.18484761267900468,
      "num_tokens": 69470615.0,
      "step": 31705
    },
    {
      "entropy": 5.688493919372559,
      "epoch": 3.147082175466455,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004027172023213803,
      "loss": 4.9757,
      "mean_token_accuracy": 0.18987265080213547,
      "num_tokens": 69481123.0,
      "step": 31710
    },
    {
      "entropy": 5.653881406784057,
      "epoch": 3.1475784041286223,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004026880306110058,
      "loss": 5.0147,
      "mean_token_accuracy": 0.18177929073572158,
      "num_tokens": 69491590.0,
      "step": 31715
    },
    {
      "entropy": 5.570854568481446,
      "epoch": 3.14807463279079,
      "grad_norm": 1.125,
      "learning_rate": 0.00040265885573424336,
      "loss": 4.9765,
      "mean_token_accuracy": 0.18853390216827393,
      "num_tokens": 69501883.0,
      "step": 31720
    },
    {
      "entropy": 5.577920866012573,
      "epoch": 3.1485708614529577,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040262967769181634,
      "loss": 4.9251,
      "mean_token_accuracy": 0.18854814618825913,
      "num_tokens": 69514556.0,
      "step": 31725
    },
    {
      "entropy": 5.687356996536255,
      "epoch": 3.149067090115125,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040260049648444836,
      "loss": 5.0498,
      "mean_token_accuracy": 0.1826518326997757,
      "num_tokens": 69525483.0,
      "step": 31730
    },
    {
      "entropy": 5.656803035736084,
      "epoch": 3.1495633187772927,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040257131211286307,
      "loss": 5.0192,
      "mean_token_accuracy": 0.1866036683320999,
      "num_tokens": 69535681.0,
      "step": 31735
    },
    {
      "entropy": 5.664316129684448,
      "epoch": 3.15005954743946,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000402542124577784,
      "loss": 5.1069,
      "mean_token_accuracy": 0.17042193710803985,
      "num_tokens": 69547825.0,
      "step": 31740
    },
    {
      "entropy": 5.645622205734253,
      "epoch": 3.1505557761016276,
      "grad_norm": 1.125,
      "learning_rate": 0.00040251293387993513,
      "loss": 4.9735,
      "mean_token_accuracy": 0.1888951539993286,
      "num_tokens": 69557922.0,
      "step": 31745
    },
    {
      "entropy": 5.670262956619263,
      "epoch": 3.151052004763795,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004024837400200403,
      "loss": 5.0358,
      "mean_token_accuracy": 0.18418613374233245,
      "num_tokens": 69568165.0,
      "step": 31750
    },
    {
      "entropy": 5.574112367630005,
      "epoch": 3.1515482334259626,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040245454299882343,
      "loss": 4.8367,
      "mean_token_accuracy": 0.19499546587467192,
      "num_tokens": 69578828.0,
      "step": 31755
    },
    {
      "entropy": 5.6352111339569095,
      "epoch": 3.15204446208813,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040242534281700843,
      "loss": 5.0339,
      "mean_token_accuracy": 0.18633735179901123,
      "num_tokens": 69590527.0,
      "step": 31760
    },
    {
      "entropy": 5.622867202758789,
      "epoch": 3.152540690750298,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040239613947531965,
      "loss": 4.9836,
      "mean_token_accuracy": 0.18874476850032806,
      "num_tokens": 69602032.0,
      "step": 31765
    },
    {
      "entropy": 5.598103284835815,
      "epoch": 3.1530369194124654,
      "grad_norm": 1.125,
      "learning_rate": 0.00040236693297448107,
      "loss": 5.0005,
      "mean_token_accuracy": 0.18206772059202195,
      "num_tokens": 69613490.0,
      "step": 31770
    },
    {
      "entropy": 5.675497102737427,
      "epoch": 3.153533148074633,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004023377233152169,
      "loss": 5.0958,
      "mean_token_accuracy": 0.17801947742700577,
      "num_tokens": 69624623.0,
      "step": 31775
    },
    {
      "entropy": 5.661495494842529,
      "epoch": 3.1540293767368004,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040230851049825173,
      "loss": 5.0495,
      "mean_token_accuracy": 0.18577795326709748,
      "num_tokens": 69634956.0,
      "step": 31780
    },
    {
      "entropy": 5.658870029449463,
      "epoch": 3.154525605398968,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040227929452430974,
      "loss": 4.9547,
      "mean_token_accuracy": 0.1931302949786186,
      "num_tokens": 69645370.0,
      "step": 31785
    },
    {
      "entropy": 5.704684543609619,
      "epoch": 3.1550218340611353,
      "grad_norm": 1.0,
      "learning_rate": 0.00040225007539411553,
      "loss": 5.042,
      "mean_token_accuracy": 0.17059522569179536,
      "num_tokens": 69656328.0,
      "step": 31790
    },
    {
      "entropy": 5.593551683425903,
      "epoch": 3.155518062723303,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004022208531083936,
      "loss": 4.9797,
      "mean_token_accuracy": 0.18946101814508437,
      "num_tokens": 69668584.0,
      "step": 31795
    },
    {
      "entropy": 5.614719486236572,
      "epoch": 3.1560142913854703,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004021916276678688,
      "loss": 5.0171,
      "mean_token_accuracy": 0.18275225907564163,
      "num_tokens": 69677977.0,
      "step": 31800
    },
    {
      "entropy": 5.608735609054565,
      "epoch": 3.1565105200476378,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040216239907326555,
      "loss": 4.9911,
      "mean_token_accuracy": 0.1850469723343849,
      "num_tokens": 69688979.0,
      "step": 31805
    },
    {
      "entropy": 5.622328376770019,
      "epoch": 3.1570067487098052,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004021331673253089,
      "loss": 5.0261,
      "mean_token_accuracy": 0.19125883281230927,
      "num_tokens": 69698567.0,
      "step": 31810
    },
    {
      "entropy": 5.668766450881958,
      "epoch": 3.157502977371973,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004021039324247236,
      "loss": 4.943,
      "mean_token_accuracy": 0.18461458384990692,
      "num_tokens": 69708623.0,
      "step": 31815
    },
    {
      "entropy": 5.614979314804077,
      "epoch": 3.1579992060341406,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040207469437223466,
      "loss": 5.0113,
      "mean_token_accuracy": 0.1829611524939537,
      "num_tokens": 69719613.0,
      "step": 31820
    },
    {
      "entropy": 5.608452463150025,
      "epoch": 3.158495434696308,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040204545316856714,
      "loss": 5.0045,
      "mean_token_accuracy": 0.18467500060796738,
      "num_tokens": 69731407.0,
      "step": 31825
    },
    {
      "entropy": 5.597069120407104,
      "epoch": 3.1589916633584756,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004020162088144461,
      "loss": 4.9408,
      "mean_token_accuracy": 0.1920430466532707,
      "num_tokens": 69742363.0,
      "step": 31830
    },
    {
      "entropy": 5.603009271621704,
      "epoch": 3.159487892020643,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040198696131059684,
      "loss": 4.947,
      "mean_token_accuracy": 0.1921755850315094,
      "num_tokens": 69754878.0,
      "step": 31835
    },
    {
      "entropy": 5.597185468673706,
      "epoch": 3.1599841206828105,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040195771065774455,
      "loss": 4.9254,
      "mean_token_accuracy": 0.1925751596689224,
      "num_tokens": 69764986.0,
      "step": 31840
    },
    {
      "entropy": 5.581723880767822,
      "epoch": 3.160480349344978,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004019284568566146,
      "loss": 4.9721,
      "mean_token_accuracy": 0.19255149960517884,
      "num_tokens": 69775762.0,
      "step": 31845
    },
    {
      "entropy": 5.6206221103668215,
      "epoch": 3.1609765780071455,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040189919990793243,
      "loss": 5.0363,
      "mean_token_accuracy": 0.18756873160600662,
      "num_tokens": 69786411.0,
      "step": 31850
    },
    {
      "entropy": 5.644739866256714,
      "epoch": 3.1614728066693134,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040186993981242353,
      "loss": 4.9852,
      "mean_token_accuracy": 0.187366347014904,
      "num_tokens": 69797140.0,
      "step": 31855
    },
    {
      "entropy": 5.572748804092408,
      "epoch": 3.161969035331481,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004018406765708135,
      "loss": 4.8716,
      "mean_token_accuracy": 0.1947699025273323,
      "num_tokens": 69807452.0,
      "step": 31860
    },
    {
      "entropy": 5.582936525344849,
      "epoch": 3.1624652639936484,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040181141018382806,
      "loss": 4.9691,
      "mean_token_accuracy": 0.18646900951862336,
      "num_tokens": 69819650.0,
      "step": 31865
    },
    {
      "entropy": 5.602268505096435,
      "epoch": 3.162961492655816,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040178214065219285,
      "loss": 4.9611,
      "mean_token_accuracy": 0.18649509698152542,
      "num_tokens": 69830411.0,
      "step": 31870
    },
    {
      "entropy": 5.597790288925171,
      "epoch": 3.1634577213179833,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004017528679766337,
      "loss": 4.9495,
      "mean_token_accuracy": 0.19465593695640565,
      "num_tokens": 69841795.0,
      "step": 31875
    },
    {
      "entropy": 5.580732631683349,
      "epoch": 3.163953949980151,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004017235921578767,
      "loss": 4.9637,
      "mean_token_accuracy": 0.18540598154067994,
      "num_tokens": 69851633.0,
      "step": 31880
    },
    {
      "entropy": 5.587359476089477,
      "epoch": 3.1644501786423183,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040169431319664756,
      "loss": 4.8897,
      "mean_token_accuracy": 0.19515779614448547,
      "num_tokens": 69862398.0,
      "step": 31885
    },
    {
      "entropy": 5.609009408950806,
      "epoch": 3.1649464073044857,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004016650310936725,
      "loss": 4.9789,
      "mean_token_accuracy": 0.18953635841608046,
      "num_tokens": 69874397.0,
      "step": 31890
    },
    {
      "entropy": 5.621585941314697,
      "epoch": 3.165442635966653,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004016357458496776,
      "loss": 4.9402,
      "mean_token_accuracy": 0.19890543520450593,
      "num_tokens": 69883950.0,
      "step": 31895
    },
    {
      "entropy": 5.593611812591552,
      "epoch": 3.165938864628821,
      "grad_norm": 1.0625,
      "learning_rate": 0.000401606457465389,
      "loss": 4.9083,
      "mean_token_accuracy": 0.19156728386878968,
      "num_tokens": 69895586.0,
      "step": 31900
    },
    {
      "entropy": 5.619146490097046,
      "epoch": 3.1664350932909886,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004015771659415332,
      "loss": 4.9484,
      "mean_token_accuracy": 0.19717443585395814,
      "num_tokens": 69906093.0,
      "step": 31905
    },
    {
      "entropy": 5.592075824737549,
      "epoch": 3.166931321953156,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004015478712788364,
      "loss": 4.9254,
      "mean_token_accuracy": 0.18684421628713607,
      "num_tokens": 69916381.0,
      "step": 31910
    },
    {
      "entropy": 5.586249828338623,
      "epoch": 3.1674275506153236,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040151857347802506,
      "loss": 4.958,
      "mean_token_accuracy": 0.19631178081035613,
      "num_tokens": 69926059.0,
      "step": 31915
    },
    {
      "entropy": 5.604129648208618,
      "epoch": 3.167923779277491,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040148927253982577,
      "loss": 4.969,
      "mean_token_accuracy": 0.18585592955350877,
      "num_tokens": 69937677.0,
      "step": 31920
    },
    {
      "entropy": 5.602210187911988,
      "epoch": 3.1684200079396585,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040145996846496497,
      "loss": 4.9492,
      "mean_token_accuracy": 0.18969205021858215,
      "num_tokens": 69949399.0,
      "step": 31925
    },
    {
      "entropy": 5.609351730346679,
      "epoch": 3.168916236601826,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004014306612541695,
      "loss": 5.0133,
      "mean_token_accuracy": 0.1947203531861305,
      "num_tokens": 69960595.0,
      "step": 31930
    },
    {
      "entropy": 5.627058792114258,
      "epoch": 3.1694124652639935,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004014013509081661,
      "loss": 4.9789,
      "mean_token_accuracy": 0.19075428694486618,
      "num_tokens": 69971018.0,
      "step": 31935
    },
    {
      "entropy": 5.6553551197052006,
      "epoch": 3.1699086939261614,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040137203742768155,
      "loss": 4.9768,
      "mean_token_accuracy": 0.18997565358877183,
      "num_tokens": 69982507.0,
      "step": 31940
    },
    {
      "entropy": 5.624846458435059,
      "epoch": 3.170404922588329,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004013427208134427,
      "loss": 4.9483,
      "mean_token_accuracy": 0.19494226276874543,
      "num_tokens": 69994190.0,
      "step": 31945
    },
    {
      "entropy": 5.661511373519898,
      "epoch": 3.1709011512504963,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040131340106617673,
      "loss": 5.1331,
      "mean_token_accuracy": 0.18136721700429917,
      "num_tokens": 70005522.0,
      "step": 31950
    },
    {
      "entropy": 5.715204238891602,
      "epoch": 3.171397379912664,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004012840781866106,
      "loss": 5.0382,
      "mean_token_accuracy": 0.19402097761631013,
      "num_tokens": 70016706.0,
      "step": 31955
    },
    {
      "entropy": 5.641490173339844,
      "epoch": 3.1718936085748313,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040125475217547135,
      "loss": 4.9505,
      "mean_token_accuracy": 0.18913732320070267,
      "num_tokens": 70028577.0,
      "step": 31960
    },
    {
      "entropy": 5.596637439727783,
      "epoch": 3.1723898372369987,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040122542303348634,
      "loss": 4.9518,
      "mean_token_accuracy": 0.18900090903043748,
      "num_tokens": 70040258.0,
      "step": 31965
    },
    {
      "entropy": 5.637865829467773,
      "epoch": 3.172886065899166,
      "grad_norm": 1.0,
      "learning_rate": 0.0004011960907613828,
      "loss": 5.0756,
      "mean_token_accuracy": 0.18048044741153718,
      "num_tokens": 70052105.0,
      "step": 31970
    },
    {
      "entropy": 5.617564916610718,
      "epoch": 3.1733822945613337,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040116675535988825,
      "loss": 4.9469,
      "mean_token_accuracy": 0.19090326130390167,
      "num_tokens": 70062272.0,
      "step": 31975
    },
    {
      "entropy": 5.614713573455811,
      "epoch": 3.173878523223501,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040113741682972994,
      "loss": 4.9679,
      "mean_token_accuracy": 0.1907571703195572,
      "num_tokens": 70073614.0,
      "step": 31980
    },
    {
      "entropy": 5.614317464828491,
      "epoch": 3.174374751885669,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004011080751716356,
      "loss": 4.9827,
      "mean_token_accuracy": 0.1823333278298378,
      "num_tokens": 70085191.0,
      "step": 31985
    },
    {
      "entropy": 5.633924341201782,
      "epoch": 3.1748709805478366,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040107873038633255,
      "loss": 4.9686,
      "mean_token_accuracy": 0.18664226233959197,
      "num_tokens": 70097610.0,
      "step": 31990
    },
    {
      "entropy": 5.576182794570923,
      "epoch": 3.175367209210004,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004010493824745487,
      "loss": 4.921,
      "mean_token_accuracy": 0.19046338349580766,
      "num_tokens": 70109040.0,
      "step": 31995
    },
    {
      "entropy": 5.613501405715942,
      "epoch": 3.1758634378721715,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040102003143701194,
      "loss": 4.9928,
      "mean_token_accuracy": 0.18119859844446182,
      "num_tokens": 70120472.0,
      "step": 32000
    },
    {
      "entropy": 5.5960907459259035,
      "epoch": 3.176359666534339,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040099067727444984,
      "loss": 4.8967,
      "mean_token_accuracy": 0.1919969141483307,
      "num_tokens": 70130481.0,
      "step": 32005
    },
    {
      "entropy": 5.5943409442901615,
      "epoch": 3.1768558951965065,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004009613199875904,
      "loss": 4.9093,
      "mean_token_accuracy": 0.1912522241473198,
      "num_tokens": 70140783.0,
      "step": 32010
    },
    {
      "entropy": 5.638874816894531,
      "epoch": 3.177352123858674,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004009319595771617,
      "loss": 5.0156,
      "mean_token_accuracy": 0.1868119016289711,
      "num_tokens": 70151624.0,
      "step": 32015
    },
    {
      "entropy": 5.652014875411988,
      "epoch": 3.1778483525208414,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004009025960438918,
      "loss": 5.0025,
      "mean_token_accuracy": 0.18255004286766052,
      "num_tokens": 70162600.0,
      "step": 32020
    },
    {
      "entropy": 5.672184944152832,
      "epoch": 3.1783445811830093,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040087322938850867,
      "loss": 5.0226,
      "mean_token_accuracy": 0.1891914799809456,
      "num_tokens": 70175563.0,
      "step": 32025
    },
    {
      "entropy": 5.637819099426269,
      "epoch": 3.178840809845177,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004008438596117408,
      "loss": 4.9328,
      "mean_token_accuracy": 0.1852874457836151,
      "num_tokens": 70186358.0,
      "step": 32030
    },
    {
      "entropy": 5.65258207321167,
      "epoch": 3.1793370385073443,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040081448671431635,
      "loss": 4.9753,
      "mean_token_accuracy": 0.1864565908908844,
      "num_tokens": 70196872.0,
      "step": 32035
    },
    {
      "entropy": 5.59209041595459,
      "epoch": 3.1798332671695118,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040078511069696377,
      "loss": 4.9516,
      "mean_token_accuracy": 0.18363064378499985,
      "num_tokens": 70208136.0,
      "step": 32040
    },
    {
      "entropy": 5.69734525680542,
      "epoch": 3.1803294958316792,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040075573156041146,
      "loss": 5.1988,
      "mean_token_accuracy": 0.17146552503108978,
      "num_tokens": 70219716.0,
      "step": 32045
    },
    {
      "entropy": 5.650466537475586,
      "epoch": 3.1808257244938467,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040072634930538796,
      "loss": 5.0204,
      "mean_token_accuracy": 0.18252875953912734,
      "num_tokens": 70230502.0,
      "step": 32050
    },
    {
      "entropy": 5.668938302993775,
      "epoch": 3.181321953156014,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040069696393262196,
      "loss": 5.0119,
      "mean_token_accuracy": 0.18443358987569808,
      "num_tokens": 70241820.0,
      "step": 32055
    },
    {
      "entropy": 5.62600622177124,
      "epoch": 3.1818181818181817,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004006675754428421,
      "loss": 4.9365,
      "mean_token_accuracy": 0.19097733944654466,
      "num_tokens": 70252243.0,
      "step": 32060
    },
    {
      "entropy": 5.5770610809326175,
      "epoch": 3.182314410480349,
      "grad_norm": 1.125,
      "learning_rate": 0.00040063818383677716,
      "loss": 4.8764,
      "mean_token_accuracy": 0.18952387869358062,
      "num_tokens": 70261835.0,
      "step": 32065
    },
    {
      "entropy": 5.660823583602905,
      "epoch": 3.182810639142517,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000400608789115156,
      "loss": 5.0358,
      "mean_token_accuracy": 0.18494364023208618,
      "num_tokens": 70273210.0,
      "step": 32070
    },
    {
      "entropy": 5.641028690338135,
      "epoch": 3.1833068678046845,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040057939127870754,
      "loss": 4.9953,
      "mean_token_accuracy": 0.1854668453335762,
      "num_tokens": 70283904.0,
      "step": 32075
    },
    {
      "entropy": 5.674900960922241,
      "epoch": 3.183803096466852,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004005499903281607,
      "loss": 4.9772,
      "mean_token_accuracy": 0.19153794646263123,
      "num_tokens": 70295199.0,
      "step": 32080
    },
    {
      "entropy": 5.56585168838501,
      "epoch": 3.1842993251290195,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040052058626424464,
      "loss": 4.8947,
      "mean_token_accuracy": 0.19071387499570847,
      "num_tokens": 70305418.0,
      "step": 32085
    },
    {
      "entropy": 5.6188576221466064,
      "epoch": 3.184795553791187,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004004911790876886,
      "loss": 5.0598,
      "mean_token_accuracy": 0.18545716553926467,
      "num_tokens": 70316131.0,
      "step": 32090
    },
    {
      "entropy": 5.60106782913208,
      "epoch": 3.1852917824533544,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040046176879922156,
      "loss": 4.9014,
      "mean_token_accuracy": 0.18956612795591354,
      "num_tokens": 70326804.0,
      "step": 32095
    },
    {
      "entropy": 5.730264616012573,
      "epoch": 3.185788011115522,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040043235539957315,
      "loss": 5.1069,
      "mean_token_accuracy": 0.17729222327470778,
      "num_tokens": 70338125.0,
      "step": 32100
    },
    {
      "entropy": 5.651753520965576,
      "epoch": 3.1862842397776894,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004004029388894726,
      "loss": 4.9676,
      "mean_token_accuracy": 0.18736709356307985,
      "num_tokens": 70348614.0,
      "step": 32105
    },
    {
      "entropy": 5.60921893119812,
      "epoch": 3.1867804684398573,
      "grad_norm": 1.125,
      "learning_rate": 0.0004003735192696494,
      "loss": 5.0272,
      "mean_token_accuracy": 0.1834728166460991,
      "num_tokens": 70358735.0,
      "step": 32110
    },
    {
      "entropy": 5.53709044456482,
      "epoch": 3.1872766971020248,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000400344096540833,
      "loss": 4.8916,
      "mean_token_accuracy": 0.1954600229859352,
      "num_tokens": 70369448.0,
      "step": 32115
    },
    {
      "entropy": 5.573430061340332,
      "epoch": 3.1877729257641922,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040031467070375315,
      "loss": 4.9485,
      "mean_token_accuracy": 0.19237327575683594,
      "num_tokens": 70379462.0,
      "step": 32120
    },
    {
      "entropy": 5.6164586544036865,
      "epoch": 3.1882691544263597,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004002852417591395,
      "loss": 4.9896,
      "mean_token_accuracy": 0.18352630585432053,
      "num_tokens": 70389283.0,
      "step": 32125
    },
    {
      "entropy": 5.606248474121093,
      "epoch": 3.188765383088527,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004002558097077218,
      "loss": 5.0143,
      "mean_token_accuracy": 0.18903428614139556,
      "num_tokens": 70400058.0,
      "step": 32130
    },
    {
      "entropy": 5.562234497070312,
      "epoch": 3.1892616117506947,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004002263745502299,
      "loss": 4.9089,
      "mean_token_accuracy": 0.1922511115670204,
      "num_tokens": 70411614.0,
      "step": 32135
    },
    {
      "entropy": 5.609171628952026,
      "epoch": 3.189757840412862,
      "grad_norm": 1.125,
      "learning_rate": 0.00040019693628739377,
      "loss": 4.9642,
      "mean_token_accuracy": 0.18372986018657683,
      "num_tokens": 70422404.0,
      "step": 32140
    },
    {
      "entropy": 5.695654344558716,
      "epoch": 3.1902540690750296,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004001674949199434,
      "loss": 5.0373,
      "mean_token_accuracy": 0.184920471906662,
      "num_tokens": 70432219.0,
      "step": 32145
    },
    {
      "entropy": 5.619348621368408,
      "epoch": 3.190750297737197,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004001380504486089,
      "loss": 5.0045,
      "mean_token_accuracy": 0.1893002673983574,
      "num_tokens": 70443129.0,
      "step": 32150
    },
    {
      "entropy": 5.608421087265015,
      "epoch": 3.191246526399365,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004001086028741204,
      "loss": 5.0082,
      "mean_token_accuracy": 0.19213293343782425,
      "num_tokens": 70453225.0,
      "step": 32155
    },
    {
      "entropy": 5.6015526294708256,
      "epoch": 3.1917427550615325,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004000791521972081,
      "loss": 4.9789,
      "mean_token_accuracy": 0.18547270447015762,
      "num_tokens": 70464909.0,
      "step": 32160
    },
    {
      "entropy": 5.638247776031494,
      "epoch": 3.1922389837237,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040004969841860233,
      "loss": 4.9939,
      "mean_token_accuracy": 0.18866123259067535,
      "num_tokens": 70475211.0,
      "step": 32165
    },
    {
      "entropy": 5.712134742736817,
      "epoch": 3.1927352123858674,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040002024153903347,
      "loss": 4.9921,
      "mean_token_accuracy": 0.19054341018199922,
      "num_tokens": 70485603.0,
      "step": 32170
    },
    {
      "entropy": 5.590348100662231,
      "epoch": 3.193231441048035,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003999907815592321,
      "loss": 4.9169,
      "mean_token_accuracy": 0.1926991432905197,
      "num_tokens": 70497163.0,
      "step": 32175
    },
    {
      "entropy": 5.607549667358398,
      "epoch": 3.1937276697102024,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039996131847992863,
      "loss": 4.9621,
      "mean_token_accuracy": 0.19287292957305907,
      "num_tokens": 70509323.0,
      "step": 32180
    },
    {
      "entropy": 5.6592480659484865,
      "epoch": 3.19422389837237,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003999318523018537,
      "loss": 5.0779,
      "mean_token_accuracy": 0.18152988106012344,
      "num_tokens": 70520767.0,
      "step": 32185
    },
    {
      "entropy": 5.597926950454712,
      "epoch": 3.1947201270345373,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003999023830257382,
      "loss": 4.9321,
      "mean_token_accuracy": 0.18799193501472472,
      "num_tokens": 70531671.0,
      "step": 32190
    },
    {
      "entropy": 5.608534717559815,
      "epoch": 3.1952163556967053,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003998729106523126,
      "loss": 4.9642,
      "mean_token_accuracy": 0.18877976536750793,
      "num_tokens": 70542309.0,
      "step": 32195
    },
    {
      "entropy": 5.681344795227051,
      "epoch": 3.1957125843588727,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003998434351823078,
      "loss": 5.0431,
      "mean_token_accuracy": 0.18252351880073547,
      "num_tokens": 70553003.0,
      "step": 32200
    },
    {
      "entropy": 5.58217978477478,
      "epoch": 3.19620881302104,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000399813956616455,
      "loss": 4.9783,
      "mean_token_accuracy": 0.18537816405296326,
      "num_tokens": 70564161.0,
      "step": 32205
    },
    {
      "entropy": 5.639946985244751,
      "epoch": 3.1967050416832077,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039978447495548497,
      "loss": 4.9397,
      "mean_token_accuracy": 0.19479253888130188,
      "num_tokens": 70575891.0,
      "step": 32210
    },
    {
      "entropy": 5.66501030921936,
      "epoch": 3.197201270345375,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039975499020012886,
      "loss": 5.0947,
      "mean_token_accuracy": 0.1836167350411415,
      "num_tokens": 70586691.0,
      "step": 32215
    },
    {
      "entropy": 5.610972261428833,
      "epoch": 3.1976974990075426,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039972550235111776,
      "loss": 4.9214,
      "mean_token_accuracy": 0.18857578486204146,
      "num_tokens": 70597423.0,
      "step": 32220
    },
    {
      "entropy": 5.641131067276001,
      "epoch": 3.19819372766971,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000399696011409183,
      "loss": 4.9784,
      "mean_token_accuracy": 0.1912684455513954,
      "num_tokens": 70609828.0,
      "step": 32225
    },
    {
      "entropy": 5.53939642906189,
      "epoch": 3.1986899563318776,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039966651737505585,
      "loss": 4.9099,
      "mean_token_accuracy": 0.19129652827978133,
      "num_tokens": 70620480.0,
      "step": 32230
    },
    {
      "entropy": 5.6101501941680905,
      "epoch": 3.199186184994045,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003996370202494677,
      "loss": 4.9812,
      "mean_token_accuracy": 0.18981295973062515,
      "num_tokens": 70630045.0,
      "step": 32235
    },
    {
      "entropy": 5.611341857910157,
      "epoch": 3.199682413656213,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039960752003315,
      "loss": 4.8942,
      "mean_token_accuracy": 0.19801639765501022,
      "num_tokens": 70640142.0,
      "step": 32240
    },
    {
      "entropy": 5.582249450683594,
      "epoch": 3.2001786423183805,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003995780167268343,
      "loss": 4.9599,
      "mean_token_accuracy": 0.18824424147605895,
      "num_tokens": 70652062.0,
      "step": 32245
    },
    {
      "entropy": 5.584763145446777,
      "epoch": 3.200674870980548,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003995485103312523,
      "loss": 4.9839,
      "mean_token_accuracy": 0.18563077747821807,
      "num_tokens": 70666187.0,
      "step": 32250
    },
    {
      "entropy": 5.6821364402771,
      "epoch": 3.2011710996427154,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039951900084713555,
      "loss": 5.0483,
      "mean_token_accuracy": 0.1873306080698967,
      "num_tokens": 70677432.0,
      "step": 32255
    },
    {
      "entropy": 5.63507571220398,
      "epoch": 3.201667328304883,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003994894882752159,
      "loss": 4.9454,
      "mean_token_accuracy": 0.1883583977818489,
      "num_tokens": 70686963.0,
      "step": 32260
    },
    {
      "entropy": 5.694299030303955,
      "epoch": 3.2021635569670504,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039945997261622526,
      "loss": 5.1253,
      "mean_token_accuracy": 0.17825454026460646,
      "num_tokens": 70698345.0,
      "step": 32265
    },
    {
      "entropy": 5.615434837341309,
      "epoch": 3.202659785629218,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003994304538708954,
      "loss": 4.9727,
      "mean_token_accuracy": 0.18645173609256743,
      "num_tokens": 70709695.0,
      "step": 32270
    },
    {
      "entropy": 5.591451406478882,
      "epoch": 3.2031560142913853,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003994009320399584,
      "loss": 4.9128,
      "mean_token_accuracy": 0.18863255977630616,
      "num_tokens": 70720843.0,
      "step": 32275
    },
    {
      "entropy": 5.5955455780029295,
      "epoch": 3.203652242953553,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039937140712414644,
      "loss": 5.0098,
      "mean_token_accuracy": 0.18719736784696578,
      "num_tokens": 70731886.0,
      "step": 32280
    },
    {
      "entropy": 5.6227654933929445,
      "epoch": 3.2041484716157207,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003993418791241915,
      "loss": 4.9878,
      "mean_token_accuracy": 0.190148463845253,
      "num_tokens": 70741838.0,
      "step": 32285
    },
    {
      "entropy": 5.608957052230835,
      "epoch": 3.204644700277888,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003993123480408259,
      "loss": 4.8997,
      "mean_token_accuracy": 0.19919877648353576,
      "num_tokens": 70751454.0,
      "step": 32290
    },
    {
      "entropy": 5.643386316299439,
      "epoch": 3.2051409289400556,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039928281387478207,
      "loss": 4.9977,
      "mean_token_accuracy": 0.1835695058107376,
      "num_tokens": 70762294.0,
      "step": 32295
    },
    {
      "entropy": 5.599501037597657,
      "epoch": 3.205637157602223,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003992532766267921,
      "loss": 4.9489,
      "mean_token_accuracy": 0.195822174847126,
      "num_tokens": 70773186.0,
      "step": 32300
    },
    {
      "entropy": 5.63258638381958,
      "epoch": 3.2061333862643906,
      "grad_norm": 1.125,
      "learning_rate": 0.0003992237362975887,
      "loss": 4.9677,
      "mean_token_accuracy": 0.18964232355356217,
      "num_tokens": 70783984.0,
      "step": 32305
    },
    {
      "entropy": 5.587663555145264,
      "epoch": 3.206629614926558,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039919419288790423,
      "loss": 4.9442,
      "mean_token_accuracy": 0.19034005999565123,
      "num_tokens": 70795779.0,
      "step": 32310
    },
    {
      "entropy": 5.61511926651001,
      "epoch": 3.2071258435887255,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039916464639847145,
      "loss": 4.9323,
      "mean_token_accuracy": 0.18781168162822723,
      "num_tokens": 70807247.0,
      "step": 32315
    },
    {
      "entropy": 5.656467723846435,
      "epoch": 3.207622072250893,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003991350968300229,
      "loss": 4.9932,
      "mean_token_accuracy": 0.18517060130834578,
      "num_tokens": 70818507.0,
      "step": 32320
    },
    {
      "entropy": 5.626264190673828,
      "epoch": 3.208118300913061,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003991055441832916,
      "loss": 4.9988,
      "mean_token_accuracy": 0.18560095727443696,
      "num_tokens": 70830055.0,
      "step": 32325
    },
    {
      "entropy": 5.690177297592163,
      "epoch": 3.2086145295752284,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039907598845901016,
      "loss": 5.0999,
      "mean_token_accuracy": 0.18281847834587098,
      "num_tokens": 70841052.0,
      "step": 32330
    },
    {
      "entropy": 5.696987581253052,
      "epoch": 3.209110758237396,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003990464296579114,
      "loss": 5.0201,
      "mean_token_accuracy": 0.18820379674434662,
      "num_tokens": 70852148.0,
      "step": 32335
    },
    {
      "entropy": 5.603126764297485,
      "epoch": 3.2096069868995634,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003990168677807287,
      "loss": 5.0079,
      "mean_token_accuracy": 0.1853021815419197,
      "num_tokens": 70863998.0,
      "step": 32340
    },
    {
      "entropy": 5.602583932876587,
      "epoch": 3.210103215561731,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003989873028281948,
      "loss": 5.0702,
      "mean_token_accuracy": 0.18047043234109877,
      "num_tokens": 70874013.0,
      "step": 32345
    },
    {
      "entropy": 5.604137754440307,
      "epoch": 3.2105994442238983,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000398957734801043,
      "loss": 4.9138,
      "mean_token_accuracy": 0.19575486034154893,
      "num_tokens": 70885600.0,
      "step": 32350
    },
    {
      "entropy": 5.555959320068359,
      "epoch": 3.211095672886066,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039892816370000657,
      "loss": 4.8769,
      "mean_token_accuracy": 0.20349669456481934,
      "num_tokens": 70896447.0,
      "step": 32355
    },
    {
      "entropy": 5.702511358261108,
      "epoch": 3.2115919015482333,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003988985895258186,
      "loss": 5.0411,
      "mean_token_accuracy": 0.18799518048763275,
      "num_tokens": 70907558.0,
      "step": 32360
    },
    {
      "entropy": 5.630418491363526,
      "epoch": 3.212088130210401,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039886901227921263,
      "loss": 4.9756,
      "mean_token_accuracy": 0.1853855162858963,
      "num_tokens": 70919687.0,
      "step": 32365
    },
    {
      "entropy": 5.6199102878570555,
      "epoch": 3.2125843588725687,
      "grad_norm": 1.125,
      "learning_rate": 0.0003988394319609221,
      "loss": 4.9498,
      "mean_token_accuracy": 0.18865047842264177,
      "num_tokens": 70930417.0,
      "step": 32370
    },
    {
      "entropy": 5.601237392425537,
      "epoch": 3.213080587534736,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003988098485716804,
      "loss": 4.9394,
      "mean_token_accuracy": 0.1922930970788002,
      "num_tokens": 70941721.0,
      "step": 32375
    },
    {
      "entropy": 5.609778213500976,
      "epoch": 3.2135768161969036,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039878026211222136,
      "loss": 4.9401,
      "mean_token_accuracy": 0.19523605108261108,
      "num_tokens": 70952616.0,
      "step": 32380
    },
    {
      "entropy": 5.595232629776001,
      "epoch": 3.214073044859071,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039875067258327853,
      "loss": 4.9454,
      "mean_token_accuracy": 0.1942280575633049,
      "num_tokens": 70963236.0,
      "step": 32385
    },
    {
      "entropy": 5.641554355621338,
      "epoch": 3.2145692735212386,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003987210799855856,
      "loss": 5.0203,
      "mean_token_accuracy": 0.18428304344415664,
      "num_tokens": 70973376.0,
      "step": 32390
    },
    {
      "entropy": 5.630133438110351,
      "epoch": 3.215065502183406,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039869148431987653,
      "loss": 5.0059,
      "mean_token_accuracy": 0.18432640582323073,
      "num_tokens": 70984937.0,
      "step": 32395
    },
    {
      "entropy": 5.720550441741944,
      "epoch": 3.2155617308455735,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003986618855868852,
      "loss": 5.0589,
      "mean_token_accuracy": 0.1876903772354126,
      "num_tokens": 70997023.0,
      "step": 32400
    },
    {
      "entropy": 5.639428520202637,
      "epoch": 3.216057959507741,
      "grad_norm": 1.125,
      "learning_rate": 0.0003986322837873455,
      "loss": 4.9201,
      "mean_token_accuracy": 0.19353375285863877,
      "num_tokens": 71008304.0,
      "step": 32405
    },
    {
      "entropy": 5.572101783752442,
      "epoch": 3.216554188169909,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039860267892199167,
      "loss": 4.9915,
      "mean_token_accuracy": 0.1819755420088768,
      "num_tokens": 71020160.0,
      "step": 32410
    },
    {
      "entropy": 5.635685014724731,
      "epoch": 3.2170504168320764,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039857307099155773,
      "loss": 5.0305,
      "mean_token_accuracy": 0.19031138420104982,
      "num_tokens": 71031701.0,
      "step": 32415
    },
    {
      "entropy": 5.633632802963257,
      "epoch": 3.217546645494244,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039854345999677777,
      "loss": 5.0373,
      "mean_token_accuracy": 0.19144735485315323,
      "num_tokens": 71041955.0,
      "step": 32420
    },
    {
      "entropy": 5.657895421981811,
      "epoch": 3.2180428741564113,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039851384593838637,
      "loss": 5.0822,
      "mean_token_accuracy": 0.18105100095272064,
      "num_tokens": 71052884.0,
      "step": 32425
    },
    {
      "entropy": 5.628983831405639,
      "epoch": 3.218539102818579,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039848422881711765,
      "loss": 4.9479,
      "mean_token_accuracy": 0.19346642196178437,
      "num_tokens": 71065091.0,
      "step": 32430
    },
    {
      "entropy": 5.659644412994385,
      "epoch": 3.2190353314807463,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003984546086337061,
      "loss": 5.0271,
      "mean_token_accuracy": 0.1876540392637253,
      "num_tokens": 71075823.0,
      "step": 32435
    },
    {
      "entropy": 5.647852420806885,
      "epoch": 3.2195315601429138,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039842498538888646,
      "loss": 4.9797,
      "mean_token_accuracy": 0.19238557368516923,
      "num_tokens": 71087408.0,
      "step": 32440
    },
    {
      "entropy": 5.6474120140075685,
      "epoch": 3.2200277888050812,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039839535908339297,
      "loss": 4.9709,
      "mean_token_accuracy": 0.18721475899219514,
      "num_tokens": 71097474.0,
      "step": 32445
    },
    {
      "entropy": 5.5780675411224365,
      "epoch": 3.2205240174672487,
      "grad_norm": 1.125,
      "learning_rate": 0.00039836572971796045,
      "loss": 4.9276,
      "mean_token_accuracy": 0.1952017918229103,
      "num_tokens": 71107064.0,
      "step": 32450
    },
    {
      "entropy": 5.5755524158477785,
      "epoch": 3.2210202461294166,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003983360972933239,
      "loss": 4.9879,
      "mean_token_accuracy": 0.18461889177560806,
      "num_tokens": 71118233.0,
      "step": 32455
    },
    {
      "entropy": 5.620985412597657,
      "epoch": 3.221516474791584,
      "grad_norm": 1.125,
      "learning_rate": 0.0003983064618102177,
      "loss": 4.9068,
      "mean_token_accuracy": 0.19475916922092437,
      "num_tokens": 71129922.0,
      "step": 32460
    },
    {
      "entropy": 5.634201335906982,
      "epoch": 3.2220127034537516,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003982768232693769,
      "loss": 5.0088,
      "mean_token_accuracy": 0.18490295559167863,
      "num_tokens": 71139073.0,
      "step": 32465
    },
    {
      "entropy": 5.608627223968506,
      "epoch": 3.222508932115919,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039824718167153666,
      "loss": 4.8965,
      "mean_token_accuracy": 0.18756671696901323,
      "num_tokens": 71152240.0,
      "step": 32470
    },
    {
      "entropy": 5.620100736618042,
      "epoch": 3.2230051607780865,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003982175370174319,
      "loss": 4.9696,
      "mean_token_accuracy": 0.18766519278287888,
      "num_tokens": 71163001.0,
      "step": 32475
    },
    {
      "entropy": 5.623642921447754,
      "epoch": 3.223501389440254,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003981878893077976,
      "loss": 4.9549,
      "mean_token_accuracy": 0.19235700964927674,
      "num_tokens": 71172811.0,
      "step": 32480
    },
    {
      "entropy": 5.600978231430053,
      "epoch": 3.2239976181024215,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039815823854336916,
      "loss": 4.8848,
      "mean_token_accuracy": 0.20045609027147293,
      "num_tokens": 71182276.0,
      "step": 32485
    },
    {
      "entropy": 5.580028486251831,
      "epoch": 3.224493846764589,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003981285847248818,
      "loss": 4.8768,
      "mean_token_accuracy": 0.20269994884729386,
      "num_tokens": 71193154.0,
      "step": 32490
    },
    {
      "entropy": 5.655219173431396,
      "epoch": 3.2249900754267564,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003980989278530709,
      "loss": 5.0453,
      "mean_token_accuracy": 0.18232607394456862,
      "num_tokens": 71204394.0,
      "step": 32495
    },
    {
      "entropy": 5.585089302062988,
      "epoch": 3.2254863040889243,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003980692679286718,
      "loss": 4.8669,
      "mean_token_accuracy": 0.1962199404835701,
      "num_tokens": 71215311.0,
      "step": 32500
    },
    {
      "entropy": 5.60144419670105,
      "epoch": 3.225982532751092,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039803960495241996,
      "loss": 4.9857,
      "mean_token_accuracy": 0.19393552839756012,
      "num_tokens": 71225800.0,
      "step": 32505
    },
    {
      "entropy": 5.649920034408569,
      "epoch": 3.2264787614132593,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003980099389250511,
      "loss": 4.9664,
      "mean_token_accuracy": 0.1887841299176216,
      "num_tokens": 71237846.0,
      "step": 32510
    },
    {
      "entropy": 5.639518404006958,
      "epoch": 3.2269749900754268,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039798026984730084,
      "loss": 4.9734,
      "mean_token_accuracy": 0.1901155471801758,
      "num_tokens": 71248221.0,
      "step": 32515
    },
    {
      "entropy": 5.621606159210205,
      "epoch": 3.2274712187375942,
      "grad_norm": 1.125,
      "learning_rate": 0.0003979505977199049,
      "loss": 4.9952,
      "mean_token_accuracy": 0.18635255396366118,
      "num_tokens": 71259950.0,
      "step": 32520
    },
    {
      "entropy": 5.6252388000488285,
      "epoch": 3.2279674473997617,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039792092254359903,
      "loss": 4.9903,
      "mean_token_accuracy": 0.1904289096593857,
      "num_tokens": 71270541.0,
      "step": 32525
    },
    {
      "entropy": 5.634328699111938,
      "epoch": 3.228463676061929,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039789124431911923,
      "loss": 4.9684,
      "mean_token_accuracy": 0.18554224222898483,
      "num_tokens": 71280545.0,
      "step": 32530
    },
    {
      "entropy": 5.687854385375976,
      "epoch": 3.2289599047240967,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039786156304720126,
      "loss": 5.0546,
      "mean_token_accuracy": 0.1845954477787018,
      "num_tokens": 71292368.0,
      "step": 32535
    },
    {
      "entropy": 5.68046464920044,
      "epoch": 3.2294561333862646,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003978318787285814,
      "loss": 4.9675,
      "mean_token_accuracy": 0.18802372366189957,
      "num_tokens": 71302570.0,
      "step": 32540
    },
    {
      "entropy": 5.646098566055298,
      "epoch": 3.229952362048432,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039780219136399556,
      "loss": 4.9861,
      "mean_token_accuracy": 0.19045716375112534,
      "num_tokens": 71312629.0,
      "step": 32545
    },
    {
      "entropy": 5.677570486068726,
      "epoch": 3.2304485907105995,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039777250095418,
      "loss": 5.0347,
      "mean_token_accuracy": 0.1830051064491272,
      "num_tokens": 71324313.0,
      "step": 32550
    },
    {
      "entropy": 5.679774951934815,
      "epoch": 3.230944819372767,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003977428074998709,
      "loss": 5.057,
      "mean_token_accuracy": 0.18521777987480165,
      "num_tokens": 71336136.0,
      "step": 32555
    },
    {
      "entropy": 5.6147801876068115,
      "epoch": 3.2314410480349345,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039771311100180476,
      "loss": 4.9344,
      "mean_token_accuracy": 0.1949429288506508,
      "num_tokens": 71347267.0,
      "step": 32560
    },
    {
      "entropy": 5.58451361656189,
      "epoch": 3.231937276697102,
      "grad_norm": 1.328125,
      "learning_rate": 0.00039768341146071784,
      "loss": 4.9399,
      "mean_token_accuracy": 0.19303765743970872,
      "num_tokens": 71357268.0,
      "step": 32565
    },
    {
      "entropy": 5.707868957519532,
      "epoch": 3.2324335053592694,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003976537088773467,
      "loss": 5.0705,
      "mean_token_accuracy": 0.17763734012842178,
      "num_tokens": 71368365.0,
      "step": 32570
    },
    {
      "entropy": 5.724432754516601,
      "epoch": 3.232929734021437,
      "grad_norm": 1.125,
      "learning_rate": 0.00039762400325242785,
      "loss": 5.057,
      "mean_token_accuracy": 0.1848060742020607,
      "num_tokens": 71379623.0,
      "step": 32575
    },
    {
      "entropy": 5.634814262390137,
      "epoch": 3.2334259626836044,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039759429458669805,
      "loss": 4.972,
      "mean_token_accuracy": 0.19539736658334733,
      "num_tokens": 71388522.0,
      "step": 32580
    },
    {
      "entropy": 5.600765419006348,
      "epoch": 3.2339221913457723,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039756458288089377,
      "loss": 4.9566,
      "mean_token_accuracy": 0.19018155932426453,
      "num_tokens": 71399127.0,
      "step": 32585
    },
    {
      "entropy": 5.7460216045379635,
      "epoch": 3.2344184200079398,
      "grad_norm": 1.25,
      "learning_rate": 0.000397534868135752,
      "loss": 5.0629,
      "mean_token_accuracy": 0.1857113540172577,
      "num_tokens": 71410780.0,
      "step": 32590
    },
    {
      "entropy": 5.683740854263306,
      "epoch": 3.2349146486701073,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039750515035200947,
      "loss": 5.0822,
      "mean_token_accuracy": 0.17661726325750352,
      "num_tokens": 71422887.0,
      "step": 32595
    },
    {
      "entropy": 5.623574733734131,
      "epoch": 3.2354108773322747,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039747542953040327,
      "loss": 4.9975,
      "mean_token_accuracy": 0.1869657665491104,
      "num_tokens": 71432525.0,
      "step": 32600
    },
    {
      "entropy": 5.6240317821502686,
      "epoch": 3.235907105994442,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003974457056716703,
      "loss": 5.0093,
      "mean_token_accuracy": 0.1887815475463867,
      "num_tokens": 71444802.0,
      "step": 32605
    },
    {
      "entropy": 5.582541227340698,
      "epoch": 3.2364033346566097,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003974159787765477,
      "loss": 4.9214,
      "mean_token_accuracy": 0.1893117681145668,
      "num_tokens": 71455978.0,
      "step": 32610
    },
    {
      "entropy": 5.560724592208862,
      "epoch": 3.236899563318777,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003973862488457726,
      "loss": 4.856,
      "mean_token_accuracy": 0.1916472688317299,
      "num_tokens": 71466981.0,
      "step": 32615
    },
    {
      "entropy": 5.643742036819458,
      "epoch": 3.2373957919809446,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003973565158800822,
      "loss": 5.0174,
      "mean_token_accuracy": 0.18276199251413344,
      "num_tokens": 71478856.0,
      "step": 32620
    },
    {
      "entropy": 5.689678001403808,
      "epoch": 3.2378920206431125,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039732677988021394,
      "loss": 5.0406,
      "mean_token_accuracy": 0.18251554518938065,
      "num_tokens": 71489449.0,
      "step": 32625
    },
    {
      "entropy": 5.695577144622803,
      "epoch": 3.23838824930528,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003972970408469051,
      "loss": 5.0242,
      "mean_token_accuracy": 0.1843670815229416,
      "num_tokens": 71499687.0,
      "step": 32630
    },
    {
      "entropy": 5.698038244247437,
      "epoch": 3.2388844779674475,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003972672987808932,
      "loss": 4.9882,
      "mean_token_accuracy": 0.19059767425060273,
      "num_tokens": 71510087.0,
      "step": 32635
    },
    {
      "entropy": 5.6171205043792725,
      "epoch": 3.239380706629615,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039723755368291564,
      "loss": 4.9673,
      "mean_token_accuracy": 0.18808334618806838,
      "num_tokens": 71520504.0,
      "step": 32640
    },
    {
      "entropy": 5.652860927581787,
      "epoch": 3.2398769352917824,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003972078055537102,
      "loss": 5.0512,
      "mean_token_accuracy": 0.18145074546337128,
      "num_tokens": 71532342.0,
      "step": 32645
    },
    {
      "entropy": 5.707574653625488,
      "epoch": 3.24037316395395,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003971780543940146,
      "loss": 4.9933,
      "mean_token_accuracy": 0.18430403620004654,
      "num_tokens": 71544523.0,
      "step": 32650
    },
    {
      "entropy": 5.657123851776123,
      "epoch": 3.2408693926161174,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003971483002045665,
      "loss": 4.9931,
      "mean_token_accuracy": 0.19067687541246414,
      "num_tokens": 71555162.0,
      "step": 32655
    },
    {
      "entropy": 5.6308189868927006,
      "epoch": 3.241365621278285,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003971185429861037,
      "loss": 4.9633,
      "mean_token_accuracy": 0.19043891429901122,
      "num_tokens": 71566407.0,
      "step": 32660
    },
    {
      "entropy": 5.678513288497925,
      "epoch": 3.2418618499404523,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039708878273936424,
      "loss": 5.0154,
      "mean_token_accuracy": 0.19045694023370743,
      "num_tokens": 71577328.0,
      "step": 32665
    },
    {
      "entropy": 5.7049390316009525,
      "epoch": 3.2423580786026203,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003970590194650859,
      "loss": 5.0977,
      "mean_token_accuracy": 0.1816805124282837,
      "num_tokens": 71587716.0,
      "step": 32670
    },
    {
      "entropy": 5.620063352584839,
      "epoch": 3.2428543072647877,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003970292531640071,
      "loss": 4.9595,
      "mean_token_accuracy": 0.1884297326207161,
      "num_tokens": 71599575.0,
      "step": 32675
    },
    {
      "entropy": 5.663733339309692,
      "epoch": 3.243350535926955,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00039699948383686563,
      "loss": 5.0399,
      "mean_token_accuracy": 0.18383996039628983,
      "num_tokens": 71610925.0,
      "step": 32680
    },
    {
      "entropy": 5.65597767829895,
      "epoch": 3.2438467645891227,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003969697114843998,
      "loss": 4.9757,
      "mean_token_accuracy": 0.19054765105247498,
      "num_tokens": 71620250.0,
      "step": 32685
    },
    {
      "entropy": 5.669224071502685,
      "epoch": 3.24434299325129,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039693993610734807,
      "loss": 5.078,
      "mean_token_accuracy": 0.18225873857736588,
      "num_tokens": 71632633.0,
      "step": 32690
    },
    {
      "entropy": 5.665524673461914,
      "epoch": 3.2448392219134576,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039691015770644875,
      "loss": 5.0492,
      "mean_token_accuracy": 0.18278649300336838,
      "num_tokens": 71644031.0,
      "step": 32695
    },
    {
      "entropy": 5.6647789478302,
      "epoch": 3.245335450575625,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039688037628244,
      "loss": 5.0533,
      "mean_token_accuracy": 0.19336239397525787,
      "num_tokens": 71654785.0,
      "step": 32700
    },
    {
      "entropy": 5.637358236312866,
      "epoch": 3.2458316792377926,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039685059183606065,
      "loss": 4.9631,
      "mean_token_accuracy": 0.19313973784446717,
      "num_tokens": 71667014.0,
      "step": 32705
    },
    {
      "entropy": 5.686600303649902,
      "epoch": 3.2463279078999605,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039682080436804926,
      "loss": 5.0045,
      "mean_token_accuracy": 0.1962541550397873,
      "num_tokens": 71677157.0,
      "step": 32710
    },
    {
      "entropy": 5.7001378536224365,
      "epoch": 3.246824136562128,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003967910138791442,
      "loss": 5.0521,
      "mean_token_accuracy": 0.18085170090198516,
      "num_tokens": 71688542.0,
      "step": 32715
    },
    {
      "entropy": 5.591046762466431,
      "epoch": 3.2473203652242955,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039676122037008473,
      "loss": 4.9821,
      "mean_token_accuracy": 0.1860143467783928,
      "num_tokens": 71699364.0,
      "step": 32720
    },
    {
      "entropy": 5.661150550842285,
      "epoch": 3.247816593886463,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003967314238416092,
      "loss": 5.0185,
      "mean_token_accuracy": 0.18539552092552186,
      "num_tokens": 71710248.0,
      "step": 32725
    },
    {
      "entropy": 5.647542572021484,
      "epoch": 3.2483128225486304,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003967016242944567,
      "loss": 5.014,
      "mean_token_accuracy": 0.18224809169769288,
      "num_tokens": 71720616.0,
      "step": 32730
    },
    {
      "entropy": 5.7379731178283695,
      "epoch": 3.248809051210798,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003966718217293661,
      "loss": 5.0063,
      "mean_token_accuracy": 0.18916695564985275,
      "num_tokens": 71731536.0,
      "step": 32735
    },
    {
      "entropy": 5.723284578323364,
      "epoch": 3.2493052798729654,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039664201614707657,
      "loss": 5.0273,
      "mean_token_accuracy": 0.1890692576766014,
      "num_tokens": 71742459.0,
      "step": 32740
    },
    {
      "entropy": 5.5646727085113525,
      "epoch": 3.249801508535133,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003966122075483271,
      "loss": 4.9876,
      "mean_token_accuracy": 0.18424903303384782,
      "num_tokens": 71753611.0,
      "step": 32745
    },
    {
      "entropy": 5.634530401229858,
      "epoch": 3.2502977371973003,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039658239593385697,
      "loss": 4.9581,
      "mean_token_accuracy": 0.19442649334669113,
      "num_tokens": 71764924.0,
      "step": 32750
    },
    {
      "entropy": 5.644928312301635,
      "epoch": 3.2507939658594682,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039655258130440543,
      "loss": 5.0403,
      "mean_token_accuracy": 0.17666929662227632,
      "num_tokens": 71776078.0,
      "step": 32755
    },
    {
      "entropy": 5.657573509216308,
      "epoch": 3.2512901945216357,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003965227636607117,
      "loss": 4.9742,
      "mean_token_accuracy": 0.19067583829164506,
      "num_tokens": 71785988.0,
      "step": 32760
    },
    {
      "entropy": 5.599913024902344,
      "epoch": 3.251786423183803,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039649294300351547,
      "loss": 4.9708,
      "mean_token_accuracy": 0.189044351875782,
      "num_tokens": 71796382.0,
      "step": 32765
    },
    {
      "entropy": 5.609207725524902,
      "epoch": 3.2522826518459707,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003964631193335559,
      "loss": 4.9419,
      "mean_token_accuracy": 0.18478693813085556,
      "num_tokens": 71806533.0,
      "step": 32770
    },
    {
      "entropy": 5.639515686035156,
      "epoch": 3.252778880508138,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039643329265157275,
      "loss": 4.9777,
      "mean_token_accuracy": 0.19382440447807311,
      "num_tokens": 71816893.0,
      "step": 32775
    },
    {
      "entropy": 5.6923168182373045,
      "epoch": 3.2532751091703056,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039640346295830557,
      "loss": 4.9953,
      "mean_token_accuracy": 0.1916329562664032,
      "num_tokens": 71827366.0,
      "step": 32780
    },
    {
      "entropy": 5.7061302185058596,
      "epoch": 3.253771337832473,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003963736302544942,
      "loss": 5.0259,
      "mean_token_accuracy": 0.18261752873659134,
      "num_tokens": 71837896.0,
      "step": 32785
    },
    {
      "entropy": 5.626705169677734,
      "epoch": 3.2542675664946406,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039634379454087814,
      "loss": 5.0044,
      "mean_token_accuracy": 0.1821553647518158,
      "num_tokens": 71850583.0,
      "step": 32790
    },
    {
      "entropy": 5.6402898788452145,
      "epoch": 3.2547637951568085,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003963139558181977,
      "loss": 5.0317,
      "mean_token_accuracy": 0.18675476759672166,
      "num_tokens": 71860392.0,
      "step": 32795
    },
    {
      "entropy": 5.630014705657959,
      "epoch": 3.255260023818976,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003962841140871924,
      "loss": 4.9858,
      "mean_token_accuracy": 0.19161614775657654,
      "num_tokens": 71871323.0,
      "step": 32800
    },
    {
      "entropy": 5.5880162715911865,
      "epoch": 3.2557562524811434,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003962542693486024,
      "loss": 4.9357,
      "mean_token_accuracy": 0.1911772072315216,
      "num_tokens": 71881612.0,
      "step": 32805
    },
    {
      "entropy": 5.6669121265411375,
      "epoch": 3.256252481143311,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003962244216031679,
      "loss": 5.0412,
      "mean_token_accuracy": 0.1886057063937187,
      "num_tokens": 71893266.0,
      "step": 32810
    },
    {
      "entropy": 5.662734174728394,
      "epoch": 3.2567487098054784,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000396194570851629,
      "loss": 4.9654,
      "mean_token_accuracy": 0.1909381166100502,
      "num_tokens": 71904978.0,
      "step": 32815
    },
    {
      "entropy": 5.723815679550171,
      "epoch": 3.257244938467646,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039616471709472573,
      "loss": 5.0387,
      "mean_token_accuracy": 0.18430548012256623,
      "num_tokens": 71915097.0,
      "step": 32820
    },
    {
      "entropy": 5.690797185897827,
      "epoch": 3.2577411671298133,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003961348603331987,
      "loss": 4.9878,
      "mean_token_accuracy": 0.18123599737882615,
      "num_tokens": 71924672.0,
      "step": 32825
    },
    {
      "entropy": 5.528461599349976,
      "epoch": 3.258237395791981,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039610500056778817,
      "loss": 4.9018,
      "mean_token_accuracy": 0.18879947513341905,
      "num_tokens": 71936122.0,
      "step": 32830
    },
    {
      "entropy": 5.601458930969239,
      "epoch": 3.2587336244541483,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003960751377992344,
      "loss": 4.9115,
      "mean_token_accuracy": 0.19613440483808517,
      "num_tokens": 71946601.0,
      "step": 32835
    },
    {
      "entropy": 5.712706565856934,
      "epoch": 3.259229853116316,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039604527202827825,
      "loss": 5.0466,
      "mean_token_accuracy": 0.18613610416650772,
      "num_tokens": 71958892.0,
      "step": 32840
    },
    {
      "entropy": 5.6400634765625,
      "epoch": 3.2597260817784837,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039601540325566014,
      "loss": 4.9538,
      "mean_token_accuracy": 0.19036170542240144,
      "num_tokens": 71969183.0,
      "step": 32845
    },
    {
      "entropy": 5.644678115844727,
      "epoch": 3.260222310440651,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003959855314821208,
      "loss": 4.9684,
      "mean_token_accuracy": 0.18717559278011323,
      "num_tokens": 71981115.0,
      "step": 32850
    },
    {
      "entropy": 5.658401775360107,
      "epoch": 3.2607185391028186,
      "grad_norm": 1.171875,
      "learning_rate": 0.000395955656708401,
      "loss": 5.0326,
      "mean_token_accuracy": 0.19075410962104797,
      "num_tokens": 71991677.0,
      "step": 32855
    },
    {
      "entropy": 5.643086242675781,
      "epoch": 3.261214767764986,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039592577893524155,
      "loss": 5.0184,
      "mean_token_accuracy": 0.18931679725646972,
      "num_tokens": 72003995.0,
      "step": 32860
    },
    {
      "entropy": 5.670950508117675,
      "epoch": 3.2617109964271536,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039589589816338335,
      "loss": 4.9859,
      "mean_token_accuracy": 0.1906085878610611,
      "num_tokens": 72014052.0,
      "step": 32865
    },
    {
      "entropy": 5.63516902923584,
      "epoch": 3.262207225089321,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003958660143935674,
      "loss": 4.9378,
      "mean_token_accuracy": 0.1835433378815651,
      "num_tokens": 72024756.0,
      "step": 32870
    },
    {
      "entropy": 5.589356184005737,
      "epoch": 3.2627034537514885,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003958361276265347,
      "loss": 4.9448,
      "mean_token_accuracy": 0.194127593934536,
      "num_tokens": 72035883.0,
      "step": 32875
    },
    {
      "entropy": 5.621208715438843,
      "epoch": 3.2631996824136564,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039580623786302643,
      "loss": 4.9661,
      "mean_token_accuracy": 0.1883225679397583,
      "num_tokens": 72047296.0,
      "step": 32880
    },
    {
      "entropy": 5.6403257846832275,
      "epoch": 3.263695911075824,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039577634510378374,
      "loss": 4.9842,
      "mean_token_accuracy": 0.1782502129673958,
      "num_tokens": 72060163.0,
      "step": 32885
    },
    {
      "entropy": 5.696010160446167,
      "epoch": 3.2641921397379914,
      "grad_norm": 1.15625,
      "learning_rate": 0.000395746449349548,
      "loss": 5.0461,
      "mean_token_accuracy": 0.18793189525604248,
      "num_tokens": 72070384.0,
      "step": 32890
    },
    {
      "entropy": 5.659078884124756,
      "epoch": 3.264688368400159,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003957165506010605,
      "loss": 4.9624,
      "mean_token_accuracy": 0.18298074156045913,
      "num_tokens": 72080997.0,
      "step": 32895
    },
    {
      "entropy": 5.670000314712524,
      "epoch": 3.2651845970623263,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003956866488590626,
      "loss": 4.9927,
      "mean_token_accuracy": 0.196712489426136,
      "num_tokens": 72091161.0,
      "step": 32900
    },
    {
      "entropy": 5.612772655487061,
      "epoch": 3.265680825724494,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003956567441242959,
      "loss": 4.983,
      "mean_token_accuracy": 0.1876250460743904,
      "num_tokens": 72101807.0,
      "step": 32905
    },
    {
      "entropy": 5.661355257034302,
      "epoch": 3.2661770543866613,
      "grad_norm": 1.046875,
      "learning_rate": 0.000395626836397502,
      "loss": 5.0496,
      "mean_token_accuracy": 0.18354273289442063,
      "num_tokens": 72112800.0,
      "step": 32910
    },
    {
      "entropy": 5.626133441925049,
      "epoch": 3.2666732830488288,
      "grad_norm": 1.25,
      "learning_rate": 0.0003955969256794224,
      "loss": 4.9701,
      "mean_token_accuracy": 0.18571015298366547,
      "num_tokens": 72122391.0,
      "step": 32915
    },
    {
      "entropy": 5.610894060134887,
      "epoch": 3.2671695117109962,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039556701197079896,
      "loss": 5.0158,
      "mean_token_accuracy": 0.1865531548857689,
      "num_tokens": 72133145.0,
      "step": 32920
    },
    {
      "entropy": 5.730052709579468,
      "epoch": 3.2676657403731637,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039553709527237344,
      "loss": 5.0271,
      "mean_token_accuracy": 0.17935515493154525,
      "num_tokens": 72146126.0,
      "step": 32925
    },
    {
      "entropy": 5.657691764831543,
      "epoch": 3.2681619690353316,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003955071755848876,
      "loss": 4.9502,
      "mean_token_accuracy": 0.19036077260971068,
      "num_tokens": 72157325.0,
      "step": 32930
    },
    {
      "entropy": 5.580930185317993,
      "epoch": 3.268658197697499,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039547725290908347,
      "loss": 4.8566,
      "mean_token_accuracy": 0.19339626282453537,
      "num_tokens": 72168601.0,
      "step": 32935
    },
    {
      "entropy": 5.6142653465271,
      "epoch": 3.2691544263596666,
      "grad_norm": 1.984375,
      "learning_rate": 0.00039544732724570313,
      "loss": 4.9318,
      "mean_token_accuracy": 0.19749826192855835,
      "num_tokens": 72179845.0,
      "step": 32940
    },
    {
      "entropy": 5.576101779937744,
      "epoch": 3.269650655021834,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003954173985954887,
      "loss": 4.8922,
      "mean_token_accuracy": 0.19302307069301605,
      "num_tokens": 72191135.0,
      "step": 32945
    },
    {
      "entropy": 5.6370189666748045,
      "epoch": 3.2701468836840015,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003953874669591822,
      "loss": 4.9845,
      "mean_token_accuracy": 0.19125525355339051,
      "num_tokens": 72201551.0,
      "step": 32950
    },
    {
      "entropy": 5.670946407318115,
      "epoch": 3.270643112346169,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039535753233752583,
      "loss": 4.9716,
      "mean_token_accuracy": 0.18889778703451157,
      "num_tokens": 72211983.0,
      "step": 32955
    },
    {
      "entropy": 5.674146509170532,
      "epoch": 3.2711393410083365,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003953275947312621,
      "loss": 5.0091,
      "mean_token_accuracy": 0.177808541059494,
      "num_tokens": 72223374.0,
      "step": 32960
    },
    {
      "entropy": 5.553022241592407,
      "epoch": 3.2716355696705044,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003952976541411332,
      "loss": 4.9632,
      "mean_token_accuracy": 0.1942850962281227,
      "num_tokens": 72233878.0,
      "step": 32965
    },
    {
      "entropy": 5.65317440032959,
      "epoch": 3.272131798332672,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039526771056788186,
      "loss": 5.0467,
      "mean_token_accuracy": 0.18624120056629181,
      "num_tokens": 72246157.0,
      "step": 32970
    },
    {
      "entropy": 5.6377075672149655,
      "epoch": 3.2726280269948393,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039523776401225043,
      "loss": 4.9683,
      "mean_token_accuracy": 0.19579024314880372,
      "num_tokens": 72256805.0,
      "step": 32975
    },
    {
      "entropy": 5.665919303894043,
      "epoch": 3.273124255657007,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039520781447498145,
      "loss": 4.9914,
      "mean_token_accuracy": 0.18285243958234787,
      "num_tokens": 72267026.0,
      "step": 32980
    },
    {
      "entropy": 5.663517808914184,
      "epoch": 3.2736204843191743,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003951778619568177,
      "loss": 4.9628,
      "mean_token_accuracy": 0.19414419829845428,
      "num_tokens": 72277989.0,
      "step": 32985
    },
    {
      "entropy": 5.603063726425171,
      "epoch": 3.2741167129813418,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000395147906458502,
      "loss": 4.9244,
      "mean_token_accuracy": 0.18588101416826247,
      "num_tokens": 72287989.0,
      "step": 32990
    },
    {
      "entropy": 5.584480047225952,
      "epoch": 3.2746129416435092,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039511794798077705,
      "loss": 4.9557,
      "mean_token_accuracy": 0.186412113904953,
      "num_tokens": 72298718.0,
      "step": 32995
    },
    {
      "entropy": 5.60658860206604,
      "epoch": 3.2751091703056767,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003950879865243858,
      "loss": 4.9348,
      "mean_token_accuracy": 0.19279983192682265,
      "num_tokens": 72309217.0,
      "step": 33000
    },
    {
      "epoch": 3.2751091703056767,
      "eval_entropy": 5.429778606598778,
      "eval_loss": 5.124092102050781,
      "eval_mean_token_accuracy": 0.18671102006759263,
      "eval_num_tokens": 72309217.0,
      "eval_runtime": 20.2957,
      "eval_samples_per_second": 1658.92,
      "eval_steps_per_second": 207.383,
      "step": 33000
    },
    {
      "entropy": 5.6856543064117435,
      "epoch": 3.275605398967844,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039505802209007136,
      "loss": 4.9955,
      "mean_token_accuracy": 0.18817896246910096,
      "num_tokens": 72319809.0,
      "step": 33005
    },
    {
      "entropy": 5.675640296936035,
      "epoch": 3.2761016276300117,
      "grad_norm": 1.125,
      "learning_rate": 0.0003950280546785766,
      "loss": 5.0218,
      "mean_token_accuracy": 0.18842778205871583,
      "num_tokens": 72330616.0,
      "step": 33010
    },
    {
      "entropy": 5.660504865646362,
      "epoch": 3.2765978562921796,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003949980842906447,
      "loss": 5.0092,
      "mean_token_accuracy": 0.18840094655752182,
      "num_tokens": 72339749.0,
      "step": 33015
    },
    {
      "entropy": 5.619861316680908,
      "epoch": 3.277094084954347,
      "grad_norm": 1.0,
      "learning_rate": 0.000394968110927019,
      "loss": 4.966,
      "mean_token_accuracy": 0.18994018584489822,
      "num_tokens": 72350653.0,
      "step": 33020
    },
    {
      "entropy": 5.652794933319091,
      "epoch": 3.2775903136165145,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039493813458844256,
      "loss": 4.9716,
      "mean_token_accuracy": 0.19010526835918426,
      "num_tokens": 72360261.0,
      "step": 33025
    },
    {
      "entropy": 5.581762075424194,
      "epoch": 3.278086542278682,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039490815527565883,
      "loss": 4.9748,
      "mean_token_accuracy": 0.1852918431162834,
      "num_tokens": 72372180.0,
      "step": 33030
    },
    {
      "entropy": 5.606782102584839,
      "epoch": 3.2785827709408495,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039487817298941124,
      "loss": 4.9688,
      "mean_token_accuracy": 0.18238587826490402,
      "num_tokens": 72383324.0,
      "step": 33035
    },
    {
      "entropy": 5.6044878482818605,
      "epoch": 3.279078999603017,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003948481877304433,
      "loss": 4.9735,
      "mean_token_accuracy": 0.18368328511714935,
      "num_tokens": 72395014.0,
      "step": 33040
    },
    {
      "entropy": 5.708756637573242,
      "epoch": 3.2795752282651844,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003948181994994985,
      "loss": 5.0232,
      "mean_token_accuracy": 0.18755945414304734,
      "num_tokens": 72406806.0,
      "step": 33045
    },
    {
      "entropy": 5.671027803421021,
      "epoch": 3.2800714569273524,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003947882082973206,
      "loss": 4.9602,
      "mean_token_accuracy": 0.18698867559432983,
      "num_tokens": 72417002.0,
      "step": 33050
    },
    {
      "entropy": 5.632553911209106,
      "epoch": 3.28056768558952,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039475821412465317,
      "loss": 5.0137,
      "mean_token_accuracy": 0.18522715121507644,
      "num_tokens": 72427107.0,
      "step": 33055
    },
    {
      "entropy": 5.560556316375733,
      "epoch": 3.2810639142516873,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039472821698224023,
      "loss": 4.8942,
      "mean_token_accuracy": 0.19978390336036683,
      "num_tokens": 72438737.0,
      "step": 33060
    },
    {
      "entropy": 5.572113418579102,
      "epoch": 3.281560142913855,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003946982168708254,
      "loss": 4.914,
      "mean_token_accuracy": 0.1885348379611969,
      "num_tokens": 72449070.0,
      "step": 33065
    },
    {
      "entropy": 5.63898196220398,
      "epoch": 3.2820563715760223,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003946682137911527,
      "loss": 5.0162,
      "mean_token_accuracy": 0.18758868873119355,
      "num_tokens": 72459437.0,
      "step": 33070
    },
    {
      "entropy": 5.691158056259155,
      "epoch": 3.2825526002381897,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039463820774396623,
      "loss": 4.9906,
      "mean_token_accuracy": 0.18678425550460814,
      "num_tokens": 72470804.0,
      "step": 33075
    },
    {
      "entropy": 5.687959241867065,
      "epoch": 3.283048828900357,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039460819873000995,
      "loss": 5.003,
      "mean_token_accuracy": 0.18095744103193284,
      "num_tokens": 72482400.0,
      "step": 33080
    },
    {
      "entropy": 5.590799999237061,
      "epoch": 3.2835450575625247,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039457818675002807,
      "loss": 4.8682,
      "mean_token_accuracy": 0.1991534948348999,
      "num_tokens": 72492697.0,
      "step": 33085
    },
    {
      "entropy": 5.584022760391235,
      "epoch": 3.284041286224692,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003945481718047648,
      "loss": 4.9673,
      "mean_token_accuracy": 0.18807606995105744,
      "num_tokens": 72503913.0,
      "step": 33090
    },
    {
      "entropy": 5.669728708267212,
      "epoch": 3.2845375148868596,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003945181538949646,
      "loss": 4.9821,
      "mean_token_accuracy": 0.19062738567590715,
      "num_tokens": 72515018.0,
      "step": 33095
    },
    {
      "entropy": 5.629900789260864,
      "epoch": 3.2850337435490276,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003944881330213716,
      "loss": 5.0167,
      "mean_token_accuracy": 0.1811177134513855,
      "num_tokens": 72525256.0,
      "step": 33100
    },
    {
      "entropy": 5.594216775894165,
      "epoch": 3.285529972211195,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003944581091847304,
      "loss": 4.9319,
      "mean_token_accuracy": 0.19040315151214598,
      "num_tokens": 72535839.0,
      "step": 33105
    },
    {
      "entropy": 5.666877555847168,
      "epoch": 3.2860262008733625,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039442808238578553,
      "loss": 5.0517,
      "mean_token_accuracy": 0.1931983157992363,
      "num_tokens": 72546808.0,
      "step": 33110
    },
    {
      "entropy": 5.687342596054077,
      "epoch": 3.28652242953553,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039439805262528153,
      "loss": 5.0275,
      "mean_token_accuracy": 0.18878673166036605,
      "num_tokens": 72559268.0,
      "step": 33115
    },
    {
      "entropy": 5.696428298950195,
      "epoch": 3.2870186581976975,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003943680199039631,
      "loss": 5.0026,
      "mean_token_accuracy": 0.18243976384401323,
      "num_tokens": 72570466.0,
      "step": 33120
    },
    {
      "entropy": 5.665432596206665,
      "epoch": 3.287514886859865,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039433798422257505,
      "loss": 4.9528,
      "mean_token_accuracy": 0.18742617666721345,
      "num_tokens": 72581373.0,
      "step": 33125
    },
    {
      "entropy": 5.717948150634766,
      "epoch": 3.2880111155220324,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003943079455818621,
      "loss": 5.0587,
      "mean_token_accuracy": 0.18388065248727797,
      "num_tokens": 72592661.0,
      "step": 33130
    },
    {
      "entropy": 5.568375921249389,
      "epoch": 3.2885073441842003,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039427790398256925,
      "loss": 4.9109,
      "mean_token_accuracy": 0.19686773866415025,
      "num_tokens": 72603387.0,
      "step": 33135
    },
    {
      "entropy": 5.612921619415284,
      "epoch": 3.289003572846368,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039424785942544126,
      "loss": 4.9662,
      "mean_token_accuracy": 0.1860117271542549,
      "num_tokens": 72614006.0,
      "step": 33140
    },
    {
      "entropy": 5.581307601928711,
      "epoch": 3.2894998015085353,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003942178119112234,
      "loss": 4.9344,
      "mean_token_accuracy": 0.1936798095703125,
      "num_tokens": 72625447.0,
      "step": 33145
    },
    {
      "entropy": 5.60381326675415,
      "epoch": 3.2899960301707027,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003941877614406607,
      "loss": 4.9637,
      "mean_token_accuracy": 0.18465970009565352,
      "num_tokens": 72636965.0,
      "step": 33150
    },
    {
      "entropy": 5.638578748703003,
      "epoch": 3.29049225883287,
      "grad_norm": 1.125,
      "learning_rate": 0.0003941577080144984,
      "loss": 4.9438,
      "mean_token_accuracy": 0.1917504608631134,
      "num_tokens": 72648161.0,
      "step": 33155
    },
    {
      "entropy": 5.615290784835816,
      "epoch": 3.2909884874950377,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003941276516334816,
      "loss": 4.919,
      "mean_token_accuracy": 0.19320254176855087,
      "num_tokens": 72658253.0,
      "step": 33160
    },
    {
      "entropy": 5.571826457977295,
      "epoch": 3.291484716157205,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039409759229835586,
      "loss": 4.9315,
      "mean_token_accuracy": 0.19407905340194703,
      "num_tokens": 72669583.0,
      "step": 33165
    },
    {
      "entropy": 5.551540660858154,
      "epoch": 3.2919809448193726,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003940675300098663,
      "loss": 4.8895,
      "mean_token_accuracy": 0.19193111062049867,
      "num_tokens": 72680367.0,
      "step": 33170
    },
    {
      "entropy": 5.599527597427368,
      "epoch": 3.29247717348154,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003940374647687586,
      "loss": 4.9948,
      "mean_token_accuracy": 0.18407041430473328,
      "num_tokens": 72692437.0,
      "step": 33175
    },
    {
      "entropy": 5.594222927093506,
      "epoch": 3.2929734021437076,
      "grad_norm": 0.921875,
      "learning_rate": 0.0003940073965757784,
      "loss": 5.0185,
      "mean_token_accuracy": 0.1867866709828377,
      "num_tokens": 72704286.0,
      "step": 33180
    },
    {
      "entropy": 5.673228693008423,
      "epoch": 3.2934696308058755,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003939773254316711,
      "loss": 5.0449,
      "mean_token_accuracy": 0.18328275680541992,
      "num_tokens": 72714767.0,
      "step": 33185
    },
    {
      "entropy": 5.698676490783692,
      "epoch": 3.293965859468043,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003939472513371826,
      "loss": 5.0413,
      "mean_token_accuracy": 0.18855324238538743,
      "num_tokens": 72725110.0,
      "step": 33190
    },
    {
      "entropy": 5.631068706512451,
      "epoch": 3.2944620881302105,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003939171742930584,
      "loss": 4.976,
      "mean_token_accuracy": 0.19379199296236038,
      "num_tokens": 72737150.0,
      "step": 33195
    },
    {
      "entropy": 5.661468029022217,
      "epoch": 3.294958316792378,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039388709430004465,
      "loss": 4.9334,
      "mean_token_accuracy": 0.19192452728748322,
      "num_tokens": 72747523.0,
      "step": 33200
    },
    {
      "entropy": 5.664415740966797,
      "epoch": 3.2954545454545454,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039385701135888715,
      "loss": 5.0299,
      "mean_token_accuracy": 0.19570449292659758,
      "num_tokens": 72760163.0,
      "step": 33205
    },
    {
      "entropy": 5.705789613723755,
      "epoch": 3.295950774116713,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003938269254703318,
      "loss": 5.0448,
      "mean_token_accuracy": 0.17972014248371124,
      "num_tokens": 72770031.0,
      "step": 33210
    },
    {
      "entropy": 5.637001180648804,
      "epoch": 3.2964470027788804,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039379683663512475,
      "loss": 4.8957,
      "mean_token_accuracy": 0.1908359631896019,
      "num_tokens": 72780187.0,
      "step": 33215
    },
    {
      "entropy": 5.628072881698609,
      "epoch": 3.2969432314410483,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003937667448540123,
      "loss": 5.011,
      "mean_token_accuracy": 0.1845507189631462,
      "num_tokens": 72790486.0,
      "step": 33220
    },
    {
      "entropy": 5.56777777671814,
      "epoch": 3.2974394601032158,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039373665012774027,
      "loss": 4.8502,
      "mean_token_accuracy": 0.1970338374376297,
      "num_tokens": 72800086.0,
      "step": 33225
    },
    {
      "entropy": 5.672174692153931,
      "epoch": 3.2979356887653832,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039370655245705527,
      "loss": 5.0606,
      "mean_token_accuracy": 0.18482479751110076,
      "num_tokens": 72811018.0,
      "step": 33230
    },
    {
      "entropy": 5.640746355056763,
      "epoch": 3.2984319174275507,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039367645184270364,
      "loss": 4.9686,
      "mean_token_accuracy": 0.19175706654787064,
      "num_tokens": 72821792.0,
      "step": 33235
    },
    {
      "entropy": 5.660642099380493,
      "epoch": 3.298928146089718,
      "grad_norm": 1.125,
      "learning_rate": 0.0003936463482854317,
      "loss": 5.0676,
      "mean_token_accuracy": 0.1918063670396805,
      "num_tokens": 72833782.0,
      "step": 33240
    },
    {
      "entropy": 5.664888668060303,
      "epoch": 3.2994243747518857,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003936162417859858,
      "loss": 4.9837,
      "mean_token_accuracy": 0.19121472537517548,
      "num_tokens": 72844480.0,
      "step": 33245
    },
    {
      "entropy": 5.7055260181427006,
      "epoch": 3.299920603414053,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039358613234511286,
      "loss": 5.1209,
      "mean_token_accuracy": 0.18074070811271667,
      "num_tokens": 72855855.0,
      "step": 33250
    },
    {
      "entropy": 5.641639852523804,
      "epoch": 3.3004168320762206,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039355601996355944,
      "loss": 4.9465,
      "mean_token_accuracy": 0.18552816957235335,
      "num_tokens": 72866444.0,
      "step": 33255
    },
    {
      "entropy": 5.642256069183349,
      "epoch": 3.300913060738388,
      "grad_norm": 1.1875,
      "learning_rate": 0.000393525904642072,
      "loss": 4.9913,
      "mean_token_accuracy": 0.18174412548542024,
      "num_tokens": 72877083.0,
      "step": 33260
    },
    {
      "entropy": 5.5513824939727785,
      "epoch": 3.3014092894005556,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039349578638139766,
      "loss": 4.9477,
      "mean_token_accuracy": 0.1923418238759041,
      "num_tokens": 72888183.0,
      "step": 33265
    },
    {
      "entropy": 5.6816215991973875,
      "epoch": 3.3019055180627235,
      "grad_norm": 1.125,
      "learning_rate": 0.0003934656651822832,
      "loss": 5.1203,
      "mean_token_accuracy": 0.1751182794570923,
      "num_tokens": 72899171.0,
      "step": 33270
    },
    {
      "entropy": 5.666674709320068,
      "epoch": 3.302401746724891,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003934355410454754,
      "loss": 4.9842,
      "mean_token_accuracy": 0.18223138451576232,
      "num_tokens": 72909362.0,
      "step": 33275
    },
    {
      "entropy": 5.6646253108978275,
      "epoch": 3.3028979753870584,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039340541397172137,
      "loss": 4.961,
      "mean_token_accuracy": 0.19121915847063065,
      "num_tokens": 72920110.0,
      "step": 33280
    },
    {
      "entropy": 5.647672033309936,
      "epoch": 3.303394204049226,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003933752839617682,
      "loss": 5.0231,
      "mean_token_accuracy": 0.17707641273736954,
      "num_tokens": 72931006.0,
      "step": 33285
    },
    {
      "entropy": 5.6604138851165775,
      "epoch": 3.3038904327113934,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003933451510163632,
      "loss": 5.1217,
      "mean_token_accuracy": 0.17499126642942428,
      "num_tokens": 72941371.0,
      "step": 33290
    },
    {
      "entropy": 5.640499687194824,
      "epoch": 3.304386661373561,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003933150151362534,
      "loss": 4.97,
      "mean_token_accuracy": 0.18928758054971695,
      "num_tokens": 72953044.0,
      "step": 33295
    },
    {
      "entropy": 5.651807737350464,
      "epoch": 3.3048828900357283,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00039328487632218613,
      "loss": 5.0042,
      "mean_token_accuracy": 0.19146625101566314,
      "num_tokens": 72963581.0,
      "step": 33300
    },
    {
      "entropy": 5.577941751480102,
      "epoch": 3.305379118697896,
      "grad_norm": 1.125,
      "learning_rate": 0.0003932547345749088,
      "loss": 4.9303,
      "mean_token_accuracy": 0.1932595282793045,
      "num_tokens": 72974153.0,
      "step": 33305
    },
    {
      "entropy": 5.612428903579712,
      "epoch": 3.3058753473600637,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039322458989516897,
      "loss": 4.9763,
      "mean_token_accuracy": 0.1886230379343033,
      "num_tokens": 72985071.0,
      "step": 33310
    },
    {
      "entropy": 5.624249124526978,
      "epoch": 3.306371576022231,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039319444228371397,
      "loss": 4.9741,
      "mean_token_accuracy": 0.19135317504405974,
      "num_tokens": 72995592.0,
      "step": 33315
    },
    {
      "entropy": 5.605645942687988,
      "epoch": 3.3068678046843987,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003931642917412916,
      "loss": 4.9138,
      "mean_token_accuracy": 0.19392848163843154,
      "num_tokens": 73005431.0,
      "step": 33320
    },
    {
      "entropy": 5.563002395629883,
      "epoch": 3.307364033346566,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003931341382686493,
      "loss": 4.8774,
      "mean_token_accuracy": 0.1973934516310692,
      "num_tokens": 73015212.0,
      "step": 33325
    },
    {
      "entropy": 5.627389907836914,
      "epoch": 3.3078602620087336,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039310398186653497,
      "loss": 4.9421,
      "mean_token_accuracy": 0.1871776983141899,
      "num_tokens": 73026585.0,
      "step": 33330
    },
    {
      "entropy": 5.657269811630249,
      "epoch": 3.308356490670901,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039307382253569637,
      "loss": 5.0215,
      "mean_token_accuracy": 0.18489119857549668,
      "num_tokens": 73036844.0,
      "step": 33335
    },
    {
      "entropy": 5.595163202285766,
      "epoch": 3.3088527193330686,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003930436602768814,
      "loss": 4.9219,
      "mean_token_accuracy": 0.1896970272064209,
      "num_tokens": 73047881.0,
      "step": 33340
    },
    {
      "entropy": 5.589047145843506,
      "epoch": 3.309348947995236,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000393013495090838,
      "loss": 4.9679,
      "mean_token_accuracy": 0.1885521575808525,
      "num_tokens": 73058541.0,
      "step": 33345
    },
    {
      "entropy": 5.552940559387207,
      "epoch": 3.3098451766574035,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003929833269783143,
      "loss": 4.8579,
      "mean_token_accuracy": 0.20272941887378693,
      "num_tokens": 73069208.0,
      "step": 33350
    },
    {
      "entropy": 5.577508211135864,
      "epoch": 3.3103414053195714,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003929531559400582,
      "loss": 4.9182,
      "mean_token_accuracy": 0.19440184086561202,
      "num_tokens": 73079581.0,
      "step": 33355
    },
    {
      "entropy": 5.606033277511597,
      "epoch": 3.310837633981739,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039292298197681803,
      "loss": 4.9533,
      "mean_token_accuracy": 0.19160328656435013,
      "num_tokens": 73090126.0,
      "step": 33360
    },
    {
      "entropy": 5.596736907958984,
      "epoch": 3.3113338626439064,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039289280508934203,
      "loss": 4.954,
      "mean_token_accuracy": 0.188624007999897,
      "num_tokens": 73101421.0,
      "step": 33365
    },
    {
      "entropy": 5.654076671600341,
      "epoch": 3.311830091306074,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039286262527837856,
      "loss": 4.9428,
      "mean_token_accuracy": 0.19596467465162276,
      "num_tokens": 73111826.0,
      "step": 33370
    },
    {
      "entropy": 5.654988622665405,
      "epoch": 3.3123263199682413,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003928324425446759,
      "loss": 4.973,
      "mean_token_accuracy": 0.19367683976888656,
      "num_tokens": 73124195.0,
      "step": 33375
    },
    {
      "entropy": 5.582295751571655,
      "epoch": 3.312822548630409,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039280225688898256,
      "loss": 4.9597,
      "mean_token_accuracy": 0.1873883992433548,
      "num_tokens": 73135590.0,
      "step": 33380
    },
    {
      "entropy": 5.61033821105957,
      "epoch": 3.3133187772925763,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039277206831204715,
      "loss": 4.8987,
      "mean_token_accuracy": 0.1879389241337776,
      "num_tokens": 73145577.0,
      "step": 33385
    },
    {
      "entropy": 5.526025676727295,
      "epoch": 3.3138150059547438,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003927418768146182,
      "loss": 4.9214,
      "mean_token_accuracy": 0.20107753425836564,
      "num_tokens": 73156717.0,
      "step": 33390
    },
    {
      "entropy": 5.627098798751831,
      "epoch": 3.3143112346169117,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039271168239744434,
      "loss": 4.9506,
      "mean_token_accuracy": 0.18597076684236527,
      "num_tokens": 73166534.0,
      "step": 33395
    },
    {
      "entropy": 5.679745054244995,
      "epoch": 3.314807463279079,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003926814850612745,
      "loss": 4.9299,
      "mean_token_accuracy": 0.1911795988678932,
      "num_tokens": 73176182.0,
      "step": 33400
    },
    {
      "entropy": 5.659327459335327,
      "epoch": 3.3153036919412466,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003926512848068573,
      "loss": 4.9993,
      "mean_token_accuracy": 0.18345871120691298,
      "num_tokens": 73186837.0,
      "step": 33405
    },
    {
      "entropy": 5.57657642364502,
      "epoch": 3.315799920603414,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003926210816349418,
      "loss": 4.9221,
      "mean_token_accuracy": 0.197623872756958,
      "num_tokens": 73197786.0,
      "step": 33410
    },
    {
      "entropy": 5.557447147369385,
      "epoch": 3.3162961492655816,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000392590875546277,
      "loss": 4.883,
      "mean_token_accuracy": 0.19044104367494583,
      "num_tokens": 73208408.0,
      "step": 33415
    },
    {
      "entropy": 5.627487659454346,
      "epoch": 3.316792377927749,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039256066654161175,
      "loss": 5.0084,
      "mean_token_accuracy": 0.18916217237710953,
      "num_tokens": 73219596.0,
      "step": 33420
    },
    {
      "entropy": 5.668840837478638,
      "epoch": 3.3172886065899165,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003925304546216954,
      "loss": 5.0344,
      "mean_token_accuracy": 0.18634878545999528,
      "num_tokens": 73232301.0,
      "step": 33425
    },
    {
      "entropy": 5.721412944793701,
      "epoch": 3.317784835252084,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000392500239787277,
      "loss": 5.0247,
      "mean_token_accuracy": 0.18541740775108337,
      "num_tokens": 73243447.0,
      "step": 33430
    },
    {
      "entropy": 5.674263429641724,
      "epoch": 3.3182810639142515,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003924700220391058,
      "loss": 5.0483,
      "mean_token_accuracy": 0.18089107424020767,
      "num_tokens": 73254336.0,
      "step": 33435
    },
    {
      "entropy": 5.697359895706176,
      "epoch": 3.3187772925764194,
      "grad_norm": 1.0,
      "learning_rate": 0.0003924398013779312,
      "loss": 5.0122,
      "mean_token_accuracy": 0.19378573596477508,
      "num_tokens": 73266098.0,
      "step": 33440
    },
    {
      "entropy": 5.586015939712524,
      "epoch": 3.319273521238587,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003924095778045026,
      "loss": 4.9967,
      "mean_token_accuracy": 0.1945002406835556,
      "num_tokens": 73276026.0,
      "step": 33445
    },
    {
      "entropy": 5.631934976577758,
      "epoch": 3.3197697499007544,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003923793513195695,
      "loss": 4.9452,
      "mean_token_accuracy": 0.1956189215183258,
      "num_tokens": 73286151.0,
      "step": 33450
    },
    {
      "entropy": 5.6792622089385985,
      "epoch": 3.320265978562922,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039234912192388137,
      "loss": 5.0576,
      "mean_token_accuracy": 0.18159447014331817,
      "num_tokens": 73298086.0,
      "step": 33455
    },
    {
      "entropy": 5.706614351272583,
      "epoch": 3.3207622072250893,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003923188896181879,
      "loss": 5.1093,
      "mean_token_accuracy": 0.1781630501151085,
      "num_tokens": 73308732.0,
      "step": 33460
    },
    {
      "entropy": 5.656428241729737,
      "epoch": 3.3212584358872568,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003922886544032388,
      "loss": 4.974,
      "mean_token_accuracy": 0.18587277233600616,
      "num_tokens": 73318979.0,
      "step": 33465
    },
    {
      "entropy": 5.6935387134552,
      "epoch": 3.3217546645494243,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039225841627978374,
      "loss": 5.0515,
      "mean_token_accuracy": 0.1857292726635933,
      "num_tokens": 73329724.0,
      "step": 33470
    },
    {
      "entropy": 5.6207190990448,
      "epoch": 3.3222508932115917,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003922281752485727,
      "loss": 4.9821,
      "mean_token_accuracy": 0.18437851816415787,
      "num_tokens": 73340463.0,
      "step": 33475
    },
    {
      "entropy": 5.559943008422851,
      "epoch": 3.3227471218737596,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039219793131035544,
      "loss": 4.8704,
      "mean_token_accuracy": 0.19879563599824907,
      "num_tokens": 73351089.0,
      "step": 33480
    },
    {
      "entropy": 5.670496559143066,
      "epoch": 3.323243350535927,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003921676844658821,
      "loss": 5.071,
      "mean_token_accuracy": 0.17887813299894334,
      "num_tokens": 73361747.0,
      "step": 33485
    },
    {
      "entropy": 5.665007877349853,
      "epoch": 3.3237395791980946,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003921374347159026,
      "loss": 5.0019,
      "mean_token_accuracy": 0.1861254721879959,
      "num_tokens": 73372340.0,
      "step": 33490
    },
    {
      "entropy": 5.655912590026856,
      "epoch": 3.324235807860262,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003921071820611671,
      "loss": 4.9878,
      "mean_token_accuracy": 0.18712233603000641,
      "num_tokens": 73383972.0,
      "step": 33495
    },
    {
      "entropy": 5.562890148162841,
      "epoch": 3.3247320365224295,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003920769265024259,
      "loss": 4.9576,
      "mean_token_accuracy": 0.19180981069803238,
      "num_tokens": 73394346.0,
      "step": 33500
    },
    {
      "entropy": 5.6223101139068605,
      "epoch": 3.325228265184597,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039204666804042913,
      "loss": 5.0028,
      "mean_token_accuracy": 0.1810423254966736,
      "num_tokens": 73404964.0,
      "step": 33505
    },
    {
      "entropy": 5.681217432022095,
      "epoch": 3.3257244938467645,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003920164066759273,
      "loss": 5.0338,
      "mean_token_accuracy": 0.18276817202568055,
      "num_tokens": 73415487.0,
      "step": 33510
    },
    {
      "entropy": 5.705403566360474,
      "epoch": 3.326220722508932,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003919861424096707,
      "loss": 5.03,
      "mean_token_accuracy": 0.18832736909389497,
      "num_tokens": 73424909.0,
      "step": 33515
    },
    {
      "entropy": 5.719451475143432,
      "epoch": 3.3267169511710994,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003919558752424098,
      "loss": 5.0256,
      "mean_token_accuracy": 0.1828944519162178,
      "num_tokens": 73435106.0,
      "step": 33520
    },
    {
      "entropy": 5.674173069000244,
      "epoch": 3.327213179833267,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003919256051748952,
      "loss": 5.0489,
      "mean_token_accuracy": 0.1798407942056656,
      "num_tokens": 73445606.0,
      "step": 33525
    },
    {
      "entropy": 5.591159009933472,
      "epoch": 3.327709408495435,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039189533220787765,
      "loss": 4.9743,
      "mean_token_accuracy": 0.18740407675504683,
      "num_tokens": 73456309.0,
      "step": 33530
    },
    {
      "entropy": 5.601495409011841,
      "epoch": 3.3282056371576023,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003918650563421076,
      "loss": 4.9589,
      "mean_token_accuracy": 0.1912628948688507,
      "num_tokens": 73467817.0,
      "step": 33535
    },
    {
      "entropy": 5.583947038650512,
      "epoch": 3.32870186581977,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000391834777578336,
      "loss": 4.9256,
      "mean_token_accuracy": 0.1913944587111473,
      "num_tokens": 73477312.0,
      "step": 33540
    },
    {
      "entropy": 5.675993490219116,
      "epoch": 3.3291980944819373,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003918044959173138,
      "loss": 5.0787,
      "mean_token_accuracy": 0.18533368557691574,
      "num_tokens": 73488772.0,
      "step": 33545
    },
    {
      "entropy": 5.645233011245727,
      "epoch": 3.3296943231441047,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003917742113597918,
      "loss": 4.9588,
      "mean_token_accuracy": 0.191229185461998,
      "num_tokens": 73500570.0,
      "step": 33550
    },
    {
      "entropy": 5.5899004459381105,
      "epoch": 3.330190551806272,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003917439239065208,
      "loss": 4.9233,
      "mean_token_accuracy": 0.1925743341445923,
      "num_tokens": 73511679.0,
      "step": 33555
    },
    {
      "entropy": 5.584977054595948,
      "epoch": 3.3306867804684397,
      "grad_norm": 1.125,
      "learning_rate": 0.00039171363355825216,
      "loss": 4.9482,
      "mean_token_accuracy": 0.1933127835392952,
      "num_tokens": 73523312.0,
      "step": 33560
    },
    {
      "entropy": 5.665597248077392,
      "epoch": 3.3311830091306076,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039168334031573695,
      "loss": 5.0293,
      "mean_token_accuracy": 0.183438940346241,
      "num_tokens": 73534789.0,
      "step": 33565
    },
    {
      "entropy": 5.70284514427185,
      "epoch": 3.331679237792775,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003916530441797262,
      "loss": 4.9896,
      "mean_token_accuracy": 0.1888306111097336,
      "num_tokens": 73545331.0,
      "step": 33570
    },
    {
      "entropy": 5.582738399505615,
      "epoch": 3.3321754664549426,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039162274515097147,
      "loss": 4.9292,
      "mean_token_accuracy": 0.1900951623916626,
      "num_tokens": 73555347.0,
      "step": 33575
    },
    {
      "entropy": 5.573919677734375,
      "epoch": 3.33267169511711,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003915924432302239,
      "loss": 5.0004,
      "mean_token_accuracy": 0.1820691332221031,
      "num_tokens": 73567180.0,
      "step": 33580
    },
    {
      "entropy": 5.641369390487671,
      "epoch": 3.3331679237792775,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003915621384182349,
      "loss": 5.0523,
      "mean_token_accuracy": 0.18559564650058746,
      "num_tokens": 73578589.0,
      "step": 33585
    },
    {
      "entropy": 5.642815828323364,
      "epoch": 3.333664152441445,
      "grad_norm": 1.125,
      "learning_rate": 0.0003915318307157561,
      "loss": 4.9534,
      "mean_token_accuracy": 0.1870204821228981,
      "num_tokens": 73588939.0,
      "step": 33590
    },
    {
      "entropy": 5.676202774047852,
      "epoch": 3.3341603811036125,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039150152012353896,
      "loss": 5.0222,
      "mean_token_accuracy": 0.18737622648477553,
      "num_tokens": 73600476.0,
      "step": 33595
    },
    {
      "entropy": 5.685246515274048,
      "epoch": 3.33465660976578,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039147120664233525,
      "loss": 5.0282,
      "mean_token_accuracy": 0.19324573278427123,
      "num_tokens": 73610826.0,
      "step": 33600
    },
    {
      "entropy": 5.633710336685181,
      "epoch": 3.3351528384279474,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039144089027289653,
      "loss": 4.9643,
      "mean_token_accuracy": 0.18566376268863677,
      "num_tokens": 73623029.0,
      "step": 33605
    },
    {
      "entropy": 5.607221031188965,
      "epoch": 3.335649067090115,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039141057101597465,
      "loss": 4.9602,
      "mean_token_accuracy": 0.1869424194097519,
      "num_tokens": 73634092.0,
      "step": 33610
    },
    {
      "entropy": 5.630801200866699,
      "epoch": 3.336145295752283,
      "grad_norm": 1.984375,
      "learning_rate": 0.0003913802488723215,
      "loss": 4.9422,
      "mean_token_accuracy": 0.20126294791698457,
      "num_tokens": 73643297.0,
      "step": 33615
    },
    {
      "entropy": 5.603030204772949,
      "epoch": 3.3366415244144503,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003913499238426889,
      "loss": 4.9747,
      "mean_token_accuracy": 0.18817686885595322,
      "num_tokens": 73654001.0,
      "step": 33620
    },
    {
      "entropy": 5.627749872207642,
      "epoch": 3.3371377530766178,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039131959592782884,
      "loss": 4.9667,
      "mean_token_accuracy": 0.19412513971328735,
      "num_tokens": 73664509.0,
      "step": 33625
    },
    {
      "entropy": 5.565874624252319,
      "epoch": 3.3376339817387852,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039128926512849356,
      "loss": 4.8632,
      "mean_token_accuracy": 0.19601541459560395,
      "num_tokens": 73674585.0,
      "step": 33630
    },
    {
      "entropy": 5.540516138076782,
      "epoch": 3.3381302104009527,
      "grad_norm": 1.15625,
      "learning_rate": 0.000391258931445435,
      "loss": 4.8554,
      "mean_token_accuracy": 0.19040701240301133,
      "num_tokens": 73684550.0,
      "step": 33635
    },
    {
      "entropy": 5.647283458709717,
      "epoch": 3.33862643906312,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003912285948794055,
      "loss": 5.0036,
      "mean_token_accuracy": 0.18604012131690978,
      "num_tokens": 73694767.0,
      "step": 33640
    },
    {
      "entropy": 5.773872375488281,
      "epoch": 3.3391226677252877,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003911982554311573,
      "loss": 5.0553,
      "mean_token_accuracy": 0.19022490978240966,
      "num_tokens": 73707900.0,
      "step": 33645
    },
    {
      "entropy": 5.674327564239502,
      "epoch": 3.3396188963874556,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039116791310144275,
      "loss": 5.0243,
      "mean_token_accuracy": 0.18774009346961976,
      "num_tokens": 73719014.0,
      "step": 33650
    },
    {
      "entropy": 5.6858972072601315,
      "epoch": 3.340115125049623,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003911375678910142,
      "loss": 5.1139,
      "mean_token_accuracy": 0.1880767434835434,
      "num_tokens": 73730107.0,
      "step": 33655
    },
    {
      "entropy": 5.698630619049072,
      "epoch": 3.3406113537117905,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039110721980062434,
      "loss": 5.1062,
      "mean_token_accuracy": 0.18452730625867844,
      "num_tokens": 73742355.0,
      "step": 33660
    },
    {
      "entropy": 5.690839672088623,
      "epoch": 3.341107582373958,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039107686883102554,
      "loss": 5.0522,
      "mean_token_accuracy": 0.18382985442876815,
      "num_tokens": 73753372.0,
      "step": 33665
    },
    {
      "entropy": 5.6460212707519535,
      "epoch": 3.3416038110361255,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039104651498297053,
      "loss": 4.9638,
      "mean_token_accuracy": 0.19116344302892685,
      "num_tokens": 73764547.0,
      "step": 33670
    },
    {
      "entropy": 5.679820442199707,
      "epoch": 3.342100039698293,
      "grad_norm": 1.0625,
      "learning_rate": 0.000391016158257212,
      "loss": 5.0493,
      "mean_token_accuracy": 0.18340421617031097,
      "num_tokens": 73775505.0,
      "step": 33675
    },
    {
      "entropy": 5.7338536262512205,
      "epoch": 3.3425962683604604,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003909857986545028,
      "loss": 5.0799,
      "mean_token_accuracy": 0.1804432138800621,
      "num_tokens": 73785963.0,
      "step": 33680
    },
    {
      "entropy": 5.673681306838989,
      "epoch": 3.343092497022628,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039095543617559566,
      "loss": 4.9358,
      "mean_token_accuracy": 0.19177063554525375,
      "num_tokens": 73796457.0,
      "step": 33685
    },
    {
      "entropy": 5.69313793182373,
      "epoch": 3.3435887256847954,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003909250708212436,
      "loss": 4.9995,
      "mean_token_accuracy": 0.18699901252985002,
      "num_tokens": 73807662.0,
      "step": 33690
    },
    {
      "entropy": 5.621763563156128,
      "epoch": 3.344084954346963,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039089470259219956,
      "loss": 5.0096,
      "mean_token_accuracy": 0.18262848854064942,
      "num_tokens": 73818688.0,
      "step": 33695
    },
    {
      "entropy": 5.698143672943115,
      "epoch": 3.3445811830091308,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039086433148921665,
      "loss": 5.0393,
      "mean_token_accuracy": 0.18780644983053207,
      "num_tokens": 73831234.0,
      "step": 33700
    },
    {
      "entropy": 5.652841806411743,
      "epoch": 3.3450774116712982,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003908339575130479,
      "loss": 5.005,
      "mean_token_accuracy": 0.19528495520353317,
      "num_tokens": 73841505.0,
      "step": 33705
    },
    {
      "entropy": 5.6665881156921385,
      "epoch": 3.3455736403334657,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003908035806644467,
      "loss": 5.0551,
      "mean_token_accuracy": 0.1813798025250435,
      "num_tokens": 73851900.0,
      "step": 33710
    },
    {
      "entropy": 5.72101674079895,
      "epoch": 3.346069868995633,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039077320094416617,
      "loss": 4.974,
      "mean_token_accuracy": 0.19270550459623337,
      "num_tokens": 73862317.0,
      "step": 33715
    },
    {
      "entropy": 5.699909400939942,
      "epoch": 3.3465660976578007,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003907428183529598,
      "loss": 5.0078,
      "mean_token_accuracy": 0.1808699667453766,
      "num_tokens": 73872166.0,
      "step": 33720
    },
    {
      "entropy": 5.652720355987549,
      "epoch": 3.347062326319968,
      "grad_norm": 1.375,
      "learning_rate": 0.00039071243289158085,
      "loss": 5.0042,
      "mean_token_accuracy": 0.18637402653694152,
      "num_tokens": 73883064.0,
      "step": 33725
    },
    {
      "entropy": 5.656395721435547,
      "epoch": 3.3475585549821356,
      "grad_norm": 1.125,
      "learning_rate": 0.00039068204456078296,
      "loss": 4.9867,
      "mean_token_accuracy": 0.18786996603012085,
      "num_tokens": 73895032.0,
      "step": 33730
    },
    {
      "entropy": 5.65455379486084,
      "epoch": 3.3480547836443035,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039065165336131963,
      "loss": 4.9665,
      "mean_token_accuracy": 0.1897524416446686,
      "num_tokens": 73906309.0,
      "step": 33735
    },
    {
      "entropy": 5.626242637634277,
      "epoch": 3.348551012306471,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003906212592939445,
      "loss": 4.9671,
      "mean_token_accuracy": 0.18674151301383973,
      "num_tokens": 73917240.0,
      "step": 33740
    },
    {
      "entropy": 5.6043556213378904,
      "epoch": 3.3490472409686385,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039059086235941127,
      "loss": 4.961,
      "mean_token_accuracy": 0.1866143599152565,
      "num_tokens": 73928216.0,
      "step": 33745
    },
    {
      "entropy": 5.6156598091125485,
      "epoch": 3.349543469630806,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003905604625584737,
      "loss": 4.9755,
      "mean_token_accuracy": 0.1880088284611702,
      "num_tokens": 73940115.0,
      "step": 33750
    },
    {
      "entropy": 5.681036853790284,
      "epoch": 3.3500396982929734,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003905300598918857,
      "loss": 5.0723,
      "mean_token_accuracy": 0.18676788210868836,
      "num_tokens": 73950756.0,
      "step": 33755
    },
    {
      "entropy": 5.637555742263794,
      "epoch": 3.350535926955141,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039049965436040125,
      "loss": 4.9767,
      "mean_token_accuracy": 0.18870867192745208,
      "num_tokens": 73962269.0,
      "step": 33760
    },
    {
      "entropy": 5.6561534881591795,
      "epoch": 3.3510321556173084,
      "grad_norm": 1.25,
      "learning_rate": 0.0003904692459647741,
      "loss": 4.9523,
      "mean_token_accuracy": 0.18487376123666763,
      "num_tokens": 73971883.0,
      "step": 33765
    },
    {
      "entropy": 5.713154554367065,
      "epoch": 3.351528384279476,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039043883470575856,
      "loss": 5.0822,
      "mean_token_accuracy": 0.19040634632110595,
      "num_tokens": 73983801.0,
      "step": 33770
    },
    {
      "entropy": 5.6070188045501705,
      "epoch": 3.3520246129416433,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039040842058410863,
      "loss": 4.9676,
      "mean_token_accuracy": 0.19410212635993956,
      "num_tokens": 73994198.0,
      "step": 33775
    },
    {
      "entropy": 5.663856410980225,
      "epoch": 3.352520841603811,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039037800360057864,
      "loss": 5.0336,
      "mean_token_accuracy": 0.17688684165477753,
      "num_tokens": 74005521.0,
      "step": 33780
    },
    {
      "entropy": 5.70843620300293,
      "epoch": 3.3530170702659787,
      "grad_norm": 1.125,
      "learning_rate": 0.0003903475837559226,
      "loss": 5.0405,
      "mean_token_accuracy": 0.19452300816774368,
      "num_tokens": 74016606.0,
      "step": 33785
    },
    {
      "entropy": 5.643022155761718,
      "epoch": 3.353513298928146,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039031716105089516,
      "loss": 4.9967,
      "mean_token_accuracy": 0.19213480204343797,
      "num_tokens": 74027111.0,
      "step": 33790
    },
    {
      "entropy": 5.610895490646362,
      "epoch": 3.3540095275903137,
      "grad_norm": 1.25,
      "learning_rate": 0.0003902867354862506,
      "loss": 4.9303,
      "mean_token_accuracy": 0.19139216393232344,
      "num_tokens": 74037425.0,
      "step": 33795
    },
    {
      "entropy": 5.623346853256225,
      "epoch": 3.354505756252481,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003902563070627434,
      "loss": 5.0085,
      "mean_token_accuracy": 0.18884601891040803,
      "num_tokens": 74048622.0,
      "step": 33800
    },
    {
      "entropy": 5.632418251037597,
      "epoch": 3.3550019849146486,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039022587578112813,
      "loss": 4.9555,
      "mean_token_accuracy": 0.18867465257644653,
      "num_tokens": 74058430.0,
      "step": 33805
    },
    {
      "entropy": 5.684458446502686,
      "epoch": 3.355498213576816,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039019544164215956,
      "loss": 4.9779,
      "mean_token_accuracy": 0.19240285009145736,
      "num_tokens": 74069488.0,
      "step": 33810
    },
    {
      "entropy": 5.681347370147705,
      "epoch": 3.3559944422389836,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039016500464659213,
      "loss": 4.984,
      "mean_token_accuracy": 0.19225403517484665,
      "num_tokens": 74079142.0,
      "step": 33815
    },
    {
      "entropy": 5.6405035018920895,
      "epoch": 3.3564906709011515,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003901345647951807,
      "loss": 5.036,
      "mean_token_accuracy": 0.18826033622026445,
      "num_tokens": 74089984.0,
      "step": 33820
    },
    {
      "entropy": 5.617407894134521,
      "epoch": 3.356986899563319,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039010412208868026,
      "loss": 4.9998,
      "mean_token_accuracy": 0.18271030187606813,
      "num_tokens": 74101225.0,
      "step": 33825
    },
    {
      "entropy": 5.610811376571656,
      "epoch": 3.3574831282254864,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003900736765278455,
      "loss": 4.9189,
      "mean_token_accuracy": 0.1986124560236931,
      "num_tokens": 74111198.0,
      "step": 33830
    },
    {
      "entropy": 5.696923732757568,
      "epoch": 3.357979356887654,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039004322811343165,
      "loss": 5.0771,
      "mean_token_accuracy": 0.18076223134994507,
      "num_tokens": 74121458.0,
      "step": 33835
    },
    {
      "entropy": 5.651189088821411,
      "epoch": 3.3584755855498214,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039001277684619357,
      "loss": 4.9982,
      "mean_token_accuracy": 0.19254460483789443,
      "num_tokens": 74133338.0,
      "step": 33840
    },
    {
      "entropy": 5.647135400772095,
      "epoch": 3.358971814211989,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003899823227268865,
      "loss": 5.0198,
      "mean_token_accuracy": 0.18660739064216614,
      "num_tokens": 74144062.0,
      "step": 33845
    },
    {
      "entropy": 5.6202489852905275,
      "epoch": 3.3594680428741563,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038995186575626554,
      "loss": 4.9644,
      "mean_token_accuracy": 0.19039268493652345,
      "num_tokens": 74154001.0,
      "step": 33850
    },
    {
      "entropy": 5.627652788162232,
      "epoch": 3.359964271536324,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000389921405935086,
      "loss": 4.9486,
      "mean_token_accuracy": 0.18760014176368714,
      "num_tokens": 74164254.0,
      "step": 33855
    },
    {
      "entropy": 5.59906587600708,
      "epoch": 3.3604605001984913,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038989094326410327,
      "loss": 4.9408,
      "mean_token_accuracy": 0.1955254480242729,
      "num_tokens": 74173915.0,
      "step": 33860
    },
    {
      "entropy": 5.7640222072601315,
      "epoch": 3.3609567288606588,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003898604777440727,
      "loss": 5.113,
      "mean_token_accuracy": 0.17349295541644097,
      "num_tokens": 74185903.0,
      "step": 33865
    },
    {
      "entropy": 5.713067960739136,
      "epoch": 3.3614529575228267,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003898300093757499,
      "loss": 5.0217,
      "mean_token_accuracy": 0.18190884739160537,
      "num_tokens": 74198361.0,
      "step": 33870
    },
    {
      "entropy": 5.620531129837036,
      "epoch": 3.361949186184994,
      "grad_norm": 1.125,
      "learning_rate": 0.0003897995381598901,
      "loss": 4.9614,
      "mean_token_accuracy": 0.18408718854188919,
      "num_tokens": 74210467.0,
      "step": 33875
    },
    {
      "entropy": 5.59224534034729,
      "epoch": 3.3624454148471616,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003897690640972492,
      "loss": 4.925,
      "mean_token_accuracy": 0.19391926378011703,
      "num_tokens": 74221717.0,
      "step": 33880
    },
    {
      "entropy": 5.6171324253082275,
      "epoch": 3.362941643509329,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038973858718858295,
      "loss": 4.9893,
      "mean_token_accuracy": 0.183023102581501,
      "num_tokens": 74233164.0,
      "step": 33885
    },
    {
      "entropy": 5.711560535430908,
      "epoch": 3.3634378721714966,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003897081074346468,
      "loss": 5.0486,
      "mean_token_accuracy": 0.1872299462556839,
      "num_tokens": 74245693.0,
      "step": 33890
    },
    {
      "entropy": 5.610185527801514,
      "epoch": 3.363934100833664,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038967762483619684,
      "loss": 4.9277,
      "mean_token_accuracy": 0.19512321650981904,
      "num_tokens": 74256499.0,
      "step": 33895
    },
    {
      "entropy": 5.664201831817627,
      "epoch": 3.3644303294958315,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003896471393939889,
      "loss": 5.0432,
      "mean_token_accuracy": 0.18977957963943481,
      "num_tokens": 74267713.0,
      "step": 33900
    },
    {
      "entropy": 5.69452223777771,
      "epoch": 3.364926558157999,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003896166511087789,
      "loss": 5.0623,
      "mean_token_accuracy": 0.18196397721767427,
      "num_tokens": 74279149.0,
      "step": 33905
    },
    {
      "entropy": 5.630140066146851,
      "epoch": 3.365422786820167,
      "grad_norm": 1.203125,
      "learning_rate": 0.000389586159981323,
      "loss": 4.9848,
      "mean_token_accuracy": 0.1895080417394638,
      "num_tokens": 74289520.0,
      "step": 33910
    },
    {
      "entropy": 5.700515842437744,
      "epoch": 3.3659190154823344,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003895556660123771,
      "loss": 4.9956,
      "mean_token_accuracy": 0.1896233320236206,
      "num_tokens": 74300546.0,
      "step": 33915
    },
    {
      "entropy": 5.619940280914307,
      "epoch": 3.366415244144502,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038952516920269764,
      "loss": 4.9415,
      "mean_token_accuracy": 0.18487996309995652,
      "num_tokens": 74310687.0,
      "step": 33920
    },
    {
      "entropy": 5.611721515655518,
      "epoch": 3.3669114728066694,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003894946695530408,
      "loss": 4.9844,
      "mean_token_accuracy": 0.1869920387864113,
      "num_tokens": 74322251.0,
      "step": 33925
    },
    {
      "entropy": 5.690295028686523,
      "epoch": 3.367407701468837,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038946416706416275,
      "loss": 5.0319,
      "mean_token_accuracy": 0.1816723346710205,
      "num_tokens": 74333845.0,
      "step": 33930
    },
    {
      "entropy": 5.705665111541748,
      "epoch": 3.3679039301310043,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003894336617368201,
      "loss": 5.0231,
      "mean_token_accuracy": 0.1885876849293709,
      "num_tokens": 74344188.0,
      "step": 33935
    },
    {
      "entropy": 5.678337669372558,
      "epoch": 3.368400158793172,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038940315357176926,
      "loss": 5.0569,
      "mean_token_accuracy": 0.18446802347898483,
      "num_tokens": 74355174.0,
      "step": 33940
    },
    {
      "entropy": 5.690198040008545,
      "epoch": 3.3688963874553393,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038937264256976657,
      "loss": 4.9212,
      "mean_token_accuracy": 0.19322312921285628,
      "num_tokens": 74365396.0,
      "step": 33945
    },
    {
      "entropy": 5.6763458251953125,
      "epoch": 3.3693926161175067,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003893421287315689,
      "loss": 4.9852,
      "mean_token_accuracy": 0.18619803786277772,
      "num_tokens": 74377303.0,
      "step": 33950
    },
    {
      "entropy": 5.592645692825317,
      "epoch": 3.3698888447796747,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003893116120579328,
      "loss": 4.9871,
      "mean_token_accuracy": 0.18629904985427856,
      "num_tokens": 74389967.0,
      "step": 33955
    },
    {
      "entropy": 5.638763523101806,
      "epoch": 3.370385073441842,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003892810925496151,
      "loss": 4.9696,
      "mean_token_accuracy": 0.18727408051490785,
      "num_tokens": 74400252.0,
      "step": 33960
    },
    {
      "entropy": 5.62344183921814,
      "epoch": 3.3708813021040096,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003892505702073725,
      "loss": 4.9639,
      "mean_token_accuracy": 0.19344616681337357,
      "num_tokens": 74412165.0,
      "step": 33965
    },
    {
      "entropy": 5.603759479522705,
      "epoch": 3.371377530766177,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038922004503196195,
      "loss": 4.9231,
      "mean_token_accuracy": 0.19068029671907424,
      "num_tokens": 74423078.0,
      "step": 33970
    },
    {
      "entropy": 5.603022241592408,
      "epoch": 3.3718737594283446,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003891895170241405,
      "loss": 4.9051,
      "mean_token_accuracy": 0.19342643171548843,
      "num_tokens": 74434552.0,
      "step": 33975
    },
    {
      "entropy": 5.693882846832276,
      "epoch": 3.372369988090512,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038915898618466487,
      "loss": 5.0288,
      "mean_token_accuracy": 0.1850075125694275,
      "num_tokens": 74446188.0,
      "step": 33980
    },
    {
      "entropy": 5.588215446472168,
      "epoch": 3.3728662167526795,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038912845251429256,
      "loss": 4.9112,
      "mean_token_accuracy": 0.1917141154408455,
      "num_tokens": 74456358.0,
      "step": 33985
    },
    {
      "entropy": 5.546199607849121,
      "epoch": 3.373362445414847,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038909791601378054,
      "loss": 4.8798,
      "mean_token_accuracy": 0.19096497744321822,
      "num_tokens": 74466155.0,
      "step": 33990
    },
    {
      "entropy": 5.673423385620117,
      "epoch": 3.373858674077015,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038906737668388595,
      "loss": 5.0746,
      "mean_token_accuracy": 0.18138618022203445,
      "num_tokens": 74477282.0,
      "step": 33995
    },
    {
      "entropy": 5.570738315582275,
      "epoch": 3.3743549027391824,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038903683452536634,
      "loss": 4.9517,
      "mean_token_accuracy": 0.18965776562690734,
      "num_tokens": 74489736.0,
      "step": 34000
    },
    {
      "entropy": 5.590321254730225,
      "epoch": 3.37485113140135,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003890062895389789,
      "loss": 4.9607,
      "mean_token_accuracy": 0.1886538490653038,
      "num_tokens": 74500797.0,
      "step": 34005
    },
    {
      "entropy": 5.59293761253357,
      "epoch": 3.3753473600635173,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003889757417254811,
      "loss": 4.981,
      "mean_token_accuracy": 0.19061471223831178,
      "num_tokens": 74510811.0,
      "step": 34010
    },
    {
      "entropy": 5.658205604553222,
      "epoch": 3.375843588725685,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038894519108563047,
      "loss": 4.9667,
      "mean_token_accuracy": 0.19219578355550765,
      "num_tokens": 74521726.0,
      "step": 34015
    },
    {
      "entropy": 5.614997720718383,
      "epoch": 3.3763398173878523,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038891463762018476,
      "loss": 4.9699,
      "mean_token_accuracy": 0.19341856390237808,
      "num_tokens": 74532260.0,
      "step": 34020
    },
    {
      "entropy": 5.69714994430542,
      "epoch": 3.3768360460500197,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038888408132990146,
      "loss": 5.049,
      "mean_token_accuracy": 0.18034251928329467,
      "num_tokens": 74544691.0,
      "step": 34025
    },
    {
      "entropy": 5.700321865081787,
      "epoch": 3.377332274712187,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038885352221553827,
      "loss": 5.0222,
      "mean_token_accuracy": 0.1870614007115364,
      "num_tokens": 74555250.0,
      "step": 34030
    },
    {
      "entropy": 5.623258781433106,
      "epoch": 3.3778285033743547,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003888229602778532,
      "loss": 4.9365,
      "mean_token_accuracy": 0.18915922939777374,
      "num_tokens": 74566079.0,
      "step": 34035
    },
    {
      "entropy": 5.64921817779541,
      "epoch": 3.3783247320365226,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00038879239551760386,
      "loss": 4.9999,
      "mean_token_accuracy": 0.18318423628807068,
      "num_tokens": 74577741.0,
      "step": 34040
    },
    {
      "entropy": 5.5912994861602785,
      "epoch": 3.37882096069869,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003887618279355483,
      "loss": 4.9339,
      "mean_token_accuracy": 0.19381120055913925,
      "num_tokens": 74588146.0,
      "step": 34045
    },
    {
      "entropy": 5.663030672073364,
      "epoch": 3.3793171893608576,
      "grad_norm": 1.0,
      "learning_rate": 0.00038873125753244466,
      "loss": 5.085,
      "mean_token_accuracy": 0.18288507163524628,
      "num_tokens": 74599745.0,
      "step": 34050
    },
    {
      "entropy": 5.66076545715332,
      "epoch": 3.379813418023025,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003887006843090508,
      "loss": 4.9486,
      "mean_token_accuracy": 0.1886933773756027,
      "num_tokens": 74611215.0,
      "step": 34055
    },
    {
      "entropy": 5.736796569824219,
      "epoch": 3.3803096466851925,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000388670108266125,
      "loss": 5.0475,
      "mean_token_accuracy": 0.19100068360567093,
      "num_tokens": 74620676.0,
      "step": 34060
    },
    {
      "entropy": 5.696078205108643,
      "epoch": 3.38080587534736,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003886395294044255,
      "loss": 5.0678,
      "mean_token_accuracy": 0.18617491871118547,
      "num_tokens": 74633129.0,
      "step": 34065
    },
    {
      "entropy": 5.6594194889068605,
      "epoch": 3.3813021040095275,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003886089477247105,
      "loss": 4.9459,
      "mean_token_accuracy": 0.19041821658611296,
      "num_tokens": 74643125.0,
      "step": 34070
    },
    {
      "entropy": 5.731527090072632,
      "epoch": 3.381798332671695,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003885783632277384,
      "loss": 5.0788,
      "mean_token_accuracy": 0.1830654561519623,
      "num_tokens": 74653573.0,
      "step": 34075
    },
    {
      "entropy": 5.628783273696899,
      "epoch": 3.382294561333863,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003885477759142676,
      "loss": 4.9285,
      "mean_token_accuracy": 0.19517298489809037,
      "num_tokens": 74663451.0,
      "step": 34080
    },
    {
      "entropy": 5.598822450637817,
      "epoch": 3.3827907899960303,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003885171857850567,
      "loss": 4.9403,
      "mean_token_accuracy": 0.1983172968029976,
      "num_tokens": 74673698.0,
      "step": 34085
    },
    {
      "entropy": 5.593297481536865,
      "epoch": 3.383287018658198,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003884865928408642,
      "loss": 5.0351,
      "mean_token_accuracy": 0.1875879853963852,
      "num_tokens": 74685952.0,
      "step": 34090
    },
    {
      "entropy": 5.629304695129394,
      "epoch": 3.3837832473203653,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003884559970824486,
      "loss": 4.979,
      "mean_token_accuracy": 0.1896556004881859,
      "num_tokens": 74698159.0,
      "step": 34095
    },
    {
      "entropy": 5.715069913864136,
      "epoch": 3.3842794759825328,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003884253985105689,
      "loss": 5.0483,
      "mean_token_accuracy": 0.18459729850292206,
      "num_tokens": 74708718.0,
      "step": 34100
    },
    {
      "entropy": 5.643872165679932,
      "epoch": 3.3847757046447002,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003883947971259837,
      "loss": 4.9065,
      "mean_token_accuracy": 0.1921476513147354,
      "num_tokens": 74718249.0,
      "step": 34105
    },
    {
      "entropy": 5.647002029418945,
      "epoch": 3.3852719333068677,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003883641929294519,
      "loss": 4.9042,
      "mean_token_accuracy": 0.2050986960530281,
      "num_tokens": 74728856.0,
      "step": 34110
    },
    {
      "entropy": 5.61637167930603,
      "epoch": 3.385768161969035,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003883335859217323,
      "loss": 4.9708,
      "mean_token_accuracy": 0.1961294785141945,
      "num_tokens": 74740126.0,
      "step": 34115
    },
    {
      "entropy": 5.619671583175659,
      "epoch": 3.3862643906312027,
      "grad_norm": 1.296875,
      "learning_rate": 0.000388302976103584,
      "loss": 4.9778,
      "mean_token_accuracy": 0.19078768491744996,
      "num_tokens": 74751277.0,
      "step": 34120
    },
    {
      "entropy": 5.620053386688232,
      "epoch": 3.3867606192933706,
      "grad_norm": 1.125,
      "learning_rate": 0.0003882723634757661,
      "loss": 5.0128,
      "mean_token_accuracy": 0.18868768811225892,
      "num_tokens": 74761498.0,
      "step": 34125
    },
    {
      "entropy": 5.610729598999024,
      "epoch": 3.387256847955538,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003882417480390376,
      "loss": 4.9567,
      "mean_token_accuracy": 0.18567631542682647,
      "num_tokens": 74771482.0,
      "step": 34130
    },
    {
      "entropy": 5.685754060745239,
      "epoch": 3.3877530766177055,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038821112979415786,
      "loss": 5.0285,
      "mean_token_accuracy": 0.18167900294065475,
      "num_tokens": 74783649.0,
      "step": 34135
    },
    {
      "entropy": 5.648758411407471,
      "epoch": 3.388249305279873,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000388180508741886,
      "loss": 4.9467,
      "mean_token_accuracy": 0.19049005955457687,
      "num_tokens": 74794142.0,
      "step": 34140
    },
    {
      "entropy": 5.613576936721802,
      "epoch": 3.3887455339420405,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038814988488298136,
      "loss": 4.9689,
      "mean_token_accuracy": 0.18934748470783233,
      "num_tokens": 74805369.0,
      "step": 34145
    },
    {
      "entropy": 5.627156782150268,
      "epoch": 3.389241762604208,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003881192582182034,
      "loss": 4.9786,
      "mean_token_accuracy": 0.18761470764875413,
      "num_tokens": 74816558.0,
      "step": 34150
    },
    {
      "entropy": 5.654330062866211,
      "epoch": 3.3897379912663754,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038808862874831164,
      "loss": 4.9538,
      "mean_token_accuracy": 0.19338107109069824,
      "num_tokens": 74826838.0,
      "step": 34155
    },
    {
      "entropy": 5.593519258499145,
      "epoch": 3.390234219928543,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003880579964740655,
      "loss": 4.9309,
      "mean_token_accuracy": 0.18517503887414932,
      "num_tokens": 74837182.0,
      "step": 34160
    },
    {
      "entropy": 5.638775730133057,
      "epoch": 3.390730448590711,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038802736139622473,
      "loss": 4.9184,
      "mean_token_accuracy": 0.19606312066316606,
      "num_tokens": 74847973.0,
      "step": 34165
    },
    {
      "entropy": 5.661556720733643,
      "epoch": 3.3912266772528783,
      "grad_norm": 1.125,
      "learning_rate": 0.00038799672351554895,
      "loss": 4.9733,
      "mean_token_accuracy": 0.19080301523208618,
      "num_tokens": 74858952.0,
      "step": 34170
    },
    {
      "entropy": 5.712412786483765,
      "epoch": 3.3917229059150458,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038796608283279786,
      "loss": 5.0131,
      "mean_token_accuracy": 0.1814824491739273,
      "num_tokens": 74870427.0,
      "step": 34175
    },
    {
      "entropy": 5.6064958572387695,
      "epoch": 3.3922191345772132,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003879354393487314,
      "loss": 4.977,
      "mean_token_accuracy": 0.1927547425031662,
      "num_tokens": 74881502.0,
      "step": 34180
    },
    {
      "entropy": 5.659103488922119,
      "epoch": 3.3927153632393807,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038790479306410934,
      "loss": 5.0281,
      "mean_token_accuracy": 0.1833551660180092,
      "num_tokens": 74893065.0,
      "step": 34185
    },
    {
      "entropy": 5.600415134429932,
      "epoch": 3.393211591901548,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003878741439796918,
      "loss": 4.9658,
      "mean_token_accuracy": 0.19123728573322296,
      "num_tokens": 74903806.0,
      "step": 34190
    },
    {
      "entropy": 5.732788372039795,
      "epoch": 3.3937078205637157,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003878434920962386,
      "loss": 5.0919,
      "mean_token_accuracy": 0.17868174463510514,
      "num_tokens": 74915025.0,
      "step": 34195
    },
    {
      "entropy": 5.67702841758728,
      "epoch": 3.394204049225883,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038781283741451,
      "loss": 5.0284,
      "mean_token_accuracy": 0.17615251392126083,
      "num_tokens": 74926438.0,
      "step": 34200
    },
    {
      "entropy": 5.675446891784668,
      "epoch": 3.3947002778880506,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038778217993526616,
      "loss": 4.9843,
      "mean_token_accuracy": 0.18287585079669952,
      "num_tokens": 74937707.0,
      "step": 34205
    },
    {
      "entropy": 5.683353853225708,
      "epoch": 3.395196506550218,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003877515196592672,
      "loss": 4.9626,
      "mean_token_accuracy": 0.1842968389391899,
      "num_tokens": 74947436.0,
      "step": 34210
    },
    {
      "entropy": 5.690292549133301,
      "epoch": 3.395692735212386,
      "grad_norm": 1.125,
      "learning_rate": 0.00038772085658727353,
      "loss": 5.042,
      "mean_token_accuracy": 0.1874651849269867,
      "num_tokens": 74958707.0,
      "step": 34215
    },
    {
      "entropy": 5.643882131576538,
      "epoch": 3.3961889638745535,
      "grad_norm": 1.125,
      "learning_rate": 0.00038769019072004555,
      "loss": 5.0008,
      "mean_token_accuracy": 0.18966262191534042,
      "num_tokens": 74970145.0,
      "step": 34220
    },
    {
      "entropy": 5.626382684707641,
      "epoch": 3.396685192536721,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003876595220583436,
      "loss": 4.9722,
      "mean_token_accuracy": 0.1848266378045082,
      "num_tokens": 74982095.0,
      "step": 34225
    },
    {
      "entropy": 5.613835668563842,
      "epoch": 3.3971814211988884,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003876288506029283,
      "loss": 4.8901,
      "mean_token_accuracy": 0.1928909868001938,
      "num_tokens": 74993648.0,
      "step": 34230
    },
    {
      "entropy": 5.631430673599243,
      "epoch": 3.397677649861056,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003875981763545602,
      "loss": 4.9432,
      "mean_token_accuracy": 0.18985170722007752,
      "num_tokens": 75003707.0,
      "step": 34235
    },
    {
      "entropy": 5.59314193725586,
      "epoch": 3.3981738785232234,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038756749931399993,
      "loss": 4.9406,
      "mean_token_accuracy": 0.18852822929620744,
      "num_tokens": 75013029.0,
      "step": 34240
    },
    {
      "entropy": 5.6160797595977785,
      "epoch": 3.398670107185391,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003875368194820083,
      "loss": 4.977,
      "mean_token_accuracy": 0.18516345471143722,
      "num_tokens": 75024555.0,
      "step": 34245
    },
    {
      "entropy": 5.674790620803833,
      "epoch": 3.399166335847559,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000387506136859346,
      "loss": 5.0122,
      "mean_token_accuracy": 0.18437838405370713,
      "num_tokens": 75034529.0,
      "step": 34250
    },
    {
      "entropy": 5.7046715259552006,
      "epoch": 3.3996625645097263,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003874754514467739,
      "loss": 5.0696,
      "mean_token_accuracy": 0.18227149099111556,
      "num_tokens": 75047733.0,
      "step": 34255
    },
    {
      "entropy": 5.685512351989746,
      "epoch": 3.4001587931718937,
      "grad_norm": 1.09375,
      "learning_rate": 0.000387444763245053,
      "loss": 5.0103,
      "mean_token_accuracy": 0.18958115130662917,
      "num_tokens": 75058576.0,
      "step": 34260
    },
    {
      "entropy": 5.64161868095398,
      "epoch": 3.400655021834061,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003874140722549443,
      "loss": 5.0239,
      "mean_token_accuracy": 0.1901859924197197,
      "num_tokens": 75068470.0,
      "step": 34265
    },
    {
      "entropy": 5.584136867523194,
      "epoch": 3.4011512504962287,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003873833784772088,
      "loss": 4.9027,
      "mean_token_accuracy": 0.1963128298521042,
      "num_tokens": 75079383.0,
      "step": 34270
    },
    {
      "entropy": 5.677322149276733,
      "epoch": 3.401647479158396,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003873526819126077,
      "loss": 5.0291,
      "mean_token_accuracy": 0.1805056244134903,
      "num_tokens": 75089474.0,
      "step": 34275
    },
    {
      "entropy": 5.620482301712036,
      "epoch": 3.4021437078205636,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003873219825619022,
      "loss": 5.0131,
      "mean_token_accuracy": 0.18729423284530639,
      "num_tokens": 75100286.0,
      "step": 34280
    },
    {
      "entropy": 5.720506763458252,
      "epoch": 3.402639936482731,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003872912804258536,
      "loss": 5.0419,
      "mean_token_accuracy": 0.18487170040607454,
      "num_tokens": 75111916.0,
      "step": 34285
    },
    {
      "entropy": 5.680771207809448,
      "epoch": 3.4031361651448986,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038726057550522307,
      "loss": 4.9451,
      "mean_token_accuracy": 0.19462666064500808,
      "num_tokens": 75122231.0,
      "step": 34290
    },
    {
      "entropy": 5.724327802658081,
      "epoch": 3.403632393807066,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038722986780077227,
      "loss": 5.08,
      "mean_token_accuracy": 0.178535033762455,
      "num_tokens": 75133289.0,
      "step": 34295
    },
    {
      "entropy": 5.617690324783325,
      "epoch": 3.404128622469234,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038719915731326264,
      "loss": 4.9653,
      "mean_token_accuracy": 0.19372289329767228,
      "num_tokens": 75143409.0,
      "step": 34300
    },
    {
      "entropy": 5.657311725616455,
      "epoch": 3.4046248511314015,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038716844404345556,
      "loss": 5.0483,
      "mean_token_accuracy": 0.18836937546730043,
      "num_tokens": 75154848.0,
      "step": 34305
    },
    {
      "entropy": 5.63142499923706,
      "epoch": 3.405121079793569,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038713772799211284,
      "loss": 5.0518,
      "mean_token_accuracy": 0.1840214565396309,
      "num_tokens": 75166861.0,
      "step": 34310
    },
    {
      "entropy": 5.658985280990601,
      "epoch": 3.4056173084557364,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003871070091599961,
      "loss": 4.9694,
      "mean_token_accuracy": 0.1943708285689354,
      "num_tokens": 75176934.0,
      "step": 34315
    },
    {
      "entropy": 5.61309494972229,
      "epoch": 3.406113537117904,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038707628754786706,
      "loss": 4.9278,
      "mean_token_accuracy": 0.18881726264953613,
      "num_tokens": 75187604.0,
      "step": 34320
    },
    {
      "entropy": 5.6034101963043215,
      "epoch": 3.4066097657800714,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038704556315648757,
      "loss": 4.9029,
      "mean_token_accuracy": 0.1937804937362671,
      "num_tokens": 75198524.0,
      "step": 34325
    },
    {
      "entropy": 5.639032936096191,
      "epoch": 3.407105994442239,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003870148359866196,
      "loss": 4.9979,
      "mean_token_accuracy": 0.18821868002414704,
      "num_tokens": 75209857.0,
      "step": 34330
    },
    {
      "entropy": 5.641287565231323,
      "epoch": 3.4076022231044067,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000386984106039025,
      "loss": 5.0505,
      "mean_token_accuracy": 0.18513579219579696,
      "num_tokens": 75220912.0,
      "step": 34335
    },
    {
      "entropy": 5.584141445159912,
      "epoch": 3.408098451766574,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000386953373314466,
      "loss": 4.9328,
      "mean_token_accuracy": 0.197462597489357,
      "num_tokens": 75232257.0,
      "step": 34340
    },
    {
      "entropy": 5.666533946990967,
      "epoch": 3.4085946804287417,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038692263781370446,
      "loss": 5.0232,
      "mean_token_accuracy": 0.18470944166183473,
      "num_tokens": 75244147.0,
      "step": 34345
    },
    {
      "entropy": 5.713325357437133,
      "epoch": 3.409090909090909,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038689189953750275,
      "loss": 5.0018,
      "mean_token_accuracy": 0.18923439532518388,
      "num_tokens": 75255910.0,
      "step": 34350
    },
    {
      "entropy": 5.653375101089478,
      "epoch": 3.4095871377530766,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038686115848662303,
      "loss": 5.0142,
      "mean_token_accuracy": 0.18340862691402435,
      "num_tokens": 75267611.0,
      "step": 34355
    },
    {
      "entropy": 5.640475702285767,
      "epoch": 3.410083366415244,
      "grad_norm": 1.125,
      "learning_rate": 0.0003868304146618276,
      "loss": 4.986,
      "mean_token_accuracy": 0.1835049495100975,
      "num_tokens": 75278835.0,
      "step": 34360
    },
    {
      "entropy": 5.708886528015137,
      "epoch": 3.4105795950774116,
      "grad_norm": 1.125,
      "learning_rate": 0.00038679966806387886,
      "loss": 5.0373,
      "mean_token_accuracy": 0.18518112003803253,
      "num_tokens": 75289317.0,
      "step": 34365
    },
    {
      "entropy": 5.659884929656982,
      "epoch": 3.411075823739579,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003867689186935393,
      "loss": 5.0627,
      "mean_token_accuracy": 0.18973617553710936,
      "num_tokens": 75302304.0,
      "step": 34370
    },
    {
      "entropy": 5.70028247833252,
      "epoch": 3.4115720524017465,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003867381665515712,
      "loss": 5.0084,
      "mean_token_accuracy": 0.1879429742693901,
      "num_tokens": 75312081.0,
      "step": 34375
    },
    {
      "entropy": 5.594358444213867,
      "epoch": 3.412068281063914,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003867074116387376,
      "loss": 4.9164,
      "mean_token_accuracy": 0.19247720241546631,
      "num_tokens": 75322948.0,
      "step": 34380
    },
    {
      "entropy": 5.637060689926147,
      "epoch": 3.412564509726082,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038667665395580083,
      "loss": 4.9873,
      "mean_token_accuracy": 0.18705396205186844,
      "num_tokens": 75333817.0,
      "step": 34385
    },
    {
      "entropy": 5.6320751190185545,
      "epoch": 3.4130607383882494,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003866458935035236,
      "loss": 4.9888,
      "mean_token_accuracy": 0.18335274904966353,
      "num_tokens": 75344014.0,
      "step": 34390
    },
    {
      "entropy": 5.685328531265259,
      "epoch": 3.413556967050417,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038661513028266886,
      "loss": 4.9929,
      "mean_token_accuracy": 0.18692205399274825,
      "num_tokens": 75355350.0,
      "step": 34395
    },
    {
      "entropy": 5.682299709320068,
      "epoch": 3.4140531957125844,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003865843642939994,
      "loss": 5.0086,
      "mean_token_accuracy": 0.18422770202159883,
      "num_tokens": 75364633.0,
      "step": 34400
    },
    {
      "entropy": 5.629451179504395,
      "epoch": 3.414549424374752,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038655359553827807,
      "loss": 4.9207,
      "mean_token_accuracy": 0.19425778537988664,
      "num_tokens": 75375103.0,
      "step": 34405
    },
    {
      "entropy": 5.647696208953858,
      "epoch": 3.4150456530369193,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000386522824016268,
      "loss": 5.0876,
      "mean_token_accuracy": 0.18666273206472397,
      "num_tokens": 75386112.0,
      "step": 34410
    },
    {
      "entropy": 5.555466556549073,
      "epoch": 3.415541881699087,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038649204972873225,
      "loss": 4.8763,
      "mean_token_accuracy": 0.19261167347431182,
      "num_tokens": 75396744.0,
      "step": 34415
    },
    {
      "entropy": 5.683435678482056,
      "epoch": 3.4160381103612547,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003864612726764339,
      "loss": 5.0454,
      "mean_token_accuracy": 0.18538528978824614,
      "num_tokens": 75408706.0,
      "step": 34420
    },
    {
      "entropy": 5.723730945587159,
      "epoch": 3.416534339023422,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003864304928601361,
      "loss": 4.9748,
      "mean_token_accuracy": 0.1848555713891983,
      "num_tokens": 75419036.0,
      "step": 34425
    },
    {
      "entropy": 5.615254640579224,
      "epoch": 3.4170305676855897,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003863997102806023,
      "loss": 4.9646,
      "mean_token_accuracy": 0.18886629939079286,
      "num_tokens": 75429415.0,
      "step": 34430
    },
    {
      "entropy": 5.563436985015869,
      "epoch": 3.417526796347757,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003863689249385956,
      "loss": 4.8836,
      "mean_token_accuracy": 0.2009575456380844,
      "num_tokens": 75441077.0,
      "step": 34435
    },
    {
      "entropy": 5.683196640014648,
      "epoch": 3.4180230250099246,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038633813683487954,
      "loss": 5.0741,
      "mean_token_accuracy": 0.1850430980324745,
      "num_tokens": 75452516.0,
      "step": 34440
    },
    {
      "entropy": 5.6236930847167965,
      "epoch": 3.418519253672092,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003863073459702177,
      "loss": 5.0042,
      "mean_token_accuracy": 0.18857343941926957,
      "num_tokens": 75463205.0,
      "step": 34445
    },
    {
      "entropy": 5.689514780044556,
      "epoch": 3.4190154823342596,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038627655234537345,
      "loss": 5.0091,
      "mean_token_accuracy": 0.18502978086471558,
      "num_tokens": 75474201.0,
      "step": 34450
    },
    {
      "entropy": 5.639941120147705,
      "epoch": 3.419511710996427,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038624575596111056,
      "loss": 4.9565,
      "mean_token_accuracy": 0.1962774157524109,
      "num_tokens": 75484310.0,
      "step": 34455
    },
    {
      "entropy": 5.618953609466553,
      "epoch": 3.4200079396585945,
      "grad_norm": 1.125,
      "learning_rate": 0.00038621495681819256,
      "loss": 4.9231,
      "mean_token_accuracy": 0.18661653846502305,
      "num_tokens": 75494583.0,
      "step": 34460
    },
    {
      "entropy": 5.670648097991943,
      "epoch": 3.420504168320762,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003861841549173833,
      "loss": 5.0671,
      "mean_token_accuracy": 0.1910461574792862,
      "num_tokens": 75505176.0,
      "step": 34465
    },
    {
      "entropy": 5.726762580871582,
      "epoch": 3.42100039698293,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003861533502594466,
      "loss": 5.0163,
      "mean_token_accuracy": 0.18823299258947374,
      "num_tokens": 75514795.0,
      "step": 34470
    },
    {
      "entropy": 5.6932274341583256,
      "epoch": 3.4214966256450974,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003861225428451463,
      "loss": 4.983,
      "mean_token_accuracy": 0.18700832724571229,
      "num_tokens": 75525975.0,
      "step": 34475
    },
    {
      "entropy": 5.643803501129151,
      "epoch": 3.421992854307265,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003860917326752464,
      "loss": 4.957,
      "mean_token_accuracy": 0.1873541846871376,
      "num_tokens": 75536648.0,
      "step": 34480
    },
    {
      "entropy": 5.705671739578247,
      "epoch": 3.4224890829694323,
      "grad_norm": 1.125,
      "learning_rate": 0.00038606091975051097,
      "loss": 5.1615,
      "mean_token_accuracy": 0.17917805761098862,
      "num_tokens": 75547724.0,
      "step": 34485
    },
    {
      "entropy": 5.662589502334595,
      "epoch": 3.4229853116316,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038603010407170393,
      "loss": 5.0393,
      "mean_token_accuracy": 0.18521237075328828,
      "num_tokens": 75559061.0,
      "step": 34490
    },
    {
      "entropy": 5.633554267883301,
      "epoch": 3.4234815402937673,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003859992856395896,
      "loss": 4.9924,
      "mean_token_accuracy": 0.18267502039670944,
      "num_tokens": 75569744.0,
      "step": 34495
    },
    {
      "entropy": 5.719705724716187,
      "epoch": 3.4239777689559348,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003859684644549322,
      "loss": 5.0941,
      "mean_token_accuracy": 0.18838583081960678,
      "num_tokens": 75579780.0,
      "step": 34500
    },
    {
      "entropy": 5.647161865234375,
      "epoch": 3.4244739976181022,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003859376405184959,
      "loss": 4.9729,
      "mean_token_accuracy": 0.1866798594594002,
      "num_tokens": 75591526.0,
      "step": 34505
    },
    {
      "entropy": 5.69771294593811,
      "epoch": 3.42497022628027,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003859068138310453,
      "loss": 5.0096,
      "mean_token_accuracy": 0.18509475886821747,
      "num_tokens": 75602573.0,
      "step": 34510
    },
    {
      "entropy": 5.6437602043151855,
      "epoch": 3.4254664549424376,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038587598439334466,
      "loss": 5.0238,
      "mean_token_accuracy": 0.18798624575138093,
      "num_tokens": 75613774.0,
      "step": 34515
    },
    {
      "entropy": 5.630634069442749,
      "epoch": 3.425962683604605,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038584515220615843,
      "loss": 4.9403,
      "mean_token_accuracy": 0.19788335263729095,
      "num_tokens": 75624435.0,
      "step": 34520
    },
    {
      "entropy": 5.619548368453979,
      "epoch": 3.4264589122667726,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038581431727025136,
      "loss": 4.946,
      "mean_token_accuracy": 0.1875216081738472,
      "num_tokens": 75634429.0,
      "step": 34525
    },
    {
      "entropy": 5.6084263801574705,
      "epoch": 3.42695514092894,
      "grad_norm": 1.046875,
      "learning_rate": 0.000385783479586388,
      "loss": 4.9369,
      "mean_token_accuracy": 0.18965437710285188,
      "num_tokens": 75646066.0,
      "step": 34530
    },
    {
      "entropy": 5.652806186676026,
      "epoch": 3.4274513695911075,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003857526391553331,
      "loss": 4.988,
      "mean_token_accuracy": 0.19231707006692886,
      "num_tokens": 75655718.0,
      "step": 34535
    },
    {
      "entropy": 5.632390689849854,
      "epoch": 3.427947598253275,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038572179597785134,
      "loss": 4.9312,
      "mean_token_accuracy": 0.19439249485731125,
      "num_tokens": 75666581.0,
      "step": 34540
    },
    {
      "entropy": 5.588725948333741,
      "epoch": 3.4284438269154425,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038569095005470765,
      "loss": 4.9218,
      "mean_token_accuracy": 0.19904749542474748,
      "num_tokens": 75676405.0,
      "step": 34545
    },
    {
      "entropy": 5.559793901443482,
      "epoch": 3.42894005557761,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038566010138666693,
      "loss": 4.9085,
      "mean_token_accuracy": 0.19580438435077668,
      "num_tokens": 75687723.0,
      "step": 34550
    },
    {
      "entropy": 5.6381665706634525,
      "epoch": 3.429436284239778,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038562924997449414,
      "loss": 5.0114,
      "mean_token_accuracy": 0.18606104403734208,
      "num_tokens": 75699461.0,
      "step": 34555
    },
    {
      "entropy": 5.697944641113281,
      "epoch": 3.4299325129019453,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003855983958189543,
      "loss": 5.0529,
      "mean_token_accuracy": 0.17995992600917815,
      "num_tokens": 75711679.0,
      "step": 34560
    },
    {
      "entropy": 5.711954879760742,
      "epoch": 3.430428741564113,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003855675389208126,
      "loss": 5.096,
      "mean_token_accuracy": 0.17882737666368484,
      "num_tokens": 75721027.0,
      "step": 34565
    },
    {
      "entropy": 5.671922206878662,
      "epoch": 3.4309249702262803,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003855366792808341,
      "loss": 4.9734,
      "mean_token_accuracy": 0.18896681368350982,
      "num_tokens": 75731194.0,
      "step": 34570
    },
    {
      "entropy": 5.734611511230469,
      "epoch": 3.4314211988884478,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038550581689978427,
      "loss": 5.0115,
      "mean_token_accuracy": 0.18416769951581954,
      "num_tokens": 75741730.0,
      "step": 34575
    },
    {
      "entropy": 5.674300479888916,
      "epoch": 3.4319174275506152,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038547495177842817,
      "loss": 5.017,
      "mean_token_accuracy": 0.1895319402217865,
      "num_tokens": 75752980.0,
      "step": 34580
    },
    {
      "entropy": 5.664656686782837,
      "epoch": 3.4324136562127827,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003854440839175314,
      "loss": 5.0621,
      "mean_token_accuracy": 0.1785211905837059,
      "num_tokens": 75764015.0,
      "step": 34585
    },
    {
      "entropy": 5.6868030548095705,
      "epoch": 3.43290988487495,
      "grad_norm": 1.125,
      "learning_rate": 0.00038541321331785934,
      "loss": 5.0558,
      "mean_token_accuracy": 0.18093695491552353,
      "num_tokens": 75775267.0,
      "step": 34590
    },
    {
      "entropy": 5.739681816101074,
      "epoch": 3.433406113537118,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038538233998017745,
      "loss": 4.991,
      "mean_token_accuracy": 0.1912893071770668,
      "num_tokens": 75785456.0,
      "step": 34595
    },
    {
      "entropy": 5.617831087112426,
      "epoch": 3.4339023421992856,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038535146390525143,
      "loss": 4.9527,
      "mean_token_accuracy": 0.1931486412882805,
      "num_tokens": 75795563.0,
      "step": 34600
    },
    {
      "entropy": 5.610944318771362,
      "epoch": 3.434398570861453,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038532058509384694,
      "loss": 4.9755,
      "mean_token_accuracy": 0.18946031481027603,
      "num_tokens": 75807068.0,
      "step": 34605
    },
    {
      "entropy": 5.640263366699219,
      "epoch": 3.4348947995236205,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003852897035467296,
      "loss": 4.9923,
      "mean_token_accuracy": 0.18253651559352874,
      "num_tokens": 75818354.0,
      "step": 34610
    },
    {
      "entropy": 5.6857398509979244,
      "epoch": 3.435391028185788,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038525881926466525,
      "loss": 5.0021,
      "mean_token_accuracy": 0.19257672131061554,
      "num_tokens": 75829382.0,
      "step": 34615
    },
    {
      "entropy": 5.622638082504272,
      "epoch": 3.4358872568479555,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038522793224841977,
      "loss": 4.9337,
      "mean_token_accuracy": 0.19073049873113632,
      "num_tokens": 75839278.0,
      "step": 34620
    },
    {
      "entropy": 5.61766300201416,
      "epoch": 3.436383485510123,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038519704249875917,
      "loss": 4.9568,
      "mean_token_accuracy": 0.1881062537431717,
      "num_tokens": 75851074.0,
      "step": 34625
    },
    {
      "entropy": 5.651001882553101,
      "epoch": 3.4368797141722904,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003851661500164493,
      "loss": 5.0269,
      "mean_token_accuracy": 0.1848793312907219,
      "num_tokens": 75862747.0,
      "step": 34630
    },
    {
      "entropy": 5.614938974380493,
      "epoch": 3.437375942834458,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003851352548022563,
      "loss": 4.9769,
      "mean_token_accuracy": 0.1875013679265976,
      "num_tokens": 75872807.0,
      "step": 34635
    },
    {
      "entropy": 5.668186092376709,
      "epoch": 3.437872171496626,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003851043568569463,
      "loss": 5.0052,
      "mean_token_accuracy": 0.18870173543691635,
      "num_tokens": 75883630.0,
      "step": 34640
    },
    {
      "entropy": 5.70655255317688,
      "epoch": 3.4383684001587933,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003850734561812855,
      "loss": 5.024,
      "mean_token_accuracy": 0.18510021269321442,
      "num_tokens": 75895147.0,
      "step": 34645
    },
    {
      "entropy": 5.654624605178833,
      "epoch": 3.438864628820961,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038504255277604016,
      "loss": 5.004,
      "mean_token_accuracy": 0.18837653398513793,
      "num_tokens": 75906685.0,
      "step": 34650
    },
    {
      "entropy": 5.561556768417359,
      "epoch": 3.4393608574831283,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003850116466419766,
      "loss": 4.883,
      "mean_token_accuracy": 0.19461387842893602,
      "num_tokens": 75916859.0,
      "step": 34655
    },
    {
      "entropy": 5.660220670700073,
      "epoch": 3.4398570861452957,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003849807377798614,
      "loss": 5.0966,
      "mean_token_accuracy": 0.18447185009717942,
      "num_tokens": 75927138.0,
      "step": 34660
    },
    {
      "entropy": 5.6396434783935545,
      "epoch": 3.440353314807463,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038494982619046076,
      "loss": 4.9817,
      "mean_token_accuracy": 0.18407962024211882,
      "num_tokens": 75939542.0,
      "step": 34665
    },
    {
      "entropy": 5.661172819137573,
      "epoch": 3.4408495434696307,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003849189118745414,
      "loss": 5.0509,
      "mean_token_accuracy": 0.18793898820877075,
      "num_tokens": 75951252.0,
      "step": 34670
    },
    {
      "entropy": 5.673519420623779,
      "epoch": 3.441345772131798,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003848879948328699,
      "loss": 4.9977,
      "mean_token_accuracy": 0.18211790472269057,
      "num_tokens": 75962440.0,
      "step": 34675
    },
    {
      "entropy": 5.6693939685821535,
      "epoch": 3.441842000793966,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003848570750662129,
      "loss": 4.935,
      "mean_token_accuracy": 0.1895074099302292,
      "num_tokens": 75971177.0,
      "step": 34680
    },
    {
      "entropy": 5.587520933151245,
      "epoch": 3.4423382294561335,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003848261525753371,
      "loss": 5.0533,
      "mean_token_accuracy": 0.18579520881175995,
      "num_tokens": 75983112.0,
      "step": 34685
    },
    {
      "entropy": 5.6607099056243895,
      "epoch": 3.442834458118301,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003847952273610094,
      "loss": 5.003,
      "mean_token_accuracy": 0.18953387290239335,
      "num_tokens": 75994347.0,
      "step": 34690
    },
    {
      "entropy": 5.679458570480347,
      "epoch": 3.4433306867804685,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003847642994239968,
      "loss": 4.9779,
      "mean_token_accuracy": 0.1922879472374916,
      "num_tokens": 76004712.0,
      "step": 34695
    },
    {
      "entropy": 5.680532598495484,
      "epoch": 3.443826915442636,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038473336876506585,
      "loss": 5.0473,
      "mean_token_accuracy": 0.18247615844011306,
      "num_tokens": 76014586.0,
      "step": 34700
    },
    {
      "entropy": 5.668140316009522,
      "epoch": 3.4443231441048034,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038470243538498393,
      "loss": 4.9596,
      "mean_token_accuracy": 0.1941642940044403,
      "num_tokens": 76024132.0,
      "step": 34705
    },
    {
      "entropy": 5.721749019622803,
      "epoch": 3.444819372766971,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038467149928451803,
      "loss": 5.0311,
      "mean_token_accuracy": 0.1801780119538307,
      "num_tokens": 76035122.0,
      "step": 34710
    },
    {
      "entropy": 5.671889066696167,
      "epoch": 3.4453156014291384,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00038464056046443523,
      "loss": 5.075,
      "mean_token_accuracy": 0.183881913125515,
      "num_tokens": 76047027.0,
      "step": 34715
    },
    {
      "entropy": 5.585834741592407,
      "epoch": 3.445811830091306,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003846096189255027,
      "loss": 4.9721,
      "mean_token_accuracy": 0.18557219356298446,
      "num_tokens": 76057743.0,
      "step": 34720
    },
    {
      "entropy": 5.554702520370483,
      "epoch": 3.446308058753474,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038457867466848796,
      "loss": 4.8886,
      "mean_token_accuracy": 0.19531954377889632,
      "num_tokens": 76068433.0,
      "step": 34725
    },
    {
      "entropy": 5.682654619216919,
      "epoch": 3.4468042874156413,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003845477276941582,
      "loss": 5.0191,
      "mean_token_accuracy": 0.18879933506250382,
      "num_tokens": 76080681.0,
      "step": 34730
    },
    {
      "entropy": 5.6820775985717775,
      "epoch": 3.4473005160778087,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003845167780032807,
      "loss": 5.0663,
      "mean_token_accuracy": 0.1800086960196495,
      "num_tokens": 76092584.0,
      "step": 34735
    },
    {
      "entropy": 5.596852111816406,
      "epoch": 3.447796744739976,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038448582559662324,
      "loss": 4.9227,
      "mean_token_accuracy": 0.19064564555883406,
      "num_tokens": 76103629.0,
      "step": 34740
    },
    {
      "entropy": 5.617346334457397,
      "epoch": 3.4482929734021437,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038445487047495313,
      "loss": 4.9753,
      "mean_token_accuracy": 0.19088122397661209,
      "num_tokens": 76114069.0,
      "step": 34745
    },
    {
      "entropy": 5.639846754074097,
      "epoch": 3.448789202064311,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003844239126390382,
      "loss": 5.0303,
      "mean_token_accuracy": 0.19263315945863724,
      "num_tokens": 76125896.0,
      "step": 34750
    },
    {
      "entropy": 5.652878475189209,
      "epoch": 3.4492854307264786,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038439295208964595,
      "loss": 4.9921,
      "mean_token_accuracy": 0.1901623785495758,
      "num_tokens": 76137066.0,
      "step": 34755
    },
    {
      "entropy": 5.640574884414673,
      "epoch": 3.449781659388646,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003843619888275442,
      "loss": 4.9221,
      "mean_token_accuracy": 0.19792546480894088,
      "num_tokens": 76148778.0,
      "step": 34760
    },
    {
      "entropy": 5.651046657562256,
      "epoch": 3.450277888050814,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038433102285350074,
      "loss": 5.0101,
      "mean_token_accuracy": 0.18026566356420518,
      "num_tokens": 76158760.0,
      "step": 34765
    },
    {
      "entropy": 5.636438989639283,
      "epoch": 3.4507741167129815,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038430005416828354,
      "loss": 4.9407,
      "mean_token_accuracy": 0.19713598042726516,
      "num_tokens": 76169459.0,
      "step": 34770
    },
    {
      "entropy": 5.648621702194214,
      "epoch": 3.451270345375149,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038426908277266054,
      "loss": 5.0083,
      "mean_token_accuracy": 0.18880808353424072,
      "num_tokens": 76179074.0,
      "step": 34775
    },
    {
      "entropy": 5.666333961486816,
      "epoch": 3.4517665740373165,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038423810866739975,
      "loss": 5.0043,
      "mean_token_accuracy": 0.19353010654449462,
      "num_tokens": 76189470.0,
      "step": 34780
    },
    {
      "entropy": 5.641579484939575,
      "epoch": 3.452262802699484,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038420713185326926,
      "loss": 4.9727,
      "mean_token_accuracy": 0.18775412142276765,
      "num_tokens": 76200427.0,
      "step": 34785
    },
    {
      "entropy": 5.654681444168091,
      "epoch": 3.4527590313616514,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003841761523310372,
      "loss": 5.0074,
      "mean_token_accuracy": 0.19047963619232178,
      "num_tokens": 76211132.0,
      "step": 34790
    },
    {
      "entropy": 5.654520702362061,
      "epoch": 3.453255260023819,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003841451701014718,
      "loss": 5.01,
      "mean_token_accuracy": 0.1855367675423622,
      "num_tokens": 76222661.0,
      "step": 34795
    },
    {
      "entropy": 5.68079833984375,
      "epoch": 3.4537514886859864,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038411418516534135,
      "loss": 4.9984,
      "mean_token_accuracy": 0.1924520805478096,
      "num_tokens": 76232727.0,
      "step": 34800
    },
    {
      "entropy": 5.694387340545655,
      "epoch": 3.454247717348154,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038408319752341425,
      "loss": 5.0197,
      "mean_token_accuracy": 0.1929636389017105,
      "num_tokens": 76242783.0,
      "step": 34805
    },
    {
      "entropy": 5.586372184753418,
      "epoch": 3.4547439460103213,
      "grad_norm": 1.140625,
      "learning_rate": 0.000384052207176459,
      "loss": 4.8945,
      "mean_token_accuracy": 0.18573508709669112,
      "num_tokens": 76254223.0,
      "step": 34810
    },
    {
      "entropy": 5.662993764877319,
      "epoch": 3.4552401746724892,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003840212141252438,
      "loss": 5.0178,
      "mean_token_accuracy": 0.1876671642065048,
      "num_tokens": 76265094.0,
      "step": 34815
    },
    {
      "entropy": 5.6445060729980465,
      "epoch": 3.4557364033346567,
      "grad_norm": 1.125,
      "learning_rate": 0.00038399021837053754,
      "loss": 4.9897,
      "mean_token_accuracy": 0.18479665219783784,
      "num_tokens": 76276722.0,
      "step": 34820
    },
    {
      "entropy": 5.627242469787598,
      "epoch": 3.456232631996824,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003839592199131088,
      "loss": 4.9517,
      "mean_token_accuracy": 0.1877701535820961,
      "num_tokens": 76288027.0,
      "step": 34825
    },
    {
      "entropy": 5.648294305801391,
      "epoch": 3.4567288606589917,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038392821875372603,
      "loss": 4.9763,
      "mean_token_accuracy": 0.20131542831659316,
      "num_tokens": 76299147.0,
      "step": 34830
    },
    {
      "entropy": 5.674335241317749,
      "epoch": 3.457225089321159,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003838972148931582,
      "loss": 5.11,
      "mean_token_accuracy": 0.18243689984083175,
      "num_tokens": 76309375.0,
      "step": 34835
    },
    {
      "entropy": 5.638803339004516,
      "epoch": 3.4577213179833266,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003838662083321741,
      "loss": 4.9265,
      "mean_token_accuracy": 0.1894189015030861,
      "num_tokens": 76319953.0,
      "step": 34840
    },
    {
      "entropy": 5.7024730205535885,
      "epoch": 3.458217546645494,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038383519907154263,
      "loss": 5.0106,
      "mean_token_accuracy": 0.19059733748435975,
      "num_tokens": 76331525.0,
      "step": 34845
    },
    {
      "entropy": 5.591845846176147,
      "epoch": 3.458713775307662,
      "grad_norm": 1.125,
      "learning_rate": 0.00038380418711203275,
      "loss": 4.9202,
      "mean_token_accuracy": 0.20279544591903687,
      "num_tokens": 76343290.0,
      "step": 34850
    },
    {
      "entropy": 5.690984678268433,
      "epoch": 3.4592100039698295,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038377317245441345,
      "loss": 5.1285,
      "mean_token_accuracy": 0.1815943941473961,
      "num_tokens": 76354993.0,
      "step": 34855
    },
    {
      "entropy": 5.677311372756958,
      "epoch": 3.459706232631997,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038374215509945386,
      "loss": 4.9953,
      "mean_token_accuracy": 0.18823181688785554,
      "num_tokens": 76365206.0,
      "step": 34860
    },
    {
      "entropy": 5.6562896251678465,
      "epoch": 3.4602024612941644,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003837111350479232,
      "loss": 5.0268,
      "mean_token_accuracy": 0.18175290822982787,
      "num_tokens": 76376549.0,
      "step": 34865
    },
    {
      "entropy": 5.627677011489868,
      "epoch": 3.460698689956332,
      "grad_norm": 1.125,
      "learning_rate": 0.00038368011230059055,
      "loss": 4.9624,
      "mean_token_accuracy": 0.1856428414583206,
      "num_tokens": 76387654.0,
      "step": 34870
    },
    {
      "entropy": 5.748971366882325,
      "epoch": 3.4611949186184994,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038364908685822543,
      "loss": 5.0315,
      "mean_token_accuracy": 0.18635524064302444,
      "num_tokens": 76399462.0,
      "step": 34875
    },
    {
      "entropy": 5.664646053314209,
      "epoch": 3.461691147280667,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038361805872159693,
      "loss": 4.9916,
      "mean_token_accuracy": 0.19193087369203568,
      "num_tokens": 76411394.0,
      "step": 34880
    },
    {
      "entropy": 5.663694667816162,
      "epoch": 3.4621873759428343,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003835870278914747,
      "loss": 5.0422,
      "mean_token_accuracy": 0.18223525136709212,
      "num_tokens": 76423251.0,
      "step": 34885
    },
    {
      "entropy": 5.6662369728088375,
      "epoch": 3.462683604605002,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003835559943686282,
      "loss": 4.9377,
      "mean_token_accuracy": 0.19288402944803237,
      "num_tokens": 76433828.0,
      "step": 34890
    },
    {
      "entropy": 5.619930219650269,
      "epoch": 3.4631798332671693,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003835249581538269,
      "loss": 4.9167,
      "mean_token_accuracy": 0.19965933561325072,
      "num_tokens": 76445006.0,
      "step": 34895
    },
    {
      "entropy": 5.558235740661621,
      "epoch": 3.463676061929337,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038349391924784055,
      "loss": 4.9802,
      "mean_token_accuracy": 0.18766194581985474,
      "num_tokens": 76456296.0,
      "step": 34900
    },
    {
      "entropy": 5.579535055160522,
      "epoch": 3.4641722905915047,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038346287765143866,
      "loss": 4.9271,
      "mean_token_accuracy": 0.18911176770925522,
      "num_tokens": 76467330.0,
      "step": 34905
    },
    {
      "entropy": 5.74995641708374,
      "epoch": 3.464668519253672,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038343183336539123,
      "loss": 5.0616,
      "mean_token_accuracy": 0.18081556260585785,
      "num_tokens": 76478393.0,
      "step": 34910
    },
    {
      "entropy": 5.670882749557495,
      "epoch": 3.4651647479158396,
      "grad_norm": 1.0625,
      "learning_rate": 0.000383400786390468,
      "loss": 4.9667,
      "mean_token_accuracy": 0.1945633441209793,
      "num_tokens": 76489695.0,
      "step": 34915
    },
    {
      "entropy": 5.605261278152466,
      "epoch": 3.465660976578007,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038336973672743876,
      "loss": 4.986,
      "mean_token_accuracy": 0.18393988460302352,
      "num_tokens": 76500831.0,
      "step": 34920
    },
    {
      "entropy": 5.580028629302978,
      "epoch": 3.4661572052401746,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038333868437707367,
      "loss": 4.9472,
      "mean_token_accuracy": 0.19450054317712784,
      "num_tokens": 76511054.0,
      "step": 34925
    },
    {
      "entropy": 5.6436083793640135,
      "epoch": 3.466653433902342,
      "grad_norm": 1.125,
      "learning_rate": 0.0003833076293401425,
      "loss": 4.9987,
      "mean_token_accuracy": 0.18474362045526505,
      "num_tokens": 76522948.0,
      "step": 34930
    },
    {
      "entropy": 5.596654081344605,
      "epoch": 3.46714966256451,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003832765716174156,
      "loss": 4.9718,
      "mean_token_accuracy": 0.1934824541211128,
      "num_tokens": 76533827.0,
      "step": 34935
    },
    {
      "entropy": 5.725051832199097,
      "epoch": 3.4676458912266774,
      "grad_norm": 1.125,
      "learning_rate": 0.0003832455112096631,
      "loss": 5.1007,
      "mean_token_accuracy": 0.18964487612247466,
      "num_tokens": 76545252.0,
      "step": 34940
    },
    {
      "entropy": 5.6590300559997555,
      "epoch": 3.468142119888845,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003832144481176551,
      "loss": 4.9483,
      "mean_token_accuracy": 0.19375309050083162,
      "num_tokens": 76555774.0,
      "step": 34945
    },
    {
      "entropy": 5.648236179351807,
      "epoch": 3.4686383485510124,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038318338234216194,
      "loss": 4.9694,
      "mean_token_accuracy": 0.19054309129714966,
      "num_tokens": 76566403.0,
      "step": 34950
    },
    {
      "entropy": 5.636171483993531,
      "epoch": 3.46913457721318,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00038315231388395403,
      "loss": 5.007,
      "mean_token_accuracy": 0.18972312211990355,
      "num_tokens": 76578938.0,
      "step": 34955
    },
    {
      "entropy": 5.668113803863525,
      "epoch": 3.4696308058753473,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003831212427438018,
      "loss": 5.0364,
      "mean_token_accuracy": 0.18828985691070557,
      "num_tokens": 76589355.0,
      "step": 34960
    },
    {
      "entropy": 5.613582181930542,
      "epoch": 3.470127034537515,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003830901689224757,
      "loss": 4.943,
      "mean_token_accuracy": 0.19002453684806825,
      "num_tokens": 76601674.0,
      "step": 34965
    },
    {
      "entropy": 5.695480728149414,
      "epoch": 3.4706232631996823,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038305909242074645,
      "loss": 4.982,
      "mean_token_accuracy": 0.19399910271167756,
      "num_tokens": 76612814.0,
      "step": 34970
    },
    {
      "entropy": 5.608388662338257,
      "epoch": 3.4711194918618498,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003830280132393844,
      "loss": 4.9661,
      "mean_token_accuracy": 0.1868202194571495,
      "num_tokens": 76623177.0,
      "step": 34975
    },
    {
      "entropy": 5.640443801879883,
      "epoch": 3.4716157205240172,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038299693137916046,
      "loss": 4.9992,
      "mean_token_accuracy": 0.18781937509775162,
      "num_tokens": 76633286.0,
      "step": 34980
    },
    {
      "entropy": 5.708603000640869,
      "epoch": 3.472111949186185,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038296584684084536,
      "loss": 5.0297,
      "mean_token_accuracy": 0.19171081185340882,
      "num_tokens": 76644851.0,
      "step": 34985
    },
    {
      "entropy": 5.631404304504395,
      "epoch": 3.4726081778483526,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003829347596252099,
      "loss": 4.9505,
      "mean_token_accuracy": 0.18813204467296601,
      "num_tokens": 76655165.0,
      "step": 34990
    },
    {
      "entropy": 5.580813837051392,
      "epoch": 3.47310440651052,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038290366973302493,
      "loss": 4.9697,
      "mean_token_accuracy": 0.19170617759227754,
      "num_tokens": 76666282.0,
      "step": 34995
    },
    {
      "entropy": 5.6519406795501705,
      "epoch": 3.4736006351726876,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038287257716506154,
      "loss": 5.0011,
      "mean_token_accuracy": 0.18706996440887452,
      "num_tokens": 76675600.0,
      "step": 35000
    },
    {
      "entropy": 5.664982318878174,
      "epoch": 3.474096863834855,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038284148192209064,
      "loss": 4.9855,
      "mean_token_accuracy": 0.18891624212265015,
      "num_tokens": 76687412.0,
      "step": 35005
    },
    {
      "entropy": 5.705160236358642,
      "epoch": 3.4745930924970225,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038281038400488326,
      "loss": 5.0471,
      "mean_token_accuracy": 0.18125455677509308,
      "num_tokens": 76698799.0,
      "step": 35010
    },
    {
      "entropy": 5.679745769500732,
      "epoch": 3.47508932115919,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038277928341421086,
      "loss": 5.0128,
      "mean_token_accuracy": 0.18041837364435195,
      "num_tokens": 76709672.0,
      "step": 35015
    },
    {
      "entropy": 5.63100757598877,
      "epoch": 3.475585549821358,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003827481801508444,
      "loss": 4.9732,
      "mean_token_accuracy": 0.18698426634073256,
      "num_tokens": 76720641.0,
      "step": 35020
    },
    {
      "entropy": 5.610593557357788,
      "epoch": 3.4760817784835254,
      "grad_norm": 1.125,
      "learning_rate": 0.00038271707421555517,
      "loss": 4.9248,
      "mean_token_accuracy": 0.19019225090742112,
      "num_tokens": 76731173.0,
      "step": 35025
    },
    {
      "entropy": 5.694745779037476,
      "epoch": 3.476578007145693,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003826859656091147,
      "loss": 5.0415,
      "mean_token_accuracy": 0.18390370160341263,
      "num_tokens": 76742801.0,
      "step": 35030
    },
    {
      "entropy": 5.6957169532775875,
      "epoch": 3.4770742358078603,
      "grad_norm": 1.125,
      "learning_rate": 0.0003826548543322943,
      "loss": 4.9657,
      "mean_token_accuracy": 0.1895841673016548,
      "num_tokens": 76753088.0,
      "step": 35035
    },
    {
      "entropy": 5.650423240661621,
      "epoch": 3.477570464470028,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003826237403858654,
      "loss": 5.0232,
      "mean_token_accuracy": 0.1920834019780159,
      "num_tokens": 76765715.0,
      "step": 35040
    },
    {
      "entropy": 5.703303861618042,
      "epoch": 3.4780666931321953,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003825926237705998,
      "loss": 5.0255,
      "mean_token_accuracy": 0.18634145110845565,
      "num_tokens": 76775626.0,
      "step": 35045
    },
    {
      "entropy": 5.635354518890381,
      "epoch": 3.4785629217943628,
      "grad_norm": 1.109375,
      "learning_rate": 0.000382561504487269,
      "loss": 4.9002,
      "mean_token_accuracy": 0.19007923901081086,
      "num_tokens": 76785357.0,
      "step": 35050
    },
    {
      "entropy": 5.636249017715454,
      "epoch": 3.4790591504565302,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038253038253664446,
      "loss": 4.9965,
      "mean_token_accuracy": 0.18809915632009505,
      "num_tokens": 76797553.0,
      "step": 35055
    },
    {
      "entropy": 5.591836595535279,
      "epoch": 3.4795553791186977,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038249925791949827,
      "loss": 4.9548,
      "mean_token_accuracy": 0.19700469374656676,
      "num_tokens": 76807564.0,
      "step": 35060
    },
    {
      "entropy": 5.671630620956421,
      "epoch": 3.480051607780865,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003824681306366021,
      "loss": 5.0162,
      "mean_token_accuracy": 0.18760032951831818,
      "num_tokens": 76818012.0,
      "step": 35065
    },
    {
      "entropy": 5.649958753585816,
      "epoch": 3.480547836443033,
      "grad_norm": 1.25,
      "learning_rate": 0.00038243700068872784,
      "loss": 4.993,
      "mean_token_accuracy": 0.19316240698099135,
      "num_tokens": 76828552.0,
      "step": 35070
    },
    {
      "entropy": 5.730821228027343,
      "epoch": 3.4810440651052006,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038240586807664745,
      "loss": 5.0141,
      "mean_token_accuracy": 0.1932402268052101,
      "num_tokens": 76839652.0,
      "step": 35075
    },
    {
      "entropy": 5.641420364379883,
      "epoch": 3.481540293767368,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000382374732801133,
      "loss": 4.9596,
      "mean_token_accuracy": 0.19201037287712097,
      "num_tokens": 76851563.0,
      "step": 35080
    },
    {
      "entropy": 5.648260450363159,
      "epoch": 3.4820365224295355,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038234359486295663,
      "loss": 4.9336,
      "mean_token_accuracy": 0.18890601247549058,
      "num_tokens": 76862769.0,
      "step": 35085
    },
    {
      "entropy": 5.682418870925903,
      "epoch": 3.482532751091703,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003823124542628904,
      "loss": 5.0098,
      "mean_token_accuracy": 0.18806844651699067,
      "num_tokens": 76873447.0,
      "step": 35090
    },
    {
      "entropy": 5.606055688858032,
      "epoch": 3.4830289797538705,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038228131100170637,
      "loss": 4.9817,
      "mean_token_accuracy": 0.19094177782535554,
      "num_tokens": 76884187.0,
      "step": 35095
    },
    {
      "entropy": 5.6564507484436035,
      "epoch": 3.483525208416038,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038225016508017717,
      "loss": 4.9744,
      "mean_token_accuracy": 0.19204046577215195,
      "num_tokens": 76895592.0,
      "step": 35100
    },
    {
      "entropy": 5.710466814041138,
      "epoch": 3.484021437078206,
      "grad_norm": 1.09375,
      "learning_rate": 0.000382219016499075,
      "loss": 5.0385,
      "mean_token_accuracy": 0.18559669852256774,
      "num_tokens": 76907257.0,
      "step": 35105
    },
    {
      "entropy": 5.702847814559936,
      "epoch": 3.4845176657403734,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038218786525917207,
      "loss": 5.0377,
      "mean_token_accuracy": 0.18645820021629333,
      "num_tokens": 76917926.0,
      "step": 35110
    },
    {
      "entropy": 5.637886190414429,
      "epoch": 3.485013894402541,
      "grad_norm": 1.125,
      "learning_rate": 0.0003821567113612412,
      "loss": 4.9753,
      "mean_token_accuracy": 0.18984103947877884,
      "num_tokens": 76929060.0,
      "step": 35115
    },
    {
      "entropy": 5.578448677062989,
      "epoch": 3.4855101230647083,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038212555480605475,
      "loss": 4.9424,
      "mean_token_accuracy": 0.1895960494875908,
      "num_tokens": 76939694.0,
      "step": 35120
    },
    {
      "entropy": 5.636431789398193,
      "epoch": 3.486006351726876,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038209439559438535,
      "loss": 4.9447,
      "mean_token_accuracy": 0.19820937663316726,
      "num_tokens": 76951015.0,
      "step": 35125
    },
    {
      "entropy": 5.641475057601928,
      "epoch": 3.4865025803890433,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003820632337270057,
      "loss": 5.0281,
      "mean_token_accuracy": 0.18758825212717056,
      "num_tokens": 76962606.0,
      "step": 35130
    },
    {
      "entropy": 5.660440874099732,
      "epoch": 3.4869988090512107,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038203206920468867,
      "loss": 4.9934,
      "mean_token_accuracy": 0.18890464454889297,
      "num_tokens": 76972957.0,
      "step": 35135
    },
    {
      "entropy": 5.607121992111206,
      "epoch": 3.487495037713378,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038200090202820686,
      "loss": 4.9553,
      "mean_token_accuracy": 0.19582766443490982,
      "num_tokens": 76983396.0,
      "step": 35140
    },
    {
      "entropy": 5.594200658798218,
      "epoch": 3.4879912663755457,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038196973219833323,
      "loss": 4.9287,
      "mean_token_accuracy": 0.1897799253463745,
      "num_tokens": 76994340.0,
      "step": 35145
    },
    {
      "entropy": 5.6520201683044435,
      "epoch": 3.488487495037713,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038193855971584086,
      "loss": 4.9776,
      "mean_token_accuracy": 0.18831322491168975,
      "num_tokens": 77004723.0,
      "step": 35150
    },
    {
      "entropy": 5.639257001876831,
      "epoch": 3.488983723699881,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038190738458150256,
      "loss": 5.0602,
      "mean_token_accuracy": 0.18058548867702484,
      "num_tokens": 77017031.0,
      "step": 35155
    },
    {
      "entropy": 5.655799341201782,
      "epoch": 3.4894799523620486,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038187620679609144,
      "loss": 4.949,
      "mean_token_accuracy": 0.19243387877941132,
      "num_tokens": 77028629.0,
      "step": 35160
    },
    {
      "entropy": 5.658856296539307,
      "epoch": 3.489976181024216,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038184502636038074,
      "loss": 5.0098,
      "mean_token_accuracy": 0.1917240262031555,
      "num_tokens": 77037811.0,
      "step": 35165
    },
    {
      "entropy": 5.731548309326172,
      "epoch": 3.4904724096863835,
      "grad_norm": 1.125,
      "learning_rate": 0.0003818138432751436,
      "loss": 5.128,
      "mean_token_accuracy": 0.18229179978370666,
      "num_tokens": 77049213.0,
      "step": 35170
    },
    {
      "entropy": 5.699782466888427,
      "epoch": 3.490968638348551,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003817826575411533,
      "loss": 4.9993,
      "mean_token_accuracy": 0.18490667045116424,
      "num_tokens": 77059090.0,
      "step": 35175
    },
    {
      "entropy": 5.662880945205688,
      "epoch": 3.4914648670107185,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038175146915918323,
      "loss": 4.9669,
      "mean_token_accuracy": 0.19029232561588288,
      "num_tokens": 77070092.0,
      "step": 35180
    },
    {
      "entropy": 5.669156885147094,
      "epoch": 3.491961095672886,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038172027813000676,
      "loss": 4.9991,
      "mean_token_accuracy": 0.18933751732110976,
      "num_tokens": 77079808.0,
      "step": 35185
    },
    {
      "entropy": 5.619491147994995,
      "epoch": 3.4924573243350534,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038168908445439725,
      "loss": 4.9763,
      "mean_token_accuracy": 0.187417109310627,
      "num_tokens": 77090772.0,
      "step": 35190
    },
    {
      "entropy": 5.6175142288208,
      "epoch": 3.4929535529972213,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038165788813312844,
      "loss": 4.9791,
      "mean_token_accuracy": 0.1929272085428238,
      "num_tokens": 77101358.0,
      "step": 35195
    },
    {
      "entropy": 5.621081733703614,
      "epoch": 3.493449781659389,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038162668916697385,
      "loss": 4.9141,
      "mean_token_accuracy": 0.19603862911462783,
      "num_tokens": 77111198.0,
      "step": 35200
    },
    {
      "entropy": 5.65515718460083,
      "epoch": 3.4939460103215563,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038159548755670707,
      "loss": 5.057,
      "mean_token_accuracy": 0.18909060657024385,
      "num_tokens": 77121990.0,
      "step": 35205
    },
    {
      "entropy": 5.68134298324585,
      "epoch": 3.4944422389837237,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038156428330310193,
      "loss": 5.0247,
      "mean_token_accuracy": 0.18230685740709304,
      "num_tokens": 77132952.0,
      "step": 35210
    },
    {
      "entropy": 5.651999235153198,
      "epoch": 3.4949384676458912,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003815330764069322,
      "loss": 4.9801,
      "mean_token_accuracy": 0.1881551295518875,
      "num_tokens": 77142870.0,
      "step": 35215
    },
    {
      "entropy": 5.69605975151062,
      "epoch": 3.4954346963080587,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003815018668689717,
      "loss": 5.0787,
      "mean_token_accuracy": 0.18531868904829024,
      "num_tokens": 77154180.0,
      "step": 35220
    },
    {
      "entropy": 5.693450546264648,
      "epoch": 3.495930924970226,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038147065468999435,
      "loss": 5.0633,
      "mean_token_accuracy": 0.18794459849596024,
      "num_tokens": 77165083.0,
      "step": 35225
    },
    {
      "entropy": 5.734307670593262,
      "epoch": 3.4964271536323936,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038143943987077426,
      "loss": 5.0185,
      "mean_token_accuracy": 0.18158242702484131,
      "num_tokens": 77176809.0,
      "step": 35230
    },
    {
      "entropy": 5.634205722808838,
      "epoch": 3.496923382294561,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003814082224120854,
      "loss": 4.9653,
      "mean_token_accuracy": 0.18496477901935576,
      "num_tokens": 77187325.0,
      "step": 35235
    },
    {
      "entropy": 5.732226085662842,
      "epoch": 3.497419610956729,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003813770023147019,
      "loss": 5.0342,
      "mean_token_accuracy": 0.18605092763900757,
      "num_tokens": 77197403.0,
      "step": 35240
    },
    {
      "entropy": 5.626200532913208,
      "epoch": 3.4979158396188965,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038134577957939795,
      "loss": 4.9708,
      "mean_token_accuracy": 0.19037729054689406,
      "num_tokens": 77208262.0,
      "step": 35245
    },
    {
      "entropy": 5.661021280288696,
      "epoch": 3.498412068281064,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038131455420694786,
      "loss": 4.9875,
      "mean_token_accuracy": 0.19778242260217666,
      "num_tokens": 77219327.0,
      "step": 35250
    },
    {
      "entropy": 5.675337505340576,
      "epoch": 3.4989082969432315,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003812833261981259,
      "loss": 5.0959,
      "mean_token_accuracy": 0.18119840323925018,
      "num_tokens": 77230380.0,
      "step": 35255
    },
    {
      "entropy": 5.6647594451904295,
      "epoch": 3.499404525605399,
      "grad_norm": 1.125,
      "learning_rate": 0.00038125209555370644,
      "loss": 4.9499,
      "mean_token_accuracy": 0.1928519070148468,
      "num_tokens": 77241173.0,
      "step": 35260
    },
    {
      "entropy": 5.645073080062867,
      "epoch": 3.4999007542675664,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000381220862274464,
      "loss": 4.9926,
      "mean_token_accuracy": 0.18459028154611587,
      "num_tokens": 77252674.0,
      "step": 35265
    },
    {
      "entropy": 5.639178466796875,
      "epoch": 3.500396982929734,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038118962636117304,
      "loss": 5.014,
      "mean_token_accuracy": 0.19360782504081725,
      "num_tokens": 77264826.0,
      "step": 35270
    },
    {
      "entropy": 5.679399681091309,
      "epoch": 3.500893211591902,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038115838781460816,
      "loss": 4.9935,
      "mean_token_accuracy": 0.19091441631317138,
      "num_tokens": 77276068.0,
      "step": 35275
    },
    {
      "entropy": 5.679148483276367,
      "epoch": 3.5013894402540693,
      "grad_norm": 1.21875,
      "learning_rate": 0.000381127146635544,
      "loss": 5.0457,
      "mean_token_accuracy": 0.1822613298892975,
      "num_tokens": 77285394.0,
      "step": 35280
    },
    {
      "entropy": 5.662808084487915,
      "epoch": 3.5018856689162368,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003810959028247554,
      "loss": 5.005,
      "mean_token_accuracy": 0.18366749584674835,
      "num_tokens": 77295843.0,
      "step": 35285
    },
    {
      "entropy": 5.715687561035156,
      "epoch": 3.5023818975784042,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003810646563830169,
      "loss": 5.0513,
      "mean_token_accuracy": 0.18288903385400773,
      "num_tokens": 77306064.0,
      "step": 35290
    },
    {
      "entropy": 5.682353401184082,
      "epoch": 3.5028781262405717,
      "grad_norm": 1.25,
      "learning_rate": 0.00038103340731110354,
      "loss": 4.9452,
      "mean_token_accuracy": 0.19003702998161315,
      "num_tokens": 77316425.0,
      "step": 35295
    },
    {
      "entropy": 5.807770729064941,
      "epoch": 3.503374354902739,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038100215560979016,
      "loss": 5.1563,
      "mean_token_accuracy": 0.17238960266113282,
      "num_tokens": 77327185.0,
      "step": 35300
    },
    {
      "entropy": 5.668176078796387,
      "epoch": 3.5038705835649067,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038097090127985165,
      "loss": 5.0385,
      "mean_token_accuracy": 0.18248480409383774,
      "num_tokens": 77338791.0,
      "step": 35305
    },
    {
      "entropy": 5.652017211914062,
      "epoch": 3.504366812227074,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038093964432206316,
      "loss": 4.9736,
      "mean_token_accuracy": 0.18801961690187455,
      "num_tokens": 77350035.0,
      "step": 35310
    },
    {
      "entropy": 5.660821580886841,
      "epoch": 3.5048630408892416,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003809083847371999,
      "loss": 5.0126,
      "mean_token_accuracy": 0.1867121011018753,
      "num_tokens": 77362422.0,
      "step": 35315
    },
    {
      "entropy": 5.631702518463134,
      "epoch": 3.505359269551409,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003808771225260368,
      "loss": 4.9174,
      "mean_token_accuracy": 0.18957474380731582,
      "num_tokens": 77373383.0,
      "step": 35320
    },
    {
      "entropy": 5.636592626571655,
      "epoch": 3.5058554982135766,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038084585768934917,
      "loss": 5.0158,
      "mean_token_accuracy": 0.18696541488170623,
      "num_tokens": 77383782.0,
      "step": 35325
    },
    {
      "entropy": 5.612462663650513,
      "epoch": 3.5063517268757445,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038081459022791243,
      "loss": 4.977,
      "mean_token_accuracy": 0.18940356969833375,
      "num_tokens": 77394266.0,
      "step": 35330
    },
    {
      "entropy": 5.663697099685669,
      "epoch": 3.506847955537912,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038078332014250185,
      "loss": 4.9307,
      "mean_token_accuracy": 0.1900399997830391,
      "num_tokens": 77405042.0,
      "step": 35335
    },
    {
      "entropy": 5.669951009750366,
      "epoch": 3.5073441842000794,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003807520474338928,
      "loss": 4.9852,
      "mean_token_accuracy": 0.19082293808460235,
      "num_tokens": 77415411.0,
      "step": 35340
    },
    {
      "entropy": 5.634488248825074,
      "epoch": 3.507840412862247,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003807207721028609,
      "loss": 4.9278,
      "mean_token_accuracy": 0.19766213595867158,
      "num_tokens": 77423991.0,
      "step": 35345
    },
    {
      "entropy": 5.636547470092774,
      "epoch": 3.5083366415244144,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003806894941501817,
      "loss": 4.9923,
      "mean_token_accuracy": 0.1879263386130333,
      "num_tokens": 77435701.0,
      "step": 35350
    },
    {
      "entropy": 5.631517457962036,
      "epoch": 3.508832870186582,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003806582135766307,
      "loss": 4.9049,
      "mean_token_accuracy": 0.1936315968632698,
      "num_tokens": 77445618.0,
      "step": 35355
    },
    {
      "entropy": 5.637457704544067,
      "epoch": 3.5093290988487498,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038062693038298367,
      "loss": 4.9339,
      "mean_token_accuracy": 0.18817447274923324,
      "num_tokens": 77456674.0,
      "step": 35360
    },
    {
      "entropy": 5.567584562301636,
      "epoch": 3.5098253275109172,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003805956445700165,
      "loss": 4.9801,
      "mean_token_accuracy": 0.18784230947494507,
      "num_tokens": 77467841.0,
      "step": 35365
    },
    {
      "entropy": 5.624280405044556,
      "epoch": 3.5103215561730847,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003805643561385048,
      "loss": 4.8584,
      "mean_token_accuracy": 0.19349438697099686,
      "num_tokens": 77477883.0,
      "step": 35370
    },
    {
      "entropy": 5.5785047054290775,
      "epoch": 3.510817784835252,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038053306508922455,
      "loss": 4.9116,
      "mean_token_accuracy": 0.1928410455584526,
      "num_tokens": 77488714.0,
      "step": 35375
    },
    {
      "entropy": 5.5920445919036865,
      "epoch": 3.5113140134974197,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038050177142295166,
      "loss": 4.9462,
      "mean_token_accuracy": 0.19209172278642656,
      "num_tokens": 77499325.0,
      "step": 35380
    },
    {
      "entropy": 5.598775243759155,
      "epoch": 3.511810242159587,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038047047514046223,
      "loss": 4.9815,
      "mean_token_accuracy": 0.18175462186336516,
      "num_tokens": 77510682.0,
      "step": 35385
    },
    {
      "entropy": 5.650352859497071,
      "epoch": 3.5123064708217546,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003804391762425322,
      "loss": 4.9755,
      "mean_token_accuracy": 0.18863128274679183,
      "num_tokens": 77520604.0,
      "step": 35390
    },
    {
      "entropy": 5.648537683486938,
      "epoch": 3.512802699483922,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038040787472993793,
      "loss": 5.0221,
      "mean_token_accuracy": 0.18825727701187134,
      "num_tokens": 77532481.0,
      "step": 35395
    },
    {
      "entropy": 5.722338342666626,
      "epoch": 3.5132989281460896,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003803765706034554,
      "loss": 5.0691,
      "mean_token_accuracy": 0.17637378871440887,
      "num_tokens": 77543940.0,
      "step": 35400
    },
    {
      "entropy": 5.701508855819702,
      "epoch": 3.513795156808257,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038034526386386097,
      "loss": 4.9004,
      "mean_token_accuracy": 0.1979759156703949,
      "num_tokens": 77553765.0,
      "step": 35405
    },
    {
      "entropy": 5.674768543243408,
      "epoch": 3.5142913854704245,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038031395451193105,
      "loss": 4.9981,
      "mean_token_accuracy": 0.18668659627437592,
      "num_tokens": 77566132.0,
      "step": 35410
    },
    {
      "entropy": 5.631006240844727,
      "epoch": 3.5147876141325924,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038028264254844204,
      "loss": 4.9495,
      "mean_token_accuracy": 0.19423441588878632,
      "num_tokens": 77577706.0,
      "step": 35415
    },
    {
      "entropy": 5.637039995193481,
      "epoch": 3.51528384279476,
      "grad_norm": 1.25,
      "learning_rate": 0.00038025132797417024,
      "loss": 4.9616,
      "mean_token_accuracy": 0.1894553616642952,
      "num_tokens": 77589113.0,
      "step": 35420
    },
    {
      "entropy": 5.591473007202149,
      "epoch": 3.5157800714569274,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003802200107898924,
      "loss": 4.9225,
      "mean_token_accuracy": 0.1952600136399269,
      "num_tokens": 77599012.0,
      "step": 35425
    },
    {
      "entropy": 5.623215389251709,
      "epoch": 3.516276300119095,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038018869099638506,
      "loss": 5.0216,
      "mean_token_accuracy": 0.18649994879961013,
      "num_tokens": 77610389.0,
      "step": 35430
    },
    {
      "entropy": 5.627383756637573,
      "epoch": 3.5167725287812623,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003801573685944247,
      "loss": 4.9089,
      "mean_token_accuracy": 0.19258689135313034,
      "num_tokens": 77621245.0,
      "step": 35435
    },
    {
      "entropy": 5.665114068984986,
      "epoch": 3.51726875744343,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038012604358478823,
      "loss": 4.9413,
      "mean_token_accuracy": 0.19322377741336821,
      "num_tokens": 77631003.0,
      "step": 35440
    },
    {
      "entropy": 5.5845235824584964,
      "epoch": 3.5177649861055977,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038009471596825244,
      "loss": 4.8852,
      "mean_token_accuracy": 0.19586010724306108,
      "num_tokens": 77641222.0,
      "step": 35445
    },
    {
      "entropy": 5.685783624649048,
      "epoch": 3.518261214767765,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003800633857455942,
      "loss": 5.0348,
      "mean_token_accuracy": 0.1896425351500511,
      "num_tokens": 77652705.0,
      "step": 35450
    },
    {
      "entropy": 5.697826147079468,
      "epoch": 3.5187574434299327,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003800320529175903,
      "loss": 5.0458,
      "mean_token_accuracy": 0.17682048231363295,
      "num_tokens": 77662883.0,
      "step": 35455
    },
    {
      "entropy": 5.634757232666016,
      "epoch": 3.5192536720921,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038000071748501787,
      "loss": 4.9747,
      "mean_token_accuracy": 0.18708537220954896,
      "num_tokens": 77673979.0,
      "step": 35460
    },
    {
      "entropy": 5.648714971542359,
      "epoch": 3.5197499007542676,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037996937944865383,
      "loss": 4.9462,
      "mean_token_accuracy": 0.19068948924541473,
      "num_tokens": 77684387.0,
      "step": 35465
    },
    {
      "entropy": 5.687361764907837,
      "epoch": 3.520246129416435,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003799380388092754,
      "loss": 4.9996,
      "mean_token_accuracy": 0.19023842513561248,
      "num_tokens": 77695218.0,
      "step": 35470
    },
    {
      "entropy": 5.668879985809326,
      "epoch": 3.5207423580786026,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003799066955676598,
      "loss": 5.0035,
      "mean_token_accuracy": 0.1878927692770958,
      "num_tokens": 77705118.0,
      "step": 35475
    },
    {
      "entropy": 5.563799571990967,
      "epoch": 3.52123858674077,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003798753497245841,
      "loss": 4.9118,
      "mean_token_accuracy": 0.19472246915102004,
      "num_tokens": 77715440.0,
      "step": 35480
    },
    {
      "entropy": 5.666107130050659,
      "epoch": 3.5217348154029375,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037984400128082576,
      "loss": 5.0955,
      "mean_token_accuracy": 0.18261535465717316,
      "num_tokens": 77725146.0,
      "step": 35485
    },
    {
      "entropy": 5.757195329666137,
      "epoch": 3.522231044065105,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037981265023716217,
      "loss": 5.021,
      "mean_token_accuracy": 0.18731003105640412,
      "num_tokens": 77735987.0,
      "step": 35490
    },
    {
      "entropy": 5.692268323898316,
      "epoch": 3.5227272727272725,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037978129659437063,
      "loss": 4.937,
      "mean_token_accuracy": 0.1967497557401657,
      "num_tokens": 77747003.0,
      "step": 35495
    },
    {
      "entropy": 5.595126581192017,
      "epoch": 3.5232235013894404,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003797499403532287,
      "loss": 4.9405,
      "mean_token_accuracy": 0.19070257395505905,
      "num_tokens": 77757866.0,
      "step": 35500
    },
    {
      "entropy": 5.615006303787231,
      "epoch": 3.523719730051608,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037971858151451403,
      "loss": 4.9739,
      "mean_token_accuracy": 0.19234058856964112,
      "num_tokens": 77769448.0,
      "step": 35505
    },
    {
      "entropy": 5.666511726379395,
      "epoch": 3.5242159587137754,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037968722007900415,
      "loss": 4.9298,
      "mean_token_accuracy": 0.19449602514505387,
      "num_tokens": 77781207.0,
      "step": 35510
    },
    {
      "entropy": 5.702298545837403,
      "epoch": 3.524712187375943,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037965585604747684,
      "loss": 5.0232,
      "mean_token_accuracy": 0.17995355576276778,
      "num_tokens": 77791050.0,
      "step": 35515
    },
    {
      "entropy": 5.654859972000122,
      "epoch": 3.5252084160381103,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003796244894207097,
      "loss": 4.9587,
      "mean_token_accuracy": 0.19370656758546828,
      "num_tokens": 77802486.0,
      "step": 35520
    },
    {
      "entropy": 5.616877841949463,
      "epoch": 3.525704644700278,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037959312019948075,
      "loss": 4.9558,
      "mean_token_accuracy": 0.18815942704677582,
      "num_tokens": 77815581.0,
      "step": 35525
    },
    {
      "entropy": 5.630742311477661,
      "epoch": 3.5262008733624453,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003795617483845678,
      "loss": 4.9754,
      "mean_token_accuracy": 0.1969585210084915,
      "num_tokens": 77826139.0,
      "step": 35530
    },
    {
      "entropy": 5.713269758224487,
      "epoch": 3.526697102024613,
      "grad_norm": 1.25,
      "learning_rate": 0.00037953037397674876,
      "loss": 5.0712,
      "mean_token_accuracy": 0.18017461150884628,
      "num_tokens": 77837929.0,
      "step": 35535
    },
    {
      "entropy": 5.6491142272949215,
      "epoch": 3.5271933306867806,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003794989969768017,
      "loss": 4.9387,
      "mean_token_accuracy": 0.19473579823970794,
      "num_tokens": 77848118.0,
      "step": 35540
    },
    {
      "entropy": 5.635585927963257,
      "epoch": 3.527689559348948,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003794676173855048,
      "loss": 4.9705,
      "mean_token_accuracy": 0.1852649301290512,
      "num_tokens": 77859378.0,
      "step": 35545
    },
    {
      "entropy": 5.612793827056885,
      "epoch": 3.5281857880111156,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037943623520363595,
      "loss": 4.9705,
      "mean_token_accuracy": 0.18741804361343384,
      "num_tokens": 77872524.0,
      "step": 35550
    },
    {
      "entropy": 5.617019557952881,
      "epoch": 3.528682016673283,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003794048504319736,
      "loss": 4.9728,
      "mean_token_accuracy": 0.19116953611373902,
      "num_tokens": 77883053.0,
      "step": 35555
    },
    {
      "entropy": 5.6800217628479,
      "epoch": 3.5291782453354505,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037937346307129584,
      "loss": 5.0622,
      "mean_token_accuracy": 0.1894163116812706,
      "num_tokens": 77894709.0,
      "step": 35560
    },
    {
      "entropy": 5.703686714172363,
      "epoch": 3.529674473997618,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037934207312238116,
      "loss": 5.0251,
      "mean_token_accuracy": 0.1897414967417717,
      "num_tokens": 77906213.0,
      "step": 35565
    },
    {
      "entropy": 5.67121000289917,
      "epoch": 3.5301707026597855,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037931068058600785,
      "loss": 4.9768,
      "mean_token_accuracy": 0.18386073857545854,
      "num_tokens": 77917015.0,
      "step": 35570
    },
    {
      "entropy": 5.72272515296936,
      "epoch": 3.530666931321953,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037927928546295454,
      "loss": 5.0576,
      "mean_token_accuracy": 0.18774193078279494,
      "num_tokens": 77927504.0,
      "step": 35575
    },
    {
      "entropy": 5.698671007156372,
      "epoch": 3.5311631599841204,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003792478877539995,
      "loss": 4.9875,
      "mean_token_accuracy": 0.19439131766557693,
      "num_tokens": 77937656.0,
      "step": 35580
    },
    {
      "entropy": 5.728352737426758,
      "epoch": 3.5316593886462884,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003792164874599216,
      "loss": 5.1122,
      "mean_token_accuracy": 0.17816724479198456,
      "num_tokens": 77948626.0,
      "step": 35585
    },
    {
      "entropy": 5.68288893699646,
      "epoch": 3.532155617308456,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003791850845814994,
      "loss": 4.9909,
      "mean_token_accuracy": 0.1875391945242882,
      "num_tokens": 77960055.0,
      "step": 35590
    },
    {
      "entropy": 5.6335883140563965,
      "epoch": 3.5326518459706233,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037915367911951146,
      "loss": 4.9674,
      "mean_token_accuracy": 0.18242724239826202,
      "num_tokens": 77971379.0,
      "step": 35595
    },
    {
      "entropy": 5.633884477615356,
      "epoch": 3.533148074632791,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037912227107473684,
      "loss": 4.938,
      "mean_token_accuracy": 0.1912369266152382,
      "num_tokens": 77982746.0,
      "step": 35600
    },
    {
      "entropy": 5.648226308822632,
      "epoch": 3.5336443032949583,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037909086044795427,
      "loss": 5.0031,
      "mean_token_accuracy": 0.18793057650327682,
      "num_tokens": 77994561.0,
      "step": 35605
    },
    {
      "entropy": 5.72290678024292,
      "epoch": 3.5341405319571257,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037905944723994257,
      "loss": 5.06,
      "mean_token_accuracy": 0.19225911945104598,
      "num_tokens": 78004835.0,
      "step": 35610
    },
    {
      "entropy": 5.58690996170044,
      "epoch": 3.534636760619293,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037902803145148085,
      "loss": 4.8901,
      "mean_token_accuracy": 0.19305321127176284,
      "num_tokens": 78015669.0,
      "step": 35615
    },
    {
      "entropy": 5.666744041442871,
      "epoch": 3.535132989281461,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003789966130833481,
      "loss": 5.0267,
      "mean_token_accuracy": 0.181561242043972,
      "num_tokens": 78025340.0,
      "step": 35620
    },
    {
      "entropy": 5.686877965927124,
      "epoch": 3.5356292179436286,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003789651921363234,
      "loss": 4.9968,
      "mean_token_accuracy": 0.18986756056547166,
      "num_tokens": 78034790.0,
      "step": 35625
    },
    {
      "entropy": 5.6900393009185795,
      "epoch": 3.536125446605796,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037893376861118606,
      "loss": 5.0708,
      "mean_token_accuracy": 0.18713434040546417,
      "num_tokens": 78045803.0,
      "step": 35630
    },
    {
      "entropy": 5.671922063827514,
      "epoch": 3.5366216752679636,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037890234250871515,
      "loss": 4.9619,
      "mean_token_accuracy": 0.19302959591150284,
      "num_tokens": 78056690.0,
      "step": 35635
    },
    {
      "entropy": 5.675445127487182,
      "epoch": 3.537117903930131,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037887091382969,
      "loss": 4.9618,
      "mean_token_accuracy": 0.2027149796485901,
      "num_tokens": 78067888.0,
      "step": 35640
    },
    {
      "entropy": 5.647562456130982,
      "epoch": 3.5376141325922985,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003788394825748901,
      "loss": 4.9662,
      "mean_token_accuracy": 0.19279858469963074,
      "num_tokens": 78077877.0,
      "step": 35645
    },
    {
      "entropy": 5.634072017669678,
      "epoch": 3.538110361254466,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037880804874509467,
      "loss": 4.9745,
      "mean_token_accuracy": 0.18782711923122405,
      "num_tokens": 78090114.0,
      "step": 35650
    },
    {
      "entropy": 5.776245212554931,
      "epoch": 3.5386065899166335,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003787766123410834,
      "loss": 5.1336,
      "mean_token_accuracy": 0.18200803250074388,
      "num_tokens": 78099737.0,
      "step": 35655
    },
    {
      "entropy": 5.639186096191406,
      "epoch": 3.539102818578801,
      "grad_norm": 1.125,
      "learning_rate": 0.0003787451733636357,
      "loss": 4.9084,
      "mean_token_accuracy": 0.19239026308059692,
      "num_tokens": 78110979.0,
      "step": 35660
    },
    {
      "entropy": 5.660355806350708,
      "epoch": 3.5395990472409684,
      "grad_norm": 1.0,
      "learning_rate": 0.00037871373181353135,
      "loss": 4.9551,
      "mean_token_accuracy": 0.19133665412664413,
      "num_tokens": 78123107.0,
      "step": 35665
    },
    {
      "entropy": 5.6628992557525635,
      "epoch": 3.5400952759031363,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037868228769154987,
      "loss": 5.0192,
      "mean_token_accuracy": 0.18878849148750304,
      "num_tokens": 78134128.0,
      "step": 35670
    },
    {
      "entropy": 5.687253808975219,
      "epoch": 3.540591504565304,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003786508409984711,
      "loss": 5.0374,
      "mean_token_accuracy": 0.1785629138350487,
      "num_tokens": 78144561.0,
      "step": 35675
    },
    {
      "entropy": 5.622957229614258,
      "epoch": 3.5410877332274713,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037861939173507484,
      "loss": 4.9011,
      "mean_token_accuracy": 0.19746443182229995,
      "num_tokens": 78155700.0,
      "step": 35680
    },
    {
      "entropy": 5.690249633789063,
      "epoch": 3.5415839618896388,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037858793990214087,
      "loss": 5.0644,
      "mean_token_accuracy": 0.18326238989830018,
      "num_tokens": 78166904.0,
      "step": 35685
    },
    {
      "entropy": 5.6079535484313965,
      "epoch": 3.5420801905518062,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037855648550044925,
      "loss": 4.9474,
      "mean_token_accuracy": 0.1930761978030205,
      "num_tokens": 78178409.0,
      "step": 35690
    },
    {
      "entropy": 5.721602821350098,
      "epoch": 3.5425764192139737,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003785250285307799,
      "loss": 5.007,
      "mean_token_accuracy": 0.18904672414064408,
      "num_tokens": 78190639.0,
      "step": 35695
    },
    {
      "entropy": 5.678948163986206,
      "epoch": 3.543072647876141,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037849356899391295,
      "loss": 4.9972,
      "mean_token_accuracy": 0.19167122095823289,
      "num_tokens": 78202345.0,
      "step": 35700
    },
    {
      "entropy": 5.614169645309448,
      "epoch": 3.543568876538309,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003784621068906285,
      "loss": 4.8912,
      "mean_token_accuracy": 0.19334588348865508,
      "num_tokens": 78212767.0,
      "step": 35705
    },
    {
      "entropy": 5.604736232757569,
      "epoch": 3.5440651052004766,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037843064222170677,
      "loss": 4.9443,
      "mean_token_accuracy": 0.18861835598945617,
      "num_tokens": 78223338.0,
      "step": 35710
    },
    {
      "entropy": 5.606545066833496,
      "epoch": 3.544561333862644,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037839917498792787,
      "loss": 4.9713,
      "mean_token_accuracy": 0.1929942011833191,
      "num_tokens": 78234903.0,
      "step": 35715
    },
    {
      "entropy": 5.641175603866577,
      "epoch": 3.5450575625248115,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037836770519007236,
      "loss": 4.9794,
      "mean_token_accuracy": 0.1922050029039383,
      "num_tokens": 78245936.0,
      "step": 35720
    },
    {
      "entropy": 5.654769372940064,
      "epoch": 3.545553791186979,
      "grad_norm": 1.125,
      "learning_rate": 0.0003783362328289205,
      "loss": 4.9865,
      "mean_token_accuracy": 0.1924961045384407,
      "num_tokens": 78256294.0,
      "step": 35725
    },
    {
      "entropy": 5.699330520629883,
      "epoch": 3.5460500198491465,
      "grad_norm": 1.125,
      "learning_rate": 0.0003783047579052526,
      "loss": 5.0485,
      "mean_token_accuracy": 0.18612752854824066,
      "num_tokens": 78266385.0,
      "step": 35730
    },
    {
      "entropy": 5.661443758010864,
      "epoch": 3.546546248511314,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037827328041984944,
      "loss": 4.9607,
      "mean_token_accuracy": 0.19514383226633072,
      "num_tokens": 78276044.0,
      "step": 35735
    },
    {
      "entropy": 5.651828002929688,
      "epoch": 3.5470424771734814,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003782418003734914,
      "loss": 5.0091,
      "mean_token_accuracy": 0.19046106934547424,
      "num_tokens": 78286943.0,
      "step": 35740
    },
    {
      "entropy": 5.640593385696411,
      "epoch": 3.547538705835649,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003782103177669592,
      "loss": 4.935,
      "mean_token_accuracy": 0.19651647806167602,
      "num_tokens": 78298313.0,
      "step": 35745
    },
    {
      "entropy": 5.664333915710449,
      "epoch": 3.5480349344978164,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037817883260103354,
      "loss": 5.0357,
      "mean_token_accuracy": 0.18074932247400283,
      "num_tokens": 78309750.0,
      "step": 35750
    },
    {
      "entropy": 5.609053993225098,
      "epoch": 3.548531163159984,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037814734487649516,
      "loss": 4.919,
      "mean_token_accuracy": 0.19580212980508804,
      "num_tokens": 78321998.0,
      "step": 35755
    },
    {
      "entropy": 5.714913463592529,
      "epoch": 3.5490273918221518,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003781158545941248,
      "loss": 5.0942,
      "mean_token_accuracy": 0.1791258081793785,
      "num_tokens": 78333973.0,
      "step": 35760
    },
    {
      "entropy": 5.749384450912475,
      "epoch": 3.5495236204843192,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037808436175470344,
      "loss": 5.1088,
      "mean_token_accuracy": 0.18318696022033693,
      "num_tokens": 78345281.0,
      "step": 35765
    },
    {
      "entropy": 5.731606149673462,
      "epoch": 3.5500198491464867,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003780528663590122,
      "loss": 5.0726,
      "mean_token_accuracy": 0.18143400996923448,
      "num_tokens": 78356995.0,
      "step": 35770
    },
    {
      "entropy": 5.660409212112427,
      "epoch": 3.550516077808654,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003780213684078318,
      "loss": 4.9882,
      "mean_token_accuracy": 0.188995061814785,
      "num_tokens": 78367910.0,
      "step": 35775
    },
    {
      "entropy": 5.603909254074097,
      "epoch": 3.5510123064708217,
      "grad_norm": 1.125,
      "learning_rate": 0.00037798986790194347,
      "loss": 4.9274,
      "mean_token_accuracy": 0.1951853394508362,
      "num_tokens": 78378995.0,
      "step": 35780
    },
    {
      "entropy": 5.734852027893067,
      "epoch": 3.551508535132989,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037795836484212837,
      "loss": 5.0055,
      "mean_token_accuracy": 0.19098564088344575,
      "num_tokens": 78389881.0,
      "step": 35785
    },
    {
      "entropy": 5.605979871749878,
      "epoch": 3.552004763795157,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037792685922916763,
      "loss": 4.8761,
      "mean_token_accuracy": 0.19388442784547805,
      "num_tokens": 78400242.0,
      "step": 35790
    },
    {
      "entropy": 5.598036575317383,
      "epoch": 3.5525009924573245,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00037789535106384266,
      "loss": 4.886,
      "mean_token_accuracy": 0.1909690022468567,
      "num_tokens": 78410057.0,
      "step": 35795
    },
    {
      "entropy": 5.621483707427979,
      "epoch": 3.552997221119492,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003778638403469346,
      "loss": 5.0153,
      "mean_token_accuracy": 0.188338603079319,
      "num_tokens": 78419684.0,
      "step": 35800
    },
    {
      "entropy": 5.638763809204102,
      "epoch": 3.5534934497816595,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037783232707922507,
      "loss": 4.9325,
      "mean_token_accuracy": 0.18809531778097152,
      "num_tokens": 78430569.0,
      "step": 35805
    },
    {
      "entropy": 5.712477302551269,
      "epoch": 3.553989678443827,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037780081126149523,
      "loss": 4.9949,
      "mean_token_accuracy": 0.19326310902833937,
      "num_tokens": 78441290.0,
      "step": 35810
    },
    {
      "entropy": 5.658570575714111,
      "epoch": 3.5544859071059944,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037776929289452696,
      "loss": 4.9755,
      "mean_token_accuracy": 0.1944759413599968,
      "num_tokens": 78451638.0,
      "step": 35815
    },
    {
      "entropy": 5.661691093444825,
      "epoch": 3.554982135768162,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003777377719791016,
      "loss": 4.9827,
      "mean_token_accuracy": 0.1950318694114685,
      "num_tokens": 78462759.0,
      "step": 35820
    },
    {
      "entropy": 5.569708395004272,
      "epoch": 3.5554783644303294,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037770624851600086,
      "loss": 4.9559,
      "mean_token_accuracy": 0.1980915367603302,
      "num_tokens": 78475465.0,
      "step": 35825
    },
    {
      "entropy": 5.570734214782715,
      "epoch": 3.555974593092497,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003776747225060065,
      "loss": 4.8641,
      "mean_token_accuracy": 0.19248344153165817,
      "num_tokens": 78487900.0,
      "step": 35830
    },
    {
      "entropy": 5.632704210281372,
      "epoch": 3.5564708217546643,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003776431939499003,
      "loss": 4.9094,
      "mean_token_accuracy": 0.19400074779987336,
      "num_tokens": 78497538.0,
      "step": 35835
    },
    {
      "entropy": 5.651606035232544,
      "epoch": 3.556967050416832,
      "grad_norm": 1.078125,
      "learning_rate": 0.000377611662848464,
      "loss": 5.0064,
      "mean_token_accuracy": 0.18663811236619948,
      "num_tokens": 78508299.0,
      "step": 35840
    },
    {
      "entropy": 5.630247783660889,
      "epoch": 3.5574632790789997,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003775801292024796,
      "loss": 4.9115,
      "mean_token_accuracy": 0.19790136814117432,
      "num_tokens": 78518188.0,
      "step": 35845
    },
    {
      "entropy": 5.642685270309448,
      "epoch": 3.557959507741167,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037754859301272905,
      "loss": 4.9858,
      "mean_token_accuracy": 0.19573793709278106,
      "num_tokens": 78528390.0,
      "step": 35850
    },
    {
      "entropy": 5.59140944480896,
      "epoch": 3.5584557364033347,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037751705427999436,
      "loss": 5.0419,
      "mean_token_accuracy": 0.18364521116018295,
      "num_tokens": 78540266.0,
      "step": 35855
    },
    {
      "entropy": 5.649893426895142,
      "epoch": 3.558951965065502,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003774855130050576,
      "loss": 5.0316,
      "mean_token_accuracy": 0.18550532460212707,
      "num_tokens": 78550937.0,
      "step": 35860
    },
    {
      "entropy": 5.673134231567383,
      "epoch": 3.5594481937276696,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037745396918870095,
      "loss": 4.9309,
      "mean_token_accuracy": 0.19180705547332763,
      "num_tokens": 78561619.0,
      "step": 35865
    },
    {
      "entropy": 5.608037042617798,
      "epoch": 3.559944422389837,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003774224228317067,
      "loss": 4.9442,
      "mean_token_accuracy": 0.19124407023191453,
      "num_tokens": 78573840.0,
      "step": 35870
    },
    {
      "entropy": 5.60228533744812,
      "epoch": 3.560440651052005,
      "grad_norm": 1.046875,
      "learning_rate": 0.000377390873934857,
      "loss": 4.995,
      "mean_token_accuracy": 0.1867305874824524,
      "num_tokens": 78585477.0,
      "step": 35875
    },
    {
      "entropy": 5.705884885787964,
      "epoch": 3.5609368797141725,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003773593224989343,
      "loss": 5.0726,
      "mean_token_accuracy": 0.18014492094516754,
      "num_tokens": 78597281.0,
      "step": 35880
    },
    {
      "entropy": 5.674394702911377,
      "epoch": 3.56143310837634,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003773277685247209,
      "loss": 4.9737,
      "mean_token_accuracy": 0.19225439578294753,
      "num_tokens": 78609300.0,
      "step": 35885
    },
    {
      "entropy": 5.712907075881958,
      "epoch": 3.5619293370385074,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037729621201299933,
      "loss": 4.9964,
      "mean_token_accuracy": 0.19026791006326677,
      "num_tokens": 78620443.0,
      "step": 35890
    },
    {
      "entropy": 5.672220182418823,
      "epoch": 3.562425565700675,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003772646529645522,
      "loss": 4.9954,
      "mean_token_accuracy": 0.18905208706855775,
      "num_tokens": 78630113.0,
      "step": 35895
    },
    {
      "entropy": 5.649132776260376,
      "epoch": 3.5629217943628424,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003772330913801621,
      "loss": 4.9215,
      "mean_token_accuracy": 0.1929207980632782,
      "num_tokens": 78640092.0,
      "step": 35900
    },
    {
      "entropy": 5.651379776000977,
      "epoch": 3.56341802302501,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003772015272606115,
      "loss": 4.9442,
      "mean_token_accuracy": 0.18785786777734756,
      "num_tokens": 78651092.0,
      "step": 35905
    },
    {
      "entropy": 5.618453645706177,
      "epoch": 3.5639142516871773,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003771699606066833,
      "loss": 4.9786,
      "mean_token_accuracy": 0.19375218003988265,
      "num_tokens": 78662435.0,
      "step": 35910
    },
    {
      "entropy": 5.613068866729736,
      "epoch": 3.564410480349345,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003771383914191602,
      "loss": 4.956,
      "mean_token_accuracy": 0.1909740835428238,
      "num_tokens": 78673299.0,
      "step": 35915
    },
    {
      "entropy": 5.656576919555664,
      "epoch": 3.5649067090115123,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003771068196988251,
      "loss": 4.9706,
      "mean_token_accuracy": 0.18975130766630172,
      "num_tokens": 78683286.0,
      "step": 35920
    },
    {
      "entropy": 5.634697198867798,
      "epoch": 3.5654029376736798,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037707524544646094,
      "loss": 4.9278,
      "mean_token_accuracy": 0.18903656005859376,
      "num_tokens": 78695282.0,
      "step": 35925
    },
    {
      "entropy": 5.584320211410523,
      "epoch": 3.5658991663358477,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003770436686628506,
      "loss": 4.9338,
      "mean_token_accuracy": 0.189358951151371,
      "num_tokens": 78706880.0,
      "step": 35930
    },
    {
      "entropy": 5.6537939548492435,
      "epoch": 3.566395394998015,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037701208934877716,
      "loss": 5.0016,
      "mean_token_accuracy": 0.1790866881608963,
      "num_tokens": 78717515.0,
      "step": 35935
    },
    {
      "entropy": 5.65866494178772,
      "epoch": 3.5668916236601826,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037698050750502375,
      "loss": 4.9615,
      "mean_token_accuracy": 0.19008933752775192,
      "num_tokens": 78728959.0,
      "step": 35940
    },
    {
      "entropy": 5.5979150295257565,
      "epoch": 3.56738785232235,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003769489231323735,
      "loss": 4.9228,
      "mean_token_accuracy": 0.19028561562299728,
      "num_tokens": 78740580.0,
      "step": 35945
    },
    {
      "entropy": 5.631115531921386,
      "epoch": 3.5678840809845176,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003769173362316096,
      "loss": 4.9538,
      "mean_token_accuracy": 0.18972940295934676,
      "num_tokens": 78751388.0,
      "step": 35950
    },
    {
      "entropy": 5.6256309986114506,
      "epoch": 3.568380309646685,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037688574680351544,
      "loss": 4.9904,
      "mean_token_accuracy": 0.18172740191221237,
      "num_tokens": 78763474.0,
      "step": 35955
    },
    {
      "entropy": 5.5998023509979244,
      "epoch": 3.568876538308853,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037685415484887426,
      "loss": 4.965,
      "mean_token_accuracy": 0.18693299740552902,
      "num_tokens": 78775570.0,
      "step": 35960
    },
    {
      "entropy": 5.669489240646362,
      "epoch": 3.5693727669710205,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037682256036846947,
      "loss": 4.9627,
      "mean_token_accuracy": 0.1854767009615898,
      "num_tokens": 78786558.0,
      "step": 35965
    },
    {
      "entropy": 5.676292562484742,
      "epoch": 3.569868995633188,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037679096336308473,
      "loss": 4.9597,
      "mean_token_accuracy": 0.19327413737773896,
      "num_tokens": 78796872.0,
      "step": 35970
    },
    {
      "entropy": 5.675503969192505,
      "epoch": 3.5703652242953554,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037675936383350326,
      "loss": 4.9616,
      "mean_token_accuracy": 0.18498455435037614,
      "num_tokens": 78807971.0,
      "step": 35975
    },
    {
      "entropy": 5.687791442871093,
      "epoch": 3.570861452957523,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003767277617805089,
      "loss": 5.0224,
      "mean_token_accuracy": 0.18368228375911713,
      "num_tokens": 78818593.0,
      "step": 35980
    },
    {
      "entropy": 5.6470414161682125,
      "epoch": 3.5713576816196904,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003766961572048853,
      "loss": 4.9664,
      "mean_token_accuracy": 0.19072598367929458,
      "num_tokens": 78830573.0,
      "step": 35985
    },
    {
      "entropy": 5.692064809799194,
      "epoch": 3.571853910281858,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003766645501074161,
      "loss": 5.0377,
      "mean_token_accuracy": 0.1832791656255722,
      "num_tokens": 78842041.0,
      "step": 35990
    },
    {
      "entropy": 5.704593849182129,
      "epoch": 3.5723501389440253,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037663294048888513,
      "loss": 4.9975,
      "mean_token_accuracy": 0.19022760093212127,
      "num_tokens": 78852549.0,
      "step": 35995
    },
    {
      "entropy": 5.715643072128296,
      "epoch": 3.572846367606193,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037660132835007625,
      "loss": 5.0693,
      "mean_token_accuracy": 0.19278404861688614,
      "num_tokens": 78862706.0,
      "step": 36000
    },
    {
      "epoch": 3.572846367606193,
      "eval_entropy": 5.415417033721387,
      "eval_loss": 5.111225605010986,
      "eval_mean_token_accuracy": 0.18852216658512355,
      "eval_num_tokens": 78862706.0,
      "eval_runtime": 20.51,
      "eval_samples_per_second": 1641.591,
      "eval_steps_per_second": 205.217,
      "step": 36000
    },
    {
      "entropy": 5.65831298828125,
      "epoch": 3.5733425962683603,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003765697136917734,
      "loss": 5.026,
      "mean_token_accuracy": 0.1800559341907501,
      "num_tokens": 78874214.0,
      "step": 36005
    },
    {
      "entropy": 5.649332714080811,
      "epoch": 3.5738388249305277,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003765380965147604,
      "loss": 5.0699,
      "mean_token_accuracy": 0.18580592125654222,
      "num_tokens": 78886178.0,
      "step": 36010
    },
    {
      "entropy": 5.64399790763855,
      "epoch": 3.5743350535926957,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037650647681982145,
      "loss": 4.962,
      "mean_token_accuracy": 0.18936851620674133,
      "num_tokens": 78896114.0,
      "step": 36015
    },
    {
      "entropy": 5.733638000488281,
      "epoch": 3.574831282254863,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037647485460774065,
      "loss": 4.964,
      "mean_token_accuracy": 0.19334932565689086,
      "num_tokens": 78906447.0,
      "step": 36020
    },
    {
      "entropy": 5.594136047363281,
      "epoch": 3.5753275109170306,
      "grad_norm": 1.125,
      "learning_rate": 0.0003764432298793021,
      "loss": 4.8811,
      "mean_token_accuracy": 0.1907363563776016,
      "num_tokens": 78917737.0,
      "step": 36025
    },
    {
      "entropy": 5.6024720668792725,
      "epoch": 3.575823739579198,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037641160263529003,
      "loss": 4.9868,
      "mean_token_accuracy": 0.1891980528831482,
      "num_tokens": 78929682.0,
      "step": 36030
    },
    {
      "entropy": 5.661636114120483,
      "epoch": 3.5763199682413656,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037637997287648877,
      "loss": 4.9466,
      "mean_token_accuracy": 0.1966970905661583,
      "num_tokens": 78941875.0,
      "step": 36035
    },
    {
      "entropy": 5.646842193603516,
      "epoch": 3.576816196903533,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037634834060368256,
      "loss": 4.9748,
      "mean_token_accuracy": 0.18541471511125565,
      "num_tokens": 78951891.0,
      "step": 36040
    },
    {
      "entropy": 5.6915350437164305,
      "epoch": 3.577312425565701,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003763167058176559,
      "loss": 5.0724,
      "mean_token_accuracy": 0.18555215895175933,
      "num_tokens": 78964141.0,
      "step": 36045
    },
    {
      "entropy": 5.674262666702271,
      "epoch": 3.5778086542278684,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003762850685191933,
      "loss": 4.9831,
      "mean_token_accuracy": 0.19616941511631011,
      "num_tokens": 78973644.0,
      "step": 36050
    },
    {
      "entropy": 5.622214841842651,
      "epoch": 3.578304882890036,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003762534287090792,
      "loss": 4.9705,
      "mean_token_accuracy": 0.18562959432601928,
      "num_tokens": 78984666.0,
      "step": 36055
    },
    {
      "entropy": 5.614223432540894,
      "epoch": 3.5788011115522034,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003762217863880983,
      "loss": 4.9414,
      "mean_token_accuracy": 0.1866360530257225,
      "num_tokens": 78996064.0,
      "step": 36060
    },
    {
      "entropy": 5.695829725265503,
      "epoch": 3.579297340214371,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003761901415570351,
      "loss": 5.0234,
      "mean_token_accuracy": 0.18355153352022172,
      "num_tokens": 79006733.0,
      "step": 36065
    },
    {
      "entropy": 5.684617757797241,
      "epoch": 3.5797935688765383,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037615849421667466,
      "loss": 4.9989,
      "mean_token_accuracy": 0.19414881765842437,
      "num_tokens": 79018027.0,
      "step": 36070
    },
    {
      "entropy": 5.6771135330200195,
      "epoch": 3.580289797538706,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003761268443678014,
      "loss": 4.9959,
      "mean_token_accuracy": 0.1872466742992401,
      "num_tokens": 79028281.0,
      "step": 36075
    },
    {
      "entropy": 5.679145288467407,
      "epoch": 3.5807860262008733,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003760951920112003,
      "loss": 4.9708,
      "mean_token_accuracy": 0.18765779733657836,
      "num_tokens": 79039110.0,
      "step": 36080
    },
    {
      "entropy": 5.728076648712158,
      "epoch": 3.5812822548630407,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037606353714765626,
      "loss": 5.052,
      "mean_token_accuracy": 0.18773867040872574,
      "num_tokens": 79051294.0,
      "step": 36085
    },
    {
      "entropy": 5.685580253601074,
      "epoch": 3.5817784835252082,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003760318797779543,
      "loss": 5.0709,
      "mean_token_accuracy": 0.18058335036039352,
      "num_tokens": 79062790.0,
      "step": 36090
    },
    {
      "entropy": 5.6533876895904545,
      "epoch": 3.5822747121873757,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003760002199028794,
      "loss": 5.0334,
      "mean_token_accuracy": 0.18741451650857927,
      "num_tokens": 79073699.0,
      "step": 36095
    },
    {
      "entropy": 5.655515718460083,
      "epoch": 3.5827709408495436,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003759685575232168,
      "loss": 4.8968,
      "mean_token_accuracy": 0.1990821421146393,
      "num_tokens": 79084588.0,
      "step": 36100
    },
    {
      "entropy": 5.619409990310669,
      "epoch": 3.583267169511711,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003759368926397514,
      "loss": 4.9931,
      "mean_token_accuracy": 0.1866682142019272,
      "num_tokens": 79094767.0,
      "step": 36105
    },
    {
      "entropy": 5.640954923629761,
      "epoch": 3.5837633981738786,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003759052252532686,
      "loss": 4.9986,
      "mean_token_accuracy": 0.18463067710399628,
      "num_tokens": 79105186.0,
      "step": 36110
    },
    {
      "entropy": 5.647795867919922,
      "epoch": 3.584259626836046,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003758735553645537,
      "loss": 4.9713,
      "mean_token_accuracy": 0.19127165973186494,
      "num_tokens": 79116184.0,
      "step": 36115
    },
    {
      "entropy": 5.648366451263428,
      "epoch": 3.5847558554982135,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000375841882974392,
      "loss": 4.9318,
      "mean_token_accuracy": 0.18581433296203614,
      "num_tokens": 79126676.0,
      "step": 36120
    },
    {
      "entropy": 5.697343111038208,
      "epoch": 3.585252084160381,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003758102080835688,
      "loss": 5.0867,
      "mean_token_accuracy": 0.1806989684700966,
      "num_tokens": 79138090.0,
      "step": 36125
    },
    {
      "entropy": 5.6512312412261965,
      "epoch": 3.5857483128225485,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003757785306928697,
      "loss": 4.9518,
      "mean_token_accuracy": 0.19494948387145997,
      "num_tokens": 79150087.0,
      "step": 36130
    },
    {
      "entropy": 5.656663846969605,
      "epoch": 3.5862445414847164,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003757468508030803,
      "loss": 5.0235,
      "mean_token_accuracy": 0.19162874817848205,
      "num_tokens": 79162296.0,
      "step": 36135
    },
    {
      "entropy": 5.626573371887207,
      "epoch": 3.586740770146884,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003757151684149859,
      "loss": 5.0127,
      "mean_token_accuracy": 0.18652096539735794,
      "num_tokens": 79173763.0,
      "step": 36140
    },
    {
      "entropy": 5.695679140090943,
      "epoch": 3.5872369988090513,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003756834835293725,
      "loss": 5.0524,
      "mean_token_accuracy": 0.18781177401542665,
      "num_tokens": 79184323.0,
      "step": 36145
    },
    {
      "entropy": 5.66376576423645,
      "epoch": 3.587733227471219,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003756517961470256,
      "loss": 5.0362,
      "mean_token_accuracy": 0.182950197160244,
      "num_tokens": 79195811.0,
      "step": 36150
    },
    {
      "entropy": 5.7285661697387695,
      "epoch": 3.5882294561333863,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003756201062687311,
      "loss": 5.0836,
      "mean_token_accuracy": 0.1856821894645691,
      "num_tokens": 79207243.0,
      "step": 36155
    },
    {
      "entropy": 5.684700441360474,
      "epoch": 3.5887256847955538,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037558841389527466,
      "loss": 5.0229,
      "mean_token_accuracy": 0.18920276463031768,
      "num_tokens": 79218909.0,
      "step": 36160
    },
    {
      "entropy": 5.679389524459839,
      "epoch": 3.5892219134577212,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037555671902744244,
      "loss": 4.9724,
      "mean_token_accuracy": 0.18980007022619247,
      "num_tokens": 79229170.0,
      "step": 36165
    },
    {
      "entropy": 5.6475409984588625,
      "epoch": 3.5897181421198887,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037552502166602014,
      "loss": 5.0243,
      "mean_token_accuracy": 0.18580437302589417,
      "num_tokens": 79240001.0,
      "step": 36170
    },
    {
      "entropy": 5.676709699630737,
      "epoch": 3.590214370782056,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037549332181179403,
      "loss": 5.0499,
      "mean_token_accuracy": 0.1838561400771141,
      "num_tokens": 79251525.0,
      "step": 36175
    },
    {
      "entropy": 5.653306388854981,
      "epoch": 3.5907105994442237,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037546161946555004,
      "loss": 5.0274,
      "mean_token_accuracy": 0.1866101011633873,
      "num_tokens": 79264125.0,
      "step": 36180
    },
    {
      "entropy": 5.661091279983521,
      "epoch": 3.5912068281063916,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037542991462807433,
      "loss": 4.9689,
      "mean_token_accuracy": 0.19023056477308273,
      "num_tokens": 79273880.0,
      "step": 36185
    },
    {
      "entropy": 5.684352540969849,
      "epoch": 3.591703056768559,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037539820730015316,
      "loss": 5.0034,
      "mean_token_accuracy": 0.18884728103876114,
      "num_tokens": 79284542.0,
      "step": 36190
    },
    {
      "entropy": 5.566330099105835,
      "epoch": 3.5921992854307265,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037536649748257285,
      "loss": 4.8794,
      "mean_token_accuracy": 0.19524671137332916,
      "num_tokens": 79295182.0,
      "step": 36195
    },
    {
      "entropy": 5.56683874130249,
      "epoch": 3.592695514092894,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037533478517611955,
      "loss": 4.938,
      "mean_token_accuracy": 0.19840901345014572,
      "num_tokens": 79305613.0,
      "step": 36200
    },
    {
      "entropy": 5.657979726791382,
      "epoch": 3.5931917427550615,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003753030703815799,
      "loss": 4.9971,
      "mean_token_accuracy": 0.1958724558353424,
      "num_tokens": 79316566.0,
      "step": 36205
    },
    {
      "entropy": 5.6533856868743895,
      "epoch": 3.593687971417229,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003752713530997402,
      "loss": 4.9549,
      "mean_token_accuracy": 0.19662887454032899,
      "num_tokens": 79327224.0,
      "step": 36210
    },
    {
      "entropy": 5.709544992446899,
      "epoch": 3.5941842000793964,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037523963333138693,
      "loss": 5.0681,
      "mean_token_accuracy": 0.18344240486621857,
      "num_tokens": 79338069.0,
      "step": 36215
    },
    {
      "entropy": 5.76731653213501,
      "epoch": 3.5946804287415643,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003752079110773068,
      "loss": 5.0852,
      "mean_token_accuracy": 0.18479660153388977,
      "num_tokens": 79348677.0,
      "step": 36220
    },
    {
      "entropy": 5.612334060668945,
      "epoch": 3.595176657403732,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003751761863382865,
      "loss": 4.9692,
      "mean_token_accuracy": 0.19147969037294388,
      "num_tokens": 79359253.0,
      "step": 36225
    },
    {
      "entropy": 5.60349555015564,
      "epoch": 3.5956728860658993,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037514445911511257,
      "loss": 4.9456,
      "mean_token_accuracy": 0.19128594249486924,
      "num_tokens": 79370409.0,
      "step": 36230
    },
    {
      "entropy": 5.642172050476074,
      "epoch": 3.5961691147280668,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003751127294085719,
      "loss": 4.9986,
      "mean_token_accuracy": 0.18203895688056945,
      "num_tokens": 79381259.0,
      "step": 36235
    },
    {
      "entropy": 5.616181516647339,
      "epoch": 3.5966653433902342,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037508099721945113,
      "loss": 4.9458,
      "mean_token_accuracy": 0.18629637956619263,
      "num_tokens": 79391723.0,
      "step": 36240
    },
    {
      "entropy": 5.690949106216431,
      "epoch": 3.5971615720524017,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003750492625485374,
      "loss": 4.9793,
      "mean_token_accuracy": 0.1844635620713234,
      "num_tokens": 79401015.0,
      "step": 36245
    },
    {
      "entropy": 5.615077638626099,
      "epoch": 3.597657800714569,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003750175253966175,
      "loss": 4.8701,
      "mean_token_accuracy": 0.20687513351440429,
      "num_tokens": 79411923.0,
      "step": 36250
    },
    {
      "entropy": 5.6211851119995115,
      "epoch": 3.5981540293767367,
      "grad_norm": 1.0,
      "learning_rate": 0.0003749857857644785,
      "loss": 4.9718,
      "mean_token_accuracy": 0.18484655022621155,
      "num_tokens": 79423934.0,
      "step": 36255
    },
    {
      "entropy": 5.600337362289428,
      "epoch": 3.598650258038904,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003749540436529074,
      "loss": 4.9327,
      "mean_token_accuracy": 0.1933278039097786,
      "num_tokens": 79433718.0,
      "step": 36260
    },
    {
      "entropy": 5.639444208145141,
      "epoch": 3.5991464867010716,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037492229906269147,
      "loss": 4.962,
      "mean_token_accuracy": 0.18490202128887176,
      "num_tokens": 79443935.0,
      "step": 36265
    },
    {
      "entropy": 5.647246551513672,
      "epoch": 3.5996427153632395,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037489055199461783,
      "loss": 4.9432,
      "mean_token_accuracy": 0.19541993290185927,
      "num_tokens": 79455442.0,
      "step": 36270
    },
    {
      "entropy": 5.652853965759277,
      "epoch": 3.600138944025407,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003748588024494737,
      "loss": 4.9478,
      "mean_token_accuracy": 0.18824672549962998,
      "num_tokens": 79466591.0,
      "step": 36275
    },
    {
      "entropy": 5.674451589584351,
      "epoch": 3.6006351726875745,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037482705042804655,
      "loss": 5.0808,
      "mean_token_accuracy": 0.18160554617643357,
      "num_tokens": 79477064.0,
      "step": 36280
    },
    {
      "entropy": 5.702702474594116,
      "epoch": 3.601131401349742,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037479529593112355,
      "loss": 5.0474,
      "mean_token_accuracy": 0.18650759160518646,
      "num_tokens": 79488799.0,
      "step": 36285
    },
    {
      "entropy": 5.653038835525512,
      "epoch": 3.6016276300119094,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003747635389594923,
      "loss": 4.9996,
      "mean_token_accuracy": 0.19052726179361343,
      "num_tokens": 79499406.0,
      "step": 36290
    },
    {
      "entropy": 5.616351556777954,
      "epoch": 3.602123858674077,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037473177951394027,
      "loss": 4.8941,
      "mean_token_accuracy": 0.19986247569322585,
      "num_tokens": 79509347.0,
      "step": 36295
    },
    {
      "entropy": 5.716275882720947,
      "epoch": 3.6026200873362444,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037470001759525495,
      "loss": 5.039,
      "mean_token_accuracy": 0.18097063601017,
      "num_tokens": 79520515.0,
      "step": 36300
    },
    {
      "entropy": 5.59455714225769,
      "epoch": 3.6031163159984123,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037466825320422407,
      "loss": 4.8987,
      "mean_token_accuracy": 0.19412264227867126,
      "num_tokens": 79530563.0,
      "step": 36305
    },
    {
      "entropy": 5.638566493988037,
      "epoch": 3.60361254466058,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037463648634163534,
      "loss": 4.9867,
      "mean_token_accuracy": 0.1896317720413208,
      "num_tokens": 79541578.0,
      "step": 36310
    },
    {
      "entropy": 5.583996057510376,
      "epoch": 3.6041087733227473,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003746047170082764,
      "loss": 4.9097,
      "mean_token_accuracy": 0.1949674129486084,
      "num_tokens": 79552170.0,
      "step": 36315
    },
    {
      "entropy": 5.64463438987732,
      "epoch": 3.6046050019849147,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037457294520493504,
      "loss": 5.0078,
      "mean_token_accuracy": 0.18998806476593016,
      "num_tokens": 79562205.0,
      "step": 36320
    },
    {
      "entropy": 5.691317224502564,
      "epoch": 3.605101230647082,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037454117093239925,
      "loss": 5.0289,
      "mean_token_accuracy": 0.18537894487380982,
      "num_tokens": 79572683.0,
      "step": 36325
    },
    {
      "entropy": 5.7593615531921385,
      "epoch": 3.6055974593092497,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003745093941914569,
      "loss": 5.0661,
      "mean_token_accuracy": 0.17923171520233155,
      "num_tokens": 79583453.0,
      "step": 36330
    },
    {
      "entropy": 5.670732831954956,
      "epoch": 3.606093687971417,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037447761498289595,
      "loss": 4.9341,
      "mean_token_accuracy": 0.19214134365320207,
      "num_tokens": 79593902.0,
      "step": 36335
    },
    {
      "entropy": 5.606402397155762,
      "epoch": 3.6065899166335846,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003744458333075046,
      "loss": 4.9352,
      "mean_token_accuracy": 0.19588563442230225,
      "num_tokens": 79606511.0,
      "step": 36340
    },
    {
      "entropy": 5.632430553436279,
      "epoch": 3.607086145295752,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003744140491660708,
      "loss": 4.9777,
      "mean_token_accuracy": 0.18626412451267244,
      "num_tokens": 79617082.0,
      "step": 36345
    },
    {
      "entropy": 5.698844766616821,
      "epoch": 3.6075823739579196,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003743822625593829,
      "loss": 5.0546,
      "mean_token_accuracy": 0.1805280938744545,
      "num_tokens": 79627721.0,
      "step": 36350
    },
    {
      "entropy": 5.677924537658692,
      "epoch": 3.608078602620087,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000374350473488229,
      "loss": 4.9716,
      "mean_token_accuracy": 0.18980737030506134,
      "num_tokens": 79637464.0,
      "step": 36355
    },
    {
      "entropy": 5.6078393936157225,
      "epoch": 3.608574831282255,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037431868195339735,
      "loss": 4.986,
      "mean_token_accuracy": 0.18550899922847747,
      "num_tokens": 79647128.0,
      "step": 36360
    },
    {
      "entropy": 5.633811902999878,
      "epoch": 3.6090710599444225,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003742868879556764,
      "loss": 4.9281,
      "mean_token_accuracy": 0.19519178420305253,
      "num_tokens": 79658569.0,
      "step": 36365
    },
    {
      "entropy": 5.689254522323608,
      "epoch": 3.60956728860659,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037425509149585466,
      "loss": 4.9933,
      "mean_token_accuracy": 0.18845237791538239,
      "num_tokens": 79669660.0,
      "step": 36370
    },
    {
      "entropy": 5.689824056625366,
      "epoch": 3.6100635172687574,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003742232925747205,
      "loss": 5.0046,
      "mean_token_accuracy": 0.18511229008436203,
      "num_tokens": 79680017.0,
      "step": 36375
    },
    {
      "entropy": 5.579030179977417,
      "epoch": 3.610559745930925,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003741914911930625,
      "loss": 4.9339,
      "mean_token_accuracy": 0.20702509135007857,
      "num_tokens": 79691291.0,
      "step": 36380
    },
    {
      "entropy": 5.636130285263062,
      "epoch": 3.6110559745930924,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003741596873516693,
      "loss": 5.008,
      "mean_token_accuracy": 0.18590548485517502,
      "num_tokens": 79701539.0,
      "step": 36385
    },
    {
      "entropy": 5.652529764175415,
      "epoch": 3.6115522032552603,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003741278810513294,
      "loss": 4.9795,
      "mean_token_accuracy": 0.1997131809592247,
      "num_tokens": 79711869.0,
      "step": 36390
    },
    {
      "entropy": 5.645577812194825,
      "epoch": 3.6120484319174277,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037409607229283175,
      "loss": 5.0466,
      "mean_token_accuracy": 0.19003929942846298,
      "num_tokens": 79722171.0,
      "step": 36395
    },
    {
      "entropy": 5.688244819641113,
      "epoch": 3.6125446605795952,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000374064261076965,
      "loss": 5.0108,
      "mean_token_accuracy": 0.18483102023601533,
      "num_tokens": 79733877.0,
      "step": 36400
    },
    {
      "entropy": 5.6418904781341555,
      "epoch": 3.6130408892417627,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000374032447404518,
      "loss": 4.9667,
      "mean_token_accuracy": 0.18737137615680693,
      "num_tokens": 79744494.0,
      "step": 36405
    },
    {
      "entropy": 5.70879282951355,
      "epoch": 3.61353711790393,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003740006312762799,
      "loss": 5.0439,
      "mean_token_accuracy": 0.18458761870861054,
      "num_tokens": 79755989.0,
      "step": 36410
    },
    {
      "entropy": 5.690780925750732,
      "epoch": 3.6140333465660976,
      "grad_norm": 1.734375,
      "learning_rate": 0.00037396881269303924,
      "loss": 5.0497,
      "mean_token_accuracy": 0.19414433538913728,
      "num_tokens": 79767184.0,
      "step": 36415
    },
    {
      "entropy": 5.685522508621216,
      "epoch": 3.614529575228265,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037393699165558534,
      "loss": 5.0284,
      "mean_token_accuracy": 0.1872468963265419,
      "num_tokens": 79778493.0,
      "step": 36420
    },
    {
      "entropy": 5.67584981918335,
      "epoch": 3.6150258038904326,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003739051681647073,
      "loss": 5.0072,
      "mean_token_accuracy": 0.19032889157533645,
      "num_tokens": 79787998.0,
      "step": 36425
    },
    {
      "entropy": 5.661223697662353,
      "epoch": 3.6155220325526,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037387334222119424,
      "loss": 5.0542,
      "mean_token_accuracy": 0.18416759222745896,
      "num_tokens": 79799469.0,
      "step": 36430
    },
    {
      "entropy": 5.678212594985962,
      "epoch": 3.6160182612147675,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037384151382583526,
      "loss": 4.955,
      "mean_token_accuracy": 0.18510279655456544,
      "num_tokens": 79808656.0,
      "step": 36435
    },
    {
      "entropy": 5.670429277420044,
      "epoch": 3.616514489876935,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003738096829794197,
      "loss": 4.9739,
      "mean_token_accuracy": 0.18861720710992813,
      "num_tokens": 79818127.0,
      "step": 36440
    },
    {
      "entropy": 5.672072410583496,
      "epoch": 3.617010718539103,
      "grad_norm": 1.078125,
      "learning_rate": 0.000373777849682737,
      "loss": 5.0001,
      "mean_token_accuracy": 0.1902317225933075,
      "num_tokens": 79830451.0,
      "step": 36445
    },
    {
      "entropy": 5.673600530624389,
      "epoch": 3.6175069472012704,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003737460139365764,
      "loss": 5.0247,
      "mean_token_accuracy": 0.1869382381439209,
      "num_tokens": 79842623.0,
      "step": 36450
    },
    {
      "entropy": 5.684587097167968,
      "epoch": 3.618003175863438,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003737141757417275,
      "loss": 4.967,
      "mean_token_accuracy": 0.190390045940876,
      "num_tokens": 79852208.0,
      "step": 36455
    },
    {
      "entropy": 5.697500705718994,
      "epoch": 3.6184994045256054,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003736823350989797,
      "loss": 5.016,
      "mean_token_accuracy": 0.18337772488594056,
      "num_tokens": 79862356.0,
      "step": 36460
    },
    {
      "entropy": 5.622580432891846,
      "epoch": 3.618995633187773,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037365049200912274,
      "loss": 5.0019,
      "mean_token_accuracy": 0.1845598042011261,
      "num_tokens": 79873045.0,
      "step": 36465
    },
    {
      "entropy": 5.725132465362549,
      "epoch": 3.6194918618499403,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037361864647294596,
      "loss": 5.0161,
      "mean_token_accuracy": 0.18643956780433654,
      "num_tokens": 79882738.0,
      "step": 36470
    },
    {
      "entropy": 5.6930992603302,
      "epoch": 3.6199880905121082,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003735867984912394,
      "loss": 5.0169,
      "mean_token_accuracy": 0.18620431870222093,
      "num_tokens": 79894047.0,
      "step": 36475
    },
    {
      "entropy": 5.657182455062866,
      "epoch": 3.6204843191742757,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037355494806479264,
      "loss": 4.9408,
      "mean_token_accuracy": 0.19665479362010957,
      "num_tokens": 79905607.0,
      "step": 36480
    },
    {
      "entropy": 5.636180591583252,
      "epoch": 3.620980547836443,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003735230951943955,
      "loss": 5.0431,
      "mean_token_accuracy": 0.18628124445676802,
      "num_tokens": 79917995.0,
      "step": 36485
    },
    {
      "entropy": 5.666871500015259,
      "epoch": 3.6214767764986107,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037349123988083795,
      "loss": 4.9933,
      "mean_token_accuracy": 0.1903167635202408,
      "num_tokens": 79929600.0,
      "step": 36490
    },
    {
      "entropy": 5.694523477554322,
      "epoch": 3.621973005160778,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037345938212490987,
      "loss": 5.0136,
      "mean_token_accuracy": 0.1896126851439476,
      "num_tokens": 79940597.0,
      "step": 36495
    },
    {
      "entropy": 5.7488288402557375,
      "epoch": 3.6224692338229456,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037342752192740127,
      "loss": 5.009,
      "mean_token_accuracy": 0.1933494135737419,
      "num_tokens": 79951531.0,
      "step": 36500
    },
    {
      "entropy": 5.678081274032593,
      "epoch": 3.622965462485113,
      "grad_norm": 1.25,
      "learning_rate": 0.0003733956592891022,
      "loss": 5.0164,
      "mean_token_accuracy": 0.18386504501104356,
      "num_tokens": 79962651.0,
      "step": 36505
    },
    {
      "entropy": 5.67175874710083,
      "epoch": 3.6234616911472806,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037336379421080274,
      "loss": 5.0069,
      "mean_token_accuracy": 0.18853414803743362,
      "num_tokens": 79973219.0,
      "step": 36510
    },
    {
      "entropy": 5.661754703521728,
      "epoch": 3.623957919809448,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003733319266932932,
      "loss": 4.9599,
      "mean_token_accuracy": 0.1868635058403015,
      "num_tokens": 79985037.0,
      "step": 36515
    },
    {
      "entropy": 5.67141900062561,
      "epoch": 3.6244541484716155,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037330005673736385,
      "loss": 5.0507,
      "mean_token_accuracy": 0.17767749428749086,
      "num_tokens": 79994596.0,
      "step": 36520
    },
    {
      "entropy": 5.67039303779602,
      "epoch": 3.624950377133783,
      "grad_norm": 1.125,
      "learning_rate": 0.0003732681843438048,
      "loss": 4.9946,
      "mean_token_accuracy": 0.18737873882055284,
      "num_tokens": 80006045.0,
      "step": 36525
    },
    {
      "entropy": 5.656874465942383,
      "epoch": 3.625446605795951,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037323630951340654,
      "loss": 4.9757,
      "mean_token_accuracy": 0.18462798297405242,
      "num_tokens": 80016423.0,
      "step": 36530
    },
    {
      "entropy": 5.687429666519165,
      "epoch": 3.6259428344581184,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037320443224695955,
      "loss": 5.0306,
      "mean_token_accuracy": 0.18741716742515563,
      "num_tokens": 80026579.0,
      "step": 36535
    },
    {
      "entropy": 5.6406067371368405,
      "epoch": 3.626439063120286,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003731725525452542,
      "loss": 4.953,
      "mean_token_accuracy": 0.19027327746152878,
      "num_tokens": 80037227.0,
      "step": 36540
    },
    {
      "entropy": 5.681134176254273,
      "epoch": 3.6269352917824533,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003731406704090811,
      "loss": 5.0083,
      "mean_token_accuracy": 0.18345795273780824,
      "num_tokens": 80049421.0,
      "step": 36545
    },
    {
      "entropy": 5.698862028121948,
      "epoch": 3.627431520444621,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037310878583923085,
      "loss": 4.9946,
      "mean_token_accuracy": 0.19097982794046403,
      "num_tokens": 80059710.0,
      "step": 36550
    },
    {
      "entropy": 5.659502792358398,
      "epoch": 3.6279277491067883,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003730768988364941,
      "loss": 4.9732,
      "mean_token_accuracy": 0.1895521327853203,
      "num_tokens": 80071403.0,
      "step": 36555
    },
    {
      "entropy": 5.616786003112793,
      "epoch": 3.628423977768956,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037304500940166165,
      "loss": 4.8705,
      "mean_token_accuracy": 0.20634267181158067,
      "num_tokens": 80082978.0,
      "step": 36560
    },
    {
      "entropy": 5.678821802139282,
      "epoch": 3.6289202064311237,
      "grad_norm": 1.34375,
      "learning_rate": 0.00037301311753552407,
      "loss": 5.0353,
      "mean_token_accuracy": 0.19457911550998688,
      "num_tokens": 80093002.0,
      "step": 36565
    },
    {
      "entropy": 5.654574584960938,
      "epoch": 3.629416435093291,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003729812232388726,
      "loss": 5.0083,
      "mean_token_accuracy": 0.1863307386636734,
      "num_tokens": 80104738.0,
      "step": 36570
    },
    {
      "entropy": 5.649694728851318,
      "epoch": 3.6299126637554586,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003729493265124978,
      "loss": 4.953,
      "mean_token_accuracy": 0.18996195942163469,
      "num_tokens": 80113746.0,
      "step": 36575
    },
    {
      "entropy": 5.704814338684082,
      "epoch": 3.630408892417626,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037291742735719077,
      "loss": 5.0292,
      "mean_token_accuracy": 0.1834774434566498,
      "num_tokens": 80126558.0,
      "step": 36580
    },
    {
      "entropy": 5.674174213409424,
      "epoch": 3.6309051210797936,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003728855257737425,
      "loss": 5.0598,
      "mean_token_accuracy": 0.18599198311567305,
      "num_tokens": 80138174.0,
      "step": 36585
    },
    {
      "entropy": 5.6763890266418455,
      "epoch": 3.631401349741961,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003728536217629442,
      "loss": 4.9611,
      "mean_token_accuracy": 0.19212673604488373,
      "num_tokens": 80148044.0,
      "step": 36590
    },
    {
      "entropy": 5.730143070220947,
      "epoch": 3.6318975784041285,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037282171532558697,
      "loss": 5.0697,
      "mean_token_accuracy": 0.18190509229898452,
      "num_tokens": 80158725.0,
      "step": 36595
    },
    {
      "entropy": 5.764772272109985,
      "epoch": 3.632393807066296,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037278980646246186,
      "loss": 5.1196,
      "mean_token_accuracy": 0.18671546280384063,
      "num_tokens": 80170360.0,
      "step": 36600
    },
    {
      "entropy": 5.669585371017456,
      "epoch": 3.6328900357284635,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037275789517436035,
      "loss": 4.9814,
      "mean_token_accuracy": 0.18902914226055145,
      "num_tokens": 80180167.0,
      "step": 36605
    },
    {
      "entropy": 5.590351533889771,
      "epoch": 3.633386264390631,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037272598146207363,
      "loss": 4.9071,
      "mean_token_accuracy": 0.18805044144392014,
      "num_tokens": 80190867.0,
      "step": 36610
    },
    {
      "entropy": 5.665434694290161,
      "epoch": 3.633882493052799,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037269406532639324,
      "loss": 4.9388,
      "mean_token_accuracy": 0.19216042309999465,
      "num_tokens": 80201258.0,
      "step": 36615
    },
    {
      "entropy": 5.650170707702637,
      "epoch": 3.6343787217149663,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003726621467681105,
      "loss": 4.9964,
      "mean_token_accuracy": 0.1902166113257408,
      "num_tokens": 80212499.0,
      "step": 36620
    },
    {
      "entropy": 5.648655223846435,
      "epoch": 3.634874950377134,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003726302257880169,
      "loss": 4.9647,
      "mean_token_accuracy": 0.18927889615297316,
      "num_tokens": 80223611.0,
      "step": 36625
    },
    {
      "entropy": 5.634912157058716,
      "epoch": 3.6353711790393013,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003725983023869041,
      "loss": 4.9924,
      "mean_token_accuracy": 0.18796363174915315,
      "num_tokens": 80233678.0,
      "step": 36630
    },
    {
      "entropy": 5.725435066223144,
      "epoch": 3.6358674077014688,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003725663765655638,
      "loss": 5.0603,
      "mean_token_accuracy": 0.18335881531238557,
      "num_tokens": 80246307.0,
      "step": 36635
    },
    {
      "entropy": 5.634641408920288,
      "epoch": 3.6363636363636362,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003725344483247875,
      "loss": 4.8644,
      "mean_token_accuracy": 0.20288875699043274,
      "num_tokens": 80256945.0,
      "step": 36640
    },
    {
      "entropy": 5.630103206634521,
      "epoch": 3.636859865025804,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037250251766536715,
      "loss": 5.0038,
      "mean_token_accuracy": 0.1923816680908203,
      "num_tokens": 80266299.0,
      "step": 36645
    },
    {
      "entropy": 5.643756151199341,
      "epoch": 3.6373560936879716,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003724705845880944,
      "loss": 5.0001,
      "mean_token_accuracy": 0.19008296430110933,
      "num_tokens": 80276004.0,
      "step": 36650
    },
    {
      "entropy": 5.680943822860717,
      "epoch": 3.637852322350139,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003724386490937611,
      "loss": 5.0557,
      "mean_token_accuracy": 0.17859701365232467,
      "num_tokens": 80285962.0,
      "step": 36655
    },
    {
      "entropy": 5.728194761276245,
      "epoch": 3.6383485510123066,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003724067111831594,
      "loss": 5.0518,
      "mean_token_accuracy": 0.1870071694254875,
      "num_tokens": 80295457.0,
      "step": 36660
    },
    {
      "entropy": 5.6219676494598385,
      "epoch": 3.638844779674474,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003723747708570811,
      "loss": 4.9463,
      "mean_token_accuracy": 0.19691976755857468,
      "num_tokens": 80307125.0,
      "step": 36665
    },
    {
      "entropy": 5.667863273620606,
      "epoch": 3.6393410083366415,
      "grad_norm": 1.125,
      "learning_rate": 0.00037234282811631826,
      "loss": 4.978,
      "mean_token_accuracy": 0.18200020045042037,
      "num_tokens": 80318629.0,
      "step": 36670
    },
    {
      "entropy": 5.708804035186768,
      "epoch": 3.639837236998809,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003723108829616631,
      "loss": 5.057,
      "mean_token_accuracy": 0.18655727803707123,
      "num_tokens": 80330901.0,
      "step": 36675
    },
    {
      "entropy": 5.6356292247772215,
      "epoch": 3.6403334656609765,
      "grad_norm": 1.25,
      "learning_rate": 0.0003722789353939077,
      "loss": 4.9347,
      "mean_token_accuracy": 0.1937500610947609,
      "num_tokens": 80341888.0,
      "step": 36680
    },
    {
      "entropy": 5.659424924850464,
      "epoch": 3.640829694323144,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037224698541384434,
      "loss": 4.8946,
      "mean_token_accuracy": 0.19571363925933838,
      "num_tokens": 80352903.0,
      "step": 36685
    },
    {
      "entropy": 5.601434803009033,
      "epoch": 3.6413259229853114,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037221503302226533,
      "loss": 4.931,
      "mean_token_accuracy": 0.19389264583587645,
      "num_tokens": 80364609.0,
      "step": 36690
    },
    {
      "entropy": 5.613371133804321,
      "epoch": 3.641822151647479,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003721830782199629,
      "loss": 4.9928,
      "mean_token_accuracy": 0.18531924933195115,
      "num_tokens": 80375431.0,
      "step": 36695
    },
    {
      "entropy": 5.635662508010864,
      "epoch": 3.642318380309647,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003721511210077296,
      "loss": 5.0026,
      "mean_token_accuracy": 0.18832167536020278,
      "num_tokens": 80385503.0,
      "step": 36700
    },
    {
      "entropy": 5.692660903930664,
      "epoch": 3.6428146089718143,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037211916138635775,
      "loss": 5.0507,
      "mean_token_accuracy": 0.18342592120170592,
      "num_tokens": 80396752.0,
      "step": 36705
    },
    {
      "entropy": 5.647436714172363,
      "epoch": 3.643310837633982,
      "grad_norm": 1.25,
      "learning_rate": 0.00037208719935664003,
      "loss": 4.962,
      "mean_token_accuracy": 0.19091894924640657,
      "num_tokens": 80407342.0,
      "step": 36710
    },
    {
      "entropy": 5.632803726196289,
      "epoch": 3.6438070662961493,
      "grad_norm": 1.28125,
      "learning_rate": 0.000372055234919369,
      "loss": 4.943,
      "mean_token_accuracy": 0.1903953492641449,
      "num_tokens": 80418356.0,
      "step": 36715
    },
    {
      "entropy": 5.597694730758667,
      "epoch": 3.6443032949583167,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037202326807533724,
      "loss": 4.9506,
      "mean_token_accuracy": 0.19309968799352645,
      "num_tokens": 80428579.0,
      "step": 36720
    },
    {
      "entropy": 5.676651334762573,
      "epoch": 3.644799523620484,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003719912988253376,
      "loss": 5.0863,
      "mean_token_accuracy": 0.17806349247694014,
      "num_tokens": 80441494.0,
      "step": 36725
    },
    {
      "entropy": 5.682403182983398,
      "epoch": 3.6452957522826517,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037195932717016254,
      "loss": 5.0098,
      "mean_token_accuracy": 0.19286016821861268,
      "num_tokens": 80453766.0,
      "step": 36730
    },
    {
      "entropy": 5.7098071575164795,
      "epoch": 3.6457919809448196,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003719273531106053,
      "loss": 4.9208,
      "mean_token_accuracy": 0.19233210533857345,
      "num_tokens": 80464382.0,
      "step": 36735
    },
    {
      "entropy": 5.732416391372681,
      "epoch": 3.646288209606987,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037189537664745846,
      "loss": 5.0847,
      "mean_token_accuracy": 0.1802235797047615,
      "num_tokens": 80475605.0,
      "step": 36740
    },
    {
      "entropy": 5.682588529586792,
      "epoch": 3.6467844382691545,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003718633977815151,
      "loss": 4.9928,
      "mean_token_accuracy": 0.18397856801748275,
      "num_tokens": 80485612.0,
      "step": 36745
    },
    {
      "entropy": 5.688927698135376,
      "epoch": 3.647280666931322,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037183141651356814,
      "loss": 4.9871,
      "mean_token_accuracy": 0.19410325437784196,
      "num_tokens": 80497651.0,
      "step": 36750
    },
    {
      "entropy": 5.703217601776123,
      "epoch": 3.6477768955934895,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003717994328444108,
      "loss": 5.0736,
      "mean_token_accuracy": 0.19032970517873765,
      "num_tokens": 80509897.0,
      "step": 36755
    },
    {
      "entropy": 5.665568494796753,
      "epoch": 3.648273124255657,
      "grad_norm": 1.25,
      "learning_rate": 0.00037176744677483605,
      "loss": 5.0198,
      "mean_token_accuracy": 0.1894126996397972,
      "num_tokens": 80521883.0,
      "step": 36760
    },
    {
      "entropy": 5.63701024055481,
      "epoch": 3.6487693529178244,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037173545830563714,
      "loss": 4.904,
      "mean_token_accuracy": 0.19515258818864822,
      "num_tokens": 80532580.0,
      "step": 36765
    },
    {
      "entropy": 5.703812932968139,
      "epoch": 3.649265581579992,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003717034674376073,
      "loss": 5.0008,
      "mean_token_accuracy": 0.1878053218126297,
      "num_tokens": 80543163.0,
      "step": 36770
    },
    {
      "entropy": 5.655477094650268,
      "epoch": 3.6497618102421594,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037167147417153983,
      "loss": 5.0372,
      "mean_token_accuracy": 0.18858724981546401,
      "num_tokens": 80553721.0,
      "step": 36775
    },
    {
      "entropy": 5.705943632125854,
      "epoch": 3.650258038904327,
      "grad_norm": 1.5,
      "learning_rate": 0.00037163947850822816,
      "loss": 5.0253,
      "mean_token_accuracy": 0.18584948182106018,
      "num_tokens": 80564340.0,
      "step": 36780
    },
    {
      "entropy": 5.764291095733642,
      "epoch": 3.650754267566495,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037160748044846566,
      "loss": 5.0905,
      "mean_token_accuracy": 0.18812765032052994,
      "num_tokens": 80575669.0,
      "step": 36785
    },
    {
      "entropy": 5.685942029953003,
      "epoch": 3.6512504962286623,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003715754799930458,
      "loss": 4.9699,
      "mean_token_accuracy": 0.19038884490728378,
      "num_tokens": 80586946.0,
      "step": 36790
    },
    {
      "entropy": 5.613721418380737,
      "epoch": 3.6517467248908297,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003715434771427621,
      "loss": 4.9504,
      "mean_token_accuracy": 0.19326918125152587,
      "num_tokens": 80596876.0,
      "step": 36795
    },
    {
      "entropy": 5.602602338790893,
      "epoch": 3.652242953552997,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003715114718984083,
      "loss": 4.9863,
      "mean_token_accuracy": 0.18809085637331008,
      "num_tokens": 80608243.0,
      "step": 36800
    },
    {
      "entropy": 5.635700321197509,
      "epoch": 3.6527391822151647,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037147946426077785,
      "loss": 4.9876,
      "mean_token_accuracy": 0.18780736476182938,
      "num_tokens": 80618514.0,
      "step": 36805
    },
    {
      "entropy": 5.707713317871094,
      "epoch": 3.653235410877332,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003714474542306646,
      "loss": 5.0016,
      "mean_token_accuracy": 0.18129760324954985,
      "num_tokens": 80628686.0,
      "step": 36810
    },
    {
      "entropy": 5.6679493427276615,
      "epoch": 3.6537316395394996,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037141544180886233,
      "loss": 4.9156,
      "mean_token_accuracy": 0.18823148161172867,
      "num_tokens": 80639414.0,
      "step": 36815
    },
    {
      "entropy": 5.561517381668091,
      "epoch": 3.6542278682016676,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003713834269961649,
      "loss": 4.8981,
      "mean_token_accuracy": 0.19847663044929503,
      "num_tokens": 80652504.0,
      "step": 36820
    },
    {
      "entropy": 5.64625768661499,
      "epoch": 3.654724096863835,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037135140979336614,
      "loss": 5.0057,
      "mean_token_accuracy": 0.1861216977238655,
      "num_tokens": 80663923.0,
      "step": 36825
    },
    {
      "entropy": 5.6677093505859375,
      "epoch": 3.6552203255260025,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037131939020126,
      "loss": 4.9794,
      "mean_token_accuracy": 0.18999436348676682,
      "num_tokens": 80674150.0,
      "step": 36830
    },
    {
      "entropy": 5.74662938117981,
      "epoch": 3.65571655418817,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003712873682206406,
      "loss": 5.1057,
      "mean_token_accuracy": 0.18800454139709472,
      "num_tokens": 80685963.0,
      "step": 36835
    },
    {
      "entropy": 5.645763349533081,
      "epoch": 3.6562127828503375,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003712553438523019,
      "loss": 4.9216,
      "mean_token_accuracy": 0.1940694898366928,
      "num_tokens": 80696838.0,
      "step": 36840
    },
    {
      "entropy": 5.641505384445191,
      "epoch": 3.656709011512505,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037122331709703803,
      "loss": 5.0436,
      "mean_token_accuracy": 0.18788774609565734,
      "num_tokens": 80708333.0,
      "step": 36845
    },
    {
      "entropy": 5.623002672195435,
      "epoch": 3.6572052401746724,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003711912879556433,
      "loss": 4.9658,
      "mean_token_accuracy": 0.19618334621191025,
      "num_tokens": 80718875.0,
      "step": 36850
    },
    {
      "entropy": 5.74471697807312,
      "epoch": 3.65770146883684,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000371159256428912,
      "loss": 5.0161,
      "mean_token_accuracy": 0.18270941972732543,
      "num_tokens": 80728637.0,
      "step": 36855
    },
    {
      "entropy": 5.684250831604004,
      "epoch": 3.6581976974990074,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037112722251763817,
      "loss": 4.9812,
      "mean_token_accuracy": 0.18081855475902558,
      "num_tokens": 80740083.0,
      "step": 36860
    },
    {
      "entropy": 5.674335765838623,
      "epoch": 3.658693926161175,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003710951862226165,
      "loss": 5.0306,
      "mean_token_accuracy": 0.18776969909667968,
      "num_tokens": 80752412.0,
      "step": 36865
    },
    {
      "entropy": 5.5954736232757565,
      "epoch": 3.6591901548233428,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037106314754464125,
      "loss": 4.9442,
      "mean_token_accuracy": 0.19060127884149553,
      "num_tokens": 80761653.0,
      "step": 36870
    },
    {
      "entropy": 5.600051403045654,
      "epoch": 3.6596863834855102,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003710311064845069,
      "loss": 4.9112,
      "mean_token_accuracy": 0.1917671248316765,
      "num_tokens": 80771135.0,
      "step": 36875
    },
    {
      "entropy": 5.639930200576782,
      "epoch": 3.6601826121476777,
      "grad_norm": 1.40625,
      "learning_rate": 0.000370999063043008,
      "loss": 4.988,
      "mean_token_accuracy": 0.18861730098724366,
      "num_tokens": 80782119.0,
      "step": 36880
    },
    {
      "entropy": 5.659806871414185,
      "epoch": 3.660678840809845,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003709670172209393,
      "loss": 4.9825,
      "mean_token_accuracy": 0.18516145199537276,
      "num_tokens": 80793917.0,
      "step": 36885
    },
    {
      "entropy": 5.708960342407226,
      "epoch": 3.6611750694720127,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003709349690190952,
      "loss": 5.1039,
      "mean_token_accuracy": 0.1813017964363098,
      "num_tokens": 80807256.0,
      "step": 36890
    },
    {
      "entropy": 5.635494089126587,
      "epoch": 3.66167129813418,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003709029184382708,
      "loss": 4.9657,
      "mean_token_accuracy": 0.1923630177974701,
      "num_tokens": 80817224.0,
      "step": 36895
    },
    {
      "entropy": 5.634995317459106,
      "epoch": 3.6621675267963476,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003708708654792605,
      "loss": 4.9381,
      "mean_token_accuracy": 0.18657487630844116,
      "num_tokens": 80828692.0,
      "step": 36900
    },
    {
      "entropy": 5.661797046661377,
      "epoch": 3.6626637554585155,
      "grad_norm": 1.421875,
      "learning_rate": 0.00037083881014285944,
      "loss": 4.996,
      "mean_token_accuracy": 0.18986277729272844,
      "num_tokens": 80839201.0,
      "step": 36905
    },
    {
      "entropy": 5.725727224349976,
      "epoch": 3.663159984120683,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037080675242986234,
      "loss": 5.0665,
      "mean_token_accuracy": 0.18491508662700654,
      "num_tokens": 80853184.0,
      "step": 36910
    },
    {
      "entropy": 5.602328681945801,
      "epoch": 3.6636562127828505,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003707746923410642,
      "loss": 4.8961,
      "mean_token_accuracy": 0.18626351952552794,
      "num_tokens": 80862425.0,
      "step": 36915
    },
    {
      "entropy": 5.667153787612915,
      "epoch": 3.664152441445018,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037074262987726013,
      "loss": 4.9436,
      "mean_token_accuracy": 0.19019961804151536,
      "num_tokens": 80873652.0,
      "step": 36920
    },
    {
      "entropy": 5.637211179733276,
      "epoch": 3.6646486701071854,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003707105650392451,
      "loss": 4.996,
      "mean_token_accuracy": 0.1869590938091278,
      "num_tokens": 80884505.0,
      "step": 36925
    },
    {
      "entropy": 5.614148378372192,
      "epoch": 3.665144898769353,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003706784978278144,
      "loss": 4.9879,
      "mean_token_accuracy": 0.1876019224524498,
      "num_tokens": 80895728.0,
      "step": 36930
    },
    {
      "entropy": 5.7032890796661375,
      "epoch": 3.6656411274315204,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003706464282437629,
      "loss": 4.9945,
      "mean_token_accuracy": 0.18732025176286698,
      "num_tokens": 80907054.0,
      "step": 36935
    },
    {
      "entropy": 5.717277669906617,
      "epoch": 3.666137356093688,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003706143562878863,
      "loss": 4.9572,
      "mean_token_accuracy": 0.19727876782417297,
      "num_tokens": 80917331.0,
      "step": 36940
    },
    {
      "entropy": 5.6111249923706055,
      "epoch": 3.6666335847558553,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003705822819609796,
      "loss": 4.9521,
      "mean_token_accuracy": 0.1880301997065544,
      "num_tokens": 80929768.0,
      "step": 36945
    },
    {
      "entropy": 5.663305950164795,
      "epoch": 3.667129813418023,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003705502052638383,
      "loss": 5.0199,
      "mean_token_accuracy": 0.18170399367809295,
      "num_tokens": 80940226.0,
      "step": 36950
    },
    {
      "entropy": 5.603702640533447,
      "epoch": 3.6676260420801907,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037051812619725785,
      "loss": 4.9772,
      "mean_token_accuracy": 0.18922371417284012,
      "num_tokens": 80950622.0,
      "step": 36955
    },
    {
      "entropy": 5.647648477554322,
      "epoch": 3.668122270742358,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003704860447620335,
      "loss": 4.9738,
      "mean_token_accuracy": 0.18759166598320007,
      "num_tokens": 80961439.0,
      "step": 36960
    },
    {
      "entropy": 5.7066559314727785,
      "epoch": 3.6686184994045257,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003704539609589612,
      "loss": 5.0659,
      "mean_token_accuracy": 0.17921532541513444,
      "num_tokens": 80972898.0,
      "step": 36965
    },
    {
      "entropy": 5.603192949295044,
      "epoch": 3.669114728066693,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003704218747888363,
      "loss": 4.9267,
      "mean_token_accuracy": 0.198380646109581,
      "num_tokens": 80984758.0,
      "step": 36970
    },
    {
      "entropy": 5.678066062927246,
      "epoch": 3.6696109567288606,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037038978625245447,
      "loss": 4.9649,
      "mean_token_accuracy": 0.19780608266592026,
      "num_tokens": 80995987.0,
      "step": 36975
    },
    {
      "entropy": 5.684785270690918,
      "epoch": 3.670107185391028,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003703576953506115,
      "loss": 5.0426,
      "mean_token_accuracy": 0.18571450114250182,
      "num_tokens": 81006834.0,
      "step": 36980
    },
    {
      "entropy": 5.658966064453125,
      "epoch": 3.6706034140531956,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037032560208410313,
      "loss": 4.9937,
      "mean_token_accuracy": 0.18802850991487502,
      "num_tokens": 81016273.0,
      "step": 36985
    },
    {
      "entropy": 5.6285850524902346,
      "epoch": 3.6710996427153635,
      "grad_norm": 1.25,
      "learning_rate": 0.00037029350645372526,
      "loss": 4.9549,
      "mean_token_accuracy": 0.1910532683134079,
      "num_tokens": 81026292.0,
      "step": 36990
    },
    {
      "entropy": 5.636029767990112,
      "epoch": 3.671595871377531,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037026140846027383,
      "loss": 4.9861,
      "mean_token_accuracy": 0.18645793050527573,
      "num_tokens": 81036824.0,
      "step": 36995
    },
    {
      "entropy": 5.650810575485229,
      "epoch": 3.6720921000396984,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003702293081045448,
      "loss": 4.9667,
      "mean_token_accuracy": 0.1908635452389717,
      "num_tokens": 81047379.0,
      "step": 37000
    },
    {
      "entropy": 5.739211893081665,
      "epoch": 3.672588328701866,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037019720538733394,
      "loss": 5.0898,
      "mean_token_accuracy": 0.17980088740587236,
      "num_tokens": 81059097.0,
      "step": 37005
    },
    {
      "entropy": 5.694890880584717,
      "epoch": 3.6730845573640334,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003701651003094376,
      "loss": 4.995,
      "mean_token_accuracy": 0.1881688952445984,
      "num_tokens": 81069747.0,
      "step": 37010
    },
    {
      "entropy": 5.697894811630249,
      "epoch": 3.673580786026201,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003701329928716519,
      "loss": 5.0257,
      "mean_token_accuracy": 0.18798995465040208,
      "num_tokens": 81080947.0,
      "step": 37015
    },
    {
      "entropy": 5.718620109558105,
      "epoch": 3.6740770146883683,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037010088307477286,
      "loss": 4.9781,
      "mean_token_accuracy": 0.20223062187433244,
      "num_tokens": 81091736.0,
      "step": 37020
    },
    {
      "entropy": 5.660977125167847,
      "epoch": 3.674573243350536,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003700687709195969,
      "loss": 4.9768,
      "mean_token_accuracy": 0.19375938326120376,
      "num_tokens": 81103289.0,
      "step": 37025
    },
    {
      "entropy": 5.631534147262573,
      "epoch": 3.6750694720127033,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003700366564069203,
      "loss": 4.9376,
      "mean_token_accuracy": 0.20050323456525804,
      "num_tokens": 81116738.0,
      "step": 37030
    },
    {
      "entropy": 5.62336049079895,
      "epoch": 3.6755657006748708,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037000453953753943,
      "loss": 4.941,
      "mean_token_accuracy": 0.19692125171422958,
      "num_tokens": 81126642.0,
      "step": 37035
    },
    {
      "entropy": 5.7100550651550295,
      "epoch": 3.6760619293370382,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003699724203122506,
      "loss": 5.0413,
      "mean_token_accuracy": 0.1913362130522728,
      "num_tokens": 81137744.0,
      "step": 37040
    },
    {
      "entropy": 5.598317432403564,
      "epoch": 3.676558157999206,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003699402987318504,
      "loss": 4.855,
      "mean_token_accuracy": 0.19963425844907762,
      "num_tokens": 81149420.0,
      "step": 37045
    },
    {
      "entropy": 5.7236659049987795,
      "epoch": 3.6770543866613736,
      "grad_norm": 1.25,
      "learning_rate": 0.00036990817479713543,
      "loss": 5.0306,
      "mean_token_accuracy": 0.18722683936357498,
      "num_tokens": 81158516.0,
      "step": 37050
    },
    {
      "entropy": 5.610448694229126,
      "epoch": 3.677550615323541,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036987604850890224,
      "loss": 4.9704,
      "mean_token_accuracy": 0.1911156877875328,
      "num_tokens": 81170230.0,
      "step": 37055
    },
    {
      "entropy": 5.6326422691345215,
      "epoch": 3.6780468439857086,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036984391986794745,
      "loss": 4.9313,
      "mean_token_accuracy": 0.19436147063970566,
      "num_tokens": 81180841.0,
      "step": 37060
    },
    {
      "entropy": 5.679480457305909,
      "epoch": 3.678543072647876,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00036981178887506776,
      "loss": 5.022,
      "mean_token_accuracy": 0.18684101849794388,
      "num_tokens": 81191808.0,
      "step": 37065
    },
    {
      "entropy": 5.665472936630249,
      "epoch": 3.6790393013100435,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036977965553106014,
      "loss": 4.9541,
      "mean_token_accuracy": 0.18855551779270172,
      "num_tokens": 81201326.0,
      "step": 37070
    },
    {
      "entropy": 5.667396450042725,
      "epoch": 3.6795355299722114,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003697475198367212,
      "loss": 4.9419,
      "mean_token_accuracy": 0.19553165435791015,
      "num_tokens": 81213458.0,
      "step": 37075
    },
    {
      "entropy": 5.68696608543396,
      "epoch": 3.680031758634379,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000369715381792848,
      "loss": 5.0529,
      "mean_token_accuracy": 0.18658759891986848,
      "num_tokens": 81225133.0,
      "step": 37080
    },
    {
      "entropy": 5.618008232116699,
      "epoch": 3.6805279872965464,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003696832414002374,
      "loss": 5.0458,
      "mean_token_accuracy": 0.18365175873041154,
      "num_tokens": 81237592.0,
      "step": 37085
    },
    {
      "entropy": 5.728548240661621,
      "epoch": 3.681024215958714,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003696510986596864,
      "loss": 5.034,
      "mean_token_accuracy": 0.18982648551464082,
      "num_tokens": 81247527.0,
      "step": 37090
    },
    {
      "entropy": 5.741045618057251,
      "epoch": 3.6815204446208813,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036961895357199214,
      "loss": 5.1031,
      "mean_token_accuracy": 0.18062445968389512,
      "num_tokens": 81259109.0,
      "step": 37095
    },
    {
      "entropy": 5.673085069656372,
      "epoch": 3.682016673283049,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003695868061379518,
      "loss": 4.9996,
      "mean_token_accuracy": 0.18870845288038254,
      "num_tokens": 81269747.0,
      "step": 37100
    },
    {
      "entropy": 5.684997177124023,
      "epoch": 3.6825129019452163,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003695546563583623,
      "loss": 5.0389,
      "mean_token_accuracy": 0.17914542257785798,
      "num_tokens": 81279332.0,
      "step": 37105
    },
    {
      "entropy": 5.584290838241577,
      "epoch": 3.6830091306073838,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003695225042340212,
      "loss": 4.8947,
      "mean_token_accuracy": 0.20270537436008454,
      "num_tokens": 81290274.0,
      "step": 37110
    },
    {
      "entropy": 5.709970712661743,
      "epoch": 3.6835053592695512,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003694903497657256,
      "loss": 5.0029,
      "mean_token_accuracy": 0.18887594640254973,
      "num_tokens": 81301665.0,
      "step": 37115
    },
    {
      "entropy": 5.6787604808807375,
      "epoch": 3.6840015879317187,
      "grad_norm": 1.15625,
      "learning_rate": 0.000369458192954273,
      "loss": 4.968,
      "mean_token_accuracy": 0.18190814256668092,
      "num_tokens": 81311581.0,
      "step": 37120
    },
    {
      "entropy": 5.701435804367065,
      "epoch": 3.684497816593886,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003694260338004607,
      "loss": 5.0598,
      "mean_token_accuracy": 0.19222306162118913,
      "num_tokens": 81322526.0,
      "step": 37125
    },
    {
      "entropy": 5.6713306427001955,
      "epoch": 3.684994045256054,
      "grad_norm": 1.25,
      "learning_rate": 0.00036939387230508627,
      "loss": 4.9399,
      "mean_token_accuracy": 0.19097862839698793,
      "num_tokens": 81332132.0,
      "step": 37130
    },
    {
      "entropy": 5.698784208297729,
      "epoch": 3.6854902739182216,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036936170846894715,
      "loss": 4.9751,
      "mean_token_accuracy": 0.18827423453330994,
      "num_tokens": 81343425.0,
      "step": 37135
    },
    {
      "entropy": 5.642888116836548,
      "epoch": 3.685986502580389,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036932954229284106,
      "loss": 4.9364,
      "mean_token_accuracy": 0.19291832894086838,
      "num_tokens": 81354262.0,
      "step": 37140
    },
    {
      "entropy": 5.622450971603394,
      "epoch": 3.6864827312425565,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036929737377756563,
      "loss": 4.9396,
      "mean_token_accuracy": 0.18873242735862733,
      "num_tokens": 81365324.0,
      "step": 37145
    },
    {
      "entropy": 5.5862620830535885,
      "epoch": 3.686978959904724,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003692652029239184,
      "loss": 4.9275,
      "mean_token_accuracy": 0.20240933150053025,
      "num_tokens": 81376083.0,
      "step": 37150
    },
    {
      "entropy": 5.729483699798584,
      "epoch": 3.6874751885668915,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036923302973269727,
      "loss": 4.9833,
      "mean_token_accuracy": 0.18899511694908142,
      "num_tokens": 81386283.0,
      "step": 37155
    },
    {
      "entropy": 5.611904764175415,
      "epoch": 3.6879714172290594,
      "grad_norm": 1.125,
      "learning_rate": 0.0003692008542047001,
      "loss": 4.9189,
      "mean_token_accuracy": 0.19580068737268447,
      "num_tokens": 81397878.0,
      "step": 37160
    },
    {
      "entropy": 5.682695388793945,
      "epoch": 3.688467645891227,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003691686763407248,
      "loss": 5.0678,
      "mean_token_accuracy": 0.19112859964370726,
      "num_tokens": 81409464.0,
      "step": 37165
    },
    {
      "entropy": 5.732338905334473,
      "epoch": 3.6889638745533944,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003691364961415691,
      "loss": 5.1219,
      "mean_token_accuracy": 0.18510646522045135,
      "num_tokens": 81421352.0,
      "step": 37170
    },
    {
      "entropy": 5.647936248779297,
      "epoch": 3.689460103215562,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036910431360803124,
      "loss": 4.8515,
      "mean_token_accuracy": 0.19969881623983382,
      "num_tokens": 81431045.0,
      "step": 37175
    },
    {
      "entropy": 5.6485871315002445,
      "epoch": 3.6899563318777293,
      "grad_norm": 1.25,
      "learning_rate": 0.00036907212874090923,
      "loss": 4.8978,
      "mean_token_accuracy": 0.19905928820371627,
      "num_tokens": 81441772.0,
      "step": 37180
    },
    {
      "entropy": 5.539114952087402,
      "epoch": 3.690452560539897,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000369039941541001,
      "loss": 4.9029,
      "mean_token_accuracy": 0.19167352616786956,
      "num_tokens": 81451964.0,
      "step": 37185
    },
    {
      "entropy": 5.602676916122436,
      "epoch": 3.6909487892020643,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036900775200910497,
      "loss": 4.9164,
      "mean_token_accuracy": 0.19761267453432083,
      "num_tokens": 81462598.0,
      "step": 37190
    },
    {
      "entropy": 5.691830348968506,
      "epoch": 3.6914450178642317,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036897556014601926,
      "loss": 4.9825,
      "mean_token_accuracy": 0.18840638399124146,
      "num_tokens": 81474415.0,
      "step": 37195
    },
    {
      "entropy": 5.694869327545166,
      "epoch": 3.691941246526399,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036894336595254203,
      "loss": 4.9907,
      "mean_token_accuracy": 0.19155708700418472,
      "num_tokens": 81485510.0,
      "step": 37200
    },
    {
      "entropy": 5.677191495895386,
      "epoch": 3.6924374751885667,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036891116942947184,
      "loss": 5.0325,
      "mean_token_accuracy": 0.18706253170967102,
      "num_tokens": 81496103.0,
      "step": 37205
    },
    {
      "entropy": 5.689731454849243,
      "epoch": 3.692933703850734,
      "grad_norm": 1.078125,
      "learning_rate": 0.000368878970577607,
      "loss": 5.0689,
      "mean_token_accuracy": 0.1848679080605507,
      "num_tokens": 81507876.0,
      "step": 37210
    },
    {
      "entropy": 5.622617197036743,
      "epoch": 3.693429932512902,
      "grad_norm": 1.125,
      "learning_rate": 0.00036884676939774597,
      "loss": 4.9408,
      "mean_token_accuracy": 0.18931296616792678,
      "num_tokens": 81520615.0,
      "step": 37215
    },
    {
      "entropy": 5.613566970825195,
      "epoch": 3.6939261611750696,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036881456589068725,
      "loss": 4.9474,
      "mean_token_accuracy": 0.1907176896929741,
      "num_tokens": 81531614.0,
      "step": 37220
    },
    {
      "entropy": 5.665358781814575,
      "epoch": 3.694422389837237,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036878236005722954,
      "loss": 4.9624,
      "mean_token_accuracy": 0.18819052577018738,
      "num_tokens": 81541027.0,
      "step": 37225
    },
    {
      "entropy": 5.6600617408752445,
      "epoch": 3.6949186184994045,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003687501518981713,
      "loss": 4.9954,
      "mean_token_accuracy": 0.1864415779709816,
      "num_tokens": 81551569.0,
      "step": 37230
    },
    {
      "entropy": 5.657866954803467,
      "epoch": 3.695414847161572,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003687179414143114,
      "loss": 4.9408,
      "mean_token_accuracy": 0.19395449459552766,
      "num_tokens": 81561490.0,
      "step": 37235
    },
    {
      "entropy": 5.613549613952637,
      "epoch": 3.6959110758237395,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036868572860644845,
      "loss": 4.8804,
      "mean_token_accuracy": 0.19556474685668945,
      "num_tokens": 81572434.0,
      "step": 37240
    },
    {
      "entropy": 5.653287601470947,
      "epoch": 3.6964073044859074,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036865351347538123,
      "loss": 4.968,
      "mean_token_accuracy": 0.19022875130176545,
      "num_tokens": 81585135.0,
      "step": 37245
    },
    {
      "entropy": 5.597505855560303,
      "epoch": 3.696903533148075,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003686212960219088,
      "loss": 4.8729,
      "mean_token_accuracy": 0.20235977023839952,
      "num_tokens": 81596919.0,
      "step": 37250
    },
    {
      "entropy": 5.62336835861206,
      "epoch": 3.6973997618102423,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036858907624682987,
      "loss": 4.9902,
      "mean_token_accuracy": 0.18788765519857406,
      "num_tokens": 81608335.0,
      "step": 37255
    },
    {
      "entropy": 5.597444152832031,
      "epoch": 3.69789599047241,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003685568541509435,
      "loss": 4.9606,
      "mean_token_accuracy": 0.19015612155199052,
      "num_tokens": 81618940.0,
      "step": 37260
    },
    {
      "entropy": 5.618359518051148,
      "epoch": 3.6983922191345773,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036852462973504884,
      "loss": 4.9798,
      "mean_token_accuracy": 0.18893821388483048,
      "num_tokens": 81629215.0,
      "step": 37265
    },
    {
      "entropy": 5.685302495956421,
      "epoch": 3.6988884477967447,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003684924029999449,
      "loss": 5.0501,
      "mean_token_accuracy": 0.18615690618753433,
      "num_tokens": 81642201.0,
      "step": 37270
    },
    {
      "entropy": 5.668724632263183,
      "epoch": 3.6993846764589122,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036846017394643064,
      "loss": 4.8834,
      "mean_token_accuracy": 0.19696474373340606,
      "num_tokens": 81652307.0,
      "step": 37275
    },
    {
      "entropy": 5.673766613006592,
      "epoch": 3.6998809051210797,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003684279425753057,
      "loss": 5.0035,
      "mean_token_accuracy": 0.19097662419080735,
      "num_tokens": 81662904.0,
      "step": 37280
    },
    {
      "entropy": 5.667526197433472,
      "epoch": 3.700377133783247,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036839570888736893,
      "loss": 5.0601,
      "mean_token_accuracy": 0.18380693793296815,
      "num_tokens": 81674456.0,
      "step": 37285
    },
    {
      "entropy": 5.656871318817139,
      "epoch": 3.7008733624454146,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003683634728834199,
      "loss": 4.9224,
      "mean_token_accuracy": 0.19070159494876862,
      "num_tokens": 81685128.0,
      "step": 37290
    },
    {
      "entropy": 5.646929836273193,
      "epoch": 3.701369591107582,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036833123456425794,
      "loss": 4.9896,
      "mean_token_accuracy": 0.19484492838382722,
      "num_tokens": 81696661.0,
      "step": 37295
    },
    {
      "entropy": 5.710558652877808,
      "epoch": 3.70186581976975,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003682989939306824,
      "loss": 5.0434,
      "mean_token_accuracy": 0.18588415235280992,
      "num_tokens": 81708095.0,
      "step": 37300
    },
    {
      "entropy": 5.714158582687378,
      "epoch": 3.7023620484319175,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003682667509834929,
      "loss": 5.0879,
      "mean_token_accuracy": 0.18174783885478973,
      "num_tokens": 81719798.0,
      "step": 37305
    },
    {
      "entropy": 5.636020994186401,
      "epoch": 3.702858277094085,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036823450572348896,
      "loss": 4.9118,
      "mean_token_accuracy": 0.1921858161687851,
      "num_tokens": 81731041.0,
      "step": 37310
    },
    {
      "entropy": 5.701117610931396,
      "epoch": 3.7033545057562525,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003682022581514702,
      "loss": 4.9579,
      "mean_token_accuracy": 0.1884896367788315,
      "num_tokens": 81742198.0,
      "step": 37315
    },
    {
      "entropy": 5.669619178771972,
      "epoch": 3.70385073441842,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036817000826823616,
      "loss": 4.9158,
      "mean_token_accuracy": 0.1967358410358429,
      "num_tokens": 81752824.0,
      "step": 37320
    },
    {
      "entropy": 5.639733362197876,
      "epoch": 3.7043469630805874,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036813775607458677,
      "loss": 4.9593,
      "mean_token_accuracy": 0.1900804251432419,
      "num_tokens": 81764785.0,
      "step": 37325
    },
    {
      "entropy": 5.583317852020263,
      "epoch": 3.7048431917427553,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003681055015713217,
      "loss": 4.8677,
      "mean_token_accuracy": 0.2069334328174591,
      "num_tokens": 81776755.0,
      "step": 37330
    },
    {
      "entropy": 5.684888410568237,
      "epoch": 3.705339420404923,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003680732447592408,
      "loss": 4.9418,
      "mean_token_accuracy": 0.19153749644756318,
      "num_tokens": 81788088.0,
      "step": 37335
    },
    {
      "entropy": 5.647889852523804,
      "epoch": 3.7058356490670903,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036804098563914407,
      "loss": 5.03,
      "mean_token_accuracy": 0.18535881340503693,
      "num_tokens": 81800114.0,
      "step": 37340
    },
    {
      "entropy": 5.656650114059448,
      "epoch": 3.7063318777292578,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003680087242118313,
      "loss": 4.9839,
      "mean_token_accuracy": 0.18920914828777313,
      "num_tokens": 81811309.0,
      "step": 37345
    },
    {
      "entropy": 5.668011331558228,
      "epoch": 3.7068281063914252,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003679764604781026,
      "loss": 5.0426,
      "mean_token_accuracy": 0.18641431778669357,
      "num_tokens": 81821570.0,
      "step": 37350
    },
    {
      "entropy": 5.693282222747802,
      "epoch": 3.7073243350535927,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036794419443875805,
      "loss": 4.9215,
      "mean_token_accuracy": 0.1968041941523552,
      "num_tokens": 81832674.0,
      "step": 37355
    },
    {
      "entropy": 5.702622699737549,
      "epoch": 3.70782056371576,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003679119260945977,
      "loss": 5.0628,
      "mean_token_accuracy": 0.18710445165634154,
      "num_tokens": 81843463.0,
      "step": 37360
    },
    {
      "entropy": 5.634405422210693,
      "epoch": 3.7083167923779277,
      "grad_norm": 1.25,
      "learning_rate": 0.0003678796554464219,
      "loss": 4.9872,
      "mean_token_accuracy": 0.1963970646262169,
      "num_tokens": 81853512.0,
      "step": 37365
    },
    {
      "entropy": 5.7023955345153805,
      "epoch": 3.708813021040095,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036784738249503063,
      "loss": 5.0667,
      "mean_token_accuracy": 0.1830570176243782,
      "num_tokens": 81864703.0,
      "step": 37370
    },
    {
      "entropy": 5.607061862945557,
      "epoch": 3.7093092497022626,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003678151072412245,
      "loss": 4.9104,
      "mean_token_accuracy": 0.1927666649222374,
      "num_tokens": 81875387.0,
      "step": 37375
    },
    {
      "entropy": 5.614688444137573,
      "epoch": 3.70980547836443,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036778282968580355,
      "loss": 4.9143,
      "mean_token_accuracy": 0.19528497755527496,
      "num_tokens": 81885908.0,
      "step": 37380
    },
    {
      "entropy": 5.767579984664917,
      "epoch": 3.710301707026598,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003677505498295685,
      "loss": 5.0024,
      "mean_token_accuracy": 0.18755386173725128,
      "num_tokens": 81898256.0,
      "step": 37385
    },
    {
      "entropy": 5.610192918777466,
      "epoch": 3.7107979356887655,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036771826767331955,
      "loss": 4.8876,
      "mean_token_accuracy": 0.19669305831193923,
      "num_tokens": 81908993.0,
      "step": 37390
    },
    {
      "entropy": 5.553756666183472,
      "epoch": 3.711294164350933,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003676859832178575,
      "loss": 4.9443,
      "mean_token_accuracy": 0.19648912101984023,
      "num_tokens": 81919189.0,
      "step": 37395
    },
    {
      "entropy": 5.623913908004761,
      "epoch": 3.7117903930131004,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003676536964639827,
      "loss": 4.9243,
      "mean_token_accuracy": 0.19278713017702104,
      "num_tokens": 81929213.0,
      "step": 37400
    },
    {
      "entropy": 5.6715864658355715,
      "epoch": 3.712286621675268,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003676214074124958,
      "loss": 4.9749,
      "mean_token_accuracy": 0.1905434548854828,
      "num_tokens": 81940735.0,
      "step": 37405
    },
    {
      "entropy": 5.612921810150146,
      "epoch": 3.7127828503374354,
      "grad_norm": 1.125,
      "learning_rate": 0.00036758911606419773,
      "loss": 4.9285,
      "mean_token_accuracy": 0.19069941639900206,
      "num_tokens": 81951962.0,
      "step": 37410
    },
    {
      "entropy": 5.658599710464477,
      "epoch": 3.713279078999603,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003675568224198889,
      "loss": 4.9359,
      "mean_token_accuracy": 0.19442166686058043,
      "num_tokens": 81963328.0,
      "step": 37415
    },
    {
      "entropy": 5.5733623027801515,
      "epoch": 3.7137753076617708,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003675245264803705,
      "loss": 4.958,
      "mean_token_accuracy": 0.19029925912618637,
      "num_tokens": 81975318.0,
      "step": 37420
    },
    {
      "entropy": 5.630822038650512,
      "epoch": 3.7142715363239382,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003674922282464432,
      "loss": 4.9715,
      "mean_token_accuracy": 0.19181027114391327,
      "num_tokens": 81986036.0,
      "step": 37425
    },
    {
      "entropy": 5.776478910446167,
      "epoch": 3.7147677649861057,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003674599277189078,
      "loss": 5.0807,
      "mean_token_accuracy": 0.1815550521016121,
      "num_tokens": 81998709.0,
      "step": 37430
    },
    {
      "entropy": 5.709242630004883,
      "epoch": 3.715263993648273,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036742762489856554,
      "loss": 5.0448,
      "mean_token_accuracy": 0.18239523321390153,
      "num_tokens": 82010436.0,
      "step": 37435
    },
    {
      "entropy": 5.684441661834716,
      "epoch": 3.7157602223104407,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003673953197862173,
      "loss": 5.0419,
      "mean_token_accuracy": 0.18371253907680513,
      "num_tokens": 82022194.0,
      "step": 37440
    },
    {
      "entropy": 5.642509031295776,
      "epoch": 3.716256450972608,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036736301238266434,
      "loss": 4.9857,
      "mean_token_accuracy": 0.18845987617969512,
      "num_tokens": 82033105.0,
      "step": 37445
    },
    {
      "entropy": 5.619878435134888,
      "epoch": 3.7167526796347756,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036733070268870754,
      "loss": 4.922,
      "mean_token_accuracy": 0.18181778490543365,
      "num_tokens": 82043161.0,
      "step": 37450
    },
    {
      "entropy": 5.600627183914185,
      "epoch": 3.717248908296943,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003672983907051483,
      "loss": 4.8962,
      "mean_token_accuracy": 0.19573529660701752,
      "num_tokens": 82053889.0,
      "step": 37455
    },
    {
      "entropy": 5.645229291915894,
      "epoch": 3.7177451369591106,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036726607643278793,
      "loss": 4.9793,
      "mean_token_accuracy": 0.1895422652363777,
      "num_tokens": 82066230.0,
      "step": 37460
    },
    {
      "entropy": 5.654533767700196,
      "epoch": 3.718241365621278,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036723375987242763,
      "loss": 4.9506,
      "mean_token_accuracy": 0.19138676673173904,
      "num_tokens": 82077655.0,
      "step": 37465
    },
    {
      "entropy": 5.742851972579956,
      "epoch": 3.718737594283446,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036720144102486886,
      "loss": 5.0342,
      "mean_token_accuracy": 0.18159416913986207,
      "num_tokens": 82089765.0,
      "step": 37470
    },
    {
      "entropy": 5.76821985244751,
      "epoch": 3.7192338229456134,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036716911989091304,
      "loss": 5.0766,
      "mean_token_accuracy": 0.18032603859901428,
      "num_tokens": 82101095.0,
      "step": 37475
    },
    {
      "entropy": 5.602291440963745,
      "epoch": 3.719730051607781,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036713679647136163,
      "loss": 4.8845,
      "mean_token_accuracy": 0.19950062036514282,
      "num_tokens": 82111741.0,
      "step": 37480
    },
    {
      "entropy": 5.650509834289551,
      "epoch": 3.7202262802699484,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003671044707670161,
      "loss": 5.0069,
      "mean_token_accuracy": 0.18381445556879045,
      "num_tokens": 82122749.0,
      "step": 37485
    },
    {
      "entropy": 5.640545463562011,
      "epoch": 3.720722508932116,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036707214277867834,
      "loss": 4.9333,
      "mean_token_accuracy": 0.19365932792425156,
      "num_tokens": 82132032.0,
      "step": 37490
    },
    {
      "entropy": 5.642789697647094,
      "epoch": 3.7212187375942833,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003670398125071497,
      "loss": 5.0024,
      "mean_token_accuracy": 0.18716278970241546,
      "num_tokens": 82143169.0,
      "step": 37495
    },
    {
      "entropy": 5.728437662124634,
      "epoch": 3.721714966256451,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036700747995323214,
      "loss": 5.0426,
      "mean_token_accuracy": 0.18704945594072342,
      "num_tokens": 82154412.0,
      "step": 37500
    },
    {
      "entropy": 5.6068144798278805,
      "epoch": 3.7222111949186187,
      "grad_norm": 1.25,
      "learning_rate": 0.0003669751451177273,
      "loss": 4.9631,
      "mean_token_accuracy": 0.19291810989379882,
      "num_tokens": 82165473.0,
      "step": 37505
    },
    {
      "entropy": 5.634486293792724,
      "epoch": 3.722707423580786,
      "grad_norm": 1.1875,
      "learning_rate": 0.000366942808001437,
      "loss": 4.9821,
      "mean_token_accuracy": 0.18753681927919388,
      "num_tokens": 82175716.0,
      "step": 37510
    },
    {
      "entropy": 5.67425389289856,
      "epoch": 3.7232036522429537,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036691046860516327,
      "loss": 5.0115,
      "mean_token_accuracy": 0.18976165801286698,
      "num_tokens": 82185981.0,
      "step": 37515
    },
    {
      "entropy": 5.711289358139038,
      "epoch": 3.723699880905121,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036687812692970796,
      "loss": 5.0366,
      "mean_token_accuracy": 0.18299717009067534,
      "num_tokens": 82197336.0,
      "step": 37520
    },
    {
      "entropy": 5.686084938049317,
      "epoch": 3.7241961095672886,
      "grad_norm": 1.109375,
      "learning_rate": 0.000366845782975873,
      "loss": 5.0108,
      "mean_token_accuracy": 0.1901684358716011,
      "num_tokens": 82207895.0,
      "step": 37525
    },
    {
      "entropy": 5.6414327144622805,
      "epoch": 3.724692338229456,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036681343674446054,
      "loss": 4.9202,
      "mean_token_accuracy": 0.20133910328149796,
      "num_tokens": 82218618.0,
      "step": 37530
    },
    {
      "entropy": 5.628076028823853,
      "epoch": 3.7251885668916236,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003667810882362728,
      "loss": 4.9779,
      "mean_token_accuracy": 0.18582761585712432,
      "num_tokens": 82229803.0,
      "step": 37535
    },
    {
      "entropy": 5.596296739578247,
      "epoch": 3.725684795553791,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003667487374521117,
      "loss": 4.9623,
      "mean_token_accuracy": 0.18964765667915345,
      "num_tokens": 82239448.0,
      "step": 37540
    },
    {
      "entropy": 5.62379298210144,
      "epoch": 3.7261810242159585,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003667163843927797,
      "loss": 4.9788,
      "mean_token_accuracy": 0.18918479532003402,
      "num_tokens": 82250093.0,
      "step": 37545
    },
    {
      "entropy": 5.637485980987549,
      "epoch": 3.726677252878126,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036668402905907897,
      "loss": 4.9694,
      "mean_token_accuracy": 0.18695987910032272,
      "num_tokens": 82260479.0,
      "step": 37550
    },
    {
      "entropy": 5.676919937133789,
      "epoch": 3.727173481540294,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036665167145181186,
      "loss": 5.0119,
      "mean_token_accuracy": 0.18974278569221498,
      "num_tokens": 82272482.0,
      "step": 37555
    },
    {
      "entropy": 5.647672653198242,
      "epoch": 3.7276697102024614,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036661931157178085,
      "loss": 4.9717,
      "mean_token_accuracy": 0.18772079944610595,
      "num_tokens": 82282456.0,
      "step": 37560
    },
    {
      "entropy": 5.736656999588012,
      "epoch": 3.728165938864629,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036658694941978825,
      "loss": 5.0537,
      "mean_token_accuracy": 0.18780464977025985,
      "num_tokens": 82292832.0,
      "step": 37565
    },
    {
      "entropy": 5.690512371063233,
      "epoch": 3.7286621675267964,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036655458499663675,
      "loss": 4.9613,
      "mean_token_accuracy": 0.19081752896308898,
      "num_tokens": 82302841.0,
      "step": 37570
    },
    {
      "entropy": 5.661374711990357,
      "epoch": 3.729158396188964,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003665222183031288,
      "loss": 4.9843,
      "mean_token_accuracy": 0.19093142002820968,
      "num_tokens": 82314092.0,
      "step": 37575
    },
    {
      "entropy": 5.720933341979981,
      "epoch": 3.7296546248511313,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000366489849340067,
      "loss": 5.0899,
      "mean_token_accuracy": 0.18124665170907975,
      "num_tokens": 82325153.0,
      "step": 37580
    },
    {
      "entropy": 5.7517938137054445,
      "epoch": 3.730150853513299,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036645747810825415,
      "loss": 5.0147,
      "mean_token_accuracy": 0.19165251702070235,
      "num_tokens": 82336463.0,
      "step": 37585
    },
    {
      "entropy": 5.674136734008789,
      "epoch": 3.7306470821754667,
      "grad_norm": 1.125,
      "learning_rate": 0.00036642510460849285,
      "loss": 4.9812,
      "mean_token_accuracy": 0.19306952953338624,
      "num_tokens": 82347314.0,
      "step": 37590
    },
    {
      "entropy": 5.634308815002441,
      "epoch": 3.731143310837634,
      "grad_norm": 1.21875,
      "learning_rate": 0.000366392728841586,
      "loss": 4.9931,
      "mean_token_accuracy": 0.19190353900194168,
      "num_tokens": 82358582.0,
      "step": 37595
    },
    {
      "entropy": 5.661412191390991,
      "epoch": 3.7316395394998016,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003663603508083364,
      "loss": 5.0152,
      "mean_token_accuracy": 0.18669655472040175,
      "num_tokens": 82369132.0,
      "step": 37600
    },
    {
      "entropy": 5.680916404724121,
      "epoch": 3.732135768161969,
      "grad_norm": 1.265625,
      "learning_rate": 0.000366327970509547,
      "loss": 5.0499,
      "mean_token_accuracy": 0.18053992986679077,
      "num_tokens": 82380011.0,
      "step": 37605
    },
    {
      "entropy": 5.661443042755127,
      "epoch": 3.7326319968241366,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036629558794602063,
      "loss": 5.058,
      "mean_token_accuracy": 0.18762346655130385,
      "num_tokens": 82391113.0,
      "step": 37610
    },
    {
      "entropy": 5.689117908477783,
      "epoch": 3.733128225486304,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036626320311856046,
      "loss": 4.9872,
      "mean_token_accuracy": 0.18839479237794876,
      "num_tokens": 82402780.0,
      "step": 37615
    },
    {
      "entropy": 5.6337309837341305,
      "epoch": 3.7336244541484715,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036623081602796953,
      "loss": 4.9872,
      "mean_token_accuracy": 0.18465346992015838,
      "num_tokens": 82412005.0,
      "step": 37620
    },
    {
      "entropy": 5.664219331741333,
      "epoch": 3.734120682810639,
      "grad_norm": 1.25,
      "learning_rate": 0.00036619842667505096,
      "loss": 4.9641,
      "mean_token_accuracy": 0.19144028276205063,
      "num_tokens": 82420868.0,
      "step": 37625
    },
    {
      "entropy": 5.705910062789917,
      "epoch": 3.7346169114728065,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036616603506060786,
      "loss": 4.9919,
      "mean_token_accuracy": 0.1841302365064621,
      "num_tokens": 82430397.0,
      "step": 37630
    },
    {
      "entropy": 5.68662748336792,
      "epoch": 3.735113140134974,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003661336411854436,
      "loss": 4.9763,
      "mean_token_accuracy": 0.19377679377794266,
      "num_tokens": 82440567.0,
      "step": 37635
    },
    {
      "entropy": 5.684756135940551,
      "epoch": 3.7356093687971414,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003661012450503613,
      "loss": 4.9807,
      "mean_token_accuracy": 0.1872446969151497,
      "num_tokens": 82450920.0,
      "step": 37640
    },
    {
      "entropy": 5.619291353225708,
      "epoch": 3.7361055974593094,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036606884665616455,
      "loss": 4.9836,
      "mean_token_accuracy": 0.19113411158323287,
      "num_tokens": 82461741.0,
      "step": 37645
    },
    {
      "entropy": 5.653350019454956,
      "epoch": 3.736601826121477,
      "grad_norm": 1.25,
      "learning_rate": 0.0003660364460036566,
      "loss": 4.9825,
      "mean_token_accuracy": 0.1884313091635704,
      "num_tokens": 82472360.0,
      "step": 37650
    },
    {
      "entropy": 5.592975187301636,
      "epoch": 3.7370980547836443,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003660040430936409,
      "loss": 4.9323,
      "mean_token_accuracy": 0.18654619455337523,
      "num_tokens": 82482758.0,
      "step": 37655
    },
    {
      "entropy": 5.6190660953521725,
      "epoch": 3.737594283445812,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036597163792692107,
      "loss": 4.8986,
      "mean_token_accuracy": 0.19086253941059111,
      "num_tokens": 82494059.0,
      "step": 37660
    },
    {
      "entropy": 5.676612043380738,
      "epoch": 3.7380905121079793,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036593923050430067,
      "loss": 5.058,
      "mean_token_accuracy": 0.18783315271139145,
      "num_tokens": 82505503.0,
      "step": 37665
    },
    {
      "entropy": 5.71024374961853,
      "epoch": 3.7385867407701467,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036590682082658334,
      "loss": 5.0632,
      "mean_token_accuracy": 0.19160032272338867,
      "num_tokens": 82516669.0,
      "step": 37670
    },
    {
      "entropy": 5.648042821884156,
      "epoch": 3.7390829694323147,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036587440889457274,
      "loss": 4.9445,
      "mean_token_accuracy": 0.1879419282078743,
      "num_tokens": 82526717.0,
      "step": 37675
    },
    {
      "entropy": 5.723403835296631,
      "epoch": 3.739579198094482,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036584199470907254,
      "loss": 5.0475,
      "mean_token_accuracy": 0.18426769673824311,
      "num_tokens": 82536617.0,
      "step": 37680
    },
    {
      "entropy": 5.7036957263946535,
      "epoch": 3.7400754267566496,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003658095782708866,
      "loss": 5.0551,
      "mean_token_accuracy": 0.17894801646471023,
      "num_tokens": 82547858.0,
      "step": 37685
    },
    {
      "entropy": 5.643744659423828,
      "epoch": 3.740571655418817,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036577715958081885,
      "loss": 4.9874,
      "mean_token_accuracy": 0.18614918291568755,
      "num_tokens": 82559334.0,
      "step": 37690
    },
    {
      "entropy": 5.648807191848755,
      "epoch": 3.7410678840809846,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003657447386396732,
      "loss": 4.9818,
      "mean_token_accuracy": 0.18891673237085344,
      "num_tokens": 82570687.0,
      "step": 37695
    },
    {
      "entropy": 5.724458360671997,
      "epoch": 3.741564112743152,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003657123154482534,
      "loss": 5.0675,
      "mean_token_accuracy": 0.19171916097402572,
      "num_tokens": 82582030.0,
      "step": 37700
    },
    {
      "entropy": 5.584817790985108,
      "epoch": 3.7420603414053195,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036567989000736386,
      "loss": 4.8942,
      "mean_token_accuracy": 0.1971732646226883,
      "num_tokens": 82592828.0,
      "step": 37705
    },
    {
      "entropy": 5.636371040344239,
      "epoch": 3.742556570067487,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036564746231780836,
      "loss": 4.9536,
      "mean_token_accuracy": 0.19166154861450196,
      "num_tokens": 82603133.0,
      "step": 37710
    },
    {
      "entropy": 5.638922023773193,
      "epoch": 3.7430527987296545,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003656150323803911,
      "loss": 4.9316,
      "mean_token_accuracy": 0.1860197052359581,
      "num_tokens": 82612889.0,
      "step": 37715
    },
    {
      "entropy": 5.614591407775879,
      "epoch": 3.743549027391822,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036558260019591633,
      "loss": 4.9775,
      "mean_token_accuracy": 0.19120162427425386,
      "num_tokens": 82623647.0,
      "step": 37720
    },
    {
      "entropy": 5.637222242355347,
      "epoch": 3.7440452560539894,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036555016576518816,
      "loss": 5.0091,
      "mean_token_accuracy": 0.18817664980888366,
      "num_tokens": 82633875.0,
      "step": 37725
    },
    {
      "entropy": 5.669187831878662,
      "epoch": 3.7445414847161573,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003655177290890111,
      "loss": 4.9976,
      "mean_token_accuracy": 0.1932970777153969,
      "num_tokens": 82644596.0,
      "step": 37730
    },
    {
      "entropy": 5.697026252746582,
      "epoch": 3.745037713378325,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036548529016818936,
      "loss": 4.9953,
      "mean_token_accuracy": 0.1964670404791832,
      "num_tokens": 82656073.0,
      "step": 37735
    },
    {
      "entropy": 5.717899942398072,
      "epoch": 3.7455339420404923,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003654528490035274,
      "loss": 5.0338,
      "mean_token_accuracy": 0.1883445605635643,
      "num_tokens": 82666644.0,
      "step": 37740
    },
    {
      "entropy": 5.630418968200684,
      "epoch": 3.7460301707026598,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003654204055958298,
      "loss": 4.9252,
      "mean_token_accuracy": 0.19197118878364564,
      "num_tokens": 82677812.0,
      "step": 37745
    },
    {
      "entropy": 5.635228443145752,
      "epoch": 3.7465263993648272,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036538795994590093,
      "loss": 4.9911,
      "mean_token_accuracy": 0.19507923275232314,
      "num_tokens": 82689215.0,
      "step": 37750
    },
    {
      "entropy": 5.685511302947998,
      "epoch": 3.7470226280269947,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036535551205454535,
      "loss": 5.0344,
      "mean_token_accuracy": 0.1882889077067375,
      "num_tokens": 82701115.0,
      "step": 37755
    },
    {
      "entropy": 5.7028077125549315,
      "epoch": 3.7475188566891626,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036532306192256774,
      "loss": 5.008,
      "mean_token_accuracy": 0.1893739551305771,
      "num_tokens": 82712064.0,
      "step": 37760
    },
    {
      "entropy": 5.700047159194947,
      "epoch": 3.74801508535133,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003652906095507729,
      "loss": 5.0255,
      "mean_token_accuracy": 0.1860953077673912,
      "num_tokens": 82721515.0,
      "step": 37765
    },
    {
      "entropy": 5.6674213886260985,
      "epoch": 3.7485113140134976,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036525815493996557,
      "loss": 5.071,
      "mean_token_accuracy": 0.18165502101182937,
      "num_tokens": 82731185.0,
      "step": 37770
    },
    {
      "entropy": 5.6329751968383786,
      "epoch": 3.749007542675665,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036522569809095034,
      "loss": 4.9449,
      "mean_token_accuracy": 0.19501302540302276,
      "num_tokens": 82740609.0,
      "step": 37775
    },
    {
      "entropy": 5.7682188034057615,
      "epoch": 3.7495037713378325,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036519323900453235,
      "loss": 5.158,
      "mean_token_accuracy": 0.17995607554912568,
      "num_tokens": 82752737.0,
      "step": 37780
    },
    {
      "entropy": 5.7341437339782715,
      "epoch": 3.75,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036516077768151636,
      "loss": 5.0621,
      "mean_token_accuracy": 0.18080347627401352,
      "num_tokens": 82763585.0,
      "step": 37785
    },
    {
      "entropy": 5.650814580917358,
      "epoch": 3.7504962286621675,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003651283141227073,
      "loss": 4.9695,
      "mean_token_accuracy": 0.1850069984793663,
      "num_tokens": 82774173.0,
      "step": 37790
    },
    {
      "entropy": 5.635637283325195,
      "epoch": 3.750992457324335,
      "grad_norm": 1.125,
      "learning_rate": 0.0003650958483289102,
      "loss": 5.0066,
      "mean_token_accuracy": 0.19249774515628815,
      "num_tokens": 82786279.0,
      "step": 37795
    },
    {
      "entropy": 5.594544458389282,
      "epoch": 3.7514886859865024,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003650633803009303,
      "loss": 4.9356,
      "mean_token_accuracy": 0.1850703090429306,
      "num_tokens": 82798087.0,
      "step": 37800
    },
    {
      "entropy": 5.697307825088501,
      "epoch": 3.75198491464867,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036503091003957264,
      "loss": 4.989,
      "mean_token_accuracy": 0.1958964318037033,
      "num_tokens": 82809712.0,
      "step": 37805
    },
    {
      "entropy": 5.602424240112304,
      "epoch": 3.7524811433108374,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036499843754564235,
      "loss": 4.9085,
      "mean_token_accuracy": 0.20072097480297088,
      "num_tokens": 82821894.0,
      "step": 37810
    },
    {
      "entropy": 5.665308094024658,
      "epoch": 3.7529773719730053,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003649659628199448,
      "loss": 4.9629,
      "mean_token_accuracy": 0.19432020485401152,
      "num_tokens": 82833094.0,
      "step": 37815
    },
    {
      "entropy": 5.61919379234314,
      "epoch": 3.7534736006351728,
      "grad_norm": 1.125,
      "learning_rate": 0.0003649334858632852,
      "loss": 4.9521,
      "mean_token_accuracy": 0.18850786089897156,
      "num_tokens": 82843690.0,
      "step": 37820
    },
    {
      "entropy": 5.667911958694458,
      "epoch": 3.7539698292973402,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000364901006676469,
      "loss": 4.9967,
      "mean_token_accuracy": 0.19182425737380981,
      "num_tokens": 82855810.0,
      "step": 37825
    },
    {
      "entropy": 5.642170000076294,
      "epoch": 3.7544660579595077,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036486852526030147,
      "loss": 4.9079,
      "mean_token_accuracy": 0.19463543891906737,
      "num_tokens": 82866836.0,
      "step": 37830
    },
    {
      "entropy": 5.68073582649231,
      "epoch": 3.754962286621675,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003648360416155882,
      "loss": 5.0166,
      "mean_token_accuracy": 0.19363316744565964,
      "num_tokens": 82877645.0,
      "step": 37835
    },
    {
      "entropy": 5.686104965209961,
      "epoch": 3.7554585152838427,
      "grad_norm": 1.125,
      "learning_rate": 0.0003648035557431347,
      "loss": 4.9891,
      "mean_token_accuracy": 0.18841131627559662,
      "num_tokens": 82888638.0,
      "step": 37840
    },
    {
      "entropy": 5.62652702331543,
      "epoch": 3.7559547439460106,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036477106764374646,
      "loss": 4.954,
      "mean_token_accuracy": 0.19918221980333328,
      "num_tokens": 82898550.0,
      "step": 37845
    },
    {
      "entropy": 5.667391443252564,
      "epoch": 3.756450972608178,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003647385773182292,
      "loss": 5.0235,
      "mean_token_accuracy": 0.1965051844716072,
      "num_tokens": 82908926.0,
      "step": 37850
    },
    {
      "entropy": 5.71321325302124,
      "epoch": 3.7569472012703455,
      "grad_norm": 1.125,
      "learning_rate": 0.00036470608476738865,
      "loss": 5.0582,
      "mean_token_accuracy": 0.19431214928627014,
      "num_tokens": 82921294.0,
      "step": 37855
    },
    {
      "entropy": 5.713491296768188,
      "epoch": 3.757443429932513,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003646735899920304,
      "loss": 5.0291,
      "mean_token_accuracy": 0.18556933403015136,
      "num_tokens": 82931315.0,
      "step": 37860
    },
    {
      "entropy": 5.676315069198608,
      "epoch": 3.7579396585946805,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036464109299296047,
      "loss": 4.9911,
      "mean_token_accuracy": 0.18779811412096023,
      "num_tokens": 82942385.0,
      "step": 37865
    },
    {
      "entropy": 5.672364997863769,
      "epoch": 3.758435887256848,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003646085937709845,
      "loss": 5.0121,
      "mean_token_accuracy": 0.19322432726621627,
      "num_tokens": 82952803.0,
      "step": 37870
    },
    {
      "entropy": 5.665222692489624,
      "epoch": 3.7589321159190154,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003645760923269085,
      "loss": 4.9956,
      "mean_token_accuracy": 0.18257054388523103,
      "num_tokens": 82964143.0,
      "step": 37875
    },
    {
      "entropy": 5.659320545196533,
      "epoch": 3.759428344581183,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003645435886615385,
      "loss": 4.9431,
      "mean_token_accuracy": 0.18655326664447786,
      "num_tokens": 82975800.0,
      "step": 37880
    },
    {
      "entropy": 5.70634822845459,
      "epoch": 3.7599245732433504,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036451108277568044,
      "loss": 5.0375,
      "mean_token_accuracy": 0.18272067755460739,
      "num_tokens": 82987462.0,
      "step": 37885
    },
    {
      "entropy": 5.652184963226318,
      "epoch": 3.760420801905518,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036447857467014036,
      "loss": 4.9256,
      "mean_token_accuracy": 0.1945830225944519,
      "num_tokens": 82996754.0,
      "step": 37890
    },
    {
      "entropy": 5.6401759624481205,
      "epoch": 3.7609170305676853,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003644460643457245,
      "loss": 4.959,
      "mean_token_accuracy": 0.18690723776817322,
      "num_tokens": 83009026.0,
      "step": 37895
    },
    {
      "entropy": 5.733344650268554,
      "epoch": 3.7614132592298533,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036441355180323885,
      "loss": 5.0667,
      "mean_token_accuracy": 0.17935536056756973,
      "num_tokens": 83020451.0,
      "step": 37900
    },
    {
      "entropy": 5.649039268493652,
      "epoch": 3.7619094878920207,
      "grad_norm": 1.25,
      "learning_rate": 0.0003643810370434899,
      "loss": 4.9323,
      "mean_token_accuracy": 0.19630239605903627,
      "num_tokens": 83030822.0,
      "step": 37905
    },
    {
      "entropy": 5.645249366760254,
      "epoch": 3.762405716554188,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003643485200672838,
      "loss": 4.9524,
      "mean_token_accuracy": 0.19615046679973602,
      "num_tokens": 83042441.0,
      "step": 37910
    },
    {
      "entropy": 5.6549520015716555,
      "epoch": 3.7629019452163557,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036431600087542695,
      "loss": 4.9949,
      "mean_token_accuracy": 0.18324398696422578,
      "num_tokens": 83054386.0,
      "step": 37915
    },
    {
      "entropy": 5.656224632263184,
      "epoch": 3.763398173878523,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003642834794687257,
      "loss": 4.989,
      "mean_token_accuracy": 0.18745572865009308,
      "num_tokens": 83066221.0,
      "step": 37920
    },
    {
      "entropy": 5.630146646499634,
      "epoch": 3.7638944025406906,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003642509558479865,
      "loss": 5.0183,
      "mean_token_accuracy": 0.19264536947011948,
      "num_tokens": 83078080.0,
      "step": 37925
    },
    {
      "entropy": 5.68196702003479,
      "epoch": 3.7643906312028586,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036421843001401607,
      "loss": 4.9875,
      "mean_token_accuracy": 0.1980981186032295,
      "num_tokens": 83089486.0,
      "step": 37930
    },
    {
      "entropy": 5.6889416694641115,
      "epoch": 3.764886859865026,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036418590196762076,
      "loss": 5.0295,
      "mean_token_accuracy": 0.18949300199747085,
      "num_tokens": 83101253.0,
      "step": 37935
    },
    {
      "entropy": 5.634796762466431,
      "epoch": 3.7653830885271935,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036415337170960714,
      "loss": 4.9329,
      "mean_token_accuracy": 0.1964184522628784,
      "num_tokens": 83112098.0,
      "step": 37940
    },
    {
      "entropy": 5.659979629516601,
      "epoch": 3.765879317189361,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036412083924078223,
      "loss": 4.9499,
      "mean_token_accuracy": 0.18648993521928786,
      "num_tokens": 83122047.0,
      "step": 37945
    },
    {
      "entropy": 5.6602548122406,
      "epoch": 3.7663755458515285,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003640883045619523,
      "loss": 4.972,
      "mean_token_accuracy": 0.1869770348072052,
      "num_tokens": 83132525.0,
      "step": 37950
    },
    {
      "entropy": 5.684045076370239,
      "epoch": 3.766871774513696,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003640557676739245,
      "loss": 5.0241,
      "mean_token_accuracy": 0.1872224658727646,
      "num_tokens": 83143274.0,
      "step": 37955
    },
    {
      "entropy": 5.716189670562744,
      "epoch": 3.7673680031758634,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003640232285775056,
      "loss": 5.0533,
      "mean_token_accuracy": 0.19054912477731706,
      "num_tokens": 83154451.0,
      "step": 37960
    },
    {
      "entropy": 5.6436769485473635,
      "epoch": 3.767864231838031,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036399068727350245,
      "loss": 4.9848,
      "mean_token_accuracy": 0.18079931735992433,
      "num_tokens": 83165535.0,
      "step": 37965
    },
    {
      "entropy": 5.599324750900268,
      "epoch": 3.7683604605001984,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036395814376272194,
      "loss": 4.8687,
      "mean_token_accuracy": 0.2001212164759636,
      "num_tokens": 83175955.0,
      "step": 37970
    },
    {
      "entropy": 5.666354608535767,
      "epoch": 3.768856689162366,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003639255980459713,
      "loss": 5.05,
      "mean_token_accuracy": 0.18280087858438493,
      "num_tokens": 83186200.0,
      "step": 37975
    },
    {
      "entropy": 5.724642372131347,
      "epoch": 3.7693529178245333,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003638930501240572,
      "loss": 5.0123,
      "mean_token_accuracy": 0.18390716016292571,
      "num_tokens": 83197121.0,
      "step": 37980
    },
    {
      "entropy": 5.615513134002685,
      "epoch": 3.769849146486701,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036386049999778724,
      "loss": 4.935,
      "mean_token_accuracy": 0.19432902634143828,
      "num_tokens": 83208580.0,
      "step": 37985
    },
    {
      "entropy": 5.626739931106568,
      "epoch": 3.7703453751488687,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036382794766796824,
      "loss": 4.8919,
      "mean_token_accuracy": 0.20189288556575774,
      "num_tokens": 83218589.0,
      "step": 37990
    },
    {
      "entropy": 5.63201413154602,
      "epoch": 3.770841603811036,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036379539313540757,
      "loss": 4.998,
      "mean_token_accuracy": 0.18842644691467286,
      "num_tokens": 83229712.0,
      "step": 37995
    },
    {
      "entropy": 5.624884939193725,
      "epoch": 3.7713378324732036,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003637628364009125,
      "loss": 4.9326,
      "mean_token_accuracy": 0.20170795619487764,
      "num_tokens": 83240327.0,
      "step": 38000
    },
    {
      "entropy": 5.607044839859009,
      "epoch": 3.771834061135371,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036373027746529037,
      "loss": 4.9358,
      "mean_token_accuracy": 0.19402792304754257,
      "num_tokens": 83251966.0,
      "step": 38005
    },
    {
      "entropy": 5.63621768951416,
      "epoch": 3.7723302897975386,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003636977163293485,
      "loss": 4.907,
      "mean_token_accuracy": 0.18931994438171387,
      "num_tokens": 83261948.0,
      "step": 38010
    },
    {
      "entropy": 5.650218200683594,
      "epoch": 3.772826518459706,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003636651529938944,
      "loss": 5.0023,
      "mean_token_accuracy": 0.18848451972007751,
      "num_tokens": 83273172.0,
      "step": 38015
    },
    {
      "entropy": 5.610813999176026,
      "epoch": 3.773322747121874,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003636325874597356,
      "loss": 4.9176,
      "mean_token_accuracy": 0.19390421807765962,
      "num_tokens": 83283836.0,
      "step": 38020
    },
    {
      "entropy": 5.6560750007629395,
      "epoch": 3.7738189757840415,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036360001972767956,
      "loss": 4.997,
      "mean_token_accuracy": 0.18963751196861267,
      "num_tokens": 83294206.0,
      "step": 38025
    },
    {
      "entropy": 5.696073293685913,
      "epoch": 3.774315204446209,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036356744979853406,
      "loss": 4.9993,
      "mean_token_accuracy": 0.19508673399686813,
      "num_tokens": 83305066.0,
      "step": 38030
    },
    {
      "entropy": 5.62334418296814,
      "epoch": 3.7748114331083764,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036353487767310655,
      "loss": 4.9206,
      "mean_token_accuracy": 0.19412360489368438,
      "num_tokens": 83316169.0,
      "step": 38035
    },
    {
      "entropy": 5.700054168701172,
      "epoch": 3.775307661770544,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003635023033522049,
      "loss": 4.9872,
      "mean_token_accuracy": 0.19307406544685363,
      "num_tokens": 83327702.0,
      "step": 38040
    },
    {
      "entropy": 5.652945041656494,
      "epoch": 3.7758038904327114,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003634697268366368,
      "loss": 4.9618,
      "mean_token_accuracy": 0.1903967022895813,
      "num_tokens": 83339951.0,
      "step": 38045
    },
    {
      "entropy": 5.653659057617188,
      "epoch": 3.776300119094879,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003634371481272101,
      "loss": 4.9397,
      "mean_token_accuracy": 0.1958980605006218,
      "num_tokens": 83351869.0,
      "step": 38050
    },
    {
      "entropy": 5.668624496459961,
      "epoch": 3.7767963477570463,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003634045672247327,
      "loss": 5.0192,
      "mean_token_accuracy": 0.18465211242437363,
      "num_tokens": 83362715.0,
      "step": 38055
    },
    {
      "entropy": 5.643028354644775,
      "epoch": 3.777292576419214,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003633719841300126,
      "loss": 4.9453,
      "mean_token_accuracy": 0.1916283115744591,
      "num_tokens": 83372988.0,
      "step": 38060
    },
    {
      "entropy": 5.61022424697876,
      "epoch": 3.7777888050813813,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003633393988438576,
      "loss": 4.9234,
      "mean_token_accuracy": 0.196653912961483,
      "num_tokens": 83383172.0,
      "step": 38065
    },
    {
      "entropy": 5.591726160049438,
      "epoch": 3.778285033743549,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003633068113670759,
      "loss": 4.9161,
      "mean_token_accuracy": 0.1930517315864563,
      "num_tokens": 83395402.0,
      "step": 38070
    },
    {
      "entropy": 5.668253755569458,
      "epoch": 3.7787812624057167,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036327422170047565,
      "loss": 5.0784,
      "mean_token_accuracy": 0.1926574945449829,
      "num_tokens": 83406936.0,
      "step": 38075
    },
    {
      "entropy": 5.6930787563323975,
      "epoch": 3.779277491067884,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003632416298448648,
      "loss": 5.0736,
      "mean_token_accuracy": 0.1777994751930237,
      "num_tokens": 83420043.0,
      "step": 38080
    },
    {
      "entropy": 5.67606029510498,
      "epoch": 3.7797737197300516,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036320903580105176,
      "loss": 4.9604,
      "mean_token_accuracy": 0.19244359880685807,
      "num_tokens": 83430083.0,
      "step": 38085
    },
    {
      "entropy": 5.645617723464966,
      "epoch": 3.780269948392219,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036317643956984466,
      "loss": 4.8757,
      "mean_token_accuracy": 0.19896946996450424,
      "num_tokens": 83440992.0,
      "step": 38090
    },
    {
      "entropy": 5.685812473297119,
      "epoch": 3.7807661770543866,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036314384115205184,
      "loss": 4.9915,
      "mean_token_accuracy": 0.1839625507593155,
      "num_tokens": 83452534.0,
      "step": 38095
    },
    {
      "entropy": 5.593775653839112,
      "epoch": 3.781262405716554,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003631112405484817,
      "loss": 4.8932,
      "mean_token_accuracy": 0.19467955082654953,
      "num_tokens": 83463064.0,
      "step": 38100
    },
    {
      "entropy": 5.690944385528565,
      "epoch": 3.781758634378722,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036307863775994275,
      "loss": 5.0049,
      "mean_token_accuracy": 0.18571866005659105,
      "num_tokens": 83474426.0,
      "step": 38105
    },
    {
      "entropy": 5.639050006866455,
      "epoch": 3.7822548630408894,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036304603278724325,
      "loss": 5.0167,
      "mean_token_accuracy": 0.1899923101067543,
      "num_tokens": 83485896.0,
      "step": 38110
    },
    {
      "entropy": 5.682773447036743,
      "epoch": 3.782751091703057,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036301342563119194,
      "loss": 5.035,
      "mean_token_accuracy": 0.18652884066104888,
      "num_tokens": 83496257.0,
      "step": 38115
    },
    {
      "entropy": 5.671196317672729,
      "epoch": 3.7832473203652244,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003629808162925973,
      "loss": 4.9829,
      "mean_token_accuracy": 0.19008827656507493,
      "num_tokens": 83506500.0,
      "step": 38120
    },
    {
      "entropy": 5.7323596477508545,
      "epoch": 3.783743549027392,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036294820477226797,
      "loss": 4.9752,
      "mean_token_accuracy": 0.18799030929803848,
      "num_tokens": 83517226.0,
      "step": 38125
    },
    {
      "entropy": 5.639843559265136,
      "epoch": 3.7842397776895593,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003629155910710127,
      "loss": 4.9702,
      "mean_token_accuracy": 0.19414965212345123,
      "num_tokens": 83527307.0,
      "step": 38130
    },
    {
      "entropy": 5.648539733886719,
      "epoch": 3.784736006351727,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003628829751896403,
      "loss": 4.9705,
      "mean_token_accuracy": 0.19425373673439025,
      "num_tokens": 83539252.0,
      "step": 38135
    },
    {
      "entropy": 5.681300640106201,
      "epoch": 3.7852322350138943,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036285035712895934,
      "loss": 5.0046,
      "mean_token_accuracy": 0.1874298632144928,
      "num_tokens": 83551791.0,
      "step": 38140
    },
    {
      "entropy": 5.656926870346069,
      "epoch": 3.7857284636760618,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036281773688977894,
      "loss": 4.92,
      "mean_token_accuracy": 0.18976954966783524,
      "num_tokens": 83561545.0,
      "step": 38145
    },
    {
      "entropy": 5.625182962417602,
      "epoch": 3.7862246923382292,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003627851144729078,
      "loss": 4.9151,
      "mean_token_accuracy": 0.19300140142440797,
      "num_tokens": 83572050.0,
      "step": 38150
    },
    {
      "entropy": 5.639033794403076,
      "epoch": 3.786720921000397,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036275248987915497,
      "loss": 5.0233,
      "mean_token_accuracy": 0.18994447141885756,
      "num_tokens": 83583177.0,
      "step": 38155
    },
    {
      "entropy": 5.653098154067993,
      "epoch": 3.7872171496625646,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036271986310932957,
      "loss": 4.9635,
      "mean_token_accuracy": 0.18971500992774964,
      "num_tokens": 83592906.0,
      "step": 38160
    },
    {
      "entropy": 5.643007040023804,
      "epoch": 3.787713378324732,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003626872341642406,
      "loss": 5.0172,
      "mean_token_accuracy": 0.18868179023265838,
      "num_tokens": 83602658.0,
      "step": 38165
    },
    {
      "entropy": 5.659565782546997,
      "epoch": 3.7882096069868996,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003626546030446971,
      "loss": 4.9915,
      "mean_token_accuracy": 0.19438790529966354,
      "num_tokens": 83612774.0,
      "step": 38170
    },
    {
      "entropy": 5.691502571105957,
      "epoch": 3.788705835649067,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036262196975150835,
      "loss": 4.9238,
      "mean_token_accuracy": 0.20065997093915938,
      "num_tokens": 83624284.0,
      "step": 38175
    },
    {
      "entropy": 5.67800464630127,
      "epoch": 3.7892020643112345,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003625893342854836,
      "loss": 5.0335,
      "mean_token_accuracy": 0.18718450963497163,
      "num_tokens": 83636850.0,
      "step": 38180
    },
    {
      "entropy": 5.661615228652954,
      "epoch": 3.789698292973402,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036255669664743195,
      "loss": 4.9947,
      "mean_token_accuracy": 0.18499500155448914,
      "num_tokens": 83647562.0,
      "step": 38185
    },
    {
      "entropy": 5.658803939819336,
      "epoch": 3.79019452163557,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036252405683816306,
      "loss": 4.9596,
      "mean_token_accuracy": 0.19145446866750718,
      "num_tokens": 83659020.0,
      "step": 38190
    },
    {
      "entropy": 5.6743385791778564,
      "epoch": 3.7906907502977374,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003624914148584861,
      "loss": 5.0234,
      "mean_token_accuracy": 0.18676867634057998,
      "num_tokens": 83670162.0,
      "step": 38195
    },
    {
      "entropy": 5.607451725006103,
      "epoch": 3.791186978959905,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003624587707092105,
      "loss": 4.9026,
      "mean_token_accuracy": 0.19101236909627914,
      "num_tokens": 83680609.0,
      "step": 38200
    },
    {
      "entropy": 5.68869948387146,
      "epoch": 3.7916832076220723,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036242612439114593,
      "loss": 4.9364,
      "mean_token_accuracy": 0.19141850918531417,
      "num_tokens": 83690912.0,
      "step": 38205
    },
    {
      "entropy": 5.665909099578857,
      "epoch": 3.79217943628424,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036239347590510186,
      "loss": 5.0291,
      "mean_token_accuracy": 0.1842488944530487,
      "num_tokens": 83702022.0,
      "step": 38210
    },
    {
      "entropy": 5.700987768173218,
      "epoch": 3.7926756649464073,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003623608252518879,
      "loss": 5.0693,
      "mean_token_accuracy": 0.18444106876850128,
      "num_tokens": 83713836.0,
      "step": 38215
    },
    {
      "entropy": 5.700737810134887,
      "epoch": 3.7931718936085748,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003623281724323137,
      "loss": 5.042,
      "mean_token_accuracy": 0.18903270214796067,
      "num_tokens": 83725038.0,
      "step": 38220
    },
    {
      "entropy": 5.650905418395996,
      "epoch": 3.7936681222707422,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000362295517447189,
      "loss": 4.9246,
      "mean_token_accuracy": 0.19296955168247223,
      "num_tokens": 83734984.0,
      "step": 38225
    },
    {
      "entropy": 5.684618854522705,
      "epoch": 3.7941643509329097,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003622628602973236,
      "loss": 4.9916,
      "mean_token_accuracy": 0.18919550478458405,
      "num_tokens": 83746323.0,
      "step": 38230
    },
    {
      "entropy": 5.575783538818359,
      "epoch": 3.794660579595077,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036223020098352724,
      "loss": 4.9157,
      "mean_token_accuracy": 0.19179998189210892,
      "num_tokens": 83758334.0,
      "step": 38235
    },
    {
      "entropy": 5.655095005035401,
      "epoch": 3.7951568082572447,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003621975395066099,
      "loss": 4.9334,
      "mean_token_accuracy": 0.18987203985452653,
      "num_tokens": 83769022.0,
      "step": 38240
    },
    {
      "entropy": 5.690007877349854,
      "epoch": 3.7956530369194126,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00036216487586738144,
      "loss": 5.0167,
      "mean_token_accuracy": 0.19248649924993516,
      "num_tokens": 83780682.0,
      "step": 38245
    },
    {
      "entropy": 5.680341148376465,
      "epoch": 3.79614926558158,
      "grad_norm": 1.265625,
      "learning_rate": 0.000362132210066652,
      "loss": 4.9965,
      "mean_token_accuracy": 0.18782267719507217,
      "num_tokens": 83790936.0,
      "step": 38250
    },
    {
      "entropy": 5.679314613342285,
      "epoch": 3.7966454942437475,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036209954210523137,
      "loss": 5.0449,
      "mean_token_accuracy": 0.18117585629224778,
      "num_tokens": 83802498.0,
      "step": 38255
    },
    {
      "entropy": 5.596251058578491,
      "epoch": 3.797141722905915,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003620668719839298,
      "loss": 4.9549,
      "mean_token_accuracy": 0.19032669067382812,
      "num_tokens": 83813269.0,
      "step": 38260
    },
    {
      "entropy": 5.6295116424560545,
      "epoch": 3.7976379515680825,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036203419970355747,
      "loss": 4.9632,
      "mean_token_accuracy": 0.18519876301288604,
      "num_tokens": 83822692.0,
      "step": 38265
    },
    {
      "entropy": 5.668659925460815,
      "epoch": 3.79813418023025,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003620015252649245,
      "loss": 5.0486,
      "mean_token_accuracy": 0.18948860764503478,
      "num_tokens": 83834375.0,
      "step": 38270
    },
    {
      "entropy": 5.728692436218262,
      "epoch": 3.798630408892418,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036196884866884115,
      "loss": 5.0924,
      "mean_token_accuracy": 0.18607541620731355,
      "num_tokens": 83846328.0,
      "step": 38275
    },
    {
      "entropy": 5.657659196853638,
      "epoch": 3.7991266375545854,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036193616991611784,
      "loss": 4.9823,
      "mean_token_accuracy": 0.1851368710398674,
      "num_tokens": 83857086.0,
      "step": 38280
    },
    {
      "entropy": 5.726441526412964,
      "epoch": 3.799622866216753,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036190348900756475,
      "loss": 5.0507,
      "mean_token_accuracy": 0.18597138673067093,
      "num_tokens": 83867152.0,
      "step": 38285
    },
    {
      "entropy": 5.711603784561158,
      "epoch": 3.8001190948789203,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003618708059439924,
      "loss": 4.9902,
      "mean_token_accuracy": 0.19396827220916749,
      "num_tokens": 83878960.0,
      "step": 38290
    },
    {
      "entropy": 5.625694942474365,
      "epoch": 3.8006153235410878,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003618381207262113,
      "loss": 4.9162,
      "mean_token_accuracy": 0.19171128571033477,
      "num_tokens": 83890656.0,
      "step": 38295
    },
    {
      "entropy": 5.643764781951904,
      "epoch": 3.8011115522032553,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003618054333550319,
      "loss": 4.9763,
      "mean_token_accuracy": 0.1922197461128235,
      "num_tokens": 83901099.0,
      "step": 38300
    },
    {
      "entropy": 5.634800720214844,
      "epoch": 3.8016077808654227,
      "grad_norm": 1.25,
      "learning_rate": 0.00036177274383126487,
      "loss": 5.0133,
      "mean_token_accuracy": 0.1929176241159439,
      "num_tokens": 83912016.0,
      "step": 38305
    },
    {
      "entropy": 5.690559053421021,
      "epoch": 3.80210400952759,
      "grad_norm": 1.25,
      "learning_rate": 0.0003617400521557207,
      "loss": 5.0314,
      "mean_token_accuracy": 0.17923390716314316,
      "num_tokens": 83923198.0,
      "step": 38310
    },
    {
      "entropy": 5.6654235363006595,
      "epoch": 3.8026002381897577,
      "grad_norm": 1.25,
      "learning_rate": 0.0003617073583292101,
      "loss": 5.0534,
      "mean_token_accuracy": 0.1827910929918289,
      "num_tokens": 83935310.0,
      "step": 38315
    },
    {
      "entropy": 5.625455379486084,
      "epoch": 3.803096466851925,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036167466235254394,
      "loss": 4.9577,
      "mean_token_accuracy": 0.19161336719989777,
      "num_tokens": 83947104.0,
      "step": 38320
    },
    {
      "entropy": 5.652024030685425,
      "epoch": 3.8035926955140926,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036164196422653293,
      "loss": 4.9768,
      "mean_token_accuracy": 0.1838146150112152,
      "num_tokens": 83957263.0,
      "step": 38325
    },
    {
      "entropy": 5.622638893127442,
      "epoch": 3.8040889241762605,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036160926395198786,
      "loss": 4.9197,
      "mean_token_accuracy": 0.19679610282182694,
      "num_tokens": 83970680.0,
      "step": 38330
    },
    {
      "entropy": 5.726798057556152,
      "epoch": 3.804585152838428,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003615765615297197,
      "loss": 5.0303,
      "mean_token_accuracy": 0.18798332214355468,
      "num_tokens": 83982456.0,
      "step": 38335
    },
    {
      "entropy": 5.652339792251587,
      "epoch": 3.8050813815005955,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003615438569605394,
      "loss": 5.0547,
      "mean_token_accuracy": 0.18568722605705262,
      "num_tokens": 83993303.0,
      "step": 38340
    },
    {
      "entropy": 5.708068418502807,
      "epoch": 3.805577610162763,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036151115024525777,
      "loss": 5.0147,
      "mean_token_accuracy": 0.1859205648303032,
      "num_tokens": 84004701.0,
      "step": 38345
    },
    {
      "entropy": 5.714903211593628,
      "epoch": 3.8060738388249304,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036147844138468616,
      "loss": 5.0573,
      "mean_token_accuracy": 0.18815738409757615,
      "num_tokens": 84014779.0,
      "step": 38350
    },
    {
      "entropy": 5.67008638381958,
      "epoch": 3.806570067487098,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003614457303796356,
      "loss": 5.0618,
      "mean_token_accuracy": 0.1813747137784958,
      "num_tokens": 84025309.0,
      "step": 38355
    },
    {
      "entropy": 5.730411005020142,
      "epoch": 3.807066296149266,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003614130172309171,
      "loss": 5.1431,
      "mean_token_accuracy": 0.18192608654499054,
      "num_tokens": 84037406.0,
      "step": 38360
    },
    {
      "entropy": 5.729374980926513,
      "epoch": 3.8075625248114333,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003613803019393419,
      "loss": 5.0417,
      "mean_token_accuracy": 0.1868743970990181,
      "num_tokens": 84048439.0,
      "step": 38365
    },
    {
      "entropy": 5.710426378250122,
      "epoch": 3.808058753473601,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036134758450572153,
      "loss": 5.0075,
      "mean_token_accuracy": 0.18388061076402665,
      "num_tokens": 84059681.0,
      "step": 38370
    },
    {
      "entropy": 5.755184316635132,
      "epoch": 3.8085549821357683,
      "grad_norm": 1.125,
      "learning_rate": 0.000361314864930867,
      "loss": 5.0761,
      "mean_token_accuracy": 0.18559758216142655,
      "num_tokens": 84069945.0,
      "step": 38375
    },
    {
      "entropy": 5.8043914318084715,
      "epoch": 3.8090512107979357,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003612821432155898,
      "loss": 5.1428,
      "mean_token_accuracy": 0.17623318880796432,
      "num_tokens": 84082440.0,
      "step": 38380
    },
    {
      "entropy": 5.658246803283691,
      "epoch": 3.809547439460103,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003612494193607015,
      "loss": 4.9189,
      "mean_token_accuracy": 0.2011547639966011,
      "num_tokens": 84093431.0,
      "step": 38385
    },
    {
      "entropy": 5.68376407623291,
      "epoch": 3.8100436681222707,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003612166933670134,
      "loss": 5.0536,
      "mean_token_accuracy": 0.18519871532917023,
      "num_tokens": 84105904.0,
      "step": 38390
    },
    {
      "entropy": 5.688676881790161,
      "epoch": 3.810539896784438,
      "grad_norm": 1.078125,
      "learning_rate": 0.000361183965235337,
      "loss": 5.0707,
      "mean_token_accuracy": 0.1844048410654068,
      "num_tokens": 84117967.0,
      "step": 38395
    },
    {
      "entropy": 5.690168142318726,
      "epoch": 3.8110361254466056,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003611512349664841,
      "loss": 4.909,
      "mean_token_accuracy": 0.19298657029867172,
      "num_tokens": 84129287.0,
      "step": 38400
    },
    {
      "entropy": 5.782217121124267,
      "epoch": 3.811532354108773,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003611185025612661,
      "loss": 5.0437,
      "mean_token_accuracy": 0.18430473059415817,
      "num_tokens": 84140291.0,
      "step": 38405
    },
    {
      "entropy": 5.767177581787109,
      "epoch": 3.8120285827709406,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003610857680204949,
      "loss": 5.0373,
      "mean_token_accuracy": 0.1910094901919365,
      "num_tokens": 84150955.0,
      "step": 38410
    },
    {
      "entropy": 5.64302887916565,
      "epoch": 3.8125248114331085,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036105303134498207,
      "loss": 4.9224,
      "mean_token_accuracy": 0.1938304513692856,
      "num_tokens": 84161665.0,
      "step": 38415
    },
    {
      "entropy": 5.615896844863892,
      "epoch": 3.813021040095276,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036102029253553945,
      "loss": 5.0222,
      "mean_token_accuracy": 0.18804895132780075,
      "num_tokens": 84172225.0,
      "step": 38420
    },
    {
      "entropy": 5.644429492950439,
      "epoch": 3.8135172687574435,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003609875515929791,
      "loss": 4.9803,
      "mean_token_accuracy": 0.18601789474487304,
      "num_tokens": 84181877.0,
      "step": 38425
    },
    {
      "entropy": 5.625559377670288,
      "epoch": 3.814013497419611,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003609548085181126,
      "loss": 4.9567,
      "mean_token_accuracy": 0.18912603557109833,
      "num_tokens": 84193342.0,
      "step": 38430
    },
    {
      "entropy": 5.624442720413208,
      "epoch": 3.8145097260817784,
      "grad_norm": 1.125,
      "learning_rate": 0.0003609220633117521,
      "loss": 4.9418,
      "mean_token_accuracy": 0.19122058749198914,
      "num_tokens": 84204719.0,
      "step": 38435
    },
    {
      "entropy": 5.642532634735107,
      "epoch": 3.815005954743946,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003608893159747097,
      "loss": 4.9157,
      "mean_token_accuracy": 0.18971433788537978,
      "num_tokens": 84214931.0,
      "step": 38440
    },
    {
      "entropy": 5.6310759544372555,
      "epoch": 3.815502183406114,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036085656650779716,
      "loss": 4.9788,
      "mean_token_accuracy": 0.19108003824949266,
      "num_tokens": 84227429.0,
      "step": 38445
    },
    {
      "entropy": 5.644613647460938,
      "epoch": 3.8159984120682813,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036082381491182685,
      "loss": 4.9777,
      "mean_token_accuracy": 0.18911141008138657,
      "num_tokens": 84236917.0,
      "step": 38450
    },
    {
      "entropy": 5.638164424896241,
      "epoch": 3.8164946407304488,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003607910611876109,
      "loss": 4.9789,
      "mean_token_accuracy": 0.1903209090232849,
      "num_tokens": 84247964.0,
      "step": 38455
    },
    {
      "entropy": 5.693821954727173,
      "epoch": 3.8169908693926162,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003607583053359615,
      "loss": 5.0671,
      "mean_token_accuracy": 0.18305929005146027,
      "num_tokens": 84259559.0,
      "step": 38460
    },
    {
      "entropy": 5.677792882919311,
      "epoch": 3.8174870980547837,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003607255473576908,
      "loss": 4.9367,
      "mean_token_accuracy": 0.19484805464744567,
      "num_tokens": 84270849.0,
      "step": 38465
    },
    {
      "entropy": 5.715728759765625,
      "epoch": 3.817983326716951,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036069278725361144,
      "loss": 4.9357,
      "mean_token_accuracy": 0.19628948271274566,
      "num_tokens": 84281977.0,
      "step": 38470
    },
    {
      "entropy": 5.654728889465332,
      "epoch": 3.8184795553791187,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036066002502453556,
      "loss": 4.989,
      "mean_token_accuracy": 0.1853844165802002,
      "num_tokens": 84294289.0,
      "step": 38475
    },
    {
      "entropy": 5.6274512767791744,
      "epoch": 3.818975784041286,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003606272606712755,
      "loss": 4.9723,
      "mean_token_accuracy": 0.1901981458067894,
      "num_tokens": 84304114.0,
      "step": 38480
    },
    {
      "entropy": 5.603798055648804,
      "epoch": 3.8194720127034536,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036059449419464397,
      "loss": 4.9135,
      "mean_token_accuracy": 0.1881068989634514,
      "num_tokens": 84314467.0,
      "step": 38485
    },
    {
      "entropy": 5.7432328224182125,
      "epoch": 3.819968241365621,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003605617255954535,
      "loss": 5.0749,
      "mean_token_accuracy": 0.18174135386943818,
      "num_tokens": 84325565.0,
      "step": 38490
    },
    {
      "entropy": 5.654674434661866,
      "epoch": 3.8204644700277886,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003605289548745164,
      "loss": 4.9623,
      "mean_token_accuracy": 0.18646515905857086,
      "num_tokens": 84336578.0,
      "step": 38495
    },
    {
      "entropy": 5.674659299850464,
      "epoch": 3.8209606986899565,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003604961820326457,
      "loss": 4.9672,
      "mean_token_accuracy": 0.18896415680646897,
      "num_tokens": 84347447.0,
      "step": 38500
    },
    {
      "entropy": 5.615505123138428,
      "epoch": 3.821456927352124,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036046340707065387,
      "loss": 4.9542,
      "mean_token_accuracy": 0.19354594647884368,
      "num_tokens": 84359527.0,
      "step": 38505
    },
    {
      "entropy": 5.652718019485474,
      "epoch": 3.8219531560142914,
      "grad_norm": 1.25,
      "learning_rate": 0.0003604306299893537,
      "loss": 4.9888,
      "mean_token_accuracy": 0.18504593670368194,
      "num_tokens": 84369653.0,
      "step": 38510
    },
    {
      "entropy": 5.628887271881103,
      "epoch": 3.822449384676459,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036039785078955793,
      "loss": 4.8915,
      "mean_token_accuracy": 0.1979900360107422,
      "num_tokens": 84378881.0,
      "step": 38515
    },
    {
      "entropy": 5.712813949584961,
      "epoch": 3.8229456133386264,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003603650694720796,
      "loss": 5.0877,
      "mean_token_accuracy": 0.18695492744445802,
      "num_tokens": 84390382.0,
      "step": 38520
    },
    {
      "entropy": 5.676483392715454,
      "epoch": 3.823441842000794,
      "grad_norm": 1.125,
      "learning_rate": 0.00036033228603773136,
      "loss": 4.9732,
      "mean_token_accuracy": 0.1963147297501564,
      "num_tokens": 84401032.0,
      "step": 38525
    },
    {
      "entropy": 5.7419041156768795,
      "epoch": 3.8239380706629618,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036029950048732635,
      "loss": 5.1315,
      "mean_token_accuracy": 0.18304350972175598,
      "num_tokens": 84411934.0,
      "step": 38530
    },
    {
      "entropy": 5.674359130859375,
      "epoch": 3.8244342993251292,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036026671282167753,
      "loss": 4.974,
      "mean_token_accuracy": 0.18591745793819428,
      "num_tokens": 84423291.0,
      "step": 38535
    },
    {
      "entropy": 5.6544273853302,
      "epoch": 3.8249305279872967,
      "grad_norm": 1.0625,
      "learning_rate": 0.000360233923041598,
      "loss": 5.0295,
      "mean_token_accuracy": 0.18740339428186417,
      "num_tokens": 84436451.0,
      "step": 38540
    },
    {
      "entropy": 5.6756244659423825,
      "epoch": 3.825426756649464,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036020113114790085,
      "loss": 4.9873,
      "mean_token_accuracy": 0.18988667130470277,
      "num_tokens": 84447933.0,
      "step": 38545
    },
    {
      "entropy": 5.668146753311158,
      "epoch": 3.8259229853116317,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003601683371413992,
      "loss": 4.9668,
      "mean_token_accuracy": 0.1851247951388359,
      "num_tokens": 84458135.0,
      "step": 38550
    },
    {
      "entropy": 5.787652969360352,
      "epoch": 3.826419213973799,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00036013554102290636,
      "loss": 5.0313,
      "mean_token_accuracy": 0.1836543634533882,
      "num_tokens": 84468620.0,
      "step": 38555
    },
    {
      "entropy": 5.644686555862426,
      "epoch": 3.8269154426359666,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003601027427932355,
      "loss": 4.9077,
      "mean_token_accuracy": 0.19421246945858,
      "num_tokens": 84479891.0,
      "step": 38560
    },
    {
      "entropy": 5.632384395599365,
      "epoch": 3.827411671298134,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003600699424532,
      "loss": 5.0108,
      "mean_token_accuracy": 0.1938151702284813,
      "num_tokens": 84490621.0,
      "step": 38565
    },
    {
      "entropy": 5.707897138595581,
      "epoch": 3.8279078999603016,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036003714000361326,
      "loss": 4.9913,
      "mean_token_accuracy": 0.19852638691663743,
      "num_tokens": 84500940.0,
      "step": 38570
    },
    {
      "entropy": 5.662576723098755,
      "epoch": 3.828404128622469,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036000433544528866,
      "loss": 5.0053,
      "mean_token_accuracy": 0.19429794251918792,
      "num_tokens": 84511891.0,
      "step": 38575
    },
    {
      "entropy": 5.6784180164337155,
      "epoch": 3.8289003572846365,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003599715287790397,
      "loss": 5.0176,
      "mean_token_accuracy": 0.18841783553361893,
      "num_tokens": 84522546.0,
      "step": 38580
    },
    {
      "entropy": 5.70709753036499,
      "epoch": 3.8293965859468044,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035993872000567994,
      "loss": 5.0077,
      "mean_token_accuracy": 0.18076782524585724,
      "num_tokens": 84532281.0,
      "step": 38585
    },
    {
      "entropy": 5.671993160247803,
      "epoch": 3.829892814608972,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035990590912602304,
      "loss": 4.9615,
      "mean_token_accuracy": 0.1899853378534317,
      "num_tokens": 84542477.0,
      "step": 38590
    },
    {
      "entropy": 5.711135911941528,
      "epoch": 3.8303890432711394,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035987309614088245,
      "loss": 4.9687,
      "mean_token_accuracy": 0.1885972499847412,
      "num_tokens": 84552780.0,
      "step": 38595
    },
    {
      "entropy": 5.731697702407837,
      "epoch": 3.830885271933307,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035984028105107195,
      "loss": 5.0447,
      "mean_token_accuracy": 0.1859237879514694,
      "num_tokens": 84564895.0,
      "step": 38600
    },
    {
      "entropy": 5.633631801605224,
      "epoch": 3.8313815005954743,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035980746385740536,
      "loss": 4.9314,
      "mean_token_accuracy": 0.19298363626003265,
      "num_tokens": 84575680.0,
      "step": 38605
    },
    {
      "entropy": 5.618881940841675,
      "epoch": 3.831877729257642,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035977464456069636,
      "loss": 4.9724,
      "mean_token_accuracy": 0.189489509165287,
      "num_tokens": 84586865.0,
      "step": 38610
    },
    {
      "entropy": 5.629767751693725,
      "epoch": 3.8323739579198093,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035974182316175887,
      "loss": 4.9804,
      "mean_token_accuracy": 0.19351855516433716,
      "num_tokens": 84598285.0,
      "step": 38615
    },
    {
      "entropy": 5.68062834739685,
      "epoch": 3.832870186581977,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003597089996614068,
      "loss": 5.0436,
      "mean_token_accuracy": 0.18063780665397644,
      "num_tokens": 84609481.0,
      "step": 38620
    },
    {
      "entropy": 5.7727094173431395,
      "epoch": 3.8333664152441447,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035967617406045406,
      "loss": 5.1025,
      "mean_token_accuracy": 0.18124179393053055,
      "num_tokens": 84619719.0,
      "step": 38625
    },
    {
      "entropy": 5.702367067337036,
      "epoch": 3.833862643906312,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035964334635971455,
      "loss": 4.971,
      "mean_token_accuracy": 0.18545049875974656,
      "num_tokens": 84630074.0,
      "step": 38630
    },
    {
      "entropy": 5.646145486831665,
      "epoch": 3.8343588725684796,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003596105165600025,
      "loss": 4.9373,
      "mean_token_accuracy": 0.19154839515686034,
      "num_tokens": 84640508.0,
      "step": 38635
    },
    {
      "entropy": 5.664734554290772,
      "epoch": 3.834855101230647,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035957768466213203,
      "loss": 4.9395,
      "mean_token_accuracy": 0.1972239390015602,
      "num_tokens": 84651824.0,
      "step": 38640
    },
    {
      "entropy": 5.633285331726074,
      "epoch": 3.8353513298928146,
      "grad_norm": 1.125,
      "learning_rate": 0.00035954485066691717,
      "loss": 5.0547,
      "mean_token_accuracy": 0.19376676380634308,
      "num_tokens": 84662107.0,
      "step": 38645
    },
    {
      "entropy": 5.600778818130493,
      "epoch": 3.835847558554982,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003595120145751723,
      "loss": 4.9193,
      "mean_token_accuracy": 0.196476486325264,
      "num_tokens": 84674192.0,
      "step": 38650
    },
    {
      "entropy": 5.649991273880005,
      "epoch": 3.8363437872171495,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035947917638771146,
      "loss": 4.9613,
      "mean_token_accuracy": 0.19531164914369584,
      "num_tokens": 84684294.0,
      "step": 38655
    },
    {
      "entropy": 5.634979629516602,
      "epoch": 3.836840015879317,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035944633610534904,
      "loss": 4.9694,
      "mean_token_accuracy": 0.1923084169626236,
      "num_tokens": 84695934.0,
      "step": 38660
    },
    {
      "entropy": 5.732069826126098,
      "epoch": 3.8373362445414845,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035941349372889956,
      "loss": 5.019,
      "mean_token_accuracy": 0.1855941817164421,
      "num_tokens": 84707549.0,
      "step": 38665
    },
    {
      "entropy": 5.678859996795654,
      "epoch": 3.8378324732036524,
      "grad_norm": 1.375,
      "learning_rate": 0.00035938064925917726,
      "loss": 4.8889,
      "mean_token_accuracy": 0.20225688815116882,
      "num_tokens": 84717864.0,
      "step": 38670
    },
    {
      "entropy": 5.617730045318604,
      "epoch": 3.83832870186582,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003593478026969967,
      "loss": 4.9463,
      "mean_token_accuracy": 0.18320631384849548,
      "num_tokens": 84728328.0,
      "step": 38675
    },
    {
      "entropy": 5.707433462142944,
      "epoch": 3.8388249305279873,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035931495404317243,
      "loss": 5.0828,
      "mean_token_accuracy": 0.1786729097366333,
      "num_tokens": 84739037.0,
      "step": 38680
    },
    {
      "entropy": 5.624778127670288,
      "epoch": 3.839321159190155,
      "grad_norm": 1.21875,
      "learning_rate": 0.000359282103298519,
      "loss": 4.9897,
      "mean_token_accuracy": 0.18898467421531678,
      "num_tokens": 84750356.0,
      "step": 38685
    },
    {
      "entropy": 5.687098407745362,
      "epoch": 3.8398173878523223,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035924925046385085,
      "loss": 4.9258,
      "mean_token_accuracy": 0.1936037942767143,
      "num_tokens": 84759622.0,
      "step": 38690
    },
    {
      "entropy": 5.643433332443237,
      "epoch": 3.8403136165144898,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035921639553998297,
      "loss": 4.9682,
      "mean_token_accuracy": 0.18982625305652617,
      "num_tokens": 84771602.0,
      "step": 38695
    },
    {
      "entropy": 5.641772222518921,
      "epoch": 3.8408098451766572,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035918353852772996,
      "loss": 4.9951,
      "mean_token_accuracy": 0.18580806702375413,
      "num_tokens": 84781730.0,
      "step": 38700
    },
    {
      "entropy": 5.664940071105957,
      "epoch": 3.841306073838825,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003591506794279066,
      "loss": 4.9546,
      "mean_token_accuracy": 0.19204143583774566,
      "num_tokens": 84793846.0,
      "step": 38705
    },
    {
      "entropy": 5.7427534580230715,
      "epoch": 3.8418023025009926,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003591178182413276,
      "loss": 5.1054,
      "mean_token_accuracy": 0.1861721768975258,
      "num_tokens": 84804704.0,
      "step": 38710
    },
    {
      "entropy": 5.662493181228638,
      "epoch": 3.84229853116316,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035908495496880804,
      "loss": 4.9361,
      "mean_token_accuracy": 0.19651362001895906,
      "num_tokens": 84816052.0,
      "step": 38715
    },
    {
      "entropy": 5.683733940124512,
      "epoch": 3.8427947598253276,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003590520896111628,
      "loss": 5.0456,
      "mean_token_accuracy": 0.18957991003990174,
      "num_tokens": 84827412.0,
      "step": 38720
    },
    {
      "entropy": 5.731417798995972,
      "epoch": 3.843290988487495,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035901922216920683,
      "loss": 5.0254,
      "mean_token_accuracy": 0.18698367923498155,
      "num_tokens": 84840003.0,
      "step": 38725
    },
    {
      "entropy": 5.682729196548462,
      "epoch": 3.8437872171496625,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003589863526437552,
      "loss": 4.9358,
      "mean_token_accuracy": 0.19763928800821304,
      "num_tokens": 84850385.0,
      "step": 38730
    },
    {
      "entropy": 5.667112350463867,
      "epoch": 3.84428344581183,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003589534810356231,
      "loss": 4.9325,
      "mean_token_accuracy": 0.1895981177687645,
      "num_tokens": 84860628.0,
      "step": 38735
    },
    {
      "entropy": 5.620638608932495,
      "epoch": 3.8447796744739975,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035892060734562544,
      "loss": 4.91,
      "mean_token_accuracy": 0.19908052384853364,
      "num_tokens": 84870654.0,
      "step": 38740
    },
    {
      "entropy": 5.594986391067505,
      "epoch": 3.845275903136165,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035888773157457756,
      "loss": 4.9459,
      "mean_token_accuracy": 0.19017465710639953,
      "num_tokens": 84882928.0,
      "step": 38745
    },
    {
      "entropy": 5.693980121612549,
      "epoch": 3.8457721317983324,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003588548537232947,
      "loss": 5.0156,
      "mean_token_accuracy": 0.19211437553167343,
      "num_tokens": 84893626.0,
      "step": 38750
    },
    {
      "entropy": 5.661842679977417,
      "epoch": 3.8462683604605004,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035882197379259224,
      "loss": 5.0141,
      "mean_token_accuracy": 0.19077927321195604,
      "num_tokens": 84904442.0,
      "step": 38755
    },
    {
      "entropy": 5.700229167938232,
      "epoch": 3.846764589122668,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003587890917832854,
      "loss": 5.0176,
      "mean_token_accuracy": 0.19367175102233886,
      "num_tokens": 84914444.0,
      "step": 38760
    },
    {
      "entropy": 5.707896423339844,
      "epoch": 3.8472608177848353,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003587562076961897,
      "loss": 5.0377,
      "mean_token_accuracy": 0.18650231659412383,
      "num_tokens": 84925274.0,
      "step": 38765
    },
    {
      "entropy": 5.6382567405700685,
      "epoch": 3.847757046447003,
      "grad_norm": 1.25,
      "learning_rate": 0.00035872332153212046,
      "loss": 4.9344,
      "mean_token_accuracy": 0.19190746545791626,
      "num_tokens": 84936236.0,
      "step": 38770
    },
    {
      "entropy": 5.717468500137329,
      "epoch": 3.8482532751091703,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035869043329189314,
      "loss": 5.0438,
      "mean_token_accuracy": 0.17899549007415771,
      "num_tokens": 84947895.0,
      "step": 38775
    },
    {
      "entropy": 5.639207553863526,
      "epoch": 3.8487495037713377,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003586575429763236,
      "loss": 4.9615,
      "mean_token_accuracy": 0.1931676372885704,
      "num_tokens": 84958584.0,
      "step": 38780
    },
    {
      "entropy": 5.690412855148315,
      "epoch": 3.849245732433505,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003586246505862272,
      "loss": 5.0099,
      "mean_token_accuracy": 0.1981680363416672,
      "num_tokens": 84970135.0,
      "step": 38785
    },
    {
      "entropy": 5.738853883743286,
      "epoch": 3.849741961095673,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035859175612241957,
      "loss": 5.0214,
      "mean_token_accuracy": 0.1897286891937256,
      "num_tokens": 84982086.0,
      "step": 38790
    },
    {
      "entropy": 5.653521728515625,
      "epoch": 3.8502381897578406,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035855885958571666,
      "loss": 4.9857,
      "mean_token_accuracy": 0.19289616495370865,
      "num_tokens": 84994399.0,
      "step": 38795
    },
    {
      "entropy": 5.74421877861023,
      "epoch": 3.850734418420008,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000358525960976934,
      "loss": 5.0803,
      "mean_token_accuracy": 0.18443600386381148,
      "num_tokens": 85004248.0,
      "step": 38800
    },
    {
      "entropy": 5.717569303512573,
      "epoch": 3.8512306470821756,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003584930602968875,
      "loss": 5.0822,
      "mean_token_accuracy": 0.18601056188344955,
      "num_tokens": 85015306.0,
      "step": 38805
    },
    {
      "entropy": 5.648202180862427,
      "epoch": 3.851726875744343,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035846015754639304,
      "loss": 4.9118,
      "mean_token_accuracy": 0.19571878314018248,
      "num_tokens": 85025566.0,
      "step": 38810
    },
    {
      "entropy": 5.665893173217773,
      "epoch": 3.8522231044065105,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003584272527262665,
      "loss": 5.013,
      "mean_token_accuracy": 0.19308474510908127,
      "num_tokens": 85036692.0,
      "step": 38815
    },
    {
      "entropy": 5.69976315498352,
      "epoch": 3.852719333068678,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00035839434583732384,
      "loss": 5.0462,
      "mean_token_accuracy": 0.18130480945110322,
      "num_tokens": 85046712.0,
      "step": 38820
    },
    {
      "entropy": 5.712566995620728,
      "epoch": 3.8532155617308455,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035836143688038107,
      "loss": 5.0262,
      "mean_token_accuracy": 0.189892715215683,
      "num_tokens": 85056122.0,
      "step": 38825
    },
    {
      "entropy": 5.678695487976074,
      "epoch": 3.853711790393013,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035832852585625436,
      "loss": 5.0275,
      "mean_token_accuracy": 0.18527341336011888,
      "num_tokens": 85068201.0,
      "step": 38830
    },
    {
      "entropy": 5.715319585800171,
      "epoch": 3.8542080190551804,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003582956127657598,
      "loss": 5.0114,
      "mean_token_accuracy": 0.19171222448348998,
      "num_tokens": 85079248.0,
      "step": 38835
    },
    {
      "entropy": 5.593151092529297,
      "epoch": 3.8547042477173483,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003582626976097134,
      "loss": 4.8865,
      "mean_token_accuracy": 0.1927616447210312,
      "num_tokens": 85091602.0,
      "step": 38840
    },
    {
      "entropy": 5.633051776885987,
      "epoch": 3.855200476379516,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003582297803889316,
      "loss": 4.9114,
      "mean_token_accuracy": 0.19430736750364302,
      "num_tokens": 85101540.0,
      "step": 38845
    },
    {
      "entropy": 5.634538412094116,
      "epoch": 3.8556967050416833,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003581968611042306,
      "loss": 4.974,
      "mean_token_accuracy": 0.19261692315340043,
      "num_tokens": 85112856.0,
      "step": 38850
    },
    {
      "entropy": 5.653151035308838,
      "epoch": 3.8561929337038507,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035816393975642675,
      "loss": 4.9807,
      "mean_token_accuracy": 0.18712226450443267,
      "num_tokens": 85123269.0,
      "step": 38855
    },
    {
      "entropy": 5.574547863006591,
      "epoch": 3.856689162366018,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035813101634633626,
      "loss": 4.8617,
      "mean_token_accuracy": 0.20316943228244783,
      "num_tokens": 85134175.0,
      "step": 38860
    },
    {
      "entropy": 5.673233985900879,
      "epoch": 3.8571853910281857,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035809809087477586,
      "loss": 4.9976,
      "mean_token_accuracy": 0.18989120721817015,
      "num_tokens": 85146808.0,
      "step": 38865
    },
    {
      "entropy": 5.632181024551391,
      "epoch": 3.857681619690353,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003580651633425617,
      "loss": 4.9465,
      "mean_token_accuracy": 0.18529062718153,
      "num_tokens": 85158651.0,
      "step": 38870
    },
    {
      "entropy": 5.735946416854858,
      "epoch": 3.858177848352521,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003580322337505106,
      "loss": 5.0023,
      "mean_token_accuracy": 0.1891261577606201,
      "num_tokens": 85169152.0,
      "step": 38875
    },
    {
      "entropy": 5.650037002563477,
      "epoch": 3.8586740770146886,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035799930209943906,
      "loss": 5.0037,
      "mean_token_accuracy": 0.17943573743104935,
      "num_tokens": 85180516.0,
      "step": 38880
    },
    {
      "entropy": 5.67222032546997,
      "epoch": 3.859170305676856,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003579663683901636,
      "loss": 4.9799,
      "mean_token_accuracy": 0.18724625408649445,
      "num_tokens": 85192109.0,
      "step": 38885
    },
    {
      "entropy": 5.687209558486939,
      "epoch": 3.8596665343390235,
      "grad_norm": 1.125,
      "learning_rate": 0.00035793343262350094,
      "loss": 4.9958,
      "mean_token_accuracy": 0.1877191796898842,
      "num_tokens": 85203005.0,
      "step": 38890
    },
    {
      "entropy": 5.672806692123413,
      "epoch": 3.860162763001191,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003579004948002679,
      "loss": 4.9868,
      "mean_token_accuracy": 0.1926281049847603,
      "num_tokens": 85214374.0,
      "step": 38895
    },
    {
      "entropy": 5.5803498268127445,
      "epoch": 3.8606589916633585,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035786755492128134,
      "loss": 4.8995,
      "mean_token_accuracy": 0.1987432524561882,
      "num_tokens": 85225056.0,
      "step": 38900
    },
    {
      "entropy": 5.7123040676116945,
      "epoch": 3.861155220325526,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003578346129873578,
      "loss": 5.1039,
      "mean_token_accuracy": 0.179452808201313,
      "num_tokens": 85236154.0,
      "step": 38905
    },
    {
      "entropy": 5.6593975067138675,
      "epoch": 3.8616514489876934,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035780166899931437,
      "loss": 4.9459,
      "mean_token_accuracy": 0.1913157135248184,
      "num_tokens": 85246757.0,
      "step": 38910
    },
    {
      "entropy": 5.639736080169678,
      "epoch": 3.862147677649861,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003577687229579681,
      "loss": 4.925,
      "mean_token_accuracy": 0.19206766188144683,
      "num_tokens": 85258066.0,
      "step": 38915
    },
    {
      "entropy": 5.649885654449463,
      "epoch": 3.8626439063120284,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003577357748641357,
      "loss": 4.9875,
      "mean_token_accuracy": 0.1891244098544121,
      "num_tokens": 85268204.0,
      "step": 38920
    },
    {
      "entropy": 5.599319791793823,
      "epoch": 3.863140134974196,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003577028247186344,
      "loss": 4.911,
      "mean_token_accuracy": 0.19299435168504714,
      "num_tokens": 85278808.0,
      "step": 38925
    },
    {
      "entropy": 5.690190362930298,
      "epoch": 3.8636363636363638,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035766987252228133,
      "loss": 4.9818,
      "mean_token_accuracy": 0.18738336712121964,
      "num_tokens": 85290082.0,
      "step": 38930
    },
    {
      "entropy": 5.618980360031128,
      "epoch": 3.8641325922985312,
      "grad_norm": 1.25,
      "learning_rate": 0.0003576369182758935,
      "loss": 4.966,
      "mean_token_accuracy": 0.18860511034727095,
      "num_tokens": 85300313.0,
      "step": 38935
    },
    {
      "entropy": 5.595178604125977,
      "epoch": 3.8646288209606987,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003576039619802882,
      "loss": 4.862,
      "mean_token_accuracy": 0.20361003577709197,
      "num_tokens": 85310967.0,
      "step": 38940
    },
    {
      "entropy": 5.6729778289794925,
      "epoch": 3.865125049622866,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003575710036362826,
      "loss": 4.9588,
      "mean_token_accuracy": 0.18763495832681656,
      "num_tokens": 85321835.0,
      "step": 38945
    },
    {
      "entropy": 5.70135612487793,
      "epoch": 3.8656212782850337,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003575380432446941,
      "loss": 4.9951,
      "mean_token_accuracy": 0.1882750853896141,
      "num_tokens": 85334193.0,
      "step": 38950
    },
    {
      "entropy": 5.6777853012084964,
      "epoch": 3.866117506947201,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003575050808063399,
      "loss": 5.0037,
      "mean_token_accuracy": 0.1895100250840187,
      "num_tokens": 85344864.0,
      "step": 38955
    },
    {
      "entropy": 5.609857416152954,
      "epoch": 3.866613735609369,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035747211632203757,
      "loss": 4.9339,
      "mean_token_accuracy": 0.19030471742153168,
      "num_tokens": 85356183.0,
      "step": 38960
    },
    {
      "entropy": 5.664473915100098,
      "epoch": 3.8671099642715365,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003574391497926044,
      "loss": 4.8908,
      "mean_token_accuracy": 0.2020973414182663,
      "num_tokens": 85367157.0,
      "step": 38965
    },
    {
      "entropy": 5.7234477519989015,
      "epoch": 3.867606192933704,
      "grad_norm": 1.140625,
      "learning_rate": 0.000357406181218858,
      "loss": 5.0414,
      "mean_token_accuracy": 0.19273492246866225,
      "num_tokens": 85378995.0,
      "step": 38970
    },
    {
      "entropy": 5.653461027145386,
      "epoch": 3.8681024215958715,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003573732106016158,
      "loss": 4.9741,
      "mean_token_accuracy": 0.19868308305740356,
      "num_tokens": 85389949.0,
      "step": 38975
    },
    {
      "entropy": 5.621689796447754,
      "epoch": 3.868598650258039,
      "grad_norm": 1.25,
      "learning_rate": 0.0003573402379416955,
      "loss": 5.0066,
      "mean_token_accuracy": 0.18539186269044877,
      "num_tokens": 85401261.0,
      "step": 38980
    },
    {
      "entropy": 5.708710336685181,
      "epoch": 3.8690948789202064,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035730726323991474,
      "loss": 5.0506,
      "mean_token_accuracy": 0.1983409881591797,
      "num_tokens": 85413446.0,
      "step": 38985
    },
    {
      "entropy": 5.6485858917236325,
      "epoch": 3.869591107582374,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003572742864970913,
      "loss": 4.9764,
      "mean_token_accuracy": 0.1847265914082527,
      "num_tokens": 85423819.0,
      "step": 38990
    },
    {
      "entropy": 5.685491561889648,
      "epoch": 3.8700873362445414,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003572413077140427,
      "loss": 4.9587,
      "mean_token_accuracy": 0.19277655333280563,
      "num_tokens": 85435129.0,
      "step": 38995
    },
    {
      "entropy": 5.672494506835937,
      "epoch": 3.870583564906709,
      "grad_norm": 1.203125,
      "learning_rate": 0.000357208326891587,
      "loss": 4.9524,
      "mean_token_accuracy": 0.19655549228191377,
      "num_tokens": 85445845.0,
      "step": 39000
    },
    {
      "epoch": 3.870583564906709,
      "eval_entropy": 5.422245847439647,
      "eval_loss": 5.107332706451416,
      "eval_mean_token_accuracy": 0.19013103360191155,
      "eval_num_tokens": 85445845.0,
      "eval_runtime": 20.3728,
      "eval_samples_per_second": 1652.646,
      "eval_steps_per_second": 206.599,
      "step": 39000
    },
    {
      "entropy": 5.640390205383301,
      "epoch": 3.8710797935688763,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035717534403054184,
      "loss": 4.9471,
      "mean_token_accuracy": 0.18848409354686738,
      "num_tokens": 85456966.0,
      "step": 39005
    },
    {
      "entropy": 5.701126003265381,
      "epoch": 3.871576022231044,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003571423591317252,
      "loss": 5.0126,
      "mean_token_accuracy": 0.18494829684495925,
      "num_tokens": 85467135.0,
      "step": 39010
    },
    {
      "entropy": 5.668283128738404,
      "epoch": 3.8720722508932117,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035710937219595514,
      "loss": 4.9074,
      "mean_token_accuracy": 0.19210264682769776,
      "num_tokens": 85478256.0,
      "step": 39015
    },
    {
      "entropy": 5.625346374511719,
      "epoch": 3.872568479555379,
      "grad_norm": 1.78125,
      "learning_rate": 0.0003570763832240495,
      "loss": 4.9407,
      "mean_token_accuracy": 0.19250966757535934,
      "num_tokens": 85489071.0,
      "step": 39020
    },
    {
      "entropy": 5.647308301925659,
      "epoch": 3.8730647082175467,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003570433922168265,
      "loss": 4.965,
      "mean_token_accuracy": 0.19209473133087157,
      "num_tokens": 85498585.0,
      "step": 39025
    },
    {
      "entropy": 5.734799003601074,
      "epoch": 3.873560936879714,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003570103991751042,
      "loss": 5.0364,
      "mean_token_accuracy": 0.18699149638414383,
      "num_tokens": 85508759.0,
      "step": 39030
    },
    {
      "entropy": 5.683909940719604,
      "epoch": 3.8740571655418816,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003569774040997006,
      "loss": 5.0627,
      "mean_token_accuracy": 0.18235128074884416,
      "num_tokens": 85521259.0,
      "step": 39035
    },
    {
      "entropy": 5.666550779342652,
      "epoch": 3.874553394204049,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035694440699143403,
      "loss": 4.9481,
      "mean_token_accuracy": 0.18927311301231384,
      "num_tokens": 85532223.0,
      "step": 39040
    },
    {
      "entropy": 5.710571241378784,
      "epoch": 3.875049622866217,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003569114078511228,
      "loss": 4.9856,
      "mean_token_accuracy": 0.1890406534075737,
      "num_tokens": 85543013.0,
      "step": 39045
    },
    {
      "entropy": 5.677228736877441,
      "epoch": 3.8755458515283845,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035687840667958517,
      "loss": 4.9457,
      "mean_token_accuracy": 0.19020041078329086,
      "num_tokens": 85551990.0,
      "step": 39050
    },
    {
      "entropy": 5.64106125831604,
      "epoch": 3.876042080190552,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035684540347763943,
      "loss": 4.9852,
      "mean_token_accuracy": 0.18919304609298707,
      "num_tokens": 85563201.0,
      "step": 39055
    },
    {
      "entropy": 5.577176713943482,
      "epoch": 3.8765383088527194,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003568123982461041,
      "loss": 4.9082,
      "mean_token_accuracy": 0.2023670643568039,
      "num_tokens": 85574122.0,
      "step": 39060
    },
    {
      "entropy": 5.642541408538818,
      "epoch": 3.877034537514887,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003567793909857975,
      "loss": 5.0066,
      "mean_token_accuracy": 0.18499585539102553,
      "num_tokens": 85586272.0,
      "step": 39065
    },
    {
      "entropy": 5.657641983032226,
      "epoch": 3.8775307661770544,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003567463816975383,
      "loss": 4.951,
      "mean_token_accuracy": 0.19372159540653228,
      "num_tokens": 85595571.0,
      "step": 39070
    },
    {
      "entropy": 5.727561521530151,
      "epoch": 3.878026994839222,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003567133703821449,
      "loss": 5.0627,
      "mean_token_accuracy": 0.18599239885807037,
      "num_tokens": 85606352.0,
      "step": 39075
    },
    {
      "entropy": 5.702455997467041,
      "epoch": 3.8785232235013893,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035668035704043607,
      "loss": 4.9741,
      "mean_token_accuracy": 0.19140194356441498,
      "num_tokens": 85617118.0,
      "step": 39080
    },
    {
      "entropy": 5.627232885360717,
      "epoch": 3.879019452163557,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003566473416732303,
      "loss": 4.9572,
      "mean_token_accuracy": 0.19377181231975554,
      "num_tokens": 85628681.0,
      "step": 39085
    },
    {
      "entropy": 5.666601085662842,
      "epoch": 3.8795156808257243,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035661432428134656,
      "loss": 4.9722,
      "mean_token_accuracy": 0.188024839758873,
      "num_tokens": 85638381.0,
      "step": 39090
    },
    {
      "entropy": 5.614299201965332,
      "epoch": 3.8800119094878918,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003565813048656033,
      "loss": 4.9298,
      "mean_token_accuracy": 0.18941288441419601,
      "num_tokens": 85650044.0,
      "step": 39095
    },
    {
      "entropy": 5.653824996948242,
      "epoch": 3.8805081381500597,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003565482834268195,
      "loss": 4.9819,
      "mean_token_accuracy": 0.19042586982250215,
      "num_tokens": 85661754.0,
      "step": 39100
    },
    {
      "entropy": 5.653291273117065,
      "epoch": 3.881004366812227,
      "grad_norm": 1.15625,
      "learning_rate": 0.000356515259965814,
      "loss": 4.9877,
      "mean_token_accuracy": 0.1921796604990959,
      "num_tokens": 85672009.0,
      "step": 39105
    },
    {
      "entropy": 5.644158458709716,
      "epoch": 3.8815005954743946,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003564822344834057,
      "loss": 4.942,
      "mean_token_accuracy": 0.1923763021826744,
      "num_tokens": 85681680.0,
      "step": 39110
    },
    {
      "entropy": 5.65825572013855,
      "epoch": 3.881996824136562,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003564492069804136,
      "loss": 4.9798,
      "mean_token_accuracy": 0.19368819892406464,
      "num_tokens": 85692457.0,
      "step": 39115
    },
    {
      "entropy": 5.732505798339844,
      "epoch": 3.8824930527987296,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035641617745765667,
      "loss": 5.0398,
      "mean_token_accuracy": 0.18952237963676452,
      "num_tokens": 85703367.0,
      "step": 39120
    },
    {
      "entropy": 5.7686529636383055,
      "epoch": 3.882989281460897,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003563831459159539,
      "loss": 5.0199,
      "mean_token_accuracy": 0.19266377091407777,
      "num_tokens": 85714747.0,
      "step": 39125
    },
    {
      "entropy": 5.735593461990357,
      "epoch": 3.883485510123065,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035635011235612453,
      "loss": 5.0149,
      "mean_token_accuracy": 0.19092757999897003,
      "num_tokens": 85725175.0,
      "step": 39130
    },
    {
      "entropy": 5.696420049667358,
      "epoch": 3.8839817387852325,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035631707677898767,
      "loss": 4.998,
      "mean_token_accuracy": 0.19620485752820968,
      "num_tokens": 85737792.0,
      "step": 39135
    },
    {
      "entropy": 5.682000064849854,
      "epoch": 3.8844779674474,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003562840391853625,
      "loss": 5.0151,
      "mean_token_accuracy": 0.18726796954870223,
      "num_tokens": 85749673.0,
      "step": 39140
    },
    {
      "entropy": 5.675965261459351,
      "epoch": 3.8849741961095674,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003562509995760683,
      "loss": 5.0086,
      "mean_token_accuracy": 0.19031676948070525,
      "num_tokens": 85760283.0,
      "step": 39145
    },
    {
      "entropy": 5.674776554107666,
      "epoch": 3.885470424771735,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035621795795192447,
      "loss": 4.9022,
      "mean_token_accuracy": 0.20107988864183426,
      "num_tokens": 85770267.0,
      "step": 39150
    },
    {
      "entropy": 5.587831497192383,
      "epoch": 3.8859666534339024,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003561849143137502,
      "loss": 4.9144,
      "mean_token_accuracy": 0.19242354333400727,
      "num_tokens": 85780688.0,
      "step": 39155
    },
    {
      "entropy": 5.7351888656616214,
      "epoch": 3.88646288209607,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003561518686623651,
      "loss": 5.0794,
      "mean_token_accuracy": 0.18260440975427628,
      "num_tokens": 85793901.0,
      "step": 39160
    },
    {
      "entropy": 5.723265552520752,
      "epoch": 3.8869591107582373,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003561188209985884,
      "loss": 4.9751,
      "mean_token_accuracy": 0.1889612555503845,
      "num_tokens": 85804812.0,
      "step": 39165
    },
    {
      "entropy": 5.628863906860351,
      "epoch": 3.8874553394204048,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035608577132323983,
      "loss": 4.9108,
      "mean_token_accuracy": 0.19267212152481078,
      "num_tokens": 85814479.0,
      "step": 39170
    },
    {
      "entropy": 5.6406361103057865,
      "epoch": 3.8879515680825723,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003560527196371388,
      "loss": 5.0133,
      "mean_token_accuracy": 0.19456487149000168,
      "num_tokens": 85825649.0,
      "step": 39175
    },
    {
      "entropy": 5.615159606933593,
      "epoch": 3.8884477967447397,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000356019665941105,
      "loss": 4.9272,
      "mean_token_accuracy": 0.1889337942004204,
      "num_tokens": 85835717.0,
      "step": 39180
    },
    {
      "entropy": 5.630715942382812,
      "epoch": 3.8889440254069076,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003559866102359581,
      "loss": 4.9239,
      "mean_token_accuracy": 0.19231944680213928,
      "num_tokens": 85846692.0,
      "step": 39185
    },
    {
      "entropy": 5.646965456008911,
      "epoch": 3.889440254069075,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003559535525225177,
      "loss": 4.9688,
      "mean_token_accuracy": 0.1898011103272438,
      "num_tokens": 85857487.0,
      "step": 39190
    },
    {
      "entropy": 5.651376104354858,
      "epoch": 3.8899364827312426,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003559204928016037,
      "loss": 4.9243,
      "mean_token_accuracy": 0.19325806945562363,
      "num_tokens": 85867289.0,
      "step": 39195
    },
    {
      "entropy": 5.688138580322265,
      "epoch": 3.89043271139341,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035588743107403583,
      "loss": 4.9393,
      "mean_token_accuracy": 0.20082175880670547,
      "num_tokens": 85877190.0,
      "step": 39200
    },
    {
      "entropy": 5.687372922897339,
      "epoch": 3.8909289400555775,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035585436734063394,
      "loss": 4.9426,
      "mean_token_accuracy": 0.1958848625421524,
      "num_tokens": 85887018.0,
      "step": 39205
    },
    {
      "entropy": 5.650249767303467,
      "epoch": 3.891425168717745,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000355821301602218,
      "loss": 4.9779,
      "mean_token_accuracy": 0.18713044375181198,
      "num_tokens": 85898153.0,
      "step": 39210
    },
    {
      "entropy": 5.70771541595459,
      "epoch": 3.891921397379913,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035578823385960784,
      "loss": 5.0114,
      "mean_token_accuracy": 0.18612230271101,
      "num_tokens": 85907903.0,
      "step": 39215
    },
    {
      "entropy": 5.714502954483033,
      "epoch": 3.8924176260420804,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003557551641136237,
      "loss": 5.0515,
      "mean_token_accuracy": 0.19012722074985505,
      "num_tokens": 85919456.0,
      "step": 39220
    },
    {
      "entropy": 5.684949445724487,
      "epoch": 3.892913854704248,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035572209236508546,
      "loss": 4.9956,
      "mean_token_accuracy": 0.19874370396137236,
      "num_tokens": 85930215.0,
      "step": 39225
    },
    {
      "entropy": 5.641653823852539,
      "epoch": 3.8934100833664154,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00035568901861481327,
      "loss": 4.8311,
      "mean_token_accuracy": 0.20676016211509704,
      "num_tokens": 85941595.0,
      "step": 39230
    },
    {
      "entropy": 5.669657468795776,
      "epoch": 3.893906312028583,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00035565594286362726,
      "loss": 4.9523,
      "mean_token_accuracy": 0.1869064226746559,
      "num_tokens": 85950792.0,
      "step": 39235
    },
    {
      "entropy": 5.64429497718811,
      "epoch": 3.8944025406907503,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003556228651123477,
      "loss": 4.9861,
      "mean_token_accuracy": 0.1921436905860901,
      "num_tokens": 85962365.0,
      "step": 39240
    },
    {
      "entropy": 5.661208915710449,
      "epoch": 3.894898769352918,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035558978536179476,
      "loss": 5.049,
      "mean_token_accuracy": 0.18506561517715453,
      "num_tokens": 85974728.0,
      "step": 39245
    },
    {
      "entropy": 5.710791730880738,
      "epoch": 3.8953949980150853,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035555670361278884,
      "loss": 4.9871,
      "mean_token_accuracy": 0.18802233934402465,
      "num_tokens": 85986084.0,
      "step": 39250
    },
    {
      "entropy": 5.689084529876709,
      "epoch": 3.8958912266772527,
      "grad_norm": 1.125,
      "learning_rate": 0.0003555236198661503,
      "loss": 4.942,
      "mean_token_accuracy": 0.20051328837871552,
      "num_tokens": 85997212.0,
      "step": 39255
    },
    {
      "entropy": 5.710441017150879,
      "epoch": 3.89638745533942,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003554905341226994,
      "loss": 4.9815,
      "mean_token_accuracy": 0.1871919482946396,
      "num_tokens": 86006339.0,
      "step": 39260
    },
    {
      "entropy": 5.6797178268432615,
      "epoch": 3.8968836840015877,
      "grad_norm": 1.125,
      "learning_rate": 0.00035545744638325674,
      "loss": 4.9622,
      "mean_token_accuracy": 0.18499597162008286,
      "num_tokens": 86018721.0,
      "step": 39265
    },
    {
      "entropy": 5.687720012664795,
      "epoch": 3.8973799126637556,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035542435664864277,
      "loss": 4.911,
      "mean_token_accuracy": 0.19163923263549804,
      "num_tokens": 86029516.0,
      "step": 39270
    },
    {
      "entropy": 5.697329998016357,
      "epoch": 3.897876141325923,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003553912649196781,
      "loss": 4.9269,
      "mean_token_accuracy": 0.1995895802974701,
      "num_tokens": 86039921.0,
      "step": 39275
    },
    {
      "entropy": 5.650958395004272,
      "epoch": 3.8983723699880906,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035535817119718323,
      "loss": 5.0014,
      "mean_token_accuracy": 0.18301131576299667,
      "num_tokens": 86050353.0,
      "step": 39280
    },
    {
      "entropy": 5.695401954650879,
      "epoch": 3.898868598650258,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003553250754819789,
      "loss": 5.0817,
      "mean_token_accuracy": 0.18535176217556,
      "num_tokens": 86061729.0,
      "step": 39285
    },
    {
      "entropy": 5.670260858535767,
      "epoch": 3.8993648273124255,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003552919777748858,
      "loss": 4.9985,
      "mean_token_accuracy": 0.1961885631084442,
      "num_tokens": 86073717.0,
      "step": 39290
    },
    {
      "entropy": 5.669940996170044,
      "epoch": 3.899861055974593,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003552588780767246,
      "loss": 4.9242,
      "mean_token_accuracy": 0.18981776237487794,
      "num_tokens": 86083601.0,
      "step": 39295
    },
    {
      "entropy": 5.650946712493896,
      "epoch": 3.9003572846367605,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035522577638831613,
      "loss": 4.9942,
      "mean_token_accuracy": 0.19095318019390106,
      "num_tokens": 86095752.0,
      "step": 39300
    },
    {
      "entropy": 5.679824495315552,
      "epoch": 3.9008535132989284,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003551926727104814,
      "loss": 5.0386,
      "mean_token_accuracy": 0.18489453941583633,
      "num_tokens": 86106205.0,
      "step": 39305
    },
    {
      "entropy": 5.719936752319336,
      "epoch": 3.901349741961096,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003551595670440412,
      "loss": 4.9692,
      "mean_token_accuracy": 0.19036684334278106,
      "num_tokens": 86116410.0,
      "step": 39310
    },
    {
      "entropy": 5.740840482711792,
      "epoch": 3.9018459706232633,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003551264593898164,
      "loss": 5.0633,
      "mean_token_accuracy": 0.1867325097322464,
      "num_tokens": 86128254.0,
      "step": 39315
    },
    {
      "entropy": 5.623720026016235,
      "epoch": 3.902342199285431,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035509334974862807,
      "loss": 4.9351,
      "mean_token_accuracy": 0.1900808572769165,
      "num_tokens": 86139274.0,
      "step": 39320
    },
    {
      "entropy": 5.663659191131591,
      "epoch": 3.9028384279475983,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035506023812129724,
      "loss": 4.9457,
      "mean_token_accuracy": 0.1881391167640686,
      "num_tokens": 86148857.0,
      "step": 39325
    },
    {
      "entropy": 5.610370254516601,
      "epoch": 3.9033346566097658,
      "grad_norm": 1.25,
      "learning_rate": 0.00035502712450864504,
      "loss": 4.9537,
      "mean_token_accuracy": 0.1931116670370102,
      "num_tokens": 86160542.0,
      "step": 39330
    },
    {
      "entropy": 5.630419492721558,
      "epoch": 3.9038308852719332,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035499400891149255,
      "loss": 4.95,
      "mean_token_accuracy": 0.19469355344772338,
      "num_tokens": 86171922.0,
      "step": 39335
    },
    {
      "entropy": 5.665802669525147,
      "epoch": 3.9043271139341007,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035496089133066106,
      "loss": 4.9929,
      "mean_token_accuracy": 0.18585794568061828,
      "num_tokens": 86182996.0,
      "step": 39340
    },
    {
      "entropy": 5.72991681098938,
      "epoch": 3.904823342596268,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003549277717669718,
      "loss": 4.9635,
      "mean_token_accuracy": 0.1911040261387825,
      "num_tokens": 86194696.0,
      "step": 39345
    },
    {
      "entropy": 5.727438020706177,
      "epoch": 3.9053195712584357,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003548946502212459,
      "loss": 5.044,
      "mean_token_accuracy": 0.18620511293411254,
      "num_tokens": 86206601.0,
      "step": 39350
    },
    {
      "entropy": 5.672657299041748,
      "epoch": 3.9058157999206036,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035486152669430497,
      "loss": 5.0138,
      "mean_token_accuracy": 0.1873417392373085,
      "num_tokens": 86217181.0,
      "step": 39355
    },
    {
      "entropy": 5.637640285491943,
      "epoch": 3.906312028582771,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035482840118697016,
      "loss": 4.944,
      "mean_token_accuracy": 0.19190932363271712,
      "num_tokens": 86227012.0,
      "step": 39360
    },
    {
      "entropy": 5.564298439025879,
      "epoch": 3.9068082572449385,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003547952737000631,
      "loss": 4.921,
      "mean_token_accuracy": 0.19199208170175552,
      "num_tokens": 86238505.0,
      "step": 39365
    },
    {
      "entropy": 5.666429424285889,
      "epoch": 3.907304485907106,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003547621442344052,
      "loss": 4.9525,
      "mean_token_accuracy": 0.18962276875972747,
      "num_tokens": 86248681.0,
      "step": 39370
    },
    {
      "entropy": 5.596335744857788,
      "epoch": 3.9078007145692735,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035472901279081786,
      "loss": 4.9142,
      "mean_token_accuracy": 0.19292569756507874,
      "num_tokens": 86259858.0,
      "step": 39375
    },
    {
      "entropy": 5.637512826919556,
      "epoch": 3.908296943231441,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003546958793701229,
      "loss": 4.9388,
      "mean_token_accuracy": 0.19063568562269212,
      "num_tokens": 86271680.0,
      "step": 39380
    },
    {
      "entropy": 5.627861595153808,
      "epoch": 3.9087931718936084,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035466274397314194,
      "loss": 4.9955,
      "mean_token_accuracy": 0.1869961678981781,
      "num_tokens": 86282359.0,
      "step": 39385
    },
    {
      "entropy": 5.632995748519898,
      "epoch": 3.9092894005557763,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003546296066006965,
      "loss": 4.9288,
      "mean_token_accuracy": 0.19466810822486877,
      "num_tokens": 86292939.0,
      "step": 39390
    },
    {
      "entropy": 5.709110736846924,
      "epoch": 3.909785629217944,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003545964672536084,
      "loss": 4.9635,
      "mean_token_accuracy": 0.19521084427833557,
      "num_tokens": 86303895.0,
      "step": 39395
    },
    {
      "entropy": 5.624726486206055,
      "epoch": 3.9102818578801113,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003545633259326994,
      "loss": 4.9292,
      "mean_token_accuracy": 0.1933791384100914,
      "num_tokens": 86313966.0,
      "step": 39400
    },
    {
      "entropy": 5.692647695541382,
      "epoch": 3.9107780865422788,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035453018263879144,
      "loss": 5.0554,
      "mean_token_accuracy": 0.194133523106575,
      "num_tokens": 86325871.0,
      "step": 39405
    },
    {
      "entropy": 5.689583349227905,
      "epoch": 3.9112743152044462,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003544970373727063,
      "loss": 5.0028,
      "mean_token_accuracy": 0.19578723162412642,
      "num_tokens": 86335719.0,
      "step": 39410
    },
    {
      "entropy": 5.675618314743042,
      "epoch": 3.9117705438666137,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003544638901352659,
      "loss": 5.0391,
      "mean_token_accuracy": 0.18658097386360167,
      "num_tokens": 86347753.0,
      "step": 39415
    },
    {
      "entropy": 5.643827104568482,
      "epoch": 3.912266772528781,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035443074092729226,
      "loss": 4.9539,
      "mean_token_accuracy": 0.1898644521832466,
      "num_tokens": 86358313.0,
      "step": 39420
    },
    {
      "entropy": 5.6215557098388675,
      "epoch": 3.9127630011909487,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003543975897496074,
      "loss": 4.9791,
      "mean_token_accuracy": 0.18929993808269502,
      "num_tokens": 86367942.0,
      "step": 39425
    },
    {
      "entropy": 5.661205053329468,
      "epoch": 3.913259229853116,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035436443660303343,
      "loss": 4.9583,
      "mean_token_accuracy": 0.195920729637146,
      "num_tokens": 86379023.0,
      "step": 39430
    },
    {
      "entropy": 5.680375289916992,
      "epoch": 3.9137554585152836,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035433128148839243,
      "loss": 4.9474,
      "mean_token_accuracy": 0.18831120431423187,
      "num_tokens": 86389486.0,
      "step": 39435
    },
    {
      "entropy": 5.646461868286133,
      "epoch": 3.9142516871774515,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003542981244065066,
      "loss": 5.0242,
      "mean_token_accuracy": 0.18649153858423234,
      "num_tokens": 86400910.0,
      "step": 39440
    },
    {
      "entropy": 5.665470170974731,
      "epoch": 3.914747915839619,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035426496535819813,
      "loss": 5.0009,
      "mean_token_accuracy": 0.1885412335395813,
      "num_tokens": 86412598.0,
      "step": 39445
    },
    {
      "entropy": 5.635117292404175,
      "epoch": 3.9152441445017865,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035423180434428935,
      "loss": 4.9787,
      "mean_token_accuracy": 0.19479368180036544,
      "num_tokens": 86424530.0,
      "step": 39450
    },
    {
      "entropy": 5.677382564544677,
      "epoch": 3.915740373163954,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003541986413656025,
      "loss": 5.0207,
      "mean_token_accuracy": 0.19039668440818786,
      "num_tokens": 86436207.0,
      "step": 39455
    },
    {
      "entropy": 5.7104345798492435,
      "epoch": 3.9162366018261214,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003541654764229601,
      "loss": 5.0372,
      "mean_token_accuracy": 0.1912678897380829,
      "num_tokens": 86446516.0,
      "step": 39460
    },
    {
      "entropy": 5.689990282058716,
      "epoch": 3.916732830488289,
      "grad_norm": 1.125,
      "learning_rate": 0.0003541323095171844,
      "loss": 4.9709,
      "mean_token_accuracy": 0.1859905257821083,
      "num_tokens": 86458058.0,
      "step": 39465
    },
    {
      "entropy": 5.677693653106689,
      "epoch": 3.9172290591504564,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035409914064909797,
      "loss": 5.0107,
      "mean_token_accuracy": 0.18633761256933212,
      "num_tokens": 86468620.0,
      "step": 39470
    },
    {
      "entropy": 5.651381587982177,
      "epoch": 3.9177252878126243,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003540659698195233,
      "loss": 5.0023,
      "mean_token_accuracy": 0.1896158814430237,
      "num_tokens": 86480823.0,
      "step": 39475
    },
    {
      "entropy": 5.666865682601928,
      "epoch": 3.9182215164747918,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035403279702928285,
      "loss": 4.9917,
      "mean_token_accuracy": 0.1941171780228615,
      "num_tokens": 86491592.0,
      "step": 39480
    },
    {
      "entropy": 5.705988979339599,
      "epoch": 3.9187177451369593,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035399962227919944,
      "loss": 4.9969,
      "mean_token_accuracy": 0.18755383044481277,
      "num_tokens": 86501861.0,
      "step": 39485
    },
    {
      "entropy": 5.660989284515381,
      "epoch": 3.9192139737991267,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003539664455700957,
      "loss": 4.9879,
      "mean_token_accuracy": 0.19104007631540298,
      "num_tokens": 86512491.0,
      "step": 39490
    },
    {
      "entropy": 5.6476935863494875,
      "epoch": 3.919710202461294,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003539332669027941,
      "loss": 4.9777,
      "mean_token_accuracy": 0.1881310299038887,
      "num_tokens": 86523762.0,
      "step": 39495
    },
    {
      "entropy": 5.720729064941406,
      "epoch": 3.9202064311234617,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035390008627811766,
      "loss": 5.0534,
      "mean_token_accuracy": 0.1842200607061386,
      "num_tokens": 86534791.0,
      "step": 39500
    },
    {
      "entropy": 5.635831451416015,
      "epoch": 3.920702659785629,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035386690369688907,
      "loss": 4.9052,
      "mean_token_accuracy": 0.19800885170698165,
      "num_tokens": 86546631.0,
      "step": 39505
    },
    {
      "entropy": 5.633174896240234,
      "epoch": 3.9211988884477966,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035383371915993116,
      "loss": 4.8892,
      "mean_token_accuracy": 0.19786045849323272,
      "num_tokens": 86557188.0,
      "step": 39510
    },
    {
      "entropy": 5.698771858215332,
      "epoch": 3.921695117109964,
      "grad_norm": 1.125,
      "learning_rate": 0.00035380053266806697,
      "loss": 5.0312,
      "mean_token_accuracy": 0.18374637961387635,
      "num_tokens": 86566889.0,
      "step": 39515
    },
    {
      "entropy": 5.718364191055298,
      "epoch": 3.9221913457721316,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035376734422211934,
      "loss": 5.0322,
      "mean_token_accuracy": 0.18295149654150009,
      "num_tokens": 86577942.0,
      "step": 39520
    },
    {
      "entropy": 5.682042169570923,
      "epoch": 3.922687574434299,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035373415382291136,
      "loss": 4.9189,
      "mean_token_accuracy": 0.19906968176364898,
      "num_tokens": 86589094.0,
      "step": 39525
    },
    {
      "entropy": 5.688246631622315,
      "epoch": 3.923183803096467,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003537009614712659,
      "loss": 5.0269,
      "mean_token_accuracy": 0.18278248012065887,
      "num_tokens": 86599987.0,
      "step": 39530
    },
    {
      "entropy": 5.620616006851196,
      "epoch": 3.9236800317586344,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035366776716800626,
      "loss": 4.9403,
      "mean_token_accuracy": 0.18412559926509858,
      "num_tokens": 86610617.0,
      "step": 39535
    },
    {
      "entropy": 5.652143049240112,
      "epoch": 3.924176260420802,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035363457091395546,
      "loss": 4.9646,
      "mean_token_accuracy": 0.1920250803232193,
      "num_tokens": 86620792.0,
      "step": 39540
    },
    {
      "entropy": 5.63747124671936,
      "epoch": 3.9246724890829694,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035360137270993674,
      "loss": 4.9697,
      "mean_token_accuracy": 0.188600791990757,
      "num_tokens": 86631391.0,
      "step": 39545
    },
    {
      "entropy": 5.734295797348023,
      "epoch": 3.925168717745137,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003535681725567734,
      "loss": 5.0469,
      "mean_token_accuracy": 0.19053923934698105,
      "num_tokens": 86641708.0,
      "step": 39550
    },
    {
      "entropy": 5.644957876205444,
      "epoch": 3.9256649464073043,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003535349704552887,
      "loss": 4.8951,
      "mean_token_accuracy": 0.20131759345531464,
      "num_tokens": 86653079.0,
      "step": 39555
    },
    {
      "entropy": 5.6890912532806395,
      "epoch": 3.9261611750694723,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003535017664063059,
      "loss": 5.019,
      "mean_token_accuracy": 0.18283439427614212,
      "num_tokens": 86662031.0,
      "step": 39560
    },
    {
      "entropy": 5.682589149475097,
      "epoch": 3.9266574037316397,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035346856041064845,
      "loss": 5.0271,
      "mean_token_accuracy": 0.1881980448961258,
      "num_tokens": 86672981.0,
      "step": 39565
    },
    {
      "entropy": 5.697335433959961,
      "epoch": 3.927153632393807,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035343535246913985,
      "loss": 4.9488,
      "mean_token_accuracy": 0.19337137937545776,
      "num_tokens": 86684715.0,
      "step": 39570
    },
    {
      "entropy": 5.67241678237915,
      "epoch": 3.9276498610559747,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035340214258260344,
      "loss": 5.0208,
      "mean_token_accuracy": 0.19119084775447845,
      "num_tokens": 86696024.0,
      "step": 39575
    },
    {
      "entropy": 5.702413034439087,
      "epoch": 3.928146089718142,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003533689307518629,
      "loss": 4.9683,
      "mean_token_accuracy": 0.19198645204305648,
      "num_tokens": 86705743.0,
      "step": 39580
    },
    {
      "entropy": 5.71574764251709,
      "epoch": 3.9286423183803096,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035333571697774175,
      "loss": 5.069,
      "mean_token_accuracy": 0.18362292796373367,
      "num_tokens": 86717107.0,
      "step": 39585
    },
    {
      "entropy": 5.6781991481781,
      "epoch": 3.929138547042477,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003533025012610636,
      "loss": 4.9439,
      "mean_token_accuracy": 0.18822902739048003,
      "num_tokens": 86726931.0,
      "step": 39590
    },
    {
      "entropy": 5.677310228347778,
      "epoch": 3.9296347757046446,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035326928360265207,
      "loss": 5.0311,
      "mean_token_accuracy": 0.18738446533679962,
      "num_tokens": 86740507.0,
      "step": 39595
    },
    {
      "entropy": 5.679202747344971,
      "epoch": 3.930131004366812,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035323606400333105,
      "loss": 5.0152,
      "mean_token_accuracy": 0.19012442082166672,
      "num_tokens": 86751542.0,
      "step": 39600
    },
    {
      "entropy": 5.708238220214843,
      "epoch": 3.9306272330289795,
      "grad_norm": 1.25,
      "learning_rate": 0.00035320284246392425,
      "loss": 4.9955,
      "mean_token_accuracy": 0.19367509931325913,
      "num_tokens": 86762191.0,
      "step": 39605
    },
    {
      "entropy": 5.667545890808105,
      "epoch": 3.931123461691147,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003531696189852553,
      "loss": 4.9439,
      "mean_token_accuracy": 0.18751880824565886,
      "num_tokens": 86773684.0,
      "step": 39610
    },
    {
      "entropy": 5.654065132141113,
      "epoch": 3.931619690353315,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003531363935681485,
      "loss": 4.9889,
      "mean_token_accuracy": 0.19206566363573074,
      "num_tokens": 86784237.0,
      "step": 39615
    },
    {
      "entropy": 5.605807876586914,
      "epoch": 3.9321159190154824,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003531031662134274,
      "loss": 4.912,
      "mean_token_accuracy": 0.19789964258670806,
      "num_tokens": 86794670.0,
      "step": 39620
    },
    {
      "entropy": 5.697228622436524,
      "epoch": 3.93261214767765,
      "grad_norm": 1.1875,
      "learning_rate": 0.000353069936921916,
      "loss": 5.0489,
      "mean_token_accuracy": 0.1896022766828537,
      "num_tokens": 86806692.0,
      "step": 39625
    },
    {
      "entropy": 5.6679182052612305,
      "epoch": 3.9331083763398174,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003530367056944384,
      "loss": 5.0073,
      "mean_token_accuracy": 0.18630408495664597,
      "num_tokens": 86817797.0,
      "step": 39630
    },
    {
      "entropy": 5.659595251083374,
      "epoch": 3.933604605001985,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035300347253181875,
      "loss": 4.9481,
      "mean_token_accuracy": 0.19030743688344956,
      "num_tokens": 86830096.0,
      "step": 39635
    },
    {
      "entropy": 5.735978078842163,
      "epoch": 3.9341008336641523,
      "grad_norm": 1.234375,
      "learning_rate": 0.000352970237434881,
      "loss": 5.0322,
      "mean_token_accuracy": 0.18145765513181686,
      "num_tokens": 86840958.0,
      "step": 39640
    },
    {
      "entropy": 5.679369306564331,
      "epoch": 3.9345970623263202,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035293700040444937,
      "loss": 4.9872,
      "mean_token_accuracy": 0.19648196101188659,
      "num_tokens": 86851783.0,
      "step": 39645
    },
    {
      "entropy": 5.678199625015258,
      "epoch": 3.9350932909884877,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035290376144134813,
      "loss": 5.0762,
      "mean_token_accuracy": 0.18191715627908706,
      "num_tokens": 86863615.0,
      "step": 39650
    },
    {
      "entropy": 5.7031824588775635,
      "epoch": 3.935589519650655,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003528705205464014,
      "loss": 4.9525,
      "mean_token_accuracy": 0.1921609029173851,
      "num_tokens": 86874427.0,
      "step": 39655
    },
    {
      "entropy": 5.679954481124878,
      "epoch": 3.9360857483128227,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035283727772043356,
      "loss": 4.9707,
      "mean_token_accuracy": 0.1943221926689148,
      "num_tokens": 86886590.0,
      "step": 39660
    },
    {
      "entropy": 5.675427103042603,
      "epoch": 3.93658197697499,
      "grad_norm": 1.21875,
      "learning_rate": 0.000352804032964269,
      "loss": 4.9763,
      "mean_token_accuracy": 0.192250294983387,
      "num_tokens": 86896531.0,
      "step": 39665
    },
    {
      "entropy": 5.7129234790802,
      "epoch": 3.9370782056371576,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035277078627873204,
      "loss": 5.0789,
      "mean_token_accuracy": 0.1852406904101372,
      "num_tokens": 86907291.0,
      "step": 39670
    },
    {
      "entropy": 5.740811014175415,
      "epoch": 3.937574434299325,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003527375376646471,
      "loss": 5.0435,
      "mean_token_accuracy": 0.18511389791965485,
      "num_tokens": 86917691.0,
      "step": 39675
    },
    {
      "entropy": 5.679636859893799,
      "epoch": 3.9380706629614926,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035270428712283887,
      "loss": 4.9522,
      "mean_token_accuracy": 0.19706647992134094,
      "num_tokens": 86928899.0,
      "step": 39680
    },
    {
      "entropy": 5.6870917797088625,
      "epoch": 3.93856689162366,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035267103465413165,
      "loss": 4.9661,
      "mean_token_accuracy": 0.19044882655143738,
      "num_tokens": 86939579.0,
      "step": 39685
    },
    {
      "entropy": 5.6607890129089355,
      "epoch": 3.9390631202858275,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035263778025935023,
      "loss": 5.0171,
      "mean_token_accuracy": 0.18554819226264954,
      "num_tokens": 86951355.0,
      "step": 39690
    },
    {
      "entropy": 5.646284770965576,
      "epoch": 3.939559348947995,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003526045239393191,
      "loss": 4.9099,
      "mean_token_accuracy": 0.1946093961596489,
      "num_tokens": 86961608.0,
      "step": 39695
    },
    {
      "entropy": 5.688149547576904,
      "epoch": 3.940055577610163,
      "grad_norm": 1.21875,
      "learning_rate": 0.000352571265694863,
      "loss": 4.9886,
      "mean_token_accuracy": 0.1860163539648056,
      "num_tokens": 86972934.0,
      "step": 39700
    },
    {
      "entropy": 5.661797046661377,
      "epoch": 3.9405518062723304,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035253800552680666,
      "loss": 4.9084,
      "mean_token_accuracy": 0.1999810218811035,
      "num_tokens": 86984400.0,
      "step": 39705
    },
    {
      "entropy": 5.666205549240113,
      "epoch": 3.941048034934498,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003525047434359749,
      "loss": 5.0393,
      "mean_token_accuracy": 0.18657363653182985,
      "num_tokens": 86995532.0,
      "step": 39710
    },
    {
      "entropy": 5.7991251945495605,
      "epoch": 3.9415442635966653,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003524714794231924,
      "loss": 5.2171,
      "mean_token_accuracy": 0.17767234295606613,
      "num_tokens": 87008519.0,
      "step": 39715
    },
    {
      "entropy": 5.654892110824585,
      "epoch": 3.942040492258833,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003524382134892843,
      "loss": 4.9815,
      "mean_token_accuracy": 0.18846315890550613,
      "num_tokens": 87018764.0,
      "step": 39720
    },
    {
      "entropy": 5.622712802886963,
      "epoch": 3.9425367209210003,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003524049456350753,
      "loss": 4.8505,
      "mean_token_accuracy": 0.19718524813652039,
      "num_tokens": 87027810.0,
      "step": 39725
    },
    {
      "entropy": 5.633229207992554,
      "epoch": 3.943032949583168,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035237167586139043,
      "loss": 4.9603,
      "mean_token_accuracy": 0.19107392728328704,
      "num_tokens": 87039451.0,
      "step": 39730
    },
    {
      "entropy": 5.619491052627564,
      "epoch": 3.9435291782453357,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003523384041690547,
      "loss": 4.987,
      "mean_token_accuracy": 0.1901546001434326,
      "num_tokens": 87049852.0,
      "step": 39735
    },
    {
      "entropy": 5.643772602081299,
      "epoch": 3.944025406907503,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035230513055889325,
      "loss": 4.9537,
      "mean_token_accuracy": 0.19072720259428025,
      "num_tokens": 87060657.0,
      "step": 39740
    },
    {
      "entropy": 5.6757182598114015,
      "epoch": 3.9445216355696706,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003522718550317311,
      "loss": 5.0159,
      "mean_token_accuracy": 0.19135890007019044,
      "num_tokens": 87071402.0,
      "step": 39745
    },
    {
      "entropy": 5.62472014427185,
      "epoch": 3.945017864231838,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035223857758839345,
      "loss": 4.9547,
      "mean_token_accuracy": 0.19797438085079194,
      "num_tokens": 87082265.0,
      "step": 39750
    },
    {
      "entropy": 5.670398807525634,
      "epoch": 3.9455140928940056,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035220529822970555,
      "loss": 4.974,
      "mean_token_accuracy": 0.18596822768449783,
      "num_tokens": 87092567.0,
      "step": 39755
    },
    {
      "entropy": 5.727115440368652,
      "epoch": 3.946010321556173,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035217201695649257,
      "loss": 4.9965,
      "mean_token_accuracy": 0.18877538442611694,
      "num_tokens": 87102661.0,
      "step": 39760
    },
    {
      "entropy": 5.703086471557617,
      "epoch": 3.9465065502183405,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035213873376957987,
      "loss": 4.9366,
      "mean_token_accuracy": 0.19341776221990586,
      "num_tokens": 87113911.0,
      "step": 39765
    },
    {
      "entropy": 5.695884943008423,
      "epoch": 3.947002778880508,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00035210544866979285,
      "loss": 5.0313,
      "mean_token_accuracy": 0.18049230724573134,
      "num_tokens": 87125939.0,
      "step": 39770
    },
    {
      "entropy": 5.6258745193481445,
      "epoch": 3.9474990075426755,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035207216165795676,
      "loss": 4.9296,
      "mean_token_accuracy": 0.18971387445926666,
      "num_tokens": 87137589.0,
      "step": 39775
    },
    {
      "entropy": 5.600031948089599,
      "epoch": 3.947995236204843,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003520388727348972,
      "loss": 4.915,
      "mean_token_accuracy": 0.1941943049430847,
      "num_tokens": 87147904.0,
      "step": 39780
    },
    {
      "entropy": 5.735685253143311,
      "epoch": 3.948491464867011,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035200558190143954,
      "loss": 5.0279,
      "mean_token_accuracy": 0.18225342631340027,
      "num_tokens": 87158733.0,
      "step": 39785
    },
    {
      "entropy": 5.709411239624023,
      "epoch": 3.9489876935291783,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035197228915840944,
      "loss": 5.0212,
      "mean_token_accuracy": 0.19348036795854567,
      "num_tokens": 87170302.0,
      "step": 39790
    },
    {
      "entropy": 5.68716025352478,
      "epoch": 3.949483922191346,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003519389945066324,
      "loss": 4.9837,
      "mean_token_accuracy": 0.18513433039188384,
      "num_tokens": 87181408.0,
      "step": 39795
    },
    {
      "entropy": 5.705842638015747,
      "epoch": 3.9499801508535133,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003519056979469341,
      "loss": 4.979,
      "mean_token_accuracy": 0.1918591246008873,
      "num_tokens": 87193243.0,
      "step": 39800
    },
    {
      "entropy": 5.672225379943848,
      "epoch": 3.9504763795156808,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035187239948014014,
      "loss": 4.9883,
      "mean_token_accuracy": 0.18941735476255417,
      "num_tokens": 87203719.0,
      "step": 39805
    },
    {
      "entropy": 5.622618293762207,
      "epoch": 3.9509726081778482,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003518390991070765,
      "loss": 4.9611,
      "mean_token_accuracy": 0.1943182021379471,
      "num_tokens": 87215715.0,
      "step": 39810
    },
    {
      "entropy": 5.689743089675903,
      "epoch": 3.951468836840016,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035180579682856863,
      "loss": 5.0478,
      "mean_token_accuracy": 0.1911426454782486,
      "num_tokens": 87225547.0,
      "step": 39815
    },
    {
      "entropy": 5.710071277618408,
      "epoch": 3.9519650655021836,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003517724926454425,
      "loss": 5.0254,
      "mean_token_accuracy": 0.18448887318372725,
      "num_tokens": 87236215.0,
      "step": 39820
    },
    {
      "entropy": 5.681652784347534,
      "epoch": 3.952461294164351,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035173918655852406,
      "loss": 5.0208,
      "mean_token_accuracy": 0.18263752460479737,
      "num_tokens": 87246931.0,
      "step": 39825
    },
    {
      "entropy": 5.640718412399292,
      "epoch": 3.9529575228265186,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035170587856863907,
      "loss": 4.9395,
      "mean_token_accuracy": 0.1920606166124344,
      "num_tokens": 87258177.0,
      "step": 39830
    },
    {
      "entropy": 5.6846123218536375,
      "epoch": 3.953453751488686,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035167256867661364,
      "loss": 5.0275,
      "mean_token_accuracy": 0.188796666264534,
      "num_tokens": 87269725.0,
      "step": 39835
    },
    {
      "entropy": 5.746946954727173,
      "epoch": 3.9539499801508535,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003516392568832737,
      "loss": 5.0111,
      "mean_token_accuracy": 0.191780249774456,
      "num_tokens": 87280461.0,
      "step": 39840
    },
    {
      "entropy": 5.752814245223999,
      "epoch": 3.954446208813021,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035160594318944533,
      "loss": 5.0697,
      "mean_token_accuracy": 0.18645550757646562,
      "num_tokens": 87291102.0,
      "step": 39845
    },
    {
      "entropy": 5.689475202560425,
      "epoch": 3.9549424374751885,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035157262759595467,
      "loss": 4.9107,
      "mean_token_accuracy": 0.19888312369585037,
      "num_tokens": 87301170.0,
      "step": 39850
    },
    {
      "entropy": 5.619318342208862,
      "epoch": 3.955438666137356,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003515393101036278,
      "loss": 4.9389,
      "mean_token_accuracy": 0.18864035308361055,
      "num_tokens": 87311955.0,
      "step": 39855
    },
    {
      "entropy": 5.674220275878906,
      "epoch": 3.9559348947995234,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003515059907132909,
      "loss": 4.9737,
      "mean_token_accuracy": 0.18741790503263472,
      "num_tokens": 87322887.0,
      "step": 39860
    },
    {
      "entropy": 5.626154518127441,
      "epoch": 3.956431123461691,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003514726694257704,
      "loss": 4.9143,
      "mean_token_accuracy": 0.1938166633248329,
      "num_tokens": 87333303.0,
      "step": 39865
    },
    {
      "entropy": 5.72061185836792,
      "epoch": 3.956927352123859,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035143934624189246,
      "loss": 5.0105,
      "mean_token_accuracy": 0.1931595101952553,
      "num_tokens": 87345940.0,
      "step": 39870
    },
    {
      "entropy": 5.718320703506469,
      "epoch": 3.9574235807860263,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003514060211624834,
      "loss": 5.0428,
      "mean_token_accuracy": 0.18185298293828964,
      "num_tokens": 87356969.0,
      "step": 39875
    },
    {
      "entropy": 5.655746936798096,
      "epoch": 3.9579198094481938,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003513726941883697,
      "loss": 4.9581,
      "mean_token_accuracy": 0.1953206554055214,
      "num_tokens": 87367539.0,
      "step": 39880
    },
    {
      "entropy": 5.619280624389648,
      "epoch": 3.9584160381103612,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003513393653203777,
      "loss": 4.9588,
      "mean_token_accuracy": 0.19278037548065186,
      "num_tokens": 87378548.0,
      "step": 39885
    },
    {
      "entropy": 5.648782348632812,
      "epoch": 3.9589122667725287,
      "grad_norm": 1.125,
      "learning_rate": 0.00035130603455933403,
      "loss": 4.9916,
      "mean_token_accuracy": 0.19238712787628173,
      "num_tokens": 87390033.0,
      "step": 39890
    },
    {
      "entropy": 5.683283090591431,
      "epoch": 3.959408495434696,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035127270190606503,
      "loss": 5.0263,
      "mean_token_accuracy": 0.18921944350004197,
      "num_tokens": 87401859.0,
      "step": 39895
    },
    {
      "entropy": 5.640985679626465,
      "epoch": 3.9599047240968637,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003512393673613974,
      "loss": 4.9577,
      "mean_token_accuracy": 0.19307884573936462,
      "num_tokens": 87411649.0,
      "step": 39900
    },
    {
      "entropy": 5.629931354522705,
      "epoch": 3.9604009527590316,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035120603092615775,
      "loss": 4.9604,
      "mean_token_accuracy": 0.19074338376522065,
      "num_tokens": 87424117.0,
      "step": 39905
    },
    {
      "entropy": 5.644440174102783,
      "epoch": 3.960897181421199,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035117269260117283,
      "loss": 4.9178,
      "mean_token_accuracy": 0.19269930571317673,
      "num_tokens": 87434680.0,
      "step": 39910
    },
    {
      "entropy": 5.716622161865234,
      "epoch": 3.9613934100833665,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035113935238726925,
      "loss": 5.101,
      "mean_token_accuracy": 0.18225183784961702,
      "num_tokens": 87445579.0,
      "step": 39915
    },
    {
      "entropy": 5.7413441181182865,
      "epoch": 3.961889638745534,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003511060102852736,
      "loss": 5.0421,
      "mean_token_accuracy": 0.19360638856887818,
      "num_tokens": 87456814.0,
      "step": 39920
    },
    {
      "entropy": 5.632244539260864,
      "epoch": 3.9623858674077015,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035107266629601306,
      "loss": 5.031,
      "mean_token_accuracy": 0.18516792207956315,
      "num_tokens": 87468296.0,
      "step": 39925
    },
    {
      "entropy": 5.624480724334717,
      "epoch": 3.962882096069869,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035103932042031434,
      "loss": 4.9218,
      "mean_token_accuracy": 0.19375545680522918,
      "num_tokens": 87479167.0,
      "step": 39930
    },
    {
      "entropy": 5.64074478149414,
      "epoch": 3.9633783247320364,
      "grad_norm": 1.125,
      "learning_rate": 0.00035100597265900417,
      "loss": 4.9245,
      "mean_token_accuracy": 0.19606868773698807,
      "num_tokens": 87490131.0,
      "step": 39935
    },
    {
      "entropy": 5.659698915481568,
      "epoch": 3.963874553394204,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035097262301290983,
      "loss": 4.9967,
      "mean_token_accuracy": 0.18914442956447602,
      "num_tokens": 87500004.0,
      "step": 39940
    },
    {
      "entropy": 5.7718359470367435,
      "epoch": 3.9643707820563714,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003509392714828581,
      "loss": 5.0724,
      "mean_token_accuracy": 0.1830472394824028,
      "num_tokens": 87512614.0,
      "step": 39945
    },
    {
      "entropy": 5.768133497238159,
      "epoch": 3.964867010718539,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035090591806967604,
      "loss": 5.0986,
      "mean_token_accuracy": 0.18252113312482834,
      "num_tokens": 87523337.0,
      "step": 39950
    },
    {
      "entropy": 5.6666097164154055,
      "epoch": 3.965363239380707,
      "grad_norm": 1.359375,
      "learning_rate": 0.00035087256277419076,
      "loss": 5.0031,
      "mean_token_accuracy": 0.19077930152416228,
      "num_tokens": 87533777.0,
      "step": 39955
    },
    {
      "entropy": 5.635792303085327,
      "epoch": 3.9658594680428743,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003508392055972295,
      "loss": 4.914,
      "mean_token_accuracy": 0.1937286913394928,
      "num_tokens": 87544568.0,
      "step": 39960
    },
    {
      "entropy": 5.640694522857666,
      "epoch": 3.9663556967050417,
      "grad_norm": 1.25,
      "learning_rate": 0.0003508058465396192,
      "loss": 4.9333,
      "mean_token_accuracy": 0.19891726821660996,
      "num_tokens": 87554869.0,
      "step": 39965
    },
    {
      "entropy": 5.71570086479187,
      "epoch": 3.966851925367209,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003507724856021874,
      "loss": 5.0167,
      "mean_token_accuracy": 0.18696752637624742,
      "num_tokens": 87566114.0,
      "step": 39970
    },
    {
      "entropy": 5.7280525207519535,
      "epoch": 3.9673481540293767,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003507391227857612,
      "loss": 4.9852,
      "mean_token_accuracy": 0.1909272402524948,
      "num_tokens": 87577124.0,
      "step": 39975
    },
    {
      "entropy": 5.66514253616333,
      "epoch": 3.967844382691544,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003507057580911679,
      "loss": 4.9987,
      "mean_token_accuracy": 0.1893654376268387,
      "num_tokens": 87589355.0,
      "step": 39980
    },
    {
      "entropy": 5.61621413230896,
      "epoch": 3.9683406113537116,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035067239151923495,
      "loss": 4.9562,
      "mean_token_accuracy": 0.19200285375118256,
      "num_tokens": 87598669.0,
      "step": 39985
    },
    {
      "entropy": 5.610079765319824,
      "epoch": 3.9688368400158796,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035063902307078983,
      "loss": 5.0189,
      "mean_token_accuracy": 0.18887265622615815,
      "num_tokens": 87611261.0,
      "step": 39990
    },
    {
      "entropy": 5.647180223464966,
      "epoch": 3.969333068678047,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003506056527466598,
      "loss": 4.962,
      "mean_token_accuracy": 0.19298083931207657,
      "num_tokens": 87621560.0,
      "step": 39995
    },
    {
      "entropy": 5.647839593887329,
      "epoch": 3.9698292973402145,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003505722805476726,
      "loss": 4.9695,
      "mean_token_accuracy": 0.1922499045729637,
      "num_tokens": 87631758.0,
      "step": 40000
    },
    {
      "entropy": 5.623714780807495,
      "epoch": 3.970325526002382,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035053890647465566,
      "loss": 4.9581,
      "mean_token_accuracy": 0.1875542089343071,
      "num_tokens": 87642392.0,
      "step": 40005
    },
    {
      "entropy": 5.633951330184937,
      "epoch": 3.9708217546645495,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035050553052843654,
      "loss": 4.966,
      "mean_token_accuracy": 0.19609808325767517,
      "num_tokens": 87653102.0,
      "step": 40010
    },
    {
      "entropy": 5.65379524230957,
      "epoch": 3.971317983326717,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035047215270984304,
      "loss": 4.9697,
      "mean_token_accuracy": 0.18588090687990189,
      "num_tokens": 87662784.0,
      "step": 40015
    },
    {
      "entropy": 5.647459554672241,
      "epoch": 3.9718142119888844,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003504387730197028,
      "loss": 4.9995,
      "mean_token_accuracy": 0.19339995086193085,
      "num_tokens": 87674116.0,
      "step": 40020
    },
    {
      "entropy": 5.735450887680054,
      "epoch": 3.972310440651052,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003504053914588435,
      "loss": 4.9876,
      "mean_token_accuracy": 0.18911105692386626,
      "num_tokens": 87684679.0,
      "step": 40025
    },
    {
      "entropy": 5.712547969818115,
      "epoch": 3.9728066693132194,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000350372008028093,
      "loss": 5.0355,
      "mean_token_accuracy": 0.18944416344165801,
      "num_tokens": 87695538.0,
      "step": 40030
    },
    {
      "entropy": 5.702629709243775,
      "epoch": 3.973302897975387,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000350338622728279,
      "loss": 4.9969,
      "mean_token_accuracy": 0.19338536113500596,
      "num_tokens": 87706407.0,
      "step": 40035
    },
    {
      "entropy": 5.6786449432373045,
      "epoch": 3.9737991266375547,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035030523556022966,
      "loss": 5.0463,
      "mean_token_accuracy": 0.18434676080942153,
      "num_tokens": 87718019.0,
      "step": 40040
    },
    {
      "entropy": 5.698123979568481,
      "epoch": 3.974295355299722,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003502718465247727,
      "loss": 4.9889,
      "mean_token_accuracy": 0.18981347382068633,
      "num_tokens": 87728865.0,
      "step": 40045
    },
    {
      "entropy": 5.669926166534424,
      "epoch": 3.9747915839618897,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003502384556227361,
      "loss": 4.8496,
      "mean_token_accuracy": 0.20168451964855194,
      "num_tokens": 87739340.0,
      "step": 40050
    },
    {
      "entropy": 5.698825788497925,
      "epoch": 3.975287812624057,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003502050628549479,
      "loss": 5.0435,
      "mean_token_accuracy": 0.17881471812725067,
      "num_tokens": 87751949.0,
      "step": 40055
    },
    {
      "entropy": 5.7032825469970705,
      "epoch": 3.9757840412862246,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035017166822223626,
      "loss": 4.9825,
      "mean_token_accuracy": 0.18817174583673477,
      "num_tokens": 87760916.0,
      "step": 40060
    },
    {
      "entropy": 5.648023223876953,
      "epoch": 3.976280269948392,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003501382717254292,
      "loss": 5.0092,
      "mean_token_accuracy": 0.1876834139227867,
      "num_tokens": 87772605.0,
      "step": 40065
    },
    {
      "entropy": 5.654823064804077,
      "epoch": 3.9767764986105596,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000350104873365355,
      "loss": 4.9843,
      "mean_token_accuracy": 0.20265699177980423,
      "num_tokens": 87784244.0,
      "step": 40070
    },
    {
      "entropy": 5.723001194000244,
      "epoch": 3.9772727272727275,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035007147314284167,
      "loss": 5.0181,
      "mean_token_accuracy": 0.18514293134212495,
      "num_tokens": 87795910.0,
      "step": 40075
    },
    {
      "entropy": 5.727930879592895,
      "epoch": 3.977768955934895,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003500380710587176,
      "loss": 4.9602,
      "mean_token_accuracy": 0.19481066912412642,
      "num_tokens": 87808034.0,
      "step": 40080
    },
    {
      "entropy": 5.644664335250854,
      "epoch": 3.9782651845970625,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035000466711381103,
      "loss": 4.9474,
      "mean_token_accuracy": 0.19271131455898285,
      "num_tokens": 87819239.0,
      "step": 40085
    },
    {
      "entropy": 5.636694049835205,
      "epoch": 3.97876141325923,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034997126130895036,
      "loss": 4.9573,
      "mean_token_accuracy": 0.19394417554140092,
      "num_tokens": 87829794.0,
      "step": 40090
    },
    {
      "entropy": 5.6424013614654545,
      "epoch": 3.9792576419213974,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034993785364496394,
      "loss": 4.9056,
      "mean_token_accuracy": 0.1907037764787674,
      "num_tokens": 87839914.0,
      "step": 40095
    },
    {
      "entropy": 5.60338077545166,
      "epoch": 3.979753870583565,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034990444412268026,
      "loss": 4.9258,
      "mean_token_accuracy": 0.1931212440133095,
      "num_tokens": 87851268.0,
      "step": 40100
    },
    {
      "entropy": 5.6435212135314945,
      "epoch": 3.9802500992457324,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034987103274292775,
      "loss": 4.9378,
      "mean_token_accuracy": 0.19204432517290115,
      "num_tokens": 87862377.0,
      "step": 40105
    },
    {
      "entropy": 5.707551097869873,
      "epoch": 3.9807463279079,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034983761950653497,
      "loss": 5.0056,
      "mean_token_accuracy": 0.18861709386110306,
      "num_tokens": 87873820.0,
      "step": 40110
    },
    {
      "entropy": 5.641131401062012,
      "epoch": 3.9812425565700673,
      "grad_norm": 1.25,
      "learning_rate": 0.0003498042044143305,
      "loss": 4.9531,
      "mean_token_accuracy": 0.18901376873254777,
      "num_tokens": 87883706.0,
      "step": 40115
    },
    {
      "entropy": 5.62986388206482,
      "epoch": 3.981738785232235,
      "grad_norm": 1.140625,
      "learning_rate": 0.000349770787467143,
      "loss": 4.9625,
      "mean_token_accuracy": 0.18731945902109146,
      "num_tokens": 87895160.0,
      "step": 40120
    },
    {
      "entropy": 5.704854202270508,
      "epoch": 3.9822350138944023,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000349737368665801,
      "loss": 5.0395,
      "mean_token_accuracy": 0.1923261433839798,
      "num_tokens": 87905676.0,
      "step": 40125
    },
    {
      "entropy": 5.664074182510376,
      "epoch": 3.98273124255657,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034970394801113334,
      "loss": 4.9083,
      "mean_token_accuracy": 0.19930922687053682,
      "num_tokens": 87915264.0,
      "step": 40130
    },
    {
      "entropy": 5.7127758979797365,
      "epoch": 3.9832274712187377,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034967052550396874,
      "loss": 5.0127,
      "mean_token_accuracy": 0.19808752685785294,
      "num_tokens": 87925548.0,
      "step": 40135
    },
    {
      "entropy": 5.6333893775939945,
      "epoch": 3.983723699880905,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003496371011451359,
      "loss": 5.0022,
      "mean_token_accuracy": 0.19175055176019667,
      "num_tokens": 87936412.0,
      "step": 40140
    },
    {
      "entropy": 5.687804698944092,
      "epoch": 3.9842199285430726,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034960367493546394,
      "loss": 5.0299,
      "mean_token_accuracy": 0.18625862002372742,
      "num_tokens": 87948532.0,
      "step": 40145
    },
    {
      "entropy": 5.774885129928589,
      "epoch": 3.98471615720524,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034957024687578154,
      "loss": 5.0983,
      "mean_token_accuracy": 0.18716592192649842,
      "num_tokens": 87959990.0,
      "step": 40150
    },
    {
      "entropy": 5.650770425796509,
      "epoch": 3.9852123858674076,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034953681696691763,
      "loss": 4.9163,
      "mean_token_accuracy": 0.19491071999073029,
      "num_tokens": 87969970.0,
      "step": 40155
    },
    {
      "entropy": 5.693205833435059,
      "epoch": 3.9857086145295755,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003495033852097014,
      "loss": 5.0367,
      "mean_token_accuracy": 0.18492597639560698,
      "num_tokens": 87981424.0,
      "step": 40160
    },
    {
      "entropy": 5.668083333969117,
      "epoch": 3.986204843191743,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003494699516049617,
      "loss": 4.9674,
      "mean_token_accuracy": 0.19056334793567659,
      "num_tokens": 87992457.0,
      "step": 40165
    },
    {
      "entropy": 5.685184907913208,
      "epoch": 3.9867010718539104,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034943651615352764,
      "loss": 4.9803,
      "mean_token_accuracy": 0.18977568596601485,
      "num_tokens": 88002573.0,
      "step": 40170
    },
    {
      "entropy": 5.619118595123291,
      "epoch": 3.987197300516078,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034940307885622843,
      "loss": 4.9518,
      "mean_token_accuracy": 0.19601472318172455,
      "num_tokens": 88013432.0,
      "step": 40175
    },
    {
      "entropy": 5.6763440608978275,
      "epoch": 3.9876935291782454,
      "grad_norm": 1.125,
      "learning_rate": 0.00034936963971389317,
      "loss": 4.986,
      "mean_token_accuracy": 0.19134484231472015,
      "num_tokens": 88024401.0,
      "step": 40180
    },
    {
      "entropy": 5.71321816444397,
      "epoch": 3.988189757840413,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034933619872735104,
      "loss": 5.0357,
      "mean_token_accuracy": 0.18641890585422516,
      "num_tokens": 88036646.0,
      "step": 40185
    },
    {
      "entropy": 5.693007612228394,
      "epoch": 3.9886859865025803,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003493027558974314,
      "loss": 4.9573,
      "mean_token_accuracy": 0.19384788125753402,
      "num_tokens": 88046401.0,
      "step": 40190
    },
    {
      "entropy": 5.644950819015503,
      "epoch": 3.989182215164748,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034926931122496353,
      "loss": 4.9632,
      "mean_token_accuracy": 0.1884247690439224,
      "num_tokens": 88056835.0,
      "step": 40195
    },
    {
      "entropy": 5.701878929138184,
      "epoch": 3.9896784438269153,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034923586471077676,
      "loss": 5.028,
      "mean_token_accuracy": 0.18567515015602112,
      "num_tokens": 88068001.0,
      "step": 40200
    },
    {
      "entropy": 5.673937749862671,
      "epoch": 3.9901746724890828,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003492024163557005,
      "loss": 4.9839,
      "mean_token_accuracy": 0.19575088620185851,
      "num_tokens": 88079361.0,
      "step": 40205
    },
    {
      "entropy": 5.684457206726075,
      "epoch": 3.9906709011512502,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034916896616056433,
      "loss": 5.053,
      "mean_token_accuracy": 0.18815234005451204,
      "num_tokens": 88092693.0,
      "step": 40210
    },
    {
      "entropy": 5.715020751953125,
      "epoch": 3.991167129813418,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003491355141261975,
      "loss": 5.0316,
      "mean_token_accuracy": 0.18844382017850875,
      "num_tokens": 88105539.0,
      "step": 40215
    },
    {
      "entropy": 5.66787428855896,
      "epoch": 3.9916633584755856,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034910206025342975,
      "loss": 4.9224,
      "mean_token_accuracy": 0.19659174531698226,
      "num_tokens": 88116381.0,
      "step": 40220
    },
    {
      "entropy": 5.713368701934814,
      "epoch": 3.992159587137753,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003490686045430905,
      "loss": 4.9919,
      "mean_token_accuracy": 0.1915445700287819,
      "num_tokens": 88127354.0,
      "step": 40225
    },
    {
      "entropy": 5.727880907058716,
      "epoch": 3.9926558157999206,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003490351469960094,
      "loss": 4.9685,
      "mean_token_accuracy": 0.1916164740920067,
      "num_tokens": 88138874.0,
      "step": 40230
    },
    {
      "entropy": 5.689499378204346,
      "epoch": 3.993152044462088,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003490016876130163,
      "loss": 5.0596,
      "mean_token_accuracy": 0.1851910337805748,
      "num_tokens": 88150571.0,
      "step": 40235
    },
    {
      "entropy": 5.699102258682251,
      "epoch": 3.9936482731242555,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003489682263949408,
      "loss": 5.054,
      "mean_token_accuracy": 0.1805551066994667,
      "num_tokens": 88161315.0,
      "step": 40240
    },
    {
      "entropy": 5.7719165802001955,
      "epoch": 3.9941445017864234,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034893476334261266,
      "loss": 5.0856,
      "mean_token_accuracy": 0.18490242660045625,
      "num_tokens": 88173080.0,
      "step": 40245
    },
    {
      "entropy": 5.679038381576538,
      "epoch": 3.994640730448591,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034890129845686164,
      "loss": 4.9759,
      "mean_token_accuracy": 0.18876684159040452,
      "num_tokens": 88183729.0,
      "step": 40250
    },
    {
      "entropy": 5.633384084701538,
      "epoch": 3.9951369591107584,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003488678317385178,
      "loss": 4.9304,
      "mean_token_accuracy": 0.19184732884168626,
      "num_tokens": 88193844.0,
      "step": 40255
    },
    {
      "entropy": 5.634469366073608,
      "epoch": 3.995633187772926,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003488343631884108,
      "loss": 4.8986,
      "mean_token_accuracy": 0.2037298336625099,
      "num_tokens": 88205497.0,
      "step": 40260
    },
    {
      "entropy": 5.701921558380127,
      "epoch": 3.9961294164350933,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034880089280737064,
      "loss": 5.0215,
      "mean_token_accuracy": 0.1888105273246765,
      "num_tokens": 88216887.0,
      "step": 40265
    },
    {
      "entropy": 5.67374815940857,
      "epoch": 3.996625645097261,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034876742059622754,
      "loss": 5.0077,
      "mean_token_accuracy": 0.1862021267414093,
      "num_tokens": 88228542.0,
      "step": 40270
    },
    {
      "entropy": 5.718206834793091,
      "epoch": 3.9971218737594283,
      "grad_norm": 1.125,
      "learning_rate": 0.0003487339465558112,
      "loss": 5.0247,
      "mean_token_accuracy": 0.18820881694555283,
      "num_tokens": 88240831.0,
      "step": 40275
    },
    {
      "entropy": 5.732141160964966,
      "epoch": 3.9976181024215958,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034870047068695194,
      "loss": 5.0107,
      "mean_token_accuracy": 0.18830583691596986,
      "num_tokens": 88252483.0,
      "step": 40280
    },
    {
      "entropy": 5.681803798675537,
      "epoch": 3.9981143310837632,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003486669929904797,
      "loss": 5.0095,
      "mean_token_accuracy": 0.19237091839313508,
      "num_tokens": 88263087.0,
      "step": 40285
    },
    {
      "entropy": 5.657871580123901,
      "epoch": 3.9986105597459307,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034863351346722495,
      "loss": 4.9342,
      "mean_token_accuracy": 0.19764549136161805,
      "num_tokens": 88273609.0,
      "step": 40290
    },
    {
      "entropy": 5.6580994606018065,
      "epoch": 3.999106788408098,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003486000321180177,
      "loss": 4.9984,
      "mean_token_accuracy": 0.18929286003112794,
      "num_tokens": 88283502.0,
      "step": 40295
    },
    {
      "entropy": 5.664544534683228,
      "epoch": 3.999603017070266,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034856654894368816,
      "loss": 4.9342,
      "mean_token_accuracy": 0.18901742547750472,
      "num_tokens": 88294718.0,
      "step": 40300
    },
    {
      "entropy": 5.643552827835083,
      "epoch": 4.000099245732433,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003485330639450668,
      "loss": 5.0621,
      "mean_token_accuracy": 0.1904873341321945,
      "num_tokens": 88305865.0,
      "step": 40305
    },
    {
      "entropy": 5.617317390441895,
      "epoch": 4.000595474394601,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034849957712298384,
      "loss": 4.8643,
      "mean_token_accuracy": 0.19310030937194825,
      "num_tokens": 88317882.0,
      "step": 40310
    },
    {
      "entropy": 5.723337554931641,
      "epoch": 4.001091703056769,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034846608847826976,
      "loss": 4.9908,
      "mean_token_accuracy": 0.1909791797399521,
      "num_tokens": 88329968.0,
      "step": 40315
    },
    {
      "entropy": 5.676101636886597,
      "epoch": 4.0015879317189365,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003484325980117551,
      "loss": 4.9733,
      "mean_token_accuracy": 0.1895856514573097,
      "num_tokens": 88340736.0,
      "step": 40320
    },
    {
      "entropy": 5.57849645614624,
      "epoch": 4.002084160381104,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003483991057242702,
      "loss": 4.8417,
      "mean_token_accuracy": 0.19953030794858934,
      "num_tokens": 88351981.0,
      "step": 40325
    },
    {
      "entropy": 5.67624568939209,
      "epoch": 4.002580389043271,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034836561161664546,
      "loss": 4.9605,
      "mean_token_accuracy": 0.1920433223247528,
      "num_tokens": 88363367.0,
      "step": 40330
    },
    {
      "entropy": 5.757488870620728,
      "epoch": 4.003076617705439,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003483321156897118,
      "loss": 5.0712,
      "mean_token_accuracy": 0.18124190270900725,
      "num_tokens": 88374222.0,
      "step": 40335
    },
    {
      "entropy": 5.661091041564942,
      "epoch": 4.003572846367606,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003482986179442997,
      "loss": 4.9048,
      "mean_token_accuracy": 0.19612302929162978,
      "num_tokens": 88385618.0,
      "step": 40340
    },
    {
      "entropy": 5.684421586990356,
      "epoch": 4.004069075029774,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034826511838123976,
      "loss": 5.03,
      "mean_token_accuracy": 0.1839969888329506,
      "num_tokens": 88397940.0,
      "step": 40345
    },
    {
      "entropy": 5.6409577369689945,
      "epoch": 4.004565303691941,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003482316170013628,
      "loss": 4.9012,
      "mean_token_accuracy": 0.19424138814210892,
      "num_tokens": 88407832.0,
      "step": 40350
    },
    {
      "entropy": 5.664892292022705,
      "epoch": 4.005061532354109,
      "grad_norm": 1.453125,
      "learning_rate": 0.00034819811380549956,
      "loss": 4.9409,
      "mean_token_accuracy": 0.19944825619459153,
      "num_tokens": 88419116.0,
      "step": 40355
    },
    {
      "entropy": 5.670008850097656,
      "epoch": 4.005557761016276,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034816460879448074,
      "loss": 4.9776,
      "mean_token_accuracy": 0.19069681763648988,
      "num_tokens": 88429999.0,
      "step": 40360
    },
    {
      "entropy": 5.632705497741699,
      "epoch": 4.006053989678444,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034813110196913735,
      "loss": 4.8893,
      "mean_token_accuracy": 0.20187792629003526,
      "num_tokens": 88440636.0,
      "step": 40365
    },
    {
      "entropy": 5.643341159820556,
      "epoch": 4.006550218340611,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003480975933303002,
      "loss": 4.9105,
      "mean_token_accuracy": 0.19381000548601152,
      "num_tokens": 88451650.0,
      "step": 40370
    },
    {
      "entropy": 5.702470827102661,
      "epoch": 4.007046447002779,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003480640828788002,
      "loss": 4.9817,
      "mean_token_accuracy": 0.1954888567328453,
      "num_tokens": 88463366.0,
      "step": 40375
    },
    {
      "entropy": 5.697323131561279,
      "epoch": 4.007542675664946,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003480305706154684,
      "loss": 4.9406,
      "mean_token_accuracy": 0.19486427903175355,
      "num_tokens": 88474933.0,
      "step": 40380
    },
    {
      "entropy": 5.586291790008545,
      "epoch": 4.008038904327114,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003479970565411359,
      "loss": 4.8979,
      "mean_token_accuracy": 0.2062658429145813,
      "num_tokens": 88485239.0,
      "step": 40385
    },
    {
      "entropy": 5.662531137466431,
      "epoch": 4.008535132989281,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034796354065663363,
      "loss": 4.9765,
      "mean_token_accuracy": 0.18989480137825013,
      "num_tokens": 88496140.0,
      "step": 40390
    },
    {
      "entropy": 5.665545701980591,
      "epoch": 4.009031361651449,
      "grad_norm": 1.25,
      "learning_rate": 0.00034793002296279273,
      "loss": 4.9601,
      "mean_token_accuracy": 0.19401289373636246,
      "num_tokens": 88506642.0,
      "step": 40395
    },
    {
      "entropy": 5.678291749954224,
      "epoch": 4.009527590313617,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003478965034604445,
      "loss": 4.9319,
      "mean_token_accuracy": 0.19037131667137147,
      "num_tokens": 88517405.0,
      "step": 40400
    },
    {
      "entropy": 5.698646926879883,
      "epoch": 4.010023818975784,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034786298215041996,
      "loss": 4.9301,
      "mean_token_accuracy": 0.19248116463422776,
      "num_tokens": 88530175.0,
      "step": 40405
    },
    {
      "entropy": 5.612314605712891,
      "epoch": 4.010520047637952,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034782945903355045,
      "loss": 4.9018,
      "mean_token_accuracy": 0.20286653637886048,
      "num_tokens": 88540528.0,
      "step": 40410
    },
    {
      "entropy": 5.657964038848877,
      "epoch": 4.011016276300119,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003477959341106675,
      "loss": 4.8887,
      "mean_token_accuracy": 0.19303394854068756,
      "num_tokens": 88550293.0,
      "step": 40415
    },
    {
      "entropy": 5.6900207042694095,
      "epoch": 4.011512504962287,
      "grad_norm": 1.25,
      "learning_rate": 0.000347762407382602,
      "loss": 5.0008,
      "mean_token_accuracy": 0.18577521592378615,
      "num_tokens": 88560671.0,
      "step": 40420
    },
    {
      "entropy": 5.628305292129516,
      "epoch": 4.012008733624454,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003477288788501857,
      "loss": 4.8749,
      "mean_token_accuracy": 0.19733494371175767,
      "num_tokens": 88571178.0,
      "step": 40425
    },
    {
      "entropy": 5.656347465515137,
      "epoch": 4.012504962286622,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034769534851424993,
      "loss": 4.999,
      "mean_token_accuracy": 0.18664661943912506,
      "num_tokens": 88582544.0,
      "step": 40430
    },
    {
      "entropy": 5.662238740921021,
      "epoch": 4.013001190948789,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034766181637562615,
      "loss": 4.9743,
      "mean_token_accuracy": 0.18727030754089355,
      "num_tokens": 88594619.0,
      "step": 40435
    },
    {
      "entropy": 5.6711118698120115,
      "epoch": 4.013497419610957,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003476282824351459,
      "loss": 4.9893,
      "mean_token_accuracy": 0.19037070572376252,
      "num_tokens": 88606689.0,
      "step": 40440
    },
    {
      "entropy": 5.723898410797119,
      "epoch": 4.013993648273124,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034759474669364074,
      "loss": 4.9809,
      "mean_token_accuracy": 0.191744627058506,
      "num_tokens": 88617221.0,
      "step": 40445
    },
    {
      "entropy": 5.625821781158447,
      "epoch": 4.014489876935292,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034756120915194227,
      "loss": 4.8835,
      "mean_token_accuracy": 0.19947515577077865,
      "num_tokens": 88627806.0,
      "step": 40450
    },
    {
      "entropy": 5.6116503238677975,
      "epoch": 4.014986105597459,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003475276698108822,
      "loss": 4.8335,
      "mean_token_accuracy": 0.1997131496667862,
      "num_tokens": 88639247.0,
      "step": 40455
    },
    {
      "entropy": 5.661409330368042,
      "epoch": 4.015482334259627,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034749412867129216,
      "loss": 4.9939,
      "mean_token_accuracy": 0.19408609867095947,
      "num_tokens": 88651657.0,
      "step": 40460
    },
    {
      "entropy": 5.7104846954345705,
      "epoch": 4.015978562921794,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034746058573400395,
      "loss": 4.991,
      "mean_token_accuracy": 0.18970878422260284,
      "num_tokens": 88663066.0,
      "step": 40465
    },
    {
      "entropy": 5.704178237915039,
      "epoch": 4.016474791583962,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034742704099984944,
      "loss": 4.9855,
      "mean_token_accuracy": 0.19560023695230483,
      "num_tokens": 88673890.0,
      "step": 40470
    },
    {
      "entropy": 5.6956933498382565,
      "epoch": 4.016971020246129,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003473934944696603,
      "loss": 4.964,
      "mean_token_accuracy": 0.18931989818811418,
      "num_tokens": 88683866.0,
      "step": 40475
    },
    {
      "entropy": 5.7130903720855715,
      "epoch": 4.0174672489082965,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034735994614426845,
      "loss": 4.996,
      "mean_token_accuracy": 0.19289758503437043,
      "num_tokens": 88695344.0,
      "step": 40480
    },
    {
      "entropy": 5.662121295928955,
      "epoch": 4.017963477570465,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034732639602450596,
      "loss": 4.8771,
      "mean_token_accuracy": 0.19431998282670976,
      "num_tokens": 88706527.0,
      "step": 40485
    },
    {
      "entropy": 5.6941228866577145,
      "epoch": 4.018459706232632,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003472928441112046,
      "loss": 4.9926,
      "mean_token_accuracy": 0.18914889246225358,
      "num_tokens": 88716541.0,
      "step": 40490
    },
    {
      "entropy": 5.68125810623169,
      "epoch": 4.0189559348948,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034725929040519657,
      "loss": 4.9978,
      "mean_token_accuracy": 0.19017185866832734,
      "num_tokens": 88728240.0,
      "step": 40495
    },
    {
      "entropy": 5.709426593780518,
      "epoch": 4.019452163556967,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003472257349073138,
      "loss": 4.9545,
      "mean_token_accuracy": 0.19301152527332305,
      "num_tokens": 88738116.0,
      "step": 40500
    },
    {
      "entropy": 5.707544422149658,
      "epoch": 4.019948392219135,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034719217761838844,
      "loss": 4.9646,
      "mean_token_accuracy": 0.1978837087750435,
      "num_tokens": 88750048.0,
      "step": 40505
    },
    {
      "entropy": 5.748730754852295,
      "epoch": 4.020444620881302,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003471586185392526,
      "loss": 5.0595,
      "mean_token_accuracy": 0.18636551201343537,
      "num_tokens": 88761589.0,
      "step": 40510
    },
    {
      "entropy": 5.674753618240357,
      "epoch": 4.02094084954347,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003471250576707386,
      "loss": 4.9308,
      "mean_token_accuracy": 0.19598788321018218,
      "num_tokens": 88770542.0,
      "step": 40515
    },
    {
      "entropy": 5.68879017829895,
      "epoch": 4.021437078205637,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034709149501367854,
      "loss": 5.0583,
      "mean_token_accuracy": 0.19081955254077912,
      "num_tokens": 88780927.0,
      "step": 40520
    },
    {
      "entropy": 5.711091136932373,
      "epoch": 4.021933306867805,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034705793056890466,
      "loss": 4.9744,
      "mean_token_accuracy": 0.18915590196847915,
      "num_tokens": 88793045.0,
      "step": 40525
    },
    {
      "entropy": 5.742126560211181,
      "epoch": 4.022429535529972,
      "grad_norm": 1.25,
      "learning_rate": 0.0003470243643372495,
      "loss": 5.049,
      "mean_token_accuracy": 0.18491017073392868,
      "num_tokens": 88806021.0,
      "step": 40530
    },
    {
      "entropy": 5.715581369400025,
      "epoch": 4.02292576419214,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003469907963195452,
      "loss": 4.9685,
      "mean_token_accuracy": 0.18794668912887574,
      "num_tokens": 88817664.0,
      "step": 40535
    },
    {
      "entropy": 5.6256204605102536,
      "epoch": 4.023421992854307,
      "grad_norm": 1.25,
      "learning_rate": 0.00034695722651662436,
      "loss": 4.8804,
      "mean_token_accuracy": 0.1989100381731987,
      "num_tokens": 88828115.0,
      "step": 40540
    },
    {
      "entropy": 5.693813800811768,
      "epoch": 4.023918221516475,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034692365492931934,
      "loss": 4.9664,
      "mean_token_accuracy": 0.19074371308088303,
      "num_tokens": 88839161.0,
      "step": 40545
    },
    {
      "entropy": 5.65583610534668,
      "epoch": 4.024414450178642,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034689008155846265,
      "loss": 4.8824,
      "mean_token_accuracy": 0.2001733273267746,
      "num_tokens": 88849513.0,
      "step": 40550
    },
    {
      "entropy": 5.6183388233184814,
      "epoch": 4.0249106788408096,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034685650640488687,
      "loss": 4.9376,
      "mean_token_accuracy": 0.18975962847471237,
      "num_tokens": 88860836.0,
      "step": 40555
    },
    {
      "entropy": 5.71219835281372,
      "epoch": 4.025406907502977,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034682292946942443,
      "loss": 4.993,
      "mean_token_accuracy": 0.18867028951644899,
      "num_tokens": 88871219.0,
      "step": 40560
    },
    {
      "entropy": 5.6727019309997555,
      "epoch": 4.0259031361651445,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034678935075290823,
      "loss": 4.9632,
      "mean_token_accuracy": 0.1886887803673744,
      "num_tokens": 88882469.0,
      "step": 40565
    },
    {
      "entropy": 5.675182771682739,
      "epoch": 4.026399364827313,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003467557702561708,
      "loss": 4.9524,
      "mean_token_accuracy": 0.1905936673283577,
      "num_tokens": 88892365.0,
      "step": 40570
    },
    {
      "entropy": 5.683483600616455,
      "epoch": 4.02689559348948,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003467221879800449,
      "loss": 4.9508,
      "mean_token_accuracy": 0.19463596493005753,
      "num_tokens": 88903693.0,
      "step": 40575
    },
    {
      "entropy": 5.599474716186523,
      "epoch": 4.027391822151648,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003466886039253633,
      "loss": 4.8496,
      "mean_token_accuracy": 0.20091892629861832,
      "num_tokens": 88913824.0,
      "step": 40580
    },
    {
      "entropy": 5.658615350723267,
      "epoch": 4.027888050813815,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034665501809295885,
      "loss": 4.9922,
      "mean_token_accuracy": 0.19288252294063568,
      "num_tokens": 88925135.0,
      "step": 40585
    },
    {
      "entropy": 5.690458297729492,
      "epoch": 4.028384279475983,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003466214304836642,
      "loss": 4.9976,
      "mean_token_accuracy": 0.2002732202410698,
      "num_tokens": 88937271.0,
      "step": 40590
    },
    {
      "entropy": 5.741262531280517,
      "epoch": 4.02888050813815,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003465878410983125,
      "loss": 4.9812,
      "mean_token_accuracy": 0.18887909501791,
      "num_tokens": 88948705.0,
      "step": 40595
    },
    {
      "entropy": 5.707525968551636,
      "epoch": 4.029376736800318,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003465542499377366,
      "loss": 4.9627,
      "mean_token_accuracy": 0.19067139476537703,
      "num_tokens": 88959700.0,
      "step": 40600
    },
    {
      "entropy": 5.719354343414307,
      "epoch": 4.029872965462485,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034652065700276947,
      "loss": 4.9982,
      "mean_token_accuracy": 0.1911868005990982,
      "num_tokens": 88971032.0,
      "step": 40605
    },
    {
      "entropy": 5.682363080978393,
      "epoch": 4.030369194124653,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003464870622942441,
      "loss": 4.964,
      "mean_token_accuracy": 0.19040148854255676,
      "num_tokens": 88981025.0,
      "step": 40610
    },
    {
      "entropy": 5.748150110244751,
      "epoch": 4.03086542278682,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034645346581299374,
      "loss": 5.0462,
      "mean_token_accuracy": 0.18795938044786453,
      "num_tokens": 88991938.0,
      "step": 40615
    },
    {
      "entropy": 5.611617851257324,
      "epoch": 4.031361651448988,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003464198675598514,
      "loss": 4.9082,
      "mean_token_accuracy": 0.19404507726430892,
      "num_tokens": 89002830.0,
      "step": 40620
    },
    {
      "entropy": 5.629322528839111,
      "epoch": 4.031857880111155,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003463862675356501,
      "loss": 4.8854,
      "mean_token_accuracy": 0.19733034074306488,
      "num_tokens": 89014499.0,
      "step": 40625
    },
    {
      "entropy": 5.637065505981445,
      "epoch": 4.032354108773323,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003463526657412233,
      "loss": 4.9545,
      "mean_token_accuracy": 0.18899251222610475,
      "num_tokens": 89026238.0,
      "step": 40630
    },
    {
      "entropy": 5.73188591003418,
      "epoch": 4.03285033743549,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034631906217740416,
      "loss": 4.995,
      "mean_token_accuracy": 0.18875957578420638,
      "num_tokens": 89036737.0,
      "step": 40635
    },
    {
      "entropy": 5.685699033737182,
      "epoch": 4.0333465660976575,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003462854568450259,
      "loss": 4.9316,
      "mean_token_accuracy": 0.19848347157239915,
      "num_tokens": 89047618.0,
      "step": 40640
    },
    {
      "entropy": 5.648240852355957,
      "epoch": 4.033842794759825,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003462518497449219,
      "loss": 4.9693,
      "mean_token_accuracy": 0.19826885163784028,
      "num_tokens": 89058230.0,
      "step": 40645
    },
    {
      "entropy": 5.672377300262451,
      "epoch": 4.0343390234219925,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034621824087792565,
      "loss": 4.9355,
      "mean_token_accuracy": 0.19986315965652465,
      "num_tokens": 89068636.0,
      "step": 40650
    },
    {
      "entropy": 5.665335702896118,
      "epoch": 4.034835252084161,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003461846302448704,
      "loss": 4.941,
      "mean_token_accuracy": 0.19821031093597413,
      "num_tokens": 89079311.0,
      "step": 40655
    },
    {
      "entropy": 5.6577354907989506,
      "epoch": 4.035331480746328,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003461510178465897,
      "loss": 4.9497,
      "mean_token_accuracy": 0.19229115694761276,
      "num_tokens": 89090123.0,
      "step": 40660
    },
    {
      "entropy": 5.64031343460083,
      "epoch": 4.035827709408496,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003461174036839172,
      "loss": 4.9791,
      "mean_token_accuracy": 0.18736107796430587,
      "num_tokens": 89101306.0,
      "step": 40665
    },
    {
      "entropy": 5.702320289611817,
      "epoch": 4.036323938070663,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034608378775768624,
      "loss": 4.9618,
      "mean_token_accuracy": 0.19580505341291427,
      "num_tokens": 89111869.0,
      "step": 40670
    },
    {
      "entropy": 5.686390113830567,
      "epoch": 4.036820166732831,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034605017006873046,
      "loss": 4.9079,
      "mean_token_accuracy": 0.19357299208641052,
      "num_tokens": 89122364.0,
      "step": 40675
    },
    {
      "entropy": 5.654354238510132,
      "epoch": 4.037316395394998,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003460165506178837,
      "loss": 4.9101,
      "mean_token_accuracy": 0.19145083874464036,
      "num_tokens": 89132748.0,
      "step": 40680
    },
    {
      "entropy": 5.616783332824707,
      "epoch": 4.037812624057166,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034598292940597936,
      "loss": 4.8847,
      "mean_token_accuracy": 0.19772280752658844,
      "num_tokens": 89142585.0,
      "step": 40685
    },
    {
      "entropy": 5.741355037689209,
      "epoch": 4.038308852719333,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003459493064338514,
      "loss": 5.0243,
      "mean_token_accuracy": 0.1900045245885849,
      "num_tokens": 89153764.0,
      "step": 40690
    },
    {
      "entropy": 5.748879098892212,
      "epoch": 4.038805081381501,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034591568170233355,
      "loss": 5.0204,
      "mean_token_accuracy": 0.18810850083827974,
      "num_tokens": 89165194.0,
      "step": 40695
    },
    {
      "entropy": 5.693468999862671,
      "epoch": 4.039301310043668,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034588205521225956,
      "loss": 4.9492,
      "mean_token_accuracy": 0.1945725977420807,
      "num_tokens": 89176002.0,
      "step": 40700
    },
    {
      "entropy": 5.711300611495972,
      "epoch": 4.039797538705836,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034584842696446335,
      "loss": 4.9723,
      "mean_token_accuracy": 0.19639545679092407,
      "num_tokens": 89187059.0,
      "step": 40705
    },
    {
      "entropy": 5.611379814147949,
      "epoch": 4.040293767368003,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034581479695977883,
      "loss": 4.8469,
      "mean_token_accuracy": 0.20325925797224045,
      "num_tokens": 89198119.0,
      "step": 40710
    },
    {
      "entropy": 5.64737639427185,
      "epoch": 4.0407899960301705,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003457811651990399,
      "loss": 4.9157,
      "mean_token_accuracy": 0.19151916205883027,
      "num_tokens": 89207801.0,
      "step": 40715
    },
    {
      "entropy": 5.634268188476563,
      "epoch": 4.041286224692338,
      "grad_norm": 1.25,
      "learning_rate": 0.0003457475316830806,
      "loss": 4.9619,
      "mean_token_accuracy": 0.19246250092983247,
      "num_tokens": 89218807.0,
      "step": 40720
    },
    {
      "entropy": 5.719259214401245,
      "epoch": 4.0417824533545055,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003457138964127351,
      "loss": 5.0221,
      "mean_token_accuracy": 0.1821428880095482,
      "num_tokens": 89229987.0,
      "step": 40725
    },
    {
      "entropy": 5.632569694519043,
      "epoch": 4.042278682016673,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034568025938883713,
      "loss": 4.8442,
      "mean_token_accuracy": 0.19717966616153718,
      "num_tokens": 89239740.0,
      "step": 40730
    },
    {
      "entropy": 5.668107509613037,
      "epoch": 4.04277491067884,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003456466206122212,
      "loss": 4.9349,
      "mean_token_accuracy": 0.18993010371923447,
      "num_tokens": 89251425.0,
      "step": 40735
    },
    {
      "entropy": 5.659937810897827,
      "epoch": 4.043271139341008,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034561298008372124,
      "loss": 4.9052,
      "mean_token_accuracy": 0.1951296404004097,
      "num_tokens": 89261557.0,
      "step": 40740
    },
    {
      "entropy": 5.6050961971282955,
      "epoch": 4.043767368003176,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003455793378041715,
      "loss": 4.8283,
      "mean_token_accuracy": 0.19836435317993165,
      "num_tokens": 89273912.0,
      "step": 40745
    },
    {
      "entropy": 5.699406528472901,
      "epoch": 4.044263596665344,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003455456937744063,
      "loss": 5.021,
      "mean_token_accuracy": 0.19002495855093002,
      "num_tokens": 89285298.0,
      "step": 40750
    },
    {
      "entropy": 5.665165567398072,
      "epoch": 4.044759825327511,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034551204799525986,
      "loss": 4.9435,
      "mean_token_accuracy": 0.18457690477371216,
      "num_tokens": 89295658.0,
      "step": 40755
    },
    {
      "entropy": 5.639411973953247,
      "epoch": 4.045256053989679,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034547840046756663,
      "loss": 4.9135,
      "mean_token_accuracy": 0.1951925724744797,
      "num_tokens": 89305666.0,
      "step": 40760
    },
    {
      "entropy": 5.733706426620484,
      "epoch": 4.045752282651846,
      "grad_norm": 1.234375,
      "learning_rate": 0.000345444751192161,
      "loss": 5.0852,
      "mean_token_accuracy": 0.18325369209051132,
      "num_tokens": 89317409.0,
      "step": 40765
    },
    {
      "entropy": 5.743470859527588,
      "epoch": 4.046248511314014,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034541110016987717,
      "loss": 5.0208,
      "mean_token_accuracy": 0.18434273898601533,
      "num_tokens": 89327843.0,
      "step": 40770
    },
    {
      "entropy": 5.676656579971313,
      "epoch": 4.046744739976181,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034537744740154985,
      "loss": 4.9822,
      "mean_token_accuracy": 0.19181403517723083,
      "num_tokens": 89339220.0,
      "step": 40775
    },
    {
      "entropy": 5.710796976089478,
      "epoch": 4.047240968638349,
      "grad_norm": 1.125,
      "learning_rate": 0.0003453437928880135,
      "loss": 5.0444,
      "mean_token_accuracy": 0.18431578874588012,
      "num_tokens": 89350929.0,
      "step": 40780
    },
    {
      "entropy": 5.751556396484375,
      "epoch": 4.047737197300516,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003453101366301026,
      "loss": 5.0096,
      "mean_token_accuracy": 0.19709178805351257,
      "num_tokens": 89361756.0,
      "step": 40785
    },
    {
      "entropy": 5.718288850784302,
      "epoch": 4.0482334259626835,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034527647862865184,
      "loss": 4.9571,
      "mean_token_accuracy": 0.19332235008478166,
      "num_tokens": 89371025.0,
      "step": 40790
    },
    {
      "entropy": 5.72558069229126,
      "epoch": 4.048729654624851,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003452428188844959,
      "loss": 5.0599,
      "mean_token_accuracy": 0.18209753632545472,
      "num_tokens": 89382907.0,
      "step": 40795
    },
    {
      "entropy": 5.638120985031128,
      "epoch": 4.0492258832870185,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034520915739846944,
      "loss": 4.9617,
      "mean_token_accuracy": 0.19192172586917877,
      "num_tokens": 89393981.0,
      "step": 40800
    },
    {
      "entropy": 5.616751289367675,
      "epoch": 4.049722111949186,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003451754941714071,
      "loss": 4.9404,
      "mean_token_accuracy": 0.18758370280265807,
      "num_tokens": 89405031.0,
      "step": 40805
    },
    {
      "entropy": 5.633360815048218,
      "epoch": 4.050218340611353,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003451418292041437,
      "loss": 4.8867,
      "mean_token_accuracy": 0.20419261157512664,
      "num_tokens": 89415566.0,
      "step": 40810
    },
    {
      "entropy": 5.648554277420044,
      "epoch": 4.050714569273521,
      "grad_norm": 1.25,
      "learning_rate": 0.00034510816249751415,
      "loss": 4.9631,
      "mean_token_accuracy": 0.1908249005675316,
      "num_tokens": 89427472.0,
      "step": 40815
    },
    {
      "entropy": 5.703057670593262,
      "epoch": 4.051210797935688,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003450744940523532,
      "loss": 4.9582,
      "mean_token_accuracy": 0.18447377979755403,
      "num_tokens": 89438415.0,
      "step": 40820
    },
    {
      "entropy": 5.631769180297852,
      "epoch": 4.051707026597856,
      "grad_norm": 1.25,
      "learning_rate": 0.0003450408238694958,
      "loss": 4.8949,
      "mean_token_accuracy": 0.19530619382858277,
      "num_tokens": 89450368.0,
      "step": 40825
    },
    {
      "entropy": 5.643687295913696,
      "epoch": 4.052203255260024,
      "grad_norm": 1.25,
      "learning_rate": 0.0003450071519497769,
      "loss": 4.984,
      "mean_token_accuracy": 0.18629985600709914,
      "num_tokens": 89461739.0,
      "step": 40830
    },
    {
      "entropy": 5.687065887451172,
      "epoch": 4.052699483922192,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034497347829403143,
      "loss": 4.9732,
      "mean_token_accuracy": 0.18672031462192534,
      "num_tokens": 89472098.0,
      "step": 40835
    },
    {
      "entropy": 5.671959400177002,
      "epoch": 4.053195712584359,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003449398029030946,
      "loss": 4.9876,
      "mean_token_accuracy": 0.1898678183555603,
      "num_tokens": 89482339.0,
      "step": 40840
    },
    {
      "entropy": 5.6360830783844,
      "epoch": 4.053691941246527,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034490612577780123,
      "loss": 4.9119,
      "mean_token_accuracy": 0.1932302698493004,
      "num_tokens": 89495741.0,
      "step": 40845
    },
    {
      "entropy": 5.733791160583496,
      "epoch": 4.054188169908694,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003448724469189866,
      "loss": 5.1125,
      "mean_token_accuracy": 0.19271971732378007,
      "num_tokens": 89506899.0,
      "step": 40850
    },
    {
      "entropy": 5.710840177536011,
      "epoch": 4.054684398570862,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003448387663274859,
      "loss": 4.9806,
      "mean_token_accuracy": 0.19357494711875917,
      "num_tokens": 89517335.0,
      "step": 40855
    },
    {
      "entropy": 5.654714441299438,
      "epoch": 4.055180627233029,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003448050840041342,
      "loss": 5.0072,
      "mean_token_accuracy": 0.18456026762723923,
      "num_tokens": 89529050.0,
      "step": 40860
    },
    {
      "entropy": 5.740567016601562,
      "epoch": 4.0556768558951966,
      "grad_norm": 1.265625,
      "learning_rate": 0.000344771399949767,
      "loss": 5.0561,
      "mean_token_accuracy": 0.18407191336154938,
      "num_tokens": 89539061.0,
      "step": 40865
    },
    {
      "entropy": 5.692728710174561,
      "epoch": 4.056173084557364,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034473771416521933,
      "loss": 4.94,
      "mean_token_accuracy": 0.19366440922021866,
      "num_tokens": 89550526.0,
      "step": 40870
    },
    {
      "entropy": 5.699691247940064,
      "epoch": 4.0566693132195315,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003447040266513265,
      "loss": 4.9086,
      "mean_token_accuracy": 0.19669813066720962,
      "num_tokens": 89560875.0,
      "step": 40875
    },
    {
      "entropy": 5.6459863662719725,
      "epoch": 4.057165541881699,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003446703374089242,
      "loss": 4.8975,
      "mean_token_accuracy": 0.1948913663625717,
      "num_tokens": 89571745.0,
      "step": 40880
    },
    {
      "entropy": 5.6660243511199955,
      "epoch": 4.0576617705438665,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003446366464388475,
      "loss": 4.9367,
      "mean_token_accuracy": 0.194085094332695,
      "num_tokens": 89581580.0,
      "step": 40885
    },
    {
      "entropy": 5.64026894569397,
      "epoch": 4.058157999206034,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003446029537419321,
      "loss": 4.8988,
      "mean_token_accuracy": 0.18780339360237122,
      "num_tokens": 89591454.0,
      "step": 40890
    },
    {
      "entropy": 5.5621131420135494,
      "epoch": 4.058654227868201,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003445692593190135,
      "loss": 4.8759,
      "mean_token_accuracy": 0.19981831312179565,
      "num_tokens": 89603795.0,
      "step": 40895
    },
    {
      "entropy": 5.673361492156983,
      "epoch": 4.059150456530369,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034453556317092703,
      "loss": 5.0619,
      "mean_token_accuracy": 0.18650183826684952,
      "num_tokens": 89616371.0,
      "step": 40900
    },
    {
      "entropy": 5.698376607894898,
      "epoch": 4.059646685192536,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003445018652985085,
      "loss": 4.9813,
      "mean_token_accuracy": 0.19169520288705827,
      "num_tokens": 89630413.0,
      "step": 40905
    },
    {
      "entropy": 5.708157873153686,
      "epoch": 4.060142913854704,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034446816570259353,
      "loss": 4.9782,
      "mean_token_accuracy": 0.19513836055994033,
      "num_tokens": 89642827.0,
      "step": 40910
    },
    {
      "entropy": 5.663966941833496,
      "epoch": 4.060639142516872,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003444344643840177,
      "loss": 4.9388,
      "mean_token_accuracy": 0.19172575026750566,
      "num_tokens": 89653153.0,
      "step": 40915
    },
    {
      "entropy": 5.693213272094726,
      "epoch": 4.06113537117904,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003444007613436167,
      "loss": 5.021,
      "mean_token_accuracy": 0.19461528807878495,
      "num_tokens": 89664748.0,
      "step": 40920
    },
    {
      "entropy": 5.643749809265136,
      "epoch": 4.061631599841207,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034436705658222647,
      "loss": 4.9261,
      "mean_token_accuracy": 0.20248932391405106,
      "num_tokens": 89677314.0,
      "step": 40925
    },
    {
      "entropy": 5.702971363067627,
      "epoch": 4.062127828503375,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034433335010068265,
      "loss": 4.9882,
      "mean_token_accuracy": 0.19228726774454116,
      "num_tokens": 89687860.0,
      "step": 40930
    },
    {
      "entropy": 5.631107425689697,
      "epoch": 4.062624057165542,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003442996418998211,
      "loss": 4.8597,
      "mean_token_accuracy": 0.20127118676900863,
      "num_tokens": 89699353.0,
      "step": 40935
    },
    {
      "entropy": 5.64798903465271,
      "epoch": 4.06312028582771,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003442659319804779,
      "loss": 4.9586,
      "mean_token_accuracy": 0.1912019819021225,
      "num_tokens": 89710823.0,
      "step": 40940
    },
    {
      "entropy": 5.753216171264649,
      "epoch": 4.063616514489877,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034423222034348877,
      "loss": 5.0299,
      "mean_token_accuracy": 0.18674510717391968,
      "num_tokens": 89721120.0,
      "step": 40945
    },
    {
      "entropy": 5.572028350830078,
      "epoch": 4.0641127431520445,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003441985069896897,
      "loss": 4.9062,
      "mean_token_accuracy": 0.1961537316441536,
      "num_tokens": 89731382.0,
      "step": 40950
    },
    {
      "entropy": 5.641415786743164,
      "epoch": 4.064608971814212,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034416479191991684,
      "loss": 4.9672,
      "mean_token_accuracy": 0.19449826031923295,
      "num_tokens": 89743441.0,
      "step": 40955
    },
    {
      "entropy": 5.756848001480103,
      "epoch": 4.0651052004763795,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003441310751350062,
      "loss": 5.0939,
      "mean_token_accuracy": 0.1905396282672882,
      "num_tokens": 89753417.0,
      "step": 40960
    },
    {
      "entropy": 5.7138293266296385,
      "epoch": 4.065601429138547,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034409735663579373,
      "loss": 4.9812,
      "mean_token_accuracy": 0.18651594072580338,
      "num_tokens": 89763719.0,
      "step": 40965
    },
    {
      "entropy": 5.6702769756317135,
      "epoch": 4.066097657800714,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034406363642311584,
      "loss": 4.9228,
      "mean_token_accuracy": 0.18663577139377593,
      "num_tokens": 89774500.0,
      "step": 40970
    },
    {
      "entropy": 5.691404485702515,
      "epoch": 4.066593886462882,
      "grad_norm": 1.25,
      "learning_rate": 0.0003440299144978085,
      "loss": 4.9444,
      "mean_token_accuracy": 0.18404300212860109,
      "num_tokens": 89784744.0,
      "step": 40975
    },
    {
      "entropy": 5.688155269622802,
      "epoch": 4.067090115125049,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034399619086070805,
      "loss": 4.9423,
      "mean_token_accuracy": 0.19673338830471038,
      "num_tokens": 89796081.0,
      "step": 40980
    },
    {
      "entropy": 5.661661672592163,
      "epoch": 4.067586343787217,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003439624655126507,
      "loss": 4.9841,
      "mean_token_accuracy": 0.19189904779195785,
      "num_tokens": 89806110.0,
      "step": 40985
    },
    {
      "entropy": 5.652533006668091,
      "epoch": 4.068082572449384,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034392873845447294,
      "loss": 4.8737,
      "mean_token_accuracy": 0.1997438043355942,
      "num_tokens": 89816239.0,
      "step": 40990
    },
    {
      "entropy": 5.64741883277893,
      "epoch": 4.068578801111552,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034389500968701095,
      "loss": 4.9289,
      "mean_token_accuracy": 0.19793522357940674,
      "num_tokens": 89825907.0,
      "step": 40995
    },
    {
      "entropy": 5.758502054214477,
      "epoch": 4.06907502977372,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034386127921110106,
      "loss": 4.9467,
      "mean_token_accuracy": 0.19483443796634675,
      "num_tokens": 89835909.0,
      "step": 41000
    },
    {
      "entropy": 5.756403398513794,
      "epoch": 4.069571258435888,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034382754702757997,
      "loss": 5.0386,
      "mean_token_accuracy": 0.1863629087805748,
      "num_tokens": 89847231.0,
      "step": 41005
    },
    {
      "entropy": 5.692169046401977,
      "epoch": 4.070067487098055,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003437938131372839,
      "loss": 5.0109,
      "mean_token_accuracy": 0.18930069506168365,
      "num_tokens": 89858811.0,
      "step": 41010
    },
    {
      "entropy": 5.63442029953003,
      "epoch": 4.070563715760223,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034376007754104957,
      "loss": 4.9148,
      "mean_token_accuracy": 0.19235399067401887,
      "num_tokens": 89869028.0,
      "step": 41015
    },
    {
      "entropy": 5.637424087524414,
      "epoch": 4.07105994442239,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003437263402397135,
      "loss": 4.9189,
      "mean_token_accuracy": 0.18312270194292068,
      "num_tokens": 89879293.0,
      "step": 41020
    },
    {
      "entropy": 5.660398960113525,
      "epoch": 4.0715561730845575,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034369260123411224,
      "loss": 4.9426,
      "mean_token_accuracy": 0.19053535908460617,
      "num_tokens": 89888877.0,
      "step": 41025
    },
    {
      "entropy": 5.6640924453735355,
      "epoch": 4.072052401746725,
      "grad_norm": 1.25,
      "learning_rate": 0.00034365886052508245,
      "loss": 4.9385,
      "mean_token_accuracy": 0.1894696280360222,
      "num_tokens": 89899642.0,
      "step": 41030
    },
    {
      "entropy": 5.669953060150147,
      "epoch": 4.0725486304088925,
      "grad_norm": 1.171875,
      "learning_rate": 0.000343625118113461,
      "loss": 4.9245,
      "mean_token_accuracy": 0.19357889145612717,
      "num_tokens": 89909811.0,
      "step": 41035
    },
    {
      "entropy": 5.66796088218689,
      "epoch": 4.07304485907106,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003435913740000844,
      "loss": 5.0271,
      "mean_token_accuracy": 0.18542267829179765,
      "num_tokens": 89921226.0,
      "step": 41040
    },
    {
      "entropy": 5.6854204654693605,
      "epoch": 4.073541087733227,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034355762818578955,
      "loss": 4.9836,
      "mean_token_accuracy": 0.18853577375411987,
      "num_tokens": 89932638.0,
      "step": 41045
    },
    {
      "entropy": 5.595631217956543,
      "epoch": 4.074037316395395,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034352388067141317,
      "loss": 4.8646,
      "mean_token_accuracy": 0.19814518243074417,
      "num_tokens": 89941968.0,
      "step": 41050
    },
    {
      "entropy": 5.56805453300476,
      "epoch": 4.074533545057562,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003434901314577923,
      "loss": 4.8652,
      "mean_token_accuracy": 0.2003420487046242,
      "num_tokens": 89952526.0,
      "step": 41055
    },
    {
      "entropy": 5.651809310913086,
      "epoch": 4.07502977371973,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003434563805457638,
      "loss": 4.9201,
      "mean_token_accuracy": 0.19274004995822908,
      "num_tokens": 89963205.0,
      "step": 41060
    },
    {
      "entropy": 5.7079907894134525,
      "epoch": 4.075526002381897,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003434226279361644,
      "loss": 5.0458,
      "mean_token_accuracy": 0.19068393409252166,
      "num_tokens": 89973322.0,
      "step": 41065
    },
    {
      "entropy": 5.617147445678711,
      "epoch": 4.076022231044065,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034338887362983137,
      "loss": 4.9429,
      "mean_token_accuracy": 0.1977676898241043,
      "num_tokens": 89984209.0,
      "step": 41070
    },
    {
      "entropy": 5.746996450424194,
      "epoch": 4.076518459706232,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003433551176276017,
      "loss": 5.0289,
      "mean_token_accuracy": 0.1882155179977417,
      "num_tokens": 89994645.0,
      "step": 41075
    },
    {
      "entropy": 5.734232187271118,
      "epoch": 4.0770146883684,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034332135993031234,
      "loss": 5.0215,
      "mean_token_accuracy": 0.18997889310121535,
      "num_tokens": 90006731.0,
      "step": 41080
    },
    {
      "entropy": 5.649905490875244,
      "epoch": 4.077510917030568,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003432876005388005,
      "loss": 4.9763,
      "mean_token_accuracy": 0.1986313581466675,
      "num_tokens": 90017657.0,
      "step": 41085
    },
    {
      "entropy": 5.637042379379272,
      "epoch": 4.078007145692736,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003432538394539033,
      "loss": 5.0077,
      "mean_token_accuracy": 0.18909488767385482,
      "num_tokens": 90028131.0,
      "step": 41090
    },
    {
      "entropy": 5.691683006286621,
      "epoch": 4.078503374354903,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003432200766764579,
      "loss": 4.9728,
      "mean_token_accuracy": 0.19317446202039718,
      "num_tokens": 90037780.0,
      "step": 41095
    },
    {
      "entropy": 5.736012935638428,
      "epoch": 4.0789996030170705,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003431863122073016,
      "loss": 5.0616,
      "mean_token_accuracy": 0.18004817366600037,
      "num_tokens": 90048580.0,
      "step": 41100
    },
    {
      "entropy": 5.677513313293457,
      "epoch": 4.079495831679238,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003431525460472718,
      "loss": 4.9666,
      "mean_token_accuracy": 0.1968951106071472,
      "num_tokens": 90059799.0,
      "step": 41105
    },
    {
      "entropy": 5.648508739471436,
      "epoch": 4.0799920603414055,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003431187781972056,
      "loss": 4.8886,
      "mean_token_accuracy": 0.20079572200775148,
      "num_tokens": 90069834.0,
      "step": 41110
    },
    {
      "entropy": 5.635216760635376,
      "epoch": 4.080488289003573,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003430850086579405,
      "loss": 4.9914,
      "mean_token_accuracy": 0.18875017166137695,
      "num_tokens": 90080726.0,
      "step": 41115
    },
    {
      "entropy": 5.656982898712158,
      "epoch": 4.08098451766574,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034305123743031394,
      "loss": 4.9481,
      "mean_token_accuracy": 0.18798502832651137,
      "num_tokens": 90091657.0,
      "step": 41120
    },
    {
      "entropy": 5.607025003433227,
      "epoch": 4.081480746327908,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034301746451516335,
      "loss": 4.9181,
      "mean_token_accuracy": 0.1834754839539528,
      "num_tokens": 90102415.0,
      "step": 41125
    },
    {
      "entropy": 5.730799961090088,
      "epoch": 4.081976974990075,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003429836899133261,
      "loss": 5.0255,
      "mean_token_accuracy": 0.1867552548646927,
      "num_tokens": 90113575.0,
      "step": 41130
    },
    {
      "entropy": 5.730903673171997,
      "epoch": 4.082473203652243,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034294991362563993,
      "loss": 4.9677,
      "mean_token_accuracy": 0.19435562789440156,
      "num_tokens": 90126653.0,
      "step": 41135
    },
    {
      "entropy": 5.637128210067749,
      "epoch": 4.08296943231441,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034291613565294235,
      "loss": 4.8665,
      "mean_token_accuracy": 0.1990796595811844,
      "num_tokens": 90138319.0,
      "step": 41140
    },
    {
      "entropy": 5.655110931396484,
      "epoch": 4.083465660976578,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003428823559960708,
      "loss": 4.8543,
      "mean_token_accuracy": 0.20122159868478776,
      "num_tokens": 90150232.0,
      "step": 41145
    },
    {
      "entropy": 5.609376430511475,
      "epoch": 4.083961889638745,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034284857465586324,
      "loss": 4.9129,
      "mean_token_accuracy": 0.196977499127388,
      "num_tokens": 90160733.0,
      "step": 41150
    },
    {
      "entropy": 5.708690214157104,
      "epoch": 4.084458118300913,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003428147916331571,
      "loss": 4.9819,
      "mean_token_accuracy": 0.18961664736270906,
      "num_tokens": 90171216.0,
      "step": 41155
    },
    {
      "entropy": 5.657727575302124,
      "epoch": 4.08495434696308,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034278100692879045,
      "loss": 4.8768,
      "mean_token_accuracy": 0.20270583182573318,
      "num_tokens": 90181312.0,
      "step": 41160
    },
    {
      "entropy": 5.7121009349823,
      "epoch": 4.085450575625248,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003427472205436007,
      "loss": 4.9605,
      "mean_token_accuracy": 0.19391910284757613,
      "num_tokens": 90191560.0,
      "step": 41165
    },
    {
      "entropy": 5.568203973770141,
      "epoch": 4.085946804287415,
      "grad_norm": 1.171875,
      "learning_rate": 0.000342713432478426,
      "loss": 4.9153,
      "mean_token_accuracy": 0.19178076088428497,
      "num_tokens": 90203242.0,
      "step": 41170
    },
    {
      "entropy": 5.632547998428345,
      "epoch": 4.0864430329495836,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000342679642734104,
      "loss": 4.9678,
      "mean_token_accuracy": 0.19152954071760178,
      "num_tokens": 90213780.0,
      "step": 41175
    },
    {
      "entropy": 5.7189891815185545,
      "epoch": 4.086939261611751,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003426458513114727,
      "loss": 4.9848,
      "mean_token_accuracy": 0.18821874558925628,
      "num_tokens": 90225108.0,
      "step": 41180
    },
    {
      "entropy": 5.649827098846435,
      "epoch": 4.0874354902739185,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034261205821137003,
      "loss": 4.9215,
      "mean_token_accuracy": 0.19729012846946717,
      "num_tokens": 90236801.0,
      "step": 41185
    },
    {
      "entropy": 5.627101421356201,
      "epoch": 4.087931718936086,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003425782634346341,
      "loss": 4.9293,
      "mean_token_accuracy": 0.1851354032754898,
      "num_tokens": 90247979.0,
      "step": 41190
    },
    {
      "entropy": 5.639218473434449,
      "epoch": 4.0884279475982535,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034254446698210275,
      "loss": 4.9614,
      "mean_token_accuracy": 0.19564262628555298,
      "num_tokens": 90258673.0,
      "step": 41195
    },
    {
      "entropy": 5.636295413970947,
      "epoch": 4.088924176260421,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003425106688546142,
      "loss": 4.9625,
      "mean_token_accuracy": 0.19452748745679854,
      "num_tokens": 90269865.0,
      "step": 41200
    },
    {
      "entropy": 5.713640356063843,
      "epoch": 4.089420404922588,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034247686905300657,
      "loss": 4.9805,
      "mean_token_accuracy": 0.18513719290494918,
      "num_tokens": 90280660.0,
      "step": 41205
    },
    {
      "entropy": 5.685787868499756,
      "epoch": 4.089916633584756,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003424430675781179,
      "loss": 4.9219,
      "mean_token_accuracy": 0.19949894696474074,
      "num_tokens": 90290253.0,
      "step": 41210
    },
    {
      "entropy": 5.614272356033325,
      "epoch": 4.090412862246923,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003424092644307865,
      "loss": 4.9084,
      "mean_token_accuracy": 0.19674240350723265,
      "num_tokens": 90300928.0,
      "step": 41215
    },
    {
      "entropy": 5.571532106399536,
      "epoch": 4.090909090909091,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034237545961185076,
      "loss": 4.8807,
      "mean_token_accuracy": 0.2023369088768959,
      "num_tokens": 90311579.0,
      "step": 41220
    },
    {
      "entropy": 5.648460245132446,
      "epoch": 4.091405319571258,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034234165312214867,
      "loss": 4.9529,
      "mean_token_accuracy": 0.18895126283168792,
      "num_tokens": 90322029.0,
      "step": 41225
    },
    {
      "entropy": 5.677551937103272,
      "epoch": 4.091901548233426,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003423078449625186,
      "loss": 4.9733,
      "mean_token_accuracy": 0.19091717004776002,
      "num_tokens": 90332629.0,
      "step": 41230
    },
    {
      "entropy": 5.762178754806518,
      "epoch": 4.092397776895593,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003422740351337992,
      "loss": 5.0107,
      "mean_token_accuracy": 0.1868589073419571,
      "num_tokens": 90342823.0,
      "step": 41235
    },
    {
      "entropy": 5.644744110107422,
      "epoch": 4.092894005557761,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003422402236368286,
      "loss": 4.8656,
      "mean_token_accuracy": 0.19888791888952256,
      "num_tokens": 90353200.0,
      "step": 41240
    },
    {
      "entropy": 5.606747388839722,
      "epoch": 4.093390234219928,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034220641047244527,
      "loss": 4.8829,
      "mean_token_accuracy": 0.20386362373828887,
      "num_tokens": 90363686.0,
      "step": 41245
    },
    {
      "entropy": 5.673586368560791,
      "epoch": 4.093886462882096,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003421725956414879,
      "loss": 4.9541,
      "mean_token_accuracy": 0.19885706752538682,
      "num_tokens": 90375827.0,
      "step": 41250
    },
    {
      "entropy": 5.676612806320191,
      "epoch": 4.094382691544263,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003421387791447949,
      "loss": 4.9583,
      "mean_token_accuracy": 0.19028660506010056,
      "num_tokens": 90386820.0,
      "step": 41255
    },
    {
      "entropy": 5.685366153717041,
      "epoch": 4.0948789202064315,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034210496098320477,
      "loss": 4.9998,
      "mean_token_accuracy": 0.19155848920345306,
      "num_tokens": 90398505.0,
      "step": 41260
    },
    {
      "entropy": 5.677049827575684,
      "epoch": 4.095375148868599,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034207114115755624,
      "loss": 4.968,
      "mean_token_accuracy": 0.18866465240716934,
      "num_tokens": 90411859.0,
      "step": 41265
    },
    {
      "entropy": 5.6905537128448485,
      "epoch": 4.0958713775307665,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034203731966868796,
      "loss": 5.0108,
      "mean_token_accuracy": 0.18349012285470961,
      "num_tokens": 90423497.0,
      "step": 41270
    },
    {
      "entropy": 5.725957632064819,
      "epoch": 4.096367606192934,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003420034965174385,
      "loss": 5.0819,
      "mean_token_accuracy": 0.18111883252859115,
      "num_tokens": 90434634.0,
      "step": 41275
    },
    {
      "entropy": 5.691781902313233,
      "epoch": 4.096863834855101,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003419696717046468,
      "loss": 4.955,
      "mean_token_accuracy": 0.1914307028055191,
      "num_tokens": 90445736.0,
      "step": 41280
    },
    {
      "entropy": 5.706198406219483,
      "epoch": 4.097360063517269,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034193584523115153,
      "loss": 4.9488,
      "mean_token_accuracy": 0.1989550843834877,
      "num_tokens": 90457922.0,
      "step": 41285
    },
    {
      "entropy": 5.671479892730713,
      "epoch": 4.097856292179436,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003419020170977915,
      "loss": 5.0236,
      "mean_token_accuracy": 0.18787164986133575,
      "num_tokens": 90470280.0,
      "step": 41290
    },
    {
      "entropy": 5.682135534286499,
      "epoch": 4.098352520841604,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003418681873054057,
      "loss": 5.019,
      "mean_token_accuracy": 0.18363943547010422,
      "num_tokens": 90481164.0,
      "step": 41295
    },
    {
      "entropy": 5.683955907821655,
      "epoch": 4.098848749503771,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034183435585483286,
      "loss": 4.9383,
      "mean_token_accuracy": 0.19485785216093063,
      "num_tokens": 90492643.0,
      "step": 41300
    },
    {
      "entropy": 5.603369140625,
      "epoch": 4.099344978165939,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000341800522746912,
      "loss": 4.891,
      "mean_token_accuracy": 0.19704224169254303,
      "num_tokens": 90503417.0,
      "step": 41305
    },
    {
      "entropy": 5.635453224182129,
      "epoch": 4.099841206828106,
      "grad_norm": 1.25,
      "learning_rate": 0.00034176668798248215,
      "loss": 4.9449,
      "mean_token_accuracy": 0.20020754635334015,
      "num_tokens": 90513461.0,
      "step": 41310
    },
    {
      "entropy": 5.6848550796508786,
      "epoch": 4.100337435490274,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003417328515623823,
      "loss": 4.999,
      "mean_token_accuracy": 0.189508056640625,
      "num_tokens": 90523265.0,
      "step": 41315
    },
    {
      "entropy": 5.738196134567261,
      "epoch": 4.100833664152441,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034169901348745166,
      "loss": 5.0375,
      "mean_token_accuracy": 0.18735682368278503,
      "num_tokens": 90533320.0,
      "step": 41320
    },
    {
      "entropy": 5.658511972427368,
      "epoch": 4.101329892814609,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000341665173758529,
      "loss": 4.9407,
      "mean_token_accuracy": 0.19342105239629745,
      "num_tokens": 90544117.0,
      "step": 41325
    },
    {
      "entropy": 5.7173693656921385,
      "epoch": 4.101826121476776,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003416313323764538,
      "loss": 4.9302,
      "mean_token_accuracy": 0.1949279636144638,
      "num_tokens": 90555178.0,
      "step": 41330
    },
    {
      "entropy": 5.685364627838135,
      "epoch": 4.102322350138944,
      "grad_norm": 1.125,
      "learning_rate": 0.00034159748934206515,
      "loss": 4.9644,
      "mean_token_accuracy": 0.19606650918722152,
      "num_tokens": 90567006.0,
      "step": 41335
    },
    {
      "entropy": 5.61380672454834,
      "epoch": 4.102818578801111,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003415636446562022,
      "loss": 4.8833,
      "mean_token_accuracy": 0.1945845142006874,
      "num_tokens": 90578132.0,
      "step": 41340
    },
    {
      "entropy": 5.637407922744751,
      "epoch": 4.1033148074632795,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034152979831970426,
      "loss": 4.9279,
      "mean_token_accuracy": 0.20067830234766007,
      "num_tokens": 90589295.0,
      "step": 41345
    },
    {
      "entropy": 5.6611058712005615,
      "epoch": 4.103811036125447,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034149595033341074,
      "loss": 4.9781,
      "mean_token_accuracy": 0.19914866834878922,
      "num_tokens": 90600268.0,
      "step": 41350
    },
    {
      "entropy": 5.603460264205933,
      "epoch": 4.104307264787614,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034146210069816097,
      "loss": 4.8779,
      "mean_token_accuracy": 0.19897636771202087,
      "num_tokens": 90610430.0,
      "step": 41355
    },
    {
      "entropy": 5.684274196624756,
      "epoch": 4.104803493449782,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003414282494147943,
      "loss": 4.9201,
      "mean_token_accuracy": 0.1936122789978981,
      "num_tokens": 90622207.0,
      "step": 41360
    },
    {
      "entropy": 5.694236516952515,
      "epoch": 4.105299722111949,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003413943964841501,
      "loss": 4.9212,
      "mean_token_accuracy": 0.19416435658931733,
      "num_tokens": 90632663.0,
      "step": 41365
    },
    {
      "entropy": 5.638825464248657,
      "epoch": 4.105795950774117,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034136054190706807,
      "loss": 4.9811,
      "mean_token_accuracy": 0.19219327718019485,
      "num_tokens": 90643384.0,
      "step": 41370
    },
    {
      "entropy": 5.6503393173217775,
      "epoch": 4.106292179436284,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003413266856843875,
      "loss": 4.8911,
      "mean_token_accuracy": 0.19965369999408722,
      "num_tokens": 90653015.0,
      "step": 41375
    },
    {
      "entropy": 5.682181644439697,
      "epoch": 4.106788408098452,
      "grad_norm": 1.25,
      "learning_rate": 0.0003412928278169481,
      "loss": 4.9999,
      "mean_token_accuracy": 0.18886620104312896,
      "num_tokens": 90662848.0,
      "step": 41380
    },
    {
      "entropy": 5.665849494934082,
      "epoch": 4.107284636760619,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003412589683055894,
      "loss": 4.8914,
      "mean_token_accuracy": 0.19533362090587617,
      "num_tokens": 90673103.0,
      "step": 41385
    },
    {
      "entropy": 5.700487852096558,
      "epoch": 4.107780865422787,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003412251071511511,
      "loss": 4.9971,
      "mean_token_accuracy": 0.19589365869760514,
      "num_tokens": 90683628.0,
      "step": 41390
    },
    {
      "entropy": 5.743531560897827,
      "epoch": 4.108277094084954,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034119124435447277,
      "loss": 5.0771,
      "mean_token_accuracy": 0.1860151246190071,
      "num_tokens": 90694461.0,
      "step": 41395
    },
    {
      "entropy": 5.683261966705322,
      "epoch": 4.108773322747122,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003411573799163942,
      "loss": 4.9562,
      "mean_token_accuracy": 0.19377855062484742,
      "num_tokens": 90705488.0,
      "step": 41400
    },
    {
      "entropy": 5.722251749038696,
      "epoch": 4.109269551409289,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003411235138377552,
      "loss": 5.0309,
      "mean_token_accuracy": 0.19255347549915314,
      "num_tokens": 90716357.0,
      "step": 41405
    },
    {
      "entropy": 5.684658670425415,
      "epoch": 4.109765780071457,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034108964611939555,
      "loss": 4.8589,
      "mean_token_accuracy": 0.2002853572368622,
      "num_tokens": 90726953.0,
      "step": 41410
    },
    {
      "entropy": 5.612747049331665,
      "epoch": 4.110262008733624,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003410557767621551,
      "loss": 4.9392,
      "mean_token_accuracy": 0.19874405413866042,
      "num_tokens": 90737568.0,
      "step": 41415
    },
    {
      "entropy": 5.712411975860595,
      "epoch": 4.110758237395792,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003410219057668737,
      "loss": 4.9218,
      "mean_token_accuracy": 0.1970430865883827,
      "num_tokens": 90747843.0,
      "step": 41420
    },
    {
      "entropy": 5.687972640991211,
      "epoch": 4.111254466057959,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003409880331343913,
      "loss": 4.9692,
      "mean_token_accuracy": 0.19630108326673507,
      "num_tokens": 90758549.0,
      "step": 41425
    },
    {
      "entropy": 5.7440423488616945,
      "epoch": 4.111750694720127,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034095415886554786,
      "loss": 5.0002,
      "mean_token_accuracy": 0.18357352167367935,
      "num_tokens": 90769797.0,
      "step": 41430
    },
    {
      "entropy": 5.684551572799682,
      "epoch": 4.112246923382295,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003409202829611834,
      "loss": 4.9718,
      "mean_token_accuracy": 0.1923826664686203,
      "num_tokens": 90780733.0,
      "step": 41435
    },
    {
      "entropy": 5.702896022796631,
      "epoch": 4.112743152044462,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034088640542213796,
      "loss": 4.9763,
      "mean_token_accuracy": 0.19343501329421997,
      "num_tokens": 90792650.0,
      "step": 41440
    },
    {
      "entropy": 5.6655097007751465,
      "epoch": 4.11323938070663,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003408525262492517,
      "loss": 4.9091,
      "mean_token_accuracy": 0.19627908766269683,
      "num_tokens": 90804426.0,
      "step": 41445
    },
    {
      "entropy": 5.689194202423096,
      "epoch": 4.113735609368797,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003408186454433646,
      "loss": 4.9407,
      "mean_token_accuracy": 0.18979527205228805,
      "num_tokens": 90815972.0,
      "step": 41450
    },
    {
      "entropy": 5.704244804382324,
      "epoch": 4.114231838030965,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034078476300531703,
      "loss": 5.05,
      "mean_token_accuracy": 0.18426838964223863,
      "num_tokens": 90826818.0,
      "step": 41455
    },
    {
      "entropy": 5.690662813186646,
      "epoch": 4.114728066693132,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034075087893594905,
      "loss": 4.9175,
      "mean_token_accuracy": 0.19797505140304567,
      "num_tokens": 90837243.0,
      "step": 41460
    },
    {
      "entropy": 5.701224422454834,
      "epoch": 4.1152242953553,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003407169932361008,
      "loss": 4.9954,
      "mean_token_accuracy": 0.18943558782339096,
      "num_tokens": 90848370.0,
      "step": 41465
    },
    {
      "entropy": 5.609849071502685,
      "epoch": 4.115720524017467,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034068310590661295,
      "loss": 4.8448,
      "mean_token_accuracy": 0.19616610407829285,
      "num_tokens": 90859156.0,
      "step": 41470
    },
    {
      "entropy": 5.74413571357727,
      "epoch": 4.116216752679635,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034064921694832546,
      "loss": 4.8881,
      "mean_token_accuracy": 0.1947987213730812,
      "num_tokens": 90870018.0,
      "step": 41475
    },
    {
      "entropy": 5.630806493759155,
      "epoch": 4.116712981341802,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003406153263620789,
      "loss": 4.9358,
      "mean_token_accuracy": 0.1951359272003174,
      "num_tokens": 90881000.0,
      "step": 41480
    },
    {
      "entropy": 5.611512660980225,
      "epoch": 4.11720921000397,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034058143414871374,
      "loss": 4.8756,
      "mean_token_accuracy": 0.19859308302402495,
      "num_tokens": 90892472.0,
      "step": 41485
    },
    {
      "entropy": 5.6839171886444095,
      "epoch": 4.117705438666137,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003405475403090701,
      "loss": 4.956,
      "mean_token_accuracy": 0.19420014768838884,
      "num_tokens": 90902670.0,
      "step": 41490
    },
    {
      "entropy": 5.735025548934937,
      "epoch": 4.118201667328305,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003405136448439889,
      "loss": 5.0298,
      "mean_token_accuracy": 0.185147824883461,
      "num_tokens": 90913568.0,
      "step": 41495
    },
    {
      "entropy": 5.559654903411865,
      "epoch": 4.118697895990472,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034047974775431043,
      "loss": 4.7812,
      "mean_token_accuracy": 0.19935414642095567,
      "num_tokens": 90923923.0,
      "step": 41500
    },
    {
      "entropy": 5.618366670608521,
      "epoch": 4.11919412465264,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003404458490408753,
      "loss": 4.9212,
      "mean_token_accuracy": 0.1888945683836937,
      "num_tokens": 90934221.0,
      "step": 41505
    },
    {
      "entropy": 5.664949178695679,
      "epoch": 4.119690353314807,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003404119487045241,
      "loss": 4.8796,
      "mean_token_accuracy": 0.19879865497350693,
      "num_tokens": 90944591.0,
      "step": 41510
    },
    {
      "entropy": 5.696749639511109,
      "epoch": 4.120186581976975,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034037804674609757,
      "loss": 4.9959,
      "mean_token_accuracy": 0.19225968569517135,
      "num_tokens": 90956626.0,
      "step": 41515
    },
    {
      "entropy": 5.647229528427124,
      "epoch": 4.120682810639143,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034034414316643635,
      "loss": 4.904,
      "mean_token_accuracy": 0.19350158274173737,
      "num_tokens": 90966577.0,
      "step": 41520
    },
    {
      "entropy": 5.621485805511474,
      "epoch": 4.12117903930131,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003403102379663811,
      "loss": 4.8323,
      "mean_token_accuracy": 0.20485022217035292,
      "num_tokens": 90978508.0,
      "step": 41525
    },
    {
      "entropy": 5.686321163177491,
      "epoch": 4.121675267963478,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003402763311467728,
      "loss": 4.963,
      "mean_token_accuracy": 0.19494287073612213,
      "num_tokens": 90988828.0,
      "step": 41530
    },
    {
      "entropy": 5.75260362625122,
      "epoch": 4.122171496625645,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003402424227084521,
      "loss": 5.0764,
      "mean_token_accuracy": 0.19164266735315322,
      "num_tokens": 90999681.0,
      "step": 41535
    },
    {
      "entropy": 5.725604391098022,
      "epoch": 4.122667725287813,
      "grad_norm": 1.25,
      "learning_rate": 0.0003402085126522599,
      "loss": 4.9951,
      "mean_token_accuracy": 0.19107638597488402,
      "num_tokens": 91011478.0,
      "step": 41540
    },
    {
      "entropy": 5.623208999633789,
      "epoch": 4.12316395394998,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003401746009790371,
      "loss": 4.9297,
      "mean_token_accuracy": 0.19668405950069429,
      "num_tokens": 91023531.0,
      "step": 41545
    },
    {
      "entropy": 5.648989057540893,
      "epoch": 4.123660182612148,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034014068768962465,
      "loss": 4.9228,
      "mean_token_accuracy": 0.1933478444814682,
      "num_tokens": 91033747.0,
      "step": 41550
    },
    {
      "entropy": 5.63108263015747,
      "epoch": 4.124156411274315,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034010677278486346,
      "loss": 4.9158,
      "mean_token_accuracy": 0.196421879529953,
      "num_tokens": 91045328.0,
      "step": 41555
    },
    {
      "entropy": 5.579654884338379,
      "epoch": 4.124652639936483,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034007285626559464,
      "loss": 4.8043,
      "mean_token_accuracy": 0.20370319932699205,
      "num_tokens": 91055183.0,
      "step": 41560
    },
    {
      "entropy": 5.652181625366211,
      "epoch": 4.12514886859865,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003400389381326592,
      "loss": 4.969,
      "mean_token_accuracy": 0.18708838373422623,
      "num_tokens": 91065355.0,
      "step": 41565
    },
    {
      "entropy": 5.729584407806397,
      "epoch": 4.125645097260818,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003400050183868982,
      "loss": 5.012,
      "mean_token_accuracy": 0.18834496140480042,
      "num_tokens": 91075902.0,
      "step": 41570
    },
    {
      "entropy": 5.693599700927734,
      "epoch": 4.126141325922985,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003399710970291529,
      "loss": 4.9825,
      "mean_token_accuracy": 0.18683143109083175,
      "num_tokens": 91087283.0,
      "step": 41575
    },
    {
      "entropy": 5.712442350387573,
      "epoch": 4.126637554585153,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003399371740602643,
      "loss": 5.0201,
      "mean_token_accuracy": 0.18939029574394226,
      "num_tokens": 91098751.0,
      "step": 41580
    },
    {
      "entropy": 5.6744317531585695,
      "epoch": 4.12713378324732,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003399032494810738,
      "loss": 4.9664,
      "mean_token_accuracy": 0.19277502000331878,
      "num_tokens": 91109982.0,
      "step": 41585
    },
    {
      "entropy": 5.611943626403809,
      "epoch": 4.1276300119094875,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003398693232924225,
      "loss": 4.9692,
      "mean_token_accuracy": 0.19273831695318222,
      "num_tokens": 91121071.0,
      "step": 41590
    },
    {
      "entropy": 5.610275173187256,
      "epoch": 4.128126240571655,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033983539549515174,
      "loss": 4.9073,
      "mean_token_accuracy": 0.21035386472940446,
      "num_tokens": 91132212.0,
      "step": 41595
    },
    {
      "entropy": 5.6885905265808105,
      "epoch": 4.128622469233823,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033980146609010296,
      "loss": 4.9189,
      "mean_token_accuracy": 0.19567176699638367,
      "num_tokens": 91143199.0,
      "step": 41600
    },
    {
      "entropy": 5.686251878738403,
      "epoch": 4.129118697895991,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003397675350781175,
      "loss": 4.9812,
      "mean_token_accuracy": 0.18956715613603592,
      "num_tokens": 91154683.0,
      "step": 41605
    },
    {
      "entropy": 5.644801139831543,
      "epoch": 4.129614926558158,
      "grad_norm": 1.25,
      "learning_rate": 0.0003397336024600366,
      "loss": 4.9521,
      "mean_token_accuracy": 0.19397831112146377,
      "num_tokens": 91165472.0,
      "step": 41610
    },
    {
      "entropy": 5.720595359802246,
      "epoch": 4.130111155220326,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003396996682367019,
      "loss": 5.0033,
      "mean_token_accuracy": 0.19235837012529372,
      "num_tokens": 91176767.0,
      "step": 41615
    },
    {
      "entropy": 5.6942215919494625,
      "epoch": 4.130607383882493,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003396657324089549,
      "loss": 5.0,
      "mean_token_accuracy": 0.18996072709560394,
      "num_tokens": 91187238.0,
      "step": 41620
    },
    {
      "entropy": 5.672954130172729,
      "epoch": 4.131103612544661,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033963179497763696,
      "loss": 4.925,
      "mean_token_accuracy": 0.1980120614171028,
      "num_tokens": 91198032.0,
      "step": 41625
    },
    {
      "entropy": 5.6566407680511475,
      "epoch": 4.131599841206828,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033959785594358984,
      "loss": 5.0248,
      "mean_token_accuracy": 0.18467349410057068,
      "num_tokens": 91210486.0,
      "step": 41630
    },
    {
      "entropy": 5.733146238327026,
      "epoch": 4.132096069868996,
      "grad_norm": 1.125,
      "learning_rate": 0.00033956391530765506,
      "loss": 5.0809,
      "mean_token_accuracy": 0.1802609920501709,
      "num_tokens": 91223173.0,
      "step": 41635
    },
    {
      "entropy": 5.660583734512329,
      "epoch": 4.132592298531163,
      "grad_norm": 1.125,
      "learning_rate": 0.00033952997307067437,
      "loss": 4.9326,
      "mean_token_accuracy": 0.19904503524303435,
      "num_tokens": 91233894.0,
      "step": 41640
    },
    {
      "entropy": 5.6532721519470215,
      "epoch": 4.133088527193331,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033949602923348933,
      "loss": 4.879,
      "mean_token_accuracy": 0.2015358626842499,
      "num_tokens": 91244942.0,
      "step": 41645
    },
    {
      "entropy": 5.6948981285095215,
      "epoch": 4.133584755855498,
      "grad_norm": 1.25,
      "learning_rate": 0.00033946208379694183,
      "loss": 4.9406,
      "mean_token_accuracy": 0.1949575737118721,
      "num_tokens": 91256446.0,
      "step": 41650
    },
    {
      "entropy": 5.690626430511474,
      "epoch": 4.134080984517666,
      "grad_norm": 1.25,
      "learning_rate": 0.00033942813676187344,
      "loss": 4.9687,
      "mean_token_accuracy": 0.1903382807970047,
      "num_tokens": 91265574.0,
      "step": 41655
    },
    {
      "entropy": 5.597895097732544,
      "epoch": 4.134577213179833,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003393941881291262,
      "loss": 4.8294,
      "mean_token_accuracy": 0.20177998393774033,
      "num_tokens": 91276104.0,
      "step": 41660
    },
    {
      "entropy": 5.680838918685913,
      "epoch": 4.1350734418420005,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003393602378995418,
      "loss": 4.9793,
      "mean_token_accuracy": 0.1997721716761589,
      "num_tokens": 91287193.0,
      "step": 41665
    },
    {
      "entropy": 5.667469882965088,
      "epoch": 4.135569670504168,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003393262860739621,
      "loss": 4.9525,
      "mean_token_accuracy": 0.1902840793132782,
      "num_tokens": 91298573.0,
      "step": 41670
    },
    {
      "entropy": 5.6865133285522464,
      "epoch": 4.1360658991663355,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00033929233265322917,
      "loss": 4.8909,
      "mean_token_accuracy": 0.20011174827814102,
      "num_tokens": 91309785.0,
      "step": 41675
    },
    {
      "entropy": 5.693756103515625,
      "epoch": 4.136562127828503,
      "grad_norm": 1.171875,
      "learning_rate": 0.000339258377638185,
      "loss": 5.0322,
      "mean_token_accuracy": 0.18869142830371857,
      "num_tokens": 91321631.0,
      "step": 41680
    },
    {
      "entropy": 5.68139877319336,
      "epoch": 4.137058356490671,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033922442102967147,
      "loss": 4.9295,
      "mean_token_accuracy": 0.20441845953464508,
      "num_tokens": 91331363.0,
      "step": 41685
    },
    {
      "entropy": 5.717067432403565,
      "epoch": 4.137554585152839,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003391904628285307,
      "loss": 5.0129,
      "mean_token_accuracy": 0.18604979813098907,
      "num_tokens": 91342285.0,
      "step": 41690
    },
    {
      "entropy": 5.730100297927857,
      "epoch": 4.138050813815006,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003391565030356048,
      "loss": 5.0396,
      "mean_token_accuracy": 0.1889445260167122,
      "num_tokens": 91352494.0,
      "step": 41695
    },
    {
      "entropy": 5.630677604675293,
      "epoch": 4.138547042477174,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033912254165173586,
      "loss": 4.8751,
      "mean_token_accuracy": 0.19474526941776277,
      "num_tokens": 91363487.0,
      "step": 41700
    },
    {
      "entropy": 5.6612868309021,
      "epoch": 4.139043271139341,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000339088578677766,
      "loss": 4.9557,
      "mean_token_accuracy": 0.1884537622332573,
      "num_tokens": 91374828.0,
      "step": 41705
    },
    {
      "entropy": 5.572758722305298,
      "epoch": 4.139539499801509,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003390546141145376,
      "loss": 4.8631,
      "mean_token_accuracy": 0.19599268734455108,
      "num_tokens": 91384893.0,
      "step": 41710
    },
    {
      "entropy": 5.667367267608642,
      "epoch": 4.140035728463676,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003390206479628927,
      "loss": 5.0064,
      "mean_token_accuracy": 0.18092669248580934,
      "num_tokens": 91397183.0,
      "step": 41715
    },
    {
      "entropy": 5.666904401779175,
      "epoch": 4.140531957125844,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033898668022367366,
      "loss": 4.9218,
      "mean_token_accuracy": 0.19840321987867354,
      "num_tokens": 91409312.0,
      "step": 41720
    },
    {
      "entropy": 5.683046197891235,
      "epoch": 4.141028185788011,
      "grad_norm": 1.375,
      "learning_rate": 0.0003389527108977229,
      "loss": 4.9743,
      "mean_token_accuracy": 0.1939009979367256,
      "num_tokens": 91419349.0,
      "step": 41725
    },
    {
      "entropy": 5.677097511291504,
      "epoch": 4.141524414450179,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00033891873998588264,
      "loss": 4.9992,
      "mean_token_accuracy": 0.19177380502223967,
      "num_tokens": 91429171.0,
      "step": 41730
    },
    {
      "entropy": 5.6639586925506595,
      "epoch": 4.142020643112346,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033888476748899544,
      "loss": 4.9546,
      "mean_token_accuracy": 0.18888794779777526,
      "num_tokens": 91439045.0,
      "step": 41735
    },
    {
      "entropy": 5.646781635284424,
      "epoch": 4.1425168717745136,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033885079340790364,
      "loss": 4.912,
      "mean_token_accuracy": 0.1979934334754944,
      "num_tokens": 91449143.0,
      "step": 41740
    },
    {
      "entropy": 5.722691202163697,
      "epoch": 4.143013100436681,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033881681774344975,
      "loss": 5.0186,
      "mean_token_accuracy": 0.18344949781894684,
      "num_tokens": 91460056.0,
      "step": 41745
    },
    {
      "entropy": 5.709796524047851,
      "epoch": 4.1435093290988485,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003387828404964763,
      "loss": 4.9975,
      "mean_token_accuracy": 0.18703678250312805,
      "num_tokens": 91471439.0,
      "step": 41750
    },
    {
      "entropy": 5.6562415599823,
      "epoch": 4.144005557761016,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003387488616678259,
      "loss": 4.9577,
      "mean_token_accuracy": 0.19498451799154282,
      "num_tokens": 91482817.0,
      "step": 41755
    },
    {
      "entropy": 5.7429992198944095,
      "epoch": 4.1445017864231835,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033871488125834093,
      "loss": 5.0572,
      "mean_token_accuracy": 0.18584439903497696,
      "num_tokens": 91495383.0,
      "step": 41760
    },
    {
      "entropy": 5.6993263244628904,
      "epoch": 4.144998015085351,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033868089926886435,
      "loss": 4.9341,
      "mean_token_accuracy": 0.19020491242408752,
      "num_tokens": 91505888.0,
      "step": 41765
    },
    {
      "entropy": 5.663408374786377,
      "epoch": 4.145494243747519,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033864691570023857,
      "loss": 4.8762,
      "mean_token_accuracy": 0.20362114459276198,
      "num_tokens": 91516436.0,
      "step": 41770
    },
    {
      "entropy": 5.7150044441223145,
      "epoch": 4.145990472409687,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003386129305533066,
      "loss": 5.0273,
      "mean_token_accuracy": 0.18736574351787566,
      "num_tokens": 91527555.0,
      "step": 41775
    },
    {
      "entropy": 5.617688226699829,
      "epoch": 4.146486701071854,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003385789438289109,
      "loss": 4.8606,
      "mean_token_accuracy": 0.20212152302265168,
      "num_tokens": 91537781.0,
      "step": 41780
    },
    {
      "entropy": 5.711784887313843,
      "epoch": 4.146982929734022,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003385449555278943,
      "loss": 4.9533,
      "mean_token_accuracy": 0.19391578435897827,
      "num_tokens": 91547404.0,
      "step": 41785
    },
    {
      "entropy": 5.65819902420044,
      "epoch": 4.147479158396189,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033851096565109997,
      "loss": 4.903,
      "mean_token_accuracy": 0.20592298209667206,
      "num_tokens": 91557547.0,
      "step": 41790
    },
    {
      "entropy": 5.662837934494019,
      "epoch": 4.147975387058357,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003384769741993704,
      "loss": 5.0294,
      "mean_token_accuracy": 0.1883976250886917,
      "num_tokens": 91569142.0,
      "step": 41795
    },
    {
      "entropy": 5.6688467502594,
      "epoch": 4.148471615720524,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033844298117354865,
      "loss": 4.9032,
      "mean_token_accuracy": 0.19802957475185395,
      "num_tokens": 91581473.0,
      "step": 41800
    },
    {
      "entropy": 5.658625698089599,
      "epoch": 4.148967844382692,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003384089865744777,
      "loss": 4.9905,
      "mean_token_accuracy": 0.18795002698898317,
      "num_tokens": 91592812.0,
      "step": 41805
    },
    {
      "entropy": 5.642909049987793,
      "epoch": 4.149464073044859,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003383749904030005,
      "loss": 4.9844,
      "mean_token_accuracy": 0.18824676275253296,
      "num_tokens": 91603752.0,
      "step": 41810
    },
    {
      "entropy": 5.662720584869385,
      "epoch": 4.149960301707027,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003383409926599601,
      "loss": 5.0074,
      "mean_token_accuracy": 0.19027149379253389,
      "num_tokens": 91616036.0,
      "step": 41815
    },
    {
      "entropy": 5.723036003112793,
      "epoch": 4.150456530369194,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033830699334619964,
      "loss": 5.0678,
      "mean_token_accuracy": 0.18791987895965576,
      "num_tokens": 91628018.0,
      "step": 41820
    },
    {
      "entropy": 5.700382995605469,
      "epoch": 4.1509527590313615,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003382729924625621,
      "loss": 4.9213,
      "mean_token_accuracy": 0.19414899200201036,
      "num_tokens": 91639005.0,
      "step": 41825
    },
    {
      "entropy": 5.676981735229492,
      "epoch": 4.151448987693529,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033823899000989067,
      "loss": 4.9105,
      "mean_token_accuracy": 0.19105453044176102,
      "num_tokens": 91648894.0,
      "step": 41830
    },
    {
      "entropy": 5.618474817276001,
      "epoch": 4.1519452163556965,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033820498598902853,
      "loss": 4.9133,
      "mean_token_accuracy": 0.19697092175483705,
      "num_tokens": 91658525.0,
      "step": 41835
    },
    {
      "entropy": 5.631057548522949,
      "epoch": 4.152441445017864,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033817098040081906,
      "loss": 4.9816,
      "mean_token_accuracy": 0.1925154894590378,
      "num_tokens": 91669537.0,
      "step": 41840
    },
    {
      "entropy": 5.7103136539459225,
      "epoch": 4.152937673680031,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033813697324610524,
      "loss": 4.9321,
      "mean_token_accuracy": 0.19030825048685074,
      "num_tokens": 91679421.0,
      "step": 41845
    },
    {
      "entropy": 5.706923294067383,
      "epoch": 4.153433902342199,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003381029645257306,
      "loss": 4.9162,
      "mean_token_accuracy": 0.19805109798908233,
      "num_tokens": 91689743.0,
      "step": 41850
    },
    {
      "entropy": 5.616489839553833,
      "epoch": 4.153930131004367,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003380689542405385,
      "loss": 4.9239,
      "mean_token_accuracy": 0.19887159317731856,
      "num_tokens": 91700447.0,
      "step": 41855
    },
    {
      "entropy": 5.692416524887085,
      "epoch": 4.154426359666535,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003380349423913721,
      "loss": 4.9741,
      "mean_token_accuracy": 0.19180869311094284,
      "num_tokens": 91713283.0,
      "step": 41860
    },
    {
      "entropy": 5.777812480926514,
      "epoch": 4.154922588328702,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000338000928979075,
      "loss": 5.0162,
      "mean_token_accuracy": 0.1907222956418991,
      "num_tokens": 91724011.0,
      "step": 41865
    },
    {
      "entropy": 5.6984070301055905,
      "epoch": 4.15541881699087,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003379669140044907,
      "loss": 5.0188,
      "mean_token_accuracy": 0.1830128699541092,
      "num_tokens": 91734920.0,
      "step": 41870
    },
    {
      "entropy": 5.645921611785889,
      "epoch": 4.155915045653037,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003379328974684625,
      "loss": 4.962,
      "mean_token_accuracy": 0.19292745292186736,
      "num_tokens": 91746842.0,
      "step": 41875
    },
    {
      "entropy": 5.583125019073487,
      "epoch": 4.156411274315205,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003378988793718342,
      "loss": 4.9024,
      "mean_token_accuracy": 0.19586241245269775,
      "num_tokens": 91758203.0,
      "step": 41880
    },
    {
      "entropy": 5.628285026550293,
      "epoch": 4.156907502977372,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033786485971544906,
      "loss": 4.9407,
      "mean_token_accuracy": 0.1984089121222496,
      "num_tokens": 91768556.0,
      "step": 41885
    },
    {
      "entropy": 5.681899976730347,
      "epoch": 4.15740373163954,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003378308385001509,
      "loss": 5.0128,
      "mean_token_accuracy": 0.18748738169670104,
      "num_tokens": 91779571.0,
      "step": 41890
    },
    {
      "entropy": 5.6931641578674315,
      "epoch": 4.157899960301707,
      "grad_norm": 1.25,
      "learning_rate": 0.0003377968157267834,
      "loss": 4.8862,
      "mean_token_accuracy": 0.19681539982557297,
      "num_tokens": 91789358.0,
      "step": 41895
    },
    {
      "entropy": 5.692246437072754,
      "epoch": 4.1583961889638745,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033776279139619014,
      "loss": 4.9835,
      "mean_token_accuracy": 0.18854549378156663,
      "num_tokens": 91800490.0,
      "step": 41900
    },
    {
      "entropy": 5.69375581741333,
      "epoch": 4.158892417626042,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003377287655092148,
      "loss": 5.018,
      "mean_token_accuracy": 0.19313990622758864,
      "num_tokens": 91810723.0,
      "step": 41905
    },
    {
      "entropy": 5.744571113586426,
      "epoch": 4.1593886462882095,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033769473806670143,
      "loss": 5.0028,
      "mean_token_accuracy": 0.19670204371213912,
      "num_tokens": 91821453.0,
      "step": 41910
    },
    {
      "entropy": 5.701031923294067,
      "epoch": 4.159884874950377,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033766070906949354,
      "loss": 4.998,
      "mean_token_accuracy": 0.19092240929603577,
      "num_tokens": 91831525.0,
      "step": 41915
    },
    {
      "entropy": 5.6978191375732425,
      "epoch": 4.160381103612544,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003376266785184351,
      "loss": 4.8824,
      "mean_token_accuracy": 0.2026518017053604,
      "num_tokens": 91843140.0,
      "step": 41920
    },
    {
      "entropy": 5.685964202880859,
      "epoch": 4.160877332274712,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033759264641437006,
      "loss": 5.029,
      "mean_token_accuracy": 0.1897346630692482,
      "num_tokens": 91855452.0,
      "step": 41925
    },
    {
      "entropy": 5.725081205368042,
      "epoch": 4.161373560936879,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033755861275814214,
      "loss": 5.028,
      "mean_token_accuracy": 0.1857282266020775,
      "num_tokens": 91865942.0,
      "step": 41930
    },
    {
      "entropy": 5.672060823440551,
      "epoch": 4.161869789599047,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003375245775505955,
      "loss": 4.9642,
      "mean_token_accuracy": 0.19856845885515212,
      "num_tokens": 91878065.0,
      "step": 41935
    },
    {
      "entropy": 5.699663829803467,
      "epoch": 4.162366018261215,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000337490540792574,
      "loss": 4.9824,
      "mean_token_accuracy": 0.1941913291811943,
      "num_tokens": 91889939.0,
      "step": 41940
    },
    {
      "entropy": 5.712571430206299,
      "epoch": 4.162862246923383,
      "grad_norm": 1.25,
      "learning_rate": 0.00033745650248492187,
      "loss": 4.9983,
      "mean_token_accuracy": 0.20109695792198182,
      "num_tokens": 91901740.0,
      "step": 41945
    },
    {
      "entropy": 5.729554891586304,
      "epoch": 4.16335847558555,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003374224626284829,
      "loss": 5.0238,
      "mean_token_accuracy": 0.1932003065943718,
      "num_tokens": 91912779.0,
      "step": 41950
    },
    {
      "entropy": 5.7000542163848875,
      "epoch": 4.163854704247718,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033738842122410154,
      "loss": 4.97,
      "mean_token_accuracy": 0.19675754606723786,
      "num_tokens": 91924077.0,
      "step": 41955
    },
    {
      "entropy": 5.68048357963562,
      "epoch": 4.164350932909885,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033735437827262166,
      "loss": 4.8739,
      "mean_token_accuracy": 0.2040986180305481,
      "num_tokens": 91934281.0,
      "step": 41960
    },
    {
      "entropy": 5.651430702209472,
      "epoch": 4.164847161572053,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003373203337748875,
      "loss": 4.9098,
      "mean_token_accuracy": 0.189934241771698,
      "num_tokens": 91944015.0,
      "step": 41965
    },
    {
      "entropy": 5.5791832447052006,
      "epoch": 4.16534339023422,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033728628773174344,
      "loss": 4.9311,
      "mean_token_accuracy": 0.19700984358787538,
      "num_tokens": 91954545.0,
      "step": 41970
    },
    {
      "entropy": 5.641393423080444,
      "epoch": 4.1658396188963875,
      "grad_norm": 1.125,
      "learning_rate": 0.00033725224014403374,
      "loss": 4.9181,
      "mean_token_accuracy": 0.1953295186161995,
      "num_tokens": 91966192.0,
      "step": 41975
    },
    {
      "entropy": 5.646735286712646,
      "epoch": 4.166335847558555,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033721819101260246,
      "loss": 4.9184,
      "mean_token_accuracy": 0.19434452652931214,
      "num_tokens": 91978549.0,
      "step": 41980
    },
    {
      "entropy": 5.678297328948974,
      "epoch": 4.1668320762207225,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003371841403382942,
      "loss": 4.9794,
      "mean_token_accuracy": 0.19422127455472946,
      "num_tokens": 91989753.0,
      "step": 41985
    },
    {
      "entropy": 5.679860067367554,
      "epoch": 4.16732830488289,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033715008812195324,
      "loss": 4.9469,
      "mean_token_accuracy": 0.1932293176651001,
      "num_tokens": 91999923.0,
      "step": 41990
    },
    {
      "entropy": 5.746028757095337,
      "epoch": 4.167824533545057,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000337116034364424,
      "loss": 4.9778,
      "mean_token_accuracy": 0.18569737523794175,
      "num_tokens": 92010780.0,
      "step": 41995
    },
    {
      "entropy": 5.6327001571655275,
      "epoch": 4.168320762207225,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00033708197906655096,
      "loss": 4.9896,
      "mean_token_accuracy": 0.1924167588353157,
      "num_tokens": 92020820.0,
      "step": 42000
    },
    {
      "epoch": 4.168320762207225,
      "eval_entropy": 5.413326036961239,
      "eval_loss": 5.102902412414551,
      "eval_mean_token_accuracy": 0.19153556072795497,
      "eval_num_tokens": 92020820.0,
      "eval_runtime": 20.4001,
      "eval_samples_per_second": 1650.43,
      "eval_steps_per_second": 206.322,
      "step": 42000
    },
    {
      "entropy": 5.69464693069458,
      "epoch": 4.168816990869392,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033704792222917864,
      "loss": 4.9771,
      "mean_token_accuracy": 0.18476190119981767,
      "num_tokens": 92031889.0,
      "step": 42005
    },
    {
      "entropy": 5.638769960403442,
      "epoch": 4.16931321953156,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033701386385315155,
      "loss": 4.9146,
      "mean_token_accuracy": 0.1957152545452118,
      "num_tokens": 92044416.0,
      "step": 42010
    },
    {
      "entropy": 5.650963973999024,
      "epoch": 4.169809448193727,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003369798039393142,
      "loss": 4.9357,
      "mean_token_accuracy": 0.19626383930444719,
      "num_tokens": 92055748.0,
      "step": 42015
    },
    {
      "entropy": 5.691722297668457,
      "epoch": 4.170305676855895,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033694574248851133,
      "loss": 4.9491,
      "mean_token_accuracy": 0.1908090591430664,
      "num_tokens": 92066251.0,
      "step": 42020
    },
    {
      "entropy": 5.668644714355469,
      "epoch": 4.170801905518063,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003369116795015875,
      "loss": 4.9517,
      "mean_token_accuracy": 0.19319453835487366,
      "num_tokens": 92077651.0,
      "step": 42025
    },
    {
      "entropy": 5.741732025146485,
      "epoch": 4.171298134180231,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003368776149793875,
      "loss": 5.0902,
      "mean_token_accuracy": 0.18774141371250153,
      "num_tokens": 92088849.0,
      "step": 42030
    },
    {
      "entropy": 5.684122085571289,
      "epoch": 4.171794362842398,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003368435489227558,
      "loss": 4.9145,
      "mean_token_accuracy": 0.19547101855278015,
      "num_tokens": 92099638.0,
      "step": 42035
    },
    {
      "entropy": 5.662723112106323,
      "epoch": 4.172290591504566,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033680948133253756,
      "loss": 4.893,
      "mean_token_accuracy": 0.20299158841371537,
      "num_tokens": 92109376.0,
      "step": 42040
    },
    {
      "entropy": 5.569222545623779,
      "epoch": 4.172786820166733,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003367754122095772,
      "loss": 4.8611,
      "mean_token_accuracy": 0.19665130078792573,
      "num_tokens": 92119721.0,
      "step": 42045
    },
    {
      "entropy": 5.704179048538208,
      "epoch": 4.1732830488289006,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003367413415547199,
      "loss": 5.0196,
      "mean_token_accuracy": 0.1909657523036003,
      "num_tokens": 92131706.0,
      "step": 42050
    },
    {
      "entropy": 5.683458948135376,
      "epoch": 4.173779277491068,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033670726936881024,
      "loss": 4.8346,
      "mean_token_accuracy": 0.20216701477766036,
      "num_tokens": 92140866.0,
      "step": 42055
    },
    {
      "entropy": 5.678779602050781,
      "epoch": 4.1742755061532355,
      "grad_norm": 1.375,
      "learning_rate": 0.0003366731956526933,
      "loss": 5.0599,
      "mean_token_accuracy": 0.19018671959638594,
      "num_tokens": 92152761.0,
      "step": 42060
    },
    {
      "entropy": 5.600421619415283,
      "epoch": 4.174771734815403,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033663912040721405,
      "loss": 4.8493,
      "mean_token_accuracy": 0.19970447868108748,
      "num_tokens": 92164762.0,
      "step": 42065
    },
    {
      "entropy": 5.702650499343872,
      "epoch": 4.1752679634775705,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033660504363321745,
      "loss": 4.9668,
      "mean_token_accuracy": 0.18608393371105195,
      "num_tokens": 92175070.0,
      "step": 42070
    },
    {
      "entropy": 5.661301708221435,
      "epoch": 4.175764192139738,
      "grad_norm": 1.421875,
      "learning_rate": 0.00033657096533154843,
      "loss": 4.9728,
      "mean_token_accuracy": 0.19458572268486024,
      "num_tokens": 92185023.0,
      "step": 42075
    },
    {
      "entropy": 5.646389961242676,
      "epoch": 4.176260420801905,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033653688550305214,
      "loss": 4.963,
      "mean_token_accuracy": 0.1890743836760521,
      "num_tokens": 92194988.0,
      "step": 42080
    },
    {
      "entropy": 5.725997686386108,
      "epoch": 4.176756649464073,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003365028041485737,
      "loss": 4.9886,
      "mean_token_accuracy": 0.1942492201924324,
      "num_tokens": 92205380.0,
      "step": 42085
    },
    {
      "entropy": 5.655727672576904,
      "epoch": 4.17725287812624,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033646872126895833,
      "loss": 4.868,
      "mean_token_accuracy": 0.202152119576931,
      "num_tokens": 92215604.0,
      "step": 42090
    },
    {
      "entropy": 5.6401269912719725,
      "epoch": 4.177749106788408,
      "grad_norm": 1.171875,
      "learning_rate": 0.000336434636865051,
      "loss": 4.9389,
      "mean_token_accuracy": 0.19062060862779617,
      "num_tokens": 92227259.0,
      "step": 42095
    },
    {
      "entropy": 5.69286470413208,
      "epoch": 4.178245335450575,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0003364005509376971,
      "loss": 4.9562,
      "mean_token_accuracy": 0.19815634489059447,
      "num_tokens": 92239720.0,
      "step": 42100
    },
    {
      "entropy": 5.674062919616699,
      "epoch": 4.178741564112743,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003363664634877419,
      "loss": 4.9366,
      "mean_token_accuracy": 0.19841019809246063,
      "num_tokens": 92251318.0,
      "step": 42105
    },
    {
      "entropy": 5.655763149261475,
      "epoch": 4.17923779277491,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003363323745160306,
      "loss": 4.9777,
      "mean_token_accuracy": 0.18878892064094543,
      "num_tokens": 92261869.0,
      "step": 42110
    },
    {
      "entropy": 5.712061071395874,
      "epoch": 4.179734021437079,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00033629828402340864,
      "loss": 5.0493,
      "mean_token_accuracy": 0.18310045301914216,
      "num_tokens": 92271049.0,
      "step": 42115
    },
    {
      "entropy": 5.658508396148681,
      "epoch": 4.180230250099246,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033626419201072125,
      "loss": 4.918,
      "mean_token_accuracy": 0.19773106426000595,
      "num_tokens": 92282224.0,
      "step": 42120
    },
    {
      "entropy": 5.650800704956055,
      "epoch": 4.180726478761414,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003362300984788139,
      "loss": 4.9409,
      "mean_token_accuracy": 0.19586914777755737,
      "num_tokens": 92293612.0,
      "step": 42125
    },
    {
      "entropy": 5.618044233322143,
      "epoch": 4.181222707423581,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000336196003428532,
      "loss": 4.8725,
      "mean_token_accuracy": 0.1976321518421173,
      "num_tokens": 92304195.0,
      "step": 42130
    },
    {
      "entropy": 5.607100248336792,
      "epoch": 4.1817189360857485,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003361619068607212,
      "loss": 4.8542,
      "mean_token_accuracy": 0.2015874832868576,
      "num_tokens": 92314249.0,
      "step": 42135
    },
    {
      "entropy": 5.671411514282227,
      "epoch": 4.182215164747916,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003361278087762268,
      "loss": 4.9772,
      "mean_token_accuracy": 0.1874294623732567,
      "num_tokens": 92325067.0,
      "step": 42140
    },
    {
      "entropy": 5.642491388320923,
      "epoch": 4.1827113934100835,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003360937091758945,
      "loss": 4.9314,
      "mean_token_accuracy": 0.1953207239508629,
      "num_tokens": 92336191.0,
      "step": 42145
    },
    {
      "entropy": 5.662788057327271,
      "epoch": 4.183207622072251,
      "grad_norm": 1.125,
      "learning_rate": 0.0003360596080605699,
      "loss": 4.9245,
      "mean_token_accuracy": 0.1935669884085655,
      "num_tokens": 92347161.0,
      "step": 42150
    },
    {
      "entropy": 5.666665983200073,
      "epoch": 4.183703850734418,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033602550543109855,
      "loss": 4.9602,
      "mean_token_accuracy": 0.1863080233335495,
      "num_tokens": 92358750.0,
      "step": 42155
    },
    {
      "entropy": 5.660955190658569,
      "epoch": 4.184200079396586,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033599140128832614,
      "loss": 4.9134,
      "mean_token_accuracy": 0.19158583283424377,
      "num_tokens": 92370244.0,
      "step": 42160
    },
    {
      "entropy": 5.699640226364136,
      "epoch": 4.184696308058753,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033595729563309847,
      "loss": 4.9629,
      "mean_token_accuracy": 0.19331496506929396,
      "num_tokens": 92381011.0,
      "step": 42165
    },
    {
      "entropy": 5.64087028503418,
      "epoch": 4.185192536720921,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003359231884662611,
      "loss": 4.8824,
      "mean_token_accuracy": 0.1985494986176491,
      "num_tokens": 92391710.0,
      "step": 42170
    },
    {
      "entropy": 5.57531590461731,
      "epoch": 4.185688765383088,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033588907978866013,
      "loss": 4.8061,
      "mean_token_accuracy": 0.20802656859159468,
      "num_tokens": 92402261.0,
      "step": 42175
    },
    {
      "entropy": 5.6663671970367435,
      "epoch": 4.186184994045256,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00033585496960114104,
      "loss": 4.9613,
      "mean_token_accuracy": 0.18947280049324036,
      "num_tokens": 92413136.0,
      "step": 42180
    },
    {
      "entropy": 5.6930615425109865,
      "epoch": 4.186681222707423,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033582085790454983,
      "loss": 4.9952,
      "mean_token_accuracy": 0.18831811100244522,
      "num_tokens": 92423031.0,
      "step": 42185
    },
    {
      "entropy": 5.706475305557251,
      "epoch": 4.187177451369591,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003357867446997324,
      "loss": 4.9153,
      "mean_token_accuracy": 0.1941270798444748,
      "num_tokens": 92434802.0,
      "step": 42190
    },
    {
      "entropy": 5.660243129730224,
      "epoch": 4.187673680031758,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003357526299875348,
      "loss": 4.9567,
      "mean_token_accuracy": 0.19273675680160524,
      "num_tokens": 92445925.0,
      "step": 42195
    },
    {
      "entropy": 5.6556566715240475,
      "epoch": 4.188169908693927,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033571851376880277,
      "loss": 4.9246,
      "mean_token_accuracy": 0.1970999851822853,
      "num_tokens": 92457967.0,
      "step": 42200
    },
    {
      "entropy": 5.663995361328125,
      "epoch": 4.188666137356094,
      "grad_norm": 1.25,
      "learning_rate": 0.00033568439604438245,
      "loss": 4.9313,
      "mean_token_accuracy": 0.1866244062781334,
      "num_tokens": 92468194.0,
      "step": 42205
    },
    {
      "entropy": 5.67650203704834,
      "epoch": 4.1891623660182615,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003356502768151199,
      "loss": 4.9198,
      "mean_token_accuracy": 0.18824989050626756,
      "num_tokens": 92478271.0,
      "step": 42210
    },
    {
      "entropy": 5.720641183853149,
      "epoch": 4.189658594680429,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003356161560818612,
      "loss": 4.9775,
      "mean_token_accuracy": 0.19389821141958236,
      "num_tokens": 92489383.0,
      "step": 42215
    },
    {
      "entropy": 5.684930992126465,
      "epoch": 4.1901548233425965,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003355820338454524,
      "loss": 4.9799,
      "mean_token_accuracy": 0.19554096311330796,
      "num_tokens": 92499864.0,
      "step": 42220
    },
    {
      "entropy": 5.648079061508179,
      "epoch": 4.190651052004764,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003355479101067397,
      "loss": 4.9287,
      "mean_token_accuracy": 0.19036811739206314,
      "num_tokens": 92510065.0,
      "step": 42225
    },
    {
      "entropy": 5.602301931381225,
      "epoch": 4.191147280666931,
      "grad_norm": 1.25,
      "learning_rate": 0.0003355137848665693,
      "loss": 4.874,
      "mean_token_accuracy": 0.19302008152008057,
      "num_tokens": 92520245.0,
      "step": 42230
    },
    {
      "entropy": 5.656852722167969,
      "epoch": 4.191643509329099,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003354796581257875,
      "loss": 5.0107,
      "mean_token_accuracy": 0.18144024759531022,
      "num_tokens": 92530934.0,
      "step": 42235
    },
    {
      "entropy": 5.729251432418823,
      "epoch": 4.192139737991266,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003354455298852404,
      "loss": 5.0181,
      "mean_token_accuracy": 0.1939210444688797,
      "num_tokens": 92542311.0,
      "step": 42240
    },
    {
      "entropy": 5.616614103317261,
      "epoch": 4.192635966653434,
      "grad_norm": 1.125,
      "learning_rate": 0.0003354114001457744,
      "loss": 4.8023,
      "mean_token_accuracy": 0.20508115440607072,
      "num_tokens": 92552811.0,
      "step": 42245
    },
    {
      "entropy": 5.6686732292175295,
      "epoch": 4.193132195315601,
      "grad_norm": 1.3125,
      "learning_rate": 0.000335377268908236,
      "loss": 5.0143,
      "mean_token_accuracy": 0.18995312005281448,
      "num_tokens": 92563476.0,
      "step": 42250
    },
    {
      "entropy": 5.5954584121704105,
      "epoch": 4.193628423977769,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003353431361734714,
      "loss": 4.9209,
      "mean_token_accuracy": 0.19686306715011598,
      "num_tokens": 92573188.0,
      "step": 42255
    },
    {
      "entropy": 5.679256248474121,
      "epoch": 4.194124652639936,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033530900194232694,
      "loss": 4.9792,
      "mean_token_accuracy": 0.19305987358093263,
      "num_tokens": 92584592.0,
      "step": 42260
    },
    {
      "entropy": 5.6985289573669435,
      "epoch": 4.194620881302104,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033527486621564917,
      "loss": 5.0116,
      "mean_token_accuracy": 0.18564219027757645,
      "num_tokens": 92595024.0,
      "step": 42265
    },
    {
      "entropy": 5.642615938186646,
      "epoch": 4.195117109964271,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003352407289942847,
      "loss": 4.9034,
      "mean_token_accuracy": 0.19541548639535905,
      "num_tokens": 92605373.0,
      "step": 42270
    },
    {
      "entropy": 5.67604923248291,
      "epoch": 4.195613338626439,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003352065902790799,
      "loss": 4.9405,
      "mean_token_accuracy": 0.1980057790875435,
      "num_tokens": 92616729.0,
      "step": 42275
    },
    {
      "entropy": 5.686191511154175,
      "epoch": 4.196109567288606,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003351724500708815,
      "loss": 4.8689,
      "mean_token_accuracy": 0.19912203550338745,
      "num_tokens": 92626943.0,
      "step": 42280
    },
    {
      "entropy": 5.630735063552857,
      "epoch": 4.1966057959507745,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033513830837053583,
      "loss": 4.8802,
      "mean_token_accuracy": 0.19334673136472702,
      "num_tokens": 92637325.0,
      "step": 42285
    },
    {
      "entropy": 5.707675504684448,
      "epoch": 4.197102024612942,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033510416517888985,
      "loss": 5.0839,
      "mean_token_accuracy": 0.18250282257795333,
      "num_tokens": 92648716.0,
      "step": 42290
    },
    {
      "entropy": 5.635070371627807,
      "epoch": 4.1975982532751095,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033507002049679,
      "loss": 4.9055,
      "mean_token_accuracy": 0.19665065109729768,
      "num_tokens": 92660558.0,
      "step": 42295
    },
    {
      "entropy": 5.676728677749634,
      "epoch": 4.198094481937277,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003350358743250831,
      "loss": 4.9073,
      "mean_token_accuracy": 0.1949401780962944,
      "num_tokens": 92670335.0,
      "step": 42300
    },
    {
      "entropy": 5.721124172210693,
      "epoch": 4.198590710599444,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003350017266646159,
      "loss": 4.9954,
      "mean_token_accuracy": 0.1881068617105484,
      "num_tokens": 92681221.0,
      "step": 42305
    },
    {
      "entropy": 5.6349317073822025,
      "epoch": 4.199086939261612,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003349675775162351,
      "loss": 4.9335,
      "mean_token_accuracy": 0.18964199721813202,
      "num_tokens": 92692963.0,
      "step": 42310
    },
    {
      "entropy": 5.6444135189056395,
      "epoch": 4.199583167923779,
      "grad_norm": 1.6875,
      "learning_rate": 0.00033493342688078765,
      "loss": 4.8747,
      "mean_token_accuracy": 0.20866411924362183,
      "num_tokens": 92704346.0,
      "step": 42315
    },
    {
      "entropy": 5.7194348812103275,
      "epoch": 4.200079396585947,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003348992747591204,
      "loss": 5.0113,
      "mean_token_accuracy": 0.19085330814123153,
      "num_tokens": 92715063.0,
      "step": 42320
    },
    {
      "entropy": 5.700718069076538,
      "epoch": 4.200575625248114,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003348651211520802,
      "loss": 4.9881,
      "mean_token_accuracy": 0.19522142112255098,
      "num_tokens": 92726376.0,
      "step": 42325
    },
    {
      "entropy": 5.635822677612305,
      "epoch": 4.201071853910282,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003348309660605139,
      "loss": 4.8801,
      "mean_token_accuracy": 0.19716914594173432,
      "num_tokens": 92737484.0,
      "step": 42330
    },
    {
      "entropy": 5.6316609382629395,
      "epoch": 4.201568082572449,
      "grad_norm": 1.25,
      "learning_rate": 0.0003347968094852687,
      "loss": 4.9574,
      "mean_token_accuracy": 0.19542352110147476,
      "num_tokens": 92747293.0,
      "step": 42335
    },
    {
      "entropy": 5.6670918464660645,
      "epoch": 4.202064311234617,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003347626514271914,
      "loss": 4.9492,
      "mean_token_accuracy": 0.19496062248945237,
      "num_tokens": 92757385.0,
      "step": 42340
    },
    {
      "entropy": 5.677776575088501,
      "epoch": 4.202560539896784,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003347284918871291,
      "loss": 4.9671,
      "mean_token_accuracy": 0.19307063817977904,
      "num_tokens": 92768366.0,
      "step": 42345
    },
    {
      "entropy": 5.699206972122193,
      "epoch": 4.203056768558952,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003346943308659289,
      "loss": 5.0104,
      "mean_token_accuracy": 0.18314189016819,
      "num_tokens": 92779864.0,
      "step": 42350
    },
    {
      "entropy": 5.633301258087158,
      "epoch": 4.203552997221119,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000334660168364438,
      "loss": 4.8991,
      "mean_token_accuracy": 0.20251624137163163,
      "num_tokens": 92790556.0,
      "step": 42355
    },
    {
      "entropy": 5.66216254234314,
      "epoch": 4.204049225883287,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033462600438350336,
      "loss": 4.9184,
      "mean_token_accuracy": 0.19871649742126465,
      "num_tokens": 92802078.0,
      "step": 42360
    },
    {
      "entropy": 5.599096202850342,
      "epoch": 4.204545454545454,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003345918389239724,
      "loss": 4.8866,
      "mean_token_accuracy": 0.19707197695970535,
      "num_tokens": 92812556.0,
      "step": 42365
    },
    {
      "entropy": 5.68507719039917,
      "epoch": 4.205041683207622,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003345576719866923,
      "loss": 4.9419,
      "mean_token_accuracy": 0.19696239680051802,
      "num_tokens": 92823157.0,
      "step": 42370
    },
    {
      "entropy": 5.661397647857666,
      "epoch": 4.20553791186979,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003345235035725101,
      "loss": 4.8832,
      "mean_token_accuracy": 0.19887627959251403,
      "num_tokens": 92832534.0,
      "step": 42375
    },
    {
      "entropy": 5.643929719924927,
      "epoch": 4.2060341405319575,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003344893336822734,
      "loss": 4.9738,
      "mean_token_accuracy": 0.19364141076803207,
      "num_tokens": 92842790.0,
      "step": 42380
    },
    {
      "entropy": 5.615021276473999,
      "epoch": 4.206530369194125,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033445516231682935,
      "loss": 4.9092,
      "mean_token_accuracy": 0.19620055705308914,
      "num_tokens": 92853681.0,
      "step": 42385
    },
    {
      "entropy": 5.639362239837647,
      "epoch": 4.207026597856292,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003344209894770254,
      "loss": 4.9325,
      "mean_token_accuracy": 0.19150563180446625,
      "num_tokens": 92864634.0,
      "step": 42390
    },
    {
      "entropy": 5.590328550338745,
      "epoch": 4.20752282651846,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000334386815163709,
      "loss": 4.8332,
      "mean_token_accuracy": 0.20135397017002105,
      "num_tokens": 92875678.0,
      "step": 42395
    },
    {
      "entropy": 5.6027295112609865,
      "epoch": 4.208019055180627,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003343526393777275,
      "loss": 4.9613,
      "mean_token_accuracy": 0.19029794186353682,
      "num_tokens": 92886019.0,
      "step": 42400
    },
    {
      "entropy": 5.702456140518189,
      "epoch": 4.208515283842795,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033431846211992843,
      "loss": 4.9672,
      "mean_token_accuracy": 0.19647571742534636,
      "num_tokens": 92896854.0,
      "step": 42405
    },
    {
      "entropy": 5.689942121505737,
      "epoch": 4.209011512504962,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033428428339115935,
      "loss": 4.9712,
      "mean_token_accuracy": 0.1921294003725052,
      "num_tokens": 92907786.0,
      "step": 42410
    },
    {
      "entropy": 5.697843217849732,
      "epoch": 4.20950774116713,
      "grad_norm": 1.25,
      "learning_rate": 0.0003342501031922678,
      "loss": 4.9321,
      "mean_token_accuracy": 0.1937445729970932,
      "num_tokens": 92917929.0,
      "step": 42415
    },
    {
      "entropy": 5.655327415466308,
      "epoch": 4.210003969829297,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003342159215241013,
      "loss": 4.9243,
      "mean_token_accuracy": 0.19663559049367904,
      "num_tokens": 92929349.0,
      "step": 42420
    },
    {
      "entropy": 5.695180034637451,
      "epoch": 4.210500198491465,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003341817383875075,
      "loss": 4.9568,
      "mean_token_accuracy": 0.20206589102745057,
      "num_tokens": 92939767.0,
      "step": 42425
    },
    {
      "entropy": 5.680021286010742,
      "epoch": 4.210996427153632,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033414755378333426,
      "loss": 4.9036,
      "mean_token_accuracy": 0.19461947977542876,
      "num_tokens": 92950362.0,
      "step": 42430
    },
    {
      "entropy": 5.685160350799561,
      "epoch": 4.2114926558158,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033411336771242906,
      "loss": 4.9908,
      "mean_token_accuracy": 0.19184131622314454,
      "num_tokens": 92961975.0,
      "step": 42435
    },
    {
      "entropy": 5.665349340438842,
      "epoch": 4.211988884477967,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033407918017563975,
      "loss": 4.9435,
      "mean_token_accuracy": 0.19307685792446136,
      "num_tokens": 92975803.0,
      "step": 42440
    },
    {
      "entropy": 5.687691020965576,
      "epoch": 4.212485113140135,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003340449911738141,
      "loss": 4.96,
      "mean_token_accuracy": 0.19435350894927977,
      "num_tokens": 92987391.0,
      "step": 42445
    },
    {
      "entropy": 5.68149471282959,
      "epoch": 4.212981341802302,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003340108007077999,
      "loss": 4.9921,
      "mean_token_accuracy": 0.18973210901021959,
      "num_tokens": 92999050.0,
      "step": 42450
    },
    {
      "entropy": 5.685816955566406,
      "epoch": 4.21347757046447,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033397660877844485,
      "loss": 4.9629,
      "mean_token_accuracy": 0.18914273232221604,
      "num_tokens": 93011221.0,
      "step": 42455
    },
    {
      "entropy": 5.739120197296143,
      "epoch": 4.213973799126638,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003339424153865972,
      "loss": 4.9827,
      "mean_token_accuracy": 0.19630816876888274,
      "num_tokens": 93022249.0,
      "step": 42460
    },
    {
      "entropy": 5.69448823928833,
      "epoch": 4.214470027788805,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003339082205331045,
      "loss": 4.9463,
      "mean_token_accuracy": 0.18295053392648697,
      "num_tokens": 93033231.0,
      "step": 42465
    },
    {
      "entropy": 5.594053459167481,
      "epoch": 4.214966256450973,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003338740242188149,
      "loss": 4.8788,
      "mean_token_accuracy": 0.20529643297195435,
      "num_tokens": 93044049.0,
      "step": 42470
    },
    {
      "entropy": 5.642512035369873,
      "epoch": 4.21546248511314,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033383982644457646,
      "loss": 4.9631,
      "mean_token_accuracy": 0.19286031126976014,
      "num_tokens": 93055187.0,
      "step": 42475
    },
    {
      "entropy": 5.710315752029419,
      "epoch": 4.215958713775308,
      "grad_norm": 1.1875,
      "learning_rate": 0.000333805627211237,
      "loss": 4.9996,
      "mean_token_accuracy": 0.18902584612369538,
      "num_tokens": 93065822.0,
      "step": 42480
    },
    {
      "entropy": 5.7990086555480955,
      "epoch": 4.216454942437475,
      "grad_norm": 1.125,
      "learning_rate": 0.00033377142651964476,
      "loss": 5.1063,
      "mean_token_accuracy": 0.1899325430393219,
      "num_tokens": 93077424.0,
      "step": 42485
    },
    {
      "entropy": 5.61691837310791,
      "epoch": 4.216951171099643,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033373722437064773,
      "loss": 4.9209,
      "mean_token_accuracy": 0.192516465485096,
      "num_tokens": 93088256.0,
      "step": 42490
    },
    {
      "entropy": 5.702987575531006,
      "epoch": 4.21744739976181,
      "grad_norm": 1.25,
      "learning_rate": 0.0003337030207650941,
      "loss": 4.9945,
      "mean_token_accuracy": 0.19344065636396407,
      "num_tokens": 93098111.0,
      "step": 42495
    },
    {
      "entropy": 5.657920074462891,
      "epoch": 4.217943628423978,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033366881570383213,
      "loss": 4.8723,
      "mean_token_accuracy": 0.20010487139225006,
      "num_tokens": 93109066.0,
      "step": 42500
    },
    {
      "entropy": 5.743161392211914,
      "epoch": 4.218439857086145,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003336346091877099,
      "loss": 5.0262,
      "mean_token_accuracy": 0.1913931578397751,
      "num_tokens": 93119947.0,
      "step": 42505
    },
    {
      "entropy": 5.640786266326904,
      "epoch": 4.218936085748313,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003336004012175757,
      "loss": 4.9282,
      "mean_token_accuracy": 0.19145079851150512,
      "num_tokens": 93132068.0,
      "step": 42510
    },
    {
      "entropy": 5.713075637817383,
      "epoch": 4.21943231441048,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003335661917942778,
      "loss": 4.9614,
      "mean_token_accuracy": 0.19292321801185608,
      "num_tokens": 93142897.0,
      "step": 42515
    },
    {
      "entropy": 5.635671663284302,
      "epoch": 4.219928543072648,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003335319809186646,
      "loss": 4.8918,
      "mean_token_accuracy": 0.2005203917622566,
      "num_tokens": 93154133.0,
      "step": 42520
    },
    {
      "entropy": 5.657770872116089,
      "epoch": 4.220424771734815,
      "grad_norm": 1.25,
      "learning_rate": 0.0003334977685915843,
      "loss": 4.9828,
      "mean_token_accuracy": 0.19232072979211806,
      "num_tokens": 93165818.0,
      "step": 42525
    },
    {
      "entropy": 5.661765193939209,
      "epoch": 4.220921000396983,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003334635548138856,
      "loss": 4.9812,
      "mean_token_accuracy": 0.1936617061495781,
      "num_tokens": 93177653.0,
      "step": 42530
    },
    {
      "entropy": 5.655620050430298,
      "epoch": 4.22141722905915,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003334293395864165,
      "loss": 4.8833,
      "mean_token_accuracy": 0.19766883552074432,
      "num_tokens": 93188713.0,
      "step": 42535
    },
    {
      "entropy": 5.720939350128174,
      "epoch": 4.2219134577213175,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003333951229100257,
      "loss": 5.0533,
      "mean_token_accuracy": 0.17987369745969772,
      "num_tokens": 93199249.0,
      "step": 42540
    },
    {
      "entropy": 5.676945066452026,
      "epoch": 4.222409686383486,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003333609047855618,
      "loss": 5.0001,
      "mean_token_accuracy": 0.19709109216928483,
      "num_tokens": 93209667.0,
      "step": 42545
    },
    {
      "entropy": 5.622435617446899,
      "epoch": 4.222905915045653,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033332668521387316,
      "loss": 4.9484,
      "mean_token_accuracy": 0.18960765451192857,
      "num_tokens": 93220391.0,
      "step": 42550
    },
    {
      "entropy": 5.693364524841309,
      "epoch": 4.223402143707821,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033329246419580837,
      "loss": 4.9853,
      "mean_token_accuracy": 0.1950833320617676,
      "num_tokens": 93232155.0,
      "step": 42555
    },
    {
      "entropy": 5.6661458015441895,
      "epoch": 4.223898372369988,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003332582417322162,
      "loss": 4.8552,
      "mean_token_accuracy": 0.19503723382949828,
      "num_tokens": 93240695.0,
      "step": 42560
    },
    {
      "entropy": 5.667154932022095,
      "epoch": 4.224394601032156,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003332240178239451,
      "loss": 4.9308,
      "mean_token_accuracy": 0.1989577516913414,
      "num_tokens": 93253609.0,
      "step": 42565
    },
    {
      "entropy": 5.680525016784668,
      "epoch": 4.224890829694323,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003331897924718438,
      "loss": 4.9732,
      "mean_token_accuracy": 0.19329783469438552,
      "num_tokens": 93264853.0,
      "step": 42570
    },
    {
      "entropy": 5.660511827468872,
      "epoch": 4.225387058356491,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033315556567676104,
      "loss": 4.9403,
      "mean_token_accuracy": 0.19659728854894637,
      "num_tokens": 93275803.0,
      "step": 42575
    },
    {
      "entropy": 5.634373617172241,
      "epoch": 4.225883287018658,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003331213374395455,
      "loss": 4.939,
      "mean_token_accuracy": 0.18886950761079788,
      "num_tokens": 93286425.0,
      "step": 42580
    },
    {
      "entropy": 5.670297336578369,
      "epoch": 4.226379515680826,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033308710776104616,
      "loss": 4.9458,
      "mean_token_accuracy": 0.18963258415460588,
      "num_tokens": 93297201.0,
      "step": 42585
    },
    {
      "entropy": 5.677296447753906,
      "epoch": 4.226875744342993,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003330528766421116,
      "loss": 4.9489,
      "mean_token_accuracy": 0.1978616863489151,
      "num_tokens": 93310607.0,
      "step": 42590
    },
    {
      "entropy": 5.672346496582032,
      "epoch": 4.227371973005161,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033301864408359085,
      "loss": 4.9131,
      "mean_token_accuracy": 0.1975972294807434,
      "num_tokens": 93321869.0,
      "step": 42595
    },
    {
      "entropy": 5.614494037628174,
      "epoch": 4.227868201667328,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033298441008633277,
      "loss": 4.8632,
      "mean_token_accuracy": 0.20309093296527864,
      "num_tokens": 93333330.0,
      "step": 42600
    },
    {
      "entropy": 5.642236375808716,
      "epoch": 4.228364430329496,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033295017465118624,
      "loss": 4.9704,
      "mean_token_accuracy": 0.18900943249464036,
      "num_tokens": 93344712.0,
      "step": 42605
    },
    {
      "entropy": 5.704089546203614,
      "epoch": 4.228860658991663,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033291593777900027,
      "loss": 5.0108,
      "mean_token_accuracy": 0.18997039198875426,
      "num_tokens": 93356113.0,
      "step": 42610
    },
    {
      "entropy": 5.672695779800415,
      "epoch": 4.2293568876538306,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003328816994706238,
      "loss": 4.8914,
      "mean_token_accuracy": 0.19496446549892427,
      "num_tokens": 93367002.0,
      "step": 42615
    },
    {
      "entropy": 5.660648012161255,
      "epoch": 4.229853116315998,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033284745972690595,
      "loss": 4.9764,
      "mean_token_accuracy": 0.19088294506072997,
      "num_tokens": 93378278.0,
      "step": 42620
    },
    {
      "entropy": 5.605210971832276,
      "epoch": 4.2303493449781655,
      "grad_norm": 1.25,
      "learning_rate": 0.0003328132185486957,
      "loss": 4.8701,
      "mean_token_accuracy": 0.1975643053650856,
      "num_tokens": 93387936.0,
      "step": 42625
    },
    {
      "entropy": 5.623096704483032,
      "epoch": 4.230845573640334,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033277897593684224,
      "loss": 4.9057,
      "mean_token_accuracy": 0.1970101609826088,
      "num_tokens": 93399519.0,
      "step": 42630
    },
    {
      "entropy": 5.662504863739014,
      "epoch": 4.231341802302501,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003327447318921947,
      "loss": 4.9745,
      "mean_token_accuracy": 0.19093206524848938,
      "num_tokens": 93410528.0,
      "step": 42635
    },
    {
      "entropy": 5.6938333988189695,
      "epoch": 4.231838030964669,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003327104864156022,
      "loss": 4.9497,
      "mean_token_accuracy": 0.19646688550710678,
      "num_tokens": 93420868.0,
      "step": 42640
    },
    {
      "entropy": 5.674356126785279,
      "epoch": 4.232334259626836,
      "grad_norm": 1.15625,
      "learning_rate": 0.000332676239507914,
      "loss": 4.9422,
      "mean_token_accuracy": 0.19489924013614654,
      "num_tokens": 93433334.0,
      "step": 42645
    },
    {
      "entropy": 5.685933685302734,
      "epoch": 4.232830488289004,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003326419911699793,
      "loss": 4.982,
      "mean_token_accuracy": 0.19485826492309571,
      "num_tokens": 93443972.0,
      "step": 42650
    },
    {
      "entropy": 5.657136249542236,
      "epoch": 4.233326716951171,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003326077414026475,
      "loss": 4.976,
      "mean_token_accuracy": 0.18910789340734482,
      "num_tokens": 93456145.0,
      "step": 42655
    },
    {
      "entropy": 5.650852823257447,
      "epoch": 4.233822945613339,
      "grad_norm": 1.6875,
      "learning_rate": 0.0003325734902067679,
      "loss": 4.9209,
      "mean_token_accuracy": 0.20259502977132798,
      "num_tokens": 93467574.0,
      "step": 42660
    },
    {
      "entropy": 5.616103315353394,
      "epoch": 4.234319174275506,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033253923758318964,
      "loss": 4.9324,
      "mean_token_accuracy": 0.1972751498222351,
      "num_tokens": 93478220.0,
      "step": 42665
    },
    {
      "entropy": 5.72617826461792,
      "epoch": 4.234815402937674,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033250498353276235,
      "loss": 4.9567,
      "mean_token_accuracy": 0.1961791843175888,
      "num_tokens": 93489610.0,
      "step": 42670
    },
    {
      "entropy": 5.629354619979859,
      "epoch": 4.235311631599841,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033247072805633545,
      "loss": 4.9452,
      "mean_token_accuracy": 0.1977160781621933,
      "num_tokens": 93499950.0,
      "step": 42675
    },
    {
      "entropy": 5.640688085556031,
      "epoch": 4.235807860262009,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033243647115475825,
      "loss": 4.9102,
      "mean_token_accuracy": 0.2020642340183258,
      "num_tokens": 93512789.0,
      "step": 42680
    },
    {
      "entropy": 5.685597991943359,
      "epoch": 4.236304088924176,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033240221282888026,
      "loss": 4.9366,
      "mean_token_accuracy": 0.19641928672790526,
      "num_tokens": 93524942.0,
      "step": 42685
    },
    {
      "entropy": 5.678829097747803,
      "epoch": 4.236800317586344,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003323679530795512,
      "loss": 5.0463,
      "mean_token_accuracy": 0.18983240127563478,
      "num_tokens": 93536773.0,
      "step": 42690
    },
    {
      "entropy": 5.680329942703247,
      "epoch": 4.237296546248511,
      "grad_norm": 1.25,
      "learning_rate": 0.00033233369190762053,
      "loss": 4.9918,
      "mean_token_accuracy": 0.19371294379234313,
      "num_tokens": 93548661.0,
      "step": 42695
    },
    {
      "entropy": 5.73595781326294,
      "epoch": 4.2377927749106785,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003322994293139378,
      "loss": 5.0267,
      "mean_token_accuracy": 0.18812635391950608,
      "num_tokens": 93559513.0,
      "step": 42700
    },
    {
      "entropy": 5.650206708908081,
      "epoch": 4.238289003572846,
      "grad_norm": 1.25,
      "learning_rate": 0.0003322651652993527,
      "loss": 4.9608,
      "mean_token_accuracy": 0.19511729925870896,
      "num_tokens": 93570258.0,
      "step": 42705
    },
    {
      "entropy": 5.644403505325317,
      "epoch": 4.2387852322350135,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033223089986471495,
      "loss": 4.9549,
      "mean_token_accuracy": 0.19597870260477065,
      "num_tokens": 93580339.0,
      "step": 42710
    },
    {
      "entropy": 5.566800451278686,
      "epoch": 4.239281460897182,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033219663301087407,
      "loss": 4.9131,
      "mean_token_accuracy": 0.19344371408224106,
      "num_tokens": 93591857.0,
      "step": 42715
    },
    {
      "entropy": 5.752350282669068,
      "epoch": 4.239777689559349,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033216236473868006,
      "loss": 5.0642,
      "mean_token_accuracy": 0.1881692737340927,
      "num_tokens": 93604231.0,
      "step": 42720
    },
    {
      "entropy": 5.742434549331665,
      "epoch": 4.240273918221517,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003321280950489826,
      "loss": 4.9846,
      "mean_token_accuracy": 0.18812068700790405,
      "num_tokens": 93615589.0,
      "step": 42725
    },
    {
      "entropy": 5.761406326293946,
      "epoch": 4.240770146883684,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003320938239426314,
      "loss": 5.0454,
      "mean_token_accuracy": 0.19246952086687089,
      "num_tokens": 93627624.0,
      "step": 42730
    },
    {
      "entropy": 5.593622493743896,
      "epoch": 4.241266375545852,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003320595514204763,
      "loss": 4.8503,
      "mean_token_accuracy": 0.2057907536625862,
      "num_tokens": 93637856.0,
      "step": 42735
    },
    {
      "entropy": 5.54785041809082,
      "epoch": 4.241762604208019,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033202527748336745,
      "loss": 4.8436,
      "mean_token_accuracy": 0.19479813277721406,
      "num_tokens": 93649665.0,
      "step": 42740
    },
    {
      "entropy": 5.651029014587403,
      "epoch": 4.242258832870187,
      "grad_norm": 1.125,
      "learning_rate": 0.0003319910021321545,
      "loss": 4.9555,
      "mean_token_accuracy": 0.1860696092247963,
      "num_tokens": 93660415.0,
      "step": 42745
    },
    {
      "entropy": 5.723723602294922,
      "epoch": 4.242755061532354,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003319567253676876,
      "loss": 4.9823,
      "mean_token_accuracy": 0.18780918121337892,
      "num_tokens": 93670865.0,
      "step": 42750
    },
    {
      "entropy": 5.6797096729278564,
      "epoch": 4.243251290194522,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033192244719081656,
      "loss": 4.8886,
      "mean_token_accuracy": 0.19827924072742462,
      "num_tokens": 93681111.0,
      "step": 42755
    },
    {
      "entropy": 5.6289801597595215,
      "epoch": 4.243747518856689,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033188816760239146,
      "loss": 4.8567,
      "mean_token_accuracy": 0.20242419540882112,
      "num_tokens": 93693325.0,
      "step": 42760
    },
    {
      "entropy": 5.661604738235473,
      "epoch": 4.244243747518857,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033185388660326244,
      "loss": 4.981,
      "mean_token_accuracy": 0.1899959534406662,
      "num_tokens": 93703788.0,
      "step": 42765
    },
    {
      "entropy": 5.7017717361450195,
      "epoch": 4.244739976181024,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003318196041942795,
      "loss": 5.0027,
      "mean_token_accuracy": 0.19415080696344375,
      "num_tokens": 93713924.0,
      "step": 42770
    },
    {
      "entropy": 5.642365646362305,
      "epoch": 4.2452362048431915,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033178532037629285,
      "loss": 4.9565,
      "mean_token_accuracy": 0.19447064399719238,
      "num_tokens": 93725549.0,
      "step": 42775
    },
    {
      "entropy": 5.644383335113526,
      "epoch": 4.245732433505359,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033175103515015273,
      "loss": 4.9068,
      "mean_token_accuracy": 0.1941198945045471,
      "num_tokens": 93736703.0,
      "step": 42780
    },
    {
      "entropy": 5.612426137924194,
      "epoch": 4.2462286621675265,
      "grad_norm": 1.15625,
      "learning_rate": 0.000331716748516709,
      "loss": 4.8581,
      "mean_token_accuracy": 0.19876248389482498,
      "num_tokens": 93747795.0,
      "step": 42785
    },
    {
      "entropy": 5.709466791152954,
      "epoch": 4.246724890829694,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003316824604768122,
      "loss": 5.0024,
      "mean_token_accuracy": 0.18794546276330948,
      "num_tokens": 93758862.0,
      "step": 42790
    },
    {
      "entropy": 5.6340868949890135,
      "epoch": 4.247221119491861,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033164817103131263,
      "loss": 4.9058,
      "mean_token_accuracy": 0.19899430572986604,
      "num_tokens": 93770130.0,
      "step": 42795
    },
    {
      "entropy": 5.6864235401153564,
      "epoch": 4.24771734815403,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033161388018106033,
      "loss": 4.9537,
      "mean_token_accuracy": 0.19222231656312944,
      "num_tokens": 93781965.0,
      "step": 42800
    },
    {
      "entropy": 5.628657245635987,
      "epoch": 4.248213576816197,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033157958792690587,
      "loss": 4.963,
      "mean_token_accuracy": 0.19004397690296174,
      "num_tokens": 93793421.0,
      "step": 42805
    },
    {
      "entropy": 5.629203844070434,
      "epoch": 4.248709805478365,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003315452942696996,
      "loss": 4.941,
      "mean_token_accuracy": 0.19112565517425537,
      "num_tokens": 93804339.0,
      "step": 42810
    },
    {
      "entropy": 5.680733394622803,
      "epoch": 4.249206034140532,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033151099921029174,
      "loss": 4.9758,
      "mean_token_accuracy": 0.18894012570381163,
      "num_tokens": 93815726.0,
      "step": 42815
    },
    {
      "entropy": 5.6770265102386475,
      "epoch": 4.2497022628027,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033147670274953297,
      "loss": 4.8815,
      "mean_token_accuracy": 0.2041795328259468,
      "num_tokens": 93825978.0,
      "step": 42820
    },
    {
      "entropy": 5.729869413375854,
      "epoch": 4.250198491464867,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033144240488827367,
      "loss": 5.021,
      "mean_token_accuracy": 0.19220593422651291,
      "num_tokens": 93837135.0,
      "step": 42825
    },
    {
      "entropy": 5.698654270172119,
      "epoch": 4.250694720127035,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033140810562736427,
      "loss": 4.94,
      "mean_token_accuracy": 0.19139257967472076,
      "num_tokens": 93848254.0,
      "step": 42830
    },
    {
      "entropy": 5.679120826721191,
      "epoch": 4.251190948789202,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003313738049676555,
      "loss": 4.9386,
      "mean_token_accuracy": 0.19038213193416595,
      "num_tokens": 93859915.0,
      "step": 42835
    },
    {
      "entropy": 5.662217855453491,
      "epoch": 4.25168717745137,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003313395029099978,
      "loss": 4.8926,
      "mean_token_accuracy": 0.2003703385591507,
      "num_tokens": 93871120.0,
      "step": 42840
    },
    {
      "entropy": 5.602608871459961,
      "epoch": 4.252183406113537,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003313051994552418,
      "loss": 4.8849,
      "mean_token_accuracy": 0.19503564834594728,
      "num_tokens": 93881468.0,
      "step": 42845
    },
    {
      "entropy": 5.696030473709106,
      "epoch": 4.2526796347757045,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033127089460423823,
      "loss": 4.9323,
      "mean_token_accuracy": 0.19423094838857652,
      "num_tokens": 93892342.0,
      "step": 42850
    },
    {
      "entropy": 5.694982433319092,
      "epoch": 4.253175863437872,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033123658835783776,
      "loss": 4.9288,
      "mean_token_accuracy": 0.19556872695684432,
      "num_tokens": 93902697.0,
      "step": 42855
    },
    {
      "entropy": 5.651341485977173,
      "epoch": 4.2536720921000395,
      "grad_norm": 1.125,
      "learning_rate": 0.000331202280716891,
      "loss": 4.9332,
      "mean_token_accuracy": 0.20237401127815247,
      "num_tokens": 93914660.0,
      "step": 42860
    },
    {
      "entropy": 5.663005590438843,
      "epoch": 4.254168320762207,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033116797168224876,
      "loss": 4.9267,
      "mean_token_accuracy": 0.19577141851186752,
      "num_tokens": 93926204.0,
      "step": 42865
    },
    {
      "entropy": 5.656917428970337,
      "epoch": 4.254664549424374,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003311336612547619,
      "loss": 4.9787,
      "mean_token_accuracy": 0.19869821518659592,
      "num_tokens": 93938282.0,
      "step": 42870
    },
    {
      "entropy": 5.6386981964111325,
      "epoch": 4.255160778086542,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033109934943528123,
      "loss": 4.9245,
      "mean_token_accuracy": 0.19924566596746446,
      "num_tokens": 93948206.0,
      "step": 42875
    },
    {
      "entropy": 5.6179098129272464,
      "epoch": 4.255657006748709,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003310650362246575,
      "loss": 4.9302,
      "mean_token_accuracy": 0.19861224442720413,
      "num_tokens": 93958804.0,
      "step": 42880
    },
    {
      "entropy": 5.678801345825195,
      "epoch": 4.256153235410878,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003310307216237418,
      "loss": 4.9546,
      "mean_token_accuracy": 0.18774465769529342,
      "num_tokens": 93969500.0,
      "step": 42885
    },
    {
      "entropy": 5.723500108718872,
      "epoch": 4.256649464073045,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033099640563338486,
      "loss": 4.9165,
      "mean_token_accuracy": 0.19506644308567048,
      "num_tokens": 93980164.0,
      "step": 42890
    },
    {
      "entropy": 5.67572808265686,
      "epoch": 4.257145692735213,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033096208825443774,
      "loss": 4.8894,
      "mean_token_accuracy": 0.19694258868694306,
      "num_tokens": 93991968.0,
      "step": 42895
    },
    {
      "entropy": 5.6351007461547855,
      "epoch": 4.25764192139738,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033092776948775136,
      "loss": 4.9153,
      "mean_token_accuracy": 0.1998991385102272,
      "num_tokens": 94001836.0,
      "step": 42900
    },
    {
      "entropy": 5.663444232940674,
      "epoch": 4.258138150059548,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033089344933417687,
      "loss": 4.9612,
      "mean_token_accuracy": 0.19296840727329254,
      "num_tokens": 94011828.0,
      "step": 42905
    },
    {
      "entropy": 5.686535453796386,
      "epoch": 4.258634378721715,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033085912779456527,
      "loss": 4.9579,
      "mean_token_accuracy": 0.20026956349611283,
      "num_tokens": 94022287.0,
      "step": 42910
    },
    {
      "entropy": 5.672333097457885,
      "epoch": 4.259130607383883,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033082480486976764,
      "loss": 4.8702,
      "mean_token_accuracy": 0.20373392701148987,
      "num_tokens": 94033214.0,
      "step": 42915
    },
    {
      "entropy": 5.671464586257935,
      "epoch": 4.25962683604605,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033079048056063515,
      "loss": 4.991,
      "mean_token_accuracy": 0.19065542072057723,
      "num_tokens": 94044190.0,
      "step": 42920
    },
    {
      "entropy": 5.658810806274414,
      "epoch": 4.2601230647082176,
      "grad_norm": 1.25,
      "learning_rate": 0.00033075615486801905,
      "loss": 4.9875,
      "mean_token_accuracy": 0.19180181622505188,
      "num_tokens": 94055105.0,
      "step": 42925
    },
    {
      "entropy": 5.741708469390869,
      "epoch": 4.260619293370385,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033072182779277033,
      "loss": 5.0527,
      "mean_token_accuracy": 0.18889260441064834,
      "num_tokens": 94065807.0,
      "step": 42930
    },
    {
      "entropy": 5.748111867904663,
      "epoch": 4.2611155220325525,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033068749933574037,
      "loss": 5.0448,
      "mean_token_accuracy": 0.18579353541135787,
      "num_tokens": 94078925.0,
      "step": 42935
    },
    {
      "entropy": 5.665765285491943,
      "epoch": 4.26161175069472,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003306531694977805,
      "loss": 4.9494,
      "mean_token_accuracy": 0.19507401287555695,
      "num_tokens": 94090427.0,
      "step": 42940
    },
    {
      "entropy": 5.715259218215943,
      "epoch": 4.2621079793568875,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003306188382797418,
      "loss": 5.0646,
      "mean_token_accuracy": 0.18380051106214523,
      "num_tokens": 94100232.0,
      "step": 42945
    },
    {
      "entropy": 5.7225682735443115,
      "epoch": 4.262604208019055,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003305845056824759,
      "loss": 5.0239,
      "mean_token_accuracy": 0.18579500019550324,
      "num_tokens": 94110718.0,
      "step": 42950
    },
    {
      "entropy": 5.698736143112183,
      "epoch": 4.263100436681222,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000330550171706834,
      "loss": 4.9602,
      "mean_token_accuracy": 0.18893135488033294,
      "num_tokens": 94122148.0,
      "step": 42955
    },
    {
      "entropy": 5.719652271270752,
      "epoch": 4.26359666534339,
      "grad_norm": 1.25,
      "learning_rate": 0.0003305158363536675,
      "loss": 4.9467,
      "mean_token_accuracy": 0.19702381640672684,
      "num_tokens": 94133024.0,
      "step": 42960
    },
    {
      "entropy": 5.681160497665405,
      "epoch": 4.264092894005557,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003304814996238279,
      "loss": 4.995,
      "mean_token_accuracy": 0.18946695029735566,
      "num_tokens": 94144911.0,
      "step": 42965
    },
    {
      "entropy": 5.722539758682251,
      "epoch": 4.264589122667726,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003304471615181667,
      "loss": 4.9724,
      "mean_token_accuracy": 0.1919853135943413,
      "num_tokens": 94156475.0,
      "step": 42970
    },
    {
      "entropy": 5.630614805221557,
      "epoch": 4.265085351329893,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003304128220375354,
      "loss": 4.9359,
      "mean_token_accuracy": 0.19642397314310073,
      "num_tokens": 94166919.0,
      "step": 42975
    },
    {
      "entropy": 5.667329406738281,
      "epoch": 4.265581579992061,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003303784811827854,
      "loss": 4.9764,
      "mean_token_accuracy": 0.19172192066907884,
      "num_tokens": 94177450.0,
      "step": 42980
    },
    {
      "entropy": 5.653913068771362,
      "epoch": 4.266077808654228,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033034413895476853,
      "loss": 4.8942,
      "mean_token_accuracy": 0.1949417620897293,
      "num_tokens": 94187934.0,
      "step": 42985
    },
    {
      "entropy": 5.704659509658813,
      "epoch": 4.266574037316396,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003303097953543362,
      "loss": 5.0055,
      "mean_token_accuracy": 0.19617352038621902,
      "num_tokens": 94198599.0,
      "step": 42990
    },
    {
      "entropy": 5.692276048660278,
      "epoch": 4.267070265978563,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033027545038234003,
      "loss": 4.9772,
      "mean_token_accuracy": 0.19105376899242402,
      "num_tokens": 94210001.0,
      "step": 42995
    },
    {
      "entropy": 5.707976865768432,
      "epoch": 4.267566494640731,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033024110403963194,
      "loss": 4.9436,
      "mean_token_accuracy": 0.19334954023361206,
      "num_tokens": 94220061.0,
      "step": 43000
    },
    {
      "entropy": 5.72731351852417,
      "epoch": 4.268062723302898,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033020675632706347,
      "loss": 4.9938,
      "mean_token_accuracy": 0.19500559717416763,
      "num_tokens": 94232369.0,
      "step": 43005
    },
    {
      "entropy": 5.715361738204956,
      "epoch": 4.2685589519650655,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003301724072454864,
      "loss": 4.9484,
      "mean_token_accuracy": 0.1884622722864151,
      "num_tokens": 94242344.0,
      "step": 43010
    },
    {
      "entropy": 5.704632043838501,
      "epoch": 4.269055180627233,
      "grad_norm": 1.25,
      "learning_rate": 0.00033013805679575257,
      "loss": 4.9494,
      "mean_token_accuracy": 0.1974320411682129,
      "num_tokens": 94252511.0,
      "step": 43015
    },
    {
      "entropy": 5.694336414337158,
      "epoch": 4.2695514092894005,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033010370497871364,
      "loss": 5.0169,
      "mean_token_accuracy": 0.1866871803998947,
      "num_tokens": 94264072.0,
      "step": 43020
    },
    {
      "entropy": 5.631732749938965,
      "epoch": 4.270047637951568,
      "grad_norm": 1.25,
      "learning_rate": 0.00033006935179522164,
      "loss": 4.8858,
      "mean_token_accuracy": 0.20255052000284196,
      "num_tokens": 94274993.0,
      "step": 43025
    },
    {
      "entropy": 5.6967967510223385,
      "epoch": 4.270543866613735,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003300349972461283,
      "loss": 4.9928,
      "mean_token_accuracy": 0.19733549952507018,
      "num_tokens": 94286573.0,
      "step": 43030
    },
    {
      "entropy": 5.697701025009155,
      "epoch": 4.271040095275903,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003300006413322856,
      "loss": 5.0073,
      "mean_token_accuracy": 0.1937405839562416,
      "num_tokens": 94297210.0,
      "step": 43035
    },
    {
      "entropy": 5.751306581497192,
      "epoch": 4.27153632393807,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032996628405454557,
      "loss": 4.9868,
      "mean_token_accuracy": 0.18889281749725342,
      "num_tokens": 94307902.0,
      "step": 43040
    },
    {
      "entropy": 5.738165855407715,
      "epoch": 4.272032552600238,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032993192541376016,
      "loss": 5.0357,
      "mean_token_accuracy": 0.18585488945245743,
      "num_tokens": 94318594.0,
      "step": 43045
    },
    {
      "entropy": 5.736921119689941,
      "epoch": 4.272528781262405,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032989756541078135,
      "loss": 4.9553,
      "mean_token_accuracy": 0.19383109956979752,
      "num_tokens": 94328325.0,
      "step": 43050
    },
    {
      "entropy": 5.685820388793945,
      "epoch": 4.273025009924574,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032986320404646127,
      "loss": 4.9784,
      "mean_token_accuracy": 0.18440455198287964,
      "num_tokens": 94338778.0,
      "step": 43055
    },
    {
      "entropy": 5.774709844589234,
      "epoch": 4.273521238586741,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003298288413216519,
      "loss": 5.1007,
      "mean_token_accuracy": 0.18630993515253066,
      "num_tokens": 94350763.0,
      "step": 43060
    },
    {
      "entropy": 5.70986533164978,
      "epoch": 4.274017467248909,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032979447723720533,
      "loss": 4.9376,
      "mean_token_accuracy": 0.19304608106613158,
      "num_tokens": 94361825.0,
      "step": 43065
    },
    {
      "entropy": 5.669178152084351,
      "epoch": 4.274513695911076,
      "grad_norm": 1.25,
      "learning_rate": 0.00032976011179397394,
      "loss": 4.9094,
      "mean_token_accuracy": 0.19832629710435867,
      "num_tokens": 94372141.0,
      "step": 43070
    },
    {
      "entropy": 5.68864221572876,
      "epoch": 4.275009924573244,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032972574499280964,
      "loss": 4.983,
      "mean_token_accuracy": 0.1948707267642021,
      "num_tokens": 94383323.0,
      "step": 43075
    },
    {
      "entropy": 5.649596357345581,
      "epoch": 4.275506153235411,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003296913768345649,
      "loss": 4.9029,
      "mean_token_accuracy": 0.19994126111268998,
      "num_tokens": 94394095.0,
      "step": 43080
    },
    {
      "entropy": 5.679821634292603,
      "epoch": 4.2760023818975785,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003296570073200918,
      "loss": 5.0118,
      "mean_token_accuracy": 0.1927112191915512,
      "num_tokens": 94405920.0,
      "step": 43085
    },
    {
      "entropy": 5.746376466751099,
      "epoch": 4.276498610559746,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003296226364502428,
      "loss": 4.9482,
      "mean_token_accuracy": 0.19152594208717347,
      "num_tokens": 94416318.0,
      "step": 43090
    },
    {
      "entropy": 5.6390985488891605,
      "epoch": 4.2769948392219135,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003295882642258699,
      "loss": 4.8699,
      "mean_token_accuracy": 0.20078589916229247,
      "num_tokens": 94425665.0,
      "step": 43095
    },
    {
      "entropy": 5.631504392623901,
      "epoch": 4.277491067884081,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003295538906478259,
      "loss": 4.87,
      "mean_token_accuracy": 0.20023286789655687,
      "num_tokens": 94435660.0,
      "step": 43100
    },
    {
      "entropy": 5.632411623001099,
      "epoch": 4.277987296546248,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003295195157169628,
      "loss": 4.9314,
      "mean_token_accuracy": 0.1940179243683815,
      "num_tokens": 94445429.0,
      "step": 43105
    },
    {
      "entropy": 5.677900695800782,
      "epoch": 4.278483525208416,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003294851394341332,
      "loss": 4.9392,
      "mean_token_accuracy": 0.19840228855609893,
      "num_tokens": 94455764.0,
      "step": 43110
    },
    {
      "entropy": 5.698104476928711,
      "epoch": 4.278979753870583,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003294507618001897,
      "loss": 4.9175,
      "mean_token_accuracy": 0.1935563489794731,
      "num_tokens": 94466123.0,
      "step": 43115
    },
    {
      "entropy": 5.706913948059082,
      "epoch": 4.279475982532751,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003294163828159845,
      "loss": 5.0077,
      "mean_token_accuracy": 0.18485609591007232,
      "num_tokens": 94476493.0,
      "step": 43120
    },
    {
      "entropy": 5.718250799179077,
      "epoch": 4.279972211194918,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032938200248237035,
      "loss": 4.956,
      "mean_token_accuracy": 0.19141625016927719,
      "num_tokens": 94488629.0,
      "step": 43125
    },
    {
      "entropy": 5.699174499511718,
      "epoch": 4.280468439857086,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003293476208001996,
      "loss": 4.9517,
      "mean_token_accuracy": 0.1964233249425888,
      "num_tokens": 94500510.0,
      "step": 43130
    },
    {
      "entropy": 5.686292886734009,
      "epoch": 4.280964668519253,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003293132377703251,
      "loss": 5.0057,
      "mean_token_accuracy": 0.18604594320058823,
      "num_tokens": 94511318.0,
      "step": 43135
    },
    {
      "entropy": 5.660642528533936,
      "epoch": 4.281460897181422,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032927885339359924,
      "loss": 4.8937,
      "mean_token_accuracy": 0.2010871410369873,
      "num_tokens": 94522689.0,
      "step": 43140
    },
    {
      "entropy": 5.699308252334594,
      "epoch": 4.281957125843589,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032924446767087487,
      "loss": 4.9848,
      "mean_token_accuracy": 0.1935030922293663,
      "num_tokens": 94534324.0,
      "step": 43145
    },
    {
      "entropy": 5.6464375972747805,
      "epoch": 4.282453354505757,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003292100806030045,
      "loss": 4.9554,
      "mean_token_accuracy": 0.1992902472615242,
      "num_tokens": 94546091.0,
      "step": 43150
    },
    {
      "entropy": 5.659823417663574,
      "epoch": 4.282949583167924,
      "grad_norm": 1.390625,
      "learning_rate": 0.00032917569219084096,
      "loss": 4.9533,
      "mean_token_accuracy": 0.19326135367155076,
      "num_tokens": 94556296.0,
      "step": 43155
    },
    {
      "entropy": 5.750423860549927,
      "epoch": 4.2834458118300915,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032914130243523704,
      "loss": 5.0131,
      "mean_token_accuracy": 0.18823162019252776,
      "num_tokens": 94568232.0,
      "step": 43160
    },
    {
      "entropy": 5.647044992446899,
      "epoch": 4.283942040492259,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003291069113370454,
      "loss": 4.9075,
      "mean_token_accuracy": 0.19956208467483522,
      "num_tokens": 94577778.0,
      "step": 43165
    },
    {
      "entropy": 5.650958299636841,
      "epoch": 4.2844382691544265,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032907251889711904,
      "loss": 4.9478,
      "mean_token_accuracy": 0.19407082349061966,
      "num_tokens": 94588382.0,
      "step": 43170
    },
    {
      "entropy": 5.706186819076538,
      "epoch": 4.284934497816594,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003290381251163106,
      "loss": 4.8897,
      "mean_token_accuracy": 0.21964203119277953,
      "num_tokens": 94599933.0,
      "step": 43175
    },
    {
      "entropy": 5.732546758651734,
      "epoch": 4.285430726478761,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032900372999547327,
      "loss": 4.9988,
      "mean_token_accuracy": 0.19060284048318862,
      "num_tokens": 94613438.0,
      "step": 43180
    },
    {
      "entropy": 5.635937452316284,
      "epoch": 4.285926955140929,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003289693335354597,
      "loss": 4.8943,
      "mean_token_accuracy": 0.1988831043243408,
      "num_tokens": 94624532.0,
      "step": 43185
    },
    {
      "entropy": 5.720756244659424,
      "epoch": 4.286423183803096,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032893493573712295,
      "loss": 4.9755,
      "mean_token_accuracy": 0.18906737267971038,
      "num_tokens": 94634371.0,
      "step": 43190
    },
    {
      "entropy": 5.7193516254425045,
      "epoch": 4.286919412465264,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003289005366013161,
      "loss": 5.0623,
      "mean_token_accuracy": 0.18596380203962326,
      "num_tokens": 94646425.0,
      "step": 43195
    },
    {
      "entropy": 5.709211683273315,
      "epoch": 4.287415641127431,
      "grad_norm": 1.28125,
      "learning_rate": 0.000328866136128892,
      "loss": 4.9466,
      "mean_token_accuracy": 0.1986587330698967,
      "num_tokens": 94656075.0,
      "step": 43200
    },
    {
      "entropy": 5.7289323806762695,
      "epoch": 4.287911869789599,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003288317343207038,
      "loss": 4.9588,
      "mean_token_accuracy": 0.1950291484594345,
      "num_tokens": 94665807.0,
      "step": 43205
    },
    {
      "entropy": 5.674737405776978,
      "epoch": 4.288408098451766,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032879733117760463,
      "loss": 5.031,
      "mean_token_accuracy": 0.1919906571507454,
      "num_tokens": 94676635.0,
      "step": 43210
    },
    {
      "entropy": 5.607063865661621,
      "epoch": 4.288904327113934,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003287629267004475,
      "loss": 4.8413,
      "mean_token_accuracy": 0.20256975740194322,
      "num_tokens": 94687614.0,
      "step": 43215
    },
    {
      "entropy": 5.644339847564697,
      "epoch": 4.289400555776101,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032872852089008576,
      "loss": 4.9999,
      "mean_token_accuracy": 0.19023745507001877,
      "num_tokens": 94698890.0,
      "step": 43220
    },
    {
      "entropy": 5.745917844772339,
      "epoch": 4.28989678443827,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032869411374737234,
      "loss": 5.04,
      "mean_token_accuracy": 0.18670948296785356,
      "num_tokens": 94709554.0,
      "step": 43225
    },
    {
      "entropy": 5.726171588897705,
      "epoch": 4.290393013100437,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032865970527316065,
      "loss": 5.0661,
      "mean_token_accuracy": 0.18442486822605134,
      "num_tokens": 94719867.0,
      "step": 43230
    },
    {
      "entropy": 5.716692638397217,
      "epoch": 4.2908892417626046,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032862529546830394,
      "loss": 4.9346,
      "mean_token_accuracy": 0.19499774575233458,
      "num_tokens": 94730506.0,
      "step": 43235
    },
    {
      "entropy": 5.719853305816651,
      "epoch": 4.291385470424772,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003285908843336554,
      "loss": 5.0119,
      "mean_token_accuracy": 0.18880380541086197,
      "num_tokens": 94741279.0,
      "step": 43240
    },
    {
      "entropy": 5.633294010162354,
      "epoch": 4.2918816990869395,
      "grad_norm": 1.25,
      "learning_rate": 0.0003285564718700684,
      "loss": 4.8754,
      "mean_token_accuracy": 0.20332719683647155,
      "num_tokens": 94752056.0,
      "step": 43245
    },
    {
      "entropy": 5.693610286712646,
      "epoch": 4.292377927749107,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032852205807839635,
      "loss": 4.9292,
      "mean_token_accuracy": 0.1980213075876236,
      "num_tokens": 94762825.0,
      "step": 43250
    },
    {
      "entropy": 5.7141721725463865,
      "epoch": 4.2928741564112745,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032848764295949253,
      "loss": 4.9577,
      "mean_token_accuracy": 0.18975317478179932,
      "num_tokens": 94772945.0,
      "step": 43255
    },
    {
      "entropy": 5.67806830406189,
      "epoch": 4.293370385073442,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032845322651421047,
      "loss": 4.8944,
      "mean_token_accuracy": 0.2032913461327553,
      "num_tokens": 94783269.0,
      "step": 43260
    },
    {
      "entropy": 5.748733758926392,
      "epoch": 4.293866613735609,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003284188087434036,
      "loss": 5.0493,
      "mean_token_accuracy": 0.18800313472747804,
      "num_tokens": 94795159.0,
      "step": 43265
    },
    {
      "entropy": 5.649770450592041,
      "epoch": 4.294362842397777,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003283843896479253,
      "loss": 4.9212,
      "mean_token_accuracy": 0.19303372353315354,
      "num_tokens": 94805992.0,
      "step": 43270
    },
    {
      "entropy": 5.717065525054932,
      "epoch": 4.294859071059944,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003283499692286292,
      "loss": 5.0504,
      "mean_token_accuracy": 0.19557275176048278,
      "num_tokens": 94816336.0,
      "step": 43275
    },
    {
      "entropy": 5.716075849533081,
      "epoch": 4.295355299722112,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003283155474863688,
      "loss": 4.965,
      "mean_token_accuracy": 0.189340640604496,
      "num_tokens": 94827495.0,
      "step": 43280
    },
    {
      "entropy": 5.755176115036011,
      "epoch": 4.295851528384279,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032828112442199775,
      "loss": 4.9789,
      "mean_token_accuracy": 0.18648620545864106,
      "num_tokens": 94838652.0,
      "step": 43285
    },
    {
      "entropy": 5.709683418273926,
      "epoch": 4.296347757046447,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032824670003636956,
      "loss": 4.9151,
      "mean_token_accuracy": 0.19230325967073442,
      "num_tokens": 94848904.0,
      "step": 43290
    },
    {
      "entropy": 5.613851404190063,
      "epoch": 4.296843985708614,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032821227433033805,
      "loss": 4.9267,
      "mean_token_accuracy": 0.1951769083738327,
      "num_tokens": 94859209.0,
      "step": 43295
    },
    {
      "entropy": 5.606897497177124,
      "epoch": 4.297340214370782,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032817784730475676,
      "loss": 4.8966,
      "mean_token_accuracy": 0.1981578141450882,
      "num_tokens": 94869511.0,
      "step": 43300
    },
    {
      "entropy": 5.683113622665405,
      "epoch": 4.297836443032949,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032814341896047946,
      "loss": 4.9877,
      "mean_token_accuracy": 0.19463569521903992,
      "num_tokens": 94879612.0,
      "step": 43305
    },
    {
      "entropy": 5.623797464370727,
      "epoch": 4.298332671695118,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032810898929835984,
      "loss": 4.9178,
      "mean_token_accuracy": 0.19567287415266038,
      "num_tokens": 94890345.0,
      "step": 43310
    },
    {
      "entropy": 5.654074382781983,
      "epoch": 4.298828900357285,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003280745583192517,
      "loss": 4.9455,
      "mean_token_accuracy": 0.19639973789453508,
      "num_tokens": 94900123.0,
      "step": 43315
    },
    {
      "entropy": 5.639940977096558,
      "epoch": 4.2993251290194525,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032804012602400903,
      "loss": 4.9335,
      "mean_token_accuracy": 0.20383881628513337,
      "num_tokens": 94910030.0,
      "step": 43320
    },
    {
      "entropy": 5.730897521972656,
      "epoch": 4.29982135768162,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003280056924134853,
      "loss": 5.0249,
      "mean_token_accuracy": 0.19019870907068254,
      "num_tokens": 94920437.0,
      "step": 43325
    },
    {
      "entropy": 5.681334686279297,
      "epoch": 4.3003175863437875,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032797125748853476,
      "loss": 5.0236,
      "mean_token_accuracy": 0.1910251647233963,
      "num_tokens": 94931204.0,
      "step": 43330
    },
    {
      "entropy": 5.667814064025879,
      "epoch": 4.300813815005955,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032793682125001116,
      "loss": 4.954,
      "mean_token_accuracy": 0.19459691643714905,
      "num_tokens": 94941677.0,
      "step": 43335
    },
    {
      "entropy": 5.648542881011963,
      "epoch": 4.301310043668122,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003279023836987685,
      "loss": 4.8508,
      "mean_token_accuracy": 0.20225286036729812,
      "num_tokens": 94952074.0,
      "step": 43340
    },
    {
      "entropy": 5.670362854003907,
      "epoch": 4.30180627233029,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003278679448356606,
      "loss": 4.9719,
      "mean_token_accuracy": 0.1967293217778206,
      "num_tokens": 94964098.0,
      "step": 43345
    },
    {
      "entropy": 5.616846895217895,
      "epoch": 4.302302500992457,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003278335046615417,
      "loss": 4.9278,
      "mean_token_accuracy": 0.19718758910894393,
      "num_tokens": 94975163.0,
      "step": 43350
    },
    {
      "entropy": 5.656820631027221,
      "epoch": 4.302798729654625,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032779906317726565,
      "loss": 4.8089,
      "mean_token_accuracy": 0.2107919216156006,
      "num_tokens": 94986737.0,
      "step": 43355
    },
    {
      "entropy": 5.6764349937438965,
      "epoch": 4.303294958316792,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003277646203836866,
      "loss": 4.9343,
      "mean_token_accuracy": 0.1979324534535408,
      "num_tokens": 94998530.0,
      "step": 43360
    },
    {
      "entropy": 5.7068475723266605,
      "epoch": 4.30379118697896,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003277301762816588,
      "loss": 5.0051,
      "mean_token_accuracy": 0.19000673294067383,
      "num_tokens": 95008579.0,
      "step": 43365
    },
    {
      "entropy": 5.647618103027344,
      "epoch": 4.304287415641127,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003276957308720361,
      "loss": 5.0553,
      "mean_token_accuracy": 0.1883775770664215,
      "num_tokens": 95020792.0,
      "step": 43370
    },
    {
      "entropy": 5.70197696685791,
      "epoch": 4.304783644303295,
      "grad_norm": 1.25,
      "learning_rate": 0.0003276612841556729,
      "loss": 4.9862,
      "mean_token_accuracy": 0.1942730039358139,
      "num_tokens": 95031734.0,
      "step": 43375
    },
    {
      "entropy": 5.67957615852356,
      "epoch": 4.305279872965462,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003276268361334233,
      "loss": 4.9405,
      "mean_token_accuracy": 0.18373023420572282,
      "num_tokens": 95043551.0,
      "step": 43380
    },
    {
      "entropy": 5.641284465789795,
      "epoch": 4.30577610162763,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003275923868061416,
      "loss": 4.8916,
      "mean_token_accuracy": 0.2030310332775116,
      "num_tokens": 95053556.0,
      "step": 43385
    },
    {
      "entropy": 5.59353232383728,
      "epoch": 4.306272330289797,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000327557936174682,
      "loss": 4.906,
      "mean_token_accuracy": 0.19864476919174195,
      "num_tokens": 95065203.0,
      "step": 43390
    },
    {
      "entropy": 5.660906219482422,
      "epoch": 4.3067685589519655,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032752348423989886,
      "loss": 4.9565,
      "mean_token_accuracy": 0.18704350739717485,
      "num_tokens": 95074745.0,
      "step": 43395
    },
    {
      "entropy": 5.663943719863892,
      "epoch": 4.307264787614133,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003274890310026465,
      "loss": 4.9813,
      "mean_token_accuracy": 0.1905551955103874,
      "num_tokens": 95086224.0,
      "step": 43400
    },
    {
      "entropy": 5.7044878005981445,
      "epoch": 4.3077610162763005,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032745457646377914,
      "loss": 4.964,
      "mean_token_accuracy": 0.19453366100788116,
      "num_tokens": 95097748.0,
      "step": 43405
    },
    {
      "entropy": 5.7239360332489015,
      "epoch": 4.308257244938468,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032742012062415143,
      "loss": 4.9086,
      "mean_token_accuracy": 0.196651329100132,
      "num_tokens": 95108803.0,
      "step": 43410
    },
    {
      "entropy": 5.65060076713562,
      "epoch": 4.308753473600635,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032738566348461775,
      "loss": 4.9677,
      "mean_token_accuracy": 0.18672332167625427,
      "num_tokens": 95119485.0,
      "step": 43415
    },
    {
      "entropy": 5.6943176746368405,
      "epoch": 4.309249702262803,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003273512050460324,
      "loss": 4.9093,
      "mean_token_accuracy": 0.19943960011005402,
      "num_tokens": 95130405.0,
      "step": 43420
    },
    {
      "entropy": 5.701659059524536,
      "epoch": 4.30974593092497,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032731674530925,
      "loss": 5.005,
      "mean_token_accuracy": 0.1905459463596344,
      "num_tokens": 95140932.0,
      "step": 43425
    },
    {
      "entropy": 5.634835577011108,
      "epoch": 4.310242159587138,
      "grad_norm": 1.3125,
      "learning_rate": 0.000327282284275125,
      "loss": 4.8895,
      "mean_token_accuracy": 0.20337125957012175,
      "num_tokens": 95151420.0,
      "step": 43430
    },
    {
      "entropy": 5.6675389289855955,
      "epoch": 4.310738388249305,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003272478219445121,
      "loss": 4.964,
      "mean_token_accuracy": 0.1953003838658333,
      "num_tokens": 95163012.0,
      "step": 43435
    },
    {
      "entropy": 5.616981792449951,
      "epoch": 4.311234616911473,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003272133583182657,
      "loss": 4.851,
      "mean_token_accuracy": 0.2019875392317772,
      "num_tokens": 95174047.0,
      "step": 43440
    },
    {
      "entropy": 5.634995794296264,
      "epoch": 4.31173084557364,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003271788933972406,
      "loss": 4.9843,
      "mean_token_accuracy": 0.18705955296754836,
      "num_tokens": 95185836.0,
      "step": 43445
    },
    {
      "entropy": 5.706577301025391,
      "epoch": 4.312227074235808,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003271444271822913,
      "loss": 4.998,
      "mean_token_accuracy": 0.1903688684105873,
      "num_tokens": 95198049.0,
      "step": 43450
    },
    {
      "entropy": 5.6504356384277346,
      "epoch": 4.312723302897975,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032710995967427263,
      "loss": 4.8368,
      "mean_token_accuracy": 0.20466443449258803,
      "num_tokens": 95208612.0,
      "step": 43455
    },
    {
      "entropy": 5.625619316101075,
      "epoch": 4.313219531560143,
      "grad_norm": 1.125,
      "learning_rate": 0.00032707549087403925,
      "loss": 4.9616,
      "mean_token_accuracy": 0.1917719215154648,
      "num_tokens": 95219879.0,
      "step": 43460
    },
    {
      "entropy": 5.6984070301055905,
      "epoch": 4.31371576022231,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032704102078244583,
      "loss": 4.9924,
      "mean_token_accuracy": 0.1893153041601181,
      "num_tokens": 95230318.0,
      "step": 43465
    },
    {
      "entropy": 5.6775288581848145,
      "epoch": 4.314211988884478,
      "grad_norm": 1.40625,
      "learning_rate": 0.00032700654940034734,
      "loss": 4.9997,
      "mean_token_accuracy": 0.18290365487337112,
      "num_tokens": 95240371.0,
      "step": 43470
    },
    {
      "entropy": 5.68011794090271,
      "epoch": 4.314708217546645,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003269720767285985,
      "loss": 4.9352,
      "mean_token_accuracy": 0.19097927510738372,
      "num_tokens": 95250803.0,
      "step": 43475
    },
    {
      "entropy": 5.729736089706421,
      "epoch": 4.315204446208813,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003269376027680541,
      "loss": 4.9871,
      "mean_token_accuracy": 0.18642646223306655,
      "num_tokens": 95262746.0,
      "step": 43480
    },
    {
      "entropy": 5.712310647964477,
      "epoch": 4.31570067487098,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032690312751956904,
      "loss": 4.924,
      "mean_token_accuracy": 0.1973551705479622,
      "num_tokens": 95272532.0,
      "step": 43485
    },
    {
      "entropy": 5.67970027923584,
      "epoch": 4.316196903533148,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003268686509839983,
      "loss": 4.9704,
      "mean_token_accuracy": 0.18832489252090454,
      "num_tokens": 95283502.0,
      "step": 43490
    },
    {
      "entropy": 5.674481964111328,
      "epoch": 4.316693132195316,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032683417316219673,
      "loss": 5.0534,
      "mean_token_accuracy": 0.18922065198421478,
      "num_tokens": 95294837.0,
      "step": 43495
    },
    {
      "entropy": 5.716887092590332,
      "epoch": 4.317189360857483,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003267996940550194,
      "loss": 4.9425,
      "mean_token_accuracy": 0.19310502707958221,
      "num_tokens": 95304378.0,
      "step": 43500
    },
    {
      "entropy": 5.712764501571655,
      "epoch": 4.317685589519651,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032676521366332127,
      "loss": 5.0039,
      "mean_token_accuracy": 0.18190802335739137,
      "num_tokens": 95316539.0,
      "step": 43505
    },
    {
      "entropy": 5.7531074523925785,
      "epoch": 4.318181818181818,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003267307319879574,
      "loss": 4.9806,
      "mean_token_accuracy": 0.19621891528367996,
      "num_tokens": 95328561.0,
      "step": 43510
    },
    {
      "entropy": 5.6786786079406735,
      "epoch": 4.318678046843986,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032669624902978287,
      "loss": 4.9347,
      "mean_token_accuracy": 0.19233768284320832,
      "num_tokens": 95338789.0,
      "step": 43515
    },
    {
      "entropy": 5.634509658813476,
      "epoch": 4.319174275506153,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003266617647896526,
      "loss": 4.9169,
      "mean_token_accuracy": 0.19612209498882294,
      "num_tokens": 95350135.0,
      "step": 43520
    },
    {
      "entropy": 5.6958009719848635,
      "epoch": 4.319670504168321,
      "grad_norm": 1.203125,
      "learning_rate": 0.000326627279268422,
      "loss": 4.8862,
      "mean_token_accuracy": 0.20016470402479172,
      "num_tokens": 95359901.0,
      "step": 43525
    },
    {
      "entropy": 5.692270660400391,
      "epoch": 4.320166732830488,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003265927924669462,
      "loss": 4.9262,
      "mean_token_accuracy": 0.1997204303741455,
      "num_tokens": 95371060.0,
      "step": 43530
    },
    {
      "entropy": 5.671407032012939,
      "epoch": 4.320662961492656,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032655830438608026,
      "loss": 4.8535,
      "mean_token_accuracy": 0.20631882101297377,
      "num_tokens": 95382666.0,
      "step": 43535
    },
    {
      "entropy": 5.639203834533691,
      "epoch": 4.321159190154823,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003265238150266794,
      "loss": 4.9104,
      "mean_token_accuracy": 0.19994450509548187,
      "num_tokens": 95392629.0,
      "step": 43540
    },
    {
      "entropy": 5.646855211257934,
      "epoch": 4.321655418816991,
      "grad_norm": 1.1875,
      "learning_rate": 0.000326489324389599,
      "loss": 4.9531,
      "mean_token_accuracy": 0.19078894704580307,
      "num_tokens": 95404041.0,
      "step": 43545
    },
    {
      "entropy": 5.802254009246826,
      "epoch": 4.322151647479158,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032645483247569434,
      "loss": 5.1484,
      "mean_token_accuracy": 0.18475272208452226,
      "num_tokens": 95415347.0,
      "step": 43550
    },
    {
      "entropy": 5.613854217529297,
      "epoch": 4.322647876141326,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003264203392858206,
      "loss": 4.9081,
      "mean_token_accuracy": 0.19614568650722503,
      "num_tokens": 95425649.0,
      "step": 43555
    },
    {
      "entropy": 5.695435380935669,
      "epoch": 4.323144104803493,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003263858448208334,
      "loss": 4.9715,
      "mean_token_accuracy": 0.1936752289533615,
      "num_tokens": 95437526.0,
      "step": 43560
    },
    {
      "entropy": 5.7138659954071045,
      "epoch": 4.323640333465661,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032635134908158786,
      "loss": 4.9846,
      "mean_token_accuracy": 0.19072098135948182,
      "num_tokens": 95448030.0,
      "step": 43565
    },
    {
      "entropy": 5.714077091217041,
      "epoch": 4.324136562127828,
      "grad_norm": 1.125,
      "learning_rate": 0.0003263168520689396,
      "loss": 5.024,
      "mean_token_accuracy": 0.18918701261281967,
      "num_tokens": 95459188.0,
      "step": 43570
    },
    {
      "entropy": 5.71955509185791,
      "epoch": 4.324632790789996,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032628235378374395,
      "loss": 5.0017,
      "mean_token_accuracy": 0.1953689932823181,
      "num_tokens": 95471380.0,
      "step": 43575
    },
    {
      "entropy": 5.669311904907227,
      "epoch": 4.325129019452164,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003262478542268564,
      "loss": 4.9051,
      "mean_token_accuracy": 0.19976016581058503,
      "num_tokens": 95481695.0,
      "step": 43580
    },
    {
      "entropy": 5.709539270401001,
      "epoch": 4.325625248114331,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003262133533991325,
      "loss": 5.1016,
      "mean_token_accuracy": 0.1895965114235878,
      "num_tokens": 95492860.0,
      "step": 43585
    },
    {
      "entropy": 5.655738592147827,
      "epoch": 4.326121476776499,
      "grad_norm": 1.125,
      "learning_rate": 0.00032617885130142784,
      "loss": 4.8876,
      "mean_token_accuracy": 0.20439999550580978,
      "num_tokens": 95504519.0,
      "step": 43590
    },
    {
      "entropy": 5.68358039855957,
      "epoch": 4.326617705438666,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032614434793459785,
      "loss": 4.9615,
      "mean_token_accuracy": 0.18662925958633422,
      "num_tokens": 95515549.0,
      "step": 43595
    },
    {
      "entropy": 5.725066995620727,
      "epoch": 4.327113934100834,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003261098432994983,
      "loss": 5.0356,
      "mean_token_accuracy": 0.1914721429347992,
      "num_tokens": 95525444.0,
      "step": 43600
    },
    {
      "entropy": 5.718221759796142,
      "epoch": 4.327610162763001,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003260753373969848,
      "loss": 4.9629,
      "mean_token_accuracy": 0.196437706053257,
      "num_tokens": 95536191.0,
      "step": 43605
    },
    {
      "entropy": 5.658056211471558,
      "epoch": 4.328106391425169,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000326040830227913,
      "loss": 4.8837,
      "mean_token_accuracy": 0.19498025327920915,
      "num_tokens": 95546422.0,
      "step": 43610
    },
    {
      "entropy": 5.665006017684936,
      "epoch": 4.328602620087336,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032600632179313857,
      "loss": 4.9658,
      "mean_token_accuracy": 0.19717718809843063,
      "num_tokens": 95556344.0,
      "step": 43615
    },
    {
      "entropy": 5.6765772819519045,
      "epoch": 4.329098848749504,
      "grad_norm": 1.25,
      "learning_rate": 0.0003259718120935172,
      "loss": 4.9747,
      "mean_token_accuracy": 0.19421077519655228,
      "num_tokens": 95567693.0,
      "step": 43620
    },
    {
      "entropy": 5.7128440856933596,
      "epoch": 4.329595077411671,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032593730112990475,
      "loss": 4.9798,
      "mean_token_accuracy": 0.19382269233465194,
      "num_tokens": 95578823.0,
      "step": 43625
    },
    {
      "entropy": 5.72349910736084,
      "epoch": 4.330091306073839,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032590278890315703,
      "loss": 5.0444,
      "mean_token_accuracy": 0.18477269411087036,
      "num_tokens": 95589830.0,
      "step": 43630
    },
    {
      "entropy": 5.670898723602295,
      "epoch": 4.330587534736006,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003258682754141298,
      "loss": 4.8968,
      "mean_token_accuracy": 0.19633827358484268,
      "num_tokens": 95601622.0,
      "step": 43635
    },
    {
      "entropy": 5.662805795669556,
      "epoch": 4.331083763398174,
      "grad_norm": 1.21875,
      "learning_rate": 0.000325833760663679,
      "loss": 4.8656,
      "mean_token_accuracy": 0.20091236829757692,
      "num_tokens": 95611845.0,
      "step": 43640
    },
    {
      "entropy": 5.650281572341919,
      "epoch": 4.331579992060341,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003257992446526605,
      "loss": 4.9399,
      "mean_token_accuracy": 0.2017944037914276,
      "num_tokens": 95623540.0,
      "step": 43645
    },
    {
      "entropy": 5.729867362976075,
      "epoch": 4.3320762207225085,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032576472738193006,
      "loss": 5.0367,
      "mean_token_accuracy": 0.1854973778128624,
      "num_tokens": 95633384.0,
      "step": 43650
    },
    {
      "entropy": 5.7270478248596195,
      "epoch": 4.332572449384676,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003257302088523439,
      "loss": 4.9931,
      "mean_token_accuracy": 0.18732902705669402,
      "num_tokens": 95645297.0,
      "step": 43655
    },
    {
      "entropy": 5.669024562835693,
      "epoch": 4.333068678046844,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003256956890647578,
      "loss": 4.8914,
      "mean_token_accuracy": 0.19508918970823289,
      "num_tokens": 95655329.0,
      "step": 43660
    },
    {
      "entropy": 5.718230485916138,
      "epoch": 4.333564906709012,
      "grad_norm": 1.25,
      "learning_rate": 0.0003256611680200279,
      "loss": 4.9945,
      "mean_token_accuracy": 0.19709359258413314,
      "num_tokens": 95666166.0,
      "step": 43665
    },
    {
      "entropy": 5.691750621795654,
      "epoch": 4.334061135371179,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032562664571901014,
      "loss": 4.9486,
      "mean_token_accuracy": 0.19286076873540878,
      "num_tokens": 95676289.0,
      "step": 43670
    },
    {
      "entropy": 5.636114168167114,
      "epoch": 4.334557364033347,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032559212216256076,
      "loss": 4.9337,
      "mean_token_accuracy": 0.1963438868522644,
      "num_tokens": 95687135.0,
      "step": 43675
    },
    {
      "entropy": 5.675699281692505,
      "epoch": 4.335053592695514,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003255575973515356,
      "loss": 5.0286,
      "mean_token_accuracy": 0.1851458489894867,
      "num_tokens": 95698864.0,
      "step": 43680
    },
    {
      "entropy": 5.692305374145508,
      "epoch": 4.335549821357682,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003255230712867911,
      "loss": 4.9823,
      "mean_token_accuracy": 0.19394802749156953,
      "num_tokens": 95709075.0,
      "step": 43685
    },
    {
      "entropy": 5.628979444503784,
      "epoch": 4.336046050019849,
      "grad_norm": 1.25,
      "learning_rate": 0.00032548854396918325,
      "loss": 4.8942,
      "mean_token_accuracy": 0.19945311099290847,
      "num_tokens": 95719030.0,
      "step": 43690
    },
    {
      "entropy": 5.714851522445679,
      "epoch": 4.336542278682017,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032545401539956837,
      "loss": 4.9558,
      "mean_token_accuracy": 0.1883280724287033,
      "num_tokens": 95730714.0,
      "step": 43695
    },
    {
      "entropy": 5.710557270050049,
      "epoch": 4.337038507344184,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032541948557880245,
      "loss": 4.9069,
      "mean_token_accuracy": 0.20910223871469497,
      "num_tokens": 95741579.0,
      "step": 43700
    },
    {
      "entropy": 5.600241947174072,
      "epoch": 4.337534736006352,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003253849545077421,
      "loss": 4.9571,
      "mean_token_accuracy": 0.1991075485944748,
      "num_tokens": 95752896.0,
      "step": 43705
    },
    {
      "entropy": 5.709091424942017,
      "epoch": 4.338030964668519,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032535042218724346,
      "loss": 4.9796,
      "mean_token_accuracy": 0.1947006419301033,
      "num_tokens": 95763604.0,
      "step": 43710
    },
    {
      "entropy": 5.703115653991699,
      "epoch": 4.338527193330687,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032531588861816273,
      "loss": 4.8864,
      "mean_token_accuracy": 0.2073889821767807,
      "num_tokens": 95774568.0,
      "step": 43715
    },
    {
      "entropy": 5.7282031059265135,
      "epoch": 4.339023421992854,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003252813538013565,
      "loss": 4.9852,
      "mean_token_accuracy": 0.19445157349109649,
      "num_tokens": 95786001.0,
      "step": 43720
    },
    {
      "entropy": 5.70026044845581,
      "epoch": 4.3395196506550215,
      "grad_norm": 1.171875,
      "learning_rate": 0.000325246817737681,
      "loss": 4.9864,
      "mean_token_accuracy": 0.1928622767329216,
      "num_tokens": 95796538.0,
      "step": 43725
    },
    {
      "entropy": 5.664597606658935,
      "epoch": 4.340015879317189,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003252122804279927,
      "loss": 4.8949,
      "mean_token_accuracy": 0.19633783102035524,
      "num_tokens": 95807934.0,
      "step": 43730
    },
    {
      "entropy": 5.7169294357299805,
      "epoch": 4.3405121079793565,
      "grad_norm": 1.34375,
      "learning_rate": 0.000325177741873148,
      "loss": 5.0045,
      "mean_token_accuracy": 0.18635278791189194,
      "num_tokens": 95817708.0,
      "step": 43735
    },
    {
      "entropy": 5.682724475860596,
      "epoch": 4.341008336641524,
      "grad_norm": 1.25,
      "learning_rate": 0.00032514320207400343,
      "loss": 4.9586,
      "mean_token_accuracy": 0.19768269658088683,
      "num_tokens": 95828717.0,
      "step": 43740
    },
    {
      "entropy": 5.754802703857422,
      "epoch": 4.341504565303692,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003251086610314155,
      "loss": 4.9917,
      "mean_token_accuracy": 0.19205283373594284,
      "num_tokens": 95840382.0,
      "step": 43745
    },
    {
      "entropy": 5.687585926055908,
      "epoch": 4.34200079396586,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032507411874624077,
      "loss": 4.9412,
      "mean_token_accuracy": 0.1931057468056679,
      "num_tokens": 95851141.0,
      "step": 43750
    },
    {
      "entropy": 5.712850522994995,
      "epoch": 4.342497022628027,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032503957521933576,
      "loss": 5.0132,
      "mean_token_accuracy": 0.19594351947307587,
      "num_tokens": 95862779.0,
      "step": 43755
    },
    {
      "entropy": 5.708487558364868,
      "epoch": 4.342993251290195,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032500503045155714,
      "loss": 4.9504,
      "mean_token_accuracy": 0.19554950296878815,
      "num_tokens": 95872810.0,
      "step": 43760
    },
    {
      "entropy": 5.714484977722168,
      "epoch": 4.343489479952362,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003249704844437615,
      "loss": 4.9916,
      "mean_token_accuracy": 0.1935027524828911,
      "num_tokens": 95882895.0,
      "step": 43765
    },
    {
      "entropy": 5.67972092628479,
      "epoch": 4.34398570861453,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003249359371968055,
      "loss": 4.9926,
      "mean_token_accuracy": 0.195270274579525,
      "num_tokens": 95893780.0,
      "step": 43770
    },
    {
      "entropy": 5.665400266647339,
      "epoch": 4.344481937276697,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032490138871154597,
      "loss": 4.9352,
      "mean_token_accuracy": 0.19898581504821777,
      "num_tokens": 95905292.0,
      "step": 43775
    },
    {
      "entropy": 5.704160976409912,
      "epoch": 4.344978165938865,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032486683898883944,
      "loss": 4.9323,
      "mean_token_accuracy": 0.1960470661520958,
      "num_tokens": 95915655.0,
      "step": 43780
    },
    {
      "entropy": 5.69166407585144,
      "epoch": 4.345474394601032,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032483228802954276,
      "loss": 5.0264,
      "mean_token_accuracy": 0.1844446450471878,
      "num_tokens": 95926896.0,
      "step": 43785
    },
    {
      "entropy": 5.682055044174194,
      "epoch": 4.3459706232632,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003247977358345127,
      "loss": 4.9816,
      "mean_token_accuracy": 0.18415109366178511,
      "num_tokens": 95937830.0,
      "step": 43790
    },
    {
      "entropy": 5.683033132553101,
      "epoch": 4.346466851925367,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003247631824046061,
      "loss": 4.9657,
      "mean_token_accuracy": 0.19047604203224183,
      "num_tokens": 95948476.0,
      "step": 43795
    },
    {
      "entropy": 5.755273628234863,
      "epoch": 4.3469630805875346,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003247286277406799,
      "loss": 4.9884,
      "mean_token_accuracy": 0.1913724884390831,
      "num_tokens": 95959603.0,
      "step": 43800
    },
    {
      "entropy": 5.629755210876465,
      "epoch": 4.347459309249702,
      "grad_norm": 1.25,
      "learning_rate": 0.0003246940718435908,
      "loss": 4.8707,
      "mean_token_accuracy": 0.20518653690814972,
      "num_tokens": 95969260.0,
      "step": 43805
    },
    {
      "entropy": 5.686187267303467,
      "epoch": 4.3479555379118695,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032465951471419584,
      "loss": 4.9961,
      "mean_token_accuracy": 0.19278664141893387,
      "num_tokens": 95980852.0,
      "step": 43810
    },
    {
      "entropy": 5.63807487487793,
      "epoch": 4.348451766574037,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003246249563533519,
      "loss": 4.8935,
      "mean_token_accuracy": 0.20338543355464936,
      "num_tokens": 95992815.0,
      "step": 43815
    },
    {
      "entropy": 5.6442351818084715,
      "epoch": 4.3489479952362045,
      "grad_norm": 1.25,
      "learning_rate": 0.000324590396761916,
      "loss": 4.9081,
      "mean_token_accuracy": 0.19656796306371688,
      "num_tokens": 96003865.0,
      "step": 43820
    },
    {
      "entropy": 5.6890476703643795,
      "epoch": 4.349444223898372,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032455583594074504,
      "loss": 4.9679,
      "mean_token_accuracy": 0.19605953842401505,
      "num_tokens": 96013943.0,
      "step": 43825
    },
    {
      "entropy": 5.688915205001831,
      "epoch": 4.34994045256054,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003245212738906961,
      "loss": 4.9483,
      "mean_token_accuracy": 0.19765369594097137,
      "num_tokens": 96024514.0,
      "step": 43830
    },
    {
      "entropy": 5.670521211624146,
      "epoch": 4.350436681222708,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003244867106126265,
      "loss": 4.9775,
      "mean_token_accuracy": 0.19395572394132615,
      "num_tokens": 96034657.0,
      "step": 43835
    },
    {
      "entropy": 5.733423233032227,
      "epoch": 4.350932909884875,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000324452146107393,
      "loss": 5.006,
      "mean_token_accuracy": 0.1896560862660408,
      "num_tokens": 96045007.0,
      "step": 43840
    },
    {
      "entropy": 5.715137958526611,
      "epoch": 4.351429138547043,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003244175803758527,
      "loss": 4.9914,
      "mean_token_accuracy": 0.18521799743175507,
      "num_tokens": 96055317.0,
      "step": 43845
    },
    {
      "entropy": 5.7474143505096436,
      "epoch": 4.35192536720921,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000324383013418863,
      "loss": 5.0355,
      "mean_token_accuracy": 0.19242772907018663,
      "num_tokens": 96066904.0,
      "step": 43850
    },
    {
      "entropy": 5.637742424011231,
      "epoch": 4.352421595871378,
      "grad_norm": 1.25,
      "learning_rate": 0.00032434844523728104,
      "loss": 4.8632,
      "mean_token_accuracy": 0.1965806797146797,
      "num_tokens": 96077450.0,
      "step": 43855
    },
    {
      "entropy": 5.768032884597778,
      "epoch": 4.352917824533545,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003243138758319639,
      "loss": 5.0918,
      "mean_token_accuracy": 0.1887398600578308,
      "num_tokens": 96088601.0,
      "step": 43860
    },
    {
      "entropy": 5.594685363769531,
      "epoch": 4.353414053195713,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032427930520376885,
      "loss": 4.9418,
      "mean_token_accuracy": 0.1991427570581436,
      "num_tokens": 96100312.0,
      "step": 43865
    },
    {
      "entropy": 5.729957962036133,
      "epoch": 4.35391028185788,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032424473335355335,
      "loss": 4.9523,
      "mean_token_accuracy": 0.18989777714014053,
      "num_tokens": 96110492.0,
      "step": 43870
    },
    {
      "entropy": 5.661911535263061,
      "epoch": 4.354406510520048,
      "grad_norm": 1.125,
      "learning_rate": 0.0003242101602821745,
      "loss": 4.9606,
      "mean_token_accuracy": 0.19673848003149033,
      "num_tokens": 96121552.0,
      "step": 43875
    },
    {
      "entropy": 5.6794768333435055,
      "epoch": 4.354902739182215,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003241755859904897,
      "loss": 4.9368,
      "mean_token_accuracy": 0.20340023338794708,
      "num_tokens": 96132945.0,
      "step": 43880
    },
    {
      "entropy": 5.657744312286377,
      "epoch": 4.3553989678443825,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003241410104793564,
      "loss": 4.8897,
      "mean_token_accuracy": 0.19600522071123122,
      "num_tokens": 96142952.0,
      "step": 43885
    },
    {
      "entropy": 5.651652908325195,
      "epoch": 4.35589519650655,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003241064337496318,
      "loss": 4.9769,
      "mean_token_accuracy": 0.19265807420015335,
      "num_tokens": 96153327.0,
      "step": 43890
    },
    {
      "entropy": 5.584132528305053,
      "epoch": 4.3563914251687175,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003240718558021736,
      "loss": 4.8846,
      "mean_token_accuracy": 0.20270387977361679,
      "num_tokens": 96163783.0,
      "step": 43895
    },
    {
      "entropy": 5.680524635314941,
      "epoch": 4.356887653830885,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032403727663783905,
      "loss": 4.9852,
      "mean_token_accuracy": 0.1923016905784607,
      "num_tokens": 96174273.0,
      "step": 43900
    },
    {
      "entropy": 5.734805059432984,
      "epoch": 4.357383882493052,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003240026962574857,
      "loss": 5.0452,
      "mean_token_accuracy": 0.18738834410905839,
      "num_tokens": 96185217.0,
      "step": 43905
    },
    {
      "entropy": 5.729546737670899,
      "epoch": 4.35788011115522,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003239681146619711,
      "loss": 4.9934,
      "mean_token_accuracy": 0.19028573632240295,
      "num_tokens": 96196933.0,
      "step": 43910
    },
    {
      "entropy": 5.726847696304321,
      "epoch": 4.358376339817388,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032393353185215275,
      "loss": 5.0253,
      "mean_token_accuracy": 0.1917932450771332,
      "num_tokens": 96207249.0,
      "step": 43915
    },
    {
      "entropy": 5.648269510269165,
      "epoch": 4.358872568479556,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003238989478288883,
      "loss": 4.8635,
      "mean_token_accuracy": 0.18961500376462936,
      "num_tokens": 96217283.0,
      "step": 43920
    },
    {
      "entropy": 5.687754535675049,
      "epoch": 4.359368797141723,
      "grad_norm": 1.34375,
      "learning_rate": 0.00032386436259303523,
      "loss": 4.9606,
      "mean_token_accuracy": 0.19618556946516036,
      "num_tokens": 96227843.0,
      "step": 43925
    },
    {
      "entropy": 5.66981143951416,
      "epoch": 4.359865025803891,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003238297761454513,
      "loss": 4.9482,
      "mean_token_accuracy": 0.20082871615886688,
      "num_tokens": 96238458.0,
      "step": 43930
    },
    {
      "entropy": 5.652870798110962,
      "epoch": 4.360361254466058,
      "grad_norm": 1.125,
      "learning_rate": 0.0003237951884869941,
      "loss": 4.9429,
      "mean_token_accuracy": 0.19236539900302888,
      "num_tokens": 96249356.0,
      "step": 43935
    },
    {
      "entropy": 5.685010480880737,
      "epoch": 4.360857483128226,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003237605996185215,
      "loss": 4.8996,
      "mean_token_accuracy": 0.20142129808664322,
      "num_tokens": 96259783.0,
      "step": 43940
    },
    {
      "entropy": 5.687790489196777,
      "epoch": 4.361353711790393,
      "grad_norm": 1.140625,
      "learning_rate": 0.000323726009540891,
      "loss": 4.9843,
      "mean_token_accuracy": 0.18869914412498473,
      "num_tokens": 96270829.0,
      "step": 43945
    },
    {
      "entropy": 5.599827146530151,
      "epoch": 4.361849940452561,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003236914182549605,
      "loss": 4.9317,
      "mean_token_accuracy": 0.19491953551769256,
      "num_tokens": 96281338.0,
      "step": 43950
    },
    {
      "entropy": 5.626500654220581,
      "epoch": 4.362346169114728,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003236568257615877,
      "loss": 4.9451,
      "mean_token_accuracy": 0.1945813313126564,
      "num_tokens": 96292380.0,
      "step": 43955
    },
    {
      "entropy": 5.723575067520142,
      "epoch": 4.3628423977768955,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003236222320616305,
      "loss": 4.9876,
      "mean_token_accuracy": 0.19156788736581803,
      "num_tokens": 96304631.0,
      "step": 43960
    },
    {
      "entropy": 5.690464353561401,
      "epoch": 4.363338626439063,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003235876371559467,
      "loss": 5.0324,
      "mean_token_accuracy": 0.18636093139648438,
      "num_tokens": 96316022.0,
      "step": 43965
    },
    {
      "entropy": 5.6827606678009035,
      "epoch": 4.3638348551012305,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003235530410453942,
      "loss": 5.0199,
      "mean_token_accuracy": 0.1876733586192131,
      "num_tokens": 96326956.0,
      "step": 43970
    },
    {
      "entropy": 5.675261163711548,
      "epoch": 4.364331083763398,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003235184437308309,
      "loss": 4.9408,
      "mean_token_accuracy": 0.20179031938314437,
      "num_tokens": 96338038.0,
      "step": 43975
    },
    {
      "entropy": 5.716614675521851,
      "epoch": 4.364827312425565,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032348384521311473,
      "loss": 4.928,
      "mean_token_accuracy": 0.18559528589248658,
      "num_tokens": 96349215.0,
      "step": 43980
    },
    {
      "entropy": 5.7408922672271725,
      "epoch": 4.365323541087733,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003234492454931037,
      "loss": 5.0627,
      "mean_token_accuracy": 0.18830374628305435,
      "num_tokens": 96360160.0,
      "step": 43985
    },
    {
      "entropy": 5.640000295639038,
      "epoch": 4.3658197697499,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003234146445716558,
      "loss": 4.9303,
      "mean_token_accuracy": 0.19449035078287125,
      "num_tokens": 96370968.0,
      "step": 43990
    },
    {
      "entropy": 5.667708539962769,
      "epoch": 4.366315998412068,
      "grad_norm": 1.203125,
      "learning_rate": 0.000323380042449629,
      "loss": 5.0279,
      "mean_token_accuracy": 0.19269801676273346,
      "num_tokens": 96382142.0,
      "step": 43995
    },
    {
      "entropy": 5.77301926612854,
      "epoch": 4.366812227074236,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003233454391278814,
      "loss": 5.0351,
      "mean_token_accuracy": 0.1846872478723526,
      "num_tokens": 96392439.0,
      "step": 44000
    },
    {
      "entropy": 5.7382269382476805,
      "epoch": 4.367308455736404,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003233108346072711,
      "loss": 4.9955,
      "mean_token_accuracy": 0.1843368023633957,
      "num_tokens": 96404740.0,
      "step": 44005
    },
    {
      "entropy": 5.713052225112915,
      "epoch": 4.367804684398571,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032327622888865613,
      "loss": 5.0228,
      "mean_token_accuracy": 0.19512611776590347,
      "num_tokens": 96416043.0,
      "step": 44010
    },
    {
      "entropy": 5.6228940963745115,
      "epoch": 4.368300913060739,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003232416219728948,
      "loss": 4.8773,
      "mean_token_accuracy": 0.1974542185664177,
      "num_tokens": 96427098.0,
      "step": 44015
    },
    {
      "entropy": 5.663023567199707,
      "epoch": 4.368797141722906,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00032320701386084514,
      "loss": 4.8955,
      "mean_token_accuracy": 0.204081130027771,
      "num_tokens": 96437162.0,
      "step": 44020
    },
    {
      "entropy": 5.719414520263672,
      "epoch": 4.369293370385074,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032317240455336557,
      "loss": 4.9984,
      "mean_token_accuracy": 0.18446434885263444,
      "num_tokens": 96447725.0,
      "step": 44025
    },
    {
      "entropy": 5.691858386993408,
      "epoch": 4.369789599047241,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032313779405131394,
      "loss": 4.9422,
      "mean_token_accuracy": 0.18800624161958696,
      "num_tokens": 96458720.0,
      "step": 44030
    },
    {
      "entropy": 5.715126943588257,
      "epoch": 4.3702858277094085,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003231031823555489,
      "loss": 5.0053,
      "mean_token_accuracy": 0.19284881204366683,
      "num_tokens": 96469753.0,
      "step": 44035
    },
    {
      "entropy": 5.689345502853394,
      "epoch": 4.370782056371576,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003230685694669285,
      "loss": 4.8913,
      "mean_token_accuracy": 0.20248475968837737,
      "num_tokens": 96480304.0,
      "step": 44040
    },
    {
      "entropy": 5.720668697357178,
      "epoch": 4.3712782850337435,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003230339553863112,
      "loss": 5.0137,
      "mean_token_accuracy": 0.1881990760564804,
      "num_tokens": 96491163.0,
      "step": 44045
    },
    {
      "entropy": 5.69573278427124,
      "epoch": 4.371774513695911,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032299934011455536,
      "loss": 4.9898,
      "mean_token_accuracy": 0.1917001262307167,
      "num_tokens": 96501496.0,
      "step": 44050
    },
    {
      "entropy": 5.652791118621826,
      "epoch": 4.372270742358078,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003229647236525192,
      "loss": 4.8438,
      "mean_token_accuracy": 0.19622355103492736,
      "num_tokens": 96513154.0,
      "step": 44055
    },
    {
      "entropy": 5.6966766834259035,
      "epoch": 4.372766971020246,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003229301060010613,
      "loss": 4.9912,
      "mean_token_accuracy": 0.1921660780906677,
      "num_tokens": 96523472.0,
      "step": 44060
    },
    {
      "entropy": 5.683363437652588,
      "epoch": 4.373263199682413,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032289548716104004,
      "loss": 4.95,
      "mean_token_accuracy": 0.19322688430547713,
      "num_tokens": 96534332.0,
      "step": 44065
    },
    {
      "entropy": 5.652930974960327,
      "epoch": 4.373759428344581,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032286086713331395,
      "loss": 4.936,
      "mean_token_accuracy": 0.1965326711535454,
      "num_tokens": 96545996.0,
      "step": 44070
    },
    {
      "entropy": 5.728706502914429,
      "epoch": 4.374255657006748,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003228262459187415,
      "loss": 4.9784,
      "mean_token_accuracy": 0.2005157634615898,
      "num_tokens": 96556753.0,
      "step": 44075
    },
    {
      "entropy": 5.655083560943604,
      "epoch": 4.374751885668916,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032279162351818116,
      "loss": 4.9182,
      "mean_token_accuracy": 0.1986798956990242,
      "num_tokens": 96568418.0,
      "step": 44080
    },
    {
      "entropy": 5.649936246871948,
      "epoch": 4.375248114331084,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032275699993249157,
      "loss": 4.8965,
      "mean_token_accuracy": 0.19587279707193375,
      "num_tokens": 96579470.0,
      "step": 44085
    },
    {
      "entropy": 5.709332513809204,
      "epoch": 4.375744342993252,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003227223751625312,
      "loss": 4.9759,
      "mean_token_accuracy": 0.19573488980531692,
      "num_tokens": 96591311.0,
      "step": 44090
    },
    {
      "entropy": 5.6298188209533695,
      "epoch": 4.376240571655419,
      "grad_norm": 1.390625,
      "learning_rate": 0.00032268774920915885,
      "loss": 4.869,
      "mean_token_accuracy": 0.20172009319067002,
      "num_tokens": 96601487.0,
      "step": 44095
    },
    {
      "entropy": 5.701767301559448,
      "epoch": 4.376736800317587,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032265312207323303,
      "loss": 5.0141,
      "mean_token_accuracy": 0.1912998840212822,
      "num_tokens": 96612252.0,
      "step": 44100
    },
    {
      "entropy": 5.702988004684448,
      "epoch": 4.377233028979754,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003226184937556125,
      "loss": 4.9654,
      "mean_token_accuracy": 0.2004123106598854,
      "num_tokens": 96623212.0,
      "step": 44105
    },
    {
      "entropy": 5.658825922012329,
      "epoch": 4.377729257641922,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003225838642571559,
      "loss": 4.8874,
      "mean_token_accuracy": 0.1915363773703575,
      "num_tokens": 96633792.0,
      "step": 44110
    },
    {
      "entropy": 5.714066123962402,
      "epoch": 4.378225486304089,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032254923357872206,
      "loss": 5.0091,
      "mean_token_accuracy": 0.18465050011873246,
      "num_tokens": 96643834.0,
      "step": 44115
    },
    {
      "entropy": 5.656023740768433,
      "epoch": 4.3787217149662565,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003225146017211696,
      "loss": 4.9971,
      "mean_token_accuracy": 0.19228863418102266,
      "num_tokens": 96655017.0,
      "step": 44120
    },
    {
      "entropy": 5.720269060134887,
      "epoch": 4.379217943628424,
      "grad_norm": 1.25,
      "learning_rate": 0.00032247996868535743,
      "loss": 4.9794,
      "mean_token_accuracy": 0.19083731025457382,
      "num_tokens": 96665494.0,
      "step": 44125
    },
    {
      "entropy": 5.663931846618652,
      "epoch": 4.3797141722905915,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032244533447214446,
      "loss": 4.9924,
      "mean_token_accuracy": 0.1867418646812439,
      "num_tokens": 96677023.0,
      "step": 44130
    },
    {
      "entropy": 5.662795352935791,
      "epoch": 4.380210400952759,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032241069908238933,
      "loss": 4.9608,
      "mean_token_accuracy": 0.18488940000534057,
      "num_tokens": 96688187.0,
      "step": 44135
    },
    {
      "entropy": 5.723365592956543,
      "epoch": 4.380706629614926,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00032237606251695107,
      "loss": 5.0125,
      "mean_token_accuracy": 0.19177283793687822,
      "num_tokens": 96700598.0,
      "step": 44140
    },
    {
      "entropy": 5.703943586349487,
      "epoch": 4.381202858277094,
      "grad_norm": 1.25,
      "learning_rate": 0.0003223414247766885,
      "loss": 4.9807,
      "mean_token_accuracy": 0.1873222067952156,
      "num_tokens": 96711921.0,
      "step": 44145
    },
    {
      "entropy": 5.65576138496399,
      "epoch": 4.381699086939261,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032230678586246057,
      "loss": 4.9219,
      "mean_token_accuracy": 0.19166638255119323,
      "num_tokens": 96723648.0,
      "step": 44150
    },
    {
      "entropy": 5.712467098236084,
      "epoch": 4.382195315601429,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00032227214577512637,
      "loss": 4.9643,
      "mean_token_accuracy": 0.2028913155198097,
      "num_tokens": 96734425.0,
      "step": 44155
    },
    {
      "entropy": 5.673060321807862,
      "epoch": 4.382691544263596,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032223750451554475,
      "loss": 4.9406,
      "mean_token_accuracy": 0.19237712621688843,
      "num_tokens": 96744909.0,
      "step": 44160
    },
    {
      "entropy": 5.662000036239624,
      "epoch": 4.383187772925764,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032220286208457484,
      "loss": 4.8695,
      "mean_token_accuracy": 0.1993036985397339,
      "num_tokens": 96756339.0,
      "step": 44165
    },
    {
      "entropy": 5.688985824584961,
      "epoch": 4.383684001587932,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032216821848307566,
      "loss": 4.9646,
      "mean_token_accuracy": 0.18896400630474092,
      "num_tokens": 96766684.0,
      "step": 44170
    },
    {
      "entropy": 5.694642114639282,
      "epoch": 4.3841802302501,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032213357371190627,
      "loss": 5.0489,
      "mean_token_accuracy": 0.18441291600465776,
      "num_tokens": 96777919.0,
      "step": 44175
    },
    {
      "entropy": 5.644793939590454,
      "epoch": 4.384676458912267,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003220989277719259,
      "loss": 4.8866,
      "mean_token_accuracy": 0.19792551547288895,
      "num_tokens": 96787724.0,
      "step": 44180
    },
    {
      "entropy": 5.691301870346069,
      "epoch": 4.385172687574435,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003220642806639935,
      "loss": 4.9095,
      "mean_token_accuracy": 0.20193656980991365,
      "num_tokens": 96798994.0,
      "step": 44185
    },
    {
      "entropy": 5.601290369033814,
      "epoch": 4.385668916236602,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032202963238896844,
      "loss": 4.8664,
      "mean_token_accuracy": 0.19735607057809829,
      "num_tokens": 96809414.0,
      "step": 44190
    },
    {
      "entropy": 5.640344142913818,
      "epoch": 4.3861651448987695,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003219949829477098,
      "loss": 4.9393,
      "mean_token_accuracy": 0.19516644775867462,
      "num_tokens": 96820786.0,
      "step": 44195
    },
    {
      "entropy": 5.698918247222901,
      "epoch": 4.386661373560937,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032196033234107685,
      "loss": 4.9447,
      "mean_token_accuracy": 0.19538366794586182,
      "num_tokens": 96831144.0,
      "step": 44200
    },
    {
      "entropy": 5.7206494331359865,
      "epoch": 4.3871576022231045,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032192568056992884,
      "loss": 4.9253,
      "mean_token_accuracy": 0.20170160830020906,
      "num_tokens": 96842330.0,
      "step": 44205
    },
    {
      "entropy": 5.6134607791900635,
      "epoch": 4.387653830885272,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032189102763512513,
      "loss": 4.8887,
      "mean_token_accuracy": 0.19502053558826446,
      "num_tokens": 96853051.0,
      "step": 44210
    },
    {
      "entropy": 5.681473159790039,
      "epoch": 4.388150059547439,
      "grad_norm": 1.25,
      "learning_rate": 0.0003218563735375249,
      "loss": 4.9605,
      "mean_token_accuracy": 0.1901121497154236,
      "num_tokens": 96864399.0,
      "step": 44215
    },
    {
      "entropy": 5.765049123764038,
      "epoch": 4.388646288209607,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003218217182779876,
      "loss": 4.9712,
      "mean_token_accuracy": 0.18639393001794816,
      "num_tokens": 96876558.0,
      "step": 44220
    },
    {
      "entropy": 5.674968814849853,
      "epoch": 4.389142516871774,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003217870618573726,
      "loss": 4.918,
      "mean_token_accuracy": 0.1957247108221054,
      "num_tokens": 96888332.0,
      "step": 44225
    },
    {
      "entropy": 5.696566820144653,
      "epoch": 4.389638745533942,
      "grad_norm": 1.25,
      "learning_rate": 0.00032175240427653924,
      "loss": 4.9442,
      "mean_token_accuracy": 0.19682586789131165,
      "num_tokens": 96900007.0,
      "step": 44230
    },
    {
      "entropy": 5.655014610290527,
      "epoch": 4.390134974196109,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003217177455363471,
      "loss": 4.8891,
      "mean_token_accuracy": 0.19850936084985732,
      "num_tokens": 96910584.0,
      "step": 44235
    },
    {
      "entropy": 5.707265377044678,
      "epoch": 4.390631202858277,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032168308563765535,
      "loss": 5.0178,
      "mean_token_accuracy": 0.1890997216105461,
      "num_tokens": 96921689.0,
      "step": 44240
    },
    {
      "entropy": 5.7160069942474365,
      "epoch": 4.391127431520444,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032164842458132383,
      "loss": 4.9757,
      "mean_token_accuracy": 0.2000517413020134,
      "num_tokens": 96931935.0,
      "step": 44245
    },
    {
      "entropy": 5.721270990371704,
      "epoch": 4.391623660182612,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032161376236821185,
      "loss": 4.9507,
      "mean_token_accuracy": 0.19798182994127272,
      "num_tokens": 96941834.0,
      "step": 44250
    },
    {
      "entropy": 5.678208494186402,
      "epoch": 4.39211988884478,
      "grad_norm": 1.109375,
      "learning_rate": 0.000321579098999179,
      "loss": 4.8827,
      "mean_token_accuracy": 0.2048076421022415,
      "num_tokens": 96952628.0,
      "step": 44255
    },
    {
      "entropy": 5.6857434749603275,
      "epoch": 4.392616117506948,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003215444344750848,
      "loss": 4.9288,
      "mean_token_accuracy": 0.19368433952331543,
      "num_tokens": 96963794.0,
      "step": 44260
    },
    {
      "entropy": 5.654688167572021,
      "epoch": 4.393112346169115,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000321509768796789,
      "loss": 4.9211,
      "mean_token_accuracy": 0.20064457952976228,
      "num_tokens": 96974658.0,
      "step": 44265
    },
    {
      "entropy": 5.694620084762573,
      "epoch": 4.3936085748312825,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003214751019651511,
      "loss": 4.9547,
      "mean_token_accuracy": 0.1956883579492569,
      "num_tokens": 96985784.0,
      "step": 44270
    },
    {
      "entropy": 5.686382913589478,
      "epoch": 4.39410480349345,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003214404339810309,
      "loss": 4.9254,
      "mean_token_accuracy": 0.2002488359808922,
      "num_tokens": 96996267.0,
      "step": 44275
    },
    {
      "entropy": 5.740639877319336,
      "epoch": 4.3946010321556175,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032140576484528794,
      "loss": 5.015,
      "mean_token_accuracy": 0.19415645450353622,
      "num_tokens": 97007891.0,
      "step": 44280
    },
    {
      "entropy": 5.7188397407531735,
      "epoch": 4.395097260817785,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032137109455878207,
      "loss": 4.995,
      "mean_token_accuracy": 0.19168204665184022,
      "num_tokens": 97018342.0,
      "step": 44285
    },
    {
      "entropy": 5.686589479446411,
      "epoch": 4.395593489479952,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003213364231223729,
      "loss": 4.9298,
      "mean_token_accuracy": 0.19417645037174225,
      "num_tokens": 97029450.0,
      "step": 44290
    },
    {
      "entropy": 5.664579391479492,
      "epoch": 4.39608971814212,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032130175053692034,
      "loss": 4.9141,
      "mean_token_accuracy": 0.19626306891441345,
      "num_tokens": 97039981.0,
      "step": 44295
    },
    {
      "entropy": 5.657509183883667,
      "epoch": 4.396585946804287,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003212670768032841,
      "loss": 4.8951,
      "mean_token_accuracy": 0.1964656040072441,
      "num_tokens": 97050264.0,
      "step": 44300
    },
    {
      "entropy": 5.627505159378051,
      "epoch": 4.397082175466455,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032123240192232405,
      "loss": 4.8652,
      "mean_token_accuracy": 0.19913355857133866,
      "num_tokens": 97060638.0,
      "step": 44305
    },
    {
      "entropy": 5.633697175979615,
      "epoch": 4.397578404128622,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003211977258949001,
      "loss": 4.914,
      "mean_token_accuracy": 0.20035328567028046,
      "num_tokens": 97070875.0,
      "step": 44310
    },
    {
      "entropy": 5.724863290786743,
      "epoch": 4.39807463279079,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003211630487218721,
      "loss": 5.0441,
      "mean_token_accuracy": 0.18689889311790467,
      "num_tokens": 97082119.0,
      "step": 44315
    },
    {
      "entropy": 5.694249677658081,
      "epoch": 4.398570861452957,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032112837040409987,
      "loss": 4.9605,
      "mean_token_accuracy": 0.1931263104081154,
      "num_tokens": 97092954.0,
      "step": 44320
    },
    {
      "entropy": 5.71501989364624,
      "epoch": 4.399067090115125,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003210936909424435,
      "loss": 4.9345,
      "mean_token_accuracy": 0.19548677802085876,
      "num_tokens": 97103117.0,
      "step": 44325
    },
    {
      "entropy": 5.6645238399505615,
      "epoch": 4.399563318777292,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032105901033776293,
      "loss": 4.8806,
      "mean_token_accuracy": 0.19411989152431489,
      "num_tokens": 97114262.0,
      "step": 44330
    },
    {
      "entropy": 5.647882699966431,
      "epoch": 4.40005954743946,
      "grad_norm": 1.375,
      "learning_rate": 0.00032102432859091825,
      "loss": 4.9463,
      "mean_token_accuracy": 0.1887612134218216,
      "num_tokens": 97125335.0,
      "step": 44335
    },
    {
      "entropy": 5.587248659133911,
      "epoch": 4.400555776101628,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032098964570276923,
      "loss": 4.8399,
      "mean_token_accuracy": 0.2007406994700432,
      "num_tokens": 97136548.0,
      "step": 44340
    },
    {
      "entropy": 5.653842258453369,
      "epoch": 4.4010520047637955,
      "grad_norm": 1.25,
      "learning_rate": 0.00032095496167417614,
      "loss": 4.916,
      "mean_token_accuracy": 0.19044890850782395,
      "num_tokens": 97146475.0,
      "step": 44345
    },
    {
      "entropy": 5.675239992141724,
      "epoch": 4.401548233425963,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032092027650599907,
      "loss": 4.9781,
      "mean_token_accuracy": 0.19087711870670318,
      "num_tokens": 97156829.0,
      "step": 44350
    },
    {
      "entropy": 5.6479230403900145,
      "epoch": 4.4020444620881305,
      "grad_norm": 1.375,
      "learning_rate": 0.00032088559019909806,
      "loss": 4.8619,
      "mean_token_accuracy": 0.19966960698366165,
      "num_tokens": 97166104.0,
      "step": 44355
    },
    {
      "entropy": 5.668715000152588,
      "epoch": 4.402540690750298,
      "grad_norm": 1.59375,
      "learning_rate": 0.00032085090275433333,
      "loss": 4.9127,
      "mean_token_accuracy": 0.20265305191278457,
      "num_tokens": 97177606.0,
      "step": 44360
    },
    {
      "entropy": 5.7306441307067875,
      "epoch": 4.4030369194124654,
      "grad_norm": 1.375,
      "learning_rate": 0.00032081621417256496,
      "loss": 4.9794,
      "mean_token_accuracy": 0.19235668778419496,
      "num_tokens": 97188728.0,
      "step": 44365
    },
    {
      "entropy": 5.6889934062957765,
      "epoch": 4.403533148074633,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032078152445465317,
      "loss": 4.9935,
      "mean_token_accuracy": 0.1926726683974266,
      "num_tokens": 97199285.0,
      "step": 44370
    },
    {
      "entropy": 5.693049526214599,
      "epoch": 4.4040293767368,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032074683360145827,
      "loss": 4.935,
      "mean_token_accuracy": 0.1912872925400734,
      "num_tokens": 97210345.0,
      "step": 44375
    },
    {
      "entropy": 5.659838914871216,
      "epoch": 4.404525605398968,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032071214161384057,
      "loss": 4.9552,
      "mean_token_accuracy": 0.1937023013830185,
      "num_tokens": 97222291.0,
      "step": 44380
    },
    {
      "entropy": 5.699281406402588,
      "epoch": 4.405021834061135,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003206774484926601,
      "loss": 5.0033,
      "mean_token_accuracy": 0.19261612892150878,
      "num_tokens": 97233087.0,
      "step": 44385
    },
    {
      "entropy": 5.699745082855225,
      "epoch": 4.405518062723303,
      "grad_norm": 1.25,
      "learning_rate": 0.0003206427542387774,
      "loss": 4.9576,
      "mean_token_accuracy": 0.19794164299964906,
      "num_tokens": 97243452.0,
      "step": 44390
    },
    {
      "entropy": 5.726168012619018,
      "epoch": 4.40601429138547,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003206080588530528,
      "loss": 5.0375,
      "mean_token_accuracy": 0.19334515035152436,
      "num_tokens": 97252909.0,
      "step": 44395
    },
    {
      "entropy": 5.706876993179321,
      "epoch": 4.406510520047638,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003205733623363465,
      "loss": 4.9325,
      "mean_token_accuracy": 0.19218801856040954,
      "num_tokens": 97262902.0,
      "step": 44400
    },
    {
      "entropy": 5.661709308624268,
      "epoch": 4.407006748709805,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032053866468951907,
      "loss": 5.0574,
      "mean_token_accuracy": 0.191708005964756,
      "num_tokens": 97274146.0,
      "step": 44405
    },
    {
      "entropy": 5.643920040130615,
      "epoch": 4.407502977371973,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032050396591343103,
      "loss": 4.9324,
      "mean_token_accuracy": 0.19340555369853973,
      "num_tokens": 97285493.0,
      "step": 44410
    },
    {
      "entropy": 5.732068634033203,
      "epoch": 4.40799920603414,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003204692660089425,
      "loss": 5.0003,
      "mean_token_accuracy": 0.18792833983898163,
      "num_tokens": 97295293.0,
      "step": 44415
    },
    {
      "entropy": 5.69498291015625,
      "epoch": 4.408495434696308,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032043456497691413,
      "loss": 4.9814,
      "mean_token_accuracy": 0.1896027848124504,
      "num_tokens": 97306147.0,
      "step": 44420
    },
    {
      "entropy": 5.7057905197143555,
      "epoch": 4.408991663358476,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032039986281820655,
      "loss": 4.9052,
      "mean_token_accuracy": 0.19869273751974106,
      "num_tokens": 97317715.0,
      "step": 44425
    },
    {
      "entropy": 5.676892328262329,
      "epoch": 4.4094878920206435,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032036515953368024,
      "loss": 4.9187,
      "mean_token_accuracy": 0.19418133944272994,
      "num_tokens": 97328333.0,
      "step": 44430
    },
    {
      "entropy": 5.660077524185181,
      "epoch": 4.409984120682811,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032033045512419556,
      "loss": 5.0084,
      "mean_token_accuracy": 0.18484001159667968,
      "num_tokens": 97339072.0,
      "step": 44435
    },
    {
      "entropy": 5.625210523605347,
      "epoch": 4.4104803493449785,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032029574959061344,
      "loss": 4.9111,
      "mean_token_accuracy": 0.19768817275762557,
      "num_tokens": 97350506.0,
      "step": 44440
    },
    {
      "entropy": 5.700523138046265,
      "epoch": 4.410976578007146,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003202610429337943,
      "loss": 4.9873,
      "mean_token_accuracy": 0.1855953276157379,
      "num_tokens": 97359852.0,
      "step": 44445
    },
    {
      "entropy": 5.726206827163696,
      "epoch": 4.411472806669313,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032022633515459874,
      "loss": 5.0083,
      "mean_token_accuracy": 0.18496703058481218,
      "num_tokens": 97370848.0,
      "step": 44450
    },
    {
      "entropy": 5.7501442432403564,
      "epoch": 4.411969035331481,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032019162625388756,
      "loss": 5.0139,
      "mean_token_accuracy": 0.19675789177417755,
      "num_tokens": 97380724.0,
      "step": 44455
    },
    {
      "entropy": 5.73380856513977,
      "epoch": 4.412465263993648,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003201569162325215,
      "loss": 4.9651,
      "mean_token_accuracy": 0.19469540417194367,
      "num_tokens": 97389909.0,
      "step": 44460
    },
    {
      "entropy": 5.708972644805908,
      "epoch": 4.412961492655816,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032012220509136115,
      "loss": 5.0256,
      "mean_token_accuracy": 0.19059471487998964,
      "num_tokens": 97403149.0,
      "step": 44465
    },
    {
      "entropy": 5.619215583801269,
      "epoch": 4.413457721317983,
      "grad_norm": 1.25,
      "learning_rate": 0.0003200874928312673,
      "loss": 4.8572,
      "mean_token_accuracy": 0.19858222007751464,
      "num_tokens": 97413729.0,
      "step": 44470
    },
    {
      "entropy": 5.65548357963562,
      "epoch": 4.413953949980151,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003200527794531009,
      "loss": 4.9734,
      "mean_token_accuracy": 0.19151883721351623,
      "num_tokens": 97425217.0,
      "step": 44475
    },
    {
      "entropy": 5.666769504547119,
      "epoch": 4.414450178642318,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032001806495772256,
      "loss": 4.9337,
      "mean_token_accuracy": 0.19321740865707399,
      "num_tokens": 97435325.0,
      "step": 44480
    },
    {
      "entropy": 5.745436716079712,
      "epoch": 4.414946407304486,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003199833493459932,
      "loss": 4.9895,
      "mean_token_accuracy": 0.1956065148115158,
      "num_tokens": 97446556.0,
      "step": 44485
    },
    {
      "entropy": 5.687002563476563,
      "epoch": 4.415442635966653,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003199486326187738,
      "loss": 4.9777,
      "mean_token_accuracy": 0.19667586386203767,
      "num_tokens": 97457325.0,
      "step": 44490
    },
    {
      "entropy": 5.746423768997192,
      "epoch": 4.415938864628821,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003199139147769251,
      "loss": 5.0361,
      "mean_token_accuracy": 0.18654201179742813,
      "num_tokens": 97467397.0,
      "step": 44495
    },
    {
      "entropy": 5.664596176147461,
      "epoch": 4.416435093290988,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003198791958213081,
      "loss": 4.9173,
      "mean_token_accuracy": 0.19124016016721726,
      "num_tokens": 97478362.0,
      "step": 44500
    },
    {
      "entropy": 5.66436653137207,
      "epoch": 4.416931321953156,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031984447575278375,
      "loss": 4.9618,
      "mean_token_accuracy": 0.20324870198965073,
      "num_tokens": 97488764.0,
      "step": 44505
    },
    {
      "entropy": 5.636749744415283,
      "epoch": 4.417427550615324,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031980975457221304,
      "loss": 4.8472,
      "mean_token_accuracy": 0.19844139814376832,
      "num_tokens": 97499957.0,
      "step": 44510
    },
    {
      "entropy": 5.695339488983154,
      "epoch": 4.4179237792774915,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031977503228045696,
      "loss": 4.9412,
      "mean_token_accuracy": 0.193694669008255,
      "num_tokens": 97512137.0,
      "step": 44515
    },
    {
      "entropy": 5.690736293792725,
      "epoch": 4.418420007939659,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003197403088783766,
      "loss": 4.9811,
      "mean_token_accuracy": 0.19406118988990784,
      "num_tokens": 97522255.0,
      "step": 44520
    },
    {
      "entropy": 5.643406534194947,
      "epoch": 4.418916236601826,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031970558436683296,
      "loss": 4.9562,
      "mean_token_accuracy": 0.19599957317113875,
      "num_tokens": 97532868.0,
      "step": 44525
    },
    {
      "entropy": 5.653253173828125,
      "epoch": 4.419412465263994,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003196708587466872,
      "loss": 4.9039,
      "mean_token_accuracy": 0.20053562372922898,
      "num_tokens": 97543499.0,
      "step": 44530
    },
    {
      "entropy": 5.678692197799682,
      "epoch": 4.419908693926161,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031963613201880034,
      "loss": 4.9157,
      "mean_token_accuracy": 0.20199959874153137,
      "num_tokens": 97553081.0,
      "step": 44535
    },
    {
      "entropy": 5.655190181732178,
      "epoch": 4.420404922588329,
      "grad_norm": 1.40625,
      "learning_rate": 0.00031960140418403355,
      "loss": 4.9379,
      "mean_token_accuracy": 0.1926080673933029,
      "num_tokens": 97563037.0,
      "step": 44540
    },
    {
      "entropy": 5.624886178970337,
      "epoch": 4.420901151250496,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031956667524324817,
      "loss": 4.9222,
      "mean_token_accuracy": 0.19176679402589797,
      "num_tokens": 97574256.0,
      "step": 44545
    },
    {
      "entropy": 5.707694005966187,
      "epoch": 4.421397379912664,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031953194519730515,
      "loss": 5.0283,
      "mean_token_accuracy": 0.18541944026947021,
      "num_tokens": 97585874.0,
      "step": 44550
    },
    {
      "entropy": 5.716362190246582,
      "epoch": 4.421893608574831,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031949721404706595,
      "loss": 4.9886,
      "mean_token_accuracy": 0.1883046120405197,
      "num_tokens": 97596200.0,
      "step": 44555
    },
    {
      "entropy": 5.6355126857757565,
      "epoch": 4.422389837236999,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003194624817933917,
      "loss": 4.973,
      "mean_token_accuracy": 0.19739898890256882,
      "num_tokens": 97606961.0,
      "step": 44560
    },
    {
      "entropy": 5.630923318862915,
      "epoch": 4.422886065899166,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003194277484371437,
      "loss": 4.8875,
      "mean_token_accuracy": 0.19704245924949645,
      "num_tokens": 97616892.0,
      "step": 44565
    },
    {
      "entropy": 5.787542295455933,
      "epoch": 4.423382294561334,
      "grad_norm": 1.25,
      "learning_rate": 0.0003193930139791832,
      "loss": 5.0658,
      "mean_token_accuracy": 0.18232773542404174,
      "num_tokens": 97626274.0,
      "step": 44570
    },
    {
      "entropy": 5.687932348251342,
      "epoch": 4.423878523223501,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003193582784203717,
      "loss": 4.9443,
      "mean_token_accuracy": 0.19910501092672347,
      "num_tokens": 97637036.0,
      "step": 44575
    },
    {
      "entropy": 5.696370363235474,
      "epoch": 4.424374751885669,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003193235417615705,
      "loss": 5.0084,
      "mean_token_accuracy": 0.19401873350143434,
      "num_tokens": 97648978.0,
      "step": 44580
    },
    {
      "entropy": 5.694991588592529,
      "epoch": 4.424870980547836,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003192888040036409,
      "loss": 4.969,
      "mean_token_accuracy": 0.19467386454343796,
      "num_tokens": 97660254.0,
      "step": 44585
    },
    {
      "entropy": 5.726095247268677,
      "epoch": 4.425367209210004,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031925406514744445,
      "loss": 5.0208,
      "mean_token_accuracy": 0.19041721820831298,
      "num_tokens": 97670889.0,
      "step": 44590
    },
    {
      "entropy": 5.74200701713562,
      "epoch": 4.425863437872172,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003192193251938425,
      "loss": 5.0187,
      "mean_token_accuracy": 0.18860990256071092,
      "num_tokens": 97681807.0,
      "step": 44595
    },
    {
      "entropy": 5.640779590606689,
      "epoch": 4.426359666534339,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031918458414369655,
      "loss": 4.867,
      "mean_token_accuracy": 0.19651128798723222,
      "num_tokens": 97692844.0,
      "step": 44600
    },
    {
      "entropy": 5.662618160247803,
      "epoch": 4.426855895196507,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031914984199786815,
      "loss": 4.9062,
      "mean_token_accuracy": 0.19717353731393814,
      "num_tokens": 97703855.0,
      "step": 44605
    },
    {
      "entropy": 5.773423385620117,
      "epoch": 4.427352123858674,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031911509875721884,
      "loss": 5.1425,
      "mean_token_accuracy": 0.18246541619300843,
      "num_tokens": 97716175.0,
      "step": 44610
    },
    {
      "entropy": 5.724760675430298,
      "epoch": 4.427848352520842,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031908035442261014,
      "loss": 5.0477,
      "mean_token_accuracy": 0.18907311111688613,
      "num_tokens": 97727982.0,
      "step": 44615
    },
    {
      "entropy": 5.684545707702637,
      "epoch": 4.428344581183009,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003190456089949036,
      "loss": 4.9017,
      "mean_token_accuracy": 0.19957085996866225,
      "num_tokens": 97737594.0,
      "step": 44620
    },
    {
      "entropy": 5.693740797042847,
      "epoch": 4.428840809845177,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031901086247496086,
      "loss": 4.948,
      "mean_token_accuracy": 0.19942713528871536,
      "num_tokens": 97747914.0,
      "step": 44625
    },
    {
      "entropy": 5.740226316452026,
      "epoch": 4.429337038507344,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031897611486364367,
      "loss": 5.0053,
      "mean_token_accuracy": 0.19643132388591766,
      "num_tokens": 97760201.0,
      "step": 44630
    },
    {
      "entropy": 5.76659083366394,
      "epoch": 4.429833267169512,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00031894136616181343,
      "loss": 5.0184,
      "mean_token_accuracy": 0.18428239971399307,
      "num_tokens": 97770612.0,
      "step": 44635
    },
    {
      "entropy": 5.7003227233886715,
      "epoch": 4.430329495831679,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031890661637033204,
      "loss": 4.9423,
      "mean_token_accuracy": 0.19228028506040573,
      "num_tokens": 97781599.0,
      "step": 44640
    },
    {
      "entropy": 5.701705741882324,
      "epoch": 4.430825724493847,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003188718654900612,
      "loss": 4.9557,
      "mean_token_accuracy": 0.199404801428318,
      "num_tokens": 97792198.0,
      "step": 44645
    },
    {
      "entropy": 5.630467939376831,
      "epoch": 4.431321953156014,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031883711352186263,
      "loss": 4.865,
      "mean_token_accuracy": 0.21024141311645508,
      "num_tokens": 97802481.0,
      "step": 44650
    },
    {
      "entropy": 5.713819217681885,
      "epoch": 4.431818181818182,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003188023604665981,
      "loss": 4.9888,
      "mean_token_accuracy": 0.18937449008226395,
      "num_tokens": 97813512.0,
      "step": 44655
    },
    {
      "entropy": 5.698481941223145,
      "epoch": 4.432314410480349,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031876760632512945,
      "loss": 4.9496,
      "mean_token_accuracy": 0.19432883262634276,
      "num_tokens": 97824549.0,
      "step": 44660
    },
    {
      "entropy": 5.760406351089477,
      "epoch": 4.432810639142517,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031873285109831846,
      "loss": 5.0408,
      "mean_token_accuracy": 0.18450562208890914,
      "num_tokens": 97834585.0,
      "step": 44665
    },
    {
      "entropy": 5.695442152023316,
      "epoch": 4.433306867804684,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031869809478702693,
      "loss": 5.0047,
      "mean_token_accuracy": 0.18977962881326677,
      "num_tokens": 97845939.0,
      "step": 44670
    },
    {
      "entropy": 5.69581880569458,
      "epoch": 4.433803096466852,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031866333739211697,
      "loss": 5.0115,
      "mean_token_accuracy": 0.19063443839550018,
      "num_tokens": 97857458.0,
      "step": 44675
    },
    {
      "entropy": 5.677485704421997,
      "epoch": 4.43429932512902,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003186285789144501,
      "loss": 4.888,
      "mean_token_accuracy": 0.1986178696155548,
      "num_tokens": 97868007.0,
      "step": 44680
    },
    {
      "entropy": 5.632523012161255,
      "epoch": 4.434795553791187,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031859381935488867,
      "loss": 4.9274,
      "mean_token_accuracy": 0.19081125408411026,
      "num_tokens": 97879352.0,
      "step": 44685
    },
    {
      "entropy": 5.653643703460693,
      "epoch": 4.435291782453355,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031855905871429444,
      "loss": 4.9402,
      "mean_token_accuracy": 0.19390324950218202,
      "num_tokens": 97889647.0,
      "step": 44690
    },
    {
      "entropy": 5.668124103546143,
      "epoch": 4.435788011115522,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031852429699352937,
      "loss": 4.984,
      "mean_token_accuracy": 0.18776116669178008,
      "num_tokens": 97900811.0,
      "step": 44695
    },
    {
      "entropy": 5.730275106430054,
      "epoch": 4.43628423977769,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031848953419345545,
      "loss": 5.0261,
      "mean_token_accuracy": 0.1911591112613678,
      "num_tokens": 97912050.0,
      "step": 44700
    },
    {
      "entropy": 5.727089691162109,
      "epoch": 4.436780468439857,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003184547703149349,
      "loss": 4.9346,
      "mean_token_accuracy": 0.19838514775037766,
      "num_tokens": 97922892.0,
      "step": 44705
    },
    {
      "entropy": 5.613434076309204,
      "epoch": 4.437276697102025,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003184200053588296,
      "loss": 4.8995,
      "mean_token_accuracy": 0.20803662836551667,
      "num_tokens": 97934555.0,
      "step": 44710
    },
    {
      "entropy": 5.740463447570801,
      "epoch": 4.437772925764192,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003183852393260018,
      "loss": 5.0498,
      "mean_token_accuracy": 0.18658599257469177,
      "num_tokens": 97944677.0,
      "step": 44715
    },
    {
      "entropy": 5.699524688720703,
      "epoch": 4.43826915442636,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003183504722173135,
      "loss": 4.9615,
      "mean_token_accuracy": 0.1981745406985283,
      "num_tokens": 97954194.0,
      "step": 44720
    },
    {
      "entropy": 5.661041355133056,
      "epoch": 4.438765383088527,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000318315704033627,
      "loss": 4.9456,
      "mean_token_accuracy": 0.19201672226190566,
      "num_tokens": 97966142.0,
      "step": 44725
    },
    {
      "entropy": 5.6798285961151125,
      "epoch": 4.439261611750695,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031828093477580423,
      "loss": 4.9971,
      "mean_token_accuracy": 0.1869108185172081,
      "num_tokens": 97978912.0,
      "step": 44730
    },
    {
      "entropy": 5.717312240600586,
      "epoch": 4.439757840412862,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003182461644447077,
      "loss": 4.9766,
      "mean_token_accuracy": 0.18927800208330153,
      "num_tokens": 97988379.0,
      "step": 44735
    },
    {
      "entropy": 5.705500555038452,
      "epoch": 4.44025406907503,
      "grad_norm": 1.125,
      "learning_rate": 0.0003182113930411994,
      "loss": 4.994,
      "mean_token_accuracy": 0.1897117242217064,
      "num_tokens": 97999501.0,
      "step": 44740
    },
    {
      "entropy": 5.705121898651123,
      "epoch": 4.440750297737197,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003181766205661416,
      "loss": 4.993,
      "mean_token_accuracy": 0.18727266043424606,
      "num_tokens": 98012269.0,
      "step": 44745
    },
    {
      "entropy": 5.685342121124267,
      "epoch": 4.441246526399365,
      "grad_norm": 1.25,
      "learning_rate": 0.0003181418470203967,
      "loss": 4.816,
      "mean_token_accuracy": 0.20127199441194535,
      "num_tokens": 98023234.0,
      "step": 44750
    },
    {
      "entropy": 5.706124258041382,
      "epoch": 4.441742755061532,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031810707240482696,
      "loss": 4.9373,
      "mean_token_accuracy": 0.19209374189376832,
      "num_tokens": 98033606.0,
      "step": 44755
    },
    {
      "entropy": 5.666299962997437,
      "epoch": 4.4422389837236995,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031807229672029474,
      "loss": 4.9761,
      "mean_token_accuracy": 0.1919329583644867,
      "num_tokens": 98043751.0,
      "step": 44760
    },
    {
      "entropy": 5.723554086685181,
      "epoch": 4.442735212385867,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003180375199676623,
      "loss": 4.9864,
      "mean_token_accuracy": 0.19003251194953918,
      "num_tokens": 98054005.0,
      "step": 44765
    },
    {
      "entropy": 5.755241012573242,
      "epoch": 4.4432314410480345,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003180027421477922,
      "loss": 4.9618,
      "mean_token_accuracy": 0.1988566905260086,
      "num_tokens": 98064863.0,
      "step": 44770
    },
    {
      "entropy": 5.719689512252808,
      "epoch": 4.443727669710203,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031796796326154675,
      "loss": 4.9735,
      "mean_token_accuracy": 0.19481446892023085,
      "num_tokens": 98074981.0,
      "step": 44775
    },
    {
      "entropy": 5.747901248931885,
      "epoch": 4.44422389837237,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031793318330978825,
      "loss": 4.9912,
      "mean_token_accuracy": 0.19363478273153306,
      "num_tokens": 98085479.0,
      "step": 44780
    },
    {
      "entropy": 5.720680999755859,
      "epoch": 4.444720127034538,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003178984022933795,
      "loss": 5.0195,
      "mean_token_accuracy": 0.18831662982702255,
      "num_tokens": 98097598.0,
      "step": 44785
    },
    {
      "entropy": 5.732397937774659,
      "epoch": 4.445216355696705,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003178636202131828,
      "loss": 5.0213,
      "mean_token_accuracy": 0.18985570222139359,
      "num_tokens": 98109233.0,
      "step": 44790
    },
    {
      "entropy": 5.753763914108276,
      "epoch": 4.445712584358873,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003178288370700605,
      "loss": 4.9746,
      "mean_token_accuracy": 0.1880187064409256,
      "num_tokens": 98119656.0,
      "step": 44795
    },
    {
      "entropy": 5.678819036483764,
      "epoch": 4.44620881302104,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031779405286487544,
      "loss": 5.0216,
      "mean_token_accuracy": 0.19399440586566924,
      "num_tokens": 98130013.0,
      "step": 44800
    },
    {
      "entropy": 5.699996852874756,
      "epoch": 4.446705041683208,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003177592675984901,
      "loss": 4.9846,
      "mean_token_accuracy": 0.1895550310611725,
      "num_tokens": 98142373.0,
      "step": 44805
    },
    {
      "entropy": 5.783759880065918,
      "epoch": 4.447201270345375,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031772448127176704,
      "loss": 5.0294,
      "mean_token_accuracy": 0.18753650933504104,
      "num_tokens": 98154682.0,
      "step": 44810
    },
    {
      "entropy": 5.699734258651733,
      "epoch": 4.447697499007543,
      "grad_norm": 1.25,
      "learning_rate": 0.00031768969388556897,
      "loss": 4.9624,
      "mean_token_accuracy": 0.19836724996566774,
      "num_tokens": 98165426.0,
      "step": 44815
    },
    {
      "entropy": 5.70851411819458,
      "epoch": 4.44819372766971,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003176549054407584,
      "loss": 4.9831,
      "mean_token_accuracy": 0.19744735658168794,
      "num_tokens": 98176156.0,
      "step": 44820
    },
    {
      "entropy": 5.659303092956543,
      "epoch": 4.448689956331878,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003176201159381982,
      "loss": 4.9104,
      "mean_token_accuracy": 0.19617337137460708,
      "num_tokens": 98187239.0,
      "step": 44825
    },
    {
      "entropy": 5.616475439071655,
      "epoch": 4.449186184994045,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031758532537875075,
      "loss": 4.9127,
      "mean_token_accuracy": 0.19879430830478667,
      "num_tokens": 98199052.0,
      "step": 44830
    },
    {
      "entropy": 5.689607334136963,
      "epoch": 4.4496824136562125,
      "grad_norm": 1.125,
      "learning_rate": 0.0003175505337632791,
      "loss": 4.9665,
      "mean_token_accuracy": 0.19513090699911118,
      "num_tokens": 98210320.0,
      "step": 44835
    },
    {
      "entropy": 5.69000883102417,
      "epoch": 4.45017864231838,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000317515741092646,
      "loss": 5.0324,
      "mean_token_accuracy": 0.1889127239584923,
      "num_tokens": 98222841.0,
      "step": 44840
    },
    {
      "entropy": 5.735177659988404,
      "epoch": 4.4506748709805475,
      "grad_norm": 1.140625,
      "learning_rate": 0.000317480947367714,
      "loss": 5.0483,
      "mean_token_accuracy": 0.18510400950908662,
      "num_tokens": 98234217.0,
      "step": 44845
    },
    {
      "entropy": 5.672733402252197,
      "epoch": 4.451171099642715,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003174461525893462,
      "loss": 4.9087,
      "mean_token_accuracy": 0.20098479241132736,
      "num_tokens": 98244588.0,
      "step": 44850
    },
    {
      "entropy": 5.618319320678711,
      "epoch": 4.451667328304882,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031741135675840523,
      "loss": 4.9108,
      "mean_token_accuracy": 0.197864930331707,
      "num_tokens": 98255241.0,
      "step": 44855
    },
    {
      "entropy": 5.659652233123779,
      "epoch": 4.452163556967051,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031737655987575385,
      "loss": 4.9286,
      "mean_token_accuracy": 0.1961026132106781,
      "num_tokens": 98265696.0,
      "step": 44860
    },
    {
      "entropy": 5.711027574539185,
      "epoch": 4.452659785629218,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031734176194225523,
      "loss": 5.0371,
      "mean_token_accuracy": 0.18530966937541962,
      "num_tokens": 98277380.0,
      "step": 44865
    },
    {
      "entropy": 5.747376775741577,
      "epoch": 4.453156014291386,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003173069629587721,
      "loss": 4.9892,
      "mean_token_accuracy": 0.19507194310426712,
      "num_tokens": 98287999.0,
      "step": 44870
    },
    {
      "entropy": 5.686043930053711,
      "epoch": 4.453652242953553,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031727216292616747,
      "loss": 4.9409,
      "mean_token_accuracy": 0.19918566942214966,
      "num_tokens": 98298304.0,
      "step": 44875
    },
    {
      "entropy": 5.692992305755615,
      "epoch": 4.454148471615721,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031723736184530426,
      "loss": 5.0083,
      "mean_token_accuracy": 0.1924528881907463,
      "num_tokens": 98309350.0,
      "step": 44880
    },
    {
      "entropy": 5.670776081085205,
      "epoch": 4.454644700277888,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031720255971704554,
      "loss": 4.9319,
      "mean_token_accuracy": 0.20261043757200242,
      "num_tokens": 98321021.0,
      "step": 44885
    },
    {
      "entropy": 5.6903187274932865,
      "epoch": 4.455140928940056,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003171677565422543,
      "loss": 4.9511,
      "mean_token_accuracy": 0.19894310981035232,
      "num_tokens": 98331592.0,
      "step": 44890
    },
    {
      "entropy": 5.690471601486206,
      "epoch": 4.455637157602223,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031713295232179344,
      "loss": 4.9502,
      "mean_token_accuracy": 0.1928795635700226,
      "num_tokens": 98343303.0,
      "step": 44895
    },
    {
      "entropy": 5.680007028579712,
      "epoch": 4.456133386264391,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003170981470565263,
      "loss": 4.9689,
      "mean_token_accuracy": 0.18761209547519683,
      "num_tokens": 98354022.0,
      "step": 44900
    },
    {
      "entropy": 5.744813013076782,
      "epoch": 4.456629614926558,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031706334074731575,
      "loss": 5.0286,
      "mean_token_accuracy": 0.19080719500780105,
      "num_tokens": 98365973.0,
      "step": 44905
    },
    {
      "entropy": 5.7081560611724855,
      "epoch": 4.4571258435887255,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031702853339502494,
      "loss": 4.9479,
      "mean_token_accuracy": 0.18999970704317093,
      "num_tokens": 98376732.0,
      "step": 44910
    },
    {
      "entropy": 5.6287524700164795,
      "epoch": 4.457622072250893,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031699372500051715,
      "loss": 4.8745,
      "mean_token_accuracy": 0.19875998049974442,
      "num_tokens": 98386613.0,
      "step": 44915
    },
    {
      "entropy": 5.719545221328735,
      "epoch": 4.4581183009130605,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031695891556465545,
      "loss": 5.0222,
      "mean_token_accuracy": 0.19196128249168395,
      "num_tokens": 98397992.0,
      "step": 44920
    },
    {
      "entropy": 5.71344256401062,
      "epoch": 4.458614529575228,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000316924105088303,
      "loss": 4.9731,
      "mean_token_accuracy": 0.19322271049022674,
      "num_tokens": 98407555.0,
      "step": 44925
    },
    {
      "entropy": 5.698149490356445,
      "epoch": 4.4591107582373954,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031688929357232324,
      "loss": 4.9868,
      "mean_token_accuracy": 0.19338187724351882,
      "num_tokens": 98419292.0,
      "step": 44930
    },
    {
      "entropy": 5.757478952407837,
      "epoch": 4.459606986899563,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031685448101757905,
      "loss": 5.0594,
      "mean_token_accuracy": 0.1906697168946266,
      "num_tokens": 98430742.0,
      "step": 44935
    },
    {
      "entropy": 5.679532909393311,
      "epoch": 4.46010321556173,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031681966742493414,
      "loss": 4.9521,
      "mean_token_accuracy": 0.19753036350011827,
      "num_tokens": 98441613.0,
      "step": 44940
    },
    {
      "entropy": 5.627763557434082,
      "epoch": 4.460599444223899,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003167848527952514,
      "loss": 4.8725,
      "mean_token_accuracy": 0.19800393432378768,
      "num_tokens": 98453807.0,
      "step": 44945
    },
    {
      "entropy": 5.713572406768799,
      "epoch": 4.461095672886066,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031675003712939437,
      "loss": 4.9165,
      "mean_token_accuracy": 0.19964347332715987,
      "num_tokens": 98464438.0,
      "step": 44950
    },
    {
      "entropy": 5.618909549713135,
      "epoch": 4.461591901548234,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031671522042822647,
      "loss": 4.847,
      "mean_token_accuracy": 0.19609497338533402,
      "num_tokens": 98475645.0,
      "step": 44955
    },
    {
      "entropy": 5.674815654754639,
      "epoch": 4.462088130210401,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031668040269261093,
      "loss": 4.9749,
      "mean_token_accuracy": 0.18832778036594391,
      "num_tokens": 98488247.0,
      "step": 44960
    },
    {
      "entropy": 5.666963291168213,
      "epoch": 4.462584358872569,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031664558392341125,
      "loss": 4.9558,
      "mean_token_accuracy": 0.19211545288562776,
      "num_tokens": 98499681.0,
      "step": 44965
    },
    {
      "entropy": 5.664484071731567,
      "epoch": 4.463080587534736,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031661076412149085,
      "loss": 4.9483,
      "mean_token_accuracy": 0.19257851243019103,
      "num_tokens": 98510822.0,
      "step": 44970
    },
    {
      "entropy": 5.714825534820557,
      "epoch": 4.463576816196904,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003165759432877131,
      "loss": 4.9191,
      "mean_token_accuracy": 0.1950409457087517,
      "num_tokens": 98520862.0,
      "step": 44975
    },
    {
      "entropy": 5.712248802185059,
      "epoch": 4.464073044859071,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031654112142294157,
      "loss": 4.9802,
      "mean_token_accuracy": 0.19131913036108017,
      "num_tokens": 98531398.0,
      "step": 44980
    },
    {
      "entropy": 5.590309762954712,
      "epoch": 4.464569273521239,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003165062985280397,
      "loss": 4.8496,
      "mean_token_accuracy": 0.21009326577186585,
      "num_tokens": 98541760.0,
      "step": 44985
    },
    {
      "entropy": 5.593487977981567,
      "epoch": 4.465065502183406,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031647147460387115,
      "loss": 4.9009,
      "mean_token_accuracy": 0.196771839261055,
      "num_tokens": 98552807.0,
      "step": 44990
    },
    {
      "entropy": 5.690857553482056,
      "epoch": 4.4655617308455735,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031643664965129935,
      "loss": 4.9925,
      "mean_token_accuracy": 0.18715557008981704,
      "num_tokens": 98563477.0,
      "step": 44995
    },
    {
      "entropy": 5.698572111129761,
      "epoch": 4.466057959507741,
      "grad_norm": 1.265625,
      "learning_rate": 0.000316401823671188,
      "loss": 4.9151,
      "mean_token_accuracy": 0.19480060786008835,
      "num_tokens": 98574474.0,
      "step": 45000
    },
    {
      "epoch": 4.466057959507741,
      "eval_entropy": 5.469200632073132,
      "eval_loss": 5.096126079559326,
      "eval_mean_token_accuracy": 0.19218711925143442,
      "eval_num_tokens": 98574474.0,
      "eval_runtime": 20.4051,
      "eval_samples_per_second": 1650.027,
      "eval_steps_per_second": 206.272,
      "step": 45000
    },
    {
      "entropy": 5.6607412815094,
      "epoch": 4.4665541881699085,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031636699666440053,
      "loss": 4.9501,
      "mean_token_accuracy": 0.19050482362508775,
      "num_tokens": 98585415.0,
      "step": 45005
    },
    {
      "entropy": 5.732867193222046,
      "epoch": 4.467050416832076,
      "grad_norm": 1.375,
      "learning_rate": 0.0003163321686318007,
      "loss": 4.9838,
      "mean_token_accuracy": 0.1908270925283432,
      "num_tokens": 98596157.0,
      "step": 45010
    },
    {
      "entropy": 5.729904317855835,
      "epoch": 4.467546645494243,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003162973395742522,
      "loss": 5.072,
      "mean_token_accuracy": 0.1858454555273056,
      "num_tokens": 98607684.0,
      "step": 45015
    },
    {
      "entropy": 5.769064426422119,
      "epoch": 4.468042874156411,
      "grad_norm": 1.40625,
      "learning_rate": 0.00031626250949261866,
      "loss": 4.9766,
      "mean_token_accuracy": 0.18886830806732177,
      "num_tokens": 98617306.0,
      "step": 45020
    },
    {
      "entropy": 5.553436899185181,
      "epoch": 4.468539102818578,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031622767838776375,
      "loss": 4.7714,
      "mean_token_accuracy": 0.21186713129281998,
      "num_tokens": 98628232.0,
      "step": 45025
    },
    {
      "entropy": 5.652067708969116,
      "epoch": 4.469035331480747,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003161928462605513,
      "loss": 4.963,
      "mean_token_accuracy": 0.19044503271579744,
      "num_tokens": 98639382.0,
      "step": 45030
    },
    {
      "entropy": 5.699708795547485,
      "epoch": 4.469531560142914,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003161580131118449,
      "loss": 4.9833,
      "mean_token_accuracy": 0.1902136594057083,
      "num_tokens": 98651221.0,
      "step": 45035
    },
    {
      "entropy": 5.637936353683472,
      "epoch": 4.470027788805082,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031612317894250857,
      "loss": 4.9059,
      "mean_token_accuracy": 0.1916193202137947,
      "num_tokens": 98662809.0,
      "step": 45040
    },
    {
      "entropy": 5.732720804214478,
      "epoch": 4.470524017467249,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031608834375340607,
      "loss": 5.0001,
      "mean_token_accuracy": 0.18758404403924941,
      "num_tokens": 98673495.0,
      "step": 45045
    },
    {
      "entropy": 5.632236099243164,
      "epoch": 4.471020246129417,
      "grad_norm": 1.25,
      "learning_rate": 0.0003160535075454011,
      "loss": 4.9015,
      "mean_token_accuracy": 0.19188827723264695,
      "num_tokens": 98684699.0,
      "step": 45050
    },
    {
      "entropy": 5.7512016773223875,
      "epoch": 4.471516474791584,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031601867031935763,
      "loss": 4.9749,
      "mean_token_accuracy": 0.19026329815387727,
      "num_tokens": 98696968.0,
      "step": 45055
    },
    {
      "entropy": 5.695581293106079,
      "epoch": 4.472012703453752,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003159838320761396,
      "loss": 4.9643,
      "mean_token_accuracy": 0.1872683882713318,
      "num_tokens": 98706920.0,
      "step": 45060
    },
    {
      "entropy": 5.667397546768188,
      "epoch": 4.472508932115919,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003159489928166107,
      "loss": 4.8979,
      "mean_token_accuracy": 0.20189047753810882,
      "num_tokens": 98717349.0,
      "step": 45065
    },
    {
      "entropy": 5.72419490814209,
      "epoch": 4.4730051607780865,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003159141525416351,
      "loss": 4.9906,
      "mean_token_accuracy": 0.1970092013478279,
      "num_tokens": 98727805.0,
      "step": 45070
    },
    {
      "entropy": 5.69772367477417,
      "epoch": 4.473501389440254,
      "grad_norm": 1.4375,
      "learning_rate": 0.00031587931125207666,
      "loss": 4.9359,
      "mean_token_accuracy": 0.1942005231976509,
      "num_tokens": 98738101.0,
      "step": 45075
    },
    {
      "entropy": 5.629742956161499,
      "epoch": 4.4739976181024215,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031584446894879936,
      "loss": 4.8723,
      "mean_token_accuracy": 0.2099846288561821,
      "num_tokens": 98749082.0,
      "step": 45080
    },
    {
      "entropy": 5.660413026809692,
      "epoch": 4.474493846764589,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003158096256326674,
      "loss": 4.9209,
      "mean_token_accuracy": 0.2023952692747116,
      "num_tokens": 98759986.0,
      "step": 45085
    },
    {
      "entropy": 5.693157005310058,
      "epoch": 4.474990075426756,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031577478130454445,
      "loss": 4.993,
      "mean_token_accuracy": 0.19356928318738936,
      "num_tokens": 98769422.0,
      "step": 45090
    },
    {
      "entropy": 5.608494472503662,
      "epoch": 4.475486304088924,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031573993596529485,
      "loss": 4.8479,
      "mean_token_accuracy": 0.198573599755764,
      "num_tokens": 98779575.0,
      "step": 45095
    },
    {
      "entropy": 5.792976951599121,
      "epoch": 4.475982532751091,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003157050896157827,
      "loss": 5.0984,
      "mean_token_accuracy": 0.1887320891022682,
      "num_tokens": 98791951.0,
      "step": 45100
    },
    {
      "entropy": 5.724736166000366,
      "epoch": 4.476478761413259,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003156702422568719,
      "loss": 4.9727,
      "mean_token_accuracy": 0.18900718092918395,
      "num_tokens": 98801949.0,
      "step": 45105
    },
    {
      "entropy": 5.650322484970093,
      "epoch": 4.476974990075426,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003156353938894268,
      "loss": 5.0402,
      "mean_token_accuracy": 0.1922200247645378,
      "num_tokens": 98812892.0,
      "step": 45110
    },
    {
      "entropy": 5.6377458572387695,
      "epoch": 4.477471218737595,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031560054451431154,
      "loss": 4.9218,
      "mean_token_accuracy": 0.19170230478048325,
      "num_tokens": 98823685.0,
      "step": 45115
    },
    {
      "entropy": 5.7188036918640135,
      "epoch": 4.477967447399762,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003155656941323902,
      "loss": 4.9613,
      "mean_token_accuracy": 0.19930530339479446,
      "num_tokens": 98835601.0,
      "step": 45120
    },
    {
      "entropy": 5.748407506942749,
      "epoch": 4.47846367606193,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003155308427445271,
      "loss": 5.0243,
      "mean_token_accuracy": 0.18996721357107163,
      "num_tokens": 98845896.0,
      "step": 45125
    },
    {
      "entropy": 5.653792047500611,
      "epoch": 4.478959904724097,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031549599035158644,
      "loss": 4.9359,
      "mean_token_accuracy": 0.19884450882673263,
      "num_tokens": 98857969.0,
      "step": 45130
    },
    {
      "entropy": 5.676012849807739,
      "epoch": 4.479456133386265,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00031546113695443245,
      "loss": 4.9312,
      "mean_token_accuracy": 0.19287386685609817,
      "num_tokens": 98868406.0,
      "step": 45135
    },
    {
      "entropy": 5.606126213073731,
      "epoch": 4.479952362048432,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031542628255392947,
      "loss": 4.8596,
      "mean_token_accuracy": 0.20616815388202667,
      "num_tokens": 98880393.0,
      "step": 45140
    },
    {
      "entropy": 5.621017837524414,
      "epoch": 4.4804485907105995,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003153914271509418,
      "loss": 4.988,
      "mean_token_accuracy": 0.18897729367017746,
      "num_tokens": 98891650.0,
      "step": 45145
    },
    {
      "entropy": 5.642026662826538,
      "epoch": 4.480944819372767,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003153565707463338,
      "loss": 4.9356,
      "mean_token_accuracy": 0.19579531550407409,
      "num_tokens": 98902779.0,
      "step": 45150
    },
    {
      "entropy": 5.700755310058594,
      "epoch": 4.4814410480349345,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031532171334096965,
      "loss": 4.8637,
      "mean_token_accuracy": 0.19325446337461472,
      "num_tokens": 98912659.0,
      "step": 45155
    },
    {
      "entropy": 5.662809610366821,
      "epoch": 4.481937276697102,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003152868549357141,
      "loss": 4.935,
      "mean_token_accuracy": 0.1979398623108864,
      "num_tokens": 98922951.0,
      "step": 45160
    },
    {
      "entropy": 5.65388674736023,
      "epoch": 4.482433505359269,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003152519955314312,
      "loss": 4.8831,
      "mean_token_accuracy": 0.1967577889561653,
      "num_tokens": 98932850.0,
      "step": 45165
    },
    {
      "entropy": 5.6304810523986815,
      "epoch": 4.482929734021437,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031521713512898565,
      "loss": 4.9349,
      "mean_token_accuracy": 0.19287733286619185,
      "num_tokens": 98942765.0,
      "step": 45170
    },
    {
      "entropy": 5.6995092868804935,
      "epoch": 4.483425962683604,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031518227372924175,
      "loss": 5.0175,
      "mean_token_accuracy": 0.19661054015159607,
      "num_tokens": 98954745.0,
      "step": 45175
    },
    {
      "entropy": 5.721148109436035,
      "epoch": 4.483922191345772,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031514741133306407,
      "loss": 4.9093,
      "mean_token_accuracy": 0.19393569231033325,
      "num_tokens": 98966153.0,
      "step": 45180
    },
    {
      "entropy": 5.712535715103149,
      "epoch": 4.484418420007939,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003151125479413171,
      "loss": 4.9197,
      "mean_token_accuracy": 0.1940768539905548,
      "num_tokens": 98977148.0,
      "step": 45185
    },
    {
      "entropy": 5.661189460754395,
      "epoch": 4.484914648670107,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031507768355486536,
      "loss": 4.8995,
      "mean_token_accuracy": 0.2011092334985733,
      "num_tokens": 98987787.0,
      "step": 45190
    },
    {
      "entropy": 5.645605993270874,
      "epoch": 4.485410877332274,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031504281817457347,
      "loss": 5.0063,
      "mean_token_accuracy": 0.18657966703176498,
      "num_tokens": 99001108.0,
      "step": 45195
    },
    {
      "entropy": 5.62660231590271,
      "epoch": 4.485907105994443,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003150079518013059,
      "loss": 4.8835,
      "mean_token_accuracy": 0.20276396870613098,
      "num_tokens": 99011683.0,
      "step": 45200
    },
    {
      "entropy": 5.663097667694092,
      "epoch": 4.48640333465661,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031497308443592735,
      "loss": 4.9628,
      "mean_token_accuracy": 0.19678600132465363,
      "num_tokens": 99022999.0,
      "step": 45205
    },
    {
      "entropy": 5.717371225357056,
      "epoch": 4.486899563318778,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003149382160793026,
      "loss": 5.0467,
      "mean_token_accuracy": 0.18646065294742584,
      "num_tokens": 99033957.0,
      "step": 45210
    },
    {
      "entropy": 5.64700722694397,
      "epoch": 4.487395791980945,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031490334673229593,
      "loss": 4.8862,
      "mean_token_accuracy": 0.19661956280469894,
      "num_tokens": 99044004.0,
      "step": 45215
    },
    {
      "entropy": 5.6204580783844,
      "epoch": 4.4878920206431125,
      "grad_norm": 1.25,
      "learning_rate": 0.0003148684763957723,
      "loss": 4.8888,
      "mean_token_accuracy": 0.19359471946954726,
      "num_tokens": 99053555.0,
      "step": 45220
    },
    {
      "entropy": 5.651798391342163,
      "epoch": 4.48838824930528,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003148336050705964,
      "loss": 4.9815,
      "mean_token_accuracy": 0.19194646775722504,
      "num_tokens": 99063493.0,
      "step": 45225
    },
    {
      "entropy": 5.698086452484131,
      "epoch": 4.4888844779674475,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003147987327576329,
      "loss": 5.028,
      "mean_token_accuracy": 0.1856490805745125,
      "num_tokens": 99074843.0,
      "step": 45230
    },
    {
      "entropy": 5.719392013549805,
      "epoch": 4.489380706629615,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003147638594577467,
      "loss": 4.9536,
      "mean_token_accuracy": 0.1936052367091179,
      "num_tokens": 99086426.0,
      "step": 45235
    },
    {
      "entropy": 5.657132577896118,
      "epoch": 4.4898769352917824,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003147289851718024,
      "loss": 4.9582,
      "mean_token_accuracy": 0.19984053075313568,
      "num_tokens": 99098698.0,
      "step": 45240
    },
    {
      "entropy": 5.644944095611573,
      "epoch": 4.49037316395395,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003146941099006648,
      "loss": 4.8874,
      "mean_token_accuracy": 0.19562430530786515,
      "num_tokens": 99109343.0,
      "step": 45245
    },
    {
      "entropy": 5.670809316635132,
      "epoch": 4.490869392616117,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031465923364519896,
      "loss": 4.9152,
      "mean_token_accuracy": 0.20054084807634354,
      "num_tokens": 99120555.0,
      "step": 45250
    },
    {
      "entropy": 5.657161474227905,
      "epoch": 4.491365621278285,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003146243564062694,
      "loss": 4.8846,
      "mean_token_accuracy": 0.19981491267681123,
      "num_tokens": 99132204.0,
      "step": 45255
    },
    {
      "entropy": 5.72466082572937,
      "epoch": 4.491861849940452,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031458947818474125,
      "loss": 5.0242,
      "mean_token_accuracy": 0.19344326853752136,
      "num_tokens": 99144344.0,
      "step": 45260
    },
    {
      "entropy": 5.6644371509552,
      "epoch": 4.49235807860262,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031455459898147947,
      "loss": 4.9462,
      "mean_token_accuracy": 0.19202518165111543,
      "num_tokens": 99155247.0,
      "step": 45265
    },
    {
      "entropy": 5.599242496490478,
      "epoch": 4.492854307264787,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031451971879734866,
      "loss": 4.8964,
      "mean_token_accuracy": 0.2002279505133629,
      "num_tokens": 99165389.0,
      "step": 45270
    },
    {
      "entropy": 5.804380416870117,
      "epoch": 4.493350535926955,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003144848376332141,
      "loss": 5.0845,
      "mean_token_accuracy": 0.184085975587368,
      "num_tokens": 99176450.0,
      "step": 45275
    },
    {
      "entropy": 5.6813312530517575,
      "epoch": 4.493846764589122,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003144499554899407,
      "loss": 4.8567,
      "mean_token_accuracy": 0.2045857638120651,
      "num_tokens": 99187520.0,
      "step": 45280
    },
    {
      "entropy": 5.692455625534057,
      "epoch": 4.494342993251291,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031441507236839336,
      "loss": 4.9232,
      "mean_token_accuracy": 0.1945482924580574,
      "num_tokens": 99197749.0,
      "step": 45285
    },
    {
      "entropy": 5.643592643737793,
      "epoch": 4.494839221913458,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031438018826943716,
      "loss": 4.9337,
      "mean_token_accuracy": 0.1949060782790184,
      "num_tokens": 99209075.0,
      "step": 45290
    },
    {
      "entropy": 5.660811853408814,
      "epoch": 4.495335450575626,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031434530319393713,
      "loss": 4.9911,
      "mean_token_accuracy": 0.19387850314378738,
      "num_tokens": 99219582.0,
      "step": 45295
    },
    {
      "entropy": 5.639928722381592,
      "epoch": 4.495831679237793,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003143104171427583,
      "loss": 4.8601,
      "mean_token_accuracy": 0.19494989067316054,
      "num_tokens": 99230282.0,
      "step": 45300
    },
    {
      "entropy": 5.675172233581543,
      "epoch": 4.4963279078999605,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031427553011676595,
      "loss": 4.9949,
      "mean_token_accuracy": 0.201613213121891,
      "num_tokens": 99240333.0,
      "step": 45305
    },
    {
      "entropy": 5.686156797409057,
      "epoch": 4.496824136562128,
      "grad_norm": 1.296875,
      "learning_rate": 0.000314240642116825,
      "loss": 5.0031,
      "mean_token_accuracy": 0.18980121314525605,
      "num_tokens": 99250658.0,
      "step": 45310
    },
    {
      "entropy": 5.69260721206665,
      "epoch": 4.4973203652242955,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003142057531438007,
      "loss": 4.976,
      "mean_token_accuracy": 0.18823220878839492,
      "num_tokens": 99261236.0,
      "step": 45315
    },
    {
      "entropy": 5.715425825119018,
      "epoch": 4.497816593886463,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031417086319855827,
      "loss": 5.0337,
      "mean_token_accuracy": 0.19024209082126617,
      "num_tokens": 99273652.0,
      "step": 45320
    },
    {
      "entropy": 5.666498470306396,
      "epoch": 4.49831282254863,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003141359722819628,
      "loss": 4.925,
      "mean_token_accuracy": 0.1905510500073433,
      "num_tokens": 99285686.0,
      "step": 45325
    },
    {
      "entropy": 5.668379020690918,
      "epoch": 4.498809051210798,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003141010803948795,
      "loss": 4.9156,
      "mean_token_accuracy": 0.19674271047115327,
      "num_tokens": 99295345.0,
      "step": 45330
    },
    {
      "entropy": 5.634967517852783,
      "epoch": 4.499305279872965,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003140661875381737,
      "loss": 4.8788,
      "mean_token_accuracy": 0.19563254117965698,
      "num_tokens": 99305119.0,
      "step": 45335
    },
    {
      "entropy": 5.611312007904052,
      "epoch": 4.499801508535133,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031403129371271066,
      "loss": 4.9149,
      "mean_token_accuracy": 0.20086844265460968,
      "num_tokens": 99315507.0,
      "step": 45340
    },
    {
      "entropy": 5.661655902862549,
      "epoch": 4.5002977371973,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031399639891935554,
      "loss": 4.9431,
      "mean_token_accuracy": 0.19714855700731276,
      "num_tokens": 99325903.0,
      "step": 45345
    },
    {
      "entropy": 5.69070987701416,
      "epoch": 4.500793965859468,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003139615031589739,
      "loss": 4.9937,
      "mean_token_accuracy": 0.18453333675861358,
      "num_tokens": 99336957.0,
      "step": 45350
    },
    {
      "entropy": 5.658130836486817,
      "epoch": 4.501290194521635,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003139266064324308,
      "loss": 4.8992,
      "mean_token_accuracy": 0.1978968635201454,
      "num_tokens": 99348032.0,
      "step": 45355
    },
    {
      "entropy": 5.662095975875855,
      "epoch": 4.501786423183803,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031389170874059183,
      "loss": 4.921,
      "mean_token_accuracy": 0.19328935742378234,
      "num_tokens": 99358653.0,
      "step": 45360
    },
    {
      "entropy": 5.661599636077881,
      "epoch": 4.50228265184597,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031385681008432227,
      "loss": 4.9779,
      "mean_token_accuracy": 0.19163827747106552,
      "num_tokens": 99368464.0,
      "step": 45365
    },
    {
      "entropy": 5.706456851959229,
      "epoch": 4.502778880508139,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031382191046448754,
      "loss": 4.9945,
      "mean_token_accuracy": 0.18474865555763245,
      "num_tokens": 99379941.0,
      "step": 45370
    },
    {
      "entropy": 5.690834093093872,
      "epoch": 4.503275109170306,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031378700988195305,
      "loss": 4.9609,
      "mean_token_accuracy": 0.18803133964538574,
      "num_tokens": 99390092.0,
      "step": 45375
    },
    {
      "entropy": 5.67690806388855,
      "epoch": 4.5037713378324735,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031375210833758433,
      "loss": 4.978,
      "mean_token_accuracy": 0.1819849506020546,
      "num_tokens": 99401171.0,
      "step": 45380
    },
    {
      "entropy": 5.714222955703735,
      "epoch": 4.504267566494641,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003137172058322468,
      "loss": 5.022,
      "mean_token_accuracy": 0.18944200724363328,
      "num_tokens": 99412333.0,
      "step": 45385
    },
    {
      "entropy": 5.758955717086792,
      "epoch": 4.5047637951568085,
      "grad_norm": 1.390625,
      "learning_rate": 0.000313682302366806,
      "loss": 4.9844,
      "mean_token_accuracy": 0.19483485370874404,
      "num_tokens": 99422719.0,
      "step": 45390
    },
    {
      "entropy": 5.70520167350769,
      "epoch": 4.505260023818976,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031364739794212756,
      "loss": 5.0066,
      "mean_token_accuracy": 0.1916027694940567,
      "num_tokens": 99433007.0,
      "step": 45395
    },
    {
      "entropy": 5.700666904449463,
      "epoch": 4.505756252481143,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003136124925590768,
      "loss": 4.9789,
      "mean_token_accuracy": 0.1950517013669014,
      "num_tokens": 99443875.0,
      "step": 45400
    },
    {
      "entropy": 5.742799568176269,
      "epoch": 4.506252481143311,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003135775862185194,
      "loss": 4.9752,
      "mean_token_accuracy": 0.19950313121080399,
      "num_tokens": 99454493.0,
      "step": 45405
    },
    {
      "entropy": 5.725718307495117,
      "epoch": 4.506748709805478,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003135426789213211,
      "loss": 5.0199,
      "mean_token_accuracy": 0.19457159489393233,
      "num_tokens": 99464850.0,
      "step": 45410
    },
    {
      "entropy": 5.685205602645874,
      "epoch": 4.507244938467646,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031350777066834737,
      "loss": 4.928,
      "mean_token_accuracy": 0.19795545935630798,
      "num_tokens": 99474839.0,
      "step": 45415
    },
    {
      "entropy": 5.659351634979248,
      "epoch": 4.507741167129813,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003134728614604638,
      "loss": 4.9911,
      "mean_token_accuracy": 0.1860235407948494,
      "num_tokens": 99485929.0,
      "step": 45420
    },
    {
      "entropy": 5.686732625961303,
      "epoch": 4.508237395791981,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003134379512985364,
      "loss": 4.9306,
      "mean_token_accuracy": 0.189228156208992,
      "num_tokens": 99496630.0,
      "step": 45425
    },
    {
      "entropy": 5.708027124404907,
      "epoch": 4.508733624454148,
      "grad_norm": 1.375,
      "learning_rate": 0.0003134030401834305,
      "loss": 4.984,
      "mean_token_accuracy": 0.19995985329151153,
      "num_tokens": 99507810.0,
      "step": 45430
    },
    {
      "entropy": 5.680704355239868,
      "epoch": 4.509229853116316,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003133681281160119,
      "loss": 4.9425,
      "mean_token_accuracy": 0.19326765239238738,
      "num_tokens": 99518395.0,
      "step": 45435
    },
    {
      "entropy": 5.670649194717408,
      "epoch": 4.509726081778483,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003133332150971465,
      "loss": 4.917,
      "mean_token_accuracy": 0.1947978600859642,
      "num_tokens": 99528895.0,
      "step": 45440
    },
    {
      "entropy": 5.7307545185089115,
      "epoch": 4.510222310440651,
      "grad_norm": 1.125,
      "learning_rate": 0.0003132983011276999,
      "loss": 5.0348,
      "mean_token_accuracy": 0.19094078689813615,
      "num_tokens": 99539226.0,
      "step": 45445
    },
    {
      "entropy": 5.712174558639527,
      "epoch": 4.510718539102818,
      "grad_norm": 1.125,
      "learning_rate": 0.00031326338620853795,
      "loss": 4.9662,
      "mean_token_accuracy": 0.1921815425157547,
      "num_tokens": 99551058.0,
      "step": 45450
    },
    {
      "entropy": 5.714572811126709,
      "epoch": 4.5112147677649865,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003132284703405266,
      "loss": 5.007,
      "mean_token_accuracy": 0.19148557931184768,
      "num_tokens": 99562224.0,
      "step": 45455
    },
    {
      "entropy": 5.827270603179931,
      "epoch": 4.511710996427154,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031319355352453146,
      "loss": 5.0867,
      "mean_token_accuracy": 0.19098335802555083,
      "num_tokens": 99572780.0,
      "step": 45460
    },
    {
      "entropy": 5.731805753707886,
      "epoch": 4.5122072250893215,
      "grad_norm": 1.25,
      "learning_rate": 0.00031315863576141846,
      "loss": 4.8958,
      "mean_token_accuracy": 0.19929171353578568,
      "num_tokens": 99583263.0,
      "step": 45465
    },
    {
      "entropy": 5.729628562927246,
      "epoch": 4.512703453751489,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031312371705205353,
      "loss": 5.0328,
      "mean_token_accuracy": 0.19029954075813293,
      "num_tokens": 99595587.0,
      "step": 45470
    },
    {
      "entropy": 5.7382148742675785,
      "epoch": 4.513199682413656,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031308879739730264,
      "loss": 4.9961,
      "mean_token_accuracy": 0.19839489907026292,
      "num_tokens": 99605888.0,
      "step": 45475
    },
    {
      "entropy": 5.634766578674316,
      "epoch": 4.513695911075824,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003130538767980316,
      "loss": 4.941,
      "mean_token_accuracy": 0.1895659953355789,
      "num_tokens": 99617070.0,
      "step": 45480
    },
    {
      "entropy": 5.702187156677246,
      "epoch": 4.514192139737991,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003130189552551064,
      "loss": 4.9529,
      "mean_token_accuracy": 0.20003502517938615,
      "num_tokens": 99628834.0,
      "step": 45485
    },
    {
      "entropy": 5.7800984382629395,
      "epoch": 4.514688368400159,
      "grad_norm": 1.125,
      "learning_rate": 0.00031298403276939303,
      "loss": 5.1052,
      "mean_token_accuracy": 0.1843557819724083,
      "num_tokens": 99640411.0,
      "step": 45490
    },
    {
      "entropy": 5.679244804382324,
      "epoch": 4.515184597062326,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003129491093417575,
      "loss": 4.9395,
      "mean_token_accuracy": 0.19247380644083023,
      "num_tokens": 99650958.0,
      "step": 45495
    },
    {
      "entropy": 5.769410133361816,
      "epoch": 4.515680825724494,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031291418497306586,
      "loss": 5.0611,
      "mean_token_accuracy": 0.17940194755792618,
      "num_tokens": 99661753.0,
      "step": 45500
    },
    {
      "entropy": 5.766181087493896,
      "epoch": 4.516177054386661,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031287925966418405,
      "loss": 5.0228,
      "mean_token_accuracy": 0.19301580339670182,
      "num_tokens": 99675867.0,
      "step": 45505
    },
    {
      "entropy": 5.6726545810699465,
      "epoch": 4.516673283048829,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031284433341597837,
      "loss": 4.968,
      "mean_token_accuracy": 0.1960214301943779,
      "num_tokens": 99687670.0,
      "step": 45510
    },
    {
      "entropy": 5.666361379623413,
      "epoch": 4.517169511710996,
      "grad_norm": 1.25,
      "learning_rate": 0.00031280940622931467,
      "loss": 4.9904,
      "mean_token_accuracy": 0.18767300695180894,
      "num_tokens": 99699993.0,
      "step": 45515
    },
    {
      "entropy": 5.757976627349853,
      "epoch": 4.517665740373164,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003127744781050591,
      "loss": 5.0355,
      "mean_token_accuracy": 0.19037363678216934,
      "num_tokens": 99711062.0,
      "step": 45520
    },
    {
      "entropy": 5.703761291503906,
      "epoch": 4.518161969035331,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000312739549044078,
      "loss": 4.951,
      "mean_token_accuracy": 0.1941681385040283,
      "num_tokens": 99722378.0,
      "step": 45525
    },
    {
      "entropy": 5.632431221008301,
      "epoch": 4.518658197697499,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031270461904723743,
      "loss": 4.8505,
      "mean_token_accuracy": 0.19940924644470215,
      "num_tokens": 99732050.0,
      "step": 45530
    },
    {
      "entropy": 5.6769246578216555,
      "epoch": 4.519154426359666,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003126696881154034,
      "loss": 4.9967,
      "mean_token_accuracy": 0.1861405625939369,
      "num_tokens": 99743008.0,
      "step": 45535
    },
    {
      "entropy": 5.684788036346435,
      "epoch": 4.5196506550218345,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031263475624944236,
      "loss": 4.9465,
      "mean_token_accuracy": 0.19956525266170502,
      "num_tokens": 99753983.0,
      "step": 45540
    },
    {
      "entropy": 5.667759943008423,
      "epoch": 4.520146883684002,
      "grad_norm": 1.375,
      "learning_rate": 0.00031259982345022055,
      "loss": 4.9457,
      "mean_token_accuracy": 0.1952275350689888,
      "num_tokens": 99764203.0,
      "step": 45545
    },
    {
      "entropy": 5.644575309753418,
      "epoch": 4.5206431123461694,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003125648897186039,
      "loss": 4.9214,
      "mean_token_accuracy": 0.1987530305981636,
      "num_tokens": 99774592.0,
      "step": 45550
    },
    {
      "entropy": 5.713446807861328,
      "epoch": 4.521139341008337,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031252995505545917,
      "loss": 4.9809,
      "mean_token_accuracy": 0.1892118901014328,
      "num_tokens": 99785608.0,
      "step": 45555
    },
    {
      "entropy": 5.6095365524292,
      "epoch": 4.521635569670504,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003124950194616523,
      "loss": 4.8697,
      "mean_token_accuracy": 0.19850269109010696,
      "num_tokens": 99796728.0,
      "step": 45560
    },
    {
      "entropy": 5.670599365234375,
      "epoch": 4.522131798332672,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031246008293804974,
      "loss": 4.9339,
      "mean_token_accuracy": 0.19584375470876694,
      "num_tokens": 99807029.0,
      "step": 45565
    },
    {
      "entropy": 5.706264305114746,
      "epoch": 4.522628026994839,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003124251454855179,
      "loss": 4.971,
      "mean_token_accuracy": 0.19838771373033523,
      "num_tokens": 99816864.0,
      "step": 45570
    },
    {
      "entropy": 5.783929872512817,
      "epoch": 4.523124255657007,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003123902071049231,
      "loss": 5.0616,
      "mean_token_accuracy": 0.18664319515228273,
      "num_tokens": 99826627.0,
      "step": 45575
    },
    {
      "entropy": 5.708938884735107,
      "epoch": 4.523620484319174,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003123552677971317,
      "loss": 4.9881,
      "mean_token_accuracy": 0.19221688210964202,
      "num_tokens": 99837352.0,
      "step": 45580
    },
    {
      "entropy": 5.743427419662476,
      "epoch": 4.524116712981342,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031232032756301013,
      "loss": 5.0302,
      "mean_token_accuracy": 0.19727648794651031,
      "num_tokens": 99848449.0,
      "step": 45585
    },
    {
      "entropy": 5.635420083999634,
      "epoch": 4.524612941643509,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003122853864034249,
      "loss": 4.9345,
      "mean_token_accuracy": 0.1922683522105217,
      "num_tokens": 99859935.0,
      "step": 45590
    },
    {
      "entropy": 5.631267452239991,
      "epoch": 4.525109170305677,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031225044431924245,
      "loss": 4.9245,
      "mean_token_accuracy": 0.1997286781668663,
      "num_tokens": 99870818.0,
      "step": 45595
    },
    {
      "entropy": 5.725440120697021,
      "epoch": 4.525605398967844,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031221550131132916,
      "loss": 4.9428,
      "mean_token_accuracy": 0.1958521693944931,
      "num_tokens": 99881608.0,
      "step": 45600
    },
    {
      "entropy": 5.671808958053589,
      "epoch": 4.526101627630012,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003121805573805517,
      "loss": 4.9223,
      "mean_token_accuracy": 0.19899849444627762,
      "num_tokens": 99891108.0,
      "step": 45605
    },
    {
      "entropy": 5.640247535705567,
      "epoch": 4.526597856292179,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031214561252777655,
      "loss": 4.9227,
      "mean_token_accuracy": 0.19377525448799132,
      "num_tokens": 99901811.0,
      "step": 45610
    },
    {
      "entropy": 5.606838893890381,
      "epoch": 4.527094084954347,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031211066675387015,
      "loss": 4.8798,
      "mean_token_accuracy": 0.1981983467936516,
      "num_tokens": 99914514.0,
      "step": 45615
    },
    {
      "entropy": 5.672073793411255,
      "epoch": 4.527590313616514,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003120757200596993,
      "loss": 4.9293,
      "mean_token_accuracy": 0.1948674201965332,
      "num_tokens": 99925180.0,
      "step": 45620
    },
    {
      "entropy": 5.700728082656861,
      "epoch": 4.5280865422786825,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003120407724461304,
      "loss": 4.9647,
      "mean_token_accuracy": 0.18961406946182252,
      "num_tokens": 99935963.0,
      "step": 45625
    },
    {
      "entropy": 5.690195941925049,
      "epoch": 4.528582770940849,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003120058239140302,
      "loss": 5.0445,
      "mean_token_accuracy": 0.19244494438171386,
      "num_tokens": 99946913.0,
      "step": 45630
    },
    {
      "entropy": 5.764819383621216,
      "epoch": 4.529078999603017,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031197087446426533,
      "loss": 5.0401,
      "mean_token_accuracy": 0.1905534163117409,
      "num_tokens": 99959779.0,
      "step": 45635
    },
    {
      "entropy": 5.691868734359741,
      "epoch": 4.529575228265185,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003119359240977024,
      "loss": 4.8943,
      "mean_token_accuracy": 0.19707345068454743,
      "num_tokens": 99969818.0,
      "step": 45640
    },
    {
      "entropy": 5.645576763153076,
      "epoch": 4.530071456927352,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031190097281520805,
      "loss": 4.9196,
      "mean_token_accuracy": 0.1914479613304138,
      "num_tokens": 99980092.0,
      "step": 45645
    },
    {
      "entropy": 5.729344892501831,
      "epoch": 4.53056768558952,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031186602061764924,
      "loss": 5.0099,
      "mean_token_accuracy": 0.19757971465587615,
      "num_tokens": 99992692.0,
      "step": 45650
    },
    {
      "entropy": 5.640790796279907,
      "epoch": 4.531063914251687,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003118310675058925,
      "loss": 4.8853,
      "mean_token_accuracy": 0.19570524990558624,
      "num_tokens": 100004058.0,
      "step": 45655
    },
    {
      "entropy": 5.725592517852784,
      "epoch": 4.531560142913855,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003117961134808047,
      "loss": 4.9295,
      "mean_token_accuracy": 0.2001853883266449,
      "num_tokens": 100013741.0,
      "step": 45660
    },
    {
      "entropy": 5.648250913619995,
      "epoch": 4.532056371576022,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003117611585432525,
      "loss": 4.9247,
      "mean_token_accuracy": 0.19670977890491487,
      "num_tokens": 100024485.0,
      "step": 45665
    },
    {
      "entropy": 5.721880006790161,
      "epoch": 4.53255260023819,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003117262026941028,
      "loss": 5.0548,
      "mean_token_accuracy": 0.18582722842693328,
      "num_tokens": 100035180.0,
      "step": 45670
    },
    {
      "entropy": 5.7706187725067135,
      "epoch": 4.533048828900357,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031169124593422243,
      "loss": 5.0003,
      "mean_token_accuracy": 0.19034525752067566,
      "num_tokens": 100046471.0,
      "step": 45675
    },
    {
      "entropy": 5.646294116973877,
      "epoch": 4.533545057562525,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003116562882644782,
      "loss": 4.8287,
      "mean_token_accuracy": 0.20401245653629302,
      "num_tokens": 100057962.0,
      "step": 45680
    },
    {
      "entropy": 5.658492803573608,
      "epoch": 4.534041286224692,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031162132968573703,
      "loss": 4.9217,
      "mean_token_accuracy": 0.19123203456401824,
      "num_tokens": 100068657.0,
      "step": 45685
    },
    {
      "entropy": 5.655629396438599,
      "epoch": 4.53453751488686,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031158637019886574,
      "loss": 4.9604,
      "mean_token_accuracy": 0.19678430408239364,
      "num_tokens": 100080398.0,
      "step": 45690
    },
    {
      "entropy": 5.679422521591187,
      "epoch": 4.535033743549027,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003115514098047314,
      "loss": 4.9695,
      "mean_token_accuracy": 0.18856040835380555,
      "num_tokens": 100090972.0,
      "step": 45695
    },
    {
      "entropy": 5.696216058731079,
      "epoch": 4.535529972211195,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003115164485042008,
      "loss": 4.956,
      "mean_token_accuracy": 0.19644148349761964,
      "num_tokens": 100102474.0,
      "step": 45700
    },
    {
      "entropy": 5.667182588577271,
      "epoch": 4.536026200873362,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000311481486298141,
      "loss": 4.9841,
      "mean_token_accuracy": 0.19383835643529893,
      "num_tokens": 100113912.0,
      "step": 45705
    },
    {
      "entropy": 5.661350917816162,
      "epoch": 4.53652242953553,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003114465231874189,
      "loss": 4.9364,
      "mean_token_accuracy": 0.1964743986725807,
      "num_tokens": 100125038.0,
      "step": 45710
    },
    {
      "entropy": 5.666588068008423,
      "epoch": 4.537018658197697,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031141155917290155,
      "loss": 4.9058,
      "mean_token_accuracy": 0.20224276483058928,
      "num_tokens": 100135060.0,
      "step": 45715
    },
    {
      "entropy": 5.6648030281066895,
      "epoch": 4.537514886859865,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031137659425545613,
      "loss": 4.9269,
      "mean_token_accuracy": 0.191628997027874,
      "num_tokens": 100146617.0,
      "step": 45720
    },
    {
      "entropy": 5.651276159286499,
      "epoch": 4.538011115522033,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031134162843594936,
      "loss": 4.9215,
      "mean_token_accuracy": 0.19857758581638335,
      "num_tokens": 100157299.0,
      "step": 45725
    },
    {
      "entropy": 5.805001258850098,
      "epoch": 4.5385073441842,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031130666171524864,
      "loss": 5.0467,
      "mean_token_accuracy": 0.20350815504789352,
      "num_tokens": 100169218.0,
      "step": 45730
    },
    {
      "entropy": 5.65596752166748,
      "epoch": 4.539003572846368,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003112716940942209,
      "loss": 4.8755,
      "mean_token_accuracy": 0.19255854636430741,
      "num_tokens": 100178643.0,
      "step": 45735
    },
    {
      "entropy": 5.686674976348877,
      "epoch": 4.539499801508535,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003112367255737333,
      "loss": 4.9603,
      "mean_token_accuracy": 0.19474877566099166,
      "num_tokens": 100189777.0,
      "step": 45740
    },
    {
      "entropy": 5.6908159255981445,
      "epoch": 4.539996030170703,
      "grad_norm": 1.171875,
      "learning_rate": 0.000311201756154653,
      "loss": 4.9235,
      "mean_token_accuracy": 0.1933737128973007,
      "num_tokens": 100200855.0,
      "step": 45745
    },
    {
      "entropy": 5.751322984695435,
      "epoch": 4.54049225883287,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031116678583784725,
      "loss": 5.1088,
      "mean_token_accuracy": 0.18518935292959213,
      "num_tokens": 100212104.0,
      "step": 45750
    },
    {
      "entropy": 5.658909893035888,
      "epoch": 4.540988487495038,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031113181462418297,
      "loss": 4.9068,
      "mean_token_accuracy": 0.1925067663192749,
      "num_tokens": 100222104.0,
      "step": 45755
    },
    {
      "entropy": 5.69227819442749,
      "epoch": 4.541484716157205,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003110968425145276,
      "loss": 4.9686,
      "mean_token_accuracy": 0.19447325617074968,
      "num_tokens": 100232509.0,
      "step": 45760
    },
    {
      "entropy": 5.68696928024292,
      "epoch": 4.541980944819373,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031106186950974845,
      "loss": 5.0035,
      "mean_token_accuracy": 0.19109409004449845,
      "num_tokens": 100245099.0,
      "step": 45765
    },
    {
      "entropy": 5.732085943222046,
      "epoch": 4.54247717348154,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003110268956107125,
      "loss": 4.9726,
      "mean_token_accuracy": 0.1915372744202614,
      "num_tokens": 100256183.0,
      "step": 45770
    },
    {
      "entropy": 5.780483865737915,
      "epoch": 4.542973402143708,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003109919208182873,
      "loss": 5.1186,
      "mean_token_accuracy": 0.18317648470401765,
      "num_tokens": 100268795.0,
      "step": 45775
    },
    {
      "entropy": 5.5864866256713865,
      "epoch": 4.543469630805875,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031095694513333985,
      "loss": 4.8082,
      "mean_token_accuracy": 0.19749761670827864,
      "num_tokens": 100279827.0,
      "step": 45780
    },
    {
      "entropy": 5.608500576019287,
      "epoch": 4.5439658594680425,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031092196855673766,
      "loss": 4.8885,
      "mean_token_accuracy": 0.20072070509195328,
      "num_tokens": 100291711.0,
      "step": 45785
    },
    {
      "entropy": 5.684400606155395,
      "epoch": 4.54446208813021,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003108869910893482,
      "loss": 4.9172,
      "mean_token_accuracy": 0.1932823807001114,
      "num_tokens": 100304070.0,
      "step": 45790
    },
    {
      "entropy": 5.666882658004761,
      "epoch": 4.544958316792378,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003108520127320386,
      "loss": 4.927,
      "mean_token_accuracy": 0.2043797567486763,
      "num_tokens": 100315834.0,
      "step": 45795
    },
    {
      "entropy": 5.646906995773316,
      "epoch": 4.545454545454545,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031081703348567627,
      "loss": 4.9373,
      "mean_token_accuracy": 0.20314870923757553,
      "num_tokens": 100327135.0,
      "step": 45800
    },
    {
      "entropy": 5.630507135391236,
      "epoch": 4.545950774116713,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003107820533511288,
      "loss": 4.8595,
      "mean_token_accuracy": 0.2000133290886879,
      "num_tokens": 100338018.0,
      "step": 45805
    },
    {
      "entropy": 5.667078065872192,
      "epoch": 4.546447002778881,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031074707232926345,
      "loss": 4.874,
      "mean_token_accuracy": 0.2010720804333687,
      "num_tokens": 100347864.0,
      "step": 45810
    },
    {
      "entropy": 5.674705457687378,
      "epoch": 4.546943231441048,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003107120904209478,
      "loss": 4.952,
      "mean_token_accuracy": 0.1899154543876648,
      "num_tokens": 100359242.0,
      "step": 45815
    },
    {
      "entropy": 5.715176820755005,
      "epoch": 4.547439460103216,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003106771076270492,
      "loss": 5.0308,
      "mean_token_accuracy": 0.1849014788866043,
      "num_tokens": 100369533.0,
      "step": 45820
    },
    {
      "entropy": 5.663608598709106,
      "epoch": 4.547935688765383,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003106421239484352,
      "loss": 4.9086,
      "mean_token_accuracy": 0.19618528038263322,
      "num_tokens": 100380955.0,
      "step": 45825
    },
    {
      "entropy": 5.66740403175354,
      "epoch": 4.548431917427551,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003106071393859734,
      "loss": 4.9005,
      "mean_token_accuracy": 0.19451459646224975,
      "num_tokens": 100392498.0,
      "step": 45830
    },
    {
      "entropy": 5.76118540763855,
      "epoch": 4.548928146089718,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003105721539405312,
      "loss": 5.0019,
      "mean_token_accuracy": 0.19499248117208481,
      "num_tokens": 100402493.0,
      "step": 45835
    },
    {
      "entropy": 5.72054352760315,
      "epoch": 4.549424374751886,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031053716761297623,
      "loss": 4.9907,
      "mean_token_accuracy": 0.19557929039001465,
      "num_tokens": 100415179.0,
      "step": 45840
    },
    {
      "entropy": 5.668190622329712,
      "epoch": 4.549920603414053,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003105021804041761,
      "loss": 4.9877,
      "mean_token_accuracy": 0.19050547629594802,
      "num_tokens": 100426459.0,
      "step": 45845
    },
    {
      "entropy": 5.608189058303833,
      "epoch": 4.550416832076221,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031046719231499843,
      "loss": 4.896,
      "mean_token_accuracy": 0.1942111760377884,
      "num_tokens": 100438250.0,
      "step": 45850
    },
    {
      "entropy": 5.7227945804595945,
      "epoch": 4.550913060738388,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003104322033463108,
      "loss": 5.0444,
      "mean_token_accuracy": 0.19249965399503707,
      "num_tokens": 100449761.0,
      "step": 45855
    },
    {
      "entropy": 5.696386051177979,
      "epoch": 4.551409289400556,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003103972134989808,
      "loss": 4.9858,
      "mean_token_accuracy": 0.18971737921237947,
      "num_tokens": 100461323.0,
      "step": 45860
    },
    {
      "entropy": 5.749632787704468,
      "epoch": 4.551905518062723,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031036222277387634,
      "loss": 4.9941,
      "mean_token_accuracy": 0.19411333054304122,
      "num_tokens": 100472557.0,
      "step": 45865
    },
    {
      "entropy": 5.762730932235717,
      "epoch": 4.5524017467248905,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003103272311718649,
      "loss": 5.0169,
      "mean_token_accuracy": 0.18981083184480668,
      "num_tokens": 100483494.0,
      "step": 45870
    },
    {
      "entropy": 5.714646816253662,
      "epoch": 4.552897975387058,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003102922386938141,
      "loss": 4.9768,
      "mean_token_accuracy": 0.19543597847223282,
      "num_tokens": 100495199.0,
      "step": 45875
    },
    {
      "entropy": 5.669924068450928,
      "epoch": 4.553394204049226,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031025724534059195,
      "loss": 5.0047,
      "mean_token_accuracy": 0.19677600860595704,
      "num_tokens": 100507334.0,
      "step": 45880
    },
    {
      "entropy": 5.6371657371521,
      "epoch": 4.553890432711393,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031022225111306613,
      "loss": 4.951,
      "mean_token_accuracy": 0.19093509763479233,
      "num_tokens": 100518969.0,
      "step": 45885
    },
    {
      "entropy": 5.728603982925415,
      "epoch": 4.554386661373561,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003101872560121043,
      "loss": 4.9931,
      "mean_token_accuracy": 0.19234201163053513,
      "num_tokens": 100530031.0,
      "step": 45890
    },
    {
      "entropy": 5.720764827728272,
      "epoch": 4.554882890035729,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031015226003857434,
      "loss": 4.9469,
      "mean_token_accuracy": 0.19102080464363097,
      "num_tokens": 100540051.0,
      "step": 45895
    },
    {
      "entropy": 5.722499799728394,
      "epoch": 4.555379118697896,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003101172631933441,
      "loss": 4.9652,
      "mean_token_accuracy": 0.19015772938728331,
      "num_tokens": 100551552.0,
      "step": 45900
    },
    {
      "entropy": 5.618195199966431,
      "epoch": 4.555875347360064,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003100822654772814,
      "loss": 4.9731,
      "mean_token_accuracy": 0.18461068868637084,
      "num_tokens": 100562080.0,
      "step": 45905
    },
    {
      "entropy": 5.729573917388916,
      "epoch": 4.556371576022231,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031004726689125414,
      "loss": 5.0059,
      "mean_token_accuracy": 0.19392640590667726,
      "num_tokens": 100573860.0,
      "step": 45910
    },
    {
      "entropy": 5.69210205078125,
      "epoch": 4.556867804684399,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003100122674361302,
      "loss": 4.958,
      "mean_token_accuracy": 0.19707588851451874,
      "num_tokens": 100586579.0,
      "step": 45915
    },
    {
      "entropy": 5.76102728843689,
      "epoch": 4.557364033346566,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003099772671127774,
      "loss": 5.0259,
      "mean_token_accuracy": 0.19707594066858292,
      "num_tokens": 100599678.0,
      "step": 45920
    },
    {
      "entropy": 5.797706508636475,
      "epoch": 4.557860262008734,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003099422659220638,
      "loss": 5.0252,
      "mean_token_accuracy": 0.19000200778245926,
      "num_tokens": 100608861.0,
      "step": 45925
    },
    {
      "entropy": 5.7198402881622314,
      "epoch": 4.558356490670901,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003099072638648573,
      "loss": 4.9223,
      "mean_token_accuracy": 0.1959291622042656,
      "num_tokens": 100621276.0,
      "step": 45930
    },
    {
      "entropy": 5.67783842086792,
      "epoch": 4.558852719333069,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003098722609420258,
      "loss": 4.9681,
      "mean_token_accuracy": 0.19711932837963103,
      "num_tokens": 100632043.0,
      "step": 45935
    },
    {
      "entropy": 5.740010690689087,
      "epoch": 4.559348947995236,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003098372571544374,
      "loss": 5.0232,
      "mean_token_accuracy": 0.19329765439033508,
      "num_tokens": 100643509.0,
      "step": 45940
    },
    {
      "entropy": 5.66028618812561,
      "epoch": 4.5598451766574035,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003098022525029602,
      "loss": 4.9719,
      "mean_token_accuracy": 0.18454715758562087,
      "num_tokens": 100654298.0,
      "step": 45945
    },
    {
      "entropy": 5.7062664985656735,
      "epoch": 4.560341405319571,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000309767246988462,
      "loss": 4.9649,
      "mean_token_accuracy": 0.1949310138821602,
      "num_tokens": 100665861.0,
      "step": 45950
    },
    {
      "entropy": 5.720042133331299,
      "epoch": 4.5608376339817385,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030973224061181103,
      "loss": 5.0197,
      "mean_token_accuracy": 0.18909504115581513,
      "num_tokens": 100676583.0,
      "step": 45955
    },
    {
      "entropy": 5.621476745605468,
      "epoch": 4.561333862643906,
      "grad_norm": 1.25,
      "learning_rate": 0.00030969723337387535,
      "loss": 4.9161,
      "mean_token_accuracy": 0.20013623535633088,
      "num_tokens": 100687563.0,
      "step": 45960
    },
    {
      "entropy": 5.687926340103149,
      "epoch": 4.561830091306074,
      "grad_norm": 1.125,
      "learning_rate": 0.000309662225275523,
      "loss": 4.9758,
      "mean_token_accuracy": 0.2025429219007492,
      "num_tokens": 100698148.0,
      "step": 45965
    },
    {
      "entropy": 5.760768747329712,
      "epoch": 4.562326319968241,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030962721631762227,
      "loss": 4.986,
      "mean_token_accuracy": 0.18377840518951416,
      "num_tokens": 100708819.0,
      "step": 45970
    },
    {
      "entropy": 5.732872104644775,
      "epoch": 4.562822548630409,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030959220650104107,
      "loss": 4.9794,
      "mean_token_accuracy": 0.19131003469228744,
      "num_tokens": 100719048.0,
      "step": 45975
    },
    {
      "entropy": 5.690606546401978,
      "epoch": 4.563318777292577,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003095571958266477,
      "loss": 4.9363,
      "mean_token_accuracy": 0.18741430789232255,
      "num_tokens": 100728577.0,
      "step": 45980
    },
    {
      "entropy": 5.686436367034912,
      "epoch": 4.563815005954744,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003095221842953104,
      "loss": 4.9211,
      "mean_token_accuracy": 0.19222811460494996,
      "num_tokens": 100740018.0,
      "step": 45985
    },
    {
      "entropy": 5.686366987228394,
      "epoch": 4.564311234616912,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030948717190789726,
      "loss": 4.9219,
      "mean_token_accuracy": 0.19553787708282472,
      "num_tokens": 100751764.0,
      "step": 45990
    },
    {
      "entropy": 5.700778436660767,
      "epoch": 4.564807463279079,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003094521586652766,
      "loss": 4.9953,
      "mean_token_accuracy": 0.19059084802865983,
      "num_tokens": 100764196.0,
      "step": 45995
    },
    {
      "entropy": 5.717603349685669,
      "epoch": 4.565303691941247,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030941714456831666,
      "loss": 5.0088,
      "mean_token_accuracy": 0.18928933441638945,
      "num_tokens": 100776315.0,
      "step": 46000
    },
    {
      "entropy": 5.802991151809692,
      "epoch": 4.565799920603414,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003093821296178856,
      "loss": 5.0699,
      "mean_token_accuracy": 0.18174870163202286,
      "num_tokens": 100788816.0,
      "step": 46005
    },
    {
      "entropy": 5.680287790298462,
      "epoch": 4.566296149265582,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030934711381485193,
      "loss": 4.9859,
      "mean_token_accuracy": 0.19081977158784866,
      "num_tokens": 100799949.0,
      "step": 46010
    },
    {
      "entropy": 5.678953981399536,
      "epoch": 4.566792377927749,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030931209716008386,
      "loss": 4.9208,
      "mean_token_accuracy": 0.2032841518521309,
      "num_tokens": 100810854.0,
      "step": 46015
    },
    {
      "entropy": 5.7125983238220215,
      "epoch": 4.5672886065899165,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030927707965444965,
      "loss": 5.0381,
      "mean_token_accuracy": 0.18435539156198502,
      "num_tokens": 100822630.0,
      "step": 46020
    },
    {
      "entropy": 5.752562999725342,
      "epoch": 4.567784835252084,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030924206129881763,
      "loss": 5.0901,
      "mean_token_accuracy": 0.18350053280591966,
      "num_tokens": 100834617.0,
      "step": 46025
    },
    {
      "entropy": 5.695969486236573,
      "epoch": 4.5682810639142515,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030920704209405647,
      "loss": 4.9378,
      "mean_token_accuracy": 0.1978358268737793,
      "num_tokens": 100845890.0,
      "step": 46030
    },
    {
      "entropy": 5.672267150878906,
      "epoch": 4.568777292576419,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003091720220410343,
      "loss": 4.8898,
      "mean_token_accuracy": 0.20035441666841508,
      "num_tokens": 100855326.0,
      "step": 46035
    },
    {
      "entropy": 5.672778129577637,
      "epoch": 4.569273521238586,
      "grad_norm": 1.125,
      "learning_rate": 0.0003091370011406195,
      "loss": 5.0005,
      "mean_token_accuracy": 0.1937368020415306,
      "num_tokens": 100866894.0,
      "step": 46040
    },
    {
      "entropy": 5.678717708587646,
      "epoch": 4.569769749900754,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030910197939368084,
      "loss": 4.9488,
      "mean_token_accuracy": 0.19974270910024644,
      "num_tokens": 100877652.0,
      "step": 46045
    },
    {
      "entropy": 5.698641586303711,
      "epoch": 4.570265978562922,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003090669568010865,
      "loss": 4.8841,
      "mean_token_accuracy": 0.19838353246450424,
      "num_tokens": 100887498.0,
      "step": 46050
    },
    {
      "entropy": 5.733015394210815,
      "epoch": 4.570762207225089,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030903193336370493,
      "loss": 4.997,
      "mean_token_accuracy": 0.19465882182121277,
      "num_tokens": 100898595.0,
      "step": 46055
    },
    {
      "entropy": 5.673836517333984,
      "epoch": 4.571258435887257,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003089969090824049,
      "loss": 4.9196,
      "mean_token_accuracy": 0.19366802424192428,
      "num_tokens": 100909471.0,
      "step": 46060
    },
    {
      "entropy": 5.713471841812134,
      "epoch": 4.571754664549425,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003089618839580547,
      "loss": 5.0097,
      "mean_token_accuracy": 0.19243373572826386,
      "num_tokens": 100920668.0,
      "step": 46065
    },
    {
      "entropy": 5.700793695449829,
      "epoch": 4.572250893211592,
      "grad_norm": 1.34375,
      "learning_rate": 0.000308926857991523,
      "loss": 4.9573,
      "mean_token_accuracy": 0.1982711672782898,
      "num_tokens": 100930671.0,
      "step": 46070
    },
    {
      "entropy": 5.673440551757812,
      "epoch": 4.57274712187376,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030889183118367825,
      "loss": 4.9321,
      "mean_token_accuracy": 0.19815975725650786,
      "num_tokens": 100941539.0,
      "step": 46075
    },
    {
      "entropy": 5.62295355796814,
      "epoch": 4.573243350535927,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00030885680353538914,
      "loss": 4.9311,
      "mean_token_accuracy": 0.19495891332626342,
      "num_tokens": 100952638.0,
      "step": 46080
    },
    {
      "entropy": 5.671207332611084,
      "epoch": 4.573739579198095,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003088217750475243,
      "loss": 4.9071,
      "mean_token_accuracy": 0.1993838757276535,
      "num_tokens": 100962352.0,
      "step": 46085
    },
    {
      "entropy": 5.764818811416626,
      "epoch": 4.574235807860262,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030878674572095235,
      "loss": 4.9687,
      "mean_token_accuracy": 0.18697663247585297,
      "num_tokens": 100974824.0,
      "step": 46090
    },
    {
      "entropy": 5.665307235717774,
      "epoch": 4.5747320365224295,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030875171555654177,
      "loss": 4.916,
      "mean_token_accuracy": 0.2003108486533165,
      "num_tokens": 100985349.0,
      "step": 46095
    },
    {
      "entropy": 5.661301851272583,
      "epoch": 4.575228265184597,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003087166845551615,
      "loss": 4.9693,
      "mean_token_accuracy": 0.18711915165185927,
      "num_tokens": 100995918.0,
      "step": 46100
    },
    {
      "entropy": 5.640061569213867,
      "epoch": 4.5757244938467645,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003086816527176801,
      "loss": 4.8763,
      "mean_token_accuracy": 0.20128365159034728,
      "num_tokens": 101006460.0,
      "step": 46105
    },
    {
      "entropy": 5.721919012069702,
      "epoch": 4.576220722508932,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003086466200449662,
      "loss": 4.9813,
      "mean_token_accuracy": 0.1930655851960182,
      "num_tokens": 101017123.0,
      "step": 46110
    },
    {
      "entropy": 5.674138116836548,
      "epoch": 4.5767169511710994,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030861158653788866,
      "loss": 4.9154,
      "mean_token_accuracy": 0.18861766755580903,
      "num_tokens": 101027178.0,
      "step": 46115
    },
    {
      "entropy": 5.778035831451416,
      "epoch": 4.577213179833267,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003085765521973162,
      "loss": 5.0386,
      "mean_token_accuracy": 0.18603121042251586,
      "num_tokens": 101037864.0,
      "step": 46120
    },
    {
      "entropy": 5.714333534240723,
      "epoch": 4.577709408495434,
      "grad_norm": 1.25,
      "learning_rate": 0.00030854151702411765,
      "loss": 5.0091,
      "mean_token_accuracy": 0.19232516884803771,
      "num_tokens": 101049028.0,
      "step": 46125
    },
    {
      "entropy": 5.732700300216675,
      "epoch": 4.578205637157602,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030850648101916173,
      "loss": 5.0122,
      "mean_token_accuracy": 0.18974204063415528,
      "num_tokens": 101060846.0,
      "step": 46130
    },
    {
      "entropy": 5.726913738250732,
      "epoch": 4.57870186581977,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003084714441833173,
      "loss": 4.9151,
      "mean_token_accuracy": 0.1961391180753708,
      "num_tokens": 101069993.0,
      "step": 46135
    },
    {
      "entropy": 5.682952213287353,
      "epoch": 4.579198094481937,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030843640651745315,
      "loss": 4.9573,
      "mean_token_accuracy": 0.19870064854621888,
      "num_tokens": 101081629.0,
      "step": 46140
    },
    {
      "entropy": 5.701624345779419,
      "epoch": 4.579694323144105,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003084013680224382,
      "loss": 4.9944,
      "mean_token_accuracy": 0.19154397547245025,
      "num_tokens": 101092658.0,
      "step": 46145
    },
    {
      "entropy": 5.709900951385498,
      "epoch": 4.580190551806273,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003083663286991413,
      "loss": 5.0011,
      "mean_token_accuracy": 0.19489655941724776,
      "num_tokens": 101104241.0,
      "step": 46150
    },
    {
      "entropy": 5.720819520950317,
      "epoch": 4.58068678046844,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003083312885484312,
      "loss": 5.0249,
      "mean_token_accuracy": 0.1818838581442833,
      "num_tokens": 101115293.0,
      "step": 46155
    },
    {
      "entropy": 5.729002237319946,
      "epoch": 4.581183009130608,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030829624757117716,
      "loss": 5.0086,
      "mean_token_accuracy": 0.1949225053191185,
      "num_tokens": 101125188.0,
      "step": 46160
    },
    {
      "entropy": 5.7117352962493895,
      "epoch": 4.581679237792775,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030826120576824787,
      "loss": 4.9383,
      "mean_token_accuracy": 0.19497088193893433,
      "num_tokens": 101136152.0,
      "step": 46165
    },
    {
      "entropy": 5.690120840072632,
      "epoch": 4.582175466454943,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003082261631405122,
      "loss": 4.9626,
      "mean_token_accuracy": 0.18671380281448363,
      "num_tokens": 101146761.0,
      "step": 46170
    },
    {
      "entropy": 5.689545679092407,
      "epoch": 4.58267169511711,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030819111968883944,
      "loss": 4.9722,
      "mean_token_accuracy": 0.19550910890102385,
      "num_tokens": 101157143.0,
      "step": 46175
    },
    {
      "entropy": 5.704109001159668,
      "epoch": 4.5831679237792775,
      "grad_norm": 1.421875,
      "learning_rate": 0.00030815607541409844,
      "loss": 4.9479,
      "mean_token_accuracy": 0.18716806769371033,
      "num_tokens": 101168381.0,
      "step": 46180
    },
    {
      "entropy": 5.65775055885315,
      "epoch": 4.583664152441445,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030812103031715814,
      "loss": 4.9691,
      "mean_token_accuracy": 0.192995186150074,
      "num_tokens": 101179029.0,
      "step": 46185
    },
    {
      "entropy": 5.658078241348266,
      "epoch": 4.5841603811036125,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003080859843988877,
      "loss": 4.9679,
      "mean_token_accuracy": 0.20201851278543473,
      "num_tokens": 101189072.0,
      "step": 46190
    },
    {
      "entropy": 5.739101934432983,
      "epoch": 4.58465660976578,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030805093766015615,
      "loss": 4.9646,
      "mean_token_accuracy": 0.20064126551151276,
      "num_tokens": 101201255.0,
      "step": 46195
    },
    {
      "entropy": 5.723522901535034,
      "epoch": 4.585152838427947,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003080158901018325,
      "loss": 4.9708,
      "mean_token_accuracy": 0.192024065554142,
      "num_tokens": 101213425.0,
      "step": 46200
    },
    {
      "entropy": 5.7270917892456055,
      "epoch": 4.585649067090115,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030798084172478597,
      "loss": 4.9727,
      "mean_token_accuracy": 0.18959883153438567,
      "num_tokens": 101223627.0,
      "step": 46205
    },
    {
      "entropy": 5.665586376190186,
      "epoch": 4.586145295752282,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030794579252988564,
      "loss": 4.897,
      "mean_token_accuracy": 0.19842386096715928,
      "num_tokens": 101234055.0,
      "step": 46210
    },
    {
      "entropy": 5.655409717559815,
      "epoch": 4.58664152441445,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003079107425180006,
      "loss": 4.9447,
      "mean_token_accuracy": 0.1985083371400833,
      "num_tokens": 101245317.0,
      "step": 46215
    },
    {
      "entropy": 5.685274648666382,
      "epoch": 4.587137753076618,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003078756916900001,
      "loss": 4.9963,
      "mean_token_accuracy": 0.1934376761317253,
      "num_tokens": 101256385.0,
      "step": 46220
    },
    {
      "entropy": 5.702569150924683,
      "epoch": 4.587633981738785,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003078406400467533,
      "loss": 4.9703,
      "mean_token_accuracy": 0.18819855153560638,
      "num_tokens": 101267263.0,
      "step": 46225
    },
    {
      "entropy": 5.6921241760253904,
      "epoch": 4.588130210400953,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030780558758912934,
      "loss": 4.9905,
      "mean_token_accuracy": 0.19235614091157913,
      "num_tokens": 101278451.0,
      "step": 46230
    },
    {
      "entropy": 5.707369756698609,
      "epoch": 4.588626439063121,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003077705343179975,
      "loss": 5.0101,
      "mean_token_accuracy": 0.19100617319345475,
      "num_tokens": 101289056.0,
      "step": 46235
    },
    {
      "entropy": 5.6932861328125,
      "epoch": 4.589122667725288,
      "grad_norm": 1.328125,
      "learning_rate": 0.000307735480234227,
      "loss": 5.006,
      "mean_token_accuracy": 0.1883505776524544,
      "num_tokens": 101300272.0,
      "step": 46240
    },
    {
      "entropy": 5.689658451080322,
      "epoch": 4.589618896387456,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030770042533868723,
      "loss": 4.9846,
      "mean_token_accuracy": 0.18804891854524614,
      "num_tokens": 101311567.0,
      "step": 46245
    },
    {
      "entropy": 5.83546814918518,
      "epoch": 4.590115125049623,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003076653696322474,
      "loss": 5.0885,
      "mean_token_accuracy": 0.19250377416610717,
      "num_tokens": 101324687.0,
      "step": 46250
    },
    {
      "entropy": 5.762657928466797,
      "epoch": 4.5906113537117905,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030763031311577667,
      "loss": 4.9535,
      "mean_token_accuracy": 0.19661772400140762,
      "num_tokens": 101334633.0,
      "step": 46255
    },
    {
      "entropy": 5.721362113952637,
      "epoch": 4.591107582373958,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030759525579014457,
      "loss": 5.0229,
      "mean_token_accuracy": 0.1940428465604782,
      "num_tokens": 101346263.0,
      "step": 46260
    },
    {
      "entropy": 5.6643678665161135,
      "epoch": 4.5916038110361255,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003075601976562204,
      "loss": 4.9045,
      "mean_token_accuracy": 0.2005174994468689,
      "num_tokens": 101357699.0,
      "step": 46265
    },
    {
      "entropy": 5.654833507537842,
      "epoch": 4.592100039698293,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003075251387148733,
      "loss": 4.9533,
      "mean_token_accuracy": 0.19585829824209214,
      "num_tokens": 101368649.0,
      "step": 46270
    },
    {
      "entropy": 5.623820972442627,
      "epoch": 4.59259626836046,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000307490078966973,
      "loss": 4.9389,
      "mean_token_accuracy": 0.20096748918294907,
      "num_tokens": 101378407.0,
      "step": 46275
    },
    {
      "entropy": 5.748395586013794,
      "epoch": 4.593092497022628,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030745501841338875,
      "loss": 5.0094,
      "mean_token_accuracy": 0.1929832175374031,
      "num_tokens": 101388307.0,
      "step": 46280
    },
    {
      "entropy": 5.701421070098877,
      "epoch": 4.593588725684795,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030741995705498986,
      "loss": 4.9674,
      "mean_token_accuracy": 0.19078415036201476,
      "num_tokens": 101398895.0,
      "step": 46285
    },
    {
      "entropy": 5.672916507720947,
      "epoch": 4.594084954346963,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030738489489264605,
      "loss": 4.981,
      "mean_token_accuracy": 0.19507699608802795,
      "num_tokens": 101409150.0,
      "step": 46290
    },
    {
      "entropy": 5.6782125473022464,
      "epoch": 4.59458118300913,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030734983192722654,
      "loss": 4.9738,
      "mean_token_accuracy": 0.18719030916690826,
      "num_tokens": 101420331.0,
      "step": 46295
    },
    {
      "entropy": 5.716427326202393,
      "epoch": 4.595077411671298,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003073147681596009,
      "loss": 5.0131,
      "mean_token_accuracy": 0.19552670866250993,
      "num_tokens": 101431775.0,
      "step": 46300
    },
    {
      "entropy": 5.677390670776367,
      "epoch": 4.595573640333466,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030727970359063864,
      "loss": 4.8998,
      "mean_token_accuracy": 0.19779007732868195,
      "num_tokens": 101442365.0,
      "step": 46305
    },
    {
      "entropy": 5.740860414505005,
      "epoch": 4.596069868995633,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030724463822120936,
      "loss": 4.9911,
      "mean_token_accuracy": 0.18974662870168685,
      "num_tokens": 101452930.0,
      "step": 46310
    },
    {
      "entropy": 5.687701845169068,
      "epoch": 4.596566097657801,
      "grad_norm": 1.125,
      "learning_rate": 0.0003072095720521825,
      "loss": 4.9658,
      "mean_token_accuracy": 0.20182569473981857,
      "num_tokens": 101462732.0,
      "step": 46315
    },
    {
      "entropy": 5.668218183517456,
      "epoch": 4.597062326319969,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003071745050844276,
      "loss": 4.9281,
      "mean_token_accuracy": 0.20213375091552735,
      "num_tokens": 101473729.0,
      "step": 46320
    },
    {
      "entropy": 5.6497142791748045,
      "epoch": 4.597558554982136,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003071394373188144,
      "loss": 4.9366,
      "mean_token_accuracy": 0.19783090502023698,
      "num_tokens": 101483154.0,
      "step": 46325
    },
    {
      "entropy": 5.734755563735962,
      "epoch": 4.5980547836443035,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003071043687562123,
      "loss": 5.0089,
      "mean_token_accuracy": 0.19090940654277802,
      "num_tokens": 101494884.0,
      "step": 46330
    },
    {
      "entropy": 5.691573762893677,
      "epoch": 4.598551012306471,
      "grad_norm": 1.328125,
      "learning_rate": 0.00030706929939749105,
      "loss": 4.9769,
      "mean_token_accuracy": 0.1937495931982994,
      "num_tokens": 101505635.0,
      "step": 46335
    },
    {
      "entropy": 5.678376960754394,
      "epoch": 4.5990472409686385,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030703422924352035,
      "loss": 4.9429,
      "mean_token_accuracy": 0.19192392826080323,
      "num_tokens": 101516002.0,
      "step": 46340
    },
    {
      "entropy": 5.730712413787842,
      "epoch": 4.599543469630806,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003069991582951698,
      "loss": 4.9704,
      "mean_token_accuracy": 0.1929771199822426,
      "num_tokens": 101527183.0,
      "step": 46345
    },
    {
      "entropy": 5.642569875717163,
      "epoch": 4.600039698292973,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030696408655330896,
      "loss": 4.8539,
      "mean_token_accuracy": 0.20341167449951172,
      "num_tokens": 101538194.0,
      "step": 46350
    },
    {
      "entropy": 5.695984601974487,
      "epoch": 4.600535926955141,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030692901401880776,
      "loss": 5.0273,
      "mean_token_accuracy": 0.19709424376487733,
      "num_tokens": 101549520.0,
      "step": 46355
    },
    {
      "entropy": 5.768232965469361,
      "epoch": 4.601032155617308,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003068939406925358,
      "loss": 5.0609,
      "mean_token_accuracy": 0.18217841386795045,
      "num_tokens": 101560413.0,
      "step": 46360
    },
    {
      "entropy": 5.750452756881714,
      "epoch": 4.601528384279476,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030685886657536276,
      "loss": 5.0062,
      "mean_token_accuracy": 0.19270871430635453,
      "num_tokens": 101570356.0,
      "step": 46365
    },
    {
      "entropy": 5.730723237991333,
      "epoch": 4.602024612941643,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003068237916681586,
      "loss": 4.9701,
      "mean_token_accuracy": 0.19226278364658356,
      "num_tokens": 101580861.0,
      "step": 46370
    },
    {
      "entropy": 5.667830896377564,
      "epoch": 4.602520841603811,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030678871597179287,
      "loss": 4.9869,
      "mean_token_accuracy": 0.19216389805078507,
      "num_tokens": 101592279.0,
      "step": 46375
    },
    {
      "entropy": 5.684663724899292,
      "epoch": 4.603017070265978,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003067536394871356,
      "loss": 4.9634,
      "mean_token_accuracy": 0.1911135733127594,
      "num_tokens": 101603757.0,
      "step": 46380
    },
    {
      "entropy": 5.619799327850342,
      "epoch": 4.603513298928146,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003067185622150564,
      "loss": 4.9011,
      "mean_token_accuracy": 0.19783582985401155,
      "num_tokens": 101616511.0,
      "step": 46385
    },
    {
      "entropy": 5.737771701812744,
      "epoch": 4.604009527590313,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030668348415642516,
      "loss": 4.9514,
      "mean_token_accuracy": 0.1922409102320671,
      "num_tokens": 101627727.0,
      "step": 46390
    },
    {
      "entropy": 5.757972669601441,
      "epoch": 4.604505756252481,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030664840531211186,
      "loss": 5.0284,
      "mean_token_accuracy": 0.1901863157749176,
      "num_tokens": 101639128.0,
      "step": 46395
    },
    {
      "entropy": 5.748822402954102,
      "epoch": 4.605001984914649,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003066133256829863,
      "loss": 4.967,
      "mean_token_accuracy": 0.19108838587999344,
      "num_tokens": 101648481.0,
      "step": 46400
    },
    {
      "entropy": 5.69073486328125,
      "epoch": 4.6054982135768165,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030657824526991837,
      "loss": 4.982,
      "mean_token_accuracy": 0.19324965178966522,
      "num_tokens": 101658736.0,
      "step": 46405
    },
    {
      "entropy": 5.714685344696045,
      "epoch": 4.605994442238984,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030654316407377805,
      "loss": 5.0161,
      "mean_token_accuracy": 0.18858579099178313,
      "num_tokens": 101668731.0,
      "step": 46410
    },
    {
      "entropy": 5.69347071647644,
      "epoch": 4.6064906709011515,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003065080820954352,
      "loss": 4.8974,
      "mean_token_accuracy": 0.19725692719221116,
      "num_tokens": 101679212.0,
      "step": 46415
    },
    {
      "entropy": 5.680988597869873,
      "epoch": 4.606986899563319,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003064729993357597,
      "loss": 4.9386,
      "mean_token_accuracy": 0.1959274411201477,
      "num_tokens": 101690960.0,
      "step": 46420
    },
    {
      "entropy": 5.621916723251343,
      "epoch": 4.6074831282254864,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003064379157956218,
      "loss": 4.9077,
      "mean_token_accuracy": 0.20196886360645294,
      "num_tokens": 101701628.0,
      "step": 46425
    },
    {
      "entropy": 5.697695779800415,
      "epoch": 4.607979356887654,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003064028314758913,
      "loss": 5.0545,
      "mean_token_accuracy": 0.19044671803712845,
      "num_tokens": 101712503.0,
      "step": 46430
    },
    {
      "entropy": 5.671007966995239,
      "epoch": 4.608475585549821,
      "grad_norm": 1.125,
      "learning_rate": 0.00030636774637743814,
      "loss": 4.9261,
      "mean_token_accuracy": 0.19449508488178252,
      "num_tokens": 101724074.0,
      "step": 46435
    },
    {
      "entropy": 5.707862520217896,
      "epoch": 4.608971814211989,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030633266050113255,
      "loss": 5.0041,
      "mean_token_accuracy": 0.1861477494239807,
      "num_tokens": 101735154.0,
      "step": 46440
    },
    {
      "entropy": 5.716191673278809,
      "epoch": 4.609468042874156,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030629757384784454,
      "loss": 4.9673,
      "mean_token_accuracy": 0.1893083393573761,
      "num_tokens": 101745524.0,
      "step": 46445
    },
    {
      "entropy": 5.74163293838501,
      "epoch": 4.609964271536324,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000306262486418444,
      "loss": 5.0233,
      "mean_token_accuracy": 0.19198810309171677,
      "num_tokens": 101758619.0,
      "step": 46450
    },
    {
      "entropy": 5.698317050933838,
      "epoch": 4.610460500198491,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003062273982138013,
      "loss": 4.8973,
      "mean_token_accuracy": 0.19904808104038238,
      "num_tokens": 101768827.0,
      "step": 46455
    },
    {
      "entropy": 5.703480339050293,
      "epoch": 4.610956728860659,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030619230923478643,
      "loss": 5.0272,
      "mean_token_accuracy": 0.18549381494522094,
      "num_tokens": 101779900.0,
      "step": 46460
    },
    {
      "entropy": 5.686403369903564,
      "epoch": 4.611452957522826,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030615721948226926,
      "loss": 4.9024,
      "mean_token_accuracy": 0.19840627759695054,
      "num_tokens": 101790839.0,
      "step": 46465
    },
    {
      "entropy": 5.681330680847168,
      "epoch": 4.611949186184994,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030612212895712046,
      "loss": 4.9065,
      "mean_token_accuracy": 0.19185546040534973,
      "num_tokens": 101800556.0,
      "step": 46470
    },
    {
      "entropy": 5.7124025344848635,
      "epoch": 4.612445414847161,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003060870376602098,
      "loss": 4.9969,
      "mean_token_accuracy": 0.19106835275888442,
      "num_tokens": 101811250.0,
      "step": 46475
    },
    {
      "entropy": 5.724036598205567,
      "epoch": 4.612941643509329,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030605194559240754,
      "loss": 4.9464,
      "mean_token_accuracy": 0.19071220904588698,
      "num_tokens": 101821882.0,
      "step": 46480
    },
    {
      "entropy": 5.771833801269532,
      "epoch": 4.613437872171497,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030601685275458406,
      "loss": 5.1605,
      "mean_token_accuracy": 0.1854802131652832,
      "num_tokens": 101833549.0,
      "step": 46485
    },
    {
      "entropy": 5.725757455825805,
      "epoch": 4.6139341008336645,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003059817591476094,
      "loss": 5.0393,
      "mean_token_accuracy": 0.1885104864835739,
      "num_tokens": 101844096.0,
      "step": 46490
    },
    {
      "entropy": 5.646440029144287,
      "epoch": 4.614430329495832,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003059466647723539,
      "loss": 4.9372,
      "mean_token_accuracy": 0.1940033718943596,
      "num_tokens": 101855984.0,
      "step": 46495
    },
    {
      "entropy": 5.650190544128418,
      "epoch": 4.6149265581579995,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003059115696296878,
      "loss": 4.8273,
      "mean_token_accuracy": 0.20599254816770554,
      "num_tokens": 101866231.0,
      "step": 46500
    },
    {
      "entropy": 5.714240550994873,
      "epoch": 4.615422786820167,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003058764737204813,
      "loss": 5.0044,
      "mean_token_accuracy": 0.19214030653238295,
      "num_tokens": 101877450.0,
      "step": 46505
    },
    {
      "entropy": 5.7115638256073,
      "epoch": 4.615919015482334,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030584137704560486,
      "loss": 4.9502,
      "mean_token_accuracy": 0.18821438550949096,
      "num_tokens": 101887369.0,
      "step": 46510
    },
    {
      "entropy": 5.671628046035766,
      "epoch": 4.616415244144502,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030580627960592873,
      "loss": 4.9208,
      "mean_token_accuracy": 0.1939760059118271,
      "num_tokens": 101898554.0,
      "step": 46515
    },
    {
      "entropy": 5.601093912124634,
      "epoch": 4.616911472806669,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003057711814023232,
      "loss": 4.8567,
      "mean_token_accuracy": 0.21040789037942886,
      "num_tokens": 101910223.0,
      "step": 46520
    },
    {
      "entropy": 5.6650866031646725,
      "epoch": 4.617407701468837,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030573608243565875,
      "loss": 4.985,
      "mean_token_accuracy": 0.18529019355773926,
      "num_tokens": 101920820.0,
      "step": 46525
    },
    {
      "entropy": 5.707976341247559,
      "epoch": 4.617903930131004,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003057009827068057,
      "loss": 5.0143,
      "mean_token_accuracy": 0.1879585564136505,
      "num_tokens": 101933221.0,
      "step": 46530
    },
    {
      "entropy": 5.672554206848145,
      "epoch": 4.618400158793172,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003056658822166343,
      "loss": 4.9204,
      "mean_token_accuracy": 0.20024703443050385,
      "num_tokens": 101943977.0,
      "step": 46535
    },
    {
      "entropy": 5.728269720077515,
      "epoch": 4.618896387455339,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003056307809660152,
      "loss": 4.9854,
      "mean_token_accuracy": 0.19716002196073532,
      "num_tokens": 101955736.0,
      "step": 46540
    },
    {
      "entropy": 5.7425765037536625,
      "epoch": 4.619392616117507,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003055956789558187,
      "loss": 4.947,
      "mean_token_accuracy": 0.19481706917285918,
      "num_tokens": 101967452.0,
      "step": 46545
    },
    {
      "entropy": 5.711707353591919,
      "epoch": 4.619888844779674,
      "grad_norm": 1.484375,
      "learning_rate": 0.00030556057618691537,
      "loss": 4.9838,
      "mean_token_accuracy": 0.1960600048303604,
      "num_tokens": 101979203.0,
      "step": 46550
    },
    {
      "entropy": 5.660134553909302,
      "epoch": 4.620385073441842,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030552547266017555,
      "loss": 4.9333,
      "mean_token_accuracy": 0.19128093868494034,
      "num_tokens": 101989927.0,
      "step": 46555
    },
    {
      "entropy": 5.6601721286773685,
      "epoch": 4.620881302104009,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030549036837646974,
      "loss": 4.9613,
      "mean_token_accuracy": 0.19394566267728805,
      "num_tokens": 102000899.0,
      "step": 46560
    },
    {
      "entropy": 5.590873956680298,
      "epoch": 4.621377530766177,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003054552633366685,
      "loss": 4.8247,
      "mean_token_accuracy": 0.20167437940835953,
      "num_tokens": 102012107.0,
      "step": 46565
    },
    {
      "entropy": 5.613870429992676,
      "epoch": 4.621873759428345,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003054201575416424,
      "loss": 4.9078,
      "mean_token_accuracy": 0.20343851596117019,
      "num_tokens": 102024463.0,
      "step": 46570
    },
    {
      "entropy": 5.7365950584411625,
      "epoch": 4.6223699880905125,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000305385050992262,
      "loss": 5.0294,
      "mean_token_accuracy": 0.18845252841711044,
      "num_tokens": 102036938.0,
      "step": 46575
    },
    {
      "entropy": 5.74267463684082,
      "epoch": 4.62286621675268,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003053499436893977,
      "loss": 4.9988,
      "mean_token_accuracy": 0.18900613188743592,
      "num_tokens": 102047657.0,
      "step": 46580
    },
    {
      "entropy": 5.613963603973389,
      "epoch": 4.623362445414847,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003053148356339202,
      "loss": 4.8532,
      "mean_token_accuracy": 0.20163197666406632,
      "num_tokens": 102057736.0,
      "step": 46585
    },
    {
      "entropy": 5.683609199523926,
      "epoch": 4.623858674077015,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003052797268267001,
      "loss": 4.9842,
      "mean_token_accuracy": 0.18461976647377015,
      "num_tokens": 102068180.0,
      "step": 46590
    },
    {
      "entropy": 5.747671270370484,
      "epoch": 4.624354902739182,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030524461726860806,
      "loss": 5.0225,
      "mean_token_accuracy": 0.19724960029125213,
      "num_tokens": 102078347.0,
      "step": 46595
    },
    {
      "entropy": 5.689337348937988,
      "epoch": 4.62485113140135,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030520950696051477,
      "loss": 4.8976,
      "mean_token_accuracy": 0.1938445031642914,
      "num_tokens": 102088502.0,
      "step": 46600
    },
    {
      "entropy": 5.7265918254852295,
      "epoch": 4.625347360063517,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003051743959032907,
      "loss": 5.0225,
      "mean_token_accuracy": 0.19189313352108,
      "num_tokens": 102101820.0,
      "step": 46605
    },
    {
      "entropy": 5.682225179672241,
      "epoch": 4.625843588725685,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003051392840978066,
      "loss": 4.9227,
      "mean_token_accuracy": 0.1874893754720688,
      "num_tokens": 102112702.0,
      "step": 46610
    },
    {
      "entropy": 5.658632612228393,
      "epoch": 4.626339817387852,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030510417154493326,
      "loss": 4.9073,
      "mean_token_accuracy": 0.19488341212272645,
      "num_tokens": 102123927.0,
      "step": 46615
    },
    {
      "entropy": 5.678346920013428,
      "epoch": 4.62683604605002,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003050690582455413,
      "loss": 5.0182,
      "mean_token_accuracy": 0.19403901249170302,
      "num_tokens": 102135215.0,
      "step": 46620
    },
    {
      "entropy": 5.648607587814331,
      "epoch": 4.627332274712187,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003050339442005015,
      "loss": 4.9685,
      "mean_token_accuracy": 0.198044952750206,
      "num_tokens": 102148202.0,
      "step": 46625
    },
    {
      "entropy": 5.637993288040161,
      "epoch": 4.627828503374355,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030499882941068457,
      "loss": 4.8965,
      "mean_token_accuracy": 0.19164026379585267,
      "num_tokens": 102160839.0,
      "step": 46630
    },
    {
      "entropy": 5.689466953277588,
      "epoch": 4.628324732036522,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003049637138769614,
      "loss": 4.9645,
      "mean_token_accuracy": 0.19595326185226442,
      "num_tokens": 102171801.0,
      "step": 46635
    },
    {
      "entropy": 5.698542976379395,
      "epoch": 4.62882096069869,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003049285976002026,
      "loss": 5.016,
      "mean_token_accuracy": 0.19179847538471223,
      "num_tokens": 102184297.0,
      "step": 46640
    },
    {
      "entropy": 5.699436950683594,
      "epoch": 4.629317189360857,
      "grad_norm": 1.25,
      "learning_rate": 0.00030489348058127913,
      "loss": 4.9131,
      "mean_token_accuracy": 0.20469842702150345,
      "num_tokens": 102194703.0,
      "step": 46645
    },
    {
      "entropy": 5.723335838317871,
      "epoch": 4.629813418023025,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003048583628210618,
      "loss": 4.9805,
      "mean_token_accuracy": 0.19602201581001283,
      "num_tokens": 102206220.0,
      "step": 46650
    },
    {
      "entropy": 5.675420665740967,
      "epoch": 4.630309646685193,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003048232443204213,
      "loss": 4.9055,
      "mean_token_accuracy": 0.1962452232837677,
      "num_tokens": 102215384.0,
      "step": 46655
    },
    {
      "entropy": 5.63800163269043,
      "epoch": 4.63080587534736,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003047881250802286,
      "loss": 4.9593,
      "mean_token_accuracy": 0.1887870967388153,
      "num_tokens": 102226503.0,
      "step": 46660
    },
    {
      "entropy": 5.687651300430298,
      "epoch": 4.631302104009528,
      "grad_norm": 1.25,
      "learning_rate": 0.00030475300510135467,
      "loss": 4.9771,
      "mean_token_accuracy": 0.19151587784290314,
      "num_tokens": 102237835.0,
      "step": 46665
    },
    {
      "entropy": 5.68894944190979,
      "epoch": 4.631798332671695,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030471788438467024,
      "loss": 4.9198,
      "mean_token_accuracy": 0.19439459592103958,
      "num_tokens": 102248425.0,
      "step": 46670
    },
    {
      "entropy": 5.669542646408081,
      "epoch": 4.632294561333863,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030468276293104643,
      "loss": 4.9888,
      "mean_token_accuracy": 0.20031607747077942,
      "num_tokens": 102261185.0,
      "step": 46675
    },
    {
      "entropy": 5.722975158691407,
      "epoch": 4.63279078999603,
      "grad_norm": 1.1875,
      "learning_rate": 0.000304647640741354,
      "loss": 4.9724,
      "mean_token_accuracy": 0.1959003284573555,
      "num_tokens": 102272492.0,
      "step": 46680
    },
    {
      "entropy": 5.76371488571167,
      "epoch": 4.633287018658198,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003046125178164639,
      "loss": 4.9655,
      "mean_token_accuracy": 0.19093971103429794,
      "num_tokens": 102284036.0,
      "step": 46685
    },
    {
      "entropy": 5.6508458137512205,
      "epoch": 4.633783247320365,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030457739415724725,
      "loss": 4.9543,
      "mean_token_accuracy": 0.19263377785682678,
      "num_tokens": 102294265.0,
      "step": 46690
    },
    {
      "entropy": 5.724622678756714,
      "epoch": 4.634279475982533,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003045422697645749,
      "loss": 5.0163,
      "mean_token_accuracy": 0.18781045973300933,
      "num_tokens": 102305124.0,
      "step": 46695
    },
    {
      "entropy": 5.736504364013672,
      "epoch": 4.6347757046447,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000304507144639318,
      "loss": 5.0608,
      "mean_token_accuracy": 0.18581272214651107,
      "num_tokens": 102316092.0,
      "step": 46700
    },
    {
      "entropy": 5.699091958999634,
      "epoch": 4.635271933306868,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003044720187823475,
      "loss": 4.9993,
      "mean_token_accuracy": 0.1903443679213524,
      "num_tokens": 102326497.0,
      "step": 46705
    },
    {
      "entropy": 5.742067289352417,
      "epoch": 4.635768161969035,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003044368921945343,
      "loss": 5.035,
      "mean_token_accuracy": 0.19098056107759476,
      "num_tokens": 102337901.0,
      "step": 46710
    },
    {
      "entropy": 5.687427949905396,
      "epoch": 4.636264390631203,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003044017648767497,
      "loss": 4.9603,
      "mean_token_accuracy": 0.1987133502960205,
      "num_tokens": 102349708.0,
      "step": 46715
    },
    {
      "entropy": 5.708183908462525,
      "epoch": 4.63676061929337,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003043666368298647,
      "loss": 4.9806,
      "mean_token_accuracy": 0.1918801635503769,
      "num_tokens": 102360221.0,
      "step": 46720
    },
    {
      "entropy": 5.728805065155029,
      "epoch": 4.637256847955538,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030433150805475037,
      "loss": 5.0382,
      "mean_token_accuracy": 0.18630459755659104,
      "num_tokens": 102370874.0,
      "step": 46725
    },
    {
      "entropy": 5.658378219604492,
      "epoch": 4.637753076617705,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003042963785522778,
      "loss": 4.9367,
      "mean_token_accuracy": 0.19547373652458191,
      "num_tokens": 102381754.0,
      "step": 46730
    },
    {
      "entropy": 5.684584856033325,
      "epoch": 4.638249305279873,
      "grad_norm": 1.25,
      "learning_rate": 0.00030426124832331824,
      "loss": 5.0063,
      "mean_token_accuracy": 0.18989568203687668,
      "num_tokens": 102393362.0,
      "step": 46735
    },
    {
      "entropy": 5.691986703872681,
      "epoch": 4.638745533942041,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003042261173687427,
      "loss": 4.965,
      "mean_token_accuracy": 0.18944474458694457,
      "num_tokens": 102403599.0,
      "step": 46740
    },
    {
      "entropy": 5.694516944885254,
      "epoch": 4.639241762604208,
      "grad_norm": 1.328125,
      "learning_rate": 0.00030419098568942247,
      "loss": 4.9013,
      "mean_token_accuracy": 0.2125971332192421,
      "num_tokens": 102415802.0,
      "step": 46745
    },
    {
      "entropy": 5.688105154037475,
      "epoch": 4.639737991266376,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030415585328622875,
      "loss": 4.9017,
      "mean_token_accuracy": 0.1980184644460678,
      "num_tokens": 102425974.0,
      "step": 46750
    },
    {
      "entropy": 5.752027225494385,
      "epoch": 4.640234219928543,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030412072016003264,
      "loss": 5.0262,
      "mean_token_accuracy": 0.18908979147672653,
      "num_tokens": 102437070.0,
      "step": 46755
    },
    {
      "entropy": 5.705415153503418,
      "epoch": 4.640730448590711,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003040855863117054,
      "loss": 4.9777,
      "mean_token_accuracy": 0.18990974128246307,
      "num_tokens": 102449225.0,
      "step": 46760
    },
    {
      "entropy": 5.651110601425171,
      "epoch": 4.641226677252878,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003040504517421183,
      "loss": 4.9465,
      "mean_token_accuracy": 0.1893219083547592,
      "num_tokens": 102460623.0,
      "step": 46765
    },
    {
      "entropy": 5.665608501434326,
      "epoch": 4.641722905915046,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030401531645214263,
      "loss": 4.8726,
      "mean_token_accuracy": 0.1969660237431526,
      "num_tokens": 102471621.0,
      "step": 46770
    },
    {
      "entropy": 5.627744960784912,
      "epoch": 4.642219134577213,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030398018044264956,
      "loss": 4.8515,
      "mean_token_accuracy": 0.20053207129240036,
      "num_tokens": 102483168.0,
      "step": 46775
    },
    {
      "entropy": 5.573098278045654,
      "epoch": 4.642715363239381,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030394504371451055,
      "loss": 4.8598,
      "mean_token_accuracy": 0.20120981335639954,
      "num_tokens": 102494354.0,
      "step": 46780
    },
    {
      "entropy": 5.698077011108398,
      "epoch": 4.643211591901548,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003039099062685968,
      "loss": 4.9775,
      "mean_token_accuracy": 0.1886064201593399,
      "num_tokens": 102505099.0,
      "step": 46785
    },
    {
      "entropy": 5.667178010940551,
      "epoch": 4.643707820563716,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030387476810577965,
      "loss": 4.9514,
      "mean_token_accuracy": 0.19083693325519563,
      "num_tokens": 102516374.0,
      "step": 46790
    },
    {
      "entropy": 5.7581664562225345,
      "epoch": 4.644204049225883,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030383962922693046,
      "loss": 5.0478,
      "mean_token_accuracy": 0.1882656291127205,
      "num_tokens": 102527064.0,
      "step": 46795
    },
    {
      "entropy": 5.7043171405792235,
      "epoch": 4.644700277888051,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003038044896329206,
      "loss": 5.0269,
      "mean_token_accuracy": 0.19247740805149077,
      "num_tokens": 102537767.0,
      "step": 46800
    },
    {
      "entropy": 5.723250198364258,
      "epoch": 4.645196506550218,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003037693493246216,
      "loss": 5.0191,
      "mean_token_accuracy": 0.1942099004983902,
      "num_tokens": 102548703.0,
      "step": 46805
    },
    {
      "entropy": 5.673181819915771,
      "epoch": 4.645692735212386,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030373420830290455,
      "loss": 4.9926,
      "mean_token_accuracy": 0.18920332491397857,
      "num_tokens": 102559555.0,
      "step": 46810
    },
    {
      "entropy": 5.750560760498047,
      "epoch": 4.646188963874553,
      "grad_norm": 1.5,
      "learning_rate": 0.0003036990665686411,
      "loss": 5.0593,
      "mean_token_accuracy": 0.18654498159885408,
      "num_tokens": 102570144.0,
      "step": 46815
    },
    {
      "entropy": 5.749925899505615,
      "epoch": 4.6466851925367205,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003036639241227027,
      "loss": 5.046,
      "mean_token_accuracy": 0.18530529141426086,
      "num_tokens": 102580409.0,
      "step": 46820
    },
    {
      "entropy": 5.668395900726319,
      "epoch": 4.647181421198889,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003036287809659607,
      "loss": 4.9661,
      "mean_token_accuracy": 0.19900230765342714,
      "num_tokens": 102592110.0,
      "step": 46825
    },
    {
      "entropy": 5.71794319152832,
      "epoch": 4.647677649861056,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003035936370992867,
      "loss": 4.985,
      "mean_token_accuracy": 0.19536077678203584,
      "num_tokens": 102602184.0,
      "step": 46830
    },
    {
      "entropy": 5.682858991622925,
      "epoch": 4.648173878523224,
      "grad_norm": 1.359375,
      "learning_rate": 0.000303558492523552,
      "loss": 4.9398,
      "mean_token_accuracy": 0.20238267332315446,
      "num_tokens": 102612105.0,
      "step": 46835
    },
    {
      "entropy": 5.72243218421936,
      "epoch": 4.648670107185391,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030352334723962826,
      "loss": 4.9623,
      "mean_token_accuracy": 0.19496030062437059,
      "num_tokens": 102622346.0,
      "step": 46840
    },
    {
      "entropy": 5.667861604690552,
      "epoch": 4.649166335847559,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030348820124838697,
      "loss": 4.9555,
      "mean_token_accuracy": 0.194766104221344,
      "num_tokens": 102634972.0,
      "step": 46845
    },
    {
      "entropy": 5.6433899879455565,
      "epoch": 4.649662564509726,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003034530545506997,
      "loss": 4.8847,
      "mean_token_accuracy": 0.1973733812570572,
      "num_tokens": 102646885.0,
      "step": 46850
    },
    {
      "entropy": 5.683405828475952,
      "epoch": 4.650158793171894,
      "grad_norm": 1.21875,
      "learning_rate": 0.000303417907147438,
      "loss": 4.886,
      "mean_token_accuracy": 0.19895431995391846,
      "num_tokens": 102658480.0,
      "step": 46855
    },
    {
      "entropy": 5.7206761837005615,
      "epoch": 4.650655021834061,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030338275903947335,
      "loss": 5.0438,
      "mean_token_accuracy": 0.1902088165283203,
      "num_tokens": 102670213.0,
      "step": 46860
    },
    {
      "entropy": 5.672035360336304,
      "epoch": 4.651151250496229,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030334761022767756,
      "loss": 4.8758,
      "mean_token_accuracy": 0.2009607270359993,
      "num_tokens": 102680286.0,
      "step": 46865
    },
    {
      "entropy": 5.7065966606140135,
      "epoch": 4.651647479158396,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030331246071292205,
      "loss": 4.993,
      "mean_token_accuracy": 0.18766987770795823,
      "num_tokens": 102690424.0,
      "step": 46870
    },
    {
      "entropy": 5.6668041229248045,
      "epoch": 4.652143707820564,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030327731049607846,
      "loss": 4.9678,
      "mean_token_accuracy": 0.19281111657619476,
      "num_tokens": 102701317.0,
      "step": 46875
    },
    {
      "entropy": 5.696487236022949,
      "epoch": 4.652639936482731,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003032421595780186,
      "loss": 4.9869,
      "mean_token_accuracy": 0.18976860046386718,
      "num_tokens": 102711786.0,
      "step": 46880
    },
    {
      "entropy": 5.728278303146363,
      "epoch": 4.653136165144899,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030320700795961403,
      "loss": 5.0002,
      "mean_token_accuracy": 0.1892445996403694,
      "num_tokens": 102722686.0,
      "step": 46885
    },
    {
      "entropy": 5.716233205795288,
      "epoch": 4.653632393807066,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003031718556417364,
      "loss": 5.0013,
      "mean_token_accuracy": 0.19356813728809358,
      "num_tokens": 102733925.0,
      "step": 46890
    },
    {
      "entropy": 5.743839502334595,
      "epoch": 4.6541286224692335,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003031367026252575,
      "loss": 5.0,
      "mean_token_accuracy": 0.19242225885391234,
      "num_tokens": 102744849.0,
      "step": 46895
    },
    {
      "entropy": 5.7193807601928714,
      "epoch": 4.654624851131401,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030310154891104906,
      "loss": 4.9491,
      "mean_token_accuracy": 0.19473722577095032,
      "num_tokens": 102756361.0,
      "step": 46900
    },
    {
      "entropy": 5.767239093780518,
      "epoch": 4.6551210797935685,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030306639449998265,
      "loss": 5.1068,
      "mean_token_accuracy": 0.18821300566196442,
      "num_tokens": 102767220.0,
      "step": 46905
    },
    {
      "entropy": 5.815689420700073,
      "epoch": 4.655617308455737,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030303123939293017,
      "loss": 5.0893,
      "mean_token_accuracy": 0.18472495526075364,
      "num_tokens": 102779360.0,
      "step": 46910
    },
    {
      "entropy": 5.659828138351441,
      "epoch": 4.656113537117903,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030299608359076336,
      "loss": 4.918,
      "mean_token_accuracy": 0.20116373002529145,
      "num_tokens": 102790526.0,
      "step": 46915
    },
    {
      "entropy": 5.683747673034668,
      "epoch": 4.656609765780072,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000302960927094354,
      "loss": 4.9003,
      "mean_token_accuracy": 0.20355415642261504,
      "num_tokens": 102799619.0,
      "step": 46920
    },
    {
      "entropy": 5.6461187362670895,
      "epoch": 4.657105994442239,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003029257699045738,
      "loss": 4.9158,
      "mean_token_accuracy": 0.19240590184926987,
      "num_tokens": 102809938.0,
      "step": 46925
    },
    {
      "entropy": 5.773678827285766,
      "epoch": 4.657602223104407,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003028906120222948,
      "loss": 4.9972,
      "mean_token_accuracy": 0.18799347728490828,
      "num_tokens": 102819819.0,
      "step": 46930
    },
    {
      "entropy": 5.688858842849731,
      "epoch": 4.658098451766574,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030285545344838874,
      "loss": 5.0117,
      "mean_token_accuracy": 0.18861546963453293,
      "num_tokens": 102829738.0,
      "step": 46935
    },
    {
      "entropy": 5.672673130035401,
      "epoch": 4.658594680428742,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003028202941837274,
      "loss": 4.9429,
      "mean_token_accuracy": 0.19508117139339448,
      "num_tokens": 102839824.0,
      "step": 46940
    },
    {
      "entropy": 5.733531045913696,
      "epoch": 4.659090909090909,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030278513422918275,
      "loss": 5.0082,
      "mean_token_accuracy": 0.19149499833583833,
      "num_tokens": 102851536.0,
      "step": 46945
    },
    {
      "entropy": 5.686882162094117,
      "epoch": 4.659587137753077,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003027499735856266,
      "loss": 4.9357,
      "mean_token_accuracy": 0.1918787404894829,
      "num_tokens": 102862173.0,
      "step": 46950
    },
    {
      "entropy": 5.632356357574463,
      "epoch": 4.660083366415244,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030271481225393093,
      "loss": 4.9107,
      "mean_token_accuracy": 0.1975124016404152,
      "num_tokens": 102872225.0,
      "step": 46955
    },
    {
      "entropy": 5.692691230773926,
      "epoch": 4.660579595077412,
      "grad_norm": 1.328125,
      "learning_rate": 0.00030267965023496763,
      "loss": 4.9487,
      "mean_token_accuracy": 0.1946792408823967,
      "num_tokens": 102882677.0,
      "step": 46960
    },
    {
      "entropy": 5.779485416412354,
      "epoch": 4.661075823739579,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030264448752960867,
      "loss": 5.0548,
      "mean_token_accuracy": 0.18346035480499268,
      "num_tokens": 102892653.0,
      "step": 46965
    },
    {
      "entropy": 5.650964593887329,
      "epoch": 4.6615720524017465,
      "grad_norm": 1.5,
      "learning_rate": 0.0003026093241387259,
      "loss": 4.8725,
      "mean_token_accuracy": 0.20275321751832961,
      "num_tokens": 102904161.0,
      "step": 46970
    },
    {
      "entropy": 5.6413350105285645,
      "epoch": 4.662068281063914,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003025741600631915,
      "loss": 4.8946,
      "mean_token_accuracy": 0.19757098704576492,
      "num_tokens": 102915189.0,
      "step": 46975
    },
    {
      "entropy": 5.6026623249053955,
      "epoch": 4.6625645097260815,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003025389953038773,
      "loss": 4.8471,
      "mean_token_accuracy": 0.19815538674592972,
      "num_tokens": 102924797.0,
      "step": 46980
    },
    {
      "entropy": 5.665777015686035,
      "epoch": 4.663060738388249,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003025038298616554,
      "loss": 4.9895,
      "mean_token_accuracy": 0.19085992574691774,
      "num_tokens": 102936900.0,
      "step": 46985
    },
    {
      "entropy": 5.723062658309937,
      "epoch": 4.6635569670504164,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030246866373739764,
      "loss": 4.9348,
      "mean_token_accuracy": 0.1910008445382118,
      "num_tokens": 102948631.0,
      "step": 46990
    },
    {
      "entropy": 5.706612062454224,
      "epoch": 4.664053195712585,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030243349693197637,
      "loss": 4.9634,
      "mean_token_accuracy": 0.19714245945215225,
      "num_tokens": 102959866.0,
      "step": 46995
    },
    {
      "entropy": 5.691524410247803,
      "epoch": 4.664549424374751,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003023983294462634,
      "loss": 4.928,
      "mean_token_accuracy": 0.1966942310333252,
      "num_tokens": 102971256.0,
      "step": 47000
    },
    {
      "entropy": 5.744496059417725,
      "epoch": 4.66504565303692,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030236316128113086,
      "loss": 5.0085,
      "mean_token_accuracy": 0.1897053375840187,
      "num_tokens": 102985178.0,
      "step": 47005
    },
    {
      "entropy": 5.747043657302856,
      "epoch": 4.665541881699087,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030232799243745103,
      "loss": 4.9919,
      "mean_token_accuracy": 0.1986824244260788,
      "num_tokens": 102995405.0,
      "step": 47010
    },
    {
      "entropy": 5.766772842407226,
      "epoch": 4.666038110361255,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003022928229160958,
      "loss": 5.0373,
      "mean_token_accuracy": 0.19238607287406922,
      "num_tokens": 103005592.0,
      "step": 47015
    },
    {
      "entropy": 5.652877855300903,
      "epoch": 4.666534339023422,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003022576527179373,
      "loss": 4.9081,
      "mean_token_accuracy": 0.19537358433008195,
      "num_tokens": 103014941.0,
      "step": 47020
    },
    {
      "entropy": 5.645087432861328,
      "epoch": 4.66703056768559,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003022224818438478,
      "loss": 5.0023,
      "mean_token_accuracy": 0.19153235107660294,
      "num_tokens": 103025414.0,
      "step": 47025
    },
    {
      "entropy": 5.735861301422119,
      "epoch": 4.667526796347757,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003021873102946995,
      "loss": 5.0518,
      "mean_token_accuracy": 0.19058483541011811,
      "num_tokens": 103036934.0,
      "step": 47030
    },
    {
      "entropy": 5.6981476783752445,
      "epoch": 4.668023025009925,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003021521380713644,
      "loss": 4.9122,
      "mean_token_accuracy": 0.19526703655719757,
      "num_tokens": 103047981.0,
      "step": 47035
    },
    {
      "entropy": 5.819955730438233,
      "epoch": 4.668519253672092,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003021169651747148,
      "loss": 5.1035,
      "mean_token_accuracy": 0.18340202271938325,
      "num_tokens": 103059125.0,
      "step": 47040
    },
    {
      "entropy": 5.714672040939331,
      "epoch": 4.66901548233426,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003020817916056229,
      "loss": 4.9583,
      "mean_token_accuracy": 0.1943048432469368,
      "num_tokens": 103069968.0,
      "step": 47045
    },
    {
      "entropy": 5.668590688705445,
      "epoch": 4.669511710996427,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003020466173649608,
      "loss": 4.9624,
      "mean_token_accuracy": 0.19513297080993652,
      "num_tokens": 103080668.0,
      "step": 47050
    },
    {
      "entropy": 5.753994655609131,
      "epoch": 4.6700079396585945,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000302011442453601,
      "loss": 5.0079,
      "mean_token_accuracy": 0.19091798216104508,
      "num_tokens": 103091282.0,
      "step": 47055
    },
    {
      "entropy": 5.708614826202393,
      "epoch": 4.670504168320762,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030197626687241563,
      "loss": 4.9509,
      "mean_token_accuracy": 0.19081419706344604,
      "num_tokens": 103101792.0,
      "step": 47060
    },
    {
      "entropy": 5.731016826629639,
      "epoch": 4.6710003969829295,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000301941090622277,
      "loss": 5.0282,
      "mean_token_accuracy": 0.18942274898290634,
      "num_tokens": 103115706.0,
      "step": 47065
    },
    {
      "entropy": 5.614355611801147,
      "epoch": 4.671496625645097,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003019059137040573,
      "loss": 4.8425,
      "mean_token_accuracy": 0.19886002242565154,
      "num_tokens": 103125561.0,
      "step": 47070
    },
    {
      "entropy": 5.712661027908325,
      "epoch": 4.671992854307264,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003018707361186289,
      "loss": 4.941,
      "mean_token_accuracy": 0.19188328236341476,
      "num_tokens": 103136457.0,
      "step": 47075
    },
    {
      "entropy": 5.7149800777435305,
      "epoch": 4.672489082969433,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003018355578668642,
      "loss": 4.9314,
      "mean_token_accuracy": 0.19346631765365602,
      "num_tokens": 103147194.0,
      "step": 47080
    },
    {
      "entropy": 5.634368515014648,
      "epoch": 4.672985311631599,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030180037894963547,
      "loss": 4.9223,
      "mean_token_accuracy": 0.19539567828178406,
      "num_tokens": 103157011.0,
      "step": 47085
    },
    {
      "entropy": 5.825176143646241,
      "epoch": 4.673481540293768,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003017651993678152,
      "loss": 5.1974,
      "mean_token_accuracy": 0.1795124366879463,
      "num_tokens": 103169096.0,
      "step": 47090
    },
    {
      "entropy": 5.687720203399659,
      "epoch": 4.673977768955935,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030173001912227554,
      "loss": 4.9353,
      "mean_token_accuracy": 0.1902373880147934,
      "num_tokens": 103180432.0,
      "step": 47095
    },
    {
      "entropy": 5.681498193740845,
      "epoch": 4.674473997618103,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030169483821388906,
      "loss": 4.9547,
      "mean_token_accuracy": 0.20185620337724686,
      "num_tokens": 103191846.0,
      "step": 47100
    },
    {
      "entropy": 5.727250576019287,
      "epoch": 4.67497022628027,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003016596566435281,
      "loss": 5.0144,
      "mean_token_accuracy": 0.1976683109998703,
      "num_tokens": 103202973.0,
      "step": 47105
    },
    {
      "entropy": 5.669390487670898,
      "epoch": 4.675466454942438,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000301624474412065,
      "loss": 4.9339,
      "mean_token_accuracy": 0.19497938454151154,
      "num_tokens": 103213282.0,
      "step": 47110
    },
    {
      "entropy": 5.683595275878906,
      "epoch": 4.675962683604605,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003015892915203724,
      "loss": 4.9852,
      "mean_token_accuracy": 0.18720868080854416,
      "num_tokens": 103224251.0,
      "step": 47115
    },
    {
      "entropy": 5.702043247222901,
      "epoch": 4.676458912266773,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030155410796932266,
      "loss": 5.002,
      "mean_token_accuracy": 0.1921985149383545,
      "num_tokens": 103234348.0,
      "step": 47120
    },
    {
      "entropy": 5.653249597549438,
      "epoch": 4.67695514092894,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030151892375978825,
      "loss": 4.8489,
      "mean_token_accuracy": 0.1990855246782303,
      "num_tokens": 103245659.0,
      "step": 47125
    },
    {
      "entropy": 5.6837241649627686,
      "epoch": 4.6774513695911075,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003014837388926417,
      "loss": 4.9615,
      "mean_token_accuracy": 0.1982384741306305,
      "num_tokens": 103256153.0,
      "step": 47130
    },
    {
      "entropy": 5.640386629104614,
      "epoch": 4.677947598253275,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030144855336875545,
      "loss": 4.9502,
      "mean_token_accuracy": 0.19675214290618898,
      "num_tokens": 103266944.0,
      "step": 47135
    },
    {
      "entropy": 5.672486305236816,
      "epoch": 4.6784438269154425,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030141336718900207,
      "loss": 4.947,
      "mean_token_accuracy": 0.19193984419107438,
      "num_tokens": 103277471.0,
      "step": 47140
    },
    {
      "entropy": 5.623363065719604,
      "epoch": 4.67894005557761,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030137818035425414,
      "loss": 4.9028,
      "mean_token_accuracy": 0.2000954732298851,
      "num_tokens": 103288710.0,
      "step": 47145
    },
    {
      "entropy": 5.699303531646729,
      "epoch": 4.679436284239777,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003013429928653841,
      "loss": 4.9321,
      "mean_token_accuracy": 0.19360824227333068,
      "num_tokens": 103300028.0,
      "step": 47150
    },
    {
      "entropy": 5.690800189971924,
      "epoch": 4.679932512901945,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003013078047232647,
      "loss": 4.9087,
      "mean_token_accuracy": 0.19834667593240737,
      "num_tokens": 103310694.0,
      "step": 47155
    },
    {
      "entropy": 5.740881729125976,
      "epoch": 4.680428741564112,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003012726159287684,
      "loss": 5.0482,
      "mean_token_accuracy": 0.19063550382852554,
      "num_tokens": 103321080.0,
      "step": 47160
    },
    {
      "entropy": 5.655057954788208,
      "epoch": 4.680924970226281,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003012374264827679,
      "loss": 4.9303,
      "mean_token_accuracy": 0.19272918701171876,
      "num_tokens": 103331368.0,
      "step": 47165
    },
    {
      "entropy": 5.666156435012818,
      "epoch": 4.681421198888447,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003012022363861356,
      "loss": 4.8939,
      "mean_token_accuracy": 0.2017025575041771,
      "num_tokens": 103341767.0,
      "step": 47170
    },
    {
      "entropy": 5.693143558502197,
      "epoch": 4.681917427550616,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003011670456397444,
      "loss": 4.9722,
      "mean_token_accuracy": 0.19113762229681014,
      "num_tokens": 103353800.0,
      "step": 47175
    },
    {
      "entropy": 5.6926063060760494,
      "epoch": 4.682413656212783,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003011318542444669,
      "loss": 4.8925,
      "mean_token_accuracy": 0.19859661012887955,
      "num_tokens": 103365223.0,
      "step": 47180
    },
    {
      "entropy": 5.738640213012696,
      "epoch": 4.682909884874951,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030109666220117556,
      "loss": 5.032,
      "mean_token_accuracy": 0.1880713015794754,
      "num_tokens": 103377551.0,
      "step": 47185
    },
    {
      "entropy": 5.747289800643921,
      "epoch": 4.683406113537118,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030106146951074336,
      "loss": 5.0241,
      "mean_token_accuracy": 0.1924239844083786,
      "num_tokens": 103387740.0,
      "step": 47190
    },
    {
      "entropy": 5.6946464538574215,
      "epoch": 4.683902342199286,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030102627617404286,
      "loss": 4.9694,
      "mean_token_accuracy": 0.19318527430295945,
      "num_tokens": 103399082.0,
      "step": 47195
    },
    {
      "entropy": 5.651343965530396,
      "epoch": 4.684398570861453,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003009910821919468,
      "loss": 4.8573,
      "mean_token_accuracy": 0.21452263593673707,
      "num_tokens": 103409991.0,
      "step": 47200
    },
    {
      "entropy": 5.642293500900268,
      "epoch": 4.6848947995236205,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030095588756532787,
      "loss": 4.9424,
      "mean_token_accuracy": 0.19159155935049058,
      "num_tokens": 103421578.0,
      "step": 47205
    },
    {
      "entropy": 5.646605014801025,
      "epoch": 4.685391028185788,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003009206922950589,
      "loss": 4.8879,
      "mean_token_accuracy": 0.20028677731752395,
      "num_tokens": 103432253.0,
      "step": 47210
    },
    {
      "entropy": 5.684602975845337,
      "epoch": 4.6858872568479555,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030088549638201256,
      "loss": 4.9547,
      "mean_token_accuracy": 0.20132564455270768,
      "num_tokens": 103443065.0,
      "step": 47215
    },
    {
      "entropy": 5.685303783416748,
      "epoch": 4.686383485510123,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003008502998270617,
      "loss": 4.9418,
      "mean_token_accuracy": 0.2012705147266388,
      "num_tokens": 103453558.0,
      "step": 47220
    },
    {
      "entropy": 5.682772922515869,
      "epoch": 4.68687971417229,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030081510263107916,
      "loss": 4.9329,
      "mean_token_accuracy": 0.19291384220123292,
      "num_tokens": 103464071.0,
      "step": 47225
    },
    {
      "entropy": 5.772492218017578,
      "epoch": 4.687375942834458,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003007799047949377,
      "loss": 5.0356,
      "mean_token_accuracy": 0.19228608310222625,
      "num_tokens": 103475874.0,
      "step": 47230
    },
    {
      "entropy": 5.769063568115234,
      "epoch": 4.687872171496625,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003007447063195101,
      "loss": 5.1077,
      "mean_token_accuracy": 0.18459657877683638,
      "num_tokens": 103486570.0,
      "step": 47235
    },
    {
      "entropy": 5.726995849609375,
      "epoch": 4.688368400158793,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030070950720566925,
      "loss": 4.9582,
      "mean_token_accuracy": 0.1886611118912697,
      "num_tokens": 103498324.0,
      "step": 47240
    },
    {
      "entropy": 5.7739036083221436,
      "epoch": 4.68886462882096,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000300674307454288,
      "loss": 4.9852,
      "mean_token_accuracy": 0.1937109649181366,
      "num_tokens": 103509136.0,
      "step": 47245
    },
    {
      "entropy": 5.61134967803955,
      "epoch": 4.689360857483129,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003006391070662393,
      "loss": 4.8947,
      "mean_token_accuracy": 0.19641469269990922,
      "num_tokens": 103519634.0,
      "step": 47250
    },
    {
      "entropy": 5.597188997268677,
      "epoch": 4.689857086145295,
      "grad_norm": 1.234375,
      "learning_rate": 0.000300603906042396,
      "loss": 4.889,
      "mean_token_accuracy": 0.1950989082455635,
      "num_tokens": 103529869.0,
      "step": 47255
    },
    {
      "entropy": 5.631833410263061,
      "epoch": 4.690353314807464,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000300568704383631,
      "loss": 4.9252,
      "mean_token_accuracy": 0.19382865726947784,
      "num_tokens": 103539602.0,
      "step": 47260
    },
    {
      "entropy": 5.6451304912567135,
      "epoch": 4.690849543469631,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003005335020908172,
      "loss": 4.983,
      "mean_token_accuracy": 0.19531867504119874,
      "num_tokens": 103551272.0,
      "step": 47265
    },
    {
      "entropy": 5.6909185409545895,
      "epoch": 4.691345772131799,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003004982991648275,
      "loss": 4.9106,
      "mean_token_accuracy": 0.19404202401638032,
      "num_tokens": 103561879.0,
      "step": 47270
    },
    {
      "entropy": 5.682623958587646,
      "epoch": 4.691842000793966,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030046309560653506,
      "loss": 4.996,
      "mean_token_accuracy": 0.19256454557180405,
      "num_tokens": 103572098.0,
      "step": 47275
    },
    {
      "entropy": 5.662697887420654,
      "epoch": 4.6923382294561335,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003004278914168127,
      "loss": 4.9027,
      "mean_token_accuracy": 0.19860660582780837,
      "num_tokens": 103583818.0,
      "step": 47280
    },
    {
      "entropy": 5.683025169372558,
      "epoch": 4.692834458118301,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030039268659653334,
      "loss": 4.9541,
      "mean_token_accuracy": 0.1914561942219734,
      "num_tokens": 103595394.0,
      "step": 47285
    },
    {
      "entropy": 5.667491817474366,
      "epoch": 4.6933306867804685,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003003574811465702,
      "loss": 4.9183,
      "mean_token_accuracy": 0.1933379963040352,
      "num_tokens": 103605665.0,
      "step": 47290
    },
    {
      "entropy": 5.669033241271973,
      "epoch": 4.693826915442636,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030032227506779607,
      "loss": 4.9837,
      "mean_token_accuracy": 0.2055818870663643,
      "num_tokens": 103616710.0,
      "step": 47295
    },
    {
      "entropy": 5.694937133789063,
      "epoch": 4.6943231441048034,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030028706836108404,
      "loss": 4.8914,
      "mean_token_accuracy": 0.20944703072309495,
      "num_tokens": 103626436.0,
      "step": 47300
    },
    {
      "entropy": 5.693309307098389,
      "epoch": 4.694819372766971,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030025186102730736,
      "loss": 4.9228,
      "mean_token_accuracy": 0.1955860137939453,
      "num_tokens": 103636677.0,
      "step": 47305
    },
    {
      "entropy": 5.678955554962158,
      "epoch": 4.695315601429138,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003002166530673389,
      "loss": 4.9505,
      "mean_token_accuracy": 0.19146050214767457,
      "num_tokens": 103648173.0,
      "step": 47310
    },
    {
      "entropy": 5.694035768508911,
      "epoch": 4.695811830091306,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003001814444820516,
      "loss": 4.9723,
      "mean_token_accuracy": 0.1960570588707924,
      "num_tokens": 103659044.0,
      "step": 47315
    },
    {
      "entropy": 5.65799765586853,
      "epoch": 4.696308058753473,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003001462352723189,
      "loss": 4.9268,
      "mean_token_accuracy": 0.19633639901876448,
      "num_tokens": 103669371.0,
      "step": 47320
    },
    {
      "entropy": 5.73606538772583,
      "epoch": 4.696804287415641,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030011102543901373,
      "loss": 4.9378,
      "mean_token_accuracy": 0.19143574088811874,
      "num_tokens": 103680001.0,
      "step": 47325
    },
    {
      "entropy": 5.670007371902466,
      "epoch": 4.697300516077808,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030007581498300915,
      "loss": 4.8959,
      "mean_token_accuracy": 0.19681215435266494,
      "num_tokens": 103690542.0,
      "step": 47330
    },
    {
      "entropy": 5.622045421600342,
      "epoch": 4.697796744739977,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030004060390517853,
      "loss": 4.8693,
      "mean_token_accuracy": 0.19823072701692582,
      "num_tokens": 103701904.0,
      "step": 47335
    },
    {
      "entropy": 5.627264356613159,
      "epoch": 4.698292973402143,
      "grad_norm": 1.25,
      "learning_rate": 0.00030000539220639486,
      "loss": 4.8353,
      "mean_token_accuracy": 0.20319802612066268,
      "num_tokens": 103713204.0,
      "step": 47340
    },
    {
      "entropy": 5.695733261108399,
      "epoch": 4.698789202064312,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029997017988753124,
      "loss": 4.9522,
      "mean_token_accuracy": 0.18968218863010405,
      "num_tokens": 103723876.0,
      "step": 47345
    },
    {
      "entropy": 5.643178081512451,
      "epoch": 4.699285430726479,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029993496694946106,
      "loss": 4.9031,
      "mean_token_accuracy": 0.1962951675057411,
      "num_tokens": 103735455.0,
      "step": 47350
    },
    {
      "entropy": 5.641950464248657,
      "epoch": 4.699781659388647,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002998997533930574,
      "loss": 4.9036,
      "mean_token_accuracy": 0.20086462199687957,
      "num_tokens": 103744987.0,
      "step": 47355
    },
    {
      "entropy": 5.687230968475342,
      "epoch": 4.700277888050814,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029986453921919355,
      "loss": 4.96,
      "mean_token_accuracy": 0.1913190335035324,
      "num_tokens": 103755484.0,
      "step": 47360
    },
    {
      "entropy": 5.607392930984497,
      "epoch": 4.7007741167129815,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002998293244287426,
      "loss": 4.922,
      "mean_token_accuracy": 0.19478460550308227,
      "num_tokens": 103766983.0,
      "step": 47365
    },
    {
      "entropy": 5.720554542541504,
      "epoch": 4.701270345375149,
      "grad_norm": 1.234375,
      "learning_rate": 0.000299794109022578,
      "loss": 4.9994,
      "mean_token_accuracy": 0.1967352956533432,
      "num_tokens": 103777934.0,
      "step": 47370
    },
    {
      "entropy": 5.744562101364136,
      "epoch": 4.7017665740373165,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002997588930015729,
      "loss": 4.9842,
      "mean_token_accuracy": 0.1875744566321373,
      "num_tokens": 103788356.0,
      "step": 47375
    },
    {
      "entropy": 5.642034673690796,
      "epoch": 4.702262802699484,
      "grad_norm": 1.375,
      "learning_rate": 0.0002997236763666006,
      "loss": 4.877,
      "mean_token_accuracy": 0.20523356795310974,
      "num_tokens": 103798972.0,
      "step": 47380
    },
    {
      "entropy": 5.697215127944946,
      "epoch": 4.702759031361651,
      "grad_norm": 1.25,
      "learning_rate": 0.0002996884591185344,
      "loss": 5.0138,
      "mean_token_accuracy": 0.1885112151503563,
      "num_tokens": 103810743.0,
      "step": 47385
    },
    {
      "entropy": 5.687448501586914,
      "epoch": 4.703255260023819,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029965324125824763,
      "loss": 4.9402,
      "mean_token_accuracy": 0.19570014923810958,
      "num_tokens": 103821085.0,
      "step": 47390
    },
    {
      "entropy": 5.698429584503174,
      "epoch": 4.703751488685986,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002996180227866136,
      "loss": 4.9567,
      "mean_token_accuracy": 0.19261558651924132,
      "num_tokens": 103832562.0,
      "step": 47395
    },
    {
      "entropy": 5.678034830093384,
      "epoch": 4.704247717348154,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002995828037045056,
      "loss": 4.9763,
      "mean_token_accuracy": 0.19351196587085723,
      "num_tokens": 103843144.0,
      "step": 47400
    },
    {
      "entropy": 5.761011457443237,
      "epoch": 4.704743946010321,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002995475840127971,
      "loss": 5.0391,
      "mean_token_accuracy": 0.18622923493385315,
      "num_tokens": 103853551.0,
      "step": 47405
    },
    {
      "entropy": 5.663875579833984,
      "epoch": 4.705240174672489,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002995123637123614,
      "loss": 4.9527,
      "mean_token_accuracy": 0.19622737020254136,
      "num_tokens": 103864929.0,
      "step": 47410
    },
    {
      "entropy": 5.739672470092773,
      "epoch": 4.705736403334656,
      "grad_norm": 1.125,
      "learning_rate": 0.0002994771428040718,
      "loss": 5.0434,
      "mean_token_accuracy": 0.19254794269800185,
      "num_tokens": 103875545.0,
      "step": 47415
    },
    {
      "entropy": 5.739242839813232,
      "epoch": 4.706232631996825,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029944192128880193,
      "loss": 5.0878,
      "mean_token_accuracy": 0.18265020102262497,
      "num_tokens": 103889084.0,
      "step": 47420
    },
    {
      "entropy": 5.701719856262207,
      "epoch": 4.706728860658991,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029940669916742507,
      "loss": 4.9155,
      "mean_token_accuracy": 0.19372745156288146,
      "num_tokens": 103900043.0,
      "step": 47425
    },
    {
      "entropy": 5.790539360046386,
      "epoch": 4.70722508932116,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029937147644081463,
      "loss": 5.0425,
      "mean_token_accuracy": 0.19044874012470245,
      "num_tokens": 103911174.0,
      "step": 47430
    },
    {
      "entropy": 5.628839445114136,
      "epoch": 4.707721317983327,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002993362531098441,
      "loss": 4.8733,
      "mean_token_accuracy": 0.20352516025304795,
      "num_tokens": 103922882.0,
      "step": 47435
    },
    {
      "entropy": 5.59546012878418,
      "epoch": 4.7082175466454945,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002993010291753869,
      "loss": 4.8963,
      "mean_token_accuracy": 0.1998228520154953,
      "num_tokens": 103935695.0,
      "step": 47440
    },
    {
      "entropy": 5.689451694488525,
      "epoch": 4.708713775307662,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002992658046383166,
      "loss": 4.9157,
      "mean_token_accuracy": 0.1920287013053894,
      "num_tokens": 103945966.0,
      "step": 47445
    },
    {
      "entropy": 5.745244455337525,
      "epoch": 4.7092100039698295,
      "grad_norm": 1.375,
      "learning_rate": 0.00029923057949950653,
      "loss": 4.9765,
      "mean_token_accuracy": 0.19976810216903687,
      "num_tokens": 103956078.0,
      "step": 47450
    },
    {
      "entropy": 5.776128149032592,
      "epoch": 4.709706232631997,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029919535375983044,
      "loss": 5.0542,
      "mean_token_accuracy": 0.19692608565092087,
      "num_tokens": 103967585.0,
      "step": 47455
    },
    {
      "entropy": 5.64766035079956,
      "epoch": 4.710202461294164,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029916012742016164,
      "loss": 4.9416,
      "mean_token_accuracy": 0.19159546345472336,
      "num_tokens": 103977914.0,
      "step": 47460
    },
    {
      "entropy": 5.727771902084351,
      "epoch": 4.710698689956332,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002991249004813737,
      "loss": 4.9938,
      "mean_token_accuracy": 0.18919770270586014,
      "num_tokens": 103990291.0,
      "step": 47465
    },
    {
      "entropy": 5.781078958511353,
      "epoch": 4.711194918618499,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029908967294434033,
      "loss": 5.0225,
      "mean_token_accuracy": 0.19646597355604173,
      "num_tokens": 104002141.0,
      "step": 47470
    },
    {
      "entropy": 5.6559406280517575,
      "epoch": 4.711691147280667,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002990544448099349,
      "loss": 4.8574,
      "mean_token_accuracy": 0.20443433970212938,
      "num_tokens": 104012992.0,
      "step": 47475
    },
    {
      "entropy": 5.691263675689697,
      "epoch": 4.712187375942834,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000299019216079031,
      "loss": 4.9064,
      "mean_token_accuracy": 0.20106273740530015,
      "num_tokens": 104022548.0,
      "step": 47480
    },
    {
      "entropy": 5.684582901000977,
      "epoch": 4.712683604605002,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002989839867525024,
      "loss": 4.957,
      "mean_token_accuracy": 0.1965175077319145,
      "num_tokens": 104034461.0,
      "step": 47485
    },
    {
      "entropy": 5.733966588973999,
      "epoch": 4.713179833267169,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029894875683122264,
      "loss": 4.9447,
      "mean_token_accuracy": 0.18908920586109162,
      "num_tokens": 104044929.0,
      "step": 47490
    },
    {
      "entropy": 5.767256546020508,
      "epoch": 4.713676061929337,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002989135263160652,
      "loss": 5.0952,
      "mean_token_accuracy": 0.18586160689592363,
      "num_tokens": 104056474.0,
      "step": 47495
    },
    {
      "entropy": 5.724827146530151,
      "epoch": 4.714172290591504,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029887829520790395,
      "loss": 4.9159,
      "mean_token_accuracy": 0.2009772166609764,
      "num_tokens": 104066614.0,
      "step": 47500
    },
    {
      "entropy": 5.70154299736023,
      "epoch": 4.714668519253673,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002988430635076124,
      "loss": 5.0059,
      "mean_token_accuracy": 0.19119479656219482,
      "num_tokens": 104079519.0,
      "step": 47505
    },
    {
      "entropy": 5.6299357414245605,
      "epoch": 4.715164747915839,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029880783121606424,
      "loss": 4.8917,
      "mean_token_accuracy": 0.19975274801254272,
      "num_tokens": 104090555.0,
      "step": 47510
    },
    {
      "entropy": 5.603178453445435,
      "epoch": 4.7156609765780075,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002987725983341331,
      "loss": 4.8369,
      "mean_token_accuracy": 0.20319015383720399,
      "num_tokens": 104100453.0,
      "step": 47515
    },
    {
      "entropy": 5.578611564636231,
      "epoch": 4.716157205240175,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029873736486269283,
      "loss": 4.8868,
      "mean_token_accuracy": 0.1979964166879654,
      "num_tokens": 104112361.0,
      "step": 47520
    },
    {
      "entropy": 5.659583377838135,
      "epoch": 4.7166534339023425,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029870213080261703,
      "loss": 4.9744,
      "mean_token_accuracy": 0.18975224792957307,
      "num_tokens": 104123020.0,
      "step": 47525
    },
    {
      "entropy": 5.696064615249634,
      "epoch": 4.71714966256451,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002986668961547795,
      "loss": 4.969,
      "mean_token_accuracy": 0.1927126243710518,
      "num_tokens": 104133437.0,
      "step": 47530
    },
    {
      "entropy": 5.708628463745117,
      "epoch": 4.717645891226677,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002986316609200538,
      "loss": 4.9496,
      "mean_token_accuracy": 0.1954178273677826,
      "num_tokens": 104143493.0,
      "step": 47535
    },
    {
      "entropy": 5.607433319091797,
      "epoch": 4.718142119888845,
      "grad_norm": 1.234375,
      "learning_rate": 0.000298596425099314,
      "loss": 4.8423,
      "mean_token_accuracy": 0.2012164518237114,
      "num_tokens": 104153741.0,
      "step": 47540
    },
    {
      "entropy": 5.66590313911438,
      "epoch": 4.718638348551012,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002985611886934337,
      "loss": 4.9349,
      "mean_token_accuracy": 0.19913468211889268,
      "num_tokens": 104164184.0,
      "step": 47545
    },
    {
      "entropy": 5.654748106002808,
      "epoch": 4.71913457721318,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029852595170328655,
      "loss": 4.9142,
      "mean_token_accuracy": 0.19849701970815659,
      "num_tokens": 104175853.0,
      "step": 47550
    },
    {
      "entropy": 5.635046815872192,
      "epoch": 4.719630805875347,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002984907141297466,
      "loss": 4.9196,
      "mean_token_accuracy": 0.19603333324193956,
      "num_tokens": 104186662.0,
      "step": 47555
    },
    {
      "entropy": 5.67673544883728,
      "epoch": 4.720127034537515,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002984554759736875,
      "loss": 4.9619,
      "mean_token_accuracy": 0.20135591626167298,
      "num_tokens": 104197217.0,
      "step": 47560
    },
    {
      "entropy": 5.692944288253784,
      "epoch": 4.720623263199682,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029842023723598317,
      "loss": 4.9547,
      "mean_token_accuracy": 0.19480833262205124,
      "num_tokens": 104207752.0,
      "step": 47565
    },
    {
      "entropy": 5.661230564117432,
      "epoch": 4.72111949186185,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002983849979175074,
      "loss": 4.9274,
      "mean_token_accuracy": 0.19261330515146255,
      "num_tokens": 104218212.0,
      "step": 47570
    },
    {
      "entropy": 5.7239377975463865,
      "epoch": 4.721615720524017,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029834975801913417,
      "loss": 5.0503,
      "mean_token_accuracy": 0.1896983951330185,
      "num_tokens": 104229518.0,
      "step": 47575
    },
    {
      "entropy": 5.714883089065552,
      "epoch": 4.722111949186185,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029831451754173716,
      "loss": 4.9696,
      "mean_token_accuracy": 0.20622386485338212,
      "num_tokens": 104239549.0,
      "step": 47580
    },
    {
      "entropy": 5.703200101852417,
      "epoch": 4.722608177848352,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002982792764861904,
      "loss": 4.9359,
      "mean_token_accuracy": 0.200136935710907,
      "num_tokens": 104250541.0,
      "step": 47585
    },
    {
      "entropy": 5.663870143890381,
      "epoch": 4.72310440651052,
      "grad_norm": 1.375,
      "learning_rate": 0.0002982440348533678,
      "loss": 4.8889,
      "mean_token_accuracy": 0.1969560995697975,
      "num_tokens": 104260861.0,
      "step": 47590
    },
    {
      "entropy": 5.58651556968689,
      "epoch": 4.723600635172687,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029820879264414303,
      "loss": 4.9056,
      "mean_token_accuracy": 0.19740413576364518,
      "num_tokens": 104272329.0,
      "step": 47595
    },
    {
      "entropy": 5.678631019592285,
      "epoch": 4.7240968638348555,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002981735498593903,
      "loss": 4.9167,
      "mean_token_accuracy": 0.19819215834140777,
      "num_tokens": 104283714.0,
      "step": 47600
    },
    {
      "entropy": 5.677149295806885,
      "epoch": 4.724593092497023,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002981383064999836,
      "loss": 4.9381,
      "mean_token_accuracy": 0.20142599642276765,
      "num_tokens": 104294355.0,
      "step": 47605
    },
    {
      "entropy": 5.726099634170533,
      "epoch": 4.7250893211591904,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029810306256679655,
      "loss": 5.0509,
      "mean_token_accuracy": 0.18995645344257356,
      "num_tokens": 104304897.0,
      "step": 47610
    },
    {
      "entropy": 5.694409132003784,
      "epoch": 4.725585549821358,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002980678180607035,
      "loss": 4.9751,
      "mean_token_accuracy": 0.19242900013923644,
      "num_tokens": 104316914.0,
      "step": 47615
    },
    {
      "entropy": 5.694502258300782,
      "epoch": 4.726081778483525,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002980325729825782,
      "loss": 4.9487,
      "mean_token_accuracy": 0.1948195829987526,
      "num_tokens": 104327712.0,
      "step": 47620
    },
    {
      "entropy": 5.676952791213989,
      "epoch": 4.726578007145693,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029799732733329473,
      "loss": 4.9904,
      "mean_token_accuracy": 0.18764204680919647,
      "num_tokens": 104339884.0,
      "step": 47625
    },
    {
      "entropy": 5.653267288208008,
      "epoch": 4.72707423580786,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029796208111372704,
      "loss": 4.9616,
      "mean_token_accuracy": 0.1966905862092972,
      "num_tokens": 104351235.0,
      "step": 47630
    },
    {
      "entropy": 5.689508438110352,
      "epoch": 4.727570464470028,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002979268343247493,
      "loss": 5.0003,
      "mean_token_accuracy": 0.19306988269090652,
      "num_tokens": 104364143.0,
      "step": 47635
    },
    {
      "entropy": 5.681148672103882,
      "epoch": 4.728066693132195,
      "grad_norm": 1.25,
      "learning_rate": 0.0002978915869672354,
      "loss": 4.8639,
      "mean_token_accuracy": 0.19612168222665788,
      "num_tokens": 104373916.0,
      "step": 47640
    },
    {
      "entropy": 5.667503213882446,
      "epoch": 4.728562921794363,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002978563390420595,
      "loss": 4.8779,
      "mean_token_accuracy": 0.20351359993219376,
      "num_tokens": 104384494.0,
      "step": 47645
    },
    {
      "entropy": 5.652315711975097,
      "epoch": 4.72905915045653,
      "grad_norm": 1.25,
      "learning_rate": 0.00029782109055009567,
      "loss": 5.008,
      "mean_token_accuracy": 0.19352127462625504,
      "num_tokens": 104395911.0,
      "step": 47650
    },
    {
      "entropy": 5.707019186019897,
      "epoch": 4.729555379118698,
      "grad_norm": 1.375,
      "learning_rate": 0.00029778584149221793,
      "loss": 4.8915,
      "mean_token_accuracy": 0.2006876826286316,
      "num_tokens": 104405870.0,
      "step": 47655
    },
    {
      "entropy": 5.663105010986328,
      "epoch": 4.730051607780865,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002977505918693004,
      "loss": 4.9243,
      "mean_token_accuracy": 0.19817046970129013,
      "num_tokens": 104417308.0,
      "step": 47660
    },
    {
      "entropy": 5.705452728271484,
      "epoch": 4.730547836443033,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002977153416822173,
      "loss": 4.9937,
      "mean_token_accuracy": 0.19884045869112016,
      "num_tokens": 104429059.0,
      "step": 47665
    },
    {
      "entropy": 5.776842784881592,
      "epoch": 4.7310440651052,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002976800909318426,
      "loss": 5.0233,
      "mean_token_accuracy": 0.1941621333360672,
      "num_tokens": 104440812.0,
      "step": 47670
    },
    {
      "entropy": 5.7136718273162845,
      "epoch": 4.731540293767368,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029764483961905057,
      "loss": 4.9508,
      "mean_token_accuracy": 0.19533547312021254,
      "num_tokens": 104451944.0,
      "step": 47675
    },
    {
      "entropy": 5.6299147605896,
      "epoch": 4.732036522429535,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002976095877447153,
      "loss": 4.8805,
      "mean_token_accuracy": 0.19447574466466905,
      "num_tokens": 104462654.0,
      "step": 47680
    },
    {
      "entropy": 5.677879476547242,
      "epoch": 4.7325327510917035,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002975743353097109,
      "loss": 4.9471,
      "mean_token_accuracy": 0.1936282530426979,
      "num_tokens": 104473192.0,
      "step": 47685
    },
    {
      "entropy": 5.698440074920654,
      "epoch": 4.733028979753871,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002975390823149117,
      "loss": 4.9497,
      "mean_token_accuracy": 0.19309282153844834,
      "num_tokens": 104482669.0,
      "step": 47690
    },
    {
      "entropy": 5.738064002990723,
      "epoch": 4.733525208416038,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029750382876119185,
      "loss": 5.0027,
      "mean_token_accuracy": 0.19297091662883759,
      "num_tokens": 104493373.0,
      "step": 47695
    },
    {
      "entropy": 5.627460384368897,
      "epoch": 4.734021437078206,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029746857464942543,
      "loss": 4.8234,
      "mean_token_accuracy": 0.21019292771816253,
      "num_tokens": 104504218.0,
      "step": 47700
    },
    {
      "entropy": 5.708858823776245,
      "epoch": 4.734517665740373,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029743331998048693,
      "loss": 4.9592,
      "mean_token_accuracy": 0.1963695466518402,
      "num_tokens": 104515323.0,
      "step": 47705
    },
    {
      "entropy": 5.672468471527099,
      "epoch": 4.735013894402541,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029739806475525036,
      "loss": 5.0165,
      "mean_token_accuracy": 0.19056018441915512,
      "num_tokens": 104526926.0,
      "step": 47710
    },
    {
      "entropy": 5.750648546218872,
      "epoch": 4.735510123064708,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029736280897459,
      "loss": 5.0971,
      "mean_token_accuracy": 0.18722994178533553,
      "num_tokens": 104538605.0,
      "step": 47715
    },
    {
      "entropy": 5.713402986526489,
      "epoch": 4.736006351726876,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002973275526393803,
      "loss": 4.9958,
      "mean_token_accuracy": 0.19428921341896058,
      "num_tokens": 104551239.0,
      "step": 47720
    },
    {
      "entropy": 5.749577903747559,
      "epoch": 4.736502580389043,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029729229575049516,
      "loss": 4.9904,
      "mean_token_accuracy": 0.19128995537757873,
      "num_tokens": 104562111.0,
      "step": 47725
    },
    {
      "entropy": 5.762090539932251,
      "epoch": 4.736998809051211,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002972570383088093,
      "loss": 5.0003,
      "mean_token_accuracy": 0.18870515525341033,
      "num_tokens": 104572011.0,
      "step": 47730
    },
    {
      "entropy": 5.65158224105835,
      "epoch": 4.737495037713378,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029722178031519686,
      "loss": 4.8917,
      "mean_token_accuracy": 0.19762710034847258,
      "num_tokens": 104583416.0,
      "step": 47735
    },
    {
      "entropy": 5.60941514968872,
      "epoch": 4.737991266375546,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002971865217705321,
      "loss": 4.9502,
      "mean_token_accuracy": 0.1922362744808197,
      "num_tokens": 104594581.0,
      "step": 47740
    },
    {
      "entropy": 5.741798782348633,
      "epoch": 4.738487495037713,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029715126267568936,
      "loss": 5.0436,
      "mean_token_accuracy": 0.19292730689048768,
      "num_tokens": 104606730.0,
      "step": 47745
    },
    {
      "entropy": 5.679840755462647,
      "epoch": 4.738983723699881,
      "grad_norm": 1.25,
      "learning_rate": 0.00029711600303154316,
      "loss": 4.9334,
      "mean_token_accuracy": 0.19473861753940583,
      "num_tokens": 104617449.0,
      "step": 47750
    },
    {
      "entropy": 5.631608772277832,
      "epoch": 4.739479952362048,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002970807428389677,
      "loss": 4.8978,
      "mean_token_accuracy": 0.19914721846580505,
      "num_tokens": 104628386.0,
      "step": 47755
    },
    {
      "entropy": 5.665998649597168,
      "epoch": 4.739976181024216,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002970454820988373,
      "loss": 4.9234,
      "mean_token_accuracy": 0.1929865375161171,
      "num_tokens": 104640358.0,
      "step": 47760
    },
    {
      "entropy": 5.667052602767944,
      "epoch": 4.740472409686383,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002970102208120266,
      "loss": 4.885,
      "mean_token_accuracy": 0.1940722092986107,
      "num_tokens": 104650317.0,
      "step": 47765
    },
    {
      "entropy": 5.65966534614563,
      "epoch": 4.740968638348551,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029697495897940975,
      "loss": 4.9154,
      "mean_token_accuracy": 0.19471255987882613,
      "num_tokens": 104661133.0,
      "step": 47770
    },
    {
      "entropy": 5.750348234176636,
      "epoch": 4.741464867010719,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002969396966018613,
      "loss": 5.0152,
      "mean_token_accuracy": 0.1893475517630577,
      "num_tokens": 104671939.0,
      "step": 47775
    },
    {
      "entropy": 5.63973035812378,
      "epoch": 4.741961095672886,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002969044336802557,
      "loss": 4.9192,
      "mean_token_accuracy": 0.20162352472543715,
      "num_tokens": 104682251.0,
      "step": 47780
    },
    {
      "entropy": 5.68262939453125,
      "epoch": 4.742457324335054,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002968691702154672,
      "loss": 4.9274,
      "mean_token_accuracy": 0.19701522290706636,
      "num_tokens": 104693791.0,
      "step": 47785
    },
    {
      "entropy": 5.704510641098023,
      "epoch": 4.742953552997221,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029683390620837053,
      "loss": 4.9072,
      "mean_token_accuracy": 0.19538863897323608,
      "num_tokens": 104704262.0,
      "step": 47790
    },
    {
      "entropy": 5.790560483932495,
      "epoch": 4.743449781659389,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029679864165984,
      "loss": 5.0287,
      "mean_token_accuracy": 0.19068979173898698,
      "num_tokens": 104714194.0,
      "step": 47795
    },
    {
      "entropy": 5.688005876541138,
      "epoch": 4.743946010321556,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002967633765707501,
      "loss": 4.9748,
      "mean_token_accuracy": 0.1933895915746689,
      "num_tokens": 104725124.0,
      "step": 47800
    },
    {
      "entropy": 5.717849016189575,
      "epoch": 4.744442238983724,
      "grad_norm": 1.25,
      "learning_rate": 0.00029672811094197534,
      "loss": 4.9579,
      "mean_token_accuracy": 0.19033554196357727,
      "num_tokens": 104735827.0,
      "step": 47805
    },
    {
      "entropy": 5.707391691207886,
      "epoch": 4.744938467645891,
      "grad_norm": 1.125,
      "learning_rate": 0.00029669284477439034,
      "loss": 4.9525,
      "mean_token_accuracy": 0.19589215219020845,
      "num_tokens": 104748289.0,
      "step": 47810
    },
    {
      "entropy": 5.620173549652099,
      "epoch": 4.745434696308059,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029665757806886944,
      "loss": 4.8666,
      "mean_token_accuracy": 0.20324863344430924,
      "num_tokens": 104760228.0,
      "step": 47815
    },
    {
      "entropy": 5.662514305114746,
      "epoch": 4.745930924970226,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002966223108262873,
      "loss": 4.9121,
      "mean_token_accuracy": 0.20036996454000472,
      "num_tokens": 104769942.0,
      "step": 47820
    },
    {
      "entropy": 5.714241981506348,
      "epoch": 4.746427153632394,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029658704304751847,
      "loss": 4.9659,
      "mean_token_accuracy": 0.1939360335469246,
      "num_tokens": 104780502.0,
      "step": 47825
    },
    {
      "entropy": 5.640191078186035,
      "epoch": 4.746923382294561,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002965517747334375,
      "loss": 4.9571,
      "mean_token_accuracy": 0.1909611240029335,
      "num_tokens": 104793427.0,
      "step": 47830
    },
    {
      "entropy": 5.719949436187744,
      "epoch": 4.747419610956729,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029651650588491895,
      "loss": 4.9771,
      "mean_token_accuracy": 0.19621999114751815,
      "num_tokens": 104804123.0,
      "step": 47835
    },
    {
      "entropy": 5.692140054702759,
      "epoch": 4.747915839618896,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00029648123650283736,
      "loss": 4.9272,
      "mean_token_accuracy": 0.1912304475903511,
      "num_tokens": 104815118.0,
      "step": 47840
    },
    {
      "entropy": 5.690591526031494,
      "epoch": 4.7484120682810635,
      "grad_norm": 1.25,
      "learning_rate": 0.0002964459665880674,
      "loss": 4.973,
      "mean_token_accuracy": 0.1905797839164734,
      "num_tokens": 104824702.0,
      "step": 47845
    },
    {
      "entropy": 5.590579986572266,
      "epoch": 4.748908296943231,
      "grad_norm": 1.25,
      "learning_rate": 0.0002964106961414838,
      "loss": 4.9175,
      "mean_token_accuracy": 0.19574231952428817,
      "num_tokens": 104835356.0,
      "step": 47850
    },
    {
      "entropy": 5.765819692611695,
      "epoch": 4.749404525605399,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029637542516396093,
      "loss": 4.9799,
      "mean_token_accuracy": 0.1946987882256508,
      "num_tokens": 104847419.0,
      "step": 47855
    },
    {
      "entropy": 5.686941337585449,
      "epoch": 4.749900754267567,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002963401536563736,
      "loss": 4.8969,
      "mean_token_accuracy": 0.20520756095647813,
      "num_tokens": 104858037.0,
      "step": 47860
    },
    {
      "entropy": 5.646356439590454,
      "epoch": 4.750396982929734,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002963048816195965,
      "loss": 4.9635,
      "mean_token_accuracy": 0.18882405012845993,
      "num_tokens": 104868248.0,
      "step": 47865
    },
    {
      "entropy": 5.6286656856536865,
      "epoch": 4.750893211591902,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029626960905450425,
      "loss": 4.9033,
      "mean_token_accuracy": 0.20151637941598893,
      "num_tokens": 104879417.0,
      "step": 47870
    },
    {
      "entropy": 5.632173538208008,
      "epoch": 4.751389440254069,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029623433596197144,
      "loss": 4.8497,
      "mean_token_accuracy": 0.19992298483848572,
      "num_tokens": 104891010.0,
      "step": 47875
    },
    {
      "entropy": 5.690960645675659,
      "epoch": 4.751885668916237,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029619906234287297,
      "loss": 4.9415,
      "mean_token_accuracy": 0.19226425737142563,
      "num_tokens": 104901146.0,
      "step": 47880
    },
    {
      "entropy": 5.69531478881836,
      "epoch": 4.752381897578404,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029616378819808336,
      "loss": 4.9036,
      "mean_token_accuracy": 0.19482176154851913,
      "num_tokens": 104911147.0,
      "step": 47885
    },
    {
      "entropy": 5.689128637313843,
      "epoch": 4.752878126240572,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002961285135284774,
      "loss": 5.0129,
      "mean_token_accuracy": 0.1884959816932678,
      "num_tokens": 104922212.0,
      "step": 47890
    },
    {
      "entropy": 5.6661450386047365,
      "epoch": 4.753374354902739,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029609323833492985,
      "loss": 4.967,
      "mean_token_accuracy": 0.18870058357715608,
      "num_tokens": 104932968.0,
      "step": 47895
    },
    {
      "entropy": 5.6869409561157225,
      "epoch": 4.753870583564907,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002960579626183155,
      "loss": 4.9189,
      "mean_token_accuracy": 0.19701939970254898,
      "num_tokens": 104943231.0,
      "step": 47900
    },
    {
      "entropy": 5.698319005966186,
      "epoch": 4.754366812227074,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002960226863795089,
      "loss": 4.9577,
      "mean_token_accuracy": 0.1942179799079895,
      "num_tokens": 104954798.0,
      "step": 47905
    },
    {
      "entropy": 5.664641714096069,
      "epoch": 4.754863040889242,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002959874096193851,
      "loss": 4.9385,
      "mean_token_accuracy": 0.18858592808246613,
      "num_tokens": 104965548.0,
      "step": 47910
    },
    {
      "entropy": 5.7413414478302,
      "epoch": 4.755359269551409,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029595213233881877,
      "loss": 5.087,
      "mean_token_accuracy": 0.188378968834877,
      "num_tokens": 104977373.0,
      "step": 47915
    },
    {
      "entropy": 5.677464723587036,
      "epoch": 4.755855498213577,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002959168545386847,
      "loss": 4.9496,
      "mean_token_accuracy": 0.1924249365925789,
      "num_tokens": 104987632.0,
      "step": 47920
    },
    {
      "entropy": 5.664759635925293,
      "epoch": 4.756351726875744,
      "grad_norm": 1.25,
      "learning_rate": 0.0002958815762198577,
      "loss": 4.9354,
      "mean_token_accuracy": 0.1919388011097908,
      "num_tokens": 104998669.0,
      "step": 47925
    },
    {
      "entropy": 5.672183704376221,
      "epoch": 4.7568479555379115,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002958462973832126,
      "loss": 4.9158,
      "mean_token_accuracy": 0.1934397742152214,
      "num_tokens": 105009008.0,
      "step": 47930
    },
    {
      "entropy": 5.684236860275268,
      "epoch": 4.757344184200079,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029581101802962417,
      "loss": 4.9066,
      "mean_token_accuracy": 0.19785265773534774,
      "num_tokens": 105020559.0,
      "step": 47935
    },
    {
      "entropy": 5.738646984100342,
      "epoch": 4.757840412862247,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029577573815996747,
      "loss": 5.0226,
      "mean_token_accuracy": 0.18947485834360123,
      "num_tokens": 105032455.0,
      "step": 47940
    },
    {
      "entropy": 5.686953735351563,
      "epoch": 4.758336641524415,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002957404577751171,
      "loss": 4.9576,
      "mean_token_accuracy": 0.19717940986156463,
      "num_tokens": 105043797.0,
      "step": 47945
    },
    {
      "entropy": 5.787062406539917,
      "epoch": 4.758832870186582,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002957051768759482,
      "loss": 5.0543,
      "mean_token_accuracy": 0.18765613585710525,
      "num_tokens": 105054808.0,
      "step": 47950
    },
    {
      "entropy": 5.731529092788696,
      "epoch": 4.75932909884875,
      "grad_norm": 1.25,
      "learning_rate": 0.0002956698954633354,
      "loss": 5.0,
      "mean_token_accuracy": 0.19634663164615632,
      "num_tokens": 105065019.0,
      "step": 47955
    },
    {
      "entropy": 5.7670948028564455,
      "epoch": 4.759825327510917,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029563461353815384,
      "loss": 5.0238,
      "mean_token_accuracy": 0.18852096647024155,
      "num_tokens": 105076581.0,
      "step": 47960
    },
    {
      "entropy": 5.641065073013306,
      "epoch": 4.760321556173085,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029559933110127824,
      "loss": 4.9195,
      "mean_token_accuracy": 0.19695792347192764,
      "num_tokens": 105088296.0,
      "step": 47965
    },
    {
      "entropy": 5.569959497451782,
      "epoch": 4.760817784835252,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002955640481535836,
      "loss": 4.8794,
      "mean_token_accuracy": 0.20021962076425553,
      "num_tokens": 105100216.0,
      "step": 47970
    },
    {
      "entropy": 5.672644233703613,
      "epoch": 4.76131401349742,
      "grad_norm": 1.21875,
      "learning_rate": 0.000295528764695945,
      "loss": 4.917,
      "mean_token_accuracy": 0.19268352538347244,
      "num_tokens": 105110525.0,
      "step": 47975
    },
    {
      "entropy": 5.673283624649048,
      "epoch": 4.761810242159587,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002954934807292372,
      "loss": 4.9265,
      "mean_token_accuracy": 0.20045707523822784,
      "num_tokens": 105122132.0,
      "step": 47980
    },
    {
      "entropy": 5.7004352569580075,
      "epoch": 4.762306470821755,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029545819625433523,
      "loss": 4.9036,
      "mean_token_accuracy": 0.19902239888906478,
      "num_tokens": 105132647.0,
      "step": 47985
    },
    {
      "entropy": 5.679003190994263,
      "epoch": 4.762802699483922,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002954229112721141,
      "loss": 4.9111,
      "mean_token_accuracy": 0.19562074840068816,
      "num_tokens": 105143820.0,
      "step": 47990
    },
    {
      "entropy": 5.676233386993408,
      "epoch": 4.76329892814609,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029538762578344875,
      "loss": 4.8752,
      "mean_token_accuracy": 0.200111685693264,
      "num_tokens": 105154465.0,
      "step": 47995
    },
    {
      "entropy": 5.7222737789154055,
      "epoch": 4.763795156808257,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029535233978921426,
      "loss": 5.0002,
      "mean_token_accuracy": 0.18968479335308075,
      "num_tokens": 105165692.0,
      "step": 48000
    },
    {
      "epoch": 4.763795156808257,
      "eval_entropy": 5.44691301881913,
      "eval_loss": 5.090295314788818,
      "eval_mean_token_accuracy": 0.19339806397097442,
      "eval_num_tokens": 105165692.0,
      "eval_runtime": 20.4806,
      "eval_samples_per_second": 1643.943,
      "eval_steps_per_second": 205.511,
      "step": 48000
    },
    {
      "entropy": 5.726266193389892,
      "epoch": 4.7642913854704245,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002953170532902855,
      "loss": 4.9813,
      "mean_token_accuracy": 0.19576964527368546,
      "num_tokens": 105176509.0,
      "step": 48005
    },
    {
      "entropy": 5.658811426162719,
      "epoch": 4.764787614132592,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002952817662875377,
      "loss": 4.8608,
      "mean_token_accuracy": 0.20162795186042787,
      "num_tokens": 105187285.0,
      "step": 48010
    },
    {
      "entropy": 5.697658061981201,
      "epoch": 4.7652838427947595,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029524647878184586,
      "loss": 4.962,
      "mean_token_accuracy": 0.20115793496370316,
      "num_tokens": 105198766.0,
      "step": 48015
    },
    {
      "entropy": 5.702456283569336,
      "epoch": 4.765780071456927,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029521119077408477,
      "loss": 4.9814,
      "mean_token_accuracy": 0.19025586545467377,
      "num_tokens": 105210046.0,
      "step": 48020
    },
    {
      "entropy": 5.721616649627686,
      "epoch": 4.766276300119095,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002951759022651298,
      "loss": 5.0395,
      "mean_token_accuracy": 0.1893964871764183,
      "num_tokens": 105219940.0,
      "step": 48025
    },
    {
      "entropy": 5.726578235626221,
      "epoch": 4.766772528781263,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000295140613255856,
      "loss": 5.0069,
      "mean_token_accuracy": 0.19200116246938706,
      "num_tokens": 105232001.0,
      "step": 48030
    },
    {
      "entropy": 5.736940574645996,
      "epoch": 4.76726875744343,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002951053237471383,
      "loss": 4.9915,
      "mean_token_accuracy": 0.19154592603445053,
      "num_tokens": 105244905.0,
      "step": 48035
    },
    {
      "entropy": 5.763356924057007,
      "epoch": 4.767764986105598,
      "grad_norm": 1.25,
      "learning_rate": 0.00029507003373985193,
      "loss": 4.998,
      "mean_token_accuracy": 0.19364480525255204,
      "num_tokens": 105255989.0,
      "step": 48040
    },
    {
      "entropy": 5.658811140060425,
      "epoch": 4.768261214767765,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029503474323487193,
      "loss": 4.9046,
      "mean_token_accuracy": 0.19776091277599334,
      "num_tokens": 105265475.0,
      "step": 48045
    },
    {
      "entropy": 5.6394599914550785,
      "epoch": 4.768757443429933,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002949994522330735,
      "loss": 4.9757,
      "mean_token_accuracy": 0.18650995045900345,
      "num_tokens": 105274540.0,
      "step": 48050
    },
    {
      "entropy": 5.701470136642456,
      "epoch": 4.7692536720921,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029496416073533177,
      "loss": 4.9906,
      "mean_token_accuracy": 0.1936501145362854,
      "num_tokens": 105284757.0,
      "step": 48055
    },
    {
      "entropy": 5.812077903747559,
      "epoch": 4.769749900754268,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002949288687425219,
      "loss": 5.0414,
      "mean_token_accuracy": 0.18985655009746552,
      "num_tokens": 105296880.0,
      "step": 48060
    },
    {
      "entropy": 5.816578912734985,
      "epoch": 4.770246129416435,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029489357625551895,
      "loss": 5.0253,
      "mean_token_accuracy": 0.19856968075037001,
      "num_tokens": 105307481.0,
      "step": 48065
    },
    {
      "entropy": 5.735312557220459,
      "epoch": 4.770742358078603,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029485828327519817,
      "loss": 4.9646,
      "mean_token_accuracy": 0.20312740206718444,
      "num_tokens": 105319247.0,
      "step": 48070
    },
    {
      "entropy": 5.715062856674194,
      "epoch": 4.77123858674077,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002948229898024348,
      "loss": 5.0255,
      "mean_token_accuracy": 0.18804741948843,
      "num_tokens": 105331297.0,
      "step": 48075
    },
    {
      "entropy": 5.718420362472534,
      "epoch": 4.7717348154029375,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000294787695838104,
      "loss": 4.9834,
      "mean_token_accuracy": 0.19952573627233505,
      "num_tokens": 105341637.0,
      "step": 48080
    },
    {
      "entropy": 5.755389785766601,
      "epoch": 4.772231044065105,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000294752401383081,
      "loss": 4.9648,
      "mean_token_accuracy": 0.1981787070631981,
      "num_tokens": 105352516.0,
      "step": 48085
    },
    {
      "entropy": 5.708091449737549,
      "epoch": 4.7727272727272725,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029471710643824093,
      "loss": 4.9965,
      "mean_token_accuracy": 0.19711913913488388,
      "num_tokens": 105363114.0,
      "step": 48090
    },
    {
      "entropy": 5.724729728698731,
      "epoch": 4.77322350138944,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029468181100445915,
      "loss": 4.9194,
      "mean_token_accuracy": 0.19762780219316484,
      "num_tokens": 105374521.0,
      "step": 48095
    },
    {
      "entropy": 5.703747701644898,
      "epoch": 4.773719730051607,
      "grad_norm": 1.25,
      "learning_rate": 0.00029464651508261087,
      "loss": 4.9707,
      "mean_token_accuracy": 0.18799059242010116,
      "num_tokens": 105385070.0,
      "step": 48100
    },
    {
      "entropy": 5.713271808624268,
      "epoch": 4.774215958713775,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002946112186735714,
      "loss": 4.9724,
      "mean_token_accuracy": 0.1888252690434456,
      "num_tokens": 105395817.0,
      "step": 48105
    },
    {
      "entropy": 5.647746133804321,
      "epoch": 4.774712187375943,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002945759217782159,
      "loss": 4.9093,
      "mean_token_accuracy": 0.19441512674093248,
      "num_tokens": 105406469.0,
      "step": 48110
    },
    {
      "entropy": 5.685758972167969,
      "epoch": 4.775208416038111,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002945406243974198,
      "loss": 5.0294,
      "mean_token_accuracy": 0.1924970954656601,
      "num_tokens": 105417590.0,
      "step": 48115
    },
    {
      "entropy": 5.662922334671021,
      "epoch": 4.775704644700278,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029450532653205826,
      "loss": 4.9586,
      "mean_token_accuracy": 0.19337525367736816,
      "num_tokens": 105428647.0,
      "step": 48120
    },
    {
      "entropy": 5.727460765838623,
      "epoch": 4.776200873362446,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002944700281830066,
      "loss": 5.0154,
      "mean_token_accuracy": 0.18681089878082274,
      "num_tokens": 105439922.0,
      "step": 48125
    },
    {
      "entropy": 5.722469139099121,
      "epoch": 4.776697102024613,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002944347293511403,
      "loss": 5.0048,
      "mean_token_accuracy": 0.1882747769355774,
      "num_tokens": 105449705.0,
      "step": 48130
    },
    {
      "entropy": 5.75536584854126,
      "epoch": 4.777193330686781,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029439943003733456,
      "loss": 4.973,
      "mean_token_accuracy": 0.19975400120019912,
      "num_tokens": 105462028.0,
      "step": 48135
    },
    {
      "entropy": 5.688358545303345,
      "epoch": 4.777689559348948,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002943641302424647,
      "loss": 4.9387,
      "mean_token_accuracy": 0.1950881689786911,
      "num_tokens": 105474429.0,
      "step": 48140
    },
    {
      "entropy": 5.702763891220092,
      "epoch": 4.778185788011116,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029432882996740617,
      "loss": 4.9494,
      "mean_token_accuracy": 0.1935947746038437,
      "num_tokens": 105485933.0,
      "step": 48145
    },
    {
      "entropy": 5.713082408905029,
      "epoch": 4.778682016673283,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029429352921303434,
      "loss": 4.9948,
      "mean_token_accuracy": 0.18588154315948485,
      "num_tokens": 105496741.0,
      "step": 48150
    },
    {
      "entropy": 5.729077339172363,
      "epoch": 4.7791782453354505,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002942582279802245,
      "loss": 5.0228,
      "mean_token_accuracy": 0.18974916487932206,
      "num_tokens": 105508393.0,
      "step": 48155
    },
    {
      "entropy": 5.7486891746521,
      "epoch": 4.779674473997618,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002942229262698521,
      "loss": 4.9943,
      "mean_token_accuracy": 0.202560955286026,
      "num_tokens": 105518127.0,
      "step": 48160
    },
    {
      "entropy": 5.647482013702392,
      "epoch": 4.7801707026597855,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002941876240827926,
      "loss": 4.8945,
      "mean_token_accuracy": 0.20188500434160234,
      "num_tokens": 105527847.0,
      "step": 48165
    },
    {
      "entropy": 5.730143785476685,
      "epoch": 4.780666931321953,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002941523214199214,
      "loss": 4.9748,
      "mean_token_accuracy": 0.19398251175880432,
      "num_tokens": 105538943.0,
      "step": 48170
    },
    {
      "entropy": 5.734469175338745,
      "epoch": 4.7811631599841204,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002941170182821138,
      "loss": 4.9779,
      "mean_token_accuracy": 0.1948048561811447,
      "num_tokens": 105549327.0,
      "step": 48175
    },
    {
      "entropy": 5.674100637435913,
      "epoch": 4.781659388646288,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002940817146702454,
      "loss": 4.9393,
      "mean_token_accuracy": 0.19275760501623154,
      "num_tokens": 105559854.0,
      "step": 48180
    },
    {
      "entropy": 5.721744728088379,
      "epoch": 4.782155617308455,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002940464105851916,
      "loss": 4.9231,
      "mean_token_accuracy": 0.19753335863351823,
      "num_tokens": 105569139.0,
      "step": 48185
    },
    {
      "entropy": 5.689853000640869,
      "epoch": 4.782651845970623,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002940111060278279,
      "loss": 4.9191,
      "mean_token_accuracy": 0.19201015681028366,
      "num_tokens": 105579543.0,
      "step": 48190
    },
    {
      "entropy": 5.681257343292236,
      "epoch": 4.783148074632791,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029397580099902967,
      "loss": 5.0226,
      "mean_token_accuracy": 0.18807919919490815,
      "num_tokens": 105590748.0,
      "step": 48195
    },
    {
      "entropy": 5.675642967224121,
      "epoch": 4.783644303294958,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029394049549967246,
      "loss": 4.9274,
      "mean_token_accuracy": 0.19453067183494568,
      "num_tokens": 105603158.0,
      "step": 48200
    },
    {
      "entropy": 5.6915336608886715,
      "epoch": 4.784140531957126,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029390518953063176,
      "loss": 4.9313,
      "mean_token_accuracy": 0.1957581326365471,
      "num_tokens": 105613390.0,
      "step": 48205
    },
    {
      "entropy": 5.719400405883789,
      "epoch": 4.784636760619294,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002938698830927832,
      "loss": 4.9943,
      "mean_token_accuracy": 0.19322713911533357,
      "num_tokens": 105623671.0,
      "step": 48210
    },
    {
      "entropy": 5.710768795013427,
      "epoch": 4.785132989281461,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029383457618700206,
      "loss": 4.9435,
      "mean_token_accuracy": 0.18919123113155364,
      "num_tokens": 105634177.0,
      "step": 48215
    },
    {
      "entropy": 5.7051722526550295,
      "epoch": 4.785629217943629,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002937992688141641,
      "loss": 4.9412,
      "mean_token_accuracy": 0.1998508855700493,
      "num_tokens": 105644983.0,
      "step": 48220
    },
    {
      "entropy": 5.643399810791015,
      "epoch": 4.786125446605796,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029376396097514473,
      "loss": 4.9242,
      "mean_token_accuracy": 0.19184161573648453,
      "num_tokens": 105654662.0,
      "step": 48225
    },
    {
      "entropy": 5.714979362487793,
      "epoch": 4.786621675267964,
      "grad_norm": 1.046875,
      "learning_rate": 0.00029372865267081967,
      "loss": 4.9787,
      "mean_token_accuracy": 0.1857148990035057,
      "num_tokens": 105665888.0,
      "step": 48230
    },
    {
      "entropy": 5.737197160720825,
      "epoch": 4.787117903930131,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029369334390206425,
      "loss": 4.978,
      "mean_token_accuracy": 0.19306859076023103,
      "num_tokens": 105674738.0,
      "step": 48235
    },
    {
      "entropy": 5.665415239334107,
      "epoch": 4.7876141325922985,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029365803466975424,
      "loss": 4.9929,
      "mean_token_accuracy": 0.1938965454697609,
      "num_tokens": 105686809.0,
      "step": 48240
    },
    {
      "entropy": 5.703986358642578,
      "epoch": 4.788110361254466,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029362272497476517,
      "loss": 4.9536,
      "mean_token_accuracy": 0.19452123641967772,
      "num_tokens": 105697414.0,
      "step": 48245
    },
    {
      "entropy": 5.657271528244019,
      "epoch": 4.7886065899166335,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002935874148179726,
      "loss": 4.9206,
      "mean_token_accuracy": 0.19661246687173844,
      "num_tokens": 105707858.0,
      "step": 48250
    },
    {
      "entropy": 5.773865032196045,
      "epoch": 4.789102818578801,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002935521042002523,
      "loss": 5.085,
      "mean_token_accuracy": 0.1928754448890686,
      "num_tokens": 105718463.0,
      "step": 48255
    },
    {
      "entropy": 5.742237615585327,
      "epoch": 4.789599047240968,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029351679312247976,
      "loss": 5.0597,
      "mean_token_accuracy": 0.19009046256542206,
      "num_tokens": 105730139.0,
      "step": 48260
    },
    {
      "entropy": 5.667094802856445,
      "epoch": 4.790095275903136,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002934814815855306,
      "loss": 4.9221,
      "mean_token_accuracy": 0.19887358397245408,
      "num_tokens": 105740429.0,
      "step": 48265
    },
    {
      "entropy": 5.736084794998169,
      "epoch": 4.790591504565303,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002934461695902805,
      "loss": 5.0021,
      "mean_token_accuracy": 0.19045465290546418,
      "num_tokens": 105752081.0,
      "step": 48270
    },
    {
      "entropy": 5.737586927413941,
      "epoch": 4.791087733227471,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002934108571376052,
      "loss": 5.0166,
      "mean_token_accuracy": 0.19026533216238023,
      "num_tokens": 105763710.0,
      "step": 48275
    },
    {
      "entropy": 5.708095121383667,
      "epoch": 4.791583961889639,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029337554422838037,
      "loss": 4.9763,
      "mean_token_accuracy": 0.19574916064739228,
      "num_tokens": 105775296.0,
      "step": 48280
    },
    {
      "entropy": 5.706837749481201,
      "epoch": 4.792080190551806,
      "grad_norm": 1.5,
      "learning_rate": 0.00029334023086348155,
      "loss": 4.9642,
      "mean_token_accuracy": 0.20495292842388152,
      "num_tokens": 105785531.0,
      "step": 48285
    },
    {
      "entropy": 5.685818290710449,
      "epoch": 4.792576419213974,
      "grad_norm": 1.25,
      "learning_rate": 0.0002933049170437846,
      "loss": 4.9406,
      "mean_token_accuracy": 0.19726989269256592,
      "num_tokens": 105795938.0,
      "step": 48290
    },
    {
      "entropy": 5.643454074859619,
      "epoch": 4.793072647876142,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002932696027701651,
      "loss": 4.9299,
      "mean_token_accuracy": 0.18774772733449935,
      "num_tokens": 105805453.0,
      "step": 48295
    },
    {
      "entropy": 5.733874273300171,
      "epoch": 4.793568876538309,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002932342880434988,
      "loss": 4.9396,
      "mean_token_accuracy": 0.1973542958498001,
      "num_tokens": 105815192.0,
      "step": 48300
    },
    {
      "entropy": 5.663654899597168,
      "epoch": 4.794065105200477,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029319897286466144,
      "loss": 4.9825,
      "mean_token_accuracy": 0.18692944645881654,
      "num_tokens": 105826975.0,
      "step": 48305
    },
    {
      "entropy": 5.72221531867981,
      "epoch": 4.794561333862644,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002931636572345289,
      "loss": 4.9847,
      "mean_token_accuracy": 0.1956683501601219,
      "num_tokens": 105838928.0,
      "step": 48310
    },
    {
      "entropy": 5.747190523147583,
      "epoch": 4.7950575625248115,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029312834115397657,
      "loss": 5.0795,
      "mean_token_accuracy": 0.18447850346565248,
      "num_tokens": 105852010.0,
      "step": 48315
    },
    {
      "entropy": 5.707646751403809,
      "epoch": 4.795553791186979,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002930930246238806,
      "loss": 4.9311,
      "mean_token_accuracy": 0.19377971142530442,
      "num_tokens": 105862764.0,
      "step": 48320
    },
    {
      "entropy": 5.70206127166748,
      "epoch": 4.7960500198491465,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029305770764511665,
      "loss": 4.9672,
      "mean_token_accuracy": 0.19894929826259614,
      "num_tokens": 105874169.0,
      "step": 48325
    },
    {
      "entropy": 5.73315463066101,
      "epoch": 4.796546248511314,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002930223902185603,
      "loss": 5.0045,
      "mean_token_accuracy": 0.19483126401901246,
      "num_tokens": 105885728.0,
      "step": 48330
    },
    {
      "entropy": 5.660814332962036,
      "epoch": 4.797042477173481,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002929870723450876,
      "loss": 4.8452,
      "mean_token_accuracy": 0.21549836695194244,
      "num_tokens": 105895760.0,
      "step": 48335
    },
    {
      "entropy": 5.696483421325683,
      "epoch": 4.797538705835649,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002929517540255743,
      "loss": 4.939,
      "mean_token_accuracy": 0.19799540787935258,
      "num_tokens": 105907620.0,
      "step": 48340
    },
    {
      "entropy": 5.678500699996948,
      "epoch": 4.798034934497816,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002929164352608961,
      "loss": 5.0205,
      "mean_token_accuracy": 0.18315057903528215,
      "num_tokens": 105916973.0,
      "step": 48345
    },
    {
      "entropy": 5.649071741104126,
      "epoch": 4.798531163159984,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029288111605192896,
      "loss": 4.8849,
      "mean_token_accuracy": 0.19789620637893676,
      "num_tokens": 105926200.0,
      "step": 48350
    },
    {
      "entropy": 5.667071056365967,
      "epoch": 4.799027391822151,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029284579639954865,
      "loss": 4.8857,
      "mean_token_accuracy": 0.2035166084766388,
      "num_tokens": 105938049.0,
      "step": 48355
    },
    {
      "entropy": 5.678744554519653,
      "epoch": 4.799523620484319,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002928104763046311,
      "loss": 4.9292,
      "mean_token_accuracy": 0.1948329195380211,
      "num_tokens": 105950194.0,
      "step": 48360
    },
    {
      "entropy": 5.64761905670166,
      "epoch": 4.800019849146487,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002927751557680521,
      "loss": 4.8971,
      "mean_token_accuracy": 0.19934981912374497,
      "num_tokens": 105961490.0,
      "step": 48365
    },
    {
      "entropy": 5.760592126846314,
      "epoch": 4.800516077808654,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002927398347906875,
      "loss": 4.9898,
      "mean_token_accuracy": 0.19434868395328522,
      "num_tokens": 105972652.0,
      "step": 48370
    },
    {
      "entropy": 5.712092781066895,
      "epoch": 4.801012306470822,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002927045133734132,
      "loss": 4.9782,
      "mean_token_accuracy": 0.19528839439153672,
      "num_tokens": 105983495.0,
      "step": 48375
    },
    {
      "entropy": 5.68025312423706,
      "epoch": 4.80150853513299,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029266919151710514,
      "loss": 4.9369,
      "mean_token_accuracy": 0.2017463982105255,
      "num_tokens": 105994224.0,
      "step": 48380
    },
    {
      "entropy": 5.769813203811646,
      "epoch": 4.802004763795157,
      "grad_norm": 1.25,
      "learning_rate": 0.0002926338692226392,
      "loss": 4.9702,
      "mean_token_accuracy": 0.1907532826066017,
      "num_tokens": 106005673.0,
      "step": 48385
    },
    {
      "entropy": 5.703266668319702,
      "epoch": 4.8025009924573245,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029259854649089136,
      "loss": 4.9966,
      "mean_token_accuracy": 0.19300596714019774,
      "num_tokens": 106015771.0,
      "step": 48390
    },
    {
      "entropy": 5.6864382266998295,
      "epoch": 4.802997221119492,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029256322332273744,
      "loss": 4.9381,
      "mean_token_accuracy": 0.19894327074289322,
      "num_tokens": 106028073.0,
      "step": 48395
    },
    {
      "entropy": 5.57644829750061,
      "epoch": 4.8034934497816595,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029252789971905344,
      "loss": 4.8331,
      "mean_token_accuracy": 0.19691080898046492,
      "num_tokens": 106038451.0,
      "step": 48400
    },
    {
      "entropy": 5.737292385101318,
      "epoch": 4.803989678443827,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002924925756807153,
      "loss": 4.9849,
      "mean_token_accuracy": 0.195039564371109,
      "num_tokens": 106049109.0,
      "step": 48405
    },
    {
      "entropy": 5.697596502304077,
      "epoch": 4.804485907105994,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029245725120859893,
      "loss": 4.887,
      "mean_token_accuracy": 0.20613234341144562,
      "num_tokens": 106059114.0,
      "step": 48410
    },
    {
      "entropy": 5.69321837425232,
      "epoch": 4.804982135768162,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029242192630358045,
      "loss": 4.9891,
      "mean_token_accuracy": 0.19142155349254608,
      "num_tokens": 106070749.0,
      "step": 48415
    },
    {
      "entropy": 5.725415372848511,
      "epoch": 4.805478364430329,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029238660096653565,
      "loss": 4.9812,
      "mean_token_accuracy": 0.19959878772497178,
      "num_tokens": 106081851.0,
      "step": 48420
    },
    {
      "entropy": 5.664791202545166,
      "epoch": 4.805974593092497,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029235127519834067,
      "loss": 4.9374,
      "mean_token_accuracy": 0.19412993490695954,
      "num_tokens": 106093053.0,
      "step": 48425
    },
    {
      "entropy": 5.687638235092163,
      "epoch": 4.806470821754664,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029231594899987133,
      "loss": 4.9226,
      "mean_token_accuracy": 0.1915833130478859,
      "num_tokens": 106104405.0,
      "step": 48430
    },
    {
      "entropy": 5.690450620651245,
      "epoch": 4.806967050416832,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002922806223720039,
      "loss": 5.0182,
      "mean_token_accuracy": 0.19002685695886612,
      "num_tokens": 106115974.0,
      "step": 48435
    },
    {
      "entropy": 5.648723220825195,
      "epoch": 4.807463279078999,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002922452953156142,
      "loss": 4.8888,
      "mean_token_accuracy": 0.1918892666697502,
      "num_tokens": 106126618.0,
      "step": 48440
    },
    {
      "entropy": 5.677489185333252,
      "epoch": 4.807959507741167,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029220996783157837,
      "loss": 4.9217,
      "mean_token_accuracy": 0.20339683890342714,
      "num_tokens": 106137952.0,
      "step": 48445
    },
    {
      "entropy": 5.6667684555053714,
      "epoch": 4.808455736403335,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029217463992077243,
      "loss": 4.9609,
      "mean_token_accuracy": 0.1947764739394188,
      "num_tokens": 106148485.0,
      "step": 48450
    },
    {
      "entropy": 5.692499351501465,
      "epoch": 4.808951965065502,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002921393115840724,
      "loss": 4.9365,
      "mean_token_accuracy": 0.1932900905609131,
      "num_tokens": 106159877.0,
      "step": 48455
    },
    {
      "entropy": 5.678902816772461,
      "epoch": 4.80944819372767,
      "grad_norm": 1.25,
      "learning_rate": 0.0002921039828223544,
      "loss": 4.9225,
      "mean_token_accuracy": 0.1961925134062767,
      "num_tokens": 106170112.0,
      "step": 48460
    },
    {
      "entropy": 5.715104818344116,
      "epoch": 4.8099444223898375,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002920686536364944,
      "loss": 5.0107,
      "mean_token_accuracy": 0.18555509597063063,
      "num_tokens": 106181364.0,
      "step": 48465
    },
    {
      "entropy": 5.696612453460693,
      "epoch": 4.810440651052005,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002920333240273687,
      "loss": 4.949,
      "mean_token_accuracy": 0.19292895644903182,
      "num_tokens": 106191604.0,
      "step": 48470
    },
    {
      "entropy": 5.717963361740113,
      "epoch": 4.8109368797141725,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029199799399585324,
      "loss": 4.9829,
      "mean_token_accuracy": 0.1909063845872879,
      "num_tokens": 106204411.0,
      "step": 48475
    },
    {
      "entropy": 5.728474235534668,
      "epoch": 4.81143310837634,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002919626635428241,
      "loss": 5.0151,
      "mean_token_accuracy": 0.1854264631867409,
      "num_tokens": 106216045.0,
      "step": 48480
    },
    {
      "entropy": 5.645805358886719,
      "epoch": 4.8119293370385074,
      "grad_norm": 1.375,
      "learning_rate": 0.0002919273326691575,
      "loss": 4.8523,
      "mean_token_accuracy": 0.19671124815940857,
      "num_tokens": 106225381.0,
      "step": 48485
    },
    {
      "entropy": 5.717382526397705,
      "epoch": 4.812425565700675,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029189200137572953,
      "loss": 5.0012,
      "mean_token_accuracy": 0.19263758659362792,
      "num_tokens": 106236520.0,
      "step": 48490
    },
    {
      "entropy": 5.7271143913269045,
      "epoch": 4.812921794362842,
      "grad_norm": 1.25,
      "learning_rate": 0.00029185666966341625,
      "loss": 5.0068,
      "mean_token_accuracy": 0.1884971484541893,
      "num_tokens": 106248065.0,
      "step": 48495
    },
    {
      "entropy": 5.7075684547424315,
      "epoch": 4.81341802302501,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002918213375330939,
      "loss": 4.9097,
      "mean_token_accuracy": 0.19629336893558502,
      "num_tokens": 106258563.0,
      "step": 48500
    },
    {
      "entropy": 5.680696630477906,
      "epoch": 4.813914251687177,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002917860049856387,
      "loss": 4.9549,
      "mean_token_accuracy": 0.18868448287248613,
      "num_tokens": 106270682.0,
      "step": 48505
    },
    {
      "entropy": 5.667771100997925,
      "epoch": 4.814410480349345,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002917506720219267,
      "loss": 4.9146,
      "mean_token_accuracy": 0.19713609665632248,
      "num_tokens": 106281267.0,
      "step": 48510
    },
    {
      "entropy": 5.759541130065918,
      "epoch": 4.814906709011512,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002917153386428341,
      "loss": 4.9984,
      "mean_token_accuracy": 0.1917022630572319,
      "num_tokens": 106292381.0,
      "step": 48515
    },
    {
      "entropy": 5.739806079864502,
      "epoch": 4.81540293767368,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029168000484923717,
      "loss": 4.9873,
      "mean_token_accuracy": 0.19181099832057952,
      "num_tokens": 106302761.0,
      "step": 48520
    },
    {
      "entropy": 5.644868040084839,
      "epoch": 4.815899166335847,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029164467064201206,
      "loss": 4.8624,
      "mean_token_accuracy": 0.20488255023956298,
      "num_tokens": 106314911.0,
      "step": 48525
    },
    {
      "entropy": 5.715041828155518,
      "epoch": 4.816395394998015,
      "grad_norm": 1.03125,
      "learning_rate": 0.000291609336022035,
      "loss": 5.0458,
      "mean_token_accuracy": 0.1881086677312851,
      "num_tokens": 106326490.0,
      "step": 48530
    },
    {
      "entropy": 5.6636495113372805,
      "epoch": 4.816891623660183,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002915740009901821,
      "loss": 4.8471,
      "mean_token_accuracy": 0.2036970540881157,
      "num_tokens": 106336967.0,
      "step": 48535
    },
    {
      "entropy": 5.7757617950439455,
      "epoch": 4.81738785232235,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029153866554732985,
      "loss": 5.0212,
      "mean_token_accuracy": 0.1861901953816414,
      "num_tokens": 106348413.0,
      "step": 48540
    },
    {
      "entropy": 5.734197759628296,
      "epoch": 4.817884080984518,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029150332969435415,
      "loss": 5.0421,
      "mean_token_accuracy": 0.18277324438095094,
      "num_tokens": 106359486.0,
      "step": 48545
    },
    {
      "entropy": 5.714854764938354,
      "epoch": 4.8183803096466855,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029146799343213154,
      "loss": 5.0294,
      "mean_token_accuracy": 0.19178947806358337,
      "num_tokens": 106369889.0,
      "step": 48550
    },
    {
      "entropy": 5.706478118896484,
      "epoch": 4.818876538308853,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029143265676153816,
      "loss": 4.9591,
      "mean_token_accuracy": 0.19169008135795593,
      "num_tokens": 106381603.0,
      "step": 48555
    },
    {
      "entropy": 5.746821069717408,
      "epoch": 4.8193727669710205,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029139731968345033,
      "loss": 5.0363,
      "mean_token_accuracy": 0.1854133799672127,
      "num_tokens": 106394547.0,
      "step": 48560
    },
    {
      "entropy": 5.669528865814209,
      "epoch": 4.819868995633188,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002913619821987443,
      "loss": 4.8841,
      "mean_token_accuracy": 0.20841964781284333,
      "num_tokens": 106406248.0,
      "step": 48565
    },
    {
      "entropy": 5.607198047637939,
      "epoch": 4.820365224295355,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029132664430829634,
      "loss": 4.8332,
      "mean_token_accuracy": 0.210535492002964,
      "num_tokens": 106416234.0,
      "step": 48570
    },
    {
      "entropy": 5.719139242172242,
      "epoch": 4.820861452957523,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029129130601298275,
      "loss": 4.9048,
      "mean_token_accuracy": 0.19813740402460098,
      "num_tokens": 106428027.0,
      "step": 48575
    },
    {
      "entropy": 5.751349306106567,
      "epoch": 4.82135768161969,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002912559673136799,
      "loss": 5.0042,
      "mean_token_accuracy": 0.18910041898489,
      "num_tokens": 106438741.0,
      "step": 48580
    },
    {
      "entropy": 5.701038837432861,
      "epoch": 4.821853910281858,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029122062821126417,
      "loss": 4.9577,
      "mean_token_accuracy": 0.18675011098384858,
      "num_tokens": 106448951.0,
      "step": 48585
    },
    {
      "entropy": 5.636749649047852,
      "epoch": 4.822350138944025,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002911852887066117,
      "loss": 4.9772,
      "mean_token_accuracy": 0.1867016851902008,
      "num_tokens": 106460321.0,
      "step": 48590
    },
    {
      "entropy": 5.683742809295654,
      "epoch": 4.822846367606193,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029114994880059895,
      "loss": 4.9932,
      "mean_token_accuracy": 0.18493081629276276,
      "num_tokens": 106470642.0,
      "step": 48595
    },
    {
      "entropy": 5.720409250259399,
      "epoch": 4.82334259626836,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029111460849410236,
      "loss": 5.0109,
      "mean_token_accuracy": 0.19748578816652299,
      "num_tokens": 106480151.0,
      "step": 48600
    },
    {
      "entropy": 5.695727777481079,
      "epoch": 4.823838824930528,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002910792677879982,
      "loss": 4.9228,
      "mean_token_accuracy": 0.1923929825425148,
      "num_tokens": 106491005.0,
      "step": 48605
    },
    {
      "entropy": 5.711050128936767,
      "epoch": 4.824335053592695,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029104392668316277,
      "loss": 4.9848,
      "mean_token_accuracy": 0.1901163086295128,
      "num_tokens": 106502053.0,
      "step": 48610
    },
    {
      "entropy": 5.671296405792236,
      "epoch": 4.824831282254863,
      "grad_norm": 1.25,
      "learning_rate": 0.0002910085851804726,
      "loss": 4.9421,
      "mean_token_accuracy": 0.1939192533493042,
      "num_tokens": 106513275.0,
      "step": 48615
    },
    {
      "entropy": 5.729354572296143,
      "epoch": 4.825327510917031,
      "grad_norm": 1.25,
      "learning_rate": 0.000290973243280804,
      "loss": 4.9866,
      "mean_token_accuracy": 0.19609472304582595,
      "num_tokens": 106523664.0,
      "step": 48620
    },
    {
      "entropy": 5.625689554214477,
      "epoch": 4.825823739579198,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029093790098503334,
      "loss": 4.882,
      "mean_token_accuracy": 0.19539916515350342,
      "num_tokens": 106533890.0,
      "step": 48625
    },
    {
      "entropy": 5.614658975601197,
      "epoch": 4.826319968241366,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029090255829403703,
      "loss": 4.8718,
      "mean_token_accuracy": 0.1980840265750885,
      "num_tokens": 106545257.0,
      "step": 48630
    },
    {
      "entropy": 5.680109214782715,
      "epoch": 4.8268161969035335,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002908672152086917,
      "loss": 4.8668,
      "mean_token_accuracy": 0.2096521183848381,
      "num_tokens": 106555724.0,
      "step": 48635
    },
    {
      "entropy": 5.685220050811767,
      "epoch": 4.827312425565701,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029083187172987353,
      "loss": 5.0145,
      "mean_token_accuracy": 0.19164867401123048,
      "num_tokens": 106566773.0,
      "step": 48640
    },
    {
      "entropy": 5.762804937362671,
      "epoch": 4.827808654227868,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000290796527858459,
      "loss": 5.0365,
      "mean_token_accuracy": 0.18600951582193376,
      "num_tokens": 106577031.0,
      "step": 48645
    },
    {
      "entropy": 5.679256725311279,
      "epoch": 4.828304882890036,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002907611835953247,
      "loss": 4.8545,
      "mean_token_accuracy": 0.20773115009069443,
      "num_tokens": 106588550.0,
      "step": 48650
    },
    {
      "entropy": 5.703868818283081,
      "epoch": 4.828801111552203,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029072583894134703,
      "loss": 5.0268,
      "mean_token_accuracy": 0.19887096136808396,
      "num_tokens": 106600106.0,
      "step": 48655
    },
    {
      "entropy": 5.734431505203247,
      "epoch": 4.829297340214371,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002906904938974024,
      "loss": 5.0411,
      "mean_token_accuracy": 0.18464177697896958,
      "num_tokens": 106610309.0,
      "step": 48660
    },
    {
      "entropy": 5.71020188331604,
      "epoch": 4.829793568876538,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002906551484643673,
      "loss": 5.0015,
      "mean_token_accuracy": 0.19255668967962264,
      "num_tokens": 106621437.0,
      "step": 48665
    },
    {
      "entropy": 5.736194181442261,
      "epoch": 4.830289797538706,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029061980264311825,
      "loss": 5.0109,
      "mean_token_accuracy": 0.1904214322566986,
      "num_tokens": 106632633.0,
      "step": 48670
    },
    {
      "entropy": 5.761097097396851,
      "epoch": 4.830786026200873,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002905844564345318,
      "loss": 4.9979,
      "mean_token_accuracy": 0.18958929181098938,
      "num_tokens": 106643982.0,
      "step": 48675
    },
    {
      "entropy": 5.646967887878418,
      "epoch": 4.831282254863041,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029054910983948443,
      "loss": 4.915,
      "mean_token_accuracy": 0.18772610276937485,
      "num_tokens": 106655315.0,
      "step": 48680
    },
    {
      "entropy": 5.655270195007324,
      "epoch": 4.831778483525208,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029051376285885253,
      "loss": 4.948,
      "mean_token_accuracy": 0.19513627737760544,
      "num_tokens": 106667106.0,
      "step": 48685
    },
    {
      "entropy": 5.669793128967285,
      "epoch": 4.832274712187376,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029047841549351285,
      "loss": 4.9915,
      "mean_token_accuracy": 0.19176144897937775,
      "num_tokens": 106678152.0,
      "step": 48690
    },
    {
      "entropy": 5.735513544082641,
      "epoch": 4.832770940849543,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002904430677443418,
      "loss": 5.0021,
      "mean_token_accuracy": 0.19154899418354035,
      "num_tokens": 106689784.0,
      "step": 48695
    },
    {
      "entropy": 5.729791355133057,
      "epoch": 4.833267169511711,
      "grad_norm": 1.015625,
      "learning_rate": 0.00029040771961221597,
      "loss": 4.911,
      "mean_token_accuracy": 0.2022022321820259,
      "num_tokens": 106702023.0,
      "step": 48700
    },
    {
      "entropy": 5.7050065994262695,
      "epoch": 4.833763398173879,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029037237109801186,
      "loss": 4.9468,
      "mean_token_accuracy": 0.1977407604455948,
      "num_tokens": 106712075.0,
      "step": 48705
    },
    {
      "entropy": 5.707368659973144,
      "epoch": 4.834259626836046,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029033702220260607,
      "loss": 4.9628,
      "mean_token_accuracy": 0.19465886652469636,
      "num_tokens": 106722548.0,
      "step": 48710
    },
    {
      "entropy": 5.63431487083435,
      "epoch": 4.834755855498214,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002903016729268751,
      "loss": 4.8657,
      "mean_token_accuracy": 0.19548039883375168,
      "num_tokens": 106732741.0,
      "step": 48715
    },
    {
      "entropy": 5.635940790176392,
      "epoch": 4.835252084160381,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002902663232716958,
      "loss": 4.9704,
      "mean_token_accuracy": 0.1864536762237549,
      "num_tokens": 106743019.0,
      "step": 48720
    },
    {
      "entropy": 5.667228555679321,
      "epoch": 4.835748312822549,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029023097323794433,
      "loss": 4.9445,
      "mean_token_accuracy": 0.1989621177315712,
      "num_tokens": 106753021.0,
      "step": 48725
    },
    {
      "entropy": 5.701305484771728,
      "epoch": 4.836244541484716,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002901956228264977,
      "loss": 4.9209,
      "mean_token_accuracy": 0.20004830807447432,
      "num_tokens": 106762611.0,
      "step": 48730
    },
    {
      "entropy": 5.754466962814331,
      "epoch": 4.836740770146884,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002901602720382324,
      "loss": 5.0616,
      "mean_token_accuracy": 0.18463184684515,
      "num_tokens": 106774310.0,
      "step": 48735
    },
    {
      "entropy": 5.6381995677948,
      "epoch": 4.837236998809051,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029012492087402493,
      "loss": 4.9002,
      "mean_token_accuracy": 0.20340545624494552,
      "num_tokens": 106785275.0,
      "step": 48740
    },
    {
      "entropy": 5.696300888061524,
      "epoch": 4.837733227471219,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029008956933475213,
      "loss": 4.9762,
      "mean_token_accuracy": 0.19618968069553375,
      "num_tokens": 106798153.0,
      "step": 48745
    },
    {
      "entropy": 5.712812185287476,
      "epoch": 4.838229456133386,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002900542174212905,
      "loss": 4.9617,
      "mean_token_accuracy": 0.19934065788984298,
      "num_tokens": 106809170.0,
      "step": 48750
    },
    {
      "entropy": 5.6748841285705565,
      "epoch": 4.838725684795554,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002900188651345166,
      "loss": 4.9275,
      "mean_token_accuracy": 0.19603839218616487,
      "num_tokens": 106820418.0,
      "step": 48755
    },
    {
      "entropy": 5.675888252258301,
      "epoch": 4.839221913457721,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028998351247530726,
      "loss": 4.9174,
      "mean_token_accuracy": 0.1943417951464653,
      "num_tokens": 106831708.0,
      "step": 48760
    },
    {
      "entropy": 5.67789659500122,
      "epoch": 4.839718142119889,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002899481594445391,
      "loss": 4.9874,
      "mean_token_accuracy": 0.1961041584610939,
      "num_tokens": 106842276.0,
      "step": 48765
    },
    {
      "entropy": 5.710296249389648,
      "epoch": 4.840214370782056,
      "grad_norm": 1.375,
      "learning_rate": 0.0002899128060430888,
      "loss": 4.9831,
      "mean_token_accuracy": 0.1965252488851547,
      "num_tokens": 106854336.0,
      "step": 48770
    },
    {
      "entropy": 5.674123287200928,
      "epoch": 4.840710599444224,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028987745227183307,
      "loss": 4.8448,
      "mean_token_accuracy": 0.19913782626390458,
      "num_tokens": 106865486.0,
      "step": 48775
    },
    {
      "entropy": 5.653342199325562,
      "epoch": 4.841206828106391,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002898420981316486,
      "loss": 4.9041,
      "mean_token_accuracy": 0.19724292159080506,
      "num_tokens": 106876471.0,
      "step": 48780
    },
    {
      "entropy": 5.746179676055908,
      "epoch": 4.841703056768559,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028980674362341196,
      "loss": 4.9825,
      "mean_token_accuracy": 0.19413406550884246,
      "num_tokens": 106889027.0,
      "step": 48785
    },
    {
      "entropy": 5.818929195404053,
      "epoch": 4.842199285430727,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002897713887480001,
      "loss": 5.0242,
      "mean_token_accuracy": 0.19182512313127517,
      "num_tokens": 106901429.0,
      "step": 48790
    },
    {
      "entropy": 5.663073301315308,
      "epoch": 4.842695514092894,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002897360335062896,
      "loss": 4.8755,
      "mean_token_accuracy": 0.20667745918035507,
      "num_tokens": 106913545.0,
      "step": 48795
    },
    {
      "entropy": 5.6455322265625,
      "epoch": 4.843191742755062,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002897006778991572,
      "loss": 4.8728,
      "mean_token_accuracy": 0.19866805672645568,
      "num_tokens": 106924087.0,
      "step": 48800
    },
    {
      "entropy": 5.577941083908081,
      "epoch": 4.843687971417229,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028966532192747967,
      "loss": 4.8658,
      "mean_token_accuracy": 0.1921868219971657,
      "num_tokens": 106936009.0,
      "step": 48805
    },
    {
      "entropy": 5.675605154037475,
      "epoch": 4.844184200079397,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028962996559213377,
      "loss": 4.8653,
      "mean_token_accuracy": 0.20821365863084793,
      "num_tokens": 106947328.0,
      "step": 48810
    },
    {
      "entropy": 5.682539081573486,
      "epoch": 4.844680428741564,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028959460889399626,
      "loss": 4.9728,
      "mean_token_accuracy": 0.19498226046562195,
      "num_tokens": 106958544.0,
      "step": 48815
    },
    {
      "entropy": 5.659663438796997,
      "epoch": 4.845176657403732,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002895592518339439,
      "loss": 4.9313,
      "mean_token_accuracy": 0.19668834805488586,
      "num_tokens": 106969994.0,
      "step": 48820
    },
    {
      "entropy": 5.7057582378387455,
      "epoch": 4.845672886065899,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002895238944128535,
      "loss": 4.9511,
      "mean_token_accuracy": 0.19451821446418763,
      "num_tokens": 106981661.0,
      "step": 48825
    },
    {
      "entropy": 5.733078527450561,
      "epoch": 4.846169114728067,
      "grad_norm": 1.375,
      "learning_rate": 0.0002894885366316017,
      "loss": 4.9636,
      "mean_token_accuracy": 0.19829095304012298,
      "num_tokens": 106992489.0,
      "step": 48830
    },
    {
      "entropy": 5.663814306259155,
      "epoch": 4.846665343390234,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002894531784910655,
      "loss": 4.9929,
      "mean_token_accuracy": 0.19557821303606032,
      "num_tokens": 107004014.0,
      "step": 48835
    },
    {
      "entropy": 5.6639330863952635,
      "epoch": 4.847161572052402,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002894178199921216,
      "loss": 4.8991,
      "mean_token_accuracy": 0.1950862944126129,
      "num_tokens": 107016413.0,
      "step": 48840
    },
    {
      "entropy": 5.6892601490020756,
      "epoch": 4.847657800714569,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028938246113564673,
      "loss": 4.9537,
      "mean_token_accuracy": 0.19415523260831832,
      "num_tokens": 107027598.0,
      "step": 48845
    },
    {
      "entropy": 5.645717000961303,
      "epoch": 4.848154029376737,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000289347101922518,
      "loss": 4.9082,
      "mean_token_accuracy": 0.1961991861462593,
      "num_tokens": 107039554.0,
      "step": 48850
    },
    {
      "entropy": 5.720627546310425,
      "epoch": 4.848650258038904,
      "grad_norm": 1.25,
      "learning_rate": 0.00028931174235361195,
      "loss": 4.9651,
      "mean_token_accuracy": 0.19159044921398163,
      "num_tokens": 107049786.0,
      "step": 48855
    },
    {
      "entropy": 5.720208644866943,
      "epoch": 4.849146486701072,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028927638242980553,
      "loss": 4.9821,
      "mean_token_accuracy": 0.18880623430013657,
      "num_tokens": 107061192.0,
      "step": 48860
    },
    {
      "entropy": 5.670402383804321,
      "epoch": 4.849642715363239,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002892410221519756,
      "loss": 4.9793,
      "mean_token_accuracy": 0.19281671643257142,
      "num_tokens": 107073541.0,
      "step": 48865
    },
    {
      "entropy": 5.717266273498535,
      "epoch": 4.850138944025407,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028920566152099906,
      "loss": 5.0088,
      "mean_token_accuracy": 0.19387130737304686,
      "num_tokens": 107084493.0,
      "step": 48870
    },
    {
      "entropy": 5.682045984268188,
      "epoch": 4.850635172687574,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028917030053775267,
      "loss": 4.8932,
      "mean_token_accuracy": 0.2032649666070938,
      "num_tokens": 107095337.0,
      "step": 48875
    },
    {
      "entropy": 5.662584543228149,
      "epoch": 4.8511314013497415,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002891349392031134,
      "loss": 4.9098,
      "mean_token_accuracy": 0.20026459395885468,
      "num_tokens": 107105709.0,
      "step": 48880
    },
    {
      "entropy": 5.757372951507568,
      "epoch": 4.85162763001191,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028909957751795806,
      "loss": 5.0116,
      "mean_token_accuracy": 0.1890079841017723,
      "num_tokens": 107116689.0,
      "step": 48885
    },
    {
      "entropy": 5.680049991607666,
      "epoch": 4.852123858674077,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028906421548316355,
      "loss": 4.846,
      "mean_token_accuracy": 0.20538572371006011,
      "num_tokens": 107125866.0,
      "step": 48890
    },
    {
      "entropy": 5.598539257049561,
      "epoch": 4.852620087336245,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002890288530996069,
      "loss": 4.9203,
      "mean_token_accuracy": 0.20079866647720337,
      "num_tokens": 107137765.0,
      "step": 48895
    },
    {
      "entropy": 5.679372835159302,
      "epoch": 4.853116315998412,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028899349036816483,
      "loss": 4.9626,
      "mean_token_accuracy": 0.19593126326799393,
      "num_tokens": 107150042.0,
      "step": 48900
    },
    {
      "entropy": 5.673805999755859,
      "epoch": 4.85361254466058,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002889581272897143,
      "loss": 4.9446,
      "mean_token_accuracy": 0.19454423189163209,
      "num_tokens": 107160967.0,
      "step": 48905
    },
    {
      "entropy": 5.704179811477661,
      "epoch": 4.854108773322747,
      "grad_norm": 1.25,
      "learning_rate": 0.00028892276386513244,
      "loss": 5.0083,
      "mean_token_accuracy": 0.20180271565914154,
      "num_tokens": 107171079.0,
      "step": 48910
    },
    {
      "entropy": 5.719214344024659,
      "epoch": 4.854605001984915,
      "grad_norm": 1.34375,
      "learning_rate": 0.000288887400095296,
      "loss": 5.0415,
      "mean_token_accuracy": 0.19795051366090774,
      "num_tokens": 107182724.0,
      "step": 48915
    },
    {
      "entropy": 5.67552342414856,
      "epoch": 4.855101230647082,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002888520359810819,
      "loss": 4.8653,
      "mean_token_accuracy": 0.1923113539814949,
      "num_tokens": 107194253.0,
      "step": 48920
    },
    {
      "entropy": 5.63688588142395,
      "epoch": 4.85559745930925,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002888166715233672,
      "loss": 4.9612,
      "mean_token_accuracy": 0.19506543427705764,
      "num_tokens": 107204948.0,
      "step": 48925
    },
    {
      "entropy": 5.691248369216919,
      "epoch": 4.856093687971417,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002887813067230288,
      "loss": 4.9761,
      "mean_token_accuracy": 0.1899684876203537,
      "num_tokens": 107215776.0,
      "step": 48930
    },
    {
      "entropy": 5.6563636302948,
      "epoch": 4.856589916633585,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002887459415809436,
      "loss": 4.9351,
      "mean_token_accuracy": 0.19045261144638062,
      "num_tokens": 107226872.0,
      "step": 48935
    },
    {
      "entropy": 5.655428218841553,
      "epoch": 4.857086145295752,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028871057609798883,
      "loss": 4.9333,
      "mean_token_accuracy": 0.18636336773633957,
      "num_tokens": 107237904.0,
      "step": 48940
    },
    {
      "entropy": 5.705328130722046,
      "epoch": 4.85758237395792,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002886752102750413,
      "loss": 4.9902,
      "mean_token_accuracy": 0.1896054521203041,
      "num_tokens": 107247699.0,
      "step": 48945
    },
    {
      "entropy": 5.720800876617432,
      "epoch": 4.858078602620087,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028863984411297796,
      "loss": 4.9677,
      "mean_token_accuracy": 0.20423263311386108,
      "num_tokens": 107259818.0,
      "step": 48950
    },
    {
      "entropy": 5.681904363632202,
      "epoch": 4.8585748312822545,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028860447761267595,
      "loss": 4.9347,
      "mean_token_accuracy": 0.19100823402404785,
      "num_tokens": 107270194.0,
      "step": 48955
    },
    {
      "entropy": 5.6904308795928955,
      "epoch": 4.859071059944422,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002885691107750121,
      "loss": 4.9447,
      "mean_token_accuracy": 0.19517115354537964,
      "num_tokens": 107280282.0,
      "step": 48960
    },
    {
      "entropy": 5.71386752128601,
      "epoch": 4.8595672886065895,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028853374360086366,
      "loss": 4.9263,
      "mean_token_accuracy": 0.19423107653856278,
      "num_tokens": 107291549.0,
      "step": 48965
    },
    {
      "entropy": 5.742815923690796,
      "epoch": 4.860063517268758,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002884983760911074,
      "loss": 4.9871,
      "mean_token_accuracy": 0.19122356474399566,
      "num_tokens": 107302302.0,
      "step": 48970
    },
    {
      "entropy": 5.65348801612854,
      "epoch": 4.860559745930925,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002884630082466206,
      "loss": 4.9001,
      "mean_token_accuracy": 0.19599101990461348,
      "num_tokens": 107312672.0,
      "step": 48975
    },
    {
      "entropy": 5.744061756134033,
      "epoch": 4.861055974593093,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002884276400682801,
      "loss": 5.0868,
      "mean_token_accuracy": 0.19078077226877213,
      "num_tokens": 107324651.0,
      "step": 48980
    },
    {
      "entropy": 5.634518814086914,
      "epoch": 4.86155220325526,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028839227155696317,
      "loss": 4.8432,
      "mean_token_accuracy": 0.20404508113861083,
      "num_tokens": 107336012.0,
      "step": 48985
    },
    {
      "entropy": 5.784119892120361,
      "epoch": 4.862048431917428,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028835690271354674,
      "loss": 5.0541,
      "mean_token_accuracy": 0.19268283396959304,
      "num_tokens": 107348156.0,
      "step": 48990
    },
    {
      "entropy": 5.686951351165772,
      "epoch": 4.862544660579595,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002883215335389079,
      "loss": 4.9268,
      "mean_token_accuracy": 0.2002396136522293,
      "num_tokens": 107359421.0,
      "step": 48995
    },
    {
      "entropy": 5.814300584793091,
      "epoch": 4.863040889241763,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002882861640339237,
      "loss": 5.0611,
      "mean_token_accuracy": 0.19241051822900773,
      "num_tokens": 107370278.0,
      "step": 49000
    },
    {
      "entropy": 5.783464813232422,
      "epoch": 4.86353711790393,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002882507941994713,
      "loss": 5.0434,
      "mean_token_accuracy": 0.19135134071111679,
      "num_tokens": 107380837.0,
      "step": 49005
    },
    {
      "entropy": 5.7202483177185055,
      "epoch": 4.864033346566098,
      "grad_norm": 1.375,
      "learning_rate": 0.00028821542403642775,
      "loss": 4.9346,
      "mean_token_accuracy": 0.19539896696805953,
      "num_tokens": 107392103.0,
      "step": 49010
    },
    {
      "entropy": 5.692255163192749,
      "epoch": 4.864529575228265,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002881800535456702,
      "loss": 5.035,
      "mean_token_accuracy": 0.18922867476940156,
      "num_tokens": 107404453.0,
      "step": 49015
    },
    {
      "entropy": 5.674978160858155,
      "epoch": 4.865025803890433,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002881446827280756,
      "loss": 4.968,
      "mean_token_accuracy": 0.19303905367851257,
      "num_tokens": 107414307.0,
      "step": 49020
    },
    {
      "entropy": 5.714182901382446,
      "epoch": 4.8655220325526,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00028810931158452136,
      "loss": 4.9629,
      "mean_token_accuracy": 0.1956544414162636,
      "num_tokens": 107423442.0,
      "step": 49025
    },
    {
      "entropy": 5.726878309249878,
      "epoch": 4.8660182612147675,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002880739401158844,
      "loss": 4.9866,
      "mean_token_accuracy": 0.19593576490879058,
      "num_tokens": 107433135.0,
      "step": 49030
    },
    {
      "entropy": 5.701929330825806,
      "epoch": 4.866514489876935,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028803856832304174,
      "loss": 4.9707,
      "mean_token_accuracy": 0.19895504266023636,
      "num_tokens": 107443492.0,
      "step": 49035
    },
    {
      "entropy": 5.759843349456787,
      "epoch": 4.8670107185391025,
      "grad_norm": 1.375,
      "learning_rate": 0.0002880031962068709,
      "loss": 4.9623,
      "mean_token_accuracy": 0.19545700252056122,
      "num_tokens": 107453556.0,
      "step": 49040
    },
    {
      "entropy": 5.739102792739868,
      "epoch": 4.86750694720127,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002879678237682487,
      "loss": 5.0043,
      "mean_token_accuracy": 0.1940777286887169,
      "num_tokens": 107464030.0,
      "step": 49045
    },
    {
      "entropy": 5.729289197921753,
      "epoch": 4.8680031758634374,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002879324510080524,
      "loss": 4.9353,
      "mean_token_accuracy": 0.198005872964859,
      "num_tokens": 107474520.0,
      "step": 49050
    },
    {
      "entropy": 5.668497800827026,
      "epoch": 4.868499404525606,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002878970779271592,
      "loss": 4.8555,
      "mean_token_accuracy": 0.19922648817300798,
      "num_tokens": 107484917.0,
      "step": 49055
    },
    {
      "entropy": 5.707340431213379,
      "epoch": 4.868995633187773,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002878617045264463,
      "loss": 4.9686,
      "mean_token_accuracy": 0.19564418941736222,
      "num_tokens": 107496179.0,
      "step": 49060
    },
    {
      "entropy": 5.6959623336792,
      "epoch": 4.869491861849941,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002878263308067908,
      "loss": 5.0617,
      "mean_token_accuracy": 0.19095615297555923,
      "num_tokens": 107506825.0,
      "step": 49065
    },
    {
      "entropy": 5.673003435134888,
      "epoch": 4.869988090512108,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002877909567690699,
      "loss": 4.8933,
      "mean_token_accuracy": 0.20414863973855973,
      "num_tokens": 107516869.0,
      "step": 49070
    },
    {
      "entropy": 5.653318929672241,
      "epoch": 4.870484319174276,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002877555824141609,
      "loss": 4.8981,
      "mean_token_accuracy": 0.19864130020141602,
      "num_tokens": 107528165.0,
      "step": 49075
    },
    {
      "entropy": 5.688119125366211,
      "epoch": 4.870980547836443,
      "grad_norm": 1.25,
      "learning_rate": 0.0002877202077429409,
      "loss": 4.9652,
      "mean_token_accuracy": 0.18970184475183488,
      "num_tokens": 107540012.0,
      "step": 49080
    },
    {
      "entropy": 5.699485206604004,
      "epoch": 4.871476776498611,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002876848327562872,
      "loss": 4.9382,
      "mean_token_accuracy": 0.19709546118974686,
      "num_tokens": 107549503.0,
      "step": 49085
    },
    {
      "entropy": 5.670751953125,
      "epoch": 4.871973005160778,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002876494574550769,
      "loss": 4.9417,
      "mean_token_accuracy": 0.20060740560293197,
      "num_tokens": 107560445.0,
      "step": 49090
    },
    {
      "entropy": 5.711910009384155,
      "epoch": 4.872469233822946,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028761408184018735,
      "loss": 4.9881,
      "mean_token_accuracy": 0.19813707023859023,
      "num_tokens": 107571566.0,
      "step": 49095
    },
    {
      "entropy": 5.612520313262939,
      "epoch": 4.872965462485113,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028757870591249575,
      "loss": 4.9312,
      "mean_token_accuracy": 0.19184693694114685,
      "num_tokens": 107582564.0,
      "step": 49100
    },
    {
      "entropy": 5.678317403793335,
      "epoch": 4.873461691147281,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002875433296728794,
      "loss": 4.9417,
      "mean_token_accuracy": 0.20275376439094545,
      "num_tokens": 107592377.0,
      "step": 49105
    },
    {
      "entropy": 5.738354063034057,
      "epoch": 4.873957919809448,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002875079531222154,
      "loss": 4.9356,
      "mean_token_accuracy": 0.1973799630999565,
      "num_tokens": 107604145.0,
      "step": 49110
    },
    {
      "entropy": 5.799729537963867,
      "epoch": 4.8744541484716155,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028747257626138116,
      "loss": 5.0764,
      "mean_token_accuracy": 0.18491120934486388,
      "num_tokens": 107616545.0,
      "step": 49115
    },
    {
      "entropy": 5.774167013168335,
      "epoch": 4.874950377133783,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028743719909125404,
      "loss": 5.029,
      "mean_token_accuracy": 0.1899004817008972,
      "num_tokens": 107629451.0,
      "step": 49120
    },
    {
      "entropy": 5.716164779663086,
      "epoch": 4.8754466057959505,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028740182161271103,
      "loss": 4.993,
      "mean_token_accuracy": 0.1928333818912506,
      "num_tokens": 107640935.0,
      "step": 49125
    },
    {
      "entropy": 5.720112419128418,
      "epoch": 4.875942834458118,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002873664438266296,
      "loss": 4.9894,
      "mean_token_accuracy": 0.18827591240406036,
      "num_tokens": 107651560.0,
      "step": 49130
    },
    {
      "entropy": 5.8005101680755615,
      "epoch": 4.876439063120285,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002873310657338871,
      "loss": 5.0896,
      "mean_token_accuracy": 0.18845883756875992,
      "num_tokens": 107662240.0,
      "step": 49135
    },
    {
      "entropy": 5.731393814086914,
      "epoch": 4.876935291782454,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002872956873353605,
      "loss": 4.9547,
      "mean_token_accuracy": 0.19402487576007843,
      "num_tokens": 107672979.0,
      "step": 49140
    },
    {
      "entropy": 5.793628358840943,
      "epoch": 4.877431520444621,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002872603086319276,
      "loss": 5.0546,
      "mean_token_accuracy": 0.18835317492485046,
      "num_tokens": 107683690.0,
      "step": 49145
    },
    {
      "entropy": 5.651963901519776,
      "epoch": 4.877927749106789,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002872249296244654,
      "loss": 4.9692,
      "mean_token_accuracy": 0.1986391633749008,
      "num_tokens": 107694823.0,
      "step": 49150
    },
    {
      "entropy": 5.721808433532715,
      "epoch": 4.878423977768956,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028718955031385127,
      "loss": 4.9761,
      "mean_token_accuracy": 0.19909488707780837,
      "num_tokens": 107706364.0,
      "step": 49155
    },
    {
      "entropy": 5.691238260269165,
      "epoch": 4.878920206431124,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002871541707009625,
      "loss": 4.9806,
      "mean_token_accuracy": 0.19488314241170884,
      "num_tokens": 107718248.0,
      "step": 49160
    },
    {
      "entropy": 5.693844318389893,
      "epoch": 4.879416435093291,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002871187907866766,
      "loss": 4.9356,
      "mean_token_accuracy": 0.19514503180980683,
      "num_tokens": 107730781.0,
      "step": 49165
    },
    {
      "entropy": 5.73176326751709,
      "epoch": 4.879912663755459,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028708341057187065,
      "loss": 5.0658,
      "mean_token_accuracy": 0.18657437562942505,
      "num_tokens": 107742183.0,
      "step": 49170
    },
    {
      "entropy": 5.765174293518067,
      "epoch": 4.880408892417626,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028704803005742224,
      "loss": 5.0033,
      "mean_token_accuracy": 0.18955680280923842,
      "num_tokens": 107753405.0,
      "step": 49175
    },
    {
      "entropy": 5.711693000793457,
      "epoch": 4.880905121079794,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028701264924420864,
      "loss": 4.9602,
      "mean_token_accuracy": 0.1936130404472351,
      "num_tokens": 107765925.0,
      "step": 49180
    },
    {
      "entropy": 5.714852285385132,
      "epoch": 4.881401349741961,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028697726813310724,
      "loss": 4.9138,
      "mean_token_accuracy": 0.2019015833735466,
      "num_tokens": 107777793.0,
      "step": 49185
    },
    {
      "entropy": 5.745180320739746,
      "epoch": 4.8818975784041285,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00028694188672499543,
      "loss": 4.9871,
      "mean_token_accuracy": 0.18605252206325532,
      "num_tokens": 107787364.0,
      "step": 49190
    },
    {
      "entropy": 5.664391231536865,
      "epoch": 4.882393807066296,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002869065050207505,
      "loss": 4.9463,
      "mean_token_accuracy": 0.19635828584432602,
      "num_tokens": 107799952.0,
      "step": 49195
    },
    {
      "entropy": 5.735071086883545,
      "epoch": 4.8828900357284635,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002868711230212499,
      "loss": 4.9567,
      "mean_token_accuracy": 0.1953534871339798,
      "num_tokens": 107811591.0,
      "step": 49200
    },
    {
      "entropy": 5.640595579147339,
      "epoch": 4.883386264390631,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028683574072737114,
      "loss": 4.8096,
      "mean_token_accuracy": 0.207683002948761,
      "num_tokens": 107822355.0,
      "step": 49205
    },
    {
      "entropy": 5.7405884742736815,
      "epoch": 4.883882493052798,
      "grad_norm": 1.25,
      "learning_rate": 0.0002868003581399914,
      "loss": 4.9355,
      "mean_token_accuracy": 0.1980319157242775,
      "num_tokens": 107832325.0,
      "step": 49210
    },
    {
      "entropy": 5.655742597579956,
      "epoch": 4.884378721714966,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028676497525998817,
      "loss": 4.9249,
      "mean_token_accuracy": 0.19247696101665496,
      "num_tokens": 107841757.0,
      "step": 49215
    },
    {
      "entropy": 5.609076690673828,
      "epoch": 4.884874950377133,
      "grad_norm": 1.328125,
      "learning_rate": 0.000286729592088239,
      "loss": 4.9574,
      "mean_token_accuracy": 0.1942150130867958,
      "num_tokens": 107851250.0,
      "step": 49220
    },
    {
      "entropy": 5.714510679244995,
      "epoch": 4.885371179039302,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028669420862562116,
      "loss": 4.9661,
      "mean_token_accuracy": 0.19246310889720916,
      "num_tokens": 107862081.0,
      "step": 49225
    },
    {
      "entropy": 5.7255456924438475,
      "epoch": 4.885867407701469,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028665882487301215,
      "loss": 4.9613,
      "mean_token_accuracy": 0.19826064258813858,
      "num_tokens": 107873060.0,
      "step": 49230
    },
    {
      "entropy": 5.701931953430176,
      "epoch": 4.886363636363637,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028662344083128945,
      "loss": 4.9097,
      "mean_token_accuracy": 0.1930331289768219,
      "num_tokens": 107883605.0,
      "step": 49235
    },
    {
      "entropy": 5.702231216430664,
      "epoch": 4.886859865025804,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028658805650133047,
      "loss": 4.9845,
      "mean_token_accuracy": 0.1905038520693779,
      "num_tokens": 107893593.0,
      "step": 49240
    },
    {
      "entropy": 5.639764928817749,
      "epoch": 4.887356093687972,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028655267188401265,
      "loss": 4.959,
      "mean_token_accuracy": 0.1933991014957428,
      "num_tokens": 107904921.0,
      "step": 49245
    },
    {
      "entropy": 5.711206245422363,
      "epoch": 4.887852322350139,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002865172869802134,
      "loss": 4.9517,
      "mean_token_accuracy": 0.19902004599571227,
      "num_tokens": 107915907.0,
      "step": 49250
    },
    {
      "entropy": 5.691832447052002,
      "epoch": 4.888348551012307,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028648190179081026,
      "loss": 4.868,
      "mean_token_accuracy": 0.19713079929351807,
      "num_tokens": 107926236.0,
      "step": 49255
    },
    {
      "entropy": 5.756290531158447,
      "epoch": 4.888844779674474,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002864465163166808,
      "loss": 5.0147,
      "mean_token_accuracy": 0.1887855902314186,
      "num_tokens": 107936636.0,
      "step": 49260
    },
    {
      "entropy": 5.598997163772583,
      "epoch": 4.8893410083366415,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002864111305587023,
      "loss": 4.8138,
      "mean_token_accuracy": 0.20091554522514343,
      "num_tokens": 107947255.0,
      "step": 49265
    },
    {
      "entropy": 5.6858827590942385,
      "epoch": 4.889837236998809,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002863757445177523,
      "loss": 4.9546,
      "mean_token_accuracy": 0.19625718593597413,
      "num_tokens": 107958897.0,
      "step": 49270
    },
    {
      "entropy": 5.686600589752198,
      "epoch": 4.8903334656609765,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002863403581947085,
      "loss": 4.9729,
      "mean_token_accuracy": 0.20026966631412507,
      "num_tokens": 107969770.0,
      "step": 49275
    },
    {
      "entropy": 5.688115978240967,
      "epoch": 4.890829694323144,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028630497159044815,
      "loss": 4.9881,
      "mean_token_accuracy": 0.1884098768234253,
      "num_tokens": 107980620.0,
      "step": 49280
    },
    {
      "entropy": 5.729470491409302,
      "epoch": 4.891325922985311,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002862695847058489,
      "loss": 4.9761,
      "mean_token_accuracy": 0.19170249551534652,
      "num_tokens": 107991041.0,
      "step": 49285
    },
    {
      "entropy": 5.736430215835571,
      "epoch": 4.891822151647479,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002862341975417883,
      "loss": 5.0426,
      "mean_token_accuracy": 0.191471865773201,
      "num_tokens": 108002213.0,
      "step": 49290
    },
    {
      "entropy": 5.666288042068482,
      "epoch": 4.892318380309646,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028619881009914373,
      "loss": 4.9569,
      "mean_token_accuracy": 0.1929419904947281,
      "num_tokens": 108014119.0,
      "step": 49295
    },
    {
      "entropy": 5.730888700485229,
      "epoch": 4.892814608971814,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002861634223787927,
      "loss": 4.9859,
      "mean_token_accuracy": 0.19555435329675674,
      "num_tokens": 108024195.0,
      "step": 49300
    },
    {
      "entropy": 5.709969520568848,
      "epoch": 4.893310837633981,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028612803438161303,
      "loss": 4.9636,
      "mean_token_accuracy": 0.19761505722999573,
      "num_tokens": 108036021.0,
      "step": 49305
    },
    {
      "entropy": 5.673152017593384,
      "epoch": 4.89380706629615,
      "grad_norm": 1.4375,
      "learning_rate": 0.00028609264610848204,
      "loss": 4.9753,
      "mean_token_accuracy": 0.19291479289531707,
      "num_tokens": 108046021.0,
      "step": 49310
    },
    {
      "entropy": 5.7253992557525635,
      "epoch": 4.894303294958317,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002860572575602773,
      "loss": 4.9762,
      "mean_token_accuracy": 0.1911287099123001,
      "num_tokens": 108057537.0,
      "step": 49315
    },
    {
      "entropy": 5.671524953842163,
      "epoch": 4.894799523620485,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002860218687378764,
      "loss": 4.8639,
      "mean_token_accuracy": 0.20242854505777358,
      "num_tokens": 108068511.0,
      "step": 49320
    },
    {
      "entropy": 5.741006898880005,
      "epoch": 4.895295752282652,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028598647964215695,
      "loss": 5.0191,
      "mean_token_accuracy": 0.19636430591344833,
      "num_tokens": 108080108.0,
      "step": 49325
    },
    {
      "entropy": 5.719200658798218,
      "epoch": 4.89579198094482,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028595109027399634,
      "loss": 4.9453,
      "mean_token_accuracy": 0.19404344856739045,
      "num_tokens": 108091202.0,
      "step": 49330
    },
    {
      "entropy": 5.6544122219085695,
      "epoch": 4.896288209606987,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002859157006342724,
      "loss": 4.9525,
      "mean_token_accuracy": 0.19966288655996323,
      "num_tokens": 108102951.0,
      "step": 49335
    },
    {
      "entropy": 5.679069757461548,
      "epoch": 4.8967844382691545,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002858803107238626,
      "loss": 4.9285,
      "mean_token_accuracy": 0.19068968743085862,
      "num_tokens": 108113215.0,
      "step": 49340
    },
    {
      "entropy": 5.638494968414307,
      "epoch": 4.897280666931322,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002858449205436445,
      "loss": 4.9418,
      "mean_token_accuracy": 0.19375107288360596,
      "num_tokens": 108125285.0,
      "step": 49345
    },
    {
      "entropy": 5.705183982849121,
      "epoch": 4.8977768955934895,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002858095300944958,
      "loss": 5.0147,
      "mean_token_accuracy": 0.19048849791288375,
      "num_tokens": 108135096.0,
      "step": 49350
    },
    {
      "entropy": 5.627990293502807,
      "epoch": 4.898273124255657,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002857741393772941,
      "loss": 4.9671,
      "mean_token_accuracy": 0.19867169708013535,
      "num_tokens": 108145975.0,
      "step": 49355
    },
    {
      "entropy": 5.661606216430664,
      "epoch": 4.8987693529178244,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002857387483929168,
      "loss": 4.9148,
      "mean_token_accuracy": 0.19715655446052552,
      "num_tokens": 108156972.0,
      "step": 49360
    },
    {
      "entropy": 5.652564620971679,
      "epoch": 4.899265581579992,
      "grad_norm": 1.25,
      "learning_rate": 0.0002857033571422418,
      "loss": 4.9625,
      "mean_token_accuracy": 0.19596672356128692,
      "num_tokens": 108168871.0,
      "step": 49365
    },
    {
      "entropy": 5.640952110290527,
      "epoch": 4.899761810242159,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028566796562614656,
      "loss": 4.9113,
      "mean_token_accuracy": 0.19643002450466157,
      "num_tokens": 108179828.0,
      "step": 49370
    },
    {
      "entropy": 5.747585725784302,
      "epoch": 4.900258038904327,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028563257384550873,
      "loss": 5.0097,
      "mean_token_accuracy": 0.18629425019025803,
      "num_tokens": 108192029.0,
      "step": 49375
    },
    {
      "entropy": 5.627841997146606,
      "epoch": 4.900754267566494,
      "grad_norm": 1.21875,
      "learning_rate": 0.000285597181801206,
      "loss": 4.8718,
      "mean_token_accuracy": 0.2008504018187523,
      "num_tokens": 108203324.0,
      "step": 49380
    },
    {
      "entropy": 5.717341566085816,
      "epoch": 4.901250496228662,
      "grad_norm": 1.125,
      "learning_rate": 0.00028556178949411606,
      "loss": 5.0084,
      "mean_token_accuracy": 0.19252465665340424,
      "num_tokens": 108214571.0,
      "step": 49385
    },
    {
      "entropy": 5.690251064300537,
      "epoch": 4.901746724890829,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002855263969251164,
      "loss": 4.9323,
      "mean_token_accuracy": 0.19449903666973115,
      "num_tokens": 108224837.0,
      "step": 49390
    },
    {
      "entropy": 5.644850969314575,
      "epoch": 4.902242953552998,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002854910040950848,
      "loss": 4.8647,
      "mean_token_accuracy": 0.1983622968196869,
      "num_tokens": 108234912.0,
      "step": 49395
    },
    {
      "entropy": 5.67661566734314,
      "epoch": 4.902739182215164,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028545561100489896,
      "loss": 4.9433,
      "mean_token_accuracy": 0.1951085537672043,
      "num_tokens": 108245746.0,
      "step": 49400
    },
    {
      "entropy": 5.6929230213165285,
      "epoch": 4.903235410877333,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028542021765543647,
      "loss": 4.9288,
      "mean_token_accuracy": 0.19158796817064286,
      "num_tokens": 108257287.0,
      "step": 49405
    },
    {
      "entropy": 5.799065256118775,
      "epoch": 4.9037316395395,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028538482404757504,
      "loss": 5.0847,
      "mean_token_accuracy": 0.1895320326089859,
      "num_tokens": 108268616.0,
      "step": 49410
    },
    {
      "entropy": 5.6632222652435305,
      "epoch": 4.904227868201668,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028534943018219234,
      "loss": 4.9319,
      "mean_token_accuracy": 0.19482507854700087,
      "num_tokens": 108279483.0,
      "step": 49415
    },
    {
      "entropy": 5.694629287719726,
      "epoch": 4.904724096863835,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028531403606016603,
      "loss": 4.9638,
      "mean_token_accuracy": 0.18986564576625825,
      "num_tokens": 108290317.0,
      "step": 49420
    },
    {
      "entropy": 5.68295693397522,
      "epoch": 4.9052203255260025,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002852786416823739,
      "loss": 4.9261,
      "mean_token_accuracy": 0.19861290007829666,
      "num_tokens": 108300688.0,
      "step": 49425
    },
    {
      "entropy": 5.711251306533813,
      "epoch": 4.90571655418817,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002852432470496936,
      "loss": 4.9824,
      "mean_token_accuracy": 0.1947169929742813,
      "num_tokens": 108310472.0,
      "step": 49430
    },
    {
      "entropy": 5.678412675857544,
      "epoch": 4.9062127828503375,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002852078521630028,
      "loss": 4.9495,
      "mean_token_accuracy": 0.19421568512916565,
      "num_tokens": 108321782.0,
      "step": 49435
    },
    {
      "entropy": 5.730966091156006,
      "epoch": 4.906709011512505,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002851724570231794,
      "loss": 5.0209,
      "mean_token_accuracy": 0.19116199761629105,
      "num_tokens": 108330995.0,
      "step": 49440
    },
    {
      "entropy": 5.693845415115357,
      "epoch": 4.907205240174672,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002851370616311009,
      "loss": 4.9363,
      "mean_token_accuracy": 0.19330329149961473,
      "num_tokens": 108341065.0,
      "step": 49445
    },
    {
      "entropy": 5.743324661254883,
      "epoch": 4.90770146883684,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002851016659876451,
      "loss": 5.0191,
      "mean_token_accuracy": 0.19471757262945175,
      "num_tokens": 108352502.0,
      "step": 49450
    },
    {
      "entropy": 5.729592418670654,
      "epoch": 4.908197697499007,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002850662700936898,
      "loss": 5.003,
      "mean_token_accuracy": 0.19312090575695037,
      "num_tokens": 108363287.0,
      "step": 49455
    },
    {
      "entropy": 5.7036927223205565,
      "epoch": 4.908693926161175,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002850308739501126,
      "loss": 4.9536,
      "mean_token_accuracy": 0.1964465320110321,
      "num_tokens": 108373883.0,
      "step": 49460
    },
    {
      "entropy": 5.665826845169067,
      "epoch": 4.909190154823342,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002849954775577914,
      "loss": 4.9043,
      "mean_token_accuracy": 0.1962430641055107,
      "num_tokens": 108384518.0,
      "step": 49465
    },
    {
      "entropy": 5.737078905105591,
      "epoch": 4.90968638348551,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002849600809176039,
      "loss": 5.0052,
      "mean_token_accuracy": 0.18915784358978271,
      "num_tokens": 108395104.0,
      "step": 49470
    },
    {
      "entropy": 5.673831605911255,
      "epoch": 4.910182612147677,
      "grad_norm": 1.375,
      "learning_rate": 0.0002849246840304278,
      "loss": 4.9303,
      "mean_token_accuracy": 0.19725598245859147,
      "num_tokens": 108406186.0,
      "step": 49475
    },
    {
      "entropy": 5.668736171722412,
      "epoch": 4.910678840809846,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028488928689714095,
      "loss": 4.9148,
      "mean_token_accuracy": 0.19688017815351486,
      "num_tokens": 108416652.0,
      "step": 49480
    },
    {
      "entropy": 5.697911310195923,
      "epoch": 4.911175069472012,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002848538895186212,
      "loss": 4.9624,
      "mean_token_accuracy": 0.19347925037145614,
      "num_tokens": 108427029.0,
      "step": 49485
    },
    {
      "entropy": 5.6564521312713625,
      "epoch": 4.911671298134181,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028481849189574603,
      "loss": 4.8938,
      "mean_token_accuracy": 0.19523970484733583,
      "num_tokens": 108437668.0,
      "step": 49490
    },
    {
      "entropy": 5.71690559387207,
      "epoch": 4.912167526796348,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002847830940293935,
      "loss": 5.0504,
      "mean_token_accuracy": 0.1911458820104599,
      "num_tokens": 108450010.0,
      "step": 49495
    },
    {
      "entropy": 5.733745145797729,
      "epoch": 4.9126637554585155,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002847476959204414,
      "loss": 4.9971,
      "mean_token_accuracy": 0.18673628568649292,
      "num_tokens": 108461599.0,
      "step": 49500
    },
    {
      "entropy": 5.762675380706787,
      "epoch": 4.913159984120683,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028471229756976734,
      "loss": 5.0785,
      "mean_token_accuracy": 0.1846625804901123,
      "num_tokens": 108473088.0,
      "step": 49505
    },
    {
      "entropy": 5.675570678710938,
      "epoch": 4.9136562127828505,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002846768989782493,
      "loss": 4.9234,
      "mean_token_accuracy": 0.20012015104293823,
      "num_tokens": 108485245.0,
      "step": 49510
    },
    {
      "entropy": 5.699549579620362,
      "epoch": 4.914152441445018,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028464150014676507,
      "loss": 5.0023,
      "mean_token_accuracy": 0.19474540054798126,
      "num_tokens": 108495017.0,
      "step": 49515
    },
    {
      "entropy": 5.65255184173584,
      "epoch": 4.914648670107185,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002846061010761923,
      "loss": 4.8631,
      "mean_token_accuracy": 0.20624422430992126,
      "num_tokens": 108504437.0,
      "step": 49520
    },
    {
      "entropy": 5.621857833862305,
      "epoch": 4.915144898769353,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002845707017674089,
      "loss": 4.8533,
      "mean_token_accuracy": 0.19838731437921525,
      "num_tokens": 108514558.0,
      "step": 49525
    },
    {
      "entropy": 5.692378950119019,
      "epoch": 4.91564112743152,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002845353022212928,
      "loss": 4.9349,
      "mean_token_accuracy": 0.196841099858284,
      "num_tokens": 108525713.0,
      "step": 49530
    },
    {
      "entropy": 5.659377384185791,
      "epoch": 4.916137356093688,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028449990243872175,
      "loss": 4.9119,
      "mean_token_accuracy": 0.191917085647583,
      "num_tokens": 108536427.0,
      "step": 49535
    },
    {
      "entropy": 5.695629215240478,
      "epoch": 4.916633584755855,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002844645024205735,
      "loss": 4.914,
      "mean_token_accuracy": 0.19926851689815522,
      "num_tokens": 108547603.0,
      "step": 49540
    },
    {
      "entropy": 5.660531949996948,
      "epoch": 4.917129813418023,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002844291021677261,
      "loss": 4.9159,
      "mean_token_accuracy": 0.1988848328590393,
      "num_tokens": 108557167.0,
      "step": 49545
    },
    {
      "entropy": 5.717413663864136,
      "epoch": 4.91762604208019,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002843937016810572,
      "loss": 4.9341,
      "mean_token_accuracy": 0.19738865196704863,
      "num_tokens": 108568510.0,
      "step": 49550
    },
    {
      "entropy": 5.642755079269409,
      "epoch": 4.918122270742358,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002843583009614448,
      "loss": 4.9263,
      "mean_token_accuracy": 0.19255252331495284,
      "num_tokens": 108578855.0,
      "step": 49555
    },
    {
      "entropy": 5.670193862915039,
      "epoch": 4.918618499404525,
      "grad_norm": 1.375,
      "learning_rate": 0.0002843229000097667,
      "loss": 4.9802,
      "mean_token_accuracy": 0.18451395034790039,
      "num_tokens": 108588605.0,
      "step": 49560
    },
    {
      "entropy": 5.701480579376221,
      "epoch": 4.919114728066694,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028428749882690067,
      "loss": 4.9324,
      "mean_token_accuracy": 0.19661615341901778,
      "num_tokens": 108601397.0,
      "step": 49565
    },
    {
      "entropy": 5.702905988693237,
      "epoch": 4.91961095672886,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002842520974137248,
      "loss": 4.9678,
      "mean_token_accuracy": 0.19009876251220703,
      "num_tokens": 108611989.0,
      "step": 49570
    },
    {
      "entropy": 5.734916687011719,
      "epoch": 4.9201071853910285,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028421669577111683,
      "loss": 4.9866,
      "mean_token_accuracy": 0.1900283232331276,
      "num_tokens": 108623660.0,
      "step": 49575
    },
    {
      "entropy": 5.651283693313599,
      "epoch": 4.920603414053196,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002841812938999546,
      "loss": 4.8775,
      "mean_token_accuracy": 0.19596490263938904,
      "num_tokens": 108635236.0,
      "step": 49580
    },
    {
      "entropy": 5.699775505065918,
      "epoch": 4.9210996427153635,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028414589180111616,
      "loss": 4.9773,
      "mean_token_accuracy": 0.19063687175512314,
      "num_tokens": 108646397.0,
      "step": 49585
    },
    {
      "entropy": 5.736976766586304,
      "epoch": 4.921595871377531,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028411048947547926,
      "loss": 5.041,
      "mean_token_accuracy": 0.19055216610431672,
      "num_tokens": 108656930.0,
      "step": 49590
    },
    {
      "entropy": 5.705876111984253,
      "epoch": 4.922092100039698,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002840750869239218,
      "loss": 4.9772,
      "mean_token_accuracy": 0.19471403807401658,
      "num_tokens": 108667965.0,
      "step": 49595
    },
    {
      "entropy": 5.708201313018799,
      "epoch": 4.922588328701866,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028403968414732184,
      "loss": 4.974,
      "mean_token_accuracy": 0.19167367219924927,
      "num_tokens": 108677003.0,
      "step": 49600
    },
    {
      "entropy": 5.7325176239013675,
      "epoch": 4.923084557364033,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002840042811465571,
      "loss": 5.0199,
      "mean_token_accuracy": 0.1903655856847763,
      "num_tokens": 108688603.0,
      "step": 49605
    },
    {
      "entropy": 5.678376054763794,
      "epoch": 4.923580786026201,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002839688779225056,
      "loss": 4.9268,
      "mean_token_accuracy": 0.18828518241643905,
      "num_tokens": 108700991.0,
      "step": 49610
    },
    {
      "entropy": 5.766035699844361,
      "epoch": 4.924077014688368,
      "grad_norm": 1.171875,
      "learning_rate": 0.00028393347447604536,
      "loss": 5.0918,
      "mean_token_accuracy": 0.18730438202619554,
      "num_tokens": 108712276.0,
      "step": 49615
    },
    {
      "entropy": 5.734998083114624,
      "epoch": 4.924573243350536,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028389807080805407,
      "loss": 5.0299,
      "mean_token_accuracy": 0.18658512085676193,
      "num_tokens": 108724500.0,
      "step": 49620
    },
    {
      "entropy": 5.736891031265259,
      "epoch": 4.925069472012703,
      "grad_norm": 1.25,
      "learning_rate": 0.00028386266691940985,
      "loss": 4.9984,
      "mean_token_accuracy": 0.1837714985013008,
      "num_tokens": 108734671.0,
      "step": 49625
    },
    {
      "entropy": 5.704759168624878,
      "epoch": 4.925565700674871,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028382726281099064,
      "loss": 4.9401,
      "mean_token_accuracy": 0.19694546312093736,
      "num_tokens": 108745131.0,
      "step": 49630
    },
    {
      "entropy": 5.720628690719605,
      "epoch": 4.926061929337038,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028379185848367427,
      "loss": 4.977,
      "mean_token_accuracy": 0.19019268155097963,
      "num_tokens": 108756183.0,
      "step": 49635
    },
    {
      "entropy": 5.720554447174072,
      "epoch": 4.926558157999206,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028375645393833877,
      "loss": 4.9594,
      "mean_token_accuracy": 0.18906435519456863,
      "num_tokens": 108766777.0,
      "step": 49640
    },
    {
      "entropy": 5.684382820129395,
      "epoch": 4.927054386661373,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028372104917586213,
      "loss": 4.9795,
      "mean_token_accuracy": 0.18408872336149215,
      "num_tokens": 108776928.0,
      "step": 49645
    },
    {
      "entropy": 5.710606098175049,
      "epoch": 4.9275506153235415,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028368564419712223,
      "loss": 4.9865,
      "mean_token_accuracy": 0.2003144606947899,
      "num_tokens": 108789636.0,
      "step": 49650
    },
    {
      "entropy": 5.734674882888794,
      "epoch": 4.928046843985708,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028365023900299696,
      "loss": 4.9868,
      "mean_token_accuracy": 0.18921787440776824,
      "num_tokens": 108800485.0,
      "step": 49655
    },
    {
      "entropy": 5.719695138931274,
      "epoch": 4.9285430726478765,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002836148335943645,
      "loss": 4.9611,
      "mean_token_accuracy": 0.19457032084465026,
      "num_tokens": 108810312.0,
      "step": 49660
    },
    {
      "entropy": 5.670450592041016,
      "epoch": 4.929039301310044,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028357942797210286,
      "loss": 4.8826,
      "mean_token_accuracy": 0.20125064998865128,
      "num_tokens": 108821137.0,
      "step": 49665
    },
    {
      "entropy": 5.595995998382568,
      "epoch": 4.9295355299722114,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002835440221370897,
      "loss": 4.8619,
      "mean_token_accuracy": 0.200641867518425,
      "num_tokens": 108832493.0,
      "step": 49670
    },
    {
      "entropy": 5.686609888076783,
      "epoch": 4.930031758634379,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028350861609020324,
      "loss": 5.0123,
      "mean_token_accuracy": 0.19497143477201462,
      "num_tokens": 108843500.0,
      "step": 49675
    },
    {
      "entropy": 5.687665605545044,
      "epoch": 4.930527987296546,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002834732098323215,
      "loss": 4.8316,
      "mean_token_accuracy": 0.19798302948474883,
      "num_tokens": 108854354.0,
      "step": 49680
    },
    {
      "entropy": 5.649653100967408,
      "epoch": 4.931024215958714,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002834378033643223,
      "loss": 4.8813,
      "mean_token_accuracy": 0.1980280876159668,
      "num_tokens": 108864925.0,
      "step": 49685
    },
    {
      "entropy": 5.727272796630859,
      "epoch": 4.931520444620881,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002834023966870838,
      "loss": 4.9861,
      "mean_token_accuracy": 0.18945867717266082,
      "num_tokens": 108876236.0,
      "step": 49690
    },
    {
      "entropy": 5.638301753997803,
      "epoch": 4.932016673283049,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028336698980148406,
      "loss": 4.8751,
      "mean_token_accuracy": 0.1980869561433792,
      "num_tokens": 108887308.0,
      "step": 49695
    },
    {
      "entropy": 5.6811628341674805,
      "epoch": 4.932512901945216,
      "grad_norm": 1.25,
      "learning_rate": 0.00028333158270840096,
      "loss": 4.9072,
      "mean_token_accuracy": 0.19724484980106355,
      "num_tokens": 108897634.0,
      "step": 49700
    },
    {
      "entropy": 5.655720615386963,
      "epoch": 4.933009130607384,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002832961754087125,
      "loss": 4.9249,
      "mean_token_accuracy": 0.19752994775772095,
      "num_tokens": 108909940.0,
      "step": 49705
    },
    {
      "entropy": 5.701943635940552,
      "epoch": 4.933505359269551,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002832607679032968,
      "loss": 4.9656,
      "mean_token_accuracy": 0.19224682748317717,
      "num_tokens": 108920421.0,
      "step": 49710
    },
    {
      "entropy": 5.611367082595825,
      "epoch": 4.934001587931719,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028322536019303194,
      "loss": 4.8896,
      "mean_token_accuracy": 0.2050833135843277,
      "num_tokens": 108931921.0,
      "step": 49715
    },
    {
      "entropy": 5.699471282958984,
      "epoch": 4.934497816593886,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028318995227879574,
      "loss": 4.9594,
      "mean_token_accuracy": 0.196262426674366,
      "num_tokens": 108942523.0,
      "step": 49720
    },
    {
      "entropy": 5.64999828338623,
      "epoch": 4.934994045256054,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002831545441614664,
      "loss": 4.9475,
      "mean_token_accuracy": 0.19832990169525147,
      "num_tokens": 108954263.0,
      "step": 49725
    },
    {
      "entropy": 5.733028888702393,
      "epoch": 4.935490273918221,
      "grad_norm": 1.21875,
      "learning_rate": 0.000283119135841922,
      "loss": 4.9893,
      "mean_token_accuracy": 0.18861502408981323,
      "num_tokens": 108965254.0,
      "step": 49730
    },
    {
      "entropy": 5.73350305557251,
      "epoch": 4.9359865025803895,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028308372732104047,
      "loss": 4.9948,
      "mean_token_accuracy": 0.1924214169383049,
      "num_tokens": 108976394.0,
      "step": 49735
    },
    {
      "entropy": 5.709481906890869,
      "epoch": 4.936482731242556,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028304831859969994,
      "loss": 4.9916,
      "mean_token_accuracy": 0.19477350562810897,
      "num_tokens": 108986847.0,
      "step": 49740
    },
    {
      "entropy": 5.618818712234497,
      "epoch": 4.9369789599047245,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002830129096787785,
      "loss": 4.8572,
      "mean_token_accuracy": 0.2026353195309639,
      "num_tokens": 108997220.0,
      "step": 49745
    },
    {
      "entropy": 5.662354326248169,
      "epoch": 4.937475188566892,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028297750055915414,
      "loss": 4.9332,
      "mean_token_accuracy": 0.19241031557321547,
      "num_tokens": 109008150.0,
      "step": 49750
    },
    {
      "entropy": 5.65978627204895,
      "epoch": 4.937971417229059,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002829420912417049,
      "loss": 4.9305,
      "mean_token_accuracy": 0.19833489656448364,
      "num_tokens": 109018004.0,
      "step": 49755
    },
    {
      "entropy": 5.6702625274658205,
      "epoch": 4.938467645891227,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002829066817273089,
      "loss": 4.9609,
      "mean_token_accuracy": 0.19543592035770416,
      "num_tokens": 109028786.0,
      "step": 49760
    },
    {
      "entropy": 5.746601676940918,
      "epoch": 4.938963874553394,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028287127201684435,
      "loss": 4.9553,
      "mean_token_accuracy": 0.20287011414766312,
      "num_tokens": 109039956.0,
      "step": 49765
    },
    {
      "entropy": 5.6851646900177,
      "epoch": 4.939460103215562,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002828358621111891,
      "loss": 4.9417,
      "mean_token_accuracy": 0.19608947038650512,
      "num_tokens": 109052008.0,
      "step": 49770
    },
    {
      "entropy": 5.738470554351807,
      "epoch": 4.939956331877729,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002828004520112214,
      "loss": 4.9424,
      "mean_token_accuracy": 0.1975746840238571,
      "num_tokens": 109064188.0,
      "step": 49775
    },
    {
      "entropy": 5.662957048416137,
      "epoch": 4.940452560539897,
      "grad_norm": 1.125,
      "learning_rate": 0.0002827650417178193,
      "loss": 4.9667,
      "mean_token_accuracy": 0.19939926117658616,
      "num_tokens": 109075891.0,
      "step": 49780
    },
    {
      "entropy": 5.740790605545044,
      "epoch": 4.940948789202064,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002827296312318609,
      "loss": 4.9607,
      "mean_token_accuracy": 0.19175893813371658,
      "num_tokens": 109086971.0,
      "step": 49785
    },
    {
      "entropy": 5.703428649902344,
      "epoch": 4.941445017864232,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002826942205542243,
      "loss": 4.9586,
      "mean_token_accuracy": 0.18892175406217576,
      "num_tokens": 109097229.0,
      "step": 49790
    },
    {
      "entropy": 5.705257368087769,
      "epoch": 4.941941246526399,
      "grad_norm": 1.375,
      "learning_rate": 0.0002826588096857877,
      "loss": 4.9598,
      "mean_token_accuracy": 0.19456608295440675,
      "num_tokens": 109107303.0,
      "step": 49795
    },
    {
      "entropy": 5.728064823150635,
      "epoch": 4.942437475188567,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028262339862742894,
      "loss": 4.9915,
      "mean_token_accuracy": 0.1935675784945488,
      "num_tokens": 109117164.0,
      "step": 49800
    },
    {
      "entropy": 5.691151428222656,
      "epoch": 4.942933703850734,
      "grad_norm": 1.25,
      "learning_rate": 0.00028258798738002637,
      "loss": 4.9664,
      "mean_token_accuracy": 0.20466688722372056,
      "num_tokens": 109129387.0,
      "step": 49805
    },
    {
      "entropy": 5.743196153640747,
      "epoch": 4.943429932512902,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002825525759444582,
      "loss": 4.9535,
      "mean_token_accuracy": 0.19888169914484025,
      "num_tokens": 109139800.0,
      "step": 49810
    },
    {
      "entropy": 5.737924385070801,
      "epoch": 4.943926161175069,
      "grad_norm": 1.25,
      "learning_rate": 0.00028251716432160226,
      "loss": 4.9484,
      "mean_token_accuracy": 0.20014910250902176,
      "num_tokens": 109149764.0,
      "step": 49815
    },
    {
      "entropy": 5.660711193084717,
      "epoch": 4.9444223898372375,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002824817525123369,
      "loss": 4.9341,
      "mean_token_accuracy": 0.19300374984741211,
      "num_tokens": 109160427.0,
      "step": 49820
    },
    {
      "entropy": 5.7367147445678714,
      "epoch": 4.944918618499404,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002824463405175402,
      "loss": 5.0388,
      "mean_token_accuracy": 0.1922060802578926,
      "num_tokens": 109170713.0,
      "step": 49825
    },
    {
      "entropy": 5.741523551940918,
      "epoch": 4.945414847161572,
      "grad_norm": 1.125,
      "learning_rate": 0.0002824109283380903,
      "loss": 5.0405,
      "mean_token_accuracy": 0.1813303306698799,
      "num_tokens": 109182142.0,
      "step": 49830
    },
    {
      "entropy": 5.692313528060913,
      "epoch": 4.94591107582374,
      "grad_norm": 1.25,
      "learning_rate": 0.00028237551597486534,
      "loss": 4.9541,
      "mean_token_accuracy": 0.19393115490674973,
      "num_tokens": 109194084.0,
      "step": 49835
    },
    {
      "entropy": 5.695133447647095,
      "epoch": 4.946407304485907,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002823401034287435,
      "loss": 4.9482,
      "mean_token_accuracy": 0.2015363022685051,
      "num_tokens": 109205131.0,
      "step": 49840
    },
    {
      "entropy": 5.662196683883667,
      "epoch": 4.946903533148075,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002823046907006029,
      "loss": 4.9412,
      "mean_token_accuracy": 0.1941485360264778,
      "num_tokens": 109217155.0,
      "step": 49845
    },
    {
      "entropy": 5.693146657943726,
      "epoch": 4.947399761810242,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028226927779132166,
      "loss": 4.9374,
      "mean_token_accuracy": 0.1924210950732231,
      "num_tokens": 109228411.0,
      "step": 49850
    },
    {
      "entropy": 5.650593185424805,
      "epoch": 4.94789599047241,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002822338647017781,
      "loss": 4.8303,
      "mean_token_accuracy": 0.20419699102640151,
      "num_tokens": 109240121.0,
      "step": 49855
    },
    {
      "entropy": 5.663986110687256,
      "epoch": 4.948392219134577,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028219845143285025,
      "loss": 4.8486,
      "mean_token_accuracy": 0.19995565861463546,
      "num_tokens": 109250371.0,
      "step": 49860
    },
    {
      "entropy": 5.682231473922729,
      "epoch": 4.948888447796745,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028216303798541634,
      "loss": 4.9402,
      "mean_token_accuracy": 0.19344524294137955,
      "num_tokens": 109261898.0,
      "step": 49865
    },
    {
      "entropy": 5.667163467407226,
      "epoch": 4.949384676458912,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002821276243603544,
      "loss": 5.0079,
      "mean_token_accuracy": 0.1895578235387802,
      "num_tokens": 109273180.0,
      "step": 49870
    },
    {
      "entropy": 5.763123846054077,
      "epoch": 4.94988090512108,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028209221055854283,
      "loss": 5.0348,
      "mean_token_accuracy": 0.18986617475748063,
      "num_tokens": 109284252.0,
      "step": 49875
    },
    {
      "entropy": 5.723844003677368,
      "epoch": 4.950377133783247,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002820567965808597,
      "loss": 4.9716,
      "mean_token_accuracy": 0.19463796466588973,
      "num_tokens": 109295284.0,
      "step": 49880
    },
    {
      "entropy": 5.665741729736328,
      "epoch": 4.950873362445415,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002820213824281832,
      "loss": 4.8862,
      "mean_token_accuracy": 0.20670579373836517,
      "num_tokens": 109307022.0,
      "step": 49885
    },
    {
      "entropy": 5.799218988418579,
      "epoch": 4.951369591107582,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028198596810139154,
      "loss": 5.1141,
      "mean_token_accuracy": 0.18716130554676055,
      "num_tokens": 109319382.0,
      "step": 49890
    },
    {
      "entropy": 5.652585458755493,
      "epoch": 4.95186581976975,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000281950553601363,
      "loss": 4.9447,
      "mean_token_accuracy": 0.1901189938187599,
      "num_tokens": 109329788.0,
      "step": 49895
    },
    {
      "entropy": 5.711454248428344,
      "epoch": 4.952362048431917,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002819151389289757,
      "loss": 5.0112,
      "mean_token_accuracy": 0.19380347132682801,
      "num_tokens": 109341364.0,
      "step": 49900
    },
    {
      "entropy": 5.650082588195801,
      "epoch": 4.952858277094085,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028187972408510775,
      "loss": 4.8484,
      "mean_token_accuracy": 0.2084743708372116,
      "num_tokens": 109351940.0,
      "step": 49905
    },
    {
      "entropy": 5.6642327308654785,
      "epoch": 4.953354505756252,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002818443090706376,
      "loss": 4.9314,
      "mean_token_accuracy": 0.2048933506011963,
      "num_tokens": 109362542.0,
      "step": 49910
    },
    {
      "entropy": 5.686332702636719,
      "epoch": 4.95385073441842,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002818088938864432,
      "loss": 4.8835,
      "mean_token_accuracy": 0.20095077455043792,
      "num_tokens": 109374296.0,
      "step": 49915
    },
    {
      "entropy": 5.688487195968628,
      "epoch": 4.954346963080588,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000281773478533403,
      "loss": 4.9963,
      "mean_token_accuracy": 0.1834004834294319,
      "num_tokens": 109384782.0,
      "step": 49920
    },
    {
      "entropy": 5.688100290298462,
      "epoch": 4.954843191742755,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028173806301239507,
      "loss": 4.9621,
      "mean_token_accuracy": 0.19760880172252654,
      "num_tokens": 109396210.0,
      "step": 49925
    },
    {
      "entropy": 5.687745809555054,
      "epoch": 4.955339420404923,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002817026473242978,
      "loss": 4.936,
      "mean_token_accuracy": 0.19676285684108735,
      "num_tokens": 109406452.0,
      "step": 49930
    },
    {
      "entropy": 5.758593797683716,
      "epoch": 4.95583564906709,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002816672314699891,
      "loss": 4.9964,
      "mean_token_accuracy": 0.19118715077638626,
      "num_tokens": 109417037.0,
      "step": 49935
    },
    {
      "entropy": 5.7078389644622805,
      "epoch": 4.956331877729258,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028163181545034767,
      "loss": 4.9222,
      "mean_token_accuracy": 0.20059109330177308,
      "num_tokens": 109427036.0,
      "step": 49940
    },
    {
      "entropy": 5.671416711807251,
      "epoch": 4.956828106391425,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002815963992662514,
      "loss": 4.9908,
      "mean_token_accuracy": 0.18615419715642928,
      "num_tokens": 109437119.0,
      "step": 49945
    },
    {
      "entropy": 5.731782913208008,
      "epoch": 4.957324335053593,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028156098291857865,
      "loss": 5.0612,
      "mean_token_accuracy": 0.18653801530599595,
      "num_tokens": 109448501.0,
      "step": 49950
    },
    {
      "entropy": 5.66275634765625,
      "epoch": 4.95782056371576,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002815255664082077,
      "loss": 4.9548,
      "mean_token_accuracy": 0.1969681352376938,
      "num_tokens": 109459288.0,
      "step": 49955
    },
    {
      "entropy": 5.6578878402709964,
      "epoch": 4.958316792377928,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002814901497360167,
      "loss": 4.99,
      "mean_token_accuracy": 0.18516166806221007,
      "num_tokens": 109470961.0,
      "step": 49960
    },
    {
      "entropy": 5.675966453552246,
      "epoch": 4.958813021040095,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028145473290288404,
      "loss": 4.9739,
      "mean_token_accuracy": 0.20091601461172104,
      "num_tokens": 109482147.0,
      "step": 49965
    },
    {
      "entropy": 5.703062534332275,
      "epoch": 4.959309249702263,
      "grad_norm": 1.25,
      "learning_rate": 0.0002814193159096878,
      "loss": 4.9472,
      "mean_token_accuracy": 0.20170707404613494,
      "num_tokens": 109493037.0,
      "step": 49970
    },
    {
      "entropy": 5.663480615615844,
      "epoch": 4.95980547836443,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002813838987573065,
      "loss": 4.8977,
      "mean_token_accuracy": 0.19932145327329637,
      "num_tokens": 109503458.0,
      "step": 49975
    },
    {
      "entropy": 5.694046449661255,
      "epoch": 4.960301707026598,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002813484814466182,
      "loss": 4.9157,
      "mean_token_accuracy": 0.19403479099273682,
      "num_tokens": 109514644.0,
      "step": 49980
    },
    {
      "entropy": 5.645309972763061,
      "epoch": 4.960797935688765,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028131306397850115,
      "loss": 4.8477,
      "mean_token_accuracy": 0.20490213483572006,
      "num_tokens": 109524918.0,
      "step": 49985
    },
    {
      "entropy": 5.657434558868408,
      "epoch": 4.961294164350933,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028127764635383387,
      "loss": 4.927,
      "mean_token_accuracy": 0.19563890099525452,
      "num_tokens": 109534790.0,
      "step": 49990
    },
    {
      "entropy": 5.684155511856079,
      "epoch": 4.9617903930131,
      "grad_norm": 1.25,
      "learning_rate": 0.00028124222857349435,
      "loss": 4.9789,
      "mean_token_accuracy": 0.19544149041175843,
      "num_tokens": 109546353.0,
      "step": 49995
    },
    {
      "entropy": 5.66235728263855,
      "epoch": 4.962286621675268,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028120681063836106,
      "loss": 4.9339,
      "mean_token_accuracy": 0.19176682382822036,
      "num_tokens": 109558062.0,
      "step": 50000
    },
    {
      "entropy": 5.696244764328003,
      "epoch": 4.962782850337436,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028117139254931226,
      "loss": 5.0406,
      "mean_token_accuracy": 0.18742683827877044,
      "num_tokens": 109570437.0,
      "step": 50005
    },
    {
      "entropy": 5.638661813735962,
      "epoch": 4.963279078999603,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028113597430722627,
      "loss": 4.9482,
      "mean_token_accuracy": 0.19538624882698058,
      "num_tokens": 109581825.0,
      "step": 50010
    },
    {
      "entropy": 5.712722063064575,
      "epoch": 4.963775307661771,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002811005559129813,
      "loss": 4.9484,
      "mean_token_accuracy": 0.19090123772621154,
      "num_tokens": 109593866.0,
      "step": 50015
    },
    {
      "entropy": 5.729470920562744,
      "epoch": 4.964271536323938,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002810651373674557,
      "loss": 4.9332,
      "mean_token_accuracy": 0.1971033126115799,
      "num_tokens": 109605331.0,
      "step": 50020
    },
    {
      "entropy": 5.681068181991577,
      "epoch": 4.964767764986106,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002810297186715277,
      "loss": 4.8657,
      "mean_token_accuracy": 0.20262503027915954,
      "num_tokens": 109616344.0,
      "step": 50025
    },
    {
      "entropy": 5.630282878875732,
      "epoch": 4.965263993648273,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002809942998260757,
      "loss": 4.8825,
      "mean_token_accuracy": 0.2007095456123352,
      "num_tokens": 109626506.0,
      "step": 50030
    },
    {
      "entropy": 5.700970792770386,
      "epoch": 4.965760222310441,
      "grad_norm": 1.140625,
      "learning_rate": 0.000280958880831978,
      "loss": 5.0012,
      "mean_token_accuracy": 0.19178980737924575,
      "num_tokens": 109637146.0,
      "step": 50035
    },
    {
      "entropy": 5.610707855224609,
      "epoch": 4.966256450972608,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028092346169011293,
      "loss": 4.8342,
      "mean_token_accuracy": 0.20643771141767503,
      "num_tokens": 109646680.0,
      "step": 50040
    },
    {
      "entropy": 5.623475360870361,
      "epoch": 4.966752679634776,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002808880424013587,
      "loss": 4.9832,
      "mean_token_accuracy": 0.1976648598909378,
      "num_tokens": 109656804.0,
      "step": 50045
    },
    {
      "entropy": 5.662008476257324,
      "epoch": 4.967248908296943,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028085262296659373,
      "loss": 4.9485,
      "mean_token_accuracy": 0.19809097200632095,
      "num_tokens": 109668496.0,
      "step": 50050
    },
    {
      "entropy": 5.705982398986817,
      "epoch": 4.967745136959111,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028081720338669637,
      "loss": 4.9607,
      "mean_token_accuracy": 0.1998879760503769,
      "num_tokens": 109679586.0,
      "step": 50055
    },
    {
      "entropy": 5.736908054351806,
      "epoch": 4.968241365621278,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028078178366254487,
      "loss": 4.9781,
      "mean_token_accuracy": 0.19916096180677414,
      "num_tokens": 109690576.0,
      "step": 50060
    },
    {
      "entropy": 5.697902202606201,
      "epoch": 4.9687375942834455,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028074636379501756,
      "loss": 4.9398,
      "mean_token_accuracy": 0.19354118257761002,
      "num_tokens": 109701387.0,
      "step": 50065
    },
    {
      "entropy": 5.6635724067687985,
      "epoch": 4.969233822945613,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028071094378499285,
      "loss": 4.9035,
      "mean_token_accuracy": 0.19788621068000795,
      "num_tokens": 109711565.0,
      "step": 50070
    },
    {
      "entropy": 5.704798555374145,
      "epoch": 4.969730051607781,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000280675523633349,
      "loss": 4.9176,
      "mean_token_accuracy": 0.20512582063674928,
      "num_tokens": 109722635.0,
      "step": 50075
    },
    {
      "entropy": 5.612300443649292,
      "epoch": 4.970226280269948,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028064010334096444,
      "loss": 4.8798,
      "mean_token_accuracy": 0.19398256242275239,
      "num_tokens": 109733606.0,
      "step": 50080
    },
    {
      "entropy": 5.681055021286011,
      "epoch": 4.970722508932116,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002806046829087175,
      "loss": 4.9265,
      "mean_token_accuracy": 0.1962448015809059,
      "num_tokens": 109743854.0,
      "step": 50085
    },
    {
      "entropy": 5.629793405532837,
      "epoch": 4.971218737594284,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028056926233748647,
      "loss": 4.8996,
      "mean_token_accuracy": 0.1998347356915474,
      "num_tokens": 109754646.0,
      "step": 50090
    },
    {
      "entropy": 5.685682582855224,
      "epoch": 4.971714966256451,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002805338416281497,
      "loss": 4.9144,
      "mean_token_accuracy": 0.20291344970464706,
      "num_tokens": 109766284.0,
      "step": 50095
    },
    {
      "entropy": 5.6166160106658936,
      "epoch": 4.972211194918619,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028049842078158553,
      "loss": 4.8371,
      "mean_token_accuracy": 0.198057422041893,
      "num_tokens": 109777135.0,
      "step": 50100
    },
    {
      "entropy": 5.655422163009644,
      "epoch": 4.972707423580786,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028046299979867244,
      "loss": 4.8943,
      "mean_token_accuracy": 0.1993760034441948,
      "num_tokens": 109788351.0,
      "step": 50105
    },
    {
      "entropy": 5.794500112533569,
      "epoch": 4.973203652242954,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002804275786802888,
      "loss": 5.0535,
      "mean_token_accuracy": 0.18537669479846955,
      "num_tokens": 109799013.0,
      "step": 50110
    },
    {
      "entropy": 5.729785442352295,
      "epoch": 4.973699880905121,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002803921574273128,
      "loss": 5.031,
      "mean_token_accuracy": 0.18915116786956787,
      "num_tokens": 109809807.0,
      "step": 50115
    },
    {
      "entropy": 5.6946327686309814,
      "epoch": 4.974196109567289,
      "grad_norm": 1.15625,
      "learning_rate": 0.000280356736040623,
      "loss": 5.0452,
      "mean_token_accuracy": 0.1846546024084091,
      "num_tokens": 109821750.0,
      "step": 50120
    },
    {
      "entropy": 5.730362367630005,
      "epoch": 4.974692338229456,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002803213145210975,
      "loss": 4.9815,
      "mean_token_accuracy": 0.19195117503404618,
      "num_tokens": 109833054.0,
      "step": 50125
    },
    {
      "entropy": 5.734809684753418,
      "epoch": 4.975188566891624,
      "grad_norm": 1.265625,
      "learning_rate": 0.000280285892869615,
      "loss": 4.9823,
      "mean_token_accuracy": 0.20018503516912461,
      "num_tokens": 109843375.0,
      "step": 50130
    },
    {
      "entropy": 5.687150144577027,
      "epoch": 4.975684795553791,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028025047108705387,
      "loss": 4.9229,
      "mean_token_accuracy": 0.19846314042806626,
      "num_tokens": 109854242.0,
      "step": 50135
    },
    {
      "entropy": 5.630066871643066,
      "epoch": 4.9761810242159585,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002802150491742922,
      "loss": 4.9351,
      "mean_token_accuracy": 0.18770833164453507,
      "num_tokens": 109865359.0,
      "step": 50140
    },
    {
      "entropy": 5.6868775367736815,
      "epoch": 4.976677252878126,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028017962713220855,
      "loss": 5.0286,
      "mean_token_accuracy": 0.19818443208932876,
      "num_tokens": 109876040.0,
      "step": 50145
    },
    {
      "entropy": 5.648627614974975,
      "epoch": 4.9771734815402935,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028014420496168144,
      "loss": 4.933,
      "mean_token_accuracy": 0.20242203176021575,
      "num_tokens": 109887179.0,
      "step": 50150
    },
    {
      "entropy": 5.784441614151001,
      "epoch": 4.977669710202461,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028010878266358904,
      "loss": 4.9983,
      "mean_token_accuracy": 0.19073008894920349,
      "num_tokens": 109897407.0,
      "step": 50155
    },
    {
      "entropy": 5.729750776290894,
      "epoch": 4.978165938864628,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028007336023880985,
      "loss": 4.9955,
      "mean_token_accuracy": 0.19134574085474015,
      "num_tokens": 109907270.0,
      "step": 50160
    },
    {
      "entropy": 5.740966367721557,
      "epoch": 4.978662167526796,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002800379376882223,
      "loss": 4.966,
      "mean_token_accuracy": 0.1879853665828705,
      "num_tokens": 109917430.0,
      "step": 50165
    },
    {
      "entropy": 5.691484498977661,
      "epoch": 4.979158396188964,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028000251501270464,
      "loss": 4.9127,
      "mean_token_accuracy": 0.19800102561712266,
      "num_tokens": 109927386.0,
      "step": 50170
    },
    {
      "entropy": 5.6695469379425045,
      "epoch": 4.979654624851132,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002799670922131355,
      "loss": 4.9405,
      "mean_token_accuracy": 0.19461729675531386,
      "num_tokens": 109937911.0,
      "step": 50175
    },
    {
      "entropy": 5.764958810806275,
      "epoch": 4.980150853513299,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002799316692903931,
      "loss": 5.0778,
      "mean_token_accuracy": 0.18924451768398284,
      "num_tokens": 109949049.0,
      "step": 50180
    },
    {
      "entropy": 5.739908838272095,
      "epoch": 4.980647082175467,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027989624624535594,
      "loss": 4.9609,
      "mean_token_accuracy": 0.1999809429049492,
      "num_tokens": 109962410.0,
      "step": 50185
    },
    {
      "entropy": 5.759980916976929,
      "epoch": 4.981143310837634,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002798608230789025,
      "loss": 5.0074,
      "mean_token_accuracy": 0.18636856526136397,
      "num_tokens": 109973087.0,
      "step": 50190
    },
    {
      "entropy": 5.713355922698975,
      "epoch": 4.981639539499802,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027982539979191104,
      "loss": 4.9536,
      "mean_token_accuracy": 0.1821708381175995,
      "num_tokens": 109982986.0,
      "step": 50195
    },
    {
      "entropy": 5.6913553237915036,
      "epoch": 4.982135768161969,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027978997638526004,
      "loss": 4.9571,
      "mean_token_accuracy": 0.1899561196565628,
      "num_tokens": 109993425.0,
      "step": 50200
    },
    {
      "entropy": 5.705378770828247,
      "epoch": 4.982631996824137,
      "grad_norm": 1.28125,
      "learning_rate": 0.000279754552859828,
      "loss": 4.9754,
      "mean_token_accuracy": 0.19122917354106903,
      "num_tokens": 110005319.0,
      "step": 50205
    },
    {
      "entropy": 5.670808124542236,
      "epoch": 4.983128225486304,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002797191292164933,
      "loss": 4.9142,
      "mean_token_accuracy": 0.20284205228090285,
      "num_tokens": 110016001.0,
      "step": 50210
    },
    {
      "entropy": 5.696332788467407,
      "epoch": 4.9836244541484715,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002796837054561343,
      "loss": 4.9048,
      "mean_token_accuracy": 0.20105994194746019,
      "num_tokens": 110027326.0,
      "step": 50215
    },
    {
      "entropy": 5.666791391372681,
      "epoch": 4.984120682810639,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027964828157962964,
      "loss": 4.8796,
      "mean_token_accuracy": 0.19647541046142578,
      "num_tokens": 110038155.0,
      "step": 50220
    },
    {
      "entropy": 5.66616907119751,
      "epoch": 4.9846169114728065,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002796128575878575,
      "loss": 4.9484,
      "mean_token_accuracy": 0.19578845649957657,
      "num_tokens": 110049182.0,
      "step": 50225
    },
    {
      "entropy": 5.742909049987793,
      "epoch": 4.985113140134974,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002795774334816964,
      "loss": 5.0724,
      "mean_token_accuracy": 0.18524757772684097,
      "num_tokens": 110060780.0,
      "step": 50230
    },
    {
      "entropy": 5.719228839874267,
      "epoch": 4.9856093687971414,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027954200926202493,
      "loss": 5.0066,
      "mean_token_accuracy": 0.1883491948246956,
      "num_tokens": 110071957.0,
      "step": 50235
    },
    {
      "entropy": 5.758763694763184,
      "epoch": 4.986105597459309,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002795065849297213,
      "loss": 5.0499,
      "mean_token_accuracy": 0.18769701719284057,
      "num_tokens": 110083264.0,
      "step": 50240
    },
    {
      "entropy": 5.728816747665405,
      "epoch": 4.986601826121476,
      "grad_norm": 1.25,
      "learning_rate": 0.00027947116048566414,
      "loss": 4.8915,
      "mean_token_accuracy": 0.20174896270036696,
      "num_tokens": 110094754.0,
      "step": 50245
    },
    {
      "entropy": 5.741538715362549,
      "epoch": 4.987098054783644,
      "grad_norm": 1.46875,
      "learning_rate": 0.00027943573593073185,
      "loss": 4.9772,
      "mean_token_accuracy": 0.2009858399629593,
      "num_tokens": 110105824.0,
      "step": 50250
    },
    {
      "entropy": 5.65765495300293,
      "epoch": 4.987594283445812,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027940031126580286,
      "loss": 4.8677,
      "mean_token_accuracy": 0.20255733877420426,
      "num_tokens": 110115264.0,
      "step": 50255
    },
    {
      "entropy": 5.6627357482910154,
      "epoch": 4.98809051210798,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002793648864917556,
      "loss": 4.9911,
      "mean_token_accuracy": 0.195909284055233,
      "num_tokens": 110126585.0,
      "step": 50260
    },
    {
      "entropy": 5.703181982040405,
      "epoch": 4.988586740770147,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002793294616094685,
      "loss": 4.9358,
      "mean_token_accuracy": 0.2031191796064377,
      "num_tokens": 110136609.0,
      "step": 50265
    },
    {
      "entropy": 5.712695646286011,
      "epoch": 4.989082969432315,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027929403661982016,
      "loss": 5.0188,
      "mean_token_accuracy": 0.1887998551130295,
      "num_tokens": 110146743.0,
      "step": 50270
    },
    {
      "entropy": 5.73708143234253,
      "epoch": 4.989579198094482,
      "grad_norm": 1.140625,
      "learning_rate": 0.00027925861152368906,
      "loss": 4.9452,
      "mean_token_accuracy": 0.19782434701919555,
      "num_tokens": 110159124.0,
      "step": 50275
    },
    {
      "entropy": 5.70025405883789,
      "epoch": 4.99007542675665,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002792231863219534,
      "loss": 4.9607,
      "mean_token_accuracy": 0.19679402858018874,
      "num_tokens": 110169921.0,
      "step": 50280
    },
    {
      "entropy": 5.685193061828613,
      "epoch": 4.990571655418817,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027918776101549186,
      "loss": 4.9497,
      "mean_token_accuracy": 0.19767000079154967,
      "num_tokens": 110180459.0,
      "step": 50285
    },
    {
      "entropy": 5.7313121318817135,
      "epoch": 4.991067884080985,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000279152335605183,
      "loss": 5.0488,
      "mean_token_accuracy": 0.1858829602599144,
      "num_tokens": 110190810.0,
      "step": 50290
    },
    {
      "entropy": 5.704129600524903,
      "epoch": 4.991564112743152,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027911691009190504,
      "loss": 4.982,
      "mean_token_accuracy": 0.19070690274238586,
      "num_tokens": 110201457.0,
      "step": 50295
    },
    {
      "entropy": 5.770195960998535,
      "epoch": 4.9920603414053195,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00027908148447653656,
      "loss": 5.0831,
      "mean_token_accuracy": 0.18819876611232758,
      "num_tokens": 110210492.0,
      "step": 50300
    },
    {
      "entropy": 5.66724214553833,
      "epoch": 4.992556570067487,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027904605875995613,
      "loss": 4.9176,
      "mean_token_accuracy": 0.1985630214214325,
      "num_tokens": 110221453.0,
      "step": 50305
    },
    {
      "entropy": 5.720960283279419,
      "epoch": 4.9930527987296545,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002790106329430421,
      "loss": 4.9799,
      "mean_token_accuracy": 0.19602624773979188,
      "num_tokens": 110232075.0,
      "step": 50310
    },
    {
      "entropy": 5.628142595291138,
      "epoch": 4.993549027391822,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002789752070266731,
      "loss": 4.885,
      "mean_token_accuracy": 0.20140552073717116,
      "num_tokens": 110242009.0,
      "step": 50315
    },
    {
      "entropy": 5.716321039199829,
      "epoch": 4.994045256053989,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002789397810117275,
      "loss": 5.0305,
      "mean_token_accuracy": 0.1874823287129402,
      "num_tokens": 110252552.0,
      "step": 50320
    },
    {
      "entropy": 5.724645805358887,
      "epoch": 4.994541484716157,
      "grad_norm": 1.25,
      "learning_rate": 0.00027890435489908377,
      "loss": 4.9876,
      "mean_token_accuracy": 0.1970227465033531,
      "num_tokens": 110262619.0,
      "step": 50325
    },
    {
      "entropy": 5.776113224029541,
      "epoch": 4.995037713378324,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027886892868962047,
      "loss": 5.0074,
      "mean_token_accuracy": 0.1933481678366661,
      "num_tokens": 110273169.0,
      "step": 50330
    },
    {
      "entropy": 5.712438488006592,
      "epoch": 4.995533942040492,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027883350238421613,
      "loss": 4.9837,
      "mean_token_accuracy": 0.19105785787105561,
      "num_tokens": 110283069.0,
      "step": 50335
    },
    {
      "entropy": 5.665274715423584,
      "epoch": 4.99603017070266,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002787980759837491,
      "loss": 4.8874,
      "mean_token_accuracy": 0.19923257678747178,
      "num_tokens": 110293891.0,
      "step": 50340
    },
    {
      "entropy": 5.761516666412353,
      "epoch": 4.996526399364828,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027876264948909803,
      "loss": 5.0483,
      "mean_token_accuracy": 0.187675903737545,
      "num_tokens": 110305437.0,
      "step": 50345
    },
    {
      "entropy": 5.712433671951294,
      "epoch": 4.997022628026995,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002787272229011414,
      "loss": 4.9629,
      "mean_token_accuracy": 0.20122834295034409,
      "num_tokens": 110316545.0,
      "step": 50350
    },
    {
      "entropy": 5.725979566574097,
      "epoch": 4.997518856689163,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027869179622075764,
      "loss": 4.945,
      "mean_token_accuracy": 0.20102826058864592,
      "num_tokens": 110327642.0,
      "step": 50355
    },
    {
      "entropy": 5.669158601760865,
      "epoch": 4.99801508535133,
      "grad_norm": 1.25,
      "learning_rate": 0.0002786563694488252,
      "loss": 4.9063,
      "mean_token_accuracy": 0.19576302021741868,
      "num_tokens": 110338971.0,
      "step": 50360
    },
    {
      "entropy": 5.673297309875489,
      "epoch": 4.998511314013498,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027862094258622276,
      "loss": 4.9601,
      "mean_token_accuracy": 0.1967178165912628,
      "num_tokens": 110349706.0,
      "step": 50365
    },
    {
      "entropy": 5.7163403034210205,
      "epoch": 4.999007542675665,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002785855156338288,
      "loss": 4.9647,
      "mean_token_accuracy": 0.19729155898094178,
      "num_tokens": 110360113.0,
      "step": 50370
    },
    {
      "entropy": 5.715429449081421,
      "epoch": 4.9995037713378325,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002785500885925217,
      "loss": 4.933,
      "mean_token_accuracy": 0.19773049652576447,
      "num_tokens": 110370009.0,
      "step": 50375
    },
    {
      "entropy": 5.7254828929901125,
      "epoch": 5.0,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00027851466146318007,
      "loss": 4.9667,
      "mean_token_accuracy": 0.1850172758102417,
      "num_tokens": 110379265.0,
      "step": 50380
    },
    {
      "entropy": 5.695610046386719,
      "epoch": 5.0004962286621675,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002784792342466824,
      "loss": 4.952,
      "mean_token_accuracy": 0.19836524724960328,
      "num_tokens": 110389940.0,
      "step": 50385
    },
    {
      "entropy": 5.637142086029053,
      "epoch": 5.000992457324335,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002784438069439073,
      "loss": 4.8593,
      "mean_token_accuracy": 0.19651132822036743,
      "num_tokens": 110400853.0,
      "step": 50390
    },
    {
      "entropy": 5.701901578903199,
      "epoch": 5.001488685986502,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027840837955573304,
      "loss": 4.9398,
      "mean_token_accuracy": 0.19855444580316545,
      "num_tokens": 110411939.0,
      "step": 50395
    },
    {
      "entropy": 5.732490301132202,
      "epoch": 5.00198491464867,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002783729520830384,
      "loss": 4.9661,
      "mean_token_accuracy": 0.19618093967437744,
      "num_tokens": 110423264.0,
      "step": 50400
    },
    {
      "entropy": 5.678282165527344,
      "epoch": 5.002481143310837,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027833752452670183,
      "loss": 4.9473,
      "mean_token_accuracy": 0.19753186106681825,
      "num_tokens": 110433663.0,
      "step": 50405
    },
    {
      "entropy": 5.7394920825958256,
      "epoch": 5.002977371973005,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027830209688760186,
      "loss": 5.0431,
      "mean_token_accuracy": 0.18546697944402696,
      "num_tokens": 110445938.0,
      "step": 50410
    },
    {
      "entropy": 5.769263458251953,
      "epoch": 5.003473600635172,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027826666916661697,
      "loss": 5.0074,
      "mean_token_accuracy": 0.18942777961492538,
      "num_tokens": 110457358.0,
      "step": 50415
    },
    {
      "entropy": 5.752118301391602,
      "epoch": 5.00396982929734,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027823124136462575,
      "loss": 4.9077,
      "mean_token_accuracy": 0.20505875945091248,
      "num_tokens": 110468442.0,
      "step": 50420
    },
    {
      "entropy": 5.712847185134888,
      "epoch": 5.004466057959508,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002781958134825067,
      "loss": 4.9407,
      "mean_token_accuracy": 0.19019874334335327,
      "num_tokens": 110478727.0,
      "step": 50425
    },
    {
      "entropy": 5.725963497161866,
      "epoch": 5.004962286621676,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00027816038552113835,
      "loss": 5.0356,
      "mean_token_accuracy": 0.19367011189460753,
      "num_tokens": 110488675.0,
      "step": 50430
    },
    {
      "entropy": 5.676634311676025,
      "epoch": 5.005458515283843,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002781249574813992,
      "loss": 4.9001,
      "mean_token_accuracy": 0.1998487278819084,
      "num_tokens": 110499338.0,
      "step": 50435
    },
    {
      "entropy": 5.7012067317962645,
      "epoch": 5.005954743946011,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027808952936416787,
      "loss": 4.914,
      "mean_token_accuracy": 0.20165184736251832,
      "num_tokens": 110508602.0,
      "step": 50440
    },
    {
      "entropy": 5.637464237213135,
      "epoch": 5.006450972608178,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027805410117032295,
      "loss": 4.8554,
      "mean_token_accuracy": 0.20364387780427934,
      "num_tokens": 110518120.0,
      "step": 50445
    },
    {
      "entropy": 5.640688848495484,
      "epoch": 5.0069472012703455,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027801867290074277,
      "loss": 4.8921,
      "mean_token_accuracy": 0.1950184315443039,
      "num_tokens": 110528892.0,
      "step": 50450
    },
    {
      "entropy": 5.717186689376831,
      "epoch": 5.007443429932513,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00027798324455630607,
      "loss": 5.0044,
      "mean_token_accuracy": 0.19278548210859298,
      "num_tokens": 110540088.0,
      "step": 50455
    },
    {
      "entropy": 5.604202222824097,
      "epoch": 5.0079396585946805,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027794781613789133,
      "loss": 4.7992,
      "mean_token_accuracy": 0.20848796665668487,
      "num_tokens": 110550563.0,
      "step": 50460
    },
    {
      "entropy": 5.6695631980896,
      "epoch": 5.008435887256848,
      "grad_norm": 1.25,
      "learning_rate": 0.0002779123876463772,
      "loss": 4.8483,
      "mean_token_accuracy": 0.19974968880414962,
      "num_tokens": 110562508.0,
      "step": 50465
    },
    {
      "entropy": 5.7594586372375485,
      "epoch": 5.008932115919015,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027787695908264206,
      "loss": 4.9918,
      "mean_token_accuracy": 0.1959464967250824,
      "num_tokens": 110575458.0,
      "step": 50470
    },
    {
      "entropy": 5.782322978973388,
      "epoch": 5.009428344581183,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002778415304475644,
      "loss": 4.9528,
      "mean_token_accuracy": 0.19888652116060257,
      "num_tokens": 110585887.0,
      "step": 50475
    },
    {
      "entropy": 5.689559364318848,
      "epoch": 5.00992457324335,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002778061017420231,
      "loss": 4.9868,
      "mean_token_accuracy": 0.18998484015464784,
      "num_tokens": 110596873.0,
      "step": 50480
    },
    {
      "entropy": 5.693735027313233,
      "epoch": 5.010420801905518,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027777067296689645,
      "loss": 5.0111,
      "mean_token_accuracy": 0.1879079148173332,
      "num_tokens": 110608183.0,
      "step": 50485
    },
    {
      "entropy": 5.745501375198364,
      "epoch": 5.010917030567685,
      "grad_norm": 1.359375,
      "learning_rate": 0.000277735244123063,
      "loss": 4.9352,
      "mean_token_accuracy": 0.19880762845277786,
      "num_tokens": 110618255.0,
      "step": 50490
    },
    {
      "entropy": 5.77590765953064,
      "epoch": 5.011413259229853,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002776998152114015,
      "loss": 4.9242,
      "mean_token_accuracy": 0.20181539058685302,
      "num_tokens": 110628515.0,
      "step": 50495
    },
    {
      "entropy": 5.690201091766357,
      "epoch": 5.01190948789202,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027766438623279044,
      "loss": 4.9572,
      "mean_token_accuracy": 0.19949508160352708,
      "num_tokens": 110640493.0,
      "step": 50500
    },
    {
      "entropy": 5.68048849105835,
      "epoch": 5.012405716554188,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002776289571881082,
      "loss": 4.9306,
      "mean_token_accuracy": 0.1994481712579727,
      "num_tokens": 110652266.0,
      "step": 50505
    },
    {
      "entropy": 5.744650030136109,
      "epoch": 5.012901945216356,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027759352807823356,
      "loss": 4.9888,
      "mean_token_accuracy": 0.19597553908824922,
      "num_tokens": 110662228.0,
      "step": 50510
    },
    {
      "entropy": 5.743806409835815,
      "epoch": 5.013398173878524,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000277558098904045,
      "loss": 4.9299,
      "mean_token_accuracy": 0.19762465506792068,
      "num_tokens": 110673449.0,
      "step": 50515
    },
    {
      "entropy": 5.729211854934692,
      "epoch": 5.013894402540691,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027752266966642116,
      "loss": 4.9373,
      "mean_token_accuracy": 0.19276179671287536,
      "num_tokens": 110685368.0,
      "step": 50520
    },
    {
      "entropy": 5.703085613250733,
      "epoch": 5.0143906312028586,
      "grad_norm": 1.25,
      "learning_rate": 0.00027748724036624046,
      "loss": 4.9694,
      "mean_token_accuracy": 0.19430783092975618,
      "num_tokens": 110696094.0,
      "step": 50525
    },
    {
      "entropy": 5.736684846878052,
      "epoch": 5.014886859865026,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027745181100438166,
      "loss": 4.933,
      "mean_token_accuracy": 0.19860538393259047,
      "num_tokens": 110708851.0,
      "step": 50530
    },
    {
      "entropy": 5.649028396606445,
      "epoch": 5.0153830885271935,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002774163815817231,
      "loss": 4.9227,
      "mean_token_accuracy": 0.18994751423597336,
      "num_tokens": 110720373.0,
      "step": 50535
    },
    {
      "entropy": 5.686967611312866,
      "epoch": 5.015879317189361,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027738095209914353,
      "loss": 4.9496,
      "mean_token_accuracy": 0.1925160691142082,
      "num_tokens": 110731701.0,
      "step": 50540
    },
    {
      "entropy": 5.656658506393432,
      "epoch": 5.0163755458515285,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027734552255752153,
      "loss": 4.8791,
      "mean_token_accuracy": 0.20552184283733368,
      "num_tokens": 110743588.0,
      "step": 50545
    },
    {
      "entropy": 5.691792964935303,
      "epoch": 5.016871774513696,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027731009295773565,
      "loss": 4.9764,
      "mean_token_accuracy": 0.19823637157678603,
      "num_tokens": 110753184.0,
      "step": 50550
    },
    {
      "entropy": 5.666642427444458,
      "epoch": 5.017368003175863,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027727466330066433,
      "loss": 4.9368,
      "mean_token_accuracy": 0.19455137252807617,
      "num_tokens": 110764191.0,
      "step": 50555
    },
    {
      "entropy": 5.710934066772461,
      "epoch": 5.017864231838031,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027723923358718636,
      "loss": 4.9536,
      "mean_token_accuracy": 0.19497276991605758,
      "num_tokens": 110775177.0,
      "step": 50560
    },
    {
      "entropy": 5.695841360092163,
      "epoch": 5.018360460500198,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002772038038181801,
      "loss": 4.9388,
      "mean_token_accuracy": 0.19567052870988846,
      "num_tokens": 110785861.0,
      "step": 50565
    },
    {
      "entropy": 5.702959203720093,
      "epoch": 5.018856689162366,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027716837399452437,
      "loss": 4.9959,
      "mean_token_accuracy": 0.18540901839733123,
      "num_tokens": 110797625.0,
      "step": 50570
    },
    {
      "entropy": 5.618745136260986,
      "epoch": 5.019352917824533,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027713294411709763,
      "loss": 4.8658,
      "mean_token_accuracy": 0.2047804370522499,
      "num_tokens": 110809123.0,
      "step": 50575
    },
    {
      "entropy": 5.6589813232421875,
      "epoch": 5.019849146486701,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002770975141867784,
      "loss": 4.9389,
      "mean_token_accuracy": 0.1981199190020561,
      "num_tokens": 110820537.0,
      "step": 50580
    },
    {
      "entropy": 5.711798858642578,
      "epoch": 5.020345375148868,
      "grad_norm": 1.25,
      "learning_rate": 0.0002770620842044454,
      "loss": 4.9083,
      "mean_token_accuracy": 0.19145447313785552,
      "num_tokens": 110831907.0,
      "step": 50585
    },
    {
      "entropy": 5.700283622741699,
      "epoch": 5.020841603811036,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002770266541709771,
      "loss": 4.8502,
      "mean_token_accuracy": 0.19926880449056625,
      "num_tokens": 110843107.0,
      "step": 50590
    },
    {
      "entropy": 5.649003505706787,
      "epoch": 5.021337832473204,
      "grad_norm": 1.265625,
      "learning_rate": 0.00027699122408725226,
      "loss": 4.9411,
      "mean_token_accuracy": 0.19180062413215637,
      "num_tokens": 110854949.0,
      "step": 50595
    },
    {
      "entropy": 5.729410791397095,
      "epoch": 5.021834061135372,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002769557939541493,
      "loss": 5.0138,
      "mean_token_accuracy": 0.19944973140954972,
      "num_tokens": 110866295.0,
      "step": 50600
    },
    {
      "entropy": 5.663148593902588,
      "epoch": 5.022330289797539,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002769203637725468,
      "loss": 4.8617,
      "mean_token_accuracy": 0.2018674999475479,
      "num_tokens": 110876531.0,
      "step": 50605
    },
    {
      "entropy": 5.752137851715088,
      "epoch": 5.0228265184597065,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002768849335433234,
      "loss": 4.9557,
      "mean_token_accuracy": 0.1921977296471596,
      "num_tokens": 110887605.0,
      "step": 50610
    },
    {
      "entropy": 5.639356994628907,
      "epoch": 5.023322747121874,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027684950326735777,
      "loss": 4.8883,
      "mean_token_accuracy": 0.19965240806341172,
      "num_tokens": 110898939.0,
      "step": 50615
    },
    {
      "entropy": 5.629901456832886,
      "epoch": 5.0238189757840415,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002768140729455284,
      "loss": 4.8246,
      "mean_token_accuracy": 0.20276155173778534,
      "num_tokens": 110908882.0,
      "step": 50620
    },
    {
      "entropy": 5.708892440795898,
      "epoch": 5.024315204446209,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027677864257871396,
      "loss": 4.9805,
      "mean_token_accuracy": 0.19511889070272445,
      "num_tokens": 110921471.0,
      "step": 50625
    },
    {
      "entropy": 5.706790399551392,
      "epoch": 5.024811433108376,
      "grad_norm": 1.25,
      "learning_rate": 0.0002767432121677931,
      "loss": 4.8961,
      "mean_token_accuracy": 0.19349356442689897,
      "num_tokens": 110931630.0,
      "step": 50630
    },
    {
      "entropy": 5.677390766143799,
      "epoch": 5.025307661770544,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027670778171364423,
      "loss": 4.9217,
      "mean_token_accuracy": 0.20027055144309996,
      "num_tokens": 110942568.0,
      "step": 50635
    },
    {
      "entropy": 5.669660902023315,
      "epoch": 5.025803890432711,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002766723512171461,
      "loss": 4.9395,
      "mean_token_accuracy": 0.20004477053880693,
      "num_tokens": 110953390.0,
      "step": 50640
    },
    {
      "entropy": 5.726735639572143,
      "epoch": 5.026300119094879,
      "grad_norm": 1.265625,
      "learning_rate": 0.00027663692067917724,
      "loss": 4.9507,
      "mean_token_accuracy": 0.20251129120588302,
      "num_tokens": 110964602.0,
      "step": 50645
    },
    {
      "entropy": 5.699330806732178,
      "epoch": 5.026796347757046,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002766014901006163,
      "loss": 4.9133,
      "mean_token_accuracy": 0.19895305782556533,
      "num_tokens": 110974119.0,
      "step": 50650
    },
    {
      "entropy": 5.64515175819397,
      "epoch": 5.027292576419214,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027656605948234175,
      "loss": 4.9488,
      "mean_token_accuracy": 0.1927759602665901,
      "num_tokens": 110984903.0,
      "step": 50655
    },
    {
      "entropy": 5.637283086776733,
      "epoch": 5.027788805081381,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027653062882523243,
      "loss": 4.8954,
      "mean_token_accuracy": 0.20592667907476425,
      "num_tokens": 110996597.0,
      "step": 50660
    },
    {
      "entropy": 5.725716066360474,
      "epoch": 5.028285033743549,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002764951981301668,
      "loss": 4.9405,
      "mean_token_accuracy": 0.19746332466602326,
      "num_tokens": 111007364.0,
      "step": 50665
    },
    {
      "entropy": 5.686441087722779,
      "epoch": 5.028781262405716,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027645976739802344,
      "loss": 4.8967,
      "mean_token_accuracy": 0.19692958742380143,
      "num_tokens": 111018628.0,
      "step": 50670
    },
    {
      "entropy": 5.7326452255249025,
      "epoch": 5.029277491067884,
      "grad_norm": 1.21875,
      "learning_rate": 0.000276424336629681,
      "loss": 4.9969,
      "mean_token_accuracy": 0.19616898596286775,
      "num_tokens": 111029809.0,
      "step": 50675
    },
    {
      "entropy": 5.672727060317993,
      "epoch": 5.029773719730052,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002763889058260181,
      "loss": 4.9377,
      "mean_token_accuracy": 0.19741588830947876,
      "num_tokens": 111040292.0,
      "step": 50680
    },
    {
      "entropy": 5.7164243221282955,
      "epoch": 5.0302699483922195,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002763534749879133,
      "loss": 4.8865,
      "mean_token_accuracy": 0.2052363336086273,
      "num_tokens": 111051880.0,
      "step": 50685
    },
    {
      "entropy": 5.686276865005493,
      "epoch": 5.030766177054387,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002763180441162452,
      "loss": 4.9223,
      "mean_token_accuracy": 0.20388993471860886,
      "num_tokens": 111063084.0,
      "step": 50690
    },
    {
      "entropy": 5.736810541152954,
      "epoch": 5.0312624057165545,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027628261321189254,
      "loss": 4.97,
      "mean_token_accuracy": 0.19626391679048538,
      "num_tokens": 111072575.0,
      "step": 50695
    },
    {
      "entropy": 5.7165216445922855,
      "epoch": 5.031758634378722,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002762471822757338,
      "loss": 4.9805,
      "mean_token_accuracy": 0.19612426459789276,
      "num_tokens": 111083775.0,
      "step": 50700
    },
    {
      "entropy": 5.7262835025787355,
      "epoch": 5.032254863040889,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002762117513086476,
      "loss": 5.0285,
      "mean_token_accuracy": 0.1840009570121765,
      "num_tokens": 111095914.0,
      "step": 50705
    },
    {
      "entropy": 5.731715488433838,
      "epoch": 5.032751091703057,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027617632031151255,
      "loss": 4.8819,
      "mean_token_accuracy": 0.20037350207567214,
      "num_tokens": 111105952.0,
      "step": 50710
    },
    {
      "entropy": 5.71926908493042,
      "epoch": 5.033247320365224,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027614088928520734,
      "loss": 4.9995,
      "mean_token_accuracy": 0.18728528320789337,
      "num_tokens": 111117280.0,
      "step": 50715
    },
    {
      "entropy": 5.693426275253296,
      "epoch": 5.033743549027392,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002761054582306106,
      "loss": 4.9436,
      "mean_token_accuracy": 0.18794937580823898,
      "num_tokens": 111128483.0,
      "step": 50720
    },
    {
      "entropy": 5.743969869613648,
      "epoch": 5.034239777689559,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027607002714860075,
      "loss": 4.9284,
      "mean_token_accuracy": 0.19531276673078538,
      "num_tokens": 111140451.0,
      "step": 50725
    },
    {
      "entropy": 5.654486417770386,
      "epoch": 5.034736006351727,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002760345960400566,
      "loss": 4.8952,
      "mean_token_accuracy": 0.1997361734509468,
      "num_tokens": 111151575.0,
      "step": 50730
    },
    {
      "entropy": 5.6831197261810305,
      "epoch": 5.035232235013894,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002759991649058567,
      "loss": 4.8884,
      "mean_token_accuracy": 0.19884148687124253,
      "num_tokens": 111161007.0,
      "step": 50735
    },
    {
      "entropy": 5.648079776763916,
      "epoch": 5.035728463676062,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002759637337468796,
      "loss": 4.9147,
      "mean_token_accuracy": 0.1936602771282196,
      "num_tokens": 111173249.0,
      "step": 50740
    },
    {
      "entropy": 5.681692981719971,
      "epoch": 5.036224692338229,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027592830256400406,
      "loss": 4.9332,
      "mean_token_accuracy": 0.1975514829158783,
      "num_tokens": 111184165.0,
      "step": 50745
    },
    {
      "entropy": 5.692002820968628,
      "epoch": 5.036720921000397,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027589287135810857,
      "loss": 4.8707,
      "mean_token_accuracy": 0.20337789505720139,
      "num_tokens": 111196116.0,
      "step": 50750
    },
    {
      "entropy": 5.691051006317139,
      "epoch": 5.037217149662564,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002758574401300718,
      "loss": 4.8885,
      "mean_token_accuracy": 0.1993911698460579,
      "num_tokens": 111206913.0,
      "step": 50755
    },
    {
      "entropy": 5.688117599487304,
      "epoch": 5.037713378324732,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002758220088807724,
      "loss": 4.9761,
      "mean_token_accuracy": 0.19950492978096007,
      "num_tokens": 111218896.0,
      "step": 50760
    },
    {
      "entropy": 5.697272205352784,
      "epoch": 5.038209606986899,
      "grad_norm": 1.3125,
      "learning_rate": 0.000275786577611089,
      "loss": 4.955,
      "mean_token_accuracy": 0.19919278025627135,
      "num_tokens": 111229392.0,
      "step": 50765
    },
    {
      "entropy": 5.71423225402832,
      "epoch": 5.0387058356490675,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027575114632190004,
      "loss": 4.9309,
      "mean_token_accuracy": 0.19450290501117706,
      "num_tokens": 111240102.0,
      "step": 50770
    },
    {
      "entropy": 5.677582740783691,
      "epoch": 5.039202064311235,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027571571501408434,
      "loss": 4.9172,
      "mean_token_accuracy": 0.19730714857578277,
      "num_tokens": 111251003.0,
      "step": 50775
    },
    {
      "entropy": 5.675706386566162,
      "epoch": 5.039698292973402,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002756802836885204,
      "loss": 4.9052,
      "mean_token_accuracy": 0.20352840125560762,
      "num_tokens": 111261461.0,
      "step": 50780
    },
    {
      "entropy": 5.682121658325196,
      "epoch": 5.04019452163557,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002756448523460869,
      "loss": 4.9014,
      "mean_token_accuracy": 0.20193222165107727,
      "num_tokens": 111272796.0,
      "step": 50785
    },
    {
      "entropy": 5.729210042953492,
      "epoch": 5.040690750297737,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027560942098766255,
      "loss": 4.9297,
      "mean_token_accuracy": 0.19743713289499282,
      "num_tokens": 111283413.0,
      "step": 50790
    },
    {
      "entropy": 5.633551073074341,
      "epoch": 5.041186978959905,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002755739896141258,
      "loss": 4.824,
      "mean_token_accuracy": 0.2110162302851677,
      "num_tokens": 111294297.0,
      "step": 50795
    },
    {
      "entropy": 5.664704847335815,
      "epoch": 5.041683207622072,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002755385582263553,
      "loss": 4.8951,
      "mean_token_accuracy": 0.19329319447278975,
      "num_tokens": 111304922.0,
      "step": 50800
    },
    {
      "entropy": 5.719072246551514,
      "epoch": 5.04217943628424,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027550312682522976,
      "loss": 4.9707,
      "mean_token_accuracy": 0.19287531673908234,
      "num_tokens": 111315761.0,
      "step": 50805
    },
    {
      "entropy": 5.750563049316407,
      "epoch": 5.042675664946407,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002754676954116278,
      "loss": 5.0352,
      "mean_token_accuracy": 0.1955152213573456,
      "num_tokens": 111327753.0,
      "step": 50810
    },
    {
      "entropy": 5.675919675827027,
      "epoch": 5.043171893608575,
      "grad_norm": 1.375,
      "learning_rate": 0.000275432263986428,
      "loss": 4.971,
      "mean_token_accuracy": 0.1873941019177437,
      "num_tokens": 111339741.0,
      "step": 50815
    },
    {
      "entropy": 5.685288619995117,
      "epoch": 5.043668122270742,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002753968325505089,
      "loss": 4.9541,
      "mean_token_accuracy": 0.1995793715119362,
      "num_tokens": 111350898.0,
      "step": 50820
    },
    {
      "entropy": 5.727639484405517,
      "epoch": 5.04416435093291,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002753614011047492,
      "loss": 5.0347,
      "mean_token_accuracy": 0.19538000971078873,
      "num_tokens": 111363227.0,
      "step": 50825
    },
    {
      "entropy": 5.731527805328369,
      "epoch": 5.044660579595077,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027532596965002757,
      "loss": 4.9299,
      "mean_token_accuracy": 0.19842186123132705,
      "num_tokens": 111374229.0,
      "step": 50830
    },
    {
      "entropy": 5.6967713832855225,
      "epoch": 5.045156808257245,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002752905381872227,
      "loss": 4.9331,
      "mean_token_accuracy": 0.19851492792367936,
      "num_tokens": 111385628.0,
      "step": 50835
    },
    {
      "entropy": 5.702032518386841,
      "epoch": 5.045653036919412,
      "grad_norm": 1.328125,
      "learning_rate": 0.000275255106717213,
      "loss": 4.9131,
      "mean_token_accuracy": 0.19504204541444778,
      "num_tokens": 111397921.0,
      "step": 50840
    },
    {
      "entropy": 5.692978382110596,
      "epoch": 5.04614926558158,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002752196752408772,
      "loss": 4.89,
      "mean_token_accuracy": 0.20228450298309325,
      "num_tokens": 111408518.0,
      "step": 50845
    },
    {
      "entropy": 5.736821842193604,
      "epoch": 5.046645494243747,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027518424375909395,
      "loss": 4.9925,
      "mean_token_accuracy": 0.1865009695291519,
      "num_tokens": 111419748.0,
      "step": 50850
    },
    {
      "entropy": 5.675156831741333,
      "epoch": 5.0471417229059155,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002751488122727419,
      "loss": 4.8835,
      "mean_token_accuracy": 0.20286983251571655,
      "num_tokens": 111429654.0,
      "step": 50855
    },
    {
      "entropy": 5.68634991645813,
      "epoch": 5.047637951568083,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027511338078269954,
      "loss": 4.9077,
      "mean_token_accuracy": 0.19800308346748352,
      "num_tokens": 111440985.0,
      "step": 50860
    },
    {
      "entropy": 5.677521228790283,
      "epoch": 5.04813418023025,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002750779492898456,
      "loss": 4.893,
      "mean_token_accuracy": 0.19669591188430785,
      "num_tokens": 111451569.0,
      "step": 50865
    },
    {
      "entropy": 5.656940078735351,
      "epoch": 5.048630408892418,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027504251779505876,
      "loss": 4.8856,
      "mean_token_accuracy": 0.2081151068210602,
      "num_tokens": 111461646.0,
      "step": 50870
    },
    {
      "entropy": 5.705598402023315,
      "epoch": 5.049126637554585,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002750070862992175,
      "loss": 4.9409,
      "mean_token_accuracy": 0.1989201545715332,
      "num_tokens": 111471824.0,
      "step": 50875
    },
    {
      "entropy": 5.745653438568115,
      "epoch": 5.049622866216753,
      "grad_norm": 1.375,
      "learning_rate": 0.0002749716548032005,
      "loss": 5.0051,
      "mean_token_accuracy": 0.19470532536506652,
      "num_tokens": 111482446.0,
      "step": 50880
    },
    {
      "entropy": 5.693439865112305,
      "epoch": 5.05011909487892,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027493622330788644,
      "loss": 4.9205,
      "mean_token_accuracy": 0.19399068653583526,
      "num_tokens": 111492914.0,
      "step": 50885
    },
    {
      "entropy": 5.708818435668945,
      "epoch": 5.050615323541088,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002749007918141539,
      "loss": 4.8944,
      "mean_token_accuracy": 0.19976255893707276,
      "num_tokens": 111503973.0,
      "step": 50890
    },
    {
      "entropy": 5.733844566345215,
      "epoch": 5.051111552203255,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002748653603228815,
      "loss": 5.0329,
      "mean_token_accuracy": 0.1897711396217346,
      "num_tokens": 111514508.0,
      "step": 50895
    },
    {
      "entropy": 5.783145761489868,
      "epoch": 5.051607780865423,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002748299288349479,
      "loss": 4.9951,
      "mean_token_accuracy": 0.1991789773106575,
      "num_tokens": 111525524.0,
      "step": 50900
    },
    {
      "entropy": 5.755983018875122,
      "epoch": 5.05210400952759,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002747944973512317,
      "loss": 5.0074,
      "mean_token_accuracy": 0.18605227768421173,
      "num_tokens": 111536303.0,
      "step": 50905
    },
    {
      "entropy": 5.734213781356812,
      "epoch": 5.052600238189758,
      "grad_norm": 1.25,
      "learning_rate": 0.0002747590658726114,
      "loss": 4.9501,
      "mean_token_accuracy": 0.19988067001104354,
      "num_tokens": 111546947.0,
      "step": 50910
    },
    {
      "entropy": 5.739991760253906,
      "epoch": 5.053096466851925,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002747236343999659,
      "loss": 5.0173,
      "mean_token_accuracy": 0.19038989394903183,
      "num_tokens": 111558193.0,
      "step": 50915
    },
    {
      "entropy": 5.686546897888183,
      "epoch": 5.053592695514093,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027468820293417357,
      "loss": 4.9396,
      "mean_token_accuracy": 0.19258256107568741,
      "num_tokens": 111568965.0,
      "step": 50920
    },
    {
      "entropy": 5.653709459304809,
      "epoch": 5.05408892417626,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002746527714761132,
      "loss": 4.8675,
      "mean_token_accuracy": 0.20633742958307266,
      "num_tokens": 111579898.0,
      "step": 50925
    },
    {
      "entropy": 5.656572580337524,
      "epoch": 5.054585152838428,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027461734002666335,
      "loss": 4.8977,
      "mean_token_accuracy": 0.1981096312403679,
      "num_tokens": 111590051.0,
      "step": 50930
    },
    {
      "entropy": 5.67128210067749,
      "epoch": 5.055081381500595,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027458190858670265,
      "loss": 4.9255,
      "mean_token_accuracy": 0.198947037756443,
      "num_tokens": 111600475.0,
      "step": 50935
    },
    {
      "entropy": 5.653863477706909,
      "epoch": 5.055577610162763,
      "grad_norm": 1.25,
      "learning_rate": 0.00027454647715710964,
      "loss": 4.8164,
      "mean_token_accuracy": 0.2062255397439003,
      "num_tokens": 111610434.0,
      "step": 50940
    },
    {
      "entropy": 5.654089689254761,
      "epoch": 5.056073838824931,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027451104573876305,
      "loss": 4.9171,
      "mean_token_accuracy": 0.19407099336385727,
      "num_tokens": 111620451.0,
      "step": 50945
    },
    {
      "entropy": 5.701867628097534,
      "epoch": 5.056570067487098,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027447561433254153,
      "loss": 5.0023,
      "mean_token_accuracy": 0.19093122333288193,
      "num_tokens": 111631698.0,
      "step": 50950
    },
    {
      "entropy": 5.652398681640625,
      "epoch": 5.057066296149266,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027444018293932366,
      "loss": 4.8303,
      "mean_token_accuracy": 0.2032418042421341,
      "num_tokens": 111642173.0,
      "step": 50955
    },
    {
      "entropy": 5.661304998397827,
      "epoch": 5.057562524811433,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027440475155998803,
      "loss": 4.913,
      "mean_token_accuracy": 0.19823730289936065,
      "num_tokens": 111652839.0,
      "step": 50960
    },
    {
      "entropy": 5.699708938598633,
      "epoch": 5.058058753473601,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027436932019541315,
      "loss": 4.9789,
      "mean_token_accuracy": 0.1962178960442543,
      "num_tokens": 111663117.0,
      "step": 50965
    },
    {
      "entropy": 5.713106966018676,
      "epoch": 5.058554982135768,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002743338888464778,
      "loss": 4.911,
      "mean_token_accuracy": 0.19725992381572724,
      "num_tokens": 111674402.0,
      "step": 50970
    },
    {
      "entropy": 5.712147903442383,
      "epoch": 5.059051210797936,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002742984575140607,
      "loss": 4.9344,
      "mean_token_accuracy": 0.19209131300449372,
      "num_tokens": 111685763.0,
      "step": 50975
    },
    {
      "entropy": 5.723453664779663,
      "epoch": 5.059547439460103,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027426302619904024,
      "loss": 4.9597,
      "mean_token_accuracy": 0.19433796554803848,
      "num_tokens": 111696672.0,
      "step": 50980
    },
    {
      "entropy": 5.754551982879638,
      "epoch": 5.060043668122271,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002742275949022952,
      "loss": 5.0667,
      "mean_token_accuracy": 0.19077221155166627,
      "num_tokens": 111707117.0,
      "step": 50985
    },
    {
      "entropy": 5.719279193878174,
      "epoch": 5.060539896784438,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027419216362470415,
      "loss": 4.9644,
      "mean_token_accuracy": 0.19586135149002076,
      "num_tokens": 111719114.0,
      "step": 50990
    },
    {
      "entropy": 5.70372953414917,
      "epoch": 5.061036125446606,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027415673236714567,
      "loss": 4.8819,
      "mean_token_accuracy": 0.19780248105525972,
      "num_tokens": 111730095.0,
      "step": 50995
    },
    {
      "entropy": 5.705722427368164,
      "epoch": 5.061532354108773,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002741213011304983,
      "loss": 4.9645,
      "mean_token_accuracy": 0.19349105954170226,
      "num_tokens": 111741841.0,
      "step": 51000
    },
    {
      "epoch": 5.061532354108773,
      "eval_entropy": 5.429529687519395,
      "eval_loss": 5.0921220779418945,
      "eval_mean_token_accuracy": 0.19425826281287886,
      "eval_num_tokens": 111741841.0,
      "eval_runtime": 20.5033,
      "eval_samples_per_second": 1642.126,
      "eval_steps_per_second": 205.284,
      "step": 51000
    },
    {
      "entropy": 5.680042123794555,
      "epoch": 5.062028582770941,
      "grad_norm": 1.375,
      "learning_rate": 0.00027408586991564096,
      "loss": 4.8826,
      "mean_token_accuracy": 0.20391671657562255,
      "num_tokens": 111752215.0,
      "step": 51005
    },
    {
      "entropy": 5.624039030075073,
      "epoch": 5.062524811433108,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000274050438723452,
      "loss": 4.8687,
      "mean_token_accuracy": 0.20417113453149796,
      "num_tokens": 111761673.0,
      "step": 51010
    },
    {
      "entropy": 5.702330684661865,
      "epoch": 5.0630210400952755,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00027401500755481014,
      "loss": 4.9218,
      "mean_token_accuracy": 0.20366492867469788,
      "num_tokens": 111773673.0,
      "step": 51015
    },
    {
      "entropy": 5.693526411056519,
      "epoch": 5.063517268757443,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027397957641059394,
      "loss": 4.9117,
      "mean_token_accuracy": 0.1956127643585205,
      "num_tokens": 111784354.0,
      "step": 51020
    },
    {
      "entropy": 5.72853364944458,
      "epoch": 5.064013497419611,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027394414529168204,
      "loss": 5.012,
      "mean_token_accuracy": 0.19249616265296937,
      "num_tokens": 111795391.0,
      "step": 51025
    },
    {
      "entropy": 5.664380598068237,
      "epoch": 5.064509726081779,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002739087141989531,
      "loss": 4.8683,
      "mean_token_accuracy": 0.20207006335258484,
      "num_tokens": 111806841.0,
      "step": 51030
    },
    {
      "entropy": 5.751510095596314,
      "epoch": 5.065005954743946,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002738732831332856,
      "loss": 5.0331,
      "mean_token_accuracy": 0.18842415064573287,
      "num_tokens": 111817762.0,
      "step": 51035
    },
    {
      "entropy": 5.653390789031983,
      "epoch": 5.065502183406114,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002738378520955583,
      "loss": 4.8878,
      "mean_token_accuracy": 0.19858685582876207,
      "num_tokens": 111828938.0,
      "step": 51040
    },
    {
      "entropy": 5.703112411499023,
      "epoch": 5.065998412068281,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027380242108664983,
      "loss": 5.0301,
      "mean_token_accuracy": 0.19428151696920395,
      "num_tokens": 111839202.0,
      "step": 51045
    },
    {
      "entropy": 5.665394687652588,
      "epoch": 5.066494640730449,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002737669901074387,
      "loss": 4.8603,
      "mean_token_accuracy": 0.20430830121040344,
      "num_tokens": 111849765.0,
      "step": 51050
    },
    {
      "entropy": 5.74689884185791,
      "epoch": 5.066990869392616,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027373155915880355,
      "loss": 4.9583,
      "mean_token_accuracy": 0.19662352353334428,
      "num_tokens": 111861747.0,
      "step": 51055
    },
    {
      "entropy": 5.7154113292694095,
      "epoch": 5.067487098054784,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002736961282416229,
      "loss": 4.9583,
      "mean_token_accuracy": 0.2003811001777649,
      "num_tokens": 111873891.0,
      "step": 51060
    },
    {
      "entropy": 5.737500429153442,
      "epoch": 5.067983326716951,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002736606973567756,
      "loss": 4.96,
      "mean_token_accuracy": 0.1896360158920288,
      "num_tokens": 111884218.0,
      "step": 51065
    },
    {
      "entropy": 5.706422805786133,
      "epoch": 5.068479555379119,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027362526650514006,
      "loss": 4.9442,
      "mean_token_accuracy": 0.1989652618765831,
      "num_tokens": 111895251.0,
      "step": 51070
    },
    {
      "entropy": 5.692791604995728,
      "epoch": 5.068975784041286,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000273589835687595,
      "loss": 4.9162,
      "mean_token_accuracy": 0.19324288815259932,
      "num_tokens": 111905380.0,
      "step": 51075
    },
    {
      "entropy": 5.6125024318695065,
      "epoch": 5.069472012703454,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027355440490501893,
      "loss": 4.8794,
      "mean_token_accuracy": 0.2029230162501335,
      "num_tokens": 111916260.0,
      "step": 51080
    },
    {
      "entropy": 5.621552276611328,
      "epoch": 5.069968241365621,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027351897415829043,
      "loss": 4.9062,
      "mean_token_accuracy": 0.2039901450276375,
      "num_tokens": 111927077.0,
      "step": 51085
    },
    {
      "entropy": 5.617238140106201,
      "epoch": 5.0704644700277886,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002734835434482883,
      "loss": 4.7744,
      "mean_token_accuracy": 0.20845568031072617,
      "num_tokens": 111937144.0,
      "step": 51090
    },
    {
      "entropy": 5.685615491867066,
      "epoch": 5.070960698689956,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000273448112775891,
      "loss": 4.8622,
      "mean_token_accuracy": 0.1989223912358284,
      "num_tokens": 111948669.0,
      "step": 51095
    },
    {
      "entropy": 5.692382049560547,
      "epoch": 5.0714569273521235,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002734126821419772,
      "loss": 4.976,
      "mean_token_accuracy": 0.19498029053211213,
      "num_tokens": 111959498.0,
      "step": 51100
    },
    {
      "entropy": 5.7016191482543945,
      "epoch": 5.071953156014291,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00027337725154742535,
      "loss": 4.8987,
      "mean_token_accuracy": 0.20534315407276155,
      "num_tokens": 111969463.0,
      "step": 51105
    },
    {
      "entropy": 5.712182474136353,
      "epoch": 5.072449384676459,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027334182099311423,
      "loss": 4.9448,
      "mean_token_accuracy": 0.19627935886383058,
      "num_tokens": 111980251.0,
      "step": 51110
    },
    {
      "entropy": 5.683175325393677,
      "epoch": 5.072945613338627,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002733063904799223,
      "loss": 4.8846,
      "mean_token_accuracy": 0.19870079755783082,
      "num_tokens": 111991022.0,
      "step": 51115
    },
    {
      "entropy": 5.680105972290039,
      "epoch": 5.073441842000794,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002732709600087283,
      "loss": 4.9621,
      "mean_token_accuracy": 0.19780575931072236,
      "num_tokens": 112001850.0,
      "step": 51120
    },
    {
      "entropy": 5.729715633392334,
      "epoch": 5.073938070662962,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002732355295804108,
      "loss": 4.9482,
      "mean_token_accuracy": 0.19985172003507615,
      "num_tokens": 112013615.0,
      "step": 51125
    },
    {
      "entropy": 5.6858877658844,
      "epoch": 5.074434299325129,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002732000991958484,
      "loss": 4.9309,
      "mean_token_accuracy": 0.20018851161003112,
      "num_tokens": 112023976.0,
      "step": 51130
    },
    {
      "entropy": 5.6781871795654295,
      "epoch": 5.074930527987297,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002731646688559196,
      "loss": 4.8799,
      "mean_token_accuracy": 0.20511150658130645,
      "num_tokens": 112034770.0,
      "step": 51135
    },
    {
      "entropy": 5.741089582443237,
      "epoch": 5.075426756649464,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00027312923856150306,
      "loss": 4.9459,
      "mean_token_accuracy": 0.19680882692337037,
      "num_tokens": 112045090.0,
      "step": 51140
    },
    {
      "entropy": 5.663630485534668,
      "epoch": 5.075922985311632,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002730938083134773,
      "loss": 4.8745,
      "mean_token_accuracy": 0.19756615459918975,
      "num_tokens": 112056873.0,
      "step": 51145
    },
    {
      "entropy": 5.710334396362304,
      "epoch": 5.076419213973799,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027305837811272105,
      "loss": 4.9531,
      "mean_token_accuracy": 0.19364677965641022,
      "num_tokens": 112068279.0,
      "step": 51150
    },
    {
      "entropy": 5.785077857971191,
      "epoch": 5.076915442635967,
      "grad_norm": 1.375,
      "learning_rate": 0.0002730229479601129,
      "loss": 5.0289,
      "mean_token_accuracy": 0.19004678279161452,
      "num_tokens": 112078710.0,
      "step": 51155
    },
    {
      "entropy": 5.661362743377685,
      "epoch": 5.077411671298134,
      "grad_norm": 1.46875,
      "learning_rate": 0.00027298751785653124,
      "loss": 4.8609,
      "mean_token_accuracy": 0.20839281231164933,
      "num_tokens": 112089213.0,
      "step": 51160
    },
    {
      "entropy": 5.6531023502349855,
      "epoch": 5.077907899960302,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027295208780285496,
      "loss": 4.928,
      "mean_token_accuracy": 0.1938678056001663,
      "num_tokens": 112099966.0,
      "step": 51165
    },
    {
      "entropy": 5.653366041183472,
      "epoch": 5.078404128622469,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027291665779996245,
      "loss": 4.9076,
      "mean_token_accuracy": 0.2000040367245674,
      "num_tokens": 112112377.0,
      "step": 51170
    },
    {
      "entropy": 5.733264350891114,
      "epoch": 5.0789003572846365,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002728812278487322,
      "loss": 4.9661,
      "mean_token_accuracy": 0.19257913082838057,
      "num_tokens": 112124855.0,
      "step": 51175
    },
    {
      "entropy": 5.650633430480957,
      "epoch": 5.079396585946804,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027284579795004303,
      "loss": 4.939,
      "mean_token_accuracy": 0.19551971256732942,
      "num_tokens": 112136274.0,
      "step": 51180
    },
    {
      "entropy": 5.696023035049438,
      "epoch": 5.0798928146089715,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027281036810477345,
      "loss": 4.8668,
      "mean_token_accuracy": 0.20189937502145766,
      "num_tokens": 112146926.0,
      "step": 51185
    },
    {
      "entropy": 5.725862455368042,
      "epoch": 5.080389043271139,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002727749383138021,
      "loss": 5.0223,
      "mean_token_accuracy": 0.19491685330867767,
      "num_tokens": 112158423.0,
      "step": 51190
    },
    {
      "entropy": 5.640985584259033,
      "epoch": 5.080885271933307,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002727395085780073,
      "loss": 4.8167,
      "mean_token_accuracy": 0.20375362038612366,
      "num_tokens": 112170587.0,
      "step": 51195
    },
    {
      "entropy": 5.695815896987915,
      "epoch": 5.081381500595475,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027270407889826793,
      "loss": 4.8815,
      "mean_token_accuracy": 0.19917480349540712,
      "num_tokens": 112181402.0,
      "step": 51200
    },
    {
      "entropy": 5.656662940979004,
      "epoch": 5.081877729257642,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002726686492754624,
      "loss": 4.954,
      "mean_token_accuracy": 0.2065192937850952,
      "num_tokens": 112192916.0,
      "step": 51205
    },
    {
      "entropy": 5.716633987426758,
      "epoch": 5.08237395791981,
      "grad_norm": 1.25,
      "learning_rate": 0.0002726332197104693,
      "loss": 4.9654,
      "mean_token_accuracy": 0.1946700096130371,
      "num_tokens": 112205214.0,
      "step": 51210
    },
    {
      "entropy": 5.704485940933227,
      "epoch": 5.082870186581977,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002725977902041674,
      "loss": 4.9835,
      "mean_token_accuracy": 0.18971103876829148,
      "num_tokens": 112214844.0,
      "step": 51215
    },
    {
      "entropy": 5.7179426670074465,
      "epoch": 5.083366415244145,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002725623607574351,
      "loss": 4.9373,
      "mean_token_accuracy": 0.20691316425800324,
      "num_tokens": 112227439.0,
      "step": 51220
    },
    {
      "entropy": 5.713520765304565,
      "epoch": 5.083862643906312,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002725269313711509,
      "loss": 4.9894,
      "mean_token_accuracy": 0.18736036568880082,
      "num_tokens": 112236811.0,
      "step": 51225
    },
    {
      "entropy": 5.680002164840698,
      "epoch": 5.08435887256848,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002724915020461936,
      "loss": 4.9644,
      "mean_token_accuracy": 0.19299093186855315,
      "num_tokens": 112246805.0,
      "step": 51230
    },
    {
      "entropy": 5.69156699180603,
      "epoch": 5.084855101230647,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002724560727834415,
      "loss": 4.9266,
      "mean_token_accuracy": 0.19957747757434846,
      "num_tokens": 112258294.0,
      "step": 51235
    },
    {
      "entropy": 5.660602998733521,
      "epoch": 5.085351329892815,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027242064358377346,
      "loss": 4.9332,
      "mean_token_accuracy": 0.19966545850038528,
      "num_tokens": 112270953.0,
      "step": 51240
    },
    {
      "entropy": 5.680034065246582,
      "epoch": 5.085847558554982,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002723852144480678,
      "loss": 4.8445,
      "mean_token_accuracy": 0.2062855139374733,
      "num_tokens": 112281889.0,
      "step": 51245
    },
    {
      "entropy": 5.649099636077881,
      "epoch": 5.0863437872171495,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027234978537720323,
      "loss": 4.8677,
      "mean_token_accuracy": 0.1962835431098938,
      "num_tokens": 112292543.0,
      "step": 51250
    },
    {
      "entropy": 5.6466137886047365,
      "epoch": 5.086840015879317,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027231435637205834,
      "loss": 4.8912,
      "mean_token_accuracy": 0.1976710245013237,
      "num_tokens": 112302164.0,
      "step": 51255
    },
    {
      "entropy": 5.640860939025879,
      "epoch": 5.0873362445414845,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002722789274335115,
      "loss": 4.8938,
      "mean_token_accuracy": 0.2004529505968094,
      "num_tokens": 112313534.0,
      "step": 51260
    },
    {
      "entropy": 5.73177285194397,
      "epoch": 5.087832473203652,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027224349856244136,
      "loss": 4.9385,
      "mean_token_accuracy": 0.20114822536706925,
      "num_tokens": 112324334.0,
      "step": 51265
    },
    {
      "entropy": 5.6904990673065186,
      "epoch": 5.088328701865819,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002722080697597266,
      "loss": 4.8904,
      "mean_token_accuracy": 0.20333772748708726,
      "num_tokens": 112335167.0,
      "step": 51270
    },
    {
      "entropy": 5.732963275909424,
      "epoch": 5.088824930527987,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002721726410262457,
      "loss": 4.9319,
      "mean_token_accuracy": 0.20215461850166322,
      "num_tokens": 112346105.0,
      "step": 51275
    },
    {
      "entropy": 5.700258207321167,
      "epoch": 5.089321159190154,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002721372123628772,
      "loss": 4.8817,
      "mean_token_accuracy": 0.1928731843829155,
      "num_tokens": 112357828.0,
      "step": 51280
    },
    {
      "entropy": 5.646162128448486,
      "epoch": 5.089817387852323,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027210178377049964,
      "loss": 4.9188,
      "mean_token_accuracy": 0.20118648409843445,
      "num_tokens": 112367609.0,
      "step": 51285
    },
    {
      "entropy": 5.691160583496094,
      "epoch": 5.09031361651449,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027206635524999165,
      "loss": 4.9664,
      "mean_token_accuracy": 0.19451892226934434,
      "num_tokens": 112378667.0,
      "step": 51290
    },
    {
      "entropy": 5.71085000038147,
      "epoch": 5.090809845176658,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002720309268022316,
      "loss": 4.886,
      "mean_token_accuracy": 0.20574795454740524,
      "num_tokens": 112389128.0,
      "step": 51295
    },
    {
      "entropy": 5.8025377750396725,
      "epoch": 5.091306073838825,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027199549842809836,
      "loss": 5.0115,
      "mean_token_accuracy": 0.18545071333646773,
      "num_tokens": 112398486.0,
      "step": 51300
    },
    {
      "entropy": 5.724757719039917,
      "epoch": 5.091802302500993,
      "grad_norm": 1.265625,
      "learning_rate": 0.00027196007012847016,
      "loss": 4.9441,
      "mean_token_accuracy": 0.1932896703481674,
      "num_tokens": 112409477.0,
      "step": 51305
    },
    {
      "entropy": 5.7295060634613035,
      "epoch": 5.09229853116316,
      "grad_norm": 1.578125,
      "learning_rate": 0.00027192464190422573,
      "loss": 5.0262,
      "mean_token_accuracy": 0.19138094484806062,
      "num_tokens": 112420564.0,
      "step": 51310
    },
    {
      "entropy": 5.6446911811828615,
      "epoch": 5.092794759825328,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027188921375624354,
      "loss": 4.9417,
      "mean_token_accuracy": 0.20004604607820511,
      "num_tokens": 112431442.0,
      "step": 51315
    },
    {
      "entropy": 5.661621189117431,
      "epoch": 5.093290988487495,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027185378568540217,
      "loss": 4.869,
      "mean_token_accuracy": 0.20261743366718293,
      "num_tokens": 112441740.0,
      "step": 51320
    },
    {
      "entropy": 5.741247606277466,
      "epoch": 5.0937872171496625,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00027181835769258,
      "loss": 5.0024,
      "mean_token_accuracy": 0.1850871041417122,
      "num_tokens": 112451819.0,
      "step": 51325
    },
    {
      "entropy": 5.814014482498169,
      "epoch": 5.09428344581183,
      "grad_norm": 1.25,
      "learning_rate": 0.0002717829297786558,
      "loss": 4.9529,
      "mean_token_accuracy": 0.19965378642082215,
      "num_tokens": 112462294.0,
      "step": 51330
    },
    {
      "entropy": 5.7263916492462155,
      "epoch": 5.0947796744739975,
      "grad_norm": 1.265625,
      "learning_rate": 0.000271747501944508,
      "loss": 4.9157,
      "mean_token_accuracy": 0.19543571025133133,
      "num_tokens": 112473258.0,
      "step": 51335
    },
    {
      "entropy": 5.639217376708984,
      "epoch": 5.095275903136165,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027171207419101516,
      "loss": 4.8499,
      "mean_token_accuracy": 0.2045545235276222,
      "num_tokens": 112482815.0,
      "step": 51340
    },
    {
      "entropy": 5.65828013420105,
      "epoch": 5.095772131798332,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027167664651905584,
      "loss": 5.006,
      "mean_token_accuracy": 0.19592061638832092,
      "num_tokens": 112493325.0,
      "step": 51345
    },
    {
      "entropy": 5.7765717029571535,
      "epoch": 5.0962683604605,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002716412189295084,
      "loss": 5.1112,
      "mean_token_accuracy": 0.18773147016763686,
      "num_tokens": 112504477.0,
      "step": 51350
    },
    {
      "entropy": 5.7495198249816895,
      "epoch": 5.096764589122667,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002716057914232516,
      "loss": 4.9086,
      "mean_token_accuracy": 0.19181365817785262,
      "num_tokens": 112514331.0,
      "step": 51355
    },
    {
      "entropy": 5.7396961688995365,
      "epoch": 5.097260817784835,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002715703640011638,
      "loss": 5.047,
      "mean_token_accuracy": 0.1914614036679268,
      "num_tokens": 112526007.0,
      "step": 51360
    },
    {
      "entropy": 5.684255170822143,
      "epoch": 5.097757046447002,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002715349366641236,
      "loss": 4.9168,
      "mean_token_accuracy": 0.19950101524591446,
      "num_tokens": 112536317.0,
      "step": 51365
    },
    {
      "entropy": 5.728051662445068,
      "epoch": 5.098253275109171,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027149950941300963,
      "loss": 4.9081,
      "mean_token_accuracy": 0.20145319998264313,
      "num_tokens": 112546613.0,
      "step": 51370
    },
    {
      "entropy": 5.7376134395599365,
      "epoch": 5.098749503771338,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002714640822487001,
      "loss": 4.9988,
      "mean_token_accuracy": 0.19733433276414872,
      "num_tokens": 112556731.0,
      "step": 51375
    },
    {
      "entropy": 5.715303993225097,
      "epoch": 5.099245732433506,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002714286551720737,
      "loss": 4.9366,
      "mean_token_accuracy": 0.196652327477932,
      "num_tokens": 112568365.0,
      "step": 51380
    },
    {
      "entropy": 5.711162328720093,
      "epoch": 5.099741961095673,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002713932281840091,
      "loss": 4.9293,
      "mean_token_accuracy": 0.19845287203788758,
      "num_tokens": 112579147.0,
      "step": 51385
    },
    {
      "entropy": 5.64820613861084,
      "epoch": 5.100238189757841,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027135780128538457,
      "loss": 4.9377,
      "mean_token_accuracy": 0.19755430221557618,
      "num_tokens": 112589773.0,
      "step": 51390
    },
    {
      "entropy": 5.648815250396728,
      "epoch": 5.100734418420008,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027132237447707874,
      "loss": 4.8593,
      "mean_token_accuracy": 0.20021719485521317,
      "num_tokens": 112599991.0,
      "step": 51395
    },
    {
      "entropy": 5.69413800239563,
      "epoch": 5.1012306470821756,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027128694775997015,
      "loss": 4.9367,
      "mean_token_accuracy": 0.19614386856555938,
      "num_tokens": 112610989.0,
      "step": 51400
    },
    {
      "entropy": 5.7290363788604735,
      "epoch": 5.101726875744343,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027125152113493726,
      "loss": 4.9412,
      "mean_token_accuracy": 0.19372900277376176,
      "num_tokens": 112622860.0,
      "step": 51405
    },
    {
      "entropy": 5.70937705039978,
      "epoch": 5.1022231044065105,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002712160946028584,
      "loss": 4.979,
      "mean_token_accuracy": 0.18986533284187318,
      "num_tokens": 112634559.0,
      "step": 51410
    },
    {
      "entropy": 5.64170389175415,
      "epoch": 5.102719333068678,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002711806681646124,
      "loss": 4.8465,
      "mean_token_accuracy": 0.19539760202169418,
      "num_tokens": 112644809.0,
      "step": 51415
    },
    {
      "entropy": 5.656359100341797,
      "epoch": 5.1032155617308455,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002711452418210776,
      "loss": 4.8871,
      "mean_token_accuracy": 0.19541347920894622,
      "num_tokens": 112655288.0,
      "step": 51420
    },
    {
      "entropy": 5.653652143478394,
      "epoch": 5.103711790393013,
      "grad_norm": 1.25,
      "learning_rate": 0.0002711098155731324,
      "loss": 4.9387,
      "mean_token_accuracy": 0.197084479033947,
      "num_tokens": 112666018.0,
      "step": 51425
    },
    {
      "entropy": 5.706056833267212,
      "epoch": 5.10420801905518,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002710743894216555,
      "loss": 4.9391,
      "mean_token_accuracy": 0.194822695851326,
      "num_tokens": 112677232.0,
      "step": 51430
    },
    {
      "entropy": 5.628639984130859,
      "epoch": 5.104704247717348,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002710389633675252,
      "loss": 4.8878,
      "mean_token_accuracy": 0.20061993598937988,
      "num_tokens": 112688051.0,
      "step": 51435
    },
    {
      "entropy": 5.704768180847168,
      "epoch": 5.105200476379515,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002710035374116201,
      "loss": 4.9583,
      "mean_token_accuracy": 0.20605881810188292,
      "num_tokens": 112700050.0,
      "step": 51440
    },
    {
      "entropy": 5.71831259727478,
      "epoch": 5.105696705041683,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027096811155481867,
      "loss": 4.9426,
      "mean_token_accuracy": 0.20456757992506028,
      "num_tokens": 112712248.0,
      "step": 51445
    },
    {
      "entropy": 5.747057628631592,
      "epoch": 5.10619293370385,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002709326857979994,
      "loss": 4.9657,
      "mean_token_accuracy": 0.18954510390758514,
      "num_tokens": 112724402.0,
      "step": 51450
    },
    {
      "entropy": 5.640089797973633,
      "epoch": 5.106689162366019,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027089726014204075,
      "loss": 4.9069,
      "mean_token_accuracy": 0.20017314553260804,
      "num_tokens": 112734709.0,
      "step": 51455
    },
    {
      "entropy": 5.701892232894897,
      "epoch": 5.107185391028186,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002708618345878212,
      "loss": 4.9831,
      "mean_token_accuracy": 0.18800555914640427,
      "num_tokens": 112743728.0,
      "step": 51460
    },
    {
      "entropy": 5.708095407485962,
      "epoch": 5.107681619690354,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027082640913621926,
      "loss": 4.9045,
      "mean_token_accuracy": 0.20008797347545623,
      "num_tokens": 112754809.0,
      "step": 51465
    },
    {
      "entropy": 5.753643131256103,
      "epoch": 5.108177848352521,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002707909837881133,
      "loss": 4.9199,
      "mean_token_accuracy": 0.2005905270576477,
      "num_tokens": 112765793.0,
      "step": 51470
    },
    {
      "entropy": 5.734467172622681,
      "epoch": 5.108674077014689,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0002707555585443819,
      "loss": 4.9082,
      "mean_token_accuracy": 0.20568899512290956,
      "num_tokens": 112776517.0,
      "step": 51475
    },
    {
      "entropy": 5.655622053146362,
      "epoch": 5.109170305676856,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002707201334059036,
      "loss": 4.9299,
      "mean_token_accuracy": 0.20936342924833298,
      "num_tokens": 112788239.0,
      "step": 51480
    },
    {
      "entropy": 5.637726497650147,
      "epoch": 5.1096665343390235,
      "grad_norm": 1.25,
      "learning_rate": 0.0002706847083735567,
      "loss": 4.9054,
      "mean_token_accuracy": 0.1920645773410797,
      "num_tokens": 112798304.0,
      "step": 51485
    },
    {
      "entropy": 5.700958728790283,
      "epoch": 5.110162763001191,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00027064928344821973,
      "loss": 4.93,
      "mean_token_accuracy": 0.20505077689886092,
      "num_tokens": 112809257.0,
      "step": 51490
    },
    {
      "entropy": 5.646776723861694,
      "epoch": 5.1106589916633585,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027061385863077114,
      "loss": 4.8182,
      "mean_token_accuracy": 0.20239027589559555,
      "num_tokens": 112819157.0,
      "step": 51495
    },
    {
      "entropy": 5.614333248138427,
      "epoch": 5.111155220325526,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027057843392208934,
      "loss": 4.8573,
      "mean_token_accuracy": 0.19533316045999527,
      "num_tokens": 112830430.0,
      "step": 51500
    },
    {
      "entropy": 5.674358224868774,
      "epoch": 5.111651448987693,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027054300932305295,
      "loss": 4.9691,
      "mean_token_accuracy": 0.1953636571764946,
      "num_tokens": 112842313.0,
      "step": 51505
    },
    {
      "entropy": 5.666084861755371,
      "epoch": 5.112147677649861,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027050758483454035,
      "loss": 4.8516,
      "mean_token_accuracy": 0.1969524547457695,
      "num_tokens": 112852200.0,
      "step": 51510
    },
    {
      "entropy": 5.709683561325074,
      "epoch": 5.112643906312028,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027047216045742996,
      "loss": 4.9362,
      "mean_token_accuracy": 0.2005935549736023,
      "num_tokens": 112863505.0,
      "step": 51515
    },
    {
      "entropy": 5.7442844867706295,
      "epoch": 5.113140134974196,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002704367361926002,
      "loss": 5.0016,
      "mean_token_accuracy": 0.18848720341920852,
      "num_tokens": 112874632.0,
      "step": 51520
    },
    {
      "entropy": 5.686966133117676,
      "epoch": 5.113636363636363,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027040131204092953,
      "loss": 4.9156,
      "mean_token_accuracy": 0.19483092576265335,
      "num_tokens": 112884870.0,
      "step": 51525
    },
    {
      "entropy": 5.666449165344238,
      "epoch": 5.114132592298531,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002703658880032964,
      "loss": 4.9766,
      "mean_token_accuracy": 0.1989713877439499,
      "num_tokens": 112895155.0,
      "step": 51530
    },
    {
      "entropy": 5.725821018218994,
      "epoch": 5.114628820960698,
      "grad_norm": 1.25,
      "learning_rate": 0.0002703304640805793,
      "loss": 4.9825,
      "mean_token_accuracy": 0.19860805869102477,
      "num_tokens": 112904391.0,
      "step": 51535
    },
    {
      "entropy": 5.684945678710937,
      "epoch": 5.115125049622867,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002702950402736566,
      "loss": 4.8995,
      "mean_token_accuracy": 0.2023262083530426,
      "num_tokens": 112916191.0,
      "step": 51540
    },
    {
      "entropy": 5.698754739761353,
      "epoch": 5.115621278285034,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002702596165834068,
      "loss": 4.9251,
      "mean_token_accuracy": 0.19770684391260146,
      "num_tokens": 112928383.0,
      "step": 51545
    },
    {
      "entropy": 5.636571073532105,
      "epoch": 5.116117506947202,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002702241930107083,
      "loss": 4.8486,
      "mean_token_accuracy": 0.2034152090549469,
      "num_tokens": 112939897.0,
      "step": 51550
    },
    {
      "entropy": 5.659470176696777,
      "epoch": 5.116613735609369,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002701887695564395,
      "loss": 4.9603,
      "mean_token_accuracy": 0.19188893139362334,
      "num_tokens": 112950965.0,
      "step": 51555
    },
    {
      "entropy": 5.699693059921264,
      "epoch": 5.1171099642715365,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027015334622147876,
      "loss": 4.928,
      "mean_token_accuracy": 0.1961871787905693,
      "num_tokens": 112961695.0,
      "step": 51560
    },
    {
      "entropy": 5.673424100875854,
      "epoch": 5.117606192933704,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027011792300670476,
      "loss": 4.8331,
      "mean_token_accuracy": 0.19599697440862657,
      "num_tokens": 112972491.0,
      "step": 51565
    },
    {
      "entropy": 5.770222091674805,
      "epoch": 5.1181024215958715,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002700824999129956,
      "loss": 5.0535,
      "mean_token_accuracy": 0.18282972425222396,
      "num_tokens": 112983056.0,
      "step": 51570
    },
    {
      "entropy": 5.712692594528198,
      "epoch": 5.118598650258039,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027004707694122996,
      "loss": 5.0004,
      "mean_token_accuracy": 0.1868721514940262,
      "num_tokens": 112995236.0,
      "step": 51575
    },
    {
      "entropy": 5.704993152618409,
      "epoch": 5.119094878920206,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002700116540922861,
      "loss": 4.8807,
      "mean_token_accuracy": 0.19927671253681184,
      "num_tokens": 113006863.0,
      "step": 51580
    },
    {
      "entropy": 5.688982105255127,
      "epoch": 5.119591107582374,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002699762313670424,
      "loss": 4.958,
      "mean_token_accuracy": 0.19877121448516846,
      "num_tokens": 113017675.0,
      "step": 51585
    },
    {
      "entropy": 5.742813158035278,
      "epoch": 5.120087336244541,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026994080876637736,
      "loss": 5.0035,
      "mean_token_accuracy": 0.18823701292276382,
      "num_tokens": 113027455.0,
      "step": 51590
    },
    {
      "entropy": 5.664602756500244,
      "epoch": 5.120583564906709,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026990538629116943,
      "loss": 4.9645,
      "mean_token_accuracy": 0.19932613223791124,
      "num_tokens": 113038879.0,
      "step": 51595
    },
    {
      "entropy": 5.754399967193604,
      "epoch": 5.121079793568876,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026986996394229697,
      "loss": 5.0006,
      "mean_token_accuracy": 0.1946472316980362,
      "num_tokens": 113050929.0,
      "step": 51600
    },
    {
      "entropy": 5.76364426612854,
      "epoch": 5.121576022231044,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026983454172063834,
      "loss": 5.01,
      "mean_token_accuracy": 0.19243616312742234,
      "num_tokens": 113063626.0,
      "step": 51605
    },
    {
      "entropy": 5.755939245223999,
      "epoch": 5.122072250893211,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002697991196270719,
      "loss": 5.0362,
      "mean_token_accuracy": 0.19199542999267577,
      "num_tokens": 113073853.0,
      "step": 51610
    },
    {
      "entropy": 5.723512506484985,
      "epoch": 5.122568479555379,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000269763697662476,
      "loss": 4.9253,
      "mean_token_accuracy": 0.19555389732122422,
      "num_tokens": 113084180.0,
      "step": 51615
    },
    {
      "entropy": 5.675548934936524,
      "epoch": 5.123064708217546,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026972827582772915,
      "loss": 4.8645,
      "mean_token_accuracy": 0.2050921693444252,
      "num_tokens": 113093801.0,
      "step": 51620
    },
    {
      "entropy": 5.768913745880127,
      "epoch": 5.123560936879715,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026969285412370983,
      "loss": 5.0016,
      "mean_token_accuracy": 0.19041577279567717,
      "num_tokens": 113104932.0,
      "step": 51625
    },
    {
      "entropy": 5.655604219436645,
      "epoch": 5.124057165541882,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026965743255129624,
      "loss": 4.8758,
      "mean_token_accuracy": 0.19955455511808395,
      "num_tokens": 113115253.0,
      "step": 51630
    },
    {
      "entropy": 5.680212688446045,
      "epoch": 5.1245533942040495,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002696220111113667,
      "loss": 4.9053,
      "mean_token_accuracy": 0.20356392115354538,
      "num_tokens": 113127027.0,
      "step": 51635
    },
    {
      "entropy": 5.674380826950073,
      "epoch": 5.125049622866217,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002695865898047998,
      "loss": 4.8752,
      "mean_token_accuracy": 0.19982656836509705,
      "num_tokens": 113137140.0,
      "step": 51640
    },
    {
      "entropy": 5.674161148071289,
      "epoch": 5.1255458515283845,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026955116863247374,
      "loss": 4.9304,
      "mean_token_accuracy": 0.19757591038942338,
      "num_tokens": 113148465.0,
      "step": 51645
    },
    {
      "entropy": 5.672725820541382,
      "epoch": 5.126042080190552,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026951574759526706,
      "loss": 4.9209,
      "mean_token_accuracy": 0.2009134218096733,
      "num_tokens": 113158551.0,
      "step": 51650
    },
    {
      "entropy": 5.726443815231323,
      "epoch": 5.126538308852719,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026948032669405793,
      "loss": 4.961,
      "mean_token_accuracy": 0.19736890643835067,
      "num_tokens": 113169573.0,
      "step": 51655
    },
    {
      "entropy": 5.701073455810547,
      "epoch": 5.127034537514887,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002694449059297249,
      "loss": 4.9346,
      "mean_token_accuracy": 0.19392581582069396,
      "num_tokens": 113181409.0,
      "step": 51660
    },
    {
      "entropy": 5.740078067779541,
      "epoch": 5.127530766177054,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00026940948530314605,
      "loss": 4.9767,
      "mean_token_accuracy": 0.18917960077524185,
      "num_tokens": 113193816.0,
      "step": 51665
    },
    {
      "entropy": 5.740933895111084,
      "epoch": 5.128026994839222,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00026937406481520015,
      "loss": 4.9621,
      "mean_token_accuracy": 0.1930857241153717,
      "num_tokens": 113204738.0,
      "step": 51670
    },
    {
      "entropy": 5.774988317489624,
      "epoch": 5.128523223501389,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002693386444667651,
      "loss": 5.05,
      "mean_token_accuracy": 0.18752111196517945,
      "num_tokens": 113216525.0,
      "step": 51675
    },
    {
      "entropy": 5.731749248504639,
      "epoch": 5.129019452163557,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002693032242587196,
      "loss": 4.9686,
      "mean_token_accuracy": 0.19697688817977904,
      "num_tokens": 113227902.0,
      "step": 51680
    },
    {
      "entropy": 5.748039817810058,
      "epoch": 5.129515680825724,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00026926780419194183,
      "loss": 4.9484,
      "mean_token_accuracy": 0.19140321910381317,
      "num_tokens": 113237576.0,
      "step": 51685
    },
    {
      "entropy": 5.6661402702331545,
      "epoch": 5.130011909487892,
      "grad_norm": 1.25,
      "learning_rate": 0.00026923238426731025,
      "loss": 4.8817,
      "mean_token_accuracy": 0.19454658031463623,
      "num_tokens": 113248470.0,
      "step": 51690
    },
    {
      "entropy": 5.68524398803711,
      "epoch": 5.130508138150059,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026919696448570303,
      "loss": 4.9358,
      "mean_token_accuracy": 0.19711555391550065,
      "num_tokens": 113259850.0,
      "step": 51695
    },
    {
      "entropy": 5.698882055282593,
      "epoch": 5.131004366812227,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002691615448479986,
      "loss": 4.9051,
      "mean_token_accuracy": 0.2020708665251732,
      "num_tokens": 113271192.0,
      "step": 51700
    },
    {
      "entropy": 5.678184127807617,
      "epoch": 5.131500595474394,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002691261253550752,
      "loss": 4.9336,
      "mean_token_accuracy": 0.19629232287406922,
      "num_tokens": 113281808.0,
      "step": 51705
    },
    {
      "entropy": 5.741545629501343,
      "epoch": 5.1319968241365626,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026909070600781134,
      "loss": 5.0409,
      "mean_token_accuracy": 0.18470851927995682,
      "num_tokens": 113294328.0,
      "step": 51710
    },
    {
      "entropy": 5.7238161087036135,
      "epoch": 5.13249305279873,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026905528680708517,
      "loss": 4.9617,
      "mean_token_accuracy": 0.19791832864284514,
      "num_tokens": 113303775.0,
      "step": 51715
    },
    {
      "entropy": 5.68429741859436,
      "epoch": 5.1329892814608975,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026901986775377515,
      "loss": 4.8394,
      "mean_token_accuracy": 0.19830746799707413,
      "num_tokens": 113313469.0,
      "step": 51720
    },
    {
      "entropy": 5.616963052749634,
      "epoch": 5.133485510123065,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002689844488487595,
      "loss": 4.8376,
      "mean_token_accuracy": 0.20660037249326707,
      "num_tokens": 113324260.0,
      "step": 51725
    },
    {
      "entropy": 5.689859819412232,
      "epoch": 5.1339817387852325,
      "grad_norm": 1.25,
      "learning_rate": 0.0002689490300929165,
      "loss": 4.921,
      "mean_token_accuracy": 0.20226403325796127,
      "num_tokens": 113335008.0,
      "step": 51730
    },
    {
      "entropy": 5.681119966506958,
      "epoch": 5.1344779674474,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026891361148712446,
      "loss": 4.9145,
      "mean_token_accuracy": 0.19683047384023666,
      "num_tokens": 113345330.0,
      "step": 51735
    },
    {
      "entropy": 5.6949718475341795,
      "epoch": 5.134974196109567,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026887819303226184,
      "loss": 4.9421,
      "mean_token_accuracy": 0.19698686003684998,
      "num_tokens": 113355444.0,
      "step": 51740
    },
    {
      "entropy": 5.741718769073486,
      "epoch": 5.135470424771735,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002688427747292069,
      "loss": 4.9237,
      "mean_token_accuracy": 0.1950722321867943,
      "num_tokens": 113366066.0,
      "step": 51745
    },
    {
      "entropy": 5.7622918605804445,
      "epoch": 5.135966653433902,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026880735657883783,
      "loss": 5.0101,
      "mean_token_accuracy": 0.1938110589981079,
      "num_tokens": 113377338.0,
      "step": 51750
    },
    {
      "entropy": 5.705008125305175,
      "epoch": 5.13646288209607,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002687719385820329,
      "loss": 4.9447,
      "mean_token_accuracy": 0.19547910392284393,
      "num_tokens": 113388925.0,
      "step": 51755
    },
    {
      "entropy": 5.698161649703979,
      "epoch": 5.136959110758237,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026873652073967056,
      "loss": 4.9883,
      "mean_token_accuracy": 0.20225582718849183,
      "num_tokens": 113400847.0,
      "step": 51760
    },
    {
      "entropy": 5.749912166595459,
      "epoch": 5.137455339420405,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026870110305262897,
      "loss": 4.9821,
      "mean_token_accuracy": 0.1945611923933029,
      "num_tokens": 113411171.0,
      "step": 51765
    },
    {
      "entropy": 5.755605173110962,
      "epoch": 5.137951568082572,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026866568552178643,
      "loss": 4.9521,
      "mean_token_accuracy": 0.20014881193637848,
      "num_tokens": 113421824.0,
      "step": 51770
    },
    {
      "entropy": 5.667949390411377,
      "epoch": 5.13844779674474,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026863026814802127,
      "loss": 4.858,
      "mean_token_accuracy": 0.20949605256319045,
      "num_tokens": 113432550.0,
      "step": 51775
    },
    {
      "entropy": 5.617610454559326,
      "epoch": 5.138944025406907,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002685948509322118,
      "loss": 4.9216,
      "mean_token_accuracy": 0.1959574967622757,
      "num_tokens": 113443477.0,
      "step": 51780
    },
    {
      "entropy": 5.702698993682861,
      "epoch": 5.139440254069075,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002685594338752361,
      "loss": 4.9437,
      "mean_token_accuracy": 0.19866181761026383,
      "num_tokens": 113454671.0,
      "step": 51785
    },
    {
      "entropy": 5.672717332839966,
      "epoch": 5.139936482731242,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002685240169779727,
      "loss": 4.9205,
      "mean_token_accuracy": 0.202023483812809,
      "num_tokens": 113467811.0,
      "step": 51790
    },
    {
      "entropy": 5.695372676849365,
      "epoch": 5.1404327113934105,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002684886002412996,
      "loss": 4.9583,
      "mean_token_accuracy": 0.18794353902339936,
      "num_tokens": 113477801.0,
      "step": 51795
    },
    {
      "entropy": 5.734488439559937,
      "epoch": 5.140928940055578,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002684531836660952,
      "loss": 4.9625,
      "mean_token_accuracy": 0.19266203939914703,
      "num_tokens": 113491073.0,
      "step": 51800
    },
    {
      "entropy": 5.644621896743774,
      "epoch": 5.1414251687177455,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026841776725323774,
      "loss": 4.9138,
      "mean_token_accuracy": 0.19837565422058107,
      "num_tokens": 113501884.0,
      "step": 51805
    },
    {
      "entropy": 5.64476637840271,
      "epoch": 5.141921397379913,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026838235100360553,
      "loss": 4.8355,
      "mean_token_accuracy": 0.199147729575634,
      "num_tokens": 113511834.0,
      "step": 51810
    },
    {
      "entropy": 5.63275179862976,
      "epoch": 5.14241762604208,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002683469349180767,
      "loss": 4.8391,
      "mean_token_accuracy": 0.20430358946323396,
      "num_tokens": 113523077.0,
      "step": 51815
    },
    {
      "entropy": 5.674518823623657,
      "epoch": 5.142913854704248,
      "grad_norm": 1.375,
      "learning_rate": 0.00026831151899752955,
      "loss": 4.8689,
      "mean_token_accuracy": 0.21055637896060944,
      "num_tokens": 113533848.0,
      "step": 51820
    },
    {
      "entropy": 5.681225299835205,
      "epoch": 5.143410083366415,
      "grad_norm": 1.375,
      "learning_rate": 0.00026827610324284235,
      "loss": 4.9355,
      "mean_token_accuracy": 0.1922462835907936,
      "num_tokens": 113544879.0,
      "step": 51825
    },
    {
      "entropy": 5.685839128494263,
      "epoch": 5.143906312028583,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002682406876548933,
      "loss": 5.0015,
      "mean_token_accuracy": 0.193099944293499,
      "num_tokens": 113557168.0,
      "step": 51830
    },
    {
      "entropy": 5.646259355545044,
      "epoch": 5.14440254069075,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002682052722345606,
      "loss": 4.8683,
      "mean_token_accuracy": 0.20198126137256622,
      "num_tokens": 113568544.0,
      "step": 51835
    },
    {
      "entropy": 5.69889464378357,
      "epoch": 5.144898769352918,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026816985698272254,
      "loss": 4.8952,
      "mean_token_accuracy": 0.19622718095779418,
      "num_tokens": 113579204.0,
      "step": 51840
    },
    {
      "entropy": 5.693043899536133,
      "epoch": 5.145394998015085,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002681344419002573,
      "loss": 4.9635,
      "mean_token_accuracy": 0.19961792081594468,
      "num_tokens": 113592428.0,
      "step": 51845
    },
    {
      "entropy": 5.777036333084107,
      "epoch": 5.145891226677253,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000268099026988043,
      "loss": 5.0609,
      "mean_token_accuracy": 0.1863499402999878,
      "num_tokens": 113603468.0,
      "step": 51850
    },
    {
      "entropy": 5.695863103866577,
      "epoch": 5.14638745533942,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026806361224695807,
      "loss": 4.9899,
      "mean_token_accuracy": 0.18965326249599457,
      "num_tokens": 113614180.0,
      "step": 51855
    },
    {
      "entropy": 5.717454481124878,
      "epoch": 5.146883684001588,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002680281976778806,
      "loss": 4.9275,
      "mean_token_accuracy": 0.19760441333055495,
      "num_tokens": 113624274.0,
      "step": 51860
    },
    {
      "entropy": 5.67931866645813,
      "epoch": 5.147379912663755,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002679927832816888,
      "loss": 4.868,
      "mean_token_accuracy": 0.20358744114637375,
      "num_tokens": 113634433.0,
      "step": 51865
    },
    {
      "entropy": 5.637179565429688,
      "epoch": 5.147876141325923,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00026795736905926094,
      "loss": 4.8531,
      "mean_token_accuracy": 0.2033387690782547,
      "num_tokens": 113645329.0,
      "step": 51870
    },
    {
      "entropy": 5.712809467315674,
      "epoch": 5.14837236998809,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026792195501147513,
      "loss": 4.9853,
      "mean_token_accuracy": 0.19718911051750182,
      "num_tokens": 113655497.0,
      "step": 51875
    },
    {
      "entropy": 5.702932119369507,
      "epoch": 5.1488685986502585,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026788654113920945,
      "loss": 4.9213,
      "mean_token_accuracy": 0.19824561178684236,
      "num_tokens": 113667544.0,
      "step": 51880
    },
    {
      "entropy": 5.720713996887207,
      "epoch": 5.149364827312426,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026785112744334233,
      "loss": 4.9697,
      "mean_token_accuracy": 0.1991900086402893,
      "num_tokens": 113677467.0,
      "step": 51885
    },
    {
      "entropy": 5.714111423492431,
      "epoch": 5.149861055974593,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026781571392475186,
      "loss": 4.9389,
      "mean_token_accuracy": 0.2056644693017006,
      "num_tokens": 113687456.0,
      "step": 51890
    },
    {
      "entropy": 5.642160272598266,
      "epoch": 5.150357284636761,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002677803005843162,
      "loss": 4.9239,
      "mean_token_accuracy": 0.1973515883088112,
      "num_tokens": 113698081.0,
      "step": 51895
    },
    {
      "entropy": 5.721411561965942,
      "epoch": 5.150853513298928,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026774488742291346,
      "loss": 4.9168,
      "mean_token_accuracy": 0.20010194629430772,
      "num_tokens": 113708705.0,
      "step": 51900
    },
    {
      "entropy": 5.676312732696533,
      "epoch": 5.151349741961096,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026770947444142195,
      "loss": 4.8622,
      "mean_token_accuracy": 0.21061503440141677,
      "num_tokens": 113720231.0,
      "step": 51905
    },
    {
      "entropy": 5.743904876708984,
      "epoch": 5.151845970623263,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002676740616407197,
      "loss": 4.9969,
      "mean_token_accuracy": 0.19754922538995742,
      "num_tokens": 113732089.0,
      "step": 51910
    },
    {
      "entropy": 5.702093887329101,
      "epoch": 5.152342199285431,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002676386490216849,
      "loss": 4.9659,
      "mean_token_accuracy": 0.1905221551656723,
      "num_tokens": 113743594.0,
      "step": 51915
    },
    {
      "entropy": 5.685471296310425,
      "epoch": 5.152838427947598,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002676032365851959,
      "loss": 4.9231,
      "mean_token_accuracy": 0.19748923927545547,
      "num_tokens": 113755396.0,
      "step": 51920
    },
    {
      "entropy": 5.672514295578003,
      "epoch": 5.153334656609766,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002675678243321306,
      "loss": 4.9044,
      "mean_token_accuracy": 0.2028145208954811,
      "num_tokens": 113765913.0,
      "step": 51925
    },
    {
      "entropy": 5.7518799781799315,
      "epoch": 5.153830885271933,
      "grad_norm": 1.25,
      "learning_rate": 0.0002675324122633673,
      "loss": 4.9643,
      "mean_token_accuracy": 0.1922922372817993,
      "num_tokens": 113775875.0,
      "step": 51930
    },
    {
      "entropy": 5.696818876266479,
      "epoch": 5.154327113934101,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026749700037978404,
      "loss": 4.9726,
      "mean_token_accuracy": 0.1882768914103508,
      "num_tokens": 113785976.0,
      "step": 51935
    },
    {
      "entropy": 5.689151477813721,
      "epoch": 5.154823342596268,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026746158868225895,
      "loss": 4.9038,
      "mean_token_accuracy": 0.206154166162014,
      "num_tokens": 113795939.0,
      "step": 51940
    },
    {
      "entropy": 5.7138909816741945,
      "epoch": 5.155319571258436,
      "grad_norm": 1.25,
      "learning_rate": 0.00026742617717167026,
      "loss": 4.9652,
      "mean_token_accuracy": 0.20507194250822067,
      "num_tokens": 113808297.0,
      "step": 51945
    },
    {
      "entropy": 5.672443914413452,
      "epoch": 5.155815799920603,
      "grad_norm": 1.265625,
      "learning_rate": 0.000267390765848896,
      "loss": 4.8683,
      "mean_token_accuracy": 0.2023366168141365,
      "num_tokens": 113820099.0,
      "step": 51950
    },
    {
      "entropy": 5.708996295928955,
      "epoch": 5.156312028582771,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002673553547148145,
      "loss": 4.8961,
      "mean_token_accuracy": 0.1989888146519661,
      "num_tokens": 113830349.0,
      "step": 51955
    },
    {
      "entropy": 5.712614870071411,
      "epoch": 5.156808257244938,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002673199437703036,
      "loss": 4.9639,
      "mean_token_accuracy": 0.19229777455329894,
      "num_tokens": 113841145.0,
      "step": 51960
    },
    {
      "entropy": 5.643172073364258,
      "epoch": 5.157304485907106,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002672845330162416,
      "loss": 4.921,
      "mean_token_accuracy": 0.19413414895534514,
      "num_tokens": 113853321.0,
      "step": 51965
    },
    {
      "entropy": 5.7831260681152346,
      "epoch": 5.157800714569274,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002672491224535065,
      "loss": 5.0177,
      "mean_token_accuracy": 0.19335806220769883,
      "num_tokens": 113863832.0,
      "step": 51970
    },
    {
      "entropy": 5.700727891921997,
      "epoch": 5.158296943231441,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002672137120829765,
      "loss": 4.9337,
      "mean_token_accuracy": 0.19523469358682632,
      "num_tokens": 113874221.0,
      "step": 51975
    },
    {
      "entropy": 5.626764726638794,
      "epoch": 5.158793171893609,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002671783019055297,
      "loss": 4.8395,
      "mean_token_accuracy": 0.20517993122339248,
      "num_tokens": 113884458.0,
      "step": 51980
    },
    {
      "entropy": 5.680896043777466,
      "epoch": 5.159289400555776,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002671428919220441,
      "loss": 4.8918,
      "mean_token_accuracy": 0.2109692096710205,
      "num_tokens": 113897015.0,
      "step": 51985
    },
    {
      "entropy": 5.711483812332153,
      "epoch": 5.159785629217944,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002671074821333979,
      "loss": 4.9429,
      "mean_token_accuracy": 0.1995614156126976,
      "num_tokens": 113907148.0,
      "step": 51990
    },
    {
      "entropy": 5.69979043006897,
      "epoch": 5.160281857880111,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002670720725404691,
      "loss": 4.9496,
      "mean_token_accuracy": 0.20006020218133927,
      "num_tokens": 113917706.0,
      "step": 51995
    },
    {
      "entropy": 5.713897323608398,
      "epoch": 5.160778086542279,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026703666314413575,
      "loss": 4.9422,
      "mean_token_accuracy": 0.19264640063047409,
      "num_tokens": 113928472.0,
      "step": 52000
    },
    {
      "entropy": 5.763761758804321,
      "epoch": 5.161274315204446,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026700125394527616,
      "loss": 4.9826,
      "mean_token_accuracy": 0.19214327931404113,
      "num_tokens": 113938834.0,
      "step": 52005
    },
    {
      "entropy": 5.715460968017578,
      "epoch": 5.161770543866614,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026696584494476815,
      "loss": 4.8422,
      "mean_token_accuracy": 0.20016802251338958,
      "num_tokens": 113948296.0,
      "step": 52010
    },
    {
      "entropy": 5.743984699249268,
      "epoch": 5.162266772528781,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026693043614348983,
      "loss": 5.0218,
      "mean_token_accuracy": 0.19129240661859512,
      "num_tokens": 113958525.0,
      "step": 52015
    },
    {
      "entropy": 5.7550116062164305,
      "epoch": 5.162763001190949,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002668950275423194,
      "loss": 5.033,
      "mean_token_accuracy": 0.19224261343479157,
      "num_tokens": 113969354.0,
      "step": 52020
    },
    {
      "entropy": 5.731942176818848,
      "epoch": 5.163259229853116,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026685961914213474,
      "loss": 4.9426,
      "mean_token_accuracy": 0.19215475171804428,
      "num_tokens": 113980948.0,
      "step": 52025
    },
    {
      "entropy": 5.685930681228638,
      "epoch": 5.163755458515284,
      "grad_norm": 1.328125,
      "learning_rate": 0.000266824210943814,
      "loss": 4.884,
      "mean_token_accuracy": 0.20360352247953414,
      "num_tokens": 113990600.0,
      "step": 52030
    },
    {
      "entropy": 5.684726285934448,
      "epoch": 5.164251687177451,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002667888029482352,
      "loss": 4.8624,
      "mean_token_accuracy": 0.20195620656013488,
      "num_tokens": 114002734.0,
      "step": 52035
    },
    {
      "entropy": 5.711514949798584,
      "epoch": 5.164747915839619,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026675339515627646,
      "loss": 4.9996,
      "mean_token_accuracy": 0.191475448012352,
      "num_tokens": 114014160.0,
      "step": 52040
    },
    {
      "entropy": 5.741572666168213,
      "epoch": 5.165244144501786,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002667179875688157,
      "loss": 4.9393,
      "mean_token_accuracy": 0.19052583128213882,
      "num_tokens": 114023401.0,
      "step": 52045
    },
    {
      "entropy": 5.695638990402221,
      "epoch": 5.165740373163954,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026668258018673103,
      "loss": 4.9024,
      "mean_token_accuracy": 0.1981418699026108,
      "num_tokens": 114035211.0,
      "step": 52050
    },
    {
      "entropy": 5.6701493740081785,
      "epoch": 5.166236601826122,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002666471730109005,
      "loss": 4.863,
      "mean_token_accuracy": 0.20477278977632524,
      "num_tokens": 114046171.0,
      "step": 52055
    },
    {
      "entropy": 5.678889513015747,
      "epoch": 5.166732830488289,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002666117660422019,
      "loss": 4.9276,
      "mean_token_accuracy": 0.1919933721423149,
      "num_tokens": 114056872.0,
      "step": 52060
    },
    {
      "entropy": 5.640029096603394,
      "epoch": 5.167229059150457,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00026657635928151353,
      "loss": 4.8541,
      "mean_token_accuracy": 0.19917783737182618,
      "num_tokens": 114068051.0,
      "step": 52065
    },
    {
      "entropy": 5.717027997970581,
      "epoch": 5.167725287812624,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002665409527297133,
      "loss": 5.0124,
      "mean_token_accuracy": 0.1892006054520607,
      "num_tokens": 114081625.0,
      "step": 52070
    },
    {
      "entropy": 5.652507543563843,
      "epoch": 5.168221516474792,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026650554638767924,
      "loss": 4.8951,
      "mean_token_accuracy": 0.19859937727451324,
      "num_tokens": 114092631.0,
      "step": 52075
    },
    {
      "entropy": 5.695351457595825,
      "epoch": 5.168717745136959,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026647014025628937,
      "loss": 4.9973,
      "mean_token_accuracy": 0.19596356004476548,
      "num_tokens": 114104199.0,
      "step": 52080
    },
    {
      "entropy": 5.7754639148712155,
      "epoch": 5.169213973799127,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026643473433642144,
      "loss": 5.0347,
      "mean_token_accuracy": 0.191243976354599,
      "num_tokens": 114115080.0,
      "step": 52085
    },
    {
      "entropy": 5.762033414840698,
      "epoch": 5.169710202461294,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002663993286289538,
      "loss": 4.9538,
      "mean_token_accuracy": 0.193619704246521,
      "num_tokens": 114125787.0,
      "step": 52090
    },
    {
      "entropy": 5.690714597702026,
      "epoch": 5.170206431123462,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026636392313476425,
      "loss": 4.8931,
      "mean_token_accuracy": 0.19811396598815917,
      "num_tokens": 114136953.0,
      "step": 52095
    },
    {
      "entropy": 5.6820320129394535,
      "epoch": 5.170702659785629,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002663285178547309,
      "loss": 4.8809,
      "mean_token_accuracy": 0.2021288111805916,
      "num_tokens": 114147434.0,
      "step": 52100
    },
    {
      "entropy": 5.641103315353393,
      "epoch": 5.171198888447797,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002662931127897315,
      "loss": 4.7836,
      "mean_token_accuracy": 0.21052864342927932,
      "num_tokens": 114157883.0,
      "step": 52105
    },
    {
      "entropy": 5.682454919815063,
      "epoch": 5.171695117109964,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002662577079406442,
      "loss": 4.9143,
      "mean_token_accuracy": 0.19600823372602463,
      "num_tokens": 114168297.0,
      "step": 52110
    },
    {
      "entropy": 5.609274673461914,
      "epoch": 5.172191345772132,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002662223033083468,
      "loss": 4.9164,
      "mean_token_accuracy": 0.19653188288211823,
      "num_tokens": 114180927.0,
      "step": 52115
    },
    {
      "entropy": 5.659705114364624,
      "epoch": 5.172687574434299,
      "grad_norm": 1.375,
      "learning_rate": 0.00026618689889371745,
      "loss": 4.9514,
      "mean_token_accuracy": 0.19262132048606873,
      "num_tokens": 114191388.0,
      "step": 52120
    },
    {
      "entropy": 5.743965578079224,
      "epoch": 5.1731838030964665,
      "grad_norm": 1.28125,
      "learning_rate": 0.000266151494697634,
      "loss": 4.9529,
      "mean_token_accuracy": 0.19910614490509032,
      "num_tokens": 114202346.0,
      "step": 52125
    },
    {
      "entropy": 5.630234146118164,
      "epoch": 5.173680031758634,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002661160907209744,
      "loss": 4.7897,
      "mean_token_accuracy": 0.2115035369992256,
      "num_tokens": 114212958.0,
      "step": 52130
    },
    {
      "entropy": 5.70890622138977,
      "epoch": 5.1741762604208015,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002660806869646167,
      "loss": 4.9197,
      "mean_token_accuracy": 0.1954601839184761,
      "num_tokens": 114223145.0,
      "step": 52135
    },
    {
      "entropy": 5.658488082885742,
      "epoch": 5.17467248908297,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026604528342943863,
      "loss": 4.9172,
      "mean_token_accuracy": 0.2015857845544815,
      "num_tokens": 114234139.0,
      "step": 52140
    },
    {
      "entropy": 5.691549634933471,
      "epoch": 5.175168717745137,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002660098801163182,
      "loss": 4.9146,
      "mean_token_accuracy": 0.20154043436050414,
      "num_tokens": 114244006.0,
      "step": 52145
    },
    {
      "entropy": 5.680216455459595,
      "epoch": 5.175664946407305,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002659744770261335,
      "loss": 4.8829,
      "mean_token_accuracy": 0.20770236998796462,
      "num_tokens": 114254520.0,
      "step": 52150
    },
    {
      "entropy": 5.73703236579895,
      "epoch": 5.176161175069472,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026593907415976224,
      "loss": 4.9476,
      "mean_token_accuracy": 0.20182544440031053,
      "num_tokens": 114264507.0,
      "step": 52155
    },
    {
      "entropy": 5.652183580398559,
      "epoch": 5.17665740373164,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026590367151808246,
      "loss": 4.8701,
      "mean_token_accuracy": 0.20572776198387147,
      "num_tokens": 114274918.0,
      "step": 52160
    },
    {
      "entropy": 5.679885673522949,
      "epoch": 5.177153632393807,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026586826910197204,
      "loss": 4.9346,
      "mean_token_accuracy": 0.1960763469338417,
      "num_tokens": 114285354.0,
      "step": 52165
    },
    {
      "entropy": 5.682433557510376,
      "epoch": 5.177649861055975,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002658328669123089,
      "loss": 4.8635,
      "mean_token_accuracy": 0.19868091642856597,
      "num_tokens": 114295754.0,
      "step": 52170
    },
    {
      "entropy": 5.708004665374756,
      "epoch": 5.178146089718142,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026579746494997074,
      "loss": 4.9861,
      "mean_token_accuracy": 0.1990187421441078,
      "num_tokens": 114306663.0,
      "step": 52175
    },
    {
      "entropy": 5.693764925003052,
      "epoch": 5.17864231838031,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026576206321583573,
      "loss": 4.8998,
      "mean_token_accuracy": 0.1957993686199188,
      "num_tokens": 114317430.0,
      "step": 52180
    },
    {
      "entropy": 5.757930898666382,
      "epoch": 5.179138547042477,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026572666171078164,
      "loss": 4.9976,
      "mean_token_accuracy": 0.19276351630687713,
      "num_tokens": 114328874.0,
      "step": 52185
    },
    {
      "entropy": 5.735913467407227,
      "epoch": 5.179634775704645,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002656912604356864,
      "loss": 4.9147,
      "mean_token_accuracy": 0.19927270263433455,
      "num_tokens": 114340130.0,
      "step": 52190
    },
    {
      "entropy": 5.686321449279785,
      "epoch": 5.180131004366812,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026565585939142776,
      "loss": 4.9223,
      "mean_token_accuracy": 0.2034304305911064,
      "num_tokens": 114349969.0,
      "step": 52195
    },
    {
      "entropy": 5.715677642822266,
      "epoch": 5.1806272330289795,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002656204585788838,
      "loss": 4.9815,
      "mean_token_accuracy": 0.19483073353767394,
      "num_tokens": 114361611.0,
      "step": 52200
    },
    {
      "entropy": 5.7127406120300295,
      "epoch": 5.181123461691147,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002655850579989321,
      "loss": 4.9603,
      "mean_token_accuracy": 0.20224171727895737,
      "num_tokens": 114373164.0,
      "step": 52205
    },
    {
      "entropy": 5.764982032775879,
      "epoch": 5.1816196903533145,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002655496576524507,
      "loss": 4.9523,
      "mean_token_accuracy": 0.1888861522078514,
      "num_tokens": 114383294.0,
      "step": 52210
    },
    {
      "entropy": 5.748249292373657,
      "epoch": 5.182115919015482,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002655142575403175,
      "loss": 4.9245,
      "mean_token_accuracy": 0.20106308162212372,
      "num_tokens": 114393391.0,
      "step": 52215
    },
    {
      "entropy": 5.656456565856933,
      "epoch": 5.182612147677649,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026547885766341027,
      "loss": 4.8704,
      "mean_token_accuracy": 0.19959028214216232,
      "num_tokens": 114402646.0,
      "step": 52220
    },
    {
      "entropy": 5.7776275157928465,
      "epoch": 5.183108376339818,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026544345802260685,
      "loss": 5.1298,
      "mean_token_accuracy": 0.18670981973409653,
      "num_tokens": 114413979.0,
      "step": 52225
    },
    {
      "entropy": 5.733904838562012,
      "epoch": 5.183604605001985,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026540805861878503,
      "loss": 4.9936,
      "mean_token_accuracy": 0.19706255495548247,
      "num_tokens": 114423810.0,
      "step": 52230
    },
    {
      "entropy": 5.6497210502624515,
      "epoch": 5.184100833664153,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002653726594528227,
      "loss": 4.9178,
      "mean_token_accuracy": 0.20101001858711243,
      "num_tokens": 114434341.0,
      "step": 52235
    },
    {
      "entropy": 5.713247966766358,
      "epoch": 5.18459706232632,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002653372605255977,
      "loss": 4.9884,
      "mean_token_accuracy": 0.18918535858392715,
      "num_tokens": 114445408.0,
      "step": 52240
    },
    {
      "entropy": 5.703995084762573,
      "epoch": 5.185093290988488,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026530186183798776,
      "loss": 4.9233,
      "mean_token_accuracy": 0.18826790601015092,
      "num_tokens": 114456263.0,
      "step": 52245
    },
    {
      "entropy": 5.670274448394776,
      "epoch": 5.185589519650655,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002652664633908708,
      "loss": 4.9197,
      "mean_token_accuracy": 0.19591300189495087,
      "num_tokens": 114467490.0,
      "step": 52250
    },
    {
      "entropy": 5.764534378051758,
      "epoch": 5.186085748312823,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002652310651851246,
      "loss": 4.9414,
      "mean_token_accuracy": 0.19412774741649627,
      "num_tokens": 114477884.0,
      "step": 52255
    },
    {
      "entropy": 5.723029851913452,
      "epoch": 5.18658197697499,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002651956672216269,
      "loss": 4.9484,
      "mean_token_accuracy": 0.19816687405109407,
      "num_tokens": 114488204.0,
      "step": 52260
    },
    {
      "entropy": 5.734092569351196,
      "epoch": 5.187078205637158,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002651602695012554,
      "loss": 4.9994,
      "mean_token_accuracy": 0.19100505560636521,
      "num_tokens": 114499645.0,
      "step": 52265
    },
    {
      "entropy": 5.739578771591186,
      "epoch": 5.187574434299325,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002651248720248882,
      "loss": 4.9873,
      "mean_token_accuracy": 0.1853511705994606,
      "num_tokens": 114509672.0,
      "step": 52270
    },
    {
      "entropy": 5.6748405456542965,
      "epoch": 5.1880706629614926,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002650894747934028,
      "loss": 4.9822,
      "mean_token_accuracy": 0.19877661764621735,
      "num_tokens": 114520622.0,
      "step": 52275
    },
    {
      "entropy": 5.6969929218292235,
      "epoch": 5.18856689162366,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026505407780767716,
      "loss": 4.9092,
      "mean_token_accuracy": 0.20274868309497834,
      "num_tokens": 114532946.0,
      "step": 52280
    },
    {
      "entropy": 5.691100311279297,
      "epoch": 5.1890631202858275,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002650186810685889,
      "loss": 4.8744,
      "mean_token_accuracy": 0.20429512709379197,
      "num_tokens": 114543601.0,
      "step": 52285
    },
    {
      "entropy": 5.658070421218872,
      "epoch": 5.189559348947995,
      "grad_norm": 1.25,
      "learning_rate": 0.0002649832845770158,
      "loss": 4.8833,
      "mean_token_accuracy": 0.2054446369409561,
      "num_tokens": 114554445.0,
      "step": 52290
    },
    {
      "entropy": 5.690872859954834,
      "epoch": 5.1900555776101625,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026494788833383566,
      "loss": 4.9675,
      "mean_token_accuracy": 0.19262619614601134,
      "num_tokens": 114564303.0,
      "step": 52295
    },
    {
      "entropy": 5.675734281539917,
      "epoch": 5.19055180627233,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026491249233992633,
      "loss": 4.9812,
      "mean_token_accuracy": 0.19443232715129852,
      "num_tokens": 114576183.0,
      "step": 52300
    },
    {
      "entropy": 5.736557054519653,
      "epoch": 5.191048034934497,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00026487709659616537,
      "loss": 4.9576,
      "mean_token_accuracy": 0.20898631364107131,
      "num_tokens": 114588559.0,
      "step": 52305
    },
    {
      "entropy": 5.760212707519531,
      "epoch": 5.191544263596666,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00026484170110343066,
      "loss": 4.9574,
      "mean_token_accuracy": 0.19685906767845154,
      "num_tokens": 114600432.0,
      "step": 52310
    },
    {
      "entropy": 5.714791202545166,
      "epoch": 5.192040492258833,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026480630586259987,
      "loss": 4.8926,
      "mean_token_accuracy": 0.2032101571559906,
      "num_tokens": 114611667.0,
      "step": 52315
    },
    {
      "entropy": 5.71402153968811,
      "epoch": 5.192536720921001,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002647709108745506,
      "loss": 5.0093,
      "mean_token_accuracy": 0.19197873622179032,
      "num_tokens": 114622923.0,
      "step": 52320
    },
    {
      "entropy": 5.707777214050293,
      "epoch": 5.193032949583168,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026473551614016077,
      "loss": 5.0443,
      "mean_token_accuracy": 0.18927768170833587,
      "num_tokens": 114635125.0,
      "step": 52325
    },
    {
      "entropy": 5.695626735687256,
      "epoch": 5.193529178245336,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002647001216603081,
      "loss": 4.9306,
      "mean_token_accuracy": 0.19383531361818312,
      "num_tokens": 114645807.0,
      "step": 52330
    },
    {
      "entropy": 5.730708646774292,
      "epoch": 5.194025406907503,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002646647274358702,
      "loss": 4.9797,
      "mean_token_accuracy": 0.18523644804954528,
      "num_tokens": 114656231.0,
      "step": 52335
    },
    {
      "entropy": 5.663681316375732,
      "epoch": 5.194521635569671,
      "grad_norm": 1.25,
      "learning_rate": 0.00026462933346772474,
      "loss": 4.8563,
      "mean_token_accuracy": 0.1974518418312073,
      "num_tokens": 114667265.0,
      "step": 52340
    },
    {
      "entropy": 5.685415124893188,
      "epoch": 5.195017864231838,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002645939397567495,
      "loss": 4.8871,
      "mean_token_accuracy": 0.19779901653528215,
      "num_tokens": 114678436.0,
      "step": 52345
    },
    {
      "entropy": 5.695677185058594,
      "epoch": 5.195514092894006,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026455854630382203,
      "loss": 4.9788,
      "mean_token_accuracy": 0.1893698453903198,
      "num_tokens": 114690678.0,
      "step": 52350
    },
    {
      "entropy": 5.780515050888061,
      "epoch": 5.196010321556173,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026452315310982025,
      "loss": 4.96,
      "mean_token_accuracy": 0.19123997539281845,
      "num_tokens": 114700739.0,
      "step": 52355
    },
    {
      "entropy": 5.7671126365661625,
      "epoch": 5.1965065502183405,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002644877601756216,
      "loss": 4.9879,
      "mean_token_accuracy": 0.19023414254188536,
      "num_tokens": 114711548.0,
      "step": 52360
    },
    {
      "entropy": 5.725074529647827,
      "epoch": 5.197002778880508,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026445236750210387,
      "loss": 4.9236,
      "mean_token_accuracy": 0.19188813865184784,
      "num_tokens": 114721931.0,
      "step": 52365
    },
    {
      "entropy": 5.610989665985107,
      "epoch": 5.1974990075426755,
      "grad_norm": 1.25,
      "learning_rate": 0.0002644169750901447,
      "loss": 4.76,
      "mean_token_accuracy": 0.20966853350400924,
      "num_tokens": 114732705.0,
      "step": 52370
    },
    {
      "entropy": 5.75299825668335,
      "epoch": 5.197995236204843,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026438158294062183,
      "loss": 5.0124,
      "mean_token_accuracy": 0.19137255996465682,
      "num_tokens": 114744695.0,
      "step": 52375
    },
    {
      "entropy": 5.7455237865447994,
      "epoch": 5.19849146486701,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002643461910544127,
      "loss": 5.0186,
      "mean_token_accuracy": 0.1904127448797226,
      "num_tokens": 114754669.0,
      "step": 52380
    },
    {
      "entropy": 5.796501636505127,
      "epoch": 5.198987693529178,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002643107994323951,
      "loss": 5.0042,
      "mean_token_accuracy": 0.19516348242759704,
      "num_tokens": 114766260.0,
      "step": 52385
    },
    {
      "entropy": 5.761862707138062,
      "epoch": 5.199483922191345,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026427540807544666,
      "loss": 4.9663,
      "mean_token_accuracy": 0.1899358332157135,
      "num_tokens": 114776815.0,
      "step": 52390
    },
    {
      "entropy": 5.7531153678894045,
      "epoch": 5.199980150853514,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000264240016984445,
      "loss": 4.9787,
      "mean_token_accuracy": 0.19389090389013292,
      "num_tokens": 114787761.0,
      "step": 52395
    },
    {
      "entropy": 5.677598285675049,
      "epoch": 5.200476379515681,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026420462616026777,
      "loss": 4.9373,
      "mean_token_accuracy": 0.19539234638214112,
      "num_tokens": 114798612.0,
      "step": 52400
    },
    {
      "entropy": 5.649934911727906,
      "epoch": 5.200972608177849,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002641692356037925,
      "loss": 4.9226,
      "mean_token_accuracy": 0.1912025898694992,
      "num_tokens": 114810102.0,
      "step": 52405
    },
    {
      "entropy": 5.740754508972168,
      "epoch": 5.201468836840016,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002641338453158968,
      "loss": 5.0141,
      "mean_token_accuracy": 0.19117387235164643,
      "num_tokens": 114823000.0,
      "step": 52410
    },
    {
      "entropy": 5.784040355682373,
      "epoch": 5.201965065502184,
      "grad_norm": 1.25,
      "learning_rate": 0.0002640984552974584,
      "loss": 4.9862,
      "mean_token_accuracy": 0.19315257519483567,
      "num_tokens": 114833226.0,
      "step": 52415
    },
    {
      "entropy": 5.683459329605102,
      "epoch": 5.202461294164351,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026406306554935477,
      "loss": 4.9268,
      "mean_token_accuracy": 0.19416276961565018,
      "num_tokens": 114846127.0,
      "step": 52420
    },
    {
      "entropy": 5.607500171661377,
      "epoch": 5.202957522826519,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002640276760724636,
      "loss": 4.8891,
      "mean_token_accuracy": 0.19992706775665284,
      "num_tokens": 114856921.0,
      "step": 52425
    },
    {
      "entropy": 5.754405879974366,
      "epoch": 5.203453751488686,
      "grad_norm": 1.375,
      "learning_rate": 0.0002639922868676624,
      "loss": 5.0166,
      "mean_token_accuracy": 0.19762951731681824,
      "num_tokens": 114869240.0,
      "step": 52430
    },
    {
      "entropy": 5.688543844223022,
      "epoch": 5.2039499801508535,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002639568979358287,
      "loss": 4.8997,
      "mean_token_accuracy": 0.2038600489497185,
      "num_tokens": 114881300.0,
      "step": 52435
    },
    {
      "entropy": 5.738577508926392,
      "epoch": 5.204446208813021,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002639215092778402,
      "loss": 4.9632,
      "mean_token_accuracy": 0.20533775091171264,
      "num_tokens": 114893598.0,
      "step": 52440
    },
    {
      "entropy": 5.732811641693115,
      "epoch": 5.2049424374751885,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002638861208945744,
      "loss": 4.9006,
      "mean_token_accuracy": 0.19739169627428055,
      "num_tokens": 114904096.0,
      "step": 52445
    },
    {
      "entropy": 5.660459804534912,
      "epoch": 5.205438666137356,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002638507327869089,
      "loss": 4.8978,
      "mean_token_accuracy": 0.20030431747436522,
      "num_tokens": 114914541.0,
      "step": 52450
    },
    {
      "entropy": 5.599860668182373,
      "epoch": 5.205934894799523,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026381534495572114,
      "loss": 4.8437,
      "mean_token_accuracy": 0.19952932596206666,
      "num_tokens": 114924989.0,
      "step": 52455
    },
    {
      "entropy": 5.685359764099121,
      "epoch": 5.206431123461691,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00026377995740188874,
      "loss": 4.9264,
      "mean_token_accuracy": 0.2020778998732567,
      "num_tokens": 114938563.0,
      "step": 52460
    },
    {
      "entropy": 5.655310153961182,
      "epoch": 5.206927352123858,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002637445701262892,
      "loss": 4.9047,
      "mean_token_accuracy": 0.19690422117710113,
      "num_tokens": 114949955.0,
      "step": 52465
    },
    {
      "entropy": 5.7139967918396,
      "epoch": 5.207423580786026,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002637091831298,
      "loss": 4.9072,
      "mean_token_accuracy": 0.19938320517539979,
      "num_tokens": 114960245.0,
      "step": 52470
    },
    {
      "entropy": 5.718798828125,
      "epoch": 5.207919809448193,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026367379641329883,
      "loss": 4.9809,
      "mean_token_accuracy": 0.18926669806241989,
      "num_tokens": 114970025.0,
      "step": 52475
    },
    {
      "entropy": 5.699759387969971,
      "epoch": 5.208416038110361,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002636384099776631,
      "loss": 4.9583,
      "mean_token_accuracy": 0.19470982402563095,
      "num_tokens": 114980682.0,
      "step": 52480
    },
    {
      "entropy": 5.652218341827393,
      "epoch": 5.208912266772529,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026360302382377027,
      "loss": 4.895,
      "mean_token_accuracy": 0.19943245947360994,
      "num_tokens": 114991234.0,
      "step": 52485
    },
    {
      "entropy": 5.688083028793335,
      "epoch": 5.209408495434697,
      "grad_norm": 1.234375,
      "learning_rate": 0.000263567637952498,
      "loss": 4.9385,
      "mean_token_accuracy": 0.19240195751190187,
      "num_tokens": 115002694.0,
      "step": 52490
    },
    {
      "entropy": 5.742900943756103,
      "epoch": 5.209904724096864,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002635322523647236,
      "loss": 4.9949,
      "mean_token_accuracy": 0.19239706546068192,
      "num_tokens": 115014280.0,
      "step": 52495
    },
    {
      "entropy": 5.648763990402221,
      "epoch": 5.210400952759032,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026349686706132454,
      "loss": 4.8809,
      "mean_token_accuracy": 0.2007625252008438,
      "num_tokens": 115025652.0,
      "step": 52500
    },
    {
      "entropy": 5.719815587997436,
      "epoch": 5.210897181421199,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002634614820431785,
      "loss": 4.9468,
      "mean_token_accuracy": 0.189817214012146,
      "num_tokens": 115035981.0,
      "step": 52505
    },
    {
      "entropy": 5.741060400009156,
      "epoch": 5.2113934100833665,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026342609731116285,
      "loss": 4.9472,
      "mean_token_accuracy": 0.1966268911957741,
      "num_tokens": 115046684.0,
      "step": 52510
    },
    {
      "entropy": 5.704141998291016,
      "epoch": 5.211889638745534,
      "grad_norm": 1.3125,
      "learning_rate": 0.000263390712866155,
      "loss": 4.9149,
      "mean_token_accuracy": 0.19894082844257355,
      "num_tokens": 115057891.0,
      "step": 52515
    },
    {
      "entropy": 5.71467957496643,
      "epoch": 5.2123858674077015,
      "grad_norm": 1.4375,
      "learning_rate": 0.00026335532870903243,
      "loss": 4.9724,
      "mean_token_accuracy": 0.19646545201539994,
      "num_tokens": 115067818.0,
      "step": 52520
    },
    {
      "entropy": 5.697423696517944,
      "epoch": 5.212882096069869,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026331994484067275,
      "loss": 4.9104,
      "mean_token_accuracy": 0.2039165586233139,
      "num_tokens": 115079357.0,
      "step": 52525
    },
    {
      "entropy": 5.662748718261719,
      "epoch": 5.213378324732036,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002632845612619531,
      "loss": 4.8876,
      "mean_token_accuracy": 0.20260397493839263,
      "num_tokens": 115089266.0,
      "step": 52530
    },
    {
      "entropy": 5.683336162567139,
      "epoch": 5.213874553394204,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026324917797375115,
      "loss": 4.9467,
      "mean_token_accuracy": 0.1950573056936264,
      "num_tokens": 115100081.0,
      "step": 52535
    },
    {
      "entropy": 5.694806051254273,
      "epoch": 5.214370782056371,
      "grad_norm": 1.484375,
      "learning_rate": 0.00026321379497694423,
      "loss": 4.9467,
      "mean_token_accuracy": 0.19556639790534974,
      "num_tokens": 115110786.0,
      "step": 52540
    },
    {
      "entropy": 5.694215250015259,
      "epoch": 5.214867010718539,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026317841227241,
      "loss": 4.9873,
      "mean_token_accuracy": 0.19427194595336914,
      "num_tokens": 115121519.0,
      "step": 52545
    },
    {
      "entropy": 5.763526916503906,
      "epoch": 5.215363239380706,
      "grad_norm": 1.375,
      "learning_rate": 0.00026314302986102546,
      "loss": 4.9418,
      "mean_token_accuracy": 0.19570097029209138,
      "num_tokens": 115133148.0,
      "step": 52550
    },
    {
      "entropy": 5.698116636276245,
      "epoch": 5.215859468042874,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002631076477436683,
      "loss": 5.0287,
      "mean_token_accuracy": 0.19409075230360032,
      "num_tokens": 115143859.0,
      "step": 52555
    },
    {
      "entropy": 5.764376640319824,
      "epoch": 5.216355696705041,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026307226592121576,
      "loss": 5.0621,
      "mean_token_accuracy": 0.19102512747049333,
      "num_tokens": 115154803.0,
      "step": 52560
    },
    {
      "entropy": 5.788033437728882,
      "epoch": 5.216851925367209,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002630368843945454,
      "loss": 4.9376,
      "mean_token_accuracy": 0.19962542057037352,
      "num_tokens": 115167551.0,
      "step": 52565
    },
    {
      "entropy": 5.694334173202515,
      "epoch": 5.217348154029377,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002630015031645345,
      "loss": 4.9204,
      "mean_token_accuracy": 0.1970287024974823,
      "num_tokens": 115178908.0,
      "step": 52570
    },
    {
      "entropy": 5.724315977096557,
      "epoch": 5.217844382691545,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002629661222320606,
      "loss": 5.0064,
      "mean_token_accuracy": 0.18759672045707704,
      "num_tokens": 115189304.0,
      "step": 52575
    },
    {
      "entropy": 5.752458667755127,
      "epoch": 5.218340611353712,
      "grad_norm": 1.46875,
      "learning_rate": 0.00026293074159800087,
      "loss": 5.0222,
      "mean_token_accuracy": 0.1925370365381241,
      "num_tokens": 115200431.0,
      "step": 52580
    },
    {
      "entropy": 5.696887445449829,
      "epoch": 5.2188368400158796,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002628953612632326,
      "loss": 4.8999,
      "mean_token_accuracy": 0.19794116765260697,
      "num_tokens": 115212331.0,
      "step": 52585
    },
    {
      "entropy": 5.74598183631897,
      "epoch": 5.219333068678047,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026285998122863334,
      "loss": 4.9829,
      "mean_token_accuracy": 0.18913692235946655,
      "num_tokens": 115223462.0,
      "step": 52590
    },
    {
      "entropy": 5.728065919876099,
      "epoch": 5.2198292973402145,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026282460149508046,
      "loss": 4.9146,
      "mean_token_accuracy": 0.19595482349395751,
      "num_tokens": 115234206.0,
      "step": 52595
    },
    {
      "entropy": 5.689486312866211,
      "epoch": 5.220325526002382,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002627892220634512,
      "loss": 4.9022,
      "mean_token_accuracy": 0.19712437242269515,
      "num_tokens": 115245374.0,
      "step": 52600
    },
    {
      "entropy": 5.7416681289672855,
      "epoch": 5.2208217546645495,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000262753842934623,
      "loss": 5.0049,
      "mean_token_accuracy": 0.19836370199918746,
      "num_tokens": 115255702.0,
      "step": 52605
    },
    {
      "entropy": 5.736236190795898,
      "epoch": 5.221317983326717,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002627184641094731,
      "loss": 5.0437,
      "mean_token_accuracy": 0.19552673548460006,
      "num_tokens": 115267615.0,
      "step": 52610
    },
    {
      "entropy": 5.757116031646729,
      "epoch": 5.221814211988884,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026268308558887864,
      "loss": 4.9934,
      "mean_token_accuracy": 0.19523344784975052,
      "num_tokens": 115278219.0,
      "step": 52615
    },
    {
      "entropy": 5.766257286071777,
      "epoch": 5.222310440651052,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026264770737371724,
      "loss": 4.9183,
      "mean_token_accuracy": 0.2032124489545822,
      "num_tokens": 115288741.0,
      "step": 52620
    },
    {
      "entropy": 5.665494346618653,
      "epoch": 5.222806669313219,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026261232946486615,
      "loss": 4.904,
      "mean_token_accuracy": 0.20158242136240007,
      "num_tokens": 115299490.0,
      "step": 52625
    },
    {
      "entropy": 5.71666579246521,
      "epoch": 5.223302897975387,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026257695186320245,
      "loss": 4.9906,
      "mean_token_accuracy": 0.1926918163895607,
      "num_tokens": 115311335.0,
      "step": 52630
    },
    {
      "entropy": 5.70409140586853,
      "epoch": 5.223799126637554,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002625415745696036,
      "loss": 4.9503,
      "mean_token_accuracy": 0.1929139658808708,
      "num_tokens": 115321176.0,
      "step": 52635
    },
    {
      "entropy": 5.683938455581665,
      "epoch": 5.224295355299722,
      "grad_norm": 1.328125,
      "learning_rate": 0.000262506197584947,
      "loss": 4.9597,
      "mean_token_accuracy": 0.1905907616019249,
      "num_tokens": 115332071.0,
      "step": 52640
    },
    {
      "entropy": 5.744533586502075,
      "epoch": 5.224791583961889,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002624708209101096,
      "loss": 4.9547,
      "mean_token_accuracy": 0.19320521652698516,
      "num_tokens": 115341935.0,
      "step": 52645
    },
    {
      "entropy": 5.726182079315185,
      "epoch": 5.225287812624057,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026243544454596904,
      "loss": 4.9116,
      "mean_token_accuracy": 0.19417412877082824,
      "num_tokens": 115353665.0,
      "step": 52650
    },
    {
      "entropy": 5.790873050689697,
      "epoch": 5.225784041286225,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002624000684934022,
      "loss": 5.0619,
      "mean_token_accuracy": 0.1915119469165802,
      "num_tokens": 115364243.0,
      "step": 52655
    },
    {
      "entropy": 5.736212491989136,
      "epoch": 5.226280269948393,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002623646927532866,
      "loss": 4.9406,
      "mean_token_accuracy": 0.19378914833068847,
      "num_tokens": 115375051.0,
      "step": 52660
    },
    {
      "entropy": 5.719085693359375,
      "epoch": 5.22677649861056,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00026232931732649933,
      "loss": 4.9879,
      "mean_token_accuracy": 0.19666277021169662,
      "num_tokens": 115386283.0,
      "step": 52665
    },
    {
      "entropy": 5.740367364883423,
      "epoch": 5.2272727272727275,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026229394221391776,
      "loss": 4.8991,
      "mean_token_accuracy": 0.19822874665260315,
      "num_tokens": 115396350.0,
      "step": 52670
    },
    {
      "entropy": 5.610419225692749,
      "epoch": 5.227768955934895,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026225856741641894,
      "loss": 4.8452,
      "mean_token_accuracy": 0.19937436133623124,
      "num_tokens": 115408224.0,
      "step": 52675
    },
    {
      "entropy": 5.701363754272461,
      "epoch": 5.2282651845970625,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002622231929348803,
      "loss": 4.9964,
      "mean_token_accuracy": 0.19852211624383925,
      "num_tokens": 115419379.0,
      "step": 52680
    },
    {
      "entropy": 5.64201250076294,
      "epoch": 5.22876141325923,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026218781877017887,
      "loss": 4.8649,
      "mean_token_accuracy": 0.19502791166305541,
      "num_tokens": 115429383.0,
      "step": 52685
    },
    {
      "entropy": 5.637674617767334,
      "epoch": 5.229257641921397,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000262152444923192,
      "loss": 4.8661,
      "mean_token_accuracy": 0.20380906909704208,
      "num_tokens": 115439432.0,
      "step": 52690
    },
    {
      "entropy": 5.685958671569824,
      "epoch": 5.229753870583565,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026211707139479684,
      "loss": 4.9418,
      "mean_token_accuracy": 0.19671493470668794,
      "num_tokens": 115450880.0,
      "step": 52695
    },
    {
      "entropy": 5.621688556671143,
      "epoch": 5.230250099245732,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026208169818587045,
      "loss": 4.8628,
      "mean_token_accuracy": 0.20830719023942948,
      "num_tokens": 115464072.0,
      "step": 52700
    },
    {
      "entropy": 5.667966032028199,
      "epoch": 5.2307463279079,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002620463252972901,
      "loss": 4.9182,
      "mean_token_accuracy": 0.20183223634958267,
      "num_tokens": 115473744.0,
      "step": 52705
    },
    {
      "entropy": 5.814383268356323,
      "epoch": 5.231242556570067,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026201095272993305,
      "loss": 5.0007,
      "mean_token_accuracy": 0.1943182587623596,
      "num_tokens": 115484276.0,
      "step": 52710
    },
    {
      "entropy": 5.717293739318848,
      "epoch": 5.231738785232235,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002619755804846763,
      "loss": 4.9343,
      "mean_token_accuracy": 0.19908336848020552,
      "num_tokens": 115495903.0,
      "step": 52715
    },
    {
      "entropy": 5.661422157287598,
      "epoch": 5.232235013894402,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002619402085623972,
      "loss": 4.8722,
      "mean_token_accuracy": 0.20367136597633362,
      "num_tokens": 115506530.0,
      "step": 52720
    },
    {
      "entropy": 5.70997257232666,
      "epoch": 5.23273124255657,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002619048369639727,
      "loss": 4.9909,
      "mean_token_accuracy": 0.19311085790395738,
      "num_tokens": 115517758.0,
      "step": 52725
    },
    {
      "entropy": 5.685317277908325,
      "epoch": 5.233227471218737,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002618694656902801,
      "loss": 4.9702,
      "mean_token_accuracy": 0.2006731688976288,
      "num_tokens": 115528570.0,
      "step": 52730
    },
    {
      "entropy": 5.730946159362793,
      "epoch": 5.233723699880905,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002618340947421963,
      "loss": 4.9704,
      "mean_token_accuracy": 0.19533158391714095,
      "num_tokens": 115539588.0,
      "step": 52735
    },
    {
      "entropy": 5.689680337905884,
      "epoch": 5.234219928543073,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002617987241205987,
      "loss": 4.9358,
      "mean_token_accuracy": 0.20235762745141983,
      "num_tokens": 115551997.0,
      "step": 52740
    },
    {
      "entropy": 5.660953092575073,
      "epoch": 5.2347161572052405,
      "grad_norm": 1.25,
      "learning_rate": 0.00026176335382636435,
      "loss": 4.8919,
      "mean_token_accuracy": 0.202590149641037,
      "num_tokens": 115563700.0,
      "step": 52745
    },
    {
      "entropy": 5.694117593765259,
      "epoch": 5.235212385867408,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002617279838603702,
      "loss": 4.9222,
      "mean_token_accuracy": 0.19316876530647278,
      "num_tokens": 115573804.0,
      "step": 52750
    },
    {
      "entropy": 5.647228145599366,
      "epoch": 5.2357086145295755,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002616926142234935,
      "loss": 4.8891,
      "mean_token_accuracy": 0.19632029980421067,
      "num_tokens": 115583926.0,
      "step": 52755
    },
    {
      "entropy": 5.666193103790283,
      "epoch": 5.236204843191743,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026165724491661124,
      "loss": 4.9205,
      "mean_token_accuracy": 0.19978583306074144,
      "num_tokens": 115594799.0,
      "step": 52760
    },
    {
      "entropy": 5.668139553070068,
      "epoch": 5.23670107185391,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002616218759406006,
      "loss": 4.9526,
      "mean_token_accuracy": 0.19318466037511825,
      "num_tokens": 115605631.0,
      "step": 52765
    },
    {
      "entropy": 5.718738412857055,
      "epoch": 5.237197300516078,
      "grad_norm": 1.25,
      "learning_rate": 0.0002615865072963386,
      "loss": 4.8896,
      "mean_token_accuracy": 0.1929086208343506,
      "num_tokens": 115616908.0,
      "step": 52770
    },
    {
      "entropy": 5.667117071151734,
      "epoch": 5.237693529178245,
      "grad_norm": 1.25,
      "learning_rate": 0.00026155113898470233,
      "loss": 4.8633,
      "mean_token_accuracy": 0.20177462697029114,
      "num_tokens": 115627232.0,
      "step": 52775
    },
    {
      "entropy": 5.698588085174561,
      "epoch": 5.238189757840413,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002615157710065688,
      "loss": 4.8916,
      "mean_token_accuracy": 0.19713615328073503,
      "num_tokens": 115638680.0,
      "step": 52780
    },
    {
      "entropy": 5.768032360076904,
      "epoch": 5.23868598650258,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002614804033628151,
      "loss": 4.9454,
      "mean_token_accuracy": 0.19800226390361786,
      "num_tokens": 115649256.0,
      "step": 52785
    },
    {
      "entropy": 5.699259996414185,
      "epoch": 5.239182215164748,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026144503605431835,
      "loss": 4.9401,
      "mean_token_accuracy": 0.18971161395311356,
      "num_tokens": 115658471.0,
      "step": 52790
    },
    {
      "entropy": 5.648785591125488,
      "epoch": 5.239678443826915,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026140966908195536,
      "loss": 4.8932,
      "mean_token_accuracy": 0.20114708244800567,
      "num_tokens": 115669949.0,
      "step": 52795
    },
    {
      "entropy": 5.628480863571167,
      "epoch": 5.240174672489083,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002613743024466034,
      "loss": 4.8435,
      "mean_token_accuracy": 0.20001846551895142,
      "num_tokens": 115681257.0,
      "step": 52800
    },
    {
      "entropy": 5.680526685714722,
      "epoch": 5.24067090115125,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026133893614913934,
      "loss": 4.9039,
      "mean_token_accuracy": 0.20114381313323976,
      "num_tokens": 115692107.0,
      "step": 52805
    },
    {
      "entropy": 5.661636114120483,
      "epoch": 5.241167129813418,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002613035701904403,
      "loss": 4.8513,
      "mean_token_accuracy": 0.20290393382310867,
      "num_tokens": 115703446.0,
      "step": 52810
    },
    {
      "entropy": 5.687546348571777,
      "epoch": 5.241663358475585,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026126820457138305,
      "loss": 4.9547,
      "mean_token_accuracy": 0.19868155419826508,
      "num_tokens": 115714264.0,
      "step": 52815
    },
    {
      "entropy": 5.70382251739502,
      "epoch": 5.242159587137753,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026123283929284477,
      "loss": 4.967,
      "mean_token_accuracy": 0.19509289264678956,
      "num_tokens": 115725744.0,
      "step": 52820
    },
    {
      "entropy": 5.659027338027954,
      "epoch": 5.242655815799921,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026119747435570246,
      "loss": 4.9076,
      "mean_token_accuracy": 0.20108177065849303,
      "num_tokens": 115735659.0,
      "step": 52825
    },
    {
      "entropy": 5.666265630722046,
      "epoch": 5.2431520444620885,
      "grad_norm": 1.28125,
      "learning_rate": 0.000261162109760833,
      "loss": 4.9499,
      "mean_token_accuracy": 0.1955770179629326,
      "num_tokens": 115745811.0,
      "step": 52830
    },
    {
      "entropy": 5.772841072082519,
      "epoch": 5.243648273124256,
      "grad_norm": 1.46875,
      "learning_rate": 0.00026112674550911346,
      "loss": 4.9828,
      "mean_token_accuracy": 0.195279099047184,
      "num_tokens": 115757711.0,
      "step": 52835
    },
    {
      "entropy": 5.728541278839112,
      "epoch": 5.244144501786423,
      "grad_norm": 1.375,
      "learning_rate": 0.0002610913816014208,
      "loss": 4.9752,
      "mean_token_accuracy": 0.1905156910419464,
      "num_tokens": 115767056.0,
      "step": 52840
    },
    {
      "entropy": 5.70828332901001,
      "epoch": 5.244640730448591,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002610560180386318,
      "loss": 4.9782,
      "mean_token_accuracy": 0.19849029332399368,
      "num_tokens": 115778064.0,
      "step": 52845
    },
    {
      "entropy": 5.766352701187134,
      "epoch": 5.245136959110758,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026102065482162346,
      "loss": 4.969,
      "mean_token_accuracy": 0.18809995651245118,
      "num_tokens": 115788236.0,
      "step": 52850
    },
    {
      "entropy": 5.693109321594238,
      "epoch": 5.245633187772926,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026098529195127287,
      "loss": 4.9026,
      "mean_token_accuracy": 0.2029440224170685,
      "num_tokens": 115800335.0,
      "step": 52855
    },
    {
      "entropy": 5.766478681564331,
      "epoch": 5.246129416435093,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002609499294284568,
      "loss": 4.9924,
      "mean_token_accuracy": 0.19359638094902037,
      "num_tokens": 115810763.0,
      "step": 52860
    },
    {
      "entropy": 5.73553581237793,
      "epoch": 5.246625645097261,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002609145672540522,
      "loss": 5.091,
      "mean_token_accuracy": 0.18415138274431228,
      "num_tokens": 115821868.0,
      "step": 52865
    },
    {
      "entropy": 5.718364191055298,
      "epoch": 5.247121873759428,
      "grad_norm": 1.421875,
      "learning_rate": 0.000260879205428936,
      "loss": 4.9357,
      "mean_token_accuracy": 0.20087843090295793,
      "num_tokens": 115833692.0,
      "step": 52870
    },
    {
      "entropy": 5.689180374145508,
      "epoch": 5.247618102421596,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026084384395398506,
      "loss": 4.9282,
      "mean_token_accuracy": 0.1894230380654335,
      "num_tokens": 115844372.0,
      "step": 52875
    },
    {
      "entropy": 5.725188398361206,
      "epoch": 5.248114331083763,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026080848283007623,
      "loss": 4.9333,
      "mean_token_accuracy": 0.19577012062072754,
      "num_tokens": 115855326.0,
      "step": 52880
    },
    {
      "entropy": 5.638523244857788,
      "epoch": 5.248610559745931,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002607731220580865,
      "loss": 4.8916,
      "mean_token_accuracy": 0.20243924409151076,
      "num_tokens": 115867166.0,
      "step": 52885
    },
    {
      "entropy": 5.745821285247803,
      "epoch": 5.249106788408098,
      "grad_norm": 1.375,
      "learning_rate": 0.0002607377616388927,
      "loss": 4.9965,
      "mean_token_accuracy": 0.19009915739297867,
      "num_tokens": 115877461.0,
      "step": 52890
    },
    {
      "entropy": 5.693234205245972,
      "epoch": 5.249603017070266,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002607024015733717,
      "loss": 4.8922,
      "mean_token_accuracy": 0.19936977475881576,
      "num_tokens": 115887778.0,
      "step": 52895
    },
    {
      "entropy": 5.665613174438477,
      "epoch": 5.250099245732433,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026066704186240034,
      "loss": 4.8503,
      "mean_token_accuracy": 0.203481225669384,
      "num_tokens": 115897820.0,
      "step": 52900
    },
    {
      "entropy": 5.682843351364136,
      "epoch": 5.250595474394601,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026063168250685544,
      "loss": 4.9259,
      "mean_token_accuracy": 0.20020405501127242,
      "num_tokens": 115908656.0,
      "step": 52905
    },
    {
      "entropy": 5.692731761932373,
      "epoch": 5.251091703056769,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026059632350761375,
      "loss": 4.9333,
      "mean_token_accuracy": 0.19552842527627945,
      "num_tokens": 115918234.0,
      "step": 52910
    },
    {
      "entropy": 5.6417741775512695,
      "epoch": 5.2515879317189365,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026056096486555227,
      "loss": 4.9017,
      "mean_token_accuracy": 0.2051575481891632,
      "num_tokens": 115930485.0,
      "step": 52915
    },
    {
      "entropy": 5.673785018920898,
      "epoch": 5.252084160381104,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002605256065815478,
      "loss": 4.8683,
      "mean_token_accuracy": 0.20300298780202866,
      "num_tokens": 115942726.0,
      "step": 52920
    },
    {
      "entropy": 5.683559226989746,
      "epoch": 5.252580389043271,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002604902486564771,
      "loss": 4.9214,
      "mean_token_accuracy": 0.19865538626909257,
      "num_tokens": 115953701.0,
      "step": 52925
    },
    {
      "entropy": 5.737521457672119,
      "epoch": 5.253076617705439,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002604548910912169,
      "loss": 4.9835,
      "mean_token_accuracy": 0.19032049030065537,
      "num_tokens": 115965123.0,
      "step": 52930
    },
    {
      "entropy": 5.679692220687866,
      "epoch": 5.253572846367606,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002604195338866441,
      "loss": 4.953,
      "mean_token_accuracy": 0.19337188750505446,
      "num_tokens": 115976098.0,
      "step": 52935
    },
    {
      "entropy": 5.678041410446167,
      "epoch": 5.254069075029774,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026038417704363535,
      "loss": 4.8896,
      "mean_token_accuracy": 0.20163468569517135,
      "num_tokens": 115986655.0,
      "step": 52940
    },
    {
      "entropy": 5.766779279708862,
      "epoch": 5.254565303691941,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026034882056306756,
      "loss": 4.9244,
      "mean_token_accuracy": 0.1996227592229843,
      "num_tokens": 115997130.0,
      "step": 52945
    },
    {
      "entropy": 5.654198265075683,
      "epoch": 5.255061532354109,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026031346444581755,
      "loss": 4.8388,
      "mean_token_accuracy": 0.20532012581825257,
      "num_tokens": 116006235.0,
      "step": 52950
    },
    {
      "entropy": 5.64051775932312,
      "epoch": 5.255557761016276,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002602781086927619,
      "loss": 4.9123,
      "mean_token_accuracy": 0.1970517873764038,
      "num_tokens": 116018587.0,
      "step": 52955
    },
    {
      "entropy": 5.692645740509033,
      "epoch": 5.256053989678444,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026024275330477744,
      "loss": 4.9111,
      "mean_token_accuracy": 0.20586899369955064,
      "num_tokens": 116029370.0,
      "step": 52960
    },
    {
      "entropy": 5.694892930984497,
      "epoch": 5.256550218340611,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002602073982827409,
      "loss": 4.9301,
      "mean_token_accuracy": 0.19530829340219497,
      "num_tokens": 116041088.0,
      "step": 52965
    },
    {
      "entropy": 5.7638613224029545,
      "epoch": 5.257046447002779,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002601720436275289,
      "loss": 5.0115,
      "mean_token_accuracy": 0.1937049686908722,
      "num_tokens": 116051749.0,
      "step": 52970
    },
    {
      "entropy": 5.730875396728516,
      "epoch": 5.257542675664946,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026013668934001845,
      "loss": 4.9361,
      "mean_token_accuracy": 0.1983526811003685,
      "num_tokens": 116062160.0,
      "step": 52975
    },
    {
      "entropy": 5.6178511619567875,
      "epoch": 5.258038904327114,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000260101335421086,
      "loss": 4.8316,
      "mean_token_accuracy": 0.2071821093559265,
      "num_tokens": 116073492.0,
      "step": 52980
    },
    {
      "entropy": 5.663135385513305,
      "epoch": 5.258535132989281,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026006598187160835,
      "loss": 4.9523,
      "mean_token_accuracy": 0.1947367161512375,
      "num_tokens": 116085592.0,
      "step": 52985
    },
    {
      "entropy": 5.7002791404724125,
      "epoch": 5.259031361651449,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002600306286924621,
      "loss": 4.9586,
      "mean_token_accuracy": 0.1947256490588188,
      "num_tokens": 116096291.0,
      "step": 52990
    },
    {
      "entropy": 5.786538743972779,
      "epoch": 5.259527590313617,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002599952758845241,
      "loss": 4.9981,
      "mean_token_accuracy": 0.19199930280447006,
      "num_tokens": 116107025.0,
      "step": 52995
    },
    {
      "entropy": 5.734166955947876,
      "epoch": 5.260023818975784,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00025995992344867076,
      "loss": 4.9515,
      "mean_token_accuracy": 0.19668067544698714,
      "num_tokens": 116118596.0,
      "step": 53000
    },
    {
      "entropy": 5.7291912078857425,
      "epoch": 5.260520047637952,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00025992457138577905,
      "loss": 4.9787,
      "mean_token_accuracy": 0.19459497630596162,
      "num_tokens": 116129513.0,
      "step": 53005
    },
    {
      "entropy": 5.653101205825806,
      "epoch": 5.261016276300119,
      "grad_norm": 1.25,
      "learning_rate": 0.00025988921969672546,
      "loss": 4.8912,
      "mean_token_accuracy": 0.19998722970485688,
      "num_tokens": 116139753.0,
      "step": 53010
    },
    {
      "entropy": 5.652755641937256,
      "epoch": 5.261512504962287,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00025985386838238666,
      "loss": 4.8733,
      "mean_token_accuracy": 0.2022975876927376,
      "num_tokens": 116149943.0,
      "step": 53015
    },
    {
      "entropy": 5.707214641571045,
      "epoch": 5.262008733624454,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002598185174436393,
      "loss": 4.9936,
      "mean_token_accuracy": 0.19503922164440154,
      "num_tokens": 116161553.0,
      "step": 53020
    },
    {
      "entropy": 5.702684497833252,
      "epoch": 5.262504962286622,
      "grad_norm": 1.375,
      "learning_rate": 0.00025978316688136,
      "loss": 4.9519,
      "mean_token_accuracy": 0.19470053911209106,
      "num_tokens": 116171324.0,
      "step": 53025
    },
    {
      "entropy": 5.710220670700073,
      "epoch": 5.263001190948789,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002597478166964253,
      "loss": 4.9335,
      "mean_token_accuracy": 0.19703718423843383,
      "num_tokens": 116182193.0,
      "step": 53030
    },
    {
      "entropy": 5.694005537033081,
      "epoch": 5.263497419610957,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025971246688971195,
      "loss": 4.9038,
      "mean_token_accuracy": 0.1956353560090065,
      "num_tokens": 116193281.0,
      "step": 53035
    },
    {
      "entropy": 5.703782320022583,
      "epoch": 5.263993648273124,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002596771174620965,
      "loss": 4.9851,
      "mean_token_accuracy": 0.19985890239477158,
      "num_tokens": 116203534.0,
      "step": 53040
    },
    {
      "entropy": 5.693829727172852,
      "epoch": 5.264489876935292,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025964176841445545,
      "loss": 4.9126,
      "mean_token_accuracy": 0.19613039940595628,
      "num_tokens": 116215196.0,
      "step": 53045
    },
    {
      "entropy": 5.666661500930786,
      "epoch": 5.264986105597459,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002596064197476655,
      "loss": 4.896,
      "mean_token_accuracy": 0.20263083577156066,
      "num_tokens": 116227377.0,
      "step": 53050
    },
    {
      "entropy": 5.652254676818847,
      "epoch": 5.265482334259627,
      "grad_norm": 1.1875,
      "learning_rate": 0.00025957107146260314,
      "loss": 4.9029,
      "mean_token_accuracy": 0.1968776538968086,
      "num_tokens": 116238868.0,
      "step": 53055
    },
    {
      "entropy": 5.744005346298218,
      "epoch": 5.265978562921794,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002595357235601449,
      "loss": 4.9746,
      "mean_token_accuracy": 0.1903601258993149,
      "num_tokens": 116250941.0,
      "step": 53060
    },
    {
      "entropy": 5.734068775177002,
      "epoch": 5.266474791583962,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002595003760411675,
      "loss": 4.951,
      "mean_token_accuracy": 0.19766826778650284,
      "num_tokens": 116261903.0,
      "step": 53065
    },
    {
      "entropy": 5.698042392730713,
      "epoch": 5.266971020246129,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00025946502890654736,
      "loss": 4.8904,
      "mean_token_accuracy": 0.2010051965713501,
      "num_tokens": 116273064.0,
      "step": 53070
    },
    {
      "entropy": 5.671633958816528,
      "epoch": 5.2674672489082965,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000259429682157161,
      "loss": 4.8428,
      "mean_token_accuracy": 0.20937437266111375,
      "num_tokens": 116284775.0,
      "step": 53075
    },
    {
      "entropy": 5.719897222518921,
      "epoch": 5.267963477570465,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025939433579388497,
      "loss": 4.9241,
      "mean_token_accuracy": 0.19683159440755843,
      "num_tokens": 116296331.0,
      "step": 53080
    },
    {
      "entropy": 5.60725736618042,
      "epoch": 5.268459706232632,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025935898981759573,
      "loss": 4.8966,
      "mean_token_accuracy": 0.199302513897419,
      "num_tokens": 116306845.0,
      "step": 53085
    },
    {
      "entropy": 5.683771467208862,
      "epoch": 5.2689559348948,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025932364422916995,
      "loss": 4.9516,
      "mean_token_accuracy": 0.19754580408334732,
      "num_tokens": 116317740.0,
      "step": 53090
    },
    {
      "entropy": 5.845434999465942,
      "epoch": 5.269452163556967,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025928829902948394,
      "loss": 5.0696,
      "mean_token_accuracy": 0.18780521899461747,
      "num_tokens": 116327892.0,
      "step": 53095
    },
    {
      "entropy": 5.677501010894775,
      "epoch": 5.269948392219135,
      "grad_norm": 1.25,
      "learning_rate": 0.00025925295421941433,
      "loss": 4.8217,
      "mean_token_accuracy": 0.20885866433382033,
      "num_tokens": 116337938.0,
      "step": 53100
    },
    {
      "entropy": 5.672218084335327,
      "epoch": 5.270444620881302,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002592176097998374,
      "loss": 4.8726,
      "mean_token_accuracy": 0.1975746750831604,
      "num_tokens": 116348034.0,
      "step": 53105
    },
    {
      "entropy": 5.707786273956299,
      "epoch": 5.27094084954347,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002591822657716299,
      "loss": 4.9208,
      "mean_token_accuracy": 0.20236126482486724,
      "num_tokens": 116360117.0,
      "step": 53110
    },
    {
      "entropy": 5.717620229721069,
      "epoch": 5.271437078205637,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000259146922135668,
      "loss": 4.9797,
      "mean_token_accuracy": 0.2005777269601822,
      "num_tokens": 116370868.0,
      "step": 53115
    },
    {
      "entropy": 5.682320594787598,
      "epoch": 5.271933306867805,
      "grad_norm": 1.140625,
      "learning_rate": 0.00025911157889282833,
      "loss": 4.9031,
      "mean_token_accuracy": 0.19762063324451445,
      "num_tokens": 116383549.0,
      "step": 53120
    },
    {
      "entropy": 5.688329267501831,
      "epoch": 5.272429535529972,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002590762360439872,
      "loss": 4.9198,
      "mean_token_accuracy": 0.19965714514255523,
      "num_tokens": 116393800.0,
      "step": 53125
    },
    {
      "entropy": 5.692928457260132,
      "epoch": 5.27292576419214,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002590408935900212,
      "loss": 4.929,
      "mean_token_accuracy": 0.19720062613487244,
      "num_tokens": 116404383.0,
      "step": 53130
    },
    {
      "entropy": 5.643713569641113,
      "epoch": 5.273421992854307,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002590055515318066,
      "loss": 4.8807,
      "mean_token_accuracy": 0.20381067544221879,
      "num_tokens": 116415993.0,
      "step": 53135
    },
    {
      "entropy": 5.603085660934449,
      "epoch": 5.273918221516475,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025897020987021986,
      "loss": 4.798,
      "mean_token_accuracy": 0.2142518863081932,
      "num_tokens": 116425570.0,
      "step": 53140
    },
    {
      "entropy": 5.733222007751465,
      "epoch": 5.274414450178642,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002589348686061373,
      "loss": 4.9393,
      "mean_token_accuracy": 0.1978315681219101,
      "num_tokens": 116437814.0,
      "step": 53145
    },
    {
      "entropy": 5.705519723892212,
      "epoch": 5.2749106788408096,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002588995277404355,
      "loss": 4.9295,
      "mean_token_accuracy": 0.1964352771639824,
      "num_tokens": 116449138.0,
      "step": 53150
    },
    {
      "entropy": 5.683134269714356,
      "epoch": 5.275406907502977,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002588641872739907,
      "loss": 4.9332,
      "mean_token_accuracy": 0.19589436203241348,
      "num_tokens": 116459321.0,
      "step": 53155
    },
    {
      "entropy": 5.683179140090942,
      "epoch": 5.2759031361651445,
      "grad_norm": 1.375,
      "learning_rate": 0.00025882884720767924,
      "loss": 4.9266,
      "mean_token_accuracy": 0.19584083259105683,
      "num_tokens": 116470254.0,
      "step": 53160
    },
    {
      "entropy": 5.698365926742554,
      "epoch": 5.276399364827313,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002587935075423776,
      "loss": 4.9008,
      "mean_token_accuracy": 0.19976085275411606,
      "num_tokens": 116480409.0,
      "step": 53165
    },
    {
      "entropy": 5.707566642761231,
      "epoch": 5.27689559348948,
      "grad_norm": 1.1875,
      "learning_rate": 0.000258758168278962,
      "loss": 4.935,
      "mean_token_accuracy": 0.19798821210861206,
      "num_tokens": 116491229.0,
      "step": 53170
    },
    {
      "entropy": 5.697476100921631,
      "epoch": 5.277391822151648,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025872282941830874,
      "loss": 4.9307,
      "mean_token_accuracy": 0.20283331871032714,
      "num_tokens": 116501253.0,
      "step": 53175
    },
    {
      "entropy": 5.69756555557251,
      "epoch": 5.277888050813815,
      "grad_norm": 1.25,
      "learning_rate": 0.00025868749096129436,
      "loss": 4.924,
      "mean_token_accuracy": 0.1962843969464302,
      "num_tokens": 116511700.0,
      "step": 53180
    },
    {
      "entropy": 5.758794832229614,
      "epoch": 5.278384279475983,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000258652152908795,
      "loss": 5.006,
      "mean_token_accuracy": 0.1926845580339432,
      "num_tokens": 116522685.0,
      "step": 53185
    },
    {
      "entropy": 5.691060876846313,
      "epoch": 5.27888050813815,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000258616815261687,
      "loss": 4.9462,
      "mean_token_accuracy": 0.1997464567422867,
      "num_tokens": 116535023.0,
      "step": 53190
    },
    {
      "entropy": 5.7247490882873535,
      "epoch": 5.279376736800318,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002585814780208467,
      "loss": 4.9044,
      "mean_token_accuracy": 0.19767589420080184,
      "num_tokens": 116545366.0,
      "step": 53195
    },
    {
      "entropy": 5.717315244674682,
      "epoch": 5.279872965462485,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025854614118715047,
      "loss": 4.9262,
      "mean_token_accuracy": 0.19731526374816893,
      "num_tokens": 116557074.0,
      "step": 53200
    },
    {
      "entropy": 5.639222049713135,
      "epoch": 5.280369194124653,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002585108047614743,
      "loss": 4.9101,
      "mean_token_accuracy": 0.20206934362649917,
      "num_tokens": 116569013.0,
      "step": 53205
    },
    {
      "entropy": 5.7354230880737305,
      "epoch": 5.28086542278682,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025847546874469473,
      "loss": 4.9204,
      "mean_token_accuracy": 0.19965833574533462,
      "num_tokens": 116579863.0,
      "step": 53210
    },
    {
      "entropy": 5.682948017120362,
      "epoch": 5.281361651448988,
      "grad_norm": 1.375,
      "learning_rate": 0.0002584401331376879,
      "loss": 4.9426,
      "mean_token_accuracy": 0.19699845761060714,
      "num_tokens": 116590227.0,
      "step": 53215
    },
    {
      "entropy": 5.734935760498047,
      "epoch": 5.281857880111155,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002584047979413302,
      "loss": 5.0127,
      "mean_token_accuracy": 0.18718192726373672,
      "num_tokens": 116599922.0,
      "step": 53220
    },
    {
      "entropy": 5.658863830566406,
      "epoch": 5.282354108773323,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002583694631564977,
      "loss": 4.8895,
      "mean_token_accuracy": 0.20259537547826767,
      "num_tokens": 116610431.0,
      "step": 53225
    },
    {
      "entropy": 5.674420356750488,
      "epoch": 5.28285033743549,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002583341287840667,
      "loss": 4.9224,
      "mean_token_accuracy": 0.20362308323383332,
      "num_tokens": 116621408.0,
      "step": 53230
    },
    {
      "entropy": 5.617874670028686,
      "epoch": 5.2833465660976575,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002582987948249132,
      "loss": 4.8555,
      "mean_token_accuracy": 0.20507685989141464,
      "num_tokens": 116631446.0,
      "step": 53235
    },
    {
      "entropy": 5.6877694606781,
      "epoch": 5.283842794759825,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025826346127991374,
      "loss": 4.957,
      "mean_token_accuracy": 0.19921682476997377,
      "num_tokens": 116642143.0,
      "step": 53240
    },
    {
      "entropy": 5.669981956481934,
      "epoch": 5.2843390234219925,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025822812814994437,
      "loss": 4.8792,
      "mean_token_accuracy": 0.20581015348434448,
      "num_tokens": 116653091.0,
      "step": 53245
    },
    {
      "entropy": 5.705580997467041,
      "epoch": 5.284835252084161,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025819279543588136,
      "loss": 4.898,
      "mean_token_accuracy": 0.20360547453165054,
      "num_tokens": 116663431.0,
      "step": 53250
    },
    {
      "entropy": 5.638727521896362,
      "epoch": 5.285331480746328,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025815746313860073,
      "loss": 4.8737,
      "mean_token_accuracy": 0.19367434978485107,
      "num_tokens": 116674447.0,
      "step": 53255
    },
    {
      "entropy": 5.682613658905029,
      "epoch": 5.285827709408496,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002581221312589788,
      "loss": 4.9814,
      "mean_token_accuracy": 0.19627880603075026,
      "num_tokens": 116685325.0,
      "step": 53260
    },
    {
      "entropy": 5.642529010772705,
      "epoch": 5.286323938070663,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025808679979789147,
      "loss": 4.8719,
      "mean_token_accuracy": 0.20133240073919295,
      "num_tokens": 116696392.0,
      "step": 53265
    },
    {
      "entropy": 5.6932119846344,
      "epoch": 5.286820166732831,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025805146875621513,
      "loss": 4.9348,
      "mean_token_accuracy": 0.19937416315078735,
      "num_tokens": 116708217.0,
      "step": 53270
    },
    {
      "entropy": 5.707905006408692,
      "epoch": 5.287316395394998,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025801613813482586,
      "loss": 4.9248,
      "mean_token_accuracy": 0.19445523023605346,
      "num_tokens": 116719129.0,
      "step": 53275
    },
    {
      "entropy": 5.708728933334351,
      "epoch": 5.287812624057166,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002579808079345998,
      "loss": 4.9741,
      "mean_token_accuracy": 0.19017140567302704,
      "num_tokens": 116730696.0,
      "step": 53280
    },
    {
      "entropy": 5.768428468704224,
      "epoch": 5.288308852719333,
      "grad_norm": 1.40625,
      "learning_rate": 0.00025794547815641305,
      "loss": 5.0295,
      "mean_token_accuracy": 0.1886778622865677,
      "num_tokens": 116742674.0,
      "step": 53285
    },
    {
      "entropy": 5.749457120895386,
      "epoch": 5.288805081381501,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025791014880114174,
      "loss": 4.956,
      "mean_token_accuracy": 0.19313136786222457,
      "num_tokens": 116753067.0,
      "step": 53290
    },
    {
      "entropy": 5.731710386276245,
      "epoch": 5.289301310043668,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002578748198696617,
      "loss": 4.9703,
      "mean_token_accuracy": 0.19629756063222886,
      "num_tokens": 116764166.0,
      "step": 53295
    },
    {
      "entropy": 5.749149513244629,
      "epoch": 5.289797538705836,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002578394913628494,
      "loss": 4.9823,
      "mean_token_accuracy": 0.19671522378921508,
      "num_tokens": 116776427.0,
      "step": 53300
    },
    {
      "entropy": 5.788914966583252,
      "epoch": 5.290293767368003,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002578041632815807,
      "loss": 5.0311,
      "mean_token_accuracy": 0.19271455705165863,
      "num_tokens": 116788281.0,
      "step": 53305
    },
    {
      "entropy": 5.730815839767456,
      "epoch": 5.2907899960301705,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002577688356267317,
      "loss": 4.9746,
      "mean_token_accuracy": 0.19846539050340653,
      "num_tokens": 116800431.0,
      "step": 53310
    },
    {
      "entropy": 5.6769743919372555,
      "epoch": 5.291286224692338,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025773350839917845,
      "loss": 4.8971,
      "mean_token_accuracy": 0.1998103529214859,
      "num_tokens": 116811226.0,
      "step": 53315
    },
    {
      "entropy": 5.652565240859985,
      "epoch": 5.2917824533545055,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002576981815997969,
      "loss": 4.8885,
      "mean_token_accuracy": 0.20164889991283416,
      "num_tokens": 116823459.0,
      "step": 53320
    },
    {
      "entropy": 5.734805345535278,
      "epoch": 5.292278682016673,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025766285522946323,
      "loss": 4.9186,
      "mean_token_accuracy": 0.1919596865773201,
      "num_tokens": 116834216.0,
      "step": 53325
    },
    {
      "entropy": 5.715456962585449,
      "epoch": 5.29277491067884,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002576275292890534,
      "loss": 4.9188,
      "mean_token_accuracy": 0.1997001901268959,
      "num_tokens": 116845320.0,
      "step": 53330
    },
    {
      "entropy": 5.6461292743682865,
      "epoch": 5.293271139341009,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025759220377944347,
      "loss": 4.8628,
      "mean_token_accuracy": 0.2090821623802185,
      "num_tokens": 116857189.0,
      "step": 53335
    },
    {
      "entropy": 5.702164793014527,
      "epoch": 5.293767368003176,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025755687870150927,
      "loss": 4.9483,
      "mean_token_accuracy": 0.19787101447582245,
      "num_tokens": 116867990.0,
      "step": 53340
    },
    {
      "entropy": 5.71453857421875,
      "epoch": 5.294263596665344,
      "grad_norm": 1.390625,
      "learning_rate": 0.000257521554056127,
      "loss": 4.9421,
      "mean_token_accuracy": 0.19772758185863495,
      "num_tokens": 116878200.0,
      "step": 53345
    },
    {
      "entropy": 5.631395769119263,
      "epoch": 5.294759825327511,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002574862298441724,
      "loss": 4.8478,
      "mean_token_accuracy": 0.20060309618711472,
      "num_tokens": 116888853.0,
      "step": 53350
    },
    {
      "entropy": 5.756799030303955,
      "epoch": 5.295256053989679,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002574509060665216,
      "loss": 4.9733,
      "mean_token_accuracy": 0.19221882224082948,
      "num_tokens": 116899335.0,
      "step": 53355
    },
    {
      "entropy": 5.721328926086426,
      "epoch": 5.295752282651846,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002574155827240506,
      "loss": 4.9953,
      "mean_token_accuracy": 0.1938980981707573,
      "num_tokens": 116911798.0,
      "step": 53360
    },
    {
      "entropy": 5.699877548217773,
      "epoch": 5.296248511314014,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002573802598176352,
      "loss": 4.9588,
      "mean_token_accuracy": 0.19806158542633057,
      "num_tokens": 116922978.0,
      "step": 53365
    },
    {
      "entropy": 5.609647989273071,
      "epoch": 5.296744739976181,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002573449373481514,
      "loss": 4.8646,
      "mean_token_accuracy": 0.20951346457004547,
      "num_tokens": 116933424.0,
      "step": 53370
    },
    {
      "entropy": 5.664544916152954,
      "epoch": 5.297240968638349,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002573096153164751,
      "loss": 4.9141,
      "mean_token_accuracy": 0.2005552500486374,
      "num_tokens": 116943264.0,
      "step": 53375
    },
    {
      "entropy": 5.716147804260254,
      "epoch": 5.297737197300516,
      "grad_norm": 1.21875,
      "learning_rate": 0.00025727429372348216,
      "loss": 4.971,
      "mean_token_accuracy": 0.18949174135923386,
      "num_tokens": 116953973.0,
      "step": 53380
    },
    {
      "entropy": 5.679554748535156,
      "epoch": 5.2982334259626835,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002572389725700486,
      "loss": 4.8578,
      "mean_token_accuracy": 0.1975160911679268,
      "num_tokens": 116965205.0,
      "step": 53385
    },
    {
      "entropy": 5.703759241104126,
      "epoch": 5.298729654624851,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025720365185705023,
      "loss": 4.9288,
      "mean_token_accuracy": 0.20124392807483674,
      "num_tokens": 116976932.0,
      "step": 53390
    },
    {
      "entropy": 5.687123966217041,
      "epoch": 5.2992258832870185,
      "grad_norm": 1.34375,
      "learning_rate": 0.000257168331585363,
      "loss": 4.9543,
      "mean_token_accuracy": 0.20009365826845169,
      "num_tokens": 116987466.0,
      "step": 53395
    },
    {
      "entropy": 5.659508752822876,
      "epoch": 5.299722111949186,
      "grad_norm": 1.25,
      "learning_rate": 0.00025713301175586266,
      "loss": 4.8953,
      "mean_token_accuracy": 0.19385118782520294,
      "num_tokens": 116999589.0,
      "step": 53400
    },
    {
      "entropy": 5.642973279953003,
      "epoch": 5.300218340611353,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002570976923694252,
      "loss": 4.8633,
      "mean_token_accuracy": 0.2052304655313492,
      "num_tokens": 117011146.0,
      "step": 53405
    },
    {
      "entropy": 5.684041595458984,
      "epoch": 5.300714569273521,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002570623734269263,
      "loss": 4.9108,
      "mean_token_accuracy": 0.20223829299211502,
      "num_tokens": 117023139.0,
      "step": 53410
    },
    {
      "entropy": 5.662950325012207,
      "epoch": 5.301210797935688,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00025702705492924193,
      "loss": 4.9518,
      "mean_token_accuracy": 0.18796442300081254,
      "num_tokens": 117034500.0,
      "step": 53415
    },
    {
      "entropy": 5.682228851318359,
      "epoch": 5.301707026597857,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025699173687724784,
      "loss": 4.8502,
      "mean_token_accuracy": 0.210307776927948,
      "num_tokens": 117045423.0,
      "step": 53420
    },
    {
      "entropy": 5.69833345413208,
      "epoch": 5.302203255260024,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025695641927182,
      "loss": 4.8745,
      "mean_token_accuracy": 0.20658834874629975,
      "num_tokens": 117055203.0,
      "step": 53425
    },
    {
      "entropy": 5.640427303314209,
      "epoch": 5.302699483922192,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00025692110211383395,
      "loss": 4.8619,
      "mean_token_accuracy": 0.20743750929832458,
      "num_tokens": 117068465.0,
      "step": 53430
    },
    {
      "entropy": 5.732236003875732,
      "epoch": 5.303195712584359,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025688578540416563,
      "loss": 4.9491,
      "mean_token_accuracy": 0.19576178044080733,
      "num_tokens": 117078362.0,
      "step": 53435
    },
    {
      "entropy": 5.771626377105713,
      "epoch": 5.303691941246527,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002568504691436908,
      "loss": 4.9408,
      "mean_token_accuracy": 0.196548393368721,
      "num_tokens": 117089945.0,
      "step": 53440
    },
    {
      "entropy": 5.664397287368774,
      "epoch": 5.304188169908694,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002568151533332853,
      "loss": 4.9387,
      "mean_token_accuracy": 0.19758054465055466,
      "num_tokens": 117101111.0,
      "step": 53445
    },
    {
      "entropy": 5.68638973236084,
      "epoch": 5.304684398570862,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002567798379738247,
      "loss": 4.9067,
      "mean_token_accuracy": 0.19508874416351318,
      "num_tokens": 117110686.0,
      "step": 53450
    },
    {
      "entropy": 5.622149658203125,
      "epoch": 5.305180627233029,
      "grad_norm": 1.3125,
      "learning_rate": 0.000256744523066185,
      "loss": 4.8935,
      "mean_token_accuracy": 0.19974257946014404,
      "num_tokens": 117121377.0,
      "step": 53455
    },
    {
      "entropy": 5.7420368671417235,
      "epoch": 5.3056768558951966,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025670920861124166,
      "loss": 4.9142,
      "mean_token_accuracy": 0.1995740681886673,
      "num_tokens": 117131932.0,
      "step": 53460
    },
    {
      "entropy": 5.70610294342041,
      "epoch": 5.306173084557364,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00025667389460987057,
      "loss": 4.8909,
      "mean_token_accuracy": 0.2068943753838539,
      "num_tokens": 117143547.0,
      "step": 53465
    },
    {
      "entropy": 5.7134559631347654,
      "epoch": 5.3066693132195315,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025663858106294733,
      "loss": 4.9211,
      "mean_token_accuracy": 0.19747270196676253,
      "num_tokens": 117154076.0,
      "step": 53470
    },
    {
      "entropy": 5.689546966552735,
      "epoch": 5.307165541881699,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002566032679713478,
      "loss": 4.9501,
      "mean_token_accuracy": 0.19811102449893953,
      "num_tokens": 117164885.0,
      "step": 53475
    },
    {
      "entropy": 5.630134773254395,
      "epoch": 5.3076617705438665,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025656795533594764,
      "loss": 4.9299,
      "mean_token_accuracy": 0.18847455829381943,
      "num_tokens": 117175322.0,
      "step": 53480
    },
    {
      "entropy": 5.712917041778565,
      "epoch": 5.308157999206034,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002565326431576224,
      "loss": 4.9811,
      "mean_token_accuracy": 0.19234413504600525,
      "num_tokens": 117185645.0,
      "step": 53485
    },
    {
      "entropy": 5.703914546966553,
      "epoch": 5.308654227868201,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002564973314372478,
      "loss": 5.0157,
      "mean_token_accuracy": 0.19729146212339402,
      "num_tokens": 117198053.0,
      "step": 53490
    },
    {
      "entropy": 5.6513671875,
      "epoch": 5.309150456530369,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002564620201756996,
      "loss": 4.8861,
      "mean_token_accuracy": 0.19667430222034454,
      "num_tokens": 117208620.0,
      "step": 53495
    },
    {
      "entropy": 5.721812152862549,
      "epoch": 5.309646685192536,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002564267093738532,
      "loss": 4.9984,
      "mean_token_accuracy": 0.19013302624225617,
      "num_tokens": 117217641.0,
      "step": 53500
    },
    {
      "entropy": 5.631202697753906,
      "epoch": 5.310142913854705,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025639139903258445,
      "loss": 4.8,
      "mean_token_accuracy": 0.20431647151708604,
      "num_tokens": 117227243.0,
      "step": 53505
    },
    {
      "entropy": 5.621537733078003,
      "epoch": 5.310639142516871,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000256356089152769,
      "loss": 4.8495,
      "mean_token_accuracy": 0.19975513219833374,
      "num_tokens": 117236830.0,
      "step": 53510
    },
    {
      "entropy": 5.652871465682983,
      "epoch": 5.31113537117904,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025632077973528224,
      "loss": 4.9035,
      "mean_token_accuracy": 0.20328988283872604,
      "num_tokens": 117245791.0,
      "step": 53515
    },
    {
      "entropy": 5.652821445465088,
      "epoch": 5.311631599841207,
      "grad_norm": 1.171875,
      "learning_rate": 0.000256285470781,
      "loss": 4.8498,
      "mean_token_accuracy": 0.20564399510622025,
      "num_tokens": 117257112.0,
      "step": 53520
    },
    {
      "entropy": 5.8002691745758055,
      "epoch": 5.312127828503375,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025625016229079774,
      "loss": 5.0757,
      "mean_token_accuracy": 0.18804512917995453,
      "num_tokens": 117268075.0,
      "step": 53525
    },
    {
      "entropy": 5.7665708541870115,
      "epoch": 5.312624057165542,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000256214854265551,
      "loss": 4.9849,
      "mean_token_accuracy": 0.19603580236434937,
      "num_tokens": 117278460.0,
      "step": 53530
    },
    {
      "entropy": 5.649889039993286,
      "epoch": 5.31312028582771,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002561795467061354,
      "loss": 4.8696,
      "mean_token_accuracy": 0.20477627217769623,
      "num_tokens": 117289219.0,
      "step": 53535
    },
    {
      "entropy": 5.674045753479004,
      "epoch": 5.313616514489877,
      "grad_norm": 1.453125,
      "learning_rate": 0.00025614423961342646,
      "loss": 4.9398,
      "mean_token_accuracy": 0.1982103019952774,
      "num_tokens": 117299838.0,
      "step": 53540
    },
    {
      "entropy": 5.6837152481079105,
      "epoch": 5.3141127431520445,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025610893298829985,
      "loss": 4.9765,
      "mean_token_accuracy": 0.1933464601635933,
      "num_tokens": 117310124.0,
      "step": 53545
    },
    {
      "entropy": 5.696647596359253,
      "epoch": 5.314608971814212,
      "grad_norm": 1.375,
      "learning_rate": 0.00025607362683163087,
      "loss": 4.973,
      "mean_token_accuracy": 0.19688974916934968,
      "num_tokens": 117321394.0,
      "step": 53550
    },
    {
      "entropy": 5.740087032318115,
      "epoch": 5.3151052004763795,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025603832114429534,
      "loss": 5.0321,
      "mean_token_accuracy": 0.18798172622919082,
      "num_tokens": 117333043.0,
      "step": 53555
    },
    {
      "entropy": 5.686317777633667,
      "epoch": 5.315601429138547,
      "grad_norm": 1.25,
      "learning_rate": 0.00025600301592716837,
      "loss": 4.9493,
      "mean_token_accuracy": 0.19358501583337784,
      "num_tokens": 117344717.0,
      "step": 53560
    },
    {
      "entropy": 5.707297277450562,
      "epoch": 5.316097657800714,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002559677111811258,
      "loss": 4.9716,
      "mean_token_accuracy": 0.1914629653096199,
      "num_tokens": 117355225.0,
      "step": 53565
    },
    {
      "entropy": 5.77981915473938,
      "epoch": 5.316593886462882,
      "grad_norm": 1.25,
      "learning_rate": 0.00025593240690704295,
      "loss": 5.0012,
      "mean_token_accuracy": 0.18902926445007323,
      "num_tokens": 117365799.0,
      "step": 53570
    },
    {
      "entropy": 5.700954627990723,
      "epoch": 5.317090115125049,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002558971031057953,
      "loss": 4.876,
      "mean_token_accuracy": 0.19887966364622117,
      "num_tokens": 117376510.0,
      "step": 53575
    },
    {
      "entropy": 5.764900588989258,
      "epoch": 5.317586343787217,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025586179977825835,
      "loss": 4.95,
      "mean_token_accuracy": 0.194374980032444,
      "num_tokens": 117387018.0,
      "step": 53580
    },
    {
      "entropy": 5.654652070999146,
      "epoch": 5.318082572449384,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025582649692530745,
      "loss": 4.8637,
      "mean_token_accuracy": 0.1961910530924797,
      "num_tokens": 117398515.0,
      "step": 53585
    },
    {
      "entropy": 5.691674900054932,
      "epoch": 5.318578801111552,
      "grad_norm": 1.15625,
      "learning_rate": 0.00025579119454781824,
      "loss": 4.9854,
      "mean_token_accuracy": 0.1886212036013603,
      "num_tokens": 117409541.0,
      "step": 53590
    },
    {
      "entropy": 5.696449518203735,
      "epoch": 5.319075029773719,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025575589264666604,
      "loss": 4.9731,
      "mean_token_accuracy": 0.19298608899116515,
      "num_tokens": 117421572.0,
      "step": 53595
    },
    {
      "entropy": 5.659212446212768,
      "epoch": 5.319571258435888,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025572059122272613,
      "loss": 4.8986,
      "mean_token_accuracy": 0.2029934749007225,
      "num_tokens": 117432301.0,
      "step": 53600
    },
    {
      "entropy": 5.743692684173584,
      "epoch": 5.320067487098055,
      "grad_norm": 1.25,
      "learning_rate": 0.00025568529027687406,
      "loss": 4.9531,
      "mean_token_accuracy": 0.2012237012386322,
      "num_tokens": 117444048.0,
      "step": 53605
    },
    {
      "entropy": 5.7081512928009035,
      "epoch": 5.320563715760223,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025564998980998516,
      "loss": 4.8865,
      "mean_token_accuracy": 0.1973327189683914,
      "num_tokens": 117454351.0,
      "step": 53610
    },
    {
      "entropy": 5.75146894454956,
      "epoch": 5.32105994442239,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002556146898229348,
      "loss": 4.9771,
      "mean_token_accuracy": 0.19122579991817473,
      "num_tokens": 117464943.0,
      "step": 53615
    },
    {
      "entropy": 5.730382919311523,
      "epoch": 5.3215561730845575,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025557939031659835,
      "loss": 4.9652,
      "mean_token_accuracy": 0.19501783549785615,
      "num_tokens": 117477124.0,
      "step": 53620
    },
    {
      "entropy": 5.67926812171936,
      "epoch": 5.322052401746725,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025554409129185127,
      "loss": 4.8739,
      "mean_token_accuracy": 0.20784209817647933,
      "num_tokens": 117487347.0,
      "step": 53625
    },
    {
      "entropy": 5.679273986816407,
      "epoch": 5.3225486304088925,
      "grad_norm": 1.375,
      "learning_rate": 0.0002555087927495687,
      "loss": 4.9062,
      "mean_token_accuracy": 0.19012643098831178,
      "num_tokens": 117497735.0,
      "step": 53630
    },
    {
      "entropy": 5.728272294998169,
      "epoch": 5.32304485907106,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025547349469062607,
      "loss": 4.9957,
      "mean_token_accuracy": 0.1907808631658554,
      "num_tokens": 117509380.0,
      "step": 53635
    },
    {
      "entropy": 5.789368724822998,
      "epoch": 5.323541087733227,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002554381971158988,
      "loss": 5.0812,
      "mean_token_accuracy": 0.1836322709918022,
      "num_tokens": 117521185.0,
      "step": 53640
    },
    {
      "entropy": 5.720382261276245,
      "epoch": 5.324037316395395,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002554029000262619,
      "loss": 4.886,
      "mean_token_accuracy": 0.19472566395998,
      "num_tokens": 117531788.0,
      "step": 53645
    },
    {
      "entropy": 5.655929183959961,
      "epoch": 5.324533545057562,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00025536760342259097,
      "loss": 4.9026,
      "mean_token_accuracy": 0.19941698014736176,
      "num_tokens": 117542033.0,
      "step": 53650
    },
    {
      "entropy": 5.7133955478668215,
      "epoch": 5.32502977371973,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002553323073057612,
      "loss": 4.9921,
      "mean_token_accuracy": 0.19248603880405427,
      "num_tokens": 117552901.0,
      "step": 53655
    },
    {
      "entropy": 5.5951367855072025,
      "epoch": 5.325526002381897,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002552970116766478,
      "loss": 4.8073,
      "mean_token_accuracy": 0.20309482961893083,
      "num_tokens": 117563027.0,
      "step": 53660
    },
    {
      "entropy": 5.700161504745483,
      "epoch": 5.326022231044065,
      "grad_norm": 1.21875,
      "learning_rate": 0.00025526171653612607,
      "loss": 4.9331,
      "mean_token_accuracy": 0.19723596274852753,
      "num_tokens": 117573611.0,
      "step": 53665
    },
    {
      "entropy": 5.665382814407349,
      "epoch": 5.326518459706232,
      "grad_norm": 1.25,
      "learning_rate": 0.0002552264218850712,
      "loss": 4.9799,
      "mean_token_accuracy": 0.20311252325773238,
      "num_tokens": 117584305.0,
      "step": 53670
    },
    {
      "entropy": 5.7029201030731205,
      "epoch": 5.3270146883684,
      "grad_norm": 1.375,
      "learning_rate": 0.0002551911277243584,
      "loss": 4.8593,
      "mean_token_accuracy": 0.20494290441274643,
      "num_tokens": 117594772.0,
      "step": 53675
    },
    {
      "entropy": 5.651905202865601,
      "epoch": 5.327510917030567,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025515583405486306,
      "loss": 4.8636,
      "mean_token_accuracy": 0.1982347771525383,
      "num_tokens": 117604788.0,
      "step": 53680
    },
    {
      "entropy": 5.6978528022766115,
      "epoch": 5.328007145692736,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002551205408774603,
      "loss": 4.8763,
      "mean_token_accuracy": 0.19828317165374756,
      "num_tokens": 117615497.0,
      "step": 53685
    },
    {
      "entropy": 5.704018783569336,
      "epoch": 5.328503374354903,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002550852481930252,
      "loss": 4.952,
      "mean_token_accuracy": 0.20433240830898286,
      "num_tokens": 117625473.0,
      "step": 53690
    },
    {
      "entropy": 5.706434965133667,
      "epoch": 5.3289996030170705,
      "grad_norm": 1.46875,
      "learning_rate": 0.00025504995600243306,
      "loss": 4.9714,
      "mean_token_accuracy": 0.19620738327503204,
      "num_tokens": 117638936.0,
      "step": 53695
    },
    {
      "entropy": 5.729836750030517,
      "epoch": 5.329495831679238,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025501466430655906,
      "loss": 4.977,
      "mean_token_accuracy": 0.1946859911084175,
      "num_tokens": 117649842.0,
      "step": 53700
    },
    {
      "entropy": 5.716239738464355,
      "epoch": 5.3299920603414055,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002549793731062782,
      "loss": 4.9449,
      "mean_token_accuracy": 0.18920841366052626,
      "num_tokens": 117661901.0,
      "step": 53705
    },
    {
      "entropy": 5.738855981826783,
      "epoch": 5.330488289003573,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025494408240246593,
      "loss": 4.9983,
      "mean_token_accuracy": 0.19943312853574752,
      "num_tokens": 117673601.0,
      "step": 53710
    },
    {
      "entropy": 5.711894989013672,
      "epoch": 5.33098451766574,
      "grad_norm": 1.375,
      "learning_rate": 0.00025490879219599704,
      "loss": 4.9271,
      "mean_token_accuracy": 0.19962091892957687,
      "num_tokens": 117684327.0,
      "step": 53715
    },
    {
      "entropy": 5.697280216217041,
      "epoch": 5.331480746327908,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002548735024877469,
      "loss": 4.9727,
      "mean_token_accuracy": 0.19642882347106932,
      "num_tokens": 117695184.0,
      "step": 53720
    },
    {
      "entropy": 5.721102666854859,
      "epoch": 5.331976974990075,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002548382132785906,
      "loss": 5.0833,
      "mean_token_accuracy": 0.18991409242153168,
      "num_tokens": 117708136.0,
      "step": 53725
    },
    {
      "entropy": 5.760117053985596,
      "epoch": 5.332473203652243,
      "grad_norm": 1.15625,
      "learning_rate": 0.000254802924569403,
      "loss": 4.9708,
      "mean_token_accuracy": 0.1993483304977417,
      "num_tokens": 117719071.0,
      "step": 53730
    },
    {
      "entropy": 5.663540077209473,
      "epoch": 5.33296943231441,
      "grad_norm": 1.375,
      "learning_rate": 0.00025476763636105937,
      "loss": 4.8612,
      "mean_token_accuracy": 0.2023977130651474,
      "num_tokens": 117728675.0,
      "step": 53735
    },
    {
      "entropy": 5.687117624282837,
      "epoch": 5.333465660976578,
      "grad_norm": 1.15625,
      "learning_rate": 0.00025473234865443483,
      "loss": 4.9594,
      "mean_token_accuracy": 0.19523421823978424,
      "num_tokens": 117740309.0,
      "step": 53740
    },
    {
      "entropy": 5.626516246795655,
      "epoch": 5.333961889638745,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025469706145040435,
      "loss": 4.8861,
      "mean_token_accuracy": 0.2005854219198227,
      "num_tokens": 117750516.0,
      "step": 53745
    },
    {
      "entropy": 5.766209363937378,
      "epoch": 5.334458118300913,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000254661774749843,
      "loss": 4.9997,
      "mean_token_accuracy": 0.19188219606876372,
      "num_tokens": 117760978.0,
      "step": 53750
    },
    {
      "entropy": 5.739898347854615,
      "epoch": 5.33495434696308,
      "grad_norm": 1.375,
      "learning_rate": 0.00025462648855362587,
      "loss": 4.9429,
      "mean_token_accuracy": 0.20108372420072557,
      "num_tokens": 117771553.0,
      "step": 53755
    },
    {
      "entropy": 5.731152057647705,
      "epoch": 5.335450575625248,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002545912028626278,
      "loss": 4.9206,
      "mean_token_accuracy": 0.19442998319864274,
      "num_tokens": 117783019.0,
      "step": 53760
    },
    {
      "entropy": 5.688221979141235,
      "epoch": 5.335946804287415,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002545559176777239,
      "loss": 4.9191,
      "mean_token_accuracy": 0.20140604078769683,
      "num_tokens": 117794490.0,
      "step": 53765
    },
    {
      "entropy": 5.693714809417725,
      "epoch": 5.3364430329495836,
      "grad_norm": 1.25,
      "learning_rate": 0.00025452063299978935,
      "loss": 4.9226,
      "mean_token_accuracy": 0.19959866553544997,
      "num_tokens": 117806669.0,
      "step": 53770
    },
    {
      "entropy": 5.676909780502319,
      "epoch": 5.336939261611751,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025448534882969895,
      "loss": 4.9219,
      "mean_token_accuracy": 0.20057138055562973,
      "num_tokens": 117817279.0,
      "step": 53775
    },
    {
      "entropy": 5.686778497695923,
      "epoch": 5.3374354902739185,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002544500651683276,
      "loss": 4.915,
      "mean_token_accuracy": 0.19406181275844575,
      "num_tokens": 117827368.0,
      "step": 53780
    },
    {
      "entropy": 5.673054790496826,
      "epoch": 5.337931718936086,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002544147820165504,
      "loss": 4.9402,
      "mean_token_accuracy": 0.1899486243724823,
      "num_tokens": 117838465.0,
      "step": 53785
    },
    {
      "entropy": 5.622534894943238,
      "epoch": 5.3384279475982535,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002543794993752423,
      "loss": 4.8867,
      "mean_token_accuracy": 0.19857388287782668,
      "num_tokens": 117848910.0,
      "step": 53790
    },
    {
      "entropy": 5.7603589534759525,
      "epoch": 5.338924176260421,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000254344217245278,
      "loss": 5.0464,
      "mean_token_accuracy": 0.1931072309613228,
      "num_tokens": 117861104.0,
      "step": 53795
    },
    {
      "entropy": 5.662733125686645,
      "epoch": 5.339420404922588,
      "grad_norm": 1.375,
      "learning_rate": 0.00025430893562753277,
      "loss": 4.8442,
      "mean_token_accuracy": 0.20355254262685776,
      "num_tokens": 117871757.0,
      "step": 53800
    },
    {
      "entropy": 5.673283529281616,
      "epoch": 5.339916633584756,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002542736545228813,
      "loss": 4.8286,
      "mean_token_accuracy": 0.2062096655368805,
      "num_tokens": 117882166.0,
      "step": 53805
    },
    {
      "entropy": 5.6430487632751465,
      "epoch": 5.340412862246923,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002542383739321985,
      "loss": 4.9054,
      "mean_token_accuracy": 0.19840016216039658,
      "num_tokens": 117893260.0,
      "step": 53810
    },
    {
      "entropy": 5.704679775238037,
      "epoch": 5.340909090909091,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025420309385635936,
      "loss": 4.9257,
      "mean_token_accuracy": 0.19584948122501372,
      "num_tokens": 117903381.0,
      "step": 53815
    },
    {
      "entropy": 5.663934755325317,
      "epoch": 5.341405319571258,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002541678142962386,
      "loss": 4.9114,
      "mean_token_accuracy": 0.1975707232952118,
      "num_tokens": 117913752.0,
      "step": 53820
    },
    {
      "entropy": 5.677017688751221,
      "epoch": 5.341901548233426,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025413253525271125,
      "loss": 4.9595,
      "mean_token_accuracy": 0.1982360765337944,
      "num_tokens": 117924272.0,
      "step": 53825
    },
    {
      "entropy": 5.664569091796875,
      "epoch": 5.342397776895593,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025409725672665204,
      "loss": 4.8725,
      "mean_token_accuracy": 0.19995699673891068,
      "num_tokens": 117934505.0,
      "step": 53830
    },
    {
      "entropy": 5.687144994735718,
      "epoch": 5.342894005557761,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002540619787189358,
      "loss": 4.935,
      "mean_token_accuracy": 0.20109648257493973,
      "num_tokens": 117944586.0,
      "step": 53835
    },
    {
      "entropy": 5.7084263324737545,
      "epoch": 5.343390234219928,
      "grad_norm": 1.25,
      "learning_rate": 0.00025402670123043735,
      "loss": 4.984,
      "mean_token_accuracy": 0.18950926065444945,
      "num_tokens": 117955355.0,
      "step": 53840
    },
    {
      "entropy": 5.680844068527222,
      "epoch": 5.343886462882096,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025399142426203155,
      "loss": 4.8991,
      "mean_token_accuracy": 0.20261409133672714,
      "num_tokens": 117965707.0,
      "step": 53845
    },
    {
      "entropy": 5.7675457954406735,
      "epoch": 5.344382691544263,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002539561478145931,
      "loss": 5.023,
      "mean_token_accuracy": 0.19009212106466294,
      "num_tokens": 117975627.0,
      "step": 53850
    },
    {
      "entropy": 5.708341073989868,
      "epoch": 5.3448789202064315,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002539208718889969,
      "loss": 4.9227,
      "mean_token_accuracy": 0.19884071499109268,
      "num_tokens": 117985519.0,
      "step": 53855
    },
    {
      "entropy": 5.733967685699463,
      "epoch": 5.345375148868599,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025388559648611763,
      "loss": 5.0438,
      "mean_token_accuracy": 0.19096201360225679,
      "num_tokens": 117997234.0,
      "step": 53860
    },
    {
      "entropy": 5.6952657222747805,
      "epoch": 5.3458713775307665,
      "grad_norm": 1.484375,
      "learning_rate": 0.00025385032160683016,
      "loss": 4.914,
      "mean_token_accuracy": 0.20048919916152955,
      "num_tokens": 118007935.0,
      "step": 53865
    },
    {
      "entropy": 5.633121728897095,
      "epoch": 5.346367606192934,
      "grad_norm": 1.25,
      "learning_rate": 0.00025381504725200906,
      "loss": 4.8504,
      "mean_token_accuracy": 0.20107623487710952,
      "num_tokens": 118018550.0,
      "step": 53870
    },
    {
      "entropy": 5.715510702133178,
      "epoch": 5.346863834855101,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025377977342252913,
      "loss": 4.9799,
      "mean_token_accuracy": 0.190916445851326,
      "num_tokens": 118028697.0,
      "step": 53875
    },
    {
      "entropy": 5.677280473709106,
      "epoch": 5.347360063517269,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000253744500119265,
      "loss": 4.8977,
      "mean_token_accuracy": 0.1987181544303894,
      "num_tokens": 118039290.0,
      "step": 53880
    },
    {
      "entropy": 5.660555505752564,
      "epoch": 5.347856292179436,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002537092273430916,
      "loss": 4.9294,
      "mean_token_accuracy": 0.19519914388656617,
      "num_tokens": 118050405.0,
      "step": 53885
    },
    {
      "entropy": 5.635214757919312,
      "epoch": 5.348352520841604,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025367395509488344,
      "loss": 4.939,
      "mean_token_accuracy": 0.19547900259494783,
      "num_tokens": 118063236.0,
      "step": 53890
    },
    {
      "entropy": 5.760878324508667,
      "epoch": 5.348848749503771,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002536386833755153,
      "loss": 4.9474,
      "mean_token_accuracy": 0.19258308112621308,
      "num_tokens": 118075173.0,
      "step": 53895
    },
    {
      "entropy": 5.725471448898316,
      "epoch": 5.349344978165939,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002536034121858617,
      "loss": 4.9819,
      "mean_token_accuracy": 0.1929209351539612,
      "num_tokens": 118086126.0,
      "step": 53900
    },
    {
      "entropy": 5.740633010864258,
      "epoch": 5.349841206828106,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025356814152679747,
      "loss": 4.9293,
      "mean_token_accuracy": 0.2030064046382904,
      "num_tokens": 118097226.0,
      "step": 53905
    },
    {
      "entropy": 5.728962707519531,
      "epoch": 5.350337435490274,
      "grad_norm": 1.296875,
      "learning_rate": 0.000253532871399197,
      "loss": 4.893,
      "mean_token_accuracy": 0.20278133004903792,
      "num_tokens": 118108754.0,
      "step": 53910
    },
    {
      "entropy": 5.738568592071533,
      "epoch": 5.350833664152441,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002534976018039352,
      "loss": 4.9479,
      "mean_token_accuracy": 0.20239098072052003,
      "num_tokens": 118119238.0,
      "step": 53915
    },
    {
      "entropy": 5.644330739974976,
      "epoch": 5.351329892814609,
      "grad_norm": 1.1875,
      "learning_rate": 0.00025346233274188643,
      "loss": 4.8684,
      "mean_token_accuracy": 0.20750463604927064,
      "num_tokens": 118130266.0,
      "step": 53920
    },
    {
      "entropy": 5.731369733810425,
      "epoch": 5.351826121476776,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002534270642139255,
      "loss": 4.9767,
      "mean_token_accuracy": 0.20106178224086763,
      "num_tokens": 118142288.0,
      "step": 53925
    },
    {
      "entropy": 5.571757078170776,
      "epoch": 5.352322350138944,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025339179622092686,
      "loss": 4.8163,
      "mean_token_accuracy": 0.21086161881685256,
      "num_tokens": 118154293.0,
      "step": 53930
    },
    {
      "entropy": 5.657099914550781,
      "epoch": 5.352818578801111,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025335652876376513,
      "loss": 4.9002,
      "mean_token_accuracy": 0.19676988571882248,
      "num_tokens": 118164829.0,
      "step": 53935
    },
    {
      "entropy": 5.6855357646942135,
      "epoch": 5.3533148074632795,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002533212618433147,
      "loss": 4.9608,
      "mean_token_accuracy": 0.19528622925281525,
      "num_tokens": 118176775.0,
      "step": 53940
    },
    {
      "entropy": 5.750494480133057,
      "epoch": 5.353811036125447,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002532859954604504,
      "loss": 5.0133,
      "mean_token_accuracy": 0.18763754814863204,
      "num_tokens": 118188646.0,
      "step": 53945
    },
    {
      "entropy": 5.711465501785279,
      "epoch": 5.354307264787614,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002532507296160466,
      "loss": 5.0103,
      "mean_token_accuracy": 0.19615420699119568,
      "num_tokens": 118202279.0,
      "step": 53950
    },
    {
      "entropy": 5.669117164611817,
      "epoch": 5.354803493449782,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002532154643109779,
      "loss": 4.8998,
      "mean_token_accuracy": 0.1985538959503174,
      "num_tokens": 118212965.0,
      "step": 53955
    },
    {
      "entropy": 5.732746028900147,
      "epoch": 5.355299722111949,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002531801995461187,
      "loss": 4.9569,
      "mean_token_accuracy": 0.19790810942649842,
      "num_tokens": 118223607.0,
      "step": 53960
    },
    {
      "entropy": 5.674152803421021,
      "epoch": 5.355795950774117,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025314493532234357,
      "loss": 4.9517,
      "mean_token_accuracy": 0.2012337788939476,
      "num_tokens": 118234733.0,
      "step": 53965
    },
    {
      "entropy": 5.644043016433716,
      "epoch": 5.356292179436284,
      "grad_norm": 1.25,
      "learning_rate": 0.0002531096716405268,
      "loss": 4.859,
      "mean_token_accuracy": 0.2013933464884758,
      "num_tokens": 118245828.0,
      "step": 53970
    },
    {
      "entropy": 5.699192762374878,
      "epoch": 5.356788408098452,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002530744085015432,
      "loss": 4.9547,
      "mean_token_accuracy": 0.19576093554496765,
      "num_tokens": 118257184.0,
      "step": 53975
    },
    {
      "entropy": 5.74273362159729,
      "epoch": 5.357284636760619,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002530391459062669,
      "loss": 4.9755,
      "mean_token_accuracy": 0.19423541724681853,
      "num_tokens": 118268039.0,
      "step": 53980
    },
    {
      "entropy": 5.78689284324646,
      "epoch": 5.357780865422787,
      "grad_norm": 1.25,
      "learning_rate": 0.0002530038838555725,
      "loss": 5.0162,
      "mean_token_accuracy": 0.1911650002002716,
      "num_tokens": 118278435.0,
      "step": 53985
    },
    {
      "entropy": 5.722899532318115,
      "epoch": 5.358277094084954,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025296862235033436,
      "loss": 4.9201,
      "mean_token_accuracy": 0.199762025475502,
      "num_tokens": 118288780.0,
      "step": 53990
    },
    {
      "entropy": 5.705032062530518,
      "epoch": 5.358773322747122,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025293336139142693,
      "loss": 4.9297,
      "mean_token_accuracy": 0.20088773518800734,
      "num_tokens": 118299157.0,
      "step": 53995
    },
    {
      "entropy": 5.7147942066192625,
      "epoch": 5.359269551409289,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002528981009797245,
      "loss": 4.9682,
      "mean_token_accuracy": 0.19973773509263992,
      "num_tokens": 118310784.0,
      "step": 54000
    },
    {
      "epoch": 5.359269551409289,
      "eval_entropy": 5.479572773886056,
      "eval_loss": 5.082475662231445,
      "eval_mean_token_accuracy": 0.19524409709665122,
      "eval_num_tokens": 118310784.0,
      "eval_runtime": 20.5557,
      "eval_samples_per_second": 1637.938,
      "eval_steps_per_second": 204.761,
      "step": 54000
    },
    {
      "entropy": 5.676473569869995,
      "epoch": 5.359765780071457,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002528628411161017,
      "loss": 4.8853,
      "mean_token_accuracy": 0.20507437288761138,
      "num_tokens": 118320987.0,
      "step": 54005
    },
    {
      "entropy": 5.681708955764771,
      "epoch": 5.360262008733624,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002528275818014326,
      "loss": 4.9367,
      "mean_token_accuracy": 0.19523832201957703,
      "num_tokens": 118331664.0,
      "step": 54010
    },
    {
      "entropy": 5.680180454254151,
      "epoch": 5.360758237395792,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025279232303659184,
      "loss": 4.9635,
      "mean_token_accuracy": 0.1918136641383171,
      "num_tokens": 118342014.0,
      "step": 54015
    },
    {
      "entropy": 5.7496977806091305,
      "epoch": 5.361254466057959,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025275706482245355,
      "loss": 5.0248,
      "mean_token_accuracy": 0.1857861742377281,
      "num_tokens": 118351290.0,
      "step": 54020
    },
    {
      "entropy": 5.66893253326416,
      "epoch": 5.361750694720127,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025272180715989206,
      "loss": 4.8864,
      "mean_token_accuracy": 0.20138483494520187,
      "num_tokens": 118363188.0,
      "step": 54025
    },
    {
      "entropy": 5.678731298446655,
      "epoch": 5.362246923382295,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002526865500497817,
      "loss": 4.8811,
      "mean_token_accuracy": 0.1986805111169815,
      "num_tokens": 118373313.0,
      "step": 54030
    },
    {
      "entropy": 5.699424982070923,
      "epoch": 5.362743152044462,
      "grad_norm": 1.390625,
      "learning_rate": 0.000252651293492997,
      "loss": 4.9427,
      "mean_token_accuracy": 0.19404683411121368,
      "num_tokens": 118384208.0,
      "step": 54035
    },
    {
      "entropy": 5.686799812316894,
      "epoch": 5.36323938070663,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000252616037490412,
      "loss": 4.9267,
      "mean_token_accuracy": 0.1972465068101883,
      "num_tokens": 118395207.0,
      "step": 54040
    },
    {
      "entropy": 5.665155220031738,
      "epoch": 5.363735609368797,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025258078204290095,
      "loss": 4.9483,
      "mean_token_accuracy": 0.192159104347229,
      "num_tokens": 118407076.0,
      "step": 54045
    },
    {
      "entropy": 5.67129225730896,
      "epoch": 5.364231838030965,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025254552715133833,
      "loss": 4.947,
      "mean_token_accuracy": 0.19444359242916107,
      "num_tokens": 118419104.0,
      "step": 54050
    },
    {
      "entropy": 5.67891035079956,
      "epoch": 5.364728066693132,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002525102728165981,
      "loss": 4.885,
      "mean_token_accuracy": 0.20201606154441834,
      "num_tokens": 118428593.0,
      "step": 54055
    },
    {
      "entropy": 5.716673898696899,
      "epoch": 5.3652242953553,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002524750190395547,
      "loss": 4.9095,
      "mean_token_accuracy": 0.20340223908424376,
      "num_tokens": 118439241.0,
      "step": 54060
    },
    {
      "entropy": 5.707439422607422,
      "epoch": 5.365720524017467,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025243976582108236,
      "loss": 4.9654,
      "mean_token_accuracy": 0.20001774430274963,
      "num_tokens": 118450787.0,
      "step": 54065
    },
    {
      "entropy": 5.666901683807373,
      "epoch": 5.366216752679635,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002524045131620552,
      "loss": 4.8783,
      "mean_token_accuracy": 0.1970638945698738,
      "num_tokens": 118460723.0,
      "step": 54070
    },
    {
      "entropy": 5.71071047782898,
      "epoch": 5.366712981341802,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002523692610633474,
      "loss": 4.9791,
      "mean_token_accuracy": 0.198141385614872,
      "num_tokens": 118473128.0,
      "step": 54075
    },
    {
      "entropy": 5.722633695602417,
      "epoch": 5.36720921000397,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025233400952583314,
      "loss": 4.9313,
      "mean_token_accuracy": 0.2016165927052498,
      "num_tokens": 118484492.0,
      "step": 54080
    },
    {
      "entropy": 5.729791831970215,
      "epoch": 5.367705438666137,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002522987585503866,
      "loss": 4.979,
      "mean_token_accuracy": 0.1984617680311203,
      "num_tokens": 118494559.0,
      "step": 54085
    },
    {
      "entropy": 5.684174919128418,
      "epoch": 5.368201667328305,
      "grad_norm": 1.203125,
      "learning_rate": 0.00025226350813788194,
      "loss": 4.9122,
      "mean_token_accuracy": 0.19992267489433288,
      "num_tokens": 118505700.0,
      "step": 54090
    },
    {
      "entropy": 5.698917865753174,
      "epoch": 5.368697895990472,
      "grad_norm": 1.125,
      "learning_rate": 0.00025222825828919327,
      "loss": 4.9305,
      "mean_token_accuracy": 0.20031990557909013,
      "num_tokens": 118517218.0,
      "step": 54095
    },
    {
      "entropy": 5.7432492733001705,
      "epoch": 5.36919412465264,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002521930090051948,
      "loss": 5.0323,
      "mean_token_accuracy": 0.19150184243917465,
      "num_tokens": 118528381.0,
      "step": 54100
    },
    {
      "entropy": 5.680529308319092,
      "epoch": 5.369690353314807,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025215776028676045,
      "loss": 4.936,
      "mean_token_accuracy": 0.19617670327425002,
      "num_tokens": 118540986.0,
      "step": 54105
    },
    {
      "entropy": 5.6880029201507565,
      "epoch": 5.370186581976975,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002521225121347646,
      "loss": 4.9194,
      "mean_token_accuracy": 0.20144339352846147,
      "num_tokens": 118551403.0,
      "step": 54110
    },
    {
      "entropy": 5.657295370101929,
      "epoch": 5.370682810639143,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002520872645500809,
      "loss": 4.8842,
      "mean_token_accuracy": 0.19726910144090654,
      "num_tokens": 118562565.0,
      "step": 54115
    },
    {
      "entropy": 5.670323896408081,
      "epoch": 5.37117903930131,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025205201753358386,
      "loss": 4.9176,
      "mean_token_accuracy": 0.1994251474738121,
      "num_tokens": 118572488.0,
      "step": 54120
    },
    {
      "entropy": 5.715347576141357,
      "epoch": 5.371675267963478,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002520167710861472,
      "loss": 4.9332,
      "mean_token_accuracy": 0.19473669677972794,
      "num_tokens": 118583072.0,
      "step": 54125
    },
    {
      "entropy": 5.644540929794312,
      "epoch": 5.372171496625645,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025198152520864526,
      "loss": 4.9167,
      "mean_token_accuracy": 0.20514025390148163,
      "num_tokens": 118593203.0,
      "step": 54130
    },
    {
      "entropy": 5.681192541122437,
      "epoch": 5.372667725287813,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00025194627990195184,
      "loss": 4.9443,
      "mean_token_accuracy": 0.19978665858507155,
      "num_tokens": 118604066.0,
      "step": 54135
    },
    {
      "entropy": 5.6214502334594725,
      "epoch": 5.37316395394998,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025191103516694093,
      "loss": 4.8681,
      "mean_token_accuracy": 0.20095293670892717,
      "num_tokens": 118614831.0,
      "step": 54140
    },
    {
      "entropy": 5.7326192378997805,
      "epoch": 5.373660182612148,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002518757910044866,
      "loss": 5.0039,
      "mean_token_accuracy": 0.19530429989099501,
      "num_tokens": 118625519.0,
      "step": 54145
    },
    {
      "entropy": 5.711036109924317,
      "epoch": 5.374156411274315,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025184054741546287,
      "loss": 4.9609,
      "mean_token_accuracy": 0.19679283797740937,
      "num_tokens": 118638149.0,
      "step": 54150
    },
    {
      "entropy": 5.692322397232056,
      "epoch": 5.374652639936483,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025180530440074367,
      "loss": 4.8633,
      "mean_token_accuracy": 0.19240207374095916,
      "num_tokens": 118649075.0,
      "step": 54155
    },
    {
      "entropy": 5.688644504547119,
      "epoch": 5.37514886859865,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00025177006196120294,
      "loss": 4.9571,
      "mean_token_accuracy": 0.20286553651094436,
      "num_tokens": 118660267.0,
      "step": 54160
    },
    {
      "entropy": 5.630155038833618,
      "epoch": 5.375645097260818,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002517348200977146,
      "loss": 4.9066,
      "mean_token_accuracy": 0.20074122846126558,
      "num_tokens": 118670790.0,
      "step": 54165
    },
    {
      "entropy": 5.643966150283814,
      "epoch": 5.376141325922985,
      "grad_norm": 1.375,
      "learning_rate": 0.0002516995788111526,
      "loss": 4.9103,
      "mean_token_accuracy": 0.20156480818986894,
      "num_tokens": 118680852.0,
      "step": 54170
    },
    {
      "entropy": 5.669248104095459,
      "epoch": 5.376637554585153,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002516643381023908,
      "loss": 4.9246,
      "mean_token_accuracy": 0.2006835460662842,
      "num_tokens": 118691322.0,
      "step": 54175
    },
    {
      "entropy": 5.683677339553833,
      "epoch": 5.37713378324732,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002516290979723031,
      "loss": 4.9028,
      "mean_token_accuracy": 0.19482124745845794,
      "num_tokens": 118702460.0,
      "step": 54180
    },
    {
      "entropy": 5.679493808746338,
      "epoch": 5.3776300119094875,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002515938584217635,
      "loss": 4.9012,
      "mean_token_accuracy": 0.19928363114595413,
      "num_tokens": 118714762.0,
      "step": 54185
    },
    {
      "entropy": 5.766822099685669,
      "epoch": 5.378126240571655,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002515586194516458,
      "loss": 5.0229,
      "mean_token_accuracy": 0.1980303034186363,
      "num_tokens": 118726136.0,
      "step": 54190
    },
    {
      "entropy": 5.737609243392944,
      "epoch": 5.378622469233823,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025152338106282374,
      "loss": 5.0275,
      "mean_token_accuracy": 0.18922768980264665,
      "num_tokens": 118737260.0,
      "step": 54195
    },
    {
      "entropy": 5.708420610427856,
      "epoch": 5.379118697895991,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025148814325617126,
      "loss": 4.9182,
      "mean_token_accuracy": 0.18813695311546325,
      "num_tokens": 118748443.0,
      "step": 54200
    },
    {
      "entropy": 5.6547482967376705,
      "epoch": 5.379614926558158,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025145290603256217,
      "loss": 4.8931,
      "mean_token_accuracy": 0.19859057664871216,
      "num_tokens": 118760290.0,
      "step": 54205
    },
    {
      "entropy": 5.699523305892944,
      "epoch": 5.380111155220326,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025141766939287027,
      "loss": 4.8974,
      "mean_token_accuracy": 0.1970726877450943,
      "num_tokens": 118771242.0,
      "step": 54210
    },
    {
      "entropy": 5.675407981872558,
      "epoch": 5.380607383882493,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002513824333379694,
      "loss": 4.9011,
      "mean_token_accuracy": 0.20116450190544127,
      "num_tokens": 118780147.0,
      "step": 54215
    },
    {
      "entropy": 5.669814872741699,
      "epoch": 5.381103612544661,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002513471978687332,
      "loss": 4.8526,
      "mean_token_accuracy": 0.20192258954048156,
      "num_tokens": 118789666.0,
      "step": 54220
    },
    {
      "entropy": 5.678431272506714,
      "epoch": 5.381599841206828,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002513119629860356,
      "loss": 4.8974,
      "mean_token_accuracy": 0.2005837857723236,
      "num_tokens": 118800259.0,
      "step": 54225
    },
    {
      "entropy": 5.708706855773926,
      "epoch": 5.382096069868996,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002512767286907503,
      "loss": 5.0042,
      "mean_token_accuracy": 0.19082169830799103,
      "num_tokens": 118810924.0,
      "step": 54230
    },
    {
      "entropy": 5.669407749176026,
      "epoch": 5.382592298531163,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025124149498375085,
      "loss": 4.9394,
      "mean_token_accuracy": 0.20078413784503937,
      "num_tokens": 118821966.0,
      "step": 54235
    },
    {
      "entropy": 5.749064493179321,
      "epoch": 5.383088527193331,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025120626186591125,
      "loss": 4.996,
      "mean_token_accuracy": 0.1952737495303154,
      "num_tokens": 118834084.0,
      "step": 54240
    },
    {
      "entropy": 5.737237405776978,
      "epoch": 5.383584755855498,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002511710293381051,
      "loss": 4.9178,
      "mean_token_accuracy": 0.19740509986877441,
      "num_tokens": 118845541.0,
      "step": 54245
    },
    {
      "entropy": 5.72983193397522,
      "epoch": 5.384080984517666,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002511357974012061,
      "loss": 5.0111,
      "mean_token_accuracy": 0.19585912823677062,
      "num_tokens": 118857602.0,
      "step": 54250
    },
    {
      "entropy": 5.653813695907592,
      "epoch": 5.384577213179833,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025110056605608783,
      "loss": 4.8628,
      "mean_token_accuracy": 0.20161721110343933,
      "num_tokens": 118868194.0,
      "step": 54255
    },
    {
      "entropy": 5.685593748092652,
      "epoch": 5.3850734418420005,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002510653353036241,
      "loss": 4.941,
      "mean_token_accuracy": 0.1977750614285469,
      "num_tokens": 118880730.0,
      "step": 54260
    },
    {
      "entropy": 5.695058870315552,
      "epoch": 5.385569670504168,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002510301051446883,
      "loss": 4.9667,
      "mean_token_accuracy": 0.19595859795808793,
      "num_tokens": 118892277.0,
      "step": 54265
    },
    {
      "entropy": 5.7478461265563965,
      "epoch": 5.3860658991663355,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002509948755801543,
      "loss": 4.9542,
      "mean_token_accuracy": 0.19207381457090378,
      "num_tokens": 118902267.0,
      "step": 54270
    },
    {
      "entropy": 5.737315893173218,
      "epoch": 5.386562127828503,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025095964661089583,
      "loss": 5.0073,
      "mean_token_accuracy": 0.18747153580188752,
      "num_tokens": 118912817.0,
      "step": 54275
    },
    {
      "entropy": 5.690695858001709,
      "epoch": 5.387058356490671,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002509244182377862,
      "loss": 4.9158,
      "mean_token_accuracy": 0.19713944792747498,
      "num_tokens": 118923039.0,
      "step": 54280
    },
    {
      "entropy": 5.7203411102294925,
      "epoch": 5.387554585152839,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025088919046169916,
      "loss": 5.0064,
      "mean_token_accuracy": 0.19363737106323242,
      "num_tokens": 118934368.0,
      "step": 54285
    },
    {
      "entropy": 5.728286552429199,
      "epoch": 5.388050813815006,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025085396328350824,
      "loss": 4.9391,
      "mean_token_accuracy": 0.19928446263074875,
      "num_tokens": 118945286.0,
      "step": 54290
    },
    {
      "entropy": 5.730503797531128,
      "epoch": 5.388547042477174,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00025081873670408693,
      "loss": 4.9427,
      "mean_token_accuracy": 0.19966192692518234,
      "num_tokens": 118956438.0,
      "step": 54295
    },
    {
      "entropy": 5.734478712081909,
      "epoch": 5.389043271139341,
      "grad_norm": 1.25,
      "learning_rate": 0.0002507835107243089,
      "loss": 4.9899,
      "mean_token_accuracy": 0.1897968754172325,
      "num_tokens": 118967667.0,
      "step": 54300
    },
    {
      "entropy": 5.689482498168945,
      "epoch": 5.389539499801509,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025074828534504763,
      "loss": 4.8912,
      "mean_token_accuracy": 0.20282009840011597,
      "num_tokens": 118977010.0,
      "step": 54305
    },
    {
      "entropy": 5.703799104690551,
      "epoch": 5.390035728463676,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002507130605671766,
      "loss": 5.0225,
      "mean_token_accuracy": 0.19203834980726242,
      "num_tokens": 118988368.0,
      "step": 54310
    },
    {
      "entropy": 5.708917140960693,
      "epoch": 5.390531957125844,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002506778363915693,
      "loss": 4.8975,
      "mean_token_accuracy": 0.20220425724983215,
      "num_tokens": 118999845.0,
      "step": 54315
    },
    {
      "entropy": 5.674118614196777,
      "epoch": 5.391028185788011,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002506426128190993,
      "loss": 4.9031,
      "mean_token_accuracy": 0.19499932378530502,
      "num_tokens": 119011041.0,
      "step": 54320
    },
    {
      "entropy": 5.725915908813477,
      "epoch": 5.391524414450179,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00025060738985064,
      "loss": 4.94,
      "mean_token_accuracy": 0.1989818826317787,
      "num_tokens": 119020983.0,
      "step": 54325
    },
    {
      "entropy": 5.6324526309967045,
      "epoch": 5.392020643112346,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002505721674870649,
      "loss": 4.8683,
      "mean_token_accuracy": 0.20279012322425843,
      "num_tokens": 119031292.0,
      "step": 54330
    },
    {
      "entropy": 5.683555221557617,
      "epoch": 5.3925168717745136,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002505369457292474,
      "loss": 5.0052,
      "mean_token_accuracy": 0.19424942284822463,
      "num_tokens": 119042244.0,
      "step": 54335
    },
    {
      "entropy": 5.719524192810058,
      "epoch": 5.393013100436681,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025050172457806094,
      "loss": 4.9666,
      "mean_token_accuracy": 0.1962016925215721,
      "num_tokens": 119055183.0,
      "step": 54340
    },
    {
      "entropy": 5.740852689743042,
      "epoch": 5.3935093290988485,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025046650403437893,
      "loss": 5.0057,
      "mean_token_accuracy": 0.1930099442601204,
      "num_tokens": 119066128.0,
      "step": 54345
    },
    {
      "entropy": 5.732615947723389,
      "epoch": 5.394005557761016,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002504312840990747,
      "loss": 4.9731,
      "mean_token_accuracy": 0.20175046771764754,
      "num_tokens": 119077147.0,
      "step": 54350
    },
    {
      "entropy": 5.77218713760376,
      "epoch": 5.3945017864231835,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002503960647730217,
      "loss": 5.0234,
      "mean_token_accuracy": 0.18881407529115676,
      "num_tokens": 119087329.0,
      "step": 54355
    },
    {
      "entropy": 5.720679092407226,
      "epoch": 5.394998015085351,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00025036084605709324,
      "loss": 4.9549,
      "mean_token_accuracy": 0.19661808162927627,
      "num_tokens": 119098681.0,
      "step": 54360
    },
    {
      "entropy": 5.678655624389648,
      "epoch": 5.395494243747519,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025032562795216284,
      "loss": 4.8029,
      "mean_token_accuracy": 0.21131485998630523,
      "num_tokens": 119108955.0,
      "step": 54365
    },
    {
      "entropy": 5.693073415756226,
      "epoch": 5.395990472409687,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025029041045910357,
      "loss": 4.9149,
      "mean_token_accuracy": 0.20012434870004653,
      "num_tokens": 119119383.0,
      "step": 54370
    },
    {
      "entropy": 5.680264568328857,
      "epoch": 5.396486701071854,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025025519357878897,
      "loss": 4.8889,
      "mean_token_accuracy": 0.2011472463607788,
      "num_tokens": 119129219.0,
      "step": 54375
    },
    {
      "entropy": 5.701133489608765,
      "epoch": 5.396982929734022,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025021997731209213,
      "loss": 4.9963,
      "mean_token_accuracy": 0.19521158635616304,
      "num_tokens": 119141843.0,
      "step": 54380
    },
    {
      "entropy": 5.654879570007324,
      "epoch": 5.397479158396189,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002501847616598865,
      "loss": 4.9346,
      "mean_token_accuracy": 0.1899394929409027,
      "num_tokens": 119152927.0,
      "step": 54385
    },
    {
      "entropy": 5.721087217330933,
      "epoch": 5.397975387058357,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025014954662304535,
      "loss": 4.9481,
      "mean_token_accuracy": 0.19298589676618577,
      "num_tokens": 119164500.0,
      "step": 54390
    },
    {
      "entropy": 5.701238918304443,
      "epoch": 5.398471615720524,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002501143322024419,
      "loss": 4.9195,
      "mean_token_accuracy": 0.19910712987184526,
      "num_tokens": 119174418.0,
      "step": 54395
    },
    {
      "entropy": 5.735565423965454,
      "epoch": 5.398967844382692,
      "grad_norm": 1.375,
      "learning_rate": 0.0002500791183989494,
      "loss": 5.0066,
      "mean_token_accuracy": 0.19276413917541504,
      "num_tokens": 119185381.0,
      "step": 54400
    },
    {
      "entropy": 5.687144088745117,
      "epoch": 5.399464073044859,
      "grad_norm": 1.359375,
      "learning_rate": 0.000250043905213441,
      "loss": 4.9238,
      "mean_token_accuracy": 0.20292137414216996,
      "num_tokens": 119195696.0,
      "step": 54405
    },
    {
      "entropy": 5.779624032974243,
      "epoch": 5.399960301707027,
      "grad_norm": 1.484375,
      "learning_rate": 0.00025000869264678994,
      "loss": 5.0503,
      "mean_token_accuracy": 0.1838023841381073,
      "num_tokens": 119206391.0,
      "step": 54410
    },
    {
      "entropy": 5.7763190269470215,
      "epoch": 5.400456530369194,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024997348069986957,
      "loss": 4.9985,
      "mean_token_accuracy": 0.1933579683303833,
      "num_tokens": 119219522.0,
      "step": 54415
    },
    {
      "entropy": 5.732142353057862,
      "epoch": 5.4009527590313615,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002499382693735529,
      "loss": 4.9023,
      "mean_token_accuracy": 0.20378483831882477,
      "num_tokens": 119230037.0,
      "step": 54420
    },
    {
      "entropy": 5.765390491485595,
      "epoch": 5.401448987693529,
      "grad_norm": 1.25,
      "learning_rate": 0.00024990305866871317,
      "loss": 4.944,
      "mean_token_accuracy": 0.19602807313203813,
      "num_tokens": 119242000.0,
      "step": 54425
    },
    {
      "entropy": 5.677294397354126,
      "epoch": 5.4019452163556965,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002498678485862234,
      "loss": 4.8973,
      "mean_token_accuracy": 0.19964162558317183,
      "num_tokens": 119253742.0,
      "step": 54430
    },
    {
      "entropy": 5.668656539916992,
      "epoch": 5.402441445017864,
      "grad_norm": 1.28125,
      "learning_rate": 0.000249832639126957,
      "loss": 4.8914,
      "mean_token_accuracy": 0.2008909836411476,
      "num_tokens": 119264894.0,
      "step": 54435
    },
    {
      "entropy": 5.792925786972046,
      "epoch": 5.402937673680031,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002497974302917867,
      "loss": 5.0432,
      "mean_token_accuracy": 0.19330069720745086,
      "num_tokens": 119276465.0,
      "step": 54440
    },
    {
      "entropy": 5.733495378494263,
      "epoch": 5.403433902342199,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000249762222081586,
      "loss": 5.0276,
      "mean_token_accuracy": 0.1888740047812462,
      "num_tokens": 119288194.0,
      "step": 54445
    },
    {
      "entropy": 5.699374008178711,
      "epoch": 5.403930131004367,
      "grad_norm": 1.25,
      "learning_rate": 0.00024972701449722773,
      "loss": 4.9102,
      "mean_token_accuracy": 0.20534184724092483,
      "num_tokens": 119298815.0,
      "step": 54450
    },
    {
      "entropy": 5.726843976974488,
      "epoch": 5.404426359666535,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024969180753958513,
      "loss": 4.9926,
      "mean_token_accuracy": 0.1925747349858284,
      "num_tokens": 119308553.0,
      "step": 54455
    },
    {
      "entropy": 5.627767705917359,
      "epoch": 5.404922588328702,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00024965660120953117,
      "loss": 4.8641,
      "mean_token_accuracy": 0.203250390291214,
      "num_tokens": 119319774.0,
      "step": 54460
    },
    {
      "entropy": 5.7172990322113035,
      "epoch": 5.40541881699087,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002496213955079388,
      "loss": 4.9448,
      "mean_token_accuracy": 0.19807505309581758,
      "num_tokens": 119330499.0,
      "step": 54465
    },
    {
      "entropy": 5.72076735496521,
      "epoch": 5.405915045653037,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024958619043568105,
      "loss": 4.9288,
      "mean_token_accuracy": 0.20036333948373794,
      "num_tokens": 119341122.0,
      "step": 54470
    },
    {
      "entropy": 5.718708324432373,
      "epoch": 5.406411274315205,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024955098599363116,
      "loss": 4.9594,
      "mean_token_accuracy": 0.19848239719867705,
      "num_tokens": 119353223.0,
      "step": 54475
    },
    {
      "entropy": 5.717133045196533,
      "epoch": 5.406907502977372,
      "grad_norm": 1.375,
      "learning_rate": 0.000249515782182662,
      "loss": 4.9135,
      "mean_token_accuracy": 0.20760058015584945,
      "num_tokens": 119364216.0,
      "step": 54480
    },
    {
      "entropy": 5.696423435211182,
      "epoch": 5.40740373163954,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024948057900364646,
      "loss": 4.8725,
      "mean_token_accuracy": 0.19690958559513091,
      "num_tokens": 119374720.0,
      "step": 54485
    },
    {
      "entropy": 5.764861011505127,
      "epoch": 5.407899960301707,
      "grad_norm": 1.140625,
      "learning_rate": 0.00024944537645745765,
      "loss": 5.0572,
      "mean_token_accuracy": 0.18888060450553895,
      "num_tokens": 119386519.0,
      "step": 54490
    },
    {
      "entropy": 5.6531538486480715,
      "epoch": 5.4083961889638745,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024941017454496836,
      "loss": 4.8889,
      "mean_token_accuracy": 0.2052697241306305,
      "num_tokens": 119396806.0,
      "step": 54495
    },
    {
      "entropy": 5.649697780609131,
      "epoch": 5.408892417626042,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002493749732670515,
      "loss": 4.9154,
      "mean_token_accuracy": 0.1938018813729286,
      "num_tokens": 119407428.0,
      "step": 54500
    },
    {
      "entropy": 5.68996033668518,
      "epoch": 5.4093886462882095,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024933977262458026,
      "loss": 4.9098,
      "mean_token_accuracy": 0.20020380467176438,
      "num_tokens": 119418403.0,
      "step": 54505
    },
    {
      "entropy": 5.822287940979004,
      "epoch": 5.409884874950377,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024930457261842734,
      "loss": 5.0401,
      "mean_token_accuracy": 0.19326225370168687,
      "num_tokens": 119429768.0,
      "step": 54510
    },
    {
      "entropy": 5.627902412414551,
      "epoch": 5.410381103612544,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024926937324946556,
      "loss": 4.7867,
      "mean_token_accuracy": 0.21024577021598817,
      "num_tokens": 119440149.0,
      "step": 54515
    },
    {
      "entropy": 5.735311651229859,
      "epoch": 5.410877332274712,
      "grad_norm": 1.296875,
      "learning_rate": 0.000249234174518568,
      "loss": 5.018,
      "mean_token_accuracy": 0.19316231459379196,
      "num_tokens": 119450157.0,
      "step": 54520
    },
    {
      "entropy": 5.722608423233032,
      "epoch": 5.411373560936879,
      "grad_norm": 1.25,
      "learning_rate": 0.0002491989764266074,
      "loss": 4.9598,
      "mean_token_accuracy": 0.18760766983032226,
      "num_tokens": 119460402.0,
      "step": 54525
    },
    {
      "entropy": 5.780021238327026,
      "epoch": 5.411869789599047,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002491637789744564,
      "loss": 5.0344,
      "mean_token_accuracy": 0.19539906680583954,
      "num_tokens": 119471427.0,
      "step": 54530
    },
    {
      "entropy": 5.7367103099823,
      "epoch": 5.412366018261215,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024912858216298816,
      "loss": 4.9638,
      "mean_token_accuracy": 0.1991478681564331,
      "num_tokens": 119482638.0,
      "step": 54535
    },
    {
      "entropy": 5.742028903961182,
      "epoch": 5.412862246923383,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002490933859930753,
      "loss": 4.953,
      "mean_token_accuracy": 0.19632037729024887,
      "num_tokens": 119493565.0,
      "step": 54540
    },
    {
      "entropy": 5.708998727798462,
      "epoch": 5.41335847558555,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002490581904655907,
      "loss": 4.8949,
      "mean_token_accuracy": 0.1943197727203369,
      "num_tokens": 119504305.0,
      "step": 54545
    },
    {
      "entropy": 5.675386428833008,
      "epoch": 5.413854704247718,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024902299558140703,
      "loss": 4.9117,
      "mean_token_accuracy": 0.19518593698740005,
      "num_tokens": 119514313.0,
      "step": 54550
    },
    {
      "entropy": 5.720715856552124,
      "epoch": 5.414350932909885,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000248987801341397,
      "loss": 4.9597,
      "mean_token_accuracy": 0.19468376338481902,
      "num_tokens": 119526019.0,
      "step": 54555
    },
    {
      "entropy": 5.717365360260009,
      "epoch": 5.414847161572053,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024895260774643357,
      "loss": 4.9827,
      "mean_token_accuracy": 0.19717285931110382,
      "num_tokens": 119536669.0,
      "step": 54560
    },
    {
      "entropy": 5.736756229400635,
      "epoch": 5.41534339023422,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002489174147973893,
      "loss": 4.8858,
      "mean_token_accuracy": 0.20145838409662248,
      "num_tokens": 119547490.0,
      "step": 54565
    },
    {
      "entropy": 5.758128976821899,
      "epoch": 5.4158396188963875,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002488822224951369,
      "loss": 5.0112,
      "mean_token_accuracy": 0.1867569386959076,
      "num_tokens": 119559519.0,
      "step": 54570
    },
    {
      "entropy": 5.708441686630249,
      "epoch": 5.416335847558555,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002488470308405492,
      "loss": 4.9553,
      "mean_token_accuracy": 0.196786268055439,
      "num_tokens": 119569429.0,
      "step": 54575
    },
    {
      "entropy": 5.738256740570068,
      "epoch": 5.4168320762207225,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024881183983449875,
      "loss": 4.9748,
      "mean_token_accuracy": 0.19036914259195328,
      "num_tokens": 119580819.0,
      "step": 54580
    },
    {
      "entropy": 5.689541006088257,
      "epoch": 5.41732830488289,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002487766494778581,
      "loss": 4.9362,
      "mean_token_accuracy": 0.1974823370575905,
      "num_tokens": 119592030.0,
      "step": 54585
    },
    {
      "entropy": 5.666128778457642,
      "epoch": 5.417824533545057,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002487414597715002,
      "loss": 4.8477,
      "mean_token_accuracy": 0.20545806139707565,
      "num_tokens": 119602996.0,
      "step": 54590
    },
    {
      "entropy": 5.767043542861939,
      "epoch": 5.418320762207225,
      "grad_norm": 1.4375,
      "learning_rate": 0.00024870627071629754,
      "loss": 4.9895,
      "mean_token_accuracy": 0.1997496962547302,
      "num_tokens": 119614600.0,
      "step": 54595
    },
    {
      "entropy": 5.732475423812867,
      "epoch": 5.418816990869392,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024867108231312263,
      "loss": 4.9562,
      "mean_token_accuracy": 0.19575147181749344,
      "num_tokens": 119625728.0,
      "step": 54600
    },
    {
      "entropy": 5.652955675125122,
      "epoch": 5.41931321953156,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024863589456284815,
      "loss": 4.8642,
      "mean_token_accuracy": 0.20422145128250122,
      "num_tokens": 119636714.0,
      "step": 54605
    },
    {
      "entropy": 5.635393190383911,
      "epoch": 5.419809448193727,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024860070746634665,
      "loss": 4.8782,
      "mean_token_accuracy": 0.20275099724531173,
      "num_tokens": 119646971.0,
      "step": 54610
    },
    {
      "entropy": 5.718974685668945,
      "epoch": 5.420305676855895,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002485655210244907,
      "loss": 5.0125,
      "mean_token_accuracy": 0.18862931579351425,
      "num_tokens": 119658963.0,
      "step": 54615
    },
    {
      "entropy": 5.718960762023926,
      "epoch": 5.420801905518063,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024853033523815295,
      "loss": 4.9921,
      "mean_token_accuracy": 0.19247393161058426,
      "num_tokens": 119670889.0,
      "step": 54620
    },
    {
      "entropy": 5.682572364807129,
      "epoch": 5.421298134180231,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002484951501082058,
      "loss": 4.8833,
      "mean_token_accuracy": 0.20030374675989152,
      "num_tokens": 119681777.0,
      "step": 54625
    },
    {
      "entropy": 5.666016340255737,
      "epoch": 5.421794362842398,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002484599656355219,
      "loss": 4.906,
      "mean_token_accuracy": 0.20176294296979905,
      "num_tokens": 119692956.0,
      "step": 54630
    },
    {
      "entropy": 5.67473464012146,
      "epoch": 5.422290591504566,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024842478182097354,
      "loss": 4.931,
      "mean_token_accuracy": 0.1983781412243843,
      "num_tokens": 119704760.0,
      "step": 54635
    },
    {
      "entropy": 5.679885530471802,
      "epoch": 5.422786820166733,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002483895986654334,
      "loss": 4.9722,
      "mean_token_accuracy": 0.20031223446130753,
      "num_tokens": 119716048.0,
      "step": 54640
    },
    {
      "entropy": 5.675173997879028,
      "epoch": 5.4232830488289006,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024835441616977383,
      "loss": 4.9297,
      "mean_token_accuracy": 0.20236769318580627,
      "num_tokens": 119726844.0,
      "step": 54645
    },
    {
      "entropy": 5.73099570274353,
      "epoch": 5.423779277491068,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002483192343348673,
      "loss": 4.9665,
      "mean_token_accuracy": 0.20016074776649476,
      "num_tokens": 119737334.0,
      "step": 54650
    },
    {
      "entropy": 5.787933683395385,
      "epoch": 5.4242755061532355,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002482840531615864,
      "loss": 4.9867,
      "mean_token_accuracy": 0.19219014495611192,
      "num_tokens": 119748324.0,
      "step": 54655
    },
    {
      "entropy": 5.692540502548217,
      "epoch": 5.424771734815403,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002482488726508033,
      "loss": 4.8858,
      "mean_token_accuracy": 0.198136168718338,
      "num_tokens": 119758858.0,
      "step": 54660
    },
    {
      "entropy": 5.72880916595459,
      "epoch": 5.4252679634775705,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024821369280339056,
      "loss": 4.973,
      "mean_token_accuracy": 0.1953107014298439,
      "num_tokens": 119770561.0,
      "step": 54665
    },
    {
      "entropy": 5.648265218734741,
      "epoch": 5.425764192139738,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002481785136202205,
      "loss": 4.8645,
      "mean_token_accuracy": 0.2003192499279976,
      "num_tokens": 119781385.0,
      "step": 54670
    },
    {
      "entropy": 5.723458290100098,
      "epoch": 5.426260420801905,
      "grad_norm": 1.171875,
      "learning_rate": 0.00024814333510216546,
      "loss": 5.0116,
      "mean_token_accuracy": 0.18257851600646974,
      "num_tokens": 119791573.0,
      "step": 54675
    },
    {
      "entropy": 5.721501731872559,
      "epoch": 5.426756649464073,
      "grad_norm": 1.25,
      "learning_rate": 0.0002481081572500979,
      "loss": 4.9738,
      "mean_token_accuracy": 0.19246212393045425,
      "num_tokens": 119803171.0,
      "step": 54680
    },
    {
      "entropy": 5.706554365158081,
      "epoch": 5.42725287812624,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024807298006489004,
      "loss": 4.9237,
      "mean_token_accuracy": 0.19574215412139892,
      "num_tokens": 119814606.0,
      "step": 54685
    },
    {
      "entropy": 5.759397792816162,
      "epoch": 5.427749106788408,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024803780354741437,
      "loss": 4.9717,
      "mean_token_accuracy": 0.1969744697213173,
      "num_tokens": 119825310.0,
      "step": 54690
    },
    {
      "entropy": 5.659353256225586,
      "epoch": 5.428245335450575,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002480026276985429,
      "loss": 4.8827,
      "mean_token_accuracy": 0.20012814551591873,
      "num_tokens": 119836447.0,
      "step": 54695
    },
    {
      "entropy": 5.675319623947144,
      "epoch": 5.428741564112743,
      "grad_norm": 1.5,
      "learning_rate": 0.00024796745251914827,
      "loss": 4.9113,
      "mean_token_accuracy": 0.1960160419344902,
      "num_tokens": 119846769.0,
      "step": 54700
    },
    {
      "entropy": 5.635169219970703,
      "epoch": 5.429237792774911,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024793227801010243,
      "loss": 4.8399,
      "mean_token_accuracy": 0.19818079173564912,
      "num_tokens": 119857113.0,
      "step": 54705
    },
    {
      "entropy": 5.6236653327941895,
      "epoch": 5.429734021437079,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024789710417227786,
      "loss": 4.8602,
      "mean_token_accuracy": 0.20425012558698655,
      "num_tokens": 119868062.0,
      "step": 54710
    },
    {
      "entropy": 5.646620273590088,
      "epoch": 5.430230250099246,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002478619310065466,
      "loss": 4.8859,
      "mean_token_accuracy": 0.20163375288248062,
      "num_tokens": 119878426.0,
      "step": 54715
    },
    {
      "entropy": 5.6709247589111325,
      "epoch": 5.430726478761414,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024782675851378106,
      "loss": 4.9052,
      "mean_token_accuracy": 0.19800780415534974,
      "num_tokens": 119888057.0,
      "step": 54720
    },
    {
      "entropy": 5.688847494125366,
      "epoch": 5.431222707423581,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024779158669485337,
      "loss": 4.9287,
      "mean_token_accuracy": 0.1943237915635109,
      "num_tokens": 119898327.0,
      "step": 54725
    },
    {
      "entropy": 5.7103653907775875,
      "epoch": 5.4317189360857485,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024775641555063564,
      "loss": 4.9257,
      "mean_token_accuracy": 0.2007109835743904,
      "num_tokens": 119909355.0,
      "step": 54730
    },
    {
      "entropy": 5.689350414276123,
      "epoch": 5.432215164747916,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024772124508200006,
      "loss": 4.9294,
      "mean_token_accuracy": 0.20180980563163758,
      "num_tokens": 119919061.0,
      "step": 54735
    },
    {
      "entropy": 5.725129175186157,
      "epoch": 5.4327113934100835,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000247686075289819,
      "loss": 4.9544,
      "mean_token_accuracy": 0.19784517586231232,
      "num_tokens": 119930631.0,
      "step": 54740
    },
    {
      "entropy": 5.662878942489624,
      "epoch": 5.433207622072251,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002476509061749642,
      "loss": 4.9133,
      "mean_token_accuracy": 0.19878947883844375,
      "num_tokens": 119941407.0,
      "step": 54745
    },
    {
      "entropy": 5.7112141132354735,
      "epoch": 5.433703850734418,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002476157377383081,
      "loss": 4.9331,
      "mean_token_accuracy": 0.19678154587745667,
      "num_tokens": 119953343.0,
      "step": 54750
    },
    {
      "entropy": 5.6951066017150875,
      "epoch": 5.434200079396586,
      "grad_norm": 1.25,
      "learning_rate": 0.00024758056998072273,
      "loss": 4.9379,
      "mean_token_accuracy": 0.19520702362060546,
      "num_tokens": 119964511.0,
      "step": 54755
    },
    {
      "entropy": 5.766644048690796,
      "epoch": 5.434696308058753,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024754540290308015,
      "loss": 4.9969,
      "mean_token_accuracy": 0.20449078232049941,
      "num_tokens": 119975413.0,
      "step": 54760
    },
    {
      "entropy": 5.697099924087524,
      "epoch": 5.435192536720921,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024751023650625224,
      "loss": 4.9056,
      "mean_token_accuracy": 0.19647884070873262,
      "num_tokens": 119987060.0,
      "step": 54765
    },
    {
      "entropy": 5.793978357315064,
      "epoch": 5.435688765383088,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002474750707911114,
      "loss": 5.0604,
      "mean_token_accuracy": 0.19181925058364868,
      "num_tokens": 119999382.0,
      "step": 54770
    },
    {
      "entropy": 5.6473558902740475,
      "epoch": 5.436184994045256,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024743990575852955,
      "loss": 4.9256,
      "mean_token_accuracy": 0.19530074894428254,
      "num_tokens": 120009740.0,
      "step": 54775
    },
    {
      "entropy": 5.711723899841308,
      "epoch": 5.436681222707423,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002474047414093785,
      "loss": 4.9371,
      "mean_token_accuracy": 0.20136825144290924,
      "num_tokens": 120020720.0,
      "step": 54780
    },
    {
      "entropy": 5.669054746627808,
      "epoch": 5.437177451369591,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002473695777445305,
      "loss": 4.9141,
      "mean_token_accuracy": 0.1986997753381729,
      "num_tokens": 120031520.0,
      "step": 54785
    },
    {
      "entropy": 5.680950880050659,
      "epoch": 5.437673680031759,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024733441476485743,
      "loss": 4.9612,
      "mean_token_accuracy": 0.19280142486095428,
      "num_tokens": 120043306.0,
      "step": 54790
    },
    {
      "entropy": 5.6819586753845215,
      "epoch": 5.438169908693926,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024729925247123113,
      "loss": 4.8896,
      "mean_token_accuracy": 0.2012135401368141,
      "num_tokens": 120054303.0,
      "step": 54795
    },
    {
      "entropy": 5.6958517074584964,
      "epoch": 5.438666137356094,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002472640908645238,
      "loss": 4.9593,
      "mean_token_accuracy": 0.1966611459851265,
      "num_tokens": 120066460.0,
      "step": 54800
    },
    {
      "entropy": 5.683529472351074,
      "epoch": 5.4391623660182615,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024722892994560717,
      "loss": 4.9306,
      "mean_token_accuracy": 0.2011274978518486,
      "num_tokens": 120076664.0,
      "step": 54805
    },
    {
      "entropy": 5.666184520721435,
      "epoch": 5.439658594680429,
      "grad_norm": 1.25,
      "learning_rate": 0.00024719376971535335,
      "loss": 4.8613,
      "mean_token_accuracy": 0.2008513793349266,
      "num_tokens": 120087436.0,
      "step": 54810
    },
    {
      "entropy": 5.6987528800964355,
      "epoch": 5.4401548233425965,
      "grad_norm": 1.375,
      "learning_rate": 0.00024715861017463405,
      "loss": 4.9138,
      "mean_token_accuracy": 0.1941932111978531,
      "num_tokens": 120098676.0,
      "step": 54815
    },
    {
      "entropy": 5.634747791290283,
      "epoch": 5.440651052004764,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002471234513243212,
      "loss": 4.8674,
      "mean_token_accuracy": 0.20170824229717255,
      "num_tokens": 120110659.0,
      "step": 54820
    },
    {
      "entropy": 5.643837356567383,
      "epoch": 5.441147280666931,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024708829316528677,
      "loss": 4.9485,
      "mean_token_accuracy": 0.1972258821129799,
      "num_tokens": 120122969.0,
      "step": 54825
    },
    {
      "entropy": 5.650237560272217,
      "epoch": 5.441643509329099,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024705313569840245,
      "loss": 4.8918,
      "mean_token_accuracy": 0.19981221705675126,
      "num_tokens": 120132900.0,
      "step": 54830
    },
    {
      "entropy": 5.766252136230468,
      "epoch": 5.442139737991266,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002470179789245402,
      "loss": 4.9851,
      "mean_token_accuracy": 0.1956326261162758,
      "num_tokens": 120144117.0,
      "step": 54835
    },
    {
      "entropy": 5.698304605484009,
      "epoch": 5.442635966653434,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024698282284457175,
      "loss": 4.924,
      "mean_token_accuracy": 0.19713876098394395,
      "num_tokens": 120154751.0,
      "step": 54840
    },
    {
      "entropy": 5.652958631515503,
      "epoch": 5.443132195315601,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024694766745936894,
      "loss": 4.8997,
      "mean_token_accuracy": 0.18809228837490083,
      "num_tokens": 120165547.0,
      "step": 54845
    },
    {
      "entropy": 5.700856876373291,
      "epoch": 5.443628423977769,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002469125127698034,
      "loss": 4.9852,
      "mean_token_accuracy": 0.19443864673376082,
      "num_tokens": 120177026.0,
      "step": 54850
    },
    {
      "entropy": 5.694537353515625,
      "epoch": 5.444124652639936,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024687735877674714,
      "loss": 4.9522,
      "mean_token_accuracy": 0.19123709201812744,
      "num_tokens": 120189117.0,
      "step": 54855
    },
    {
      "entropy": 5.751786375045777,
      "epoch": 5.444620881302104,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024684220548107177,
      "loss": 4.91,
      "mean_token_accuracy": 0.20326754748821257,
      "num_tokens": 120201358.0,
      "step": 54860
    },
    {
      "entropy": 5.639511489868164,
      "epoch": 5.445117109964271,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024680705288364905,
      "loss": 4.8424,
      "mean_token_accuracy": 0.20268211960792543,
      "num_tokens": 120213882.0,
      "step": 54865
    },
    {
      "entropy": 5.699413394927978,
      "epoch": 5.445613338626439,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024677190098535074,
      "loss": 4.9604,
      "mean_token_accuracy": 0.2012516140937805,
      "num_tokens": 120224243.0,
      "step": 54870
    },
    {
      "entropy": 5.731530094146729,
      "epoch": 5.446109567288606,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024673674978704837,
      "loss": 4.9887,
      "mean_token_accuracy": 0.18970702439546586,
      "num_tokens": 120233274.0,
      "step": 54875
    },
    {
      "entropy": 5.71637077331543,
      "epoch": 5.446605795950774,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002467015992896136,
      "loss": 4.9682,
      "mean_token_accuracy": 0.19161138981580733,
      "num_tokens": 120244550.0,
      "step": 54880
    },
    {
      "entropy": 5.718169832229615,
      "epoch": 5.447102024612942,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002466664494939183,
      "loss": 4.9492,
      "mean_token_accuracy": 0.1942100167274475,
      "num_tokens": 120255593.0,
      "step": 54885
    },
    {
      "entropy": 5.764147233963013,
      "epoch": 5.4475982532751095,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024663130040083407,
      "loss": 4.9741,
      "mean_token_accuracy": 0.19574763774871826,
      "num_tokens": 120267853.0,
      "step": 54890
    },
    {
      "entropy": 5.728455686569214,
      "epoch": 5.448094481937277,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024659615201123235,
      "loss": 4.9232,
      "mean_token_accuracy": 0.19634603410959245,
      "num_tokens": 120278836.0,
      "step": 54895
    },
    {
      "entropy": 5.8072186470031735,
      "epoch": 5.448590710599444,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024656100432598495,
      "loss": 5.0782,
      "mean_token_accuracy": 0.1925838366150856,
      "num_tokens": 120290553.0,
      "step": 54900
    },
    {
      "entropy": 5.732729911804199,
      "epoch": 5.449086939261612,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024652585734596325,
      "loss": 4.8529,
      "mean_token_accuracy": 0.20829148441553116,
      "num_tokens": 120300837.0,
      "step": 54905
    },
    {
      "entropy": 5.6828772068023685,
      "epoch": 5.449583167923779,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024649071107203896,
      "loss": 4.9524,
      "mean_token_accuracy": 0.1924799144268036,
      "num_tokens": 120311378.0,
      "step": 54910
    },
    {
      "entropy": 5.657508373260498,
      "epoch": 5.450079396585947,
      "grad_norm": 1.375,
      "learning_rate": 0.00024645556550508365,
      "loss": 4.8588,
      "mean_token_accuracy": 0.2065693810582161,
      "num_tokens": 120322671.0,
      "step": 54915
    },
    {
      "entropy": 5.687947034835815,
      "epoch": 5.450575625248114,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002464204206459688,
      "loss": 4.9487,
      "mean_token_accuracy": 0.19730460941791533,
      "num_tokens": 120333362.0,
      "step": 54920
    },
    {
      "entropy": 5.631596231460572,
      "epoch": 5.451071853910282,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002463852764955659,
      "loss": 4.8589,
      "mean_token_accuracy": 0.19833541214466094,
      "num_tokens": 120344908.0,
      "step": 54925
    },
    {
      "entropy": 5.726353359222412,
      "epoch": 5.451568082572449,
      "grad_norm": 1.25,
      "learning_rate": 0.0002463501330547465,
      "loss": 5.0077,
      "mean_token_accuracy": 0.194246868789196,
      "num_tokens": 120356388.0,
      "step": 54930
    },
    {
      "entropy": 5.702703428268433,
      "epoch": 5.452064311234617,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024631499032438215,
      "loss": 4.9118,
      "mean_token_accuracy": 0.19925121068954468,
      "num_tokens": 120367709.0,
      "step": 54935
    },
    {
      "entropy": 5.740500831604004,
      "epoch": 5.452560539896784,
      "grad_norm": 1.375,
      "learning_rate": 0.00024627984830534407,
      "loss": 4.8962,
      "mean_token_accuracy": 0.20184263586997986,
      "num_tokens": 120377056.0,
      "step": 54940
    },
    {
      "entropy": 5.690966176986694,
      "epoch": 5.453056768558952,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002462447069985039,
      "loss": 4.9183,
      "mean_token_accuracy": 0.20010547637939452,
      "num_tokens": 120388531.0,
      "step": 54945
    },
    {
      "entropy": 5.672568655014038,
      "epoch": 5.453552997221119,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002462095664047331,
      "loss": 4.9599,
      "mean_token_accuracy": 0.19624151438474655,
      "num_tokens": 120399530.0,
      "step": 54950
    },
    {
      "entropy": 5.625120449066162,
      "epoch": 5.454049225883287,
      "grad_norm": 1.34375,
      "learning_rate": 0.000246174426524903,
      "loss": 4.8436,
      "mean_token_accuracy": 0.20535083413124083,
      "num_tokens": 120409738.0,
      "step": 54955
    },
    {
      "entropy": 5.681188535690308,
      "epoch": 5.454545454545454,
      "grad_norm": 1.15625,
      "learning_rate": 0.000246139287359885,
      "loss": 4.9749,
      "mean_token_accuracy": 0.2008794903755188,
      "num_tokens": 120422844.0,
      "step": 54960
    },
    {
      "entropy": 5.697185087203979,
      "epoch": 5.455041683207622,
      "grad_norm": 1.078125,
      "learning_rate": 0.00024610414891055053,
      "loss": 4.9314,
      "mean_token_accuracy": 0.19563799500465393,
      "num_tokens": 120435371.0,
      "step": 54965
    },
    {
      "entropy": 5.761399173736573,
      "epoch": 5.45553791186979,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002460690111777708,
      "loss": 5.0098,
      "mean_token_accuracy": 0.19943617284297943,
      "num_tokens": 120447740.0,
      "step": 54970
    },
    {
      "entropy": 5.694831514358521,
      "epoch": 5.4560341405319575,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002460338741624173,
      "loss": 4.8816,
      "mean_token_accuracy": 0.19947802871465684,
      "num_tokens": 120457284.0,
      "step": 54975
    },
    {
      "entropy": 5.663963651657104,
      "epoch": 5.456530369194125,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024599873786536135,
      "loss": 4.9086,
      "mean_token_accuracy": 0.19605157077312468,
      "num_tokens": 120467949.0,
      "step": 54980
    },
    {
      "entropy": 5.715018177032471,
      "epoch": 5.457026597856292,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002459636022874742,
      "loss": 4.9583,
      "mean_token_accuracy": 0.18949471712112426,
      "num_tokens": 120478072.0,
      "step": 54985
    },
    {
      "entropy": 5.694090890884399,
      "epoch": 5.45752282651846,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002459284674296271,
      "loss": 4.95,
      "mean_token_accuracy": 0.20027918964624405,
      "num_tokens": 120489634.0,
      "step": 54990
    },
    {
      "entropy": 5.766100168228149,
      "epoch": 5.458019055180627,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002458933332926914,
      "loss": 5.0109,
      "mean_token_accuracy": 0.19351096898317338,
      "num_tokens": 120501365.0,
      "step": 54995
    },
    {
      "entropy": 5.698494958877563,
      "epoch": 5.458515283842795,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002458581998775382,
      "loss": 4.9791,
      "mean_token_accuracy": 0.1892771914601326,
      "num_tokens": 120512063.0,
      "step": 55000
    },
    {
      "entropy": 5.723287439346313,
      "epoch": 5.459011512504962,
      "grad_norm": 1.390625,
      "learning_rate": 0.000245823067185039,
      "loss": 4.9367,
      "mean_token_accuracy": 0.19774828106164932,
      "num_tokens": 120522979.0,
      "step": 55005
    },
    {
      "entropy": 5.653534650802612,
      "epoch": 5.45950774116713,
      "grad_norm": 1.15625,
      "learning_rate": 0.00024578793521606477,
      "loss": 4.8892,
      "mean_token_accuracy": 0.1992585152387619,
      "num_tokens": 120534397.0,
      "step": 55010
    },
    {
      "entropy": 5.746590852737427,
      "epoch": 5.460003969829297,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002457528039714869,
      "loss": 5.0107,
      "mean_token_accuracy": 0.19307151436805725,
      "num_tokens": 120544868.0,
      "step": 55015
    },
    {
      "entropy": 5.664212322235107,
      "epoch": 5.460500198491465,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002457176734521764,
      "loss": 4.8795,
      "mean_token_accuracy": 0.20264535397291183,
      "num_tokens": 120555326.0,
      "step": 55020
    },
    {
      "entropy": 5.657929611206055,
      "epoch": 5.460996427153632,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024568254365900453,
      "loss": 4.8469,
      "mean_token_accuracy": 0.1977274864912033,
      "num_tokens": 120565618.0,
      "step": 55025
    },
    {
      "entropy": 5.641638994216919,
      "epoch": 5.4614926558158,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024564741459284236,
      "loss": 4.8666,
      "mean_token_accuracy": 0.19769756495952606,
      "num_tokens": 120576593.0,
      "step": 55030
    },
    {
      "entropy": 5.757778835296631,
      "epoch": 5.461988884477967,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002456122862545611,
      "loss": 5.0132,
      "mean_token_accuracy": 0.1964322328567505,
      "num_tokens": 120589815.0,
      "step": 55035
    },
    {
      "entropy": 5.727233934402466,
      "epoch": 5.462485113140135,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024557715864503176,
      "loss": 4.9966,
      "mean_token_accuracy": 0.19288990944623946,
      "num_tokens": 120599726.0,
      "step": 55040
    },
    {
      "entropy": 5.652446985244751,
      "epoch": 5.462981341802302,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002455420317651256,
      "loss": 4.8672,
      "mean_token_accuracy": 0.20328531712293624,
      "num_tokens": 120610070.0,
      "step": 55045
    },
    {
      "entropy": 5.616033267974854,
      "epoch": 5.46347757046447,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002455069056157135,
      "loss": 4.8518,
      "mean_token_accuracy": 0.20158286541700363,
      "num_tokens": 120619965.0,
      "step": 55050
    },
    {
      "entropy": 5.6955406188964846,
      "epoch": 5.463973799126638,
      "grad_norm": 1.375,
      "learning_rate": 0.00024547178019766654,
      "loss": 4.9135,
      "mean_token_accuracy": 0.19853681623935698,
      "num_tokens": 120630714.0,
      "step": 55055
    },
    {
      "entropy": 5.711674404144287,
      "epoch": 5.464470027788805,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002454366555118558,
      "loss": 4.9471,
      "mean_token_accuracy": 0.1991193026304245,
      "num_tokens": 120642230.0,
      "step": 55060
    },
    {
      "entropy": 5.715573501586914,
      "epoch": 5.464966256450973,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024540153155915234,
      "loss": 4.8924,
      "mean_token_accuracy": 0.1976883128285408,
      "num_tokens": 120652201.0,
      "step": 55065
    },
    {
      "entropy": 5.705887222290039,
      "epoch": 5.46546248511314,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002453664083404271,
      "loss": 4.9076,
      "mean_token_accuracy": 0.20824453234672546,
      "num_tokens": 120662196.0,
      "step": 55070
    },
    {
      "entropy": 5.68374547958374,
      "epoch": 5.465958713775308,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024533128585655116,
      "loss": 4.94,
      "mean_token_accuracy": 0.19802937656641006,
      "num_tokens": 120673374.0,
      "step": 55075
    },
    {
      "entropy": 5.679286241531372,
      "epoch": 5.466454942437475,
      "grad_norm": 1.25,
      "learning_rate": 0.0002452961641083954,
      "loss": 4.929,
      "mean_token_accuracy": 0.19153770357370375,
      "num_tokens": 120683677.0,
      "step": 55080
    },
    {
      "entropy": 5.721199083328247,
      "epoch": 5.466951171099643,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002452610430968306,
      "loss": 4.9556,
      "mean_token_accuracy": 0.19739802777767182,
      "num_tokens": 120694456.0,
      "step": 55085
    },
    {
      "entropy": 5.654023933410644,
      "epoch": 5.46744739976181,
      "grad_norm": 1.375,
      "learning_rate": 0.00024522592282272797,
      "loss": 4.9232,
      "mean_token_accuracy": 0.19674306064844133,
      "num_tokens": 120705011.0,
      "step": 55090
    },
    {
      "entropy": 5.612892770767212,
      "epoch": 5.467943628423978,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002451908032869584,
      "loss": 4.8183,
      "mean_token_accuracy": 0.20605408996343613,
      "num_tokens": 120716484.0,
      "step": 55095
    },
    {
      "entropy": 5.683294820785522,
      "epoch": 5.468439857086145,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024515568449039253,
      "loss": 4.9263,
      "mean_token_accuracy": 0.1985497310757637,
      "num_tokens": 120728027.0,
      "step": 55100
    },
    {
      "entropy": 5.620395612716675,
      "epoch": 5.468936085748313,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002451205664339014,
      "loss": 4.8871,
      "mean_token_accuracy": 0.20115449279546738,
      "num_tokens": 120738974.0,
      "step": 55105
    },
    {
      "entropy": 5.674323081970215,
      "epoch": 5.46943231441048,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002450854491183558,
      "loss": 4.929,
      "mean_token_accuracy": 0.19741756469011307,
      "num_tokens": 120749040.0,
      "step": 55110
    },
    {
      "entropy": 5.657813978195191,
      "epoch": 5.469928543072648,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002450503325446266,
      "loss": 4.8432,
      "mean_token_accuracy": 0.20003784894943238,
      "num_tokens": 120758423.0,
      "step": 55115
    },
    {
      "entropy": 5.659582233428955,
      "epoch": 5.470424771734815,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024501521671358475,
      "loss": 4.8924,
      "mean_token_accuracy": 0.19208650141954423,
      "num_tokens": 120769059.0,
      "step": 55120
    },
    {
      "entropy": 5.692150211334228,
      "epoch": 5.470921000396983,
      "grad_norm": 1.375,
      "learning_rate": 0.00024498010162610077,
      "loss": 4.9034,
      "mean_token_accuracy": 0.1986065089702606,
      "num_tokens": 120779923.0,
      "step": 55125
    },
    {
      "entropy": 5.693135213851929,
      "epoch": 5.47141722905915,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002449449872830457,
      "loss": 5.0031,
      "mean_token_accuracy": 0.19555061012506486,
      "num_tokens": 120791208.0,
      "step": 55130
    },
    {
      "entropy": 5.759771060943604,
      "epoch": 5.4719134577213175,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002449098736852901,
      "loss": 4.9782,
      "mean_token_accuracy": 0.19633003771305085,
      "num_tokens": 120801803.0,
      "step": 55135
    },
    {
      "entropy": 5.699019432067871,
      "epoch": 5.472409686383486,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024487476083370486,
      "loss": 4.845,
      "mean_token_accuracy": 0.2038598284125328,
      "num_tokens": 120811914.0,
      "step": 55140
    },
    {
      "entropy": 5.764265775680542,
      "epoch": 5.472905915045653,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002448396487291605,
      "loss": 5.0271,
      "mean_token_accuracy": 0.18958272635936738,
      "num_tokens": 120822854.0,
      "step": 55145
    },
    {
      "entropy": 5.6705183506011965,
      "epoch": 5.473402143707821,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000244804537372528,
      "loss": 4.9457,
      "mean_token_accuracy": 0.19660018086433412,
      "num_tokens": 120833918.0,
      "step": 55150
    },
    {
      "entropy": 5.668146753311158,
      "epoch": 5.473898372369988,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002447694267646779,
      "loss": 4.8518,
      "mean_token_accuracy": 0.20088837444782257,
      "num_tokens": 120843312.0,
      "step": 55155
    },
    {
      "entropy": 5.669789361953735,
      "epoch": 5.474394601032156,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024473431690648085,
      "loss": 4.8332,
      "mean_token_accuracy": 0.2129644989967346,
      "num_tokens": 120854697.0,
      "step": 55160
    },
    {
      "entropy": 5.669523572921753,
      "epoch": 5.474890829694323,
      "grad_norm": 1.25,
      "learning_rate": 0.00024469920779880754,
      "loss": 4.8289,
      "mean_token_accuracy": 0.20745866745710373,
      "num_tokens": 120864741.0,
      "step": 55165
    },
    {
      "entropy": 5.69873046875,
      "epoch": 5.475387058356491,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024466409944252855,
      "loss": 4.9942,
      "mean_token_accuracy": 0.19571684449911117,
      "num_tokens": 120876421.0,
      "step": 55170
    },
    {
      "entropy": 5.711315441131592,
      "epoch": 5.475883287018658,
      "grad_norm": 1.375,
      "learning_rate": 0.00024462899183851447,
      "loss": 5.004,
      "mean_token_accuracy": 0.19510414451360703,
      "num_tokens": 120887067.0,
      "step": 55175
    },
    {
      "entropy": 5.702348041534424,
      "epoch": 5.476379515680826,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000244593884987636,
      "loss": 4.8857,
      "mean_token_accuracy": 0.2021487683057785,
      "num_tokens": 120897728.0,
      "step": 55180
    },
    {
      "entropy": 5.644462776184082,
      "epoch": 5.476875744342993,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024455877889076365,
      "loss": 4.8791,
      "mean_token_accuracy": 0.20391388684511186,
      "num_tokens": 120908294.0,
      "step": 55185
    },
    {
      "entropy": 5.728915405273438,
      "epoch": 5.477371973005161,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002445236735487681,
      "loss": 4.9306,
      "mean_token_accuracy": 0.19832042157649993,
      "num_tokens": 120918295.0,
      "step": 55190
    },
    {
      "entropy": 5.670880317687988,
      "epoch": 5.477868201667328,
      "grad_norm": 1.25,
      "learning_rate": 0.0002444885689625197,
      "loss": 4.8782,
      "mean_token_accuracy": 0.20417665392160417,
      "num_tokens": 120929573.0,
      "step": 55195
    },
    {
      "entropy": 5.6469025135040285,
      "epoch": 5.478364430329496,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000244453465132889,
      "loss": 4.918,
      "mean_token_accuracy": 0.20188051760196685,
      "num_tokens": 120941433.0,
      "step": 55200
    },
    {
      "entropy": 5.612486028671265,
      "epoch": 5.478860658991663,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002444183620607465,
      "loss": 4.8898,
      "mean_token_accuracy": 0.20048014670610428,
      "num_tokens": 120952480.0,
      "step": 55205
    },
    {
      "entropy": 5.7263024806976315,
      "epoch": 5.4793568876538306,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024438325974696285,
      "loss": 4.973,
      "mean_token_accuracy": 0.19466088265180587,
      "num_tokens": 120964026.0,
      "step": 55210
    },
    {
      "entropy": 5.665264225006103,
      "epoch": 5.479853116315998,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024434815819240836,
      "loss": 4.8756,
      "mean_token_accuracy": 0.20433928519487382,
      "num_tokens": 120974224.0,
      "step": 55215
    },
    {
      "entropy": 5.744495868682861,
      "epoch": 5.4803493449781655,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024431305739795343,
      "loss": 4.9603,
      "mean_token_accuracy": 0.19926463365554808,
      "num_tokens": 120985171.0,
      "step": 55220
    },
    {
      "entropy": 5.732579946517944,
      "epoch": 5.480845573640334,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024427795736446867,
      "loss": 4.9233,
      "mean_token_accuracy": 0.19898062944412231,
      "num_tokens": 120995739.0,
      "step": 55225
    },
    {
      "entropy": 5.675272655487061,
      "epoch": 5.481341802302501,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002442428580928243,
      "loss": 4.8792,
      "mean_token_accuracy": 0.19881397485733032,
      "num_tokens": 121006503.0,
      "step": 55230
    },
    {
      "entropy": 5.613413286209107,
      "epoch": 5.481838030964669,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024420775958389066,
      "loss": 4.7986,
      "mean_token_accuracy": 0.21227087825536728,
      "num_tokens": 121016349.0,
      "step": 55235
    },
    {
      "entropy": 5.706301212310791,
      "epoch": 5.482334259626836,
      "grad_norm": 1.375,
      "learning_rate": 0.00024417266183853837,
      "loss": 4.9953,
      "mean_token_accuracy": 0.19292120337486268,
      "num_tokens": 121027206.0,
      "step": 55240
    },
    {
      "entropy": 5.734980630874634,
      "epoch": 5.482830488289004,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00024413756485763762,
      "loss": 4.9412,
      "mean_token_accuracy": 0.2006149560213089,
      "num_tokens": 121036617.0,
      "step": 55245
    },
    {
      "entropy": 5.698385429382324,
      "epoch": 5.483326716951171,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024410246864205876,
      "loss": 4.9745,
      "mean_token_accuracy": 0.18938594460487365,
      "num_tokens": 121048885.0,
      "step": 55250
    },
    {
      "entropy": 5.730281829833984,
      "epoch": 5.483822945613339,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024406737319267207,
      "loss": 4.9819,
      "mean_token_accuracy": 0.19405231773853301,
      "num_tokens": 121060265.0,
      "step": 55255
    },
    {
      "entropy": 5.7501367092132565,
      "epoch": 5.484319174275506,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024403227851034793,
      "loss": 4.952,
      "mean_token_accuracy": 0.18877000063657762,
      "num_tokens": 121071127.0,
      "step": 55260
    },
    {
      "entropy": 5.756885766983032,
      "epoch": 5.484815402937674,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002439971845959565,
      "loss": 5.0066,
      "mean_token_accuracy": 0.19624752551317215,
      "num_tokens": 121082299.0,
      "step": 55265
    },
    {
      "entropy": 5.666506671905518,
      "epoch": 5.485311631599841,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024396209145036802,
      "loss": 4.8968,
      "mean_token_accuracy": 0.20119473338127136,
      "num_tokens": 121093357.0,
      "step": 55270
    },
    {
      "entropy": 5.652717018127442,
      "epoch": 5.485807860262009,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024392699907445288,
      "loss": 4.9627,
      "mean_token_accuracy": 0.18586921989917754,
      "num_tokens": 121104195.0,
      "step": 55275
    },
    {
      "entropy": 5.671758556365967,
      "epoch": 5.486304088924176,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024389190746908123,
      "loss": 4.9123,
      "mean_token_accuracy": 0.19802311807870865,
      "num_tokens": 121115873.0,
      "step": 55280
    },
    {
      "entropy": 5.733998012542725,
      "epoch": 5.486800317586344,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024385681663512322,
      "loss": 4.9407,
      "mean_token_accuracy": 0.19792796820402145,
      "num_tokens": 121125345.0,
      "step": 55285
    },
    {
      "entropy": 5.6596184253692625,
      "epoch": 5.487296546248511,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024382172657344902,
      "loss": 4.8384,
      "mean_token_accuracy": 0.20469991415739058,
      "num_tokens": 121135572.0,
      "step": 55290
    },
    {
      "entropy": 5.689822101593018,
      "epoch": 5.4877927749106785,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024378663728492878,
      "loss": 4.9881,
      "mean_token_accuracy": 0.19329280406236649,
      "num_tokens": 121145101.0,
      "step": 55295
    },
    {
      "entropy": 5.697627830505371,
      "epoch": 5.488289003572846,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024375154877043277,
      "loss": 4.891,
      "mean_token_accuracy": 0.20353902280330657,
      "num_tokens": 121155216.0,
      "step": 55300
    },
    {
      "entropy": 5.721096849441528,
      "epoch": 5.4887852322350135,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024371646103083094,
      "loss": 4.908,
      "mean_token_accuracy": 0.20230183899402618,
      "num_tokens": 121166222.0,
      "step": 55305
    },
    {
      "entropy": 5.7133002281188965,
      "epoch": 5.489281460897182,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024368137406699354,
      "loss": 4.9313,
      "mean_token_accuracy": 0.1996539369225502,
      "num_tokens": 121177119.0,
      "step": 55310
    },
    {
      "entropy": 5.6586236476898195,
      "epoch": 5.489777689559349,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00024364628787979054,
      "loss": 4.9312,
      "mean_token_accuracy": 0.196266970038414,
      "num_tokens": 121188871.0,
      "step": 55315
    },
    {
      "entropy": 5.711326742172242,
      "epoch": 5.490273918221517,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00024361120247009193,
      "loss": 4.9839,
      "mean_token_accuracy": 0.1975255399942398,
      "num_tokens": 121200778.0,
      "step": 55320
    },
    {
      "entropy": 5.694546699523926,
      "epoch": 5.490770146883684,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024357611783876805,
      "loss": 4.9356,
      "mean_token_accuracy": 0.1999092623591423,
      "num_tokens": 121211697.0,
      "step": 55325
    },
    {
      "entropy": 5.665983724594116,
      "epoch": 5.491266375545852,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024354103398668863,
      "loss": 4.9262,
      "mean_token_accuracy": 0.200292731821537,
      "num_tokens": 121223037.0,
      "step": 55330
    },
    {
      "entropy": 5.727264404296875,
      "epoch": 5.491762604208019,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024350595091472381,
      "loss": 4.9663,
      "mean_token_accuracy": 0.19686685204505922,
      "num_tokens": 121233770.0,
      "step": 55335
    },
    {
      "entropy": 5.734139680862427,
      "epoch": 5.492258832870187,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024347086862374357,
      "loss": 5.0039,
      "mean_token_accuracy": 0.19240983426570893,
      "num_tokens": 121246187.0,
      "step": 55340
    },
    {
      "entropy": 5.790848731994629,
      "epoch": 5.492755061532354,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024343578711461789,
      "loss": 5.0811,
      "mean_token_accuracy": 0.1891033411026001,
      "num_tokens": 121257699.0,
      "step": 55345
    },
    {
      "entropy": 5.691515493392944,
      "epoch": 5.493251290194522,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024340070638821654,
      "loss": 4.8618,
      "mean_token_accuracy": 0.20380217134952544,
      "num_tokens": 121268413.0,
      "step": 55350
    },
    {
      "entropy": 5.670339441299438,
      "epoch": 5.493747518856689,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002433656264454096,
      "loss": 4.9537,
      "mean_token_accuracy": 0.19001973122358323,
      "num_tokens": 121280340.0,
      "step": 55355
    },
    {
      "entropy": 5.667886447906494,
      "epoch": 5.494243747518857,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024333054728706705,
      "loss": 4.9096,
      "mean_token_accuracy": 0.2045106276869774,
      "num_tokens": 121290989.0,
      "step": 55360
    },
    {
      "entropy": 5.742640542984009,
      "epoch": 5.494739976181024,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024329546891405858,
      "loss": 4.9769,
      "mean_token_accuracy": 0.19230843633413314,
      "num_tokens": 121302208.0,
      "step": 55365
    },
    {
      "entropy": 5.820689487457275,
      "epoch": 5.4952362048431915,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024326039132725424,
      "loss": 5.0212,
      "mean_token_accuracy": 0.19170837551355363,
      "num_tokens": 121313300.0,
      "step": 55370
    },
    {
      "entropy": 5.782001161575318,
      "epoch": 5.495732433505359,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024322531452752383,
      "loss": 4.9538,
      "mean_token_accuracy": 0.19886904954910278,
      "num_tokens": 121324618.0,
      "step": 55375
    },
    {
      "entropy": 5.717888593673706,
      "epoch": 5.4962286621675265,
      "grad_norm": 1.5,
      "learning_rate": 0.00024319023851573706,
      "loss": 4.9479,
      "mean_token_accuracy": 0.1970517158508301,
      "num_tokens": 121334406.0,
      "step": 55380
    },
    {
      "entropy": 5.7385540962219235,
      "epoch": 5.496724890829694,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024315516329276389,
      "loss": 5.0052,
      "mean_token_accuracy": 0.19962085634469987,
      "num_tokens": 121346093.0,
      "step": 55385
    },
    {
      "entropy": 5.7033556461334225,
      "epoch": 5.497221119491861,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024312008885947402,
      "loss": 4.9573,
      "mean_token_accuracy": 0.19200903326272964,
      "num_tokens": 121356299.0,
      "step": 55390
    },
    {
      "entropy": 5.666078567504883,
      "epoch": 5.49771734815403,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002430850152167373,
      "loss": 4.9124,
      "mean_token_accuracy": 0.20067117512226104,
      "num_tokens": 121367211.0,
      "step": 55395
    },
    {
      "entropy": 5.710581731796265,
      "epoch": 5.498213576816197,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002430499423654234,
      "loss": 4.9234,
      "mean_token_accuracy": 0.19524622857570648,
      "num_tokens": 121378437.0,
      "step": 55400
    },
    {
      "entropy": 5.723660707473755,
      "epoch": 5.498709805478365,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00024301487030640206,
      "loss": 4.8921,
      "mean_token_accuracy": 0.1973166987299919,
      "num_tokens": 121390583.0,
      "step": 55405
    },
    {
      "entropy": 5.699514532089234,
      "epoch": 5.499206034140532,
      "grad_norm": 1.15625,
      "learning_rate": 0.000242979799040543,
      "loss": 4.9163,
      "mean_token_accuracy": 0.19611674994230271,
      "num_tokens": 121402288.0,
      "step": 55410
    },
    {
      "entropy": 5.670206928253174,
      "epoch": 5.4997022628027,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024294472856871603,
      "loss": 4.9611,
      "mean_token_accuracy": 0.19643963873386383,
      "num_tokens": 121412831.0,
      "step": 55415
    },
    {
      "entropy": 5.677396440505982,
      "epoch": 5.500198491464867,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024290965889179063,
      "loss": 4.9484,
      "mean_token_accuracy": 0.1969195395708084,
      "num_tokens": 121424833.0,
      "step": 55420
    },
    {
      "entropy": 5.73442530632019,
      "epoch": 5.500694720127035,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00024287459001063656,
      "loss": 4.9044,
      "mean_token_accuracy": 0.19993714690208436,
      "num_tokens": 121435877.0,
      "step": 55425
    },
    {
      "entropy": 5.73264889717102,
      "epoch": 5.501190948789202,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002428395219261234,
      "loss": 4.9207,
      "mean_token_accuracy": 0.20042168200016022,
      "num_tokens": 121448245.0,
      "step": 55430
    },
    {
      "entropy": 5.691449928283691,
      "epoch": 5.50168717745137,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024280445463912087,
      "loss": 4.9398,
      "mean_token_accuracy": 0.19138118624687195,
      "num_tokens": 121458320.0,
      "step": 55435
    },
    {
      "entropy": 5.700177478790283,
      "epoch": 5.502183406113537,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024276938815049842,
      "loss": 4.9165,
      "mean_token_accuracy": 0.2010437875986099,
      "num_tokens": 121468820.0,
      "step": 55440
    },
    {
      "entropy": 5.770939016342163,
      "epoch": 5.5026796347757045,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00024273432246112576,
      "loss": 4.8969,
      "mean_token_accuracy": 0.1948443681001663,
      "num_tokens": 121479863.0,
      "step": 55445
    },
    {
      "entropy": 5.697417879104615,
      "epoch": 5.503175863437872,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024269925757187233,
      "loss": 4.9212,
      "mean_token_accuracy": 0.199719075858593,
      "num_tokens": 121490947.0,
      "step": 55450
    },
    {
      "entropy": 5.6647714138031,
      "epoch": 5.5036720921000395,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024266419348360774,
      "loss": 4.8678,
      "mean_token_accuracy": 0.20166699588298798,
      "num_tokens": 121501133.0,
      "step": 55455
    },
    {
      "entropy": 5.700705814361572,
      "epoch": 5.504168320762207,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002426291301972015,
      "loss": 4.9668,
      "mean_token_accuracy": 0.19195228070020676,
      "num_tokens": 121512657.0,
      "step": 55460
    },
    {
      "entropy": 5.711902189254761,
      "epoch": 5.504664549424374,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000242594067713523,
      "loss": 4.9149,
      "mean_token_accuracy": 0.1923372521996498,
      "num_tokens": 121524080.0,
      "step": 55465
    },
    {
      "entropy": 5.660768985748291,
      "epoch": 5.505160778086542,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024255900603344178,
      "loss": 4.8614,
      "mean_token_accuracy": 0.20371750593185425,
      "num_tokens": 121533244.0,
      "step": 55470
    },
    {
      "entropy": 5.62596640586853,
      "epoch": 5.505657006748709,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024252394515782732,
      "loss": 4.8585,
      "mean_token_accuracy": 0.20705859661102294,
      "num_tokens": 121544541.0,
      "step": 55475
    },
    {
      "entropy": 5.747244548797608,
      "epoch": 5.506153235410878,
      "grad_norm": 1.25,
      "learning_rate": 0.00024248888508754914,
      "loss": 5.0238,
      "mean_token_accuracy": 0.19902598708868027,
      "num_tokens": 121556336.0,
      "step": 55480
    },
    {
      "entropy": 5.670877361297608,
      "epoch": 5.506649464073045,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024245382582347647,
      "loss": 4.8956,
      "mean_token_accuracy": 0.2057880401611328,
      "num_tokens": 121566411.0,
      "step": 55485
    },
    {
      "entropy": 5.685662698745728,
      "epoch": 5.507145692735213,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024241876736647884,
      "loss": 4.9545,
      "mean_token_accuracy": 0.19556091278791427,
      "num_tokens": 121577341.0,
      "step": 55490
    },
    {
      "entropy": 5.742689180374145,
      "epoch": 5.50764192139738,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024238370971742552,
      "loss": 5.0038,
      "mean_token_accuracy": 0.1922590434551239,
      "num_tokens": 121588885.0,
      "step": 55495
    },
    {
      "entropy": 5.718653202056885,
      "epoch": 5.508138150059548,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024234865287718589,
      "loss": 4.9381,
      "mean_token_accuracy": 0.2001113325357437,
      "num_tokens": 121599551.0,
      "step": 55500
    },
    {
      "entropy": 5.718556022644043,
      "epoch": 5.508634378721715,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024231359684662935,
      "loss": 4.9785,
      "mean_token_accuracy": 0.19175284057855607,
      "num_tokens": 121609662.0,
      "step": 55505
    },
    {
      "entropy": 5.702929162979126,
      "epoch": 5.509130607383883,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024227854162662526,
      "loss": 4.9581,
      "mean_token_accuracy": 0.19249749034643174,
      "num_tokens": 121620515.0,
      "step": 55510
    },
    {
      "entropy": 5.725630140304565,
      "epoch": 5.50962683604605,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002422434872180427,
      "loss": 4.9557,
      "mean_token_accuracy": 0.1997973844408989,
      "num_tokens": 121630910.0,
      "step": 55515
    },
    {
      "entropy": 5.686618661880493,
      "epoch": 5.5101230647082176,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024220843362175116,
      "loss": 4.9281,
      "mean_token_accuracy": 0.19801986813545228,
      "num_tokens": 121640905.0,
      "step": 55520
    },
    {
      "entropy": 5.756599855422974,
      "epoch": 5.510619293370385,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002421733808386198,
      "loss": 4.9441,
      "mean_token_accuracy": 0.2017258495092392,
      "num_tokens": 121653217.0,
      "step": 55525
    },
    {
      "entropy": 5.723674058914185,
      "epoch": 5.5111155220325525,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024213832886951775,
      "loss": 4.977,
      "mean_token_accuracy": 0.19497186988592147,
      "num_tokens": 121663772.0,
      "step": 55530
    },
    {
      "entropy": 5.722745609283447,
      "epoch": 5.51161175069472,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024210327771531442,
      "loss": 4.9452,
      "mean_token_accuracy": 0.20156626552343368,
      "num_tokens": 121673375.0,
      "step": 55535
    },
    {
      "entropy": 5.743948650360108,
      "epoch": 5.5121079793568875,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024206822737687895,
      "loss": 4.9521,
      "mean_token_accuracy": 0.19837722480297088,
      "num_tokens": 121684344.0,
      "step": 55540
    },
    {
      "entropy": 5.706476020812988,
      "epoch": 5.512604208019055,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024203317785508044,
      "loss": 4.8866,
      "mean_token_accuracy": 0.2044857308268547,
      "num_tokens": 121696885.0,
      "step": 55545
    },
    {
      "entropy": 5.7248931407928465,
      "epoch": 5.513100436681222,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024199812915078812,
      "loss": 4.94,
      "mean_token_accuracy": 0.19275151938199997,
      "num_tokens": 121708856.0,
      "step": 55550
    },
    {
      "entropy": 5.720842885971069,
      "epoch": 5.51359666534339,
      "grad_norm": 1.25,
      "learning_rate": 0.00024196308126487099,
      "loss": 4.9787,
      "mean_token_accuracy": 0.19557476937770843,
      "num_tokens": 121720640.0,
      "step": 55555
    },
    {
      "entropy": 5.750701999664306,
      "epoch": 5.514092894005557,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024192803419819831,
      "loss": 4.9849,
      "mean_token_accuracy": 0.19361279308795928,
      "num_tokens": 121731009.0,
      "step": 55560
    },
    {
      "entropy": 5.717014122009277,
      "epoch": 5.514589122667726,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002418929879516391,
      "loss": 4.9774,
      "mean_token_accuracy": 0.19384915828704835,
      "num_tokens": 121743213.0,
      "step": 55565
    },
    {
      "entropy": 5.698099279403687,
      "epoch": 5.515085351329893,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002418579425260625,
      "loss": 4.9705,
      "mean_token_accuracy": 0.19916721135377885,
      "num_tokens": 121754268.0,
      "step": 55570
    },
    {
      "entropy": 5.692887306213379,
      "epoch": 5.515581579992061,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024182289792233752,
      "loss": 4.9324,
      "mean_token_accuracy": 0.1925228700041771,
      "num_tokens": 121763711.0,
      "step": 55575
    },
    {
      "entropy": 5.643364763259887,
      "epoch": 5.516077808654228,
      "grad_norm": 1.546875,
      "learning_rate": 0.00024178785414133314,
      "loss": 4.8793,
      "mean_token_accuracy": 0.2019619181752205,
      "num_tokens": 121773043.0,
      "step": 55580
    },
    {
      "entropy": 5.7541533470153805,
      "epoch": 5.516574037316396,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024175281118391835,
      "loss": 4.9868,
      "mean_token_accuracy": 0.19588330686092376,
      "num_tokens": 121784865.0,
      "step": 55585
    },
    {
      "entropy": 5.759046936035157,
      "epoch": 5.517070265978563,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024171776905096228,
      "loss": 4.9378,
      "mean_token_accuracy": 0.19308790415525437,
      "num_tokens": 121795513.0,
      "step": 55590
    },
    {
      "entropy": 5.652027893066406,
      "epoch": 5.517566494640731,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024168272774333384,
      "loss": 4.8297,
      "mean_token_accuracy": 0.20106656849384308,
      "num_tokens": 121806513.0,
      "step": 55595
    },
    {
      "entropy": 5.667620038986206,
      "epoch": 5.518062723302898,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024164768726190196,
      "loss": 4.9755,
      "mean_token_accuracy": 0.20044564157724382,
      "num_tokens": 121817643.0,
      "step": 55600
    },
    {
      "entropy": 5.699527835845947,
      "epoch": 5.5185589519650655,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024161264760753554,
      "loss": 4.9764,
      "mean_token_accuracy": 0.20333542823791503,
      "num_tokens": 121828289.0,
      "step": 55605
    },
    {
      "entropy": 5.733957290649414,
      "epoch": 5.519055180627233,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024157760878110347,
      "loss": 4.8923,
      "mean_token_accuracy": 0.20826796740293502,
      "num_tokens": 121838750.0,
      "step": 55610
    },
    {
      "entropy": 5.658882141113281,
      "epoch": 5.5195514092894005,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024154257078347463,
      "loss": 4.9032,
      "mean_token_accuracy": 0.19785390049219131,
      "num_tokens": 121848678.0,
      "step": 55615
    },
    {
      "entropy": 5.717404699325561,
      "epoch": 5.520047637951568,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024150753361551808,
      "loss": 4.8924,
      "mean_token_accuracy": 0.20193860679864883,
      "num_tokens": 121859043.0,
      "step": 55620
    },
    {
      "entropy": 5.653809881210327,
      "epoch": 5.520543866613735,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024147249727810255,
      "loss": 4.8633,
      "mean_token_accuracy": 0.20692941695451736,
      "num_tokens": 121868999.0,
      "step": 55625
    },
    {
      "entropy": 5.690823888778686,
      "epoch": 5.521040095275903,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024143746177209674,
      "loss": 4.8994,
      "mean_token_accuracy": 0.19302715957164765,
      "num_tokens": 121880933.0,
      "step": 55630
    },
    {
      "entropy": 5.684233903884888,
      "epoch": 5.52153632393807,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024140242709836963,
      "loss": 4.9294,
      "mean_token_accuracy": 0.1948453053832054,
      "num_tokens": 121892048.0,
      "step": 55635
    },
    {
      "entropy": 5.695020532608032,
      "epoch": 5.522032552600238,
      "grad_norm": 1.375,
      "learning_rate": 0.00024136739325778984,
      "loss": 4.9373,
      "mean_token_accuracy": 0.19818575531244279,
      "num_tokens": 121902272.0,
      "step": 55640
    },
    {
      "entropy": 5.672168493270874,
      "epoch": 5.522528781262405,
      "grad_norm": 1.25,
      "learning_rate": 0.00024133236025122628,
      "loss": 4.917,
      "mean_token_accuracy": 0.19626290947198868,
      "num_tokens": 121912447.0,
      "step": 55645
    },
    {
      "entropy": 5.669863557815551,
      "epoch": 5.523025009924574,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024129732807954762,
      "loss": 4.928,
      "mean_token_accuracy": 0.19964390397071838,
      "num_tokens": 121924072.0,
      "step": 55650
    },
    {
      "entropy": 5.676111841201783,
      "epoch": 5.523521238586741,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024126229674362266,
      "loss": 4.8856,
      "mean_token_accuracy": 0.20824060440063477,
      "num_tokens": 121934851.0,
      "step": 55655
    },
    {
      "entropy": 5.689506101608276,
      "epoch": 5.524017467248909,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024122726624431997,
      "loss": 4.8943,
      "mean_token_accuracy": 0.19731239527463912,
      "num_tokens": 121944917.0,
      "step": 55660
    },
    {
      "entropy": 5.702256727218628,
      "epoch": 5.524513695911076,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00024119223658250838,
      "loss": 4.9335,
      "mean_token_accuracy": 0.20340272784233093,
      "num_tokens": 121955500.0,
      "step": 55665
    },
    {
      "entropy": 5.723450088500977,
      "epoch": 5.525009924573244,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00024115720775905641,
      "loss": 4.9705,
      "mean_token_accuracy": 0.19527547955513,
      "num_tokens": 121966876.0,
      "step": 55670
    },
    {
      "entropy": 5.705947971343994,
      "epoch": 5.525506153235411,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024112217977483274,
      "loss": 4.9609,
      "mean_token_accuracy": 0.19221167117357255,
      "num_tokens": 121976394.0,
      "step": 55675
    },
    {
      "entropy": 5.70353136062622,
      "epoch": 5.5260023818975785,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024108715263070602,
      "loss": 4.9368,
      "mean_token_accuracy": 0.1974706158041954,
      "num_tokens": 121987526.0,
      "step": 55680
    },
    {
      "entropy": 5.652562093734741,
      "epoch": 5.526498610559746,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024105212632754492,
      "loss": 4.977,
      "mean_token_accuracy": 0.19486781656742097,
      "num_tokens": 122000208.0,
      "step": 55685
    },
    {
      "entropy": 5.613819694519043,
      "epoch": 5.5269948392219135,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002410171008662179,
      "loss": 4.8114,
      "mean_token_accuracy": 0.21009310632944106,
      "num_tokens": 122010138.0,
      "step": 55690
    },
    {
      "entropy": 5.74553370475769,
      "epoch": 5.527491067884081,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002409820762475935,
      "loss": 5.0788,
      "mean_token_accuracy": 0.18677215725183488,
      "num_tokens": 122021500.0,
      "step": 55695
    },
    {
      "entropy": 5.7099387645721436,
      "epoch": 5.527987296546248,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002409470524725404,
      "loss": 4.9111,
      "mean_token_accuracy": 0.1926966905593872,
      "num_tokens": 122032744.0,
      "step": 55700
    },
    {
      "entropy": 5.7326854228973385,
      "epoch": 5.528483525208416,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024091202954192687,
      "loss": 4.9331,
      "mean_token_accuracy": 0.19748219102621078,
      "num_tokens": 122043254.0,
      "step": 55705
    },
    {
      "entropy": 5.693455553054809,
      "epoch": 5.528979753870583,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024087700745662166,
      "loss": 4.9311,
      "mean_token_accuracy": 0.20117112547159194,
      "num_tokens": 122053406.0,
      "step": 55710
    },
    {
      "entropy": 5.6520530700683596,
      "epoch": 5.529475982532751,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024084198621749314,
      "loss": 4.9662,
      "mean_token_accuracy": 0.1904517963528633,
      "num_tokens": 122064390.0,
      "step": 55715
    },
    {
      "entropy": 5.730183696746826,
      "epoch": 5.529972211194918,
      "grad_norm": 1.15625,
      "learning_rate": 0.00024080696582540979,
      "loss": 4.9604,
      "mean_token_accuracy": 0.18695453852415084,
      "num_tokens": 122076025.0,
      "step": 55720
    },
    {
      "entropy": 5.662585830688476,
      "epoch": 5.530468439857086,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024077194628123995,
      "loss": 4.7978,
      "mean_token_accuracy": 0.2108218178153038,
      "num_tokens": 122087435.0,
      "step": 55725
    },
    {
      "entropy": 5.724973869323731,
      "epoch": 5.530964668519253,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024073692758585214,
      "loss": 4.97,
      "mean_token_accuracy": 0.19741852581501007,
      "num_tokens": 122098469.0,
      "step": 55730
    },
    {
      "entropy": 5.724191093444825,
      "epoch": 5.531460897181422,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024070190974011458,
      "loss": 5.0016,
      "mean_token_accuracy": 0.19709577411413193,
      "num_tokens": 122110086.0,
      "step": 55735
    },
    {
      "entropy": 5.727166986465454,
      "epoch": 5.531957125843588,
      "grad_norm": 1.25,
      "learning_rate": 0.00024066689274489583,
      "loss": 4.9737,
      "mean_token_accuracy": 0.19231536537408828,
      "num_tokens": 122121985.0,
      "step": 55740
    },
    {
      "entropy": 5.770338439941407,
      "epoch": 5.532453354505757,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024063187660106416,
      "loss": 4.973,
      "mean_token_accuracy": 0.1910538747906685,
      "num_tokens": 122132998.0,
      "step": 55745
    },
    {
      "entropy": 5.676333808898926,
      "epoch": 5.532949583167924,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024059686130948794,
      "loss": 4.8855,
      "mean_token_accuracy": 0.20616454184055327,
      "num_tokens": 122144386.0,
      "step": 55750
    },
    {
      "entropy": 5.666630411148072,
      "epoch": 5.5334458118300915,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024056184687103532,
      "loss": 4.906,
      "mean_token_accuracy": 0.20071382373571395,
      "num_tokens": 122155570.0,
      "step": 55755
    },
    {
      "entropy": 5.646411323547364,
      "epoch": 5.533942040492259,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024052683328657477,
      "loss": 4.8776,
      "mean_token_accuracy": 0.19837324023246766,
      "num_tokens": 122164822.0,
      "step": 55760
    },
    {
      "entropy": 5.710399198532104,
      "epoch": 5.5344382691544265,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024049182055697438,
      "loss": 4.9161,
      "mean_token_accuracy": 0.19396490454673768,
      "num_tokens": 122175412.0,
      "step": 55765
    },
    {
      "entropy": 5.687149095535278,
      "epoch": 5.534934497816594,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002404568086831026,
      "loss": 4.9278,
      "mean_token_accuracy": 0.20165313482284547,
      "num_tokens": 122185442.0,
      "step": 55770
    },
    {
      "entropy": 5.726541805267334,
      "epoch": 5.535430726478761,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024042179766582745,
      "loss": 4.9832,
      "mean_token_accuracy": 0.19512067139148712,
      "num_tokens": 122197837.0,
      "step": 55775
    },
    {
      "entropy": 5.725201416015625,
      "epoch": 5.535926955140929,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024038678750601728,
      "loss": 4.9636,
      "mean_token_accuracy": 0.20114609450101853,
      "num_tokens": 122208959.0,
      "step": 55780
    },
    {
      "entropy": 5.680024242401123,
      "epoch": 5.536423183803096,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024035177820454017,
      "loss": 4.8797,
      "mean_token_accuracy": 0.20560152381658553,
      "num_tokens": 122220819.0,
      "step": 55785
    },
    {
      "entropy": 5.6354930877685545,
      "epoch": 5.536919412465264,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024031676976226423,
      "loss": 4.8307,
      "mean_token_accuracy": 0.19604478925466537,
      "num_tokens": 122232289.0,
      "step": 55790
    },
    {
      "entropy": 5.7020447731018065,
      "epoch": 5.537415641127431,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024028176218005771,
      "loss": 4.9613,
      "mean_token_accuracy": 0.19600165635347366,
      "num_tokens": 122244502.0,
      "step": 55795
    },
    {
      "entropy": 5.653635263442993,
      "epoch": 5.537911869789599,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002402467554587887,
      "loss": 4.8799,
      "mean_token_accuracy": 0.20192766189575195,
      "num_tokens": 122255331.0,
      "step": 55800
    },
    {
      "entropy": 5.707657670974731,
      "epoch": 5.538408098451766,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002402117495993253,
      "loss": 4.9192,
      "mean_token_accuracy": 0.20028993636369705,
      "num_tokens": 122266647.0,
      "step": 55805
    },
    {
      "entropy": 5.740614557266236,
      "epoch": 5.538904327113934,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024017674460253546,
      "loss": 4.9943,
      "mean_token_accuracy": 0.19505588859319686,
      "num_tokens": 122278158.0,
      "step": 55810
    },
    {
      "entropy": 5.677473640441894,
      "epoch": 5.539400555776101,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002401417404692874,
      "loss": 4.9604,
      "mean_token_accuracy": 0.19685634821653367,
      "num_tokens": 122288548.0,
      "step": 55815
    },
    {
      "entropy": 5.7056537628173825,
      "epoch": 5.53989678443827,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024010673720044896,
      "loss": 4.9933,
      "mean_token_accuracy": 0.19691300690174102,
      "num_tokens": 122299263.0,
      "step": 55820
    },
    {
      "entropy": 5.5975799560546875,
      "epoch": 5.540393013100436,
      "grad_norm": 1.15625,
      "learning_rate": 0.00024007173479688825,
      "loss": 4.779,
      "mean_token_accuracy": 0.20853162407875062,
      "num_tokens": 122309994.0,
      "step": 55825
    },
    {
      "entropy": 5.6658154964447025,
      "epoch": 5.5408892417626046,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002400367332594734,
      "loss": 4.8927,
      "mean_token_accuracy": 0.19652759283781052,
      "num_tokens": 122320553.0,
      "step": 55830
    },
    {
      "entropy": 5.673855352401733,
      "epoch": 5.541385470424772,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024000173258907216,
      "loss": 4.9498,
      "mean_token_accuracy": 0.19929772168397902,
      "num_tokens": 122331963.0,
      "step": 55835
    },
    {
      "entropy": 5.718288230895996,
      "epoch": 5.5418816990869395,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002399667327865525,
      "loss": 4.9542,
      "mean_token_accuracy": 0.19211844801902772,
      "num_tokens": 122342696.0,
      "step": 55840
    },
    {
      "entropy": 5.688075113296509,
      "epoch": 5.542377927749107,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023993173385278245,
      "loss": 4.9801,
      "mean_token_accuracy": 0.19455744624137877,
      "num_tokens": 122354212.0,
      "step": 55845
    },
    {
      "entropy": 5.631690835952758,
      "epoch": 5.5428741564112745,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002398967357886297,
      "loss": 4.795,
      "mean_token_accuracy": 0.2145850107073784,
      "num_tokens": 122365148.0,
      "step": 55850
    },
    {
      "entropy": 5.655208158493042,
      "epoch": 5.543370385073442,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023986173859496236,
      "loss": 4.9278,
      "mean_token_accuracy": 0.19601832032203675,
      "num_tokens": 122375964.0,
      "step": 55855
    },
    {
      "entropy": 5.64234414100647,
      "epoch": 5.543866613735609,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023982674227264816,
      "loss": 4.8711,
      "mean_token_accuracy": 0.19792528748512267,
      "num_tokens": 122386936.0,
      "step": 55860
    },
    {
      "entropy": 5.720178318023682,
      "epoch": 5.544362842397777,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023979174682255494,
      "loss": 4.9953,
      "mean_token_accuracy": 0.1915385529398918,
      "num_tokens": 122399010.0,
      "step": 55865
    },
    {
      "entropy": 5.741361141204834,
      "epoch": 5.544859071059944,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023975675224555067,
      "loss": 4.9936,
      "mean_token_accuracy": 0.2000487580895424,
      "num_tokens": 122408890.0,
      "step": 55870
    },
    {
      "entropy": 5.744829082489014,
      "epoch": 5.545355299722112,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023972175854250282,
      "loss": 5.0106,
      "mean_token_accuracy": 0.19732455015182496,
      "num_tokens": 122419791.0,
      "step": 55875
    },
    {
      "entropy": 5.771618318557739,
      "epoch": 5.545851528384279,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023968676571427939,
      "loss": 5.0166,
      "mean_token_accuracy": 0.1859825313091278,
      "num_tokens": 122430994.0,
      "step": 55880
    },
    {
      "entropy": 5.723647546768189,
      "epoch": 5.546347757046447,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023965177376174807,
      "loss": 4.971,
      "mean_token_accuracy": 0.19035326391458512,
      "num_tokens": 122442059.0,
      "step": 55885
    },
    {
      "entropy": 5.637291765213012,
      "epoch": 5.546843985708614,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023961678268577663,
      "loss": 4.847,
      "mean_token_accuracy": 0.21059153378009796,
      "num_tokens": 122452616.0,
      "step": 55890
    },
    {
      "entropy": 5.797655534744263,
      "epoch": 5.547340214370782,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002395817924872328,
      "loss": 5.0356,
      "mean_token_accuracy": 0.19386242628097533,
      "num_tokens": 122463458.0,
      "step": 55895
    },
    {
      "entropy": 5.657173776626587,
      "epoch": 5.547836443032949,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023954680316698408,
      "loss": 4.8613,
      "mean_token_accuracy": 0.19777487069368363,
      "num_tokens": 122473829.0,
      "step": 55900
    },
    {
      "entropy": 5.64336838722229,
      "epoch": 5.548332671695118,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023951181472589827,
      "loss": 4.906,
      "mean_token_accuracy": 0.20041175186634064,
      "num_tokens": 122484637.0,
      "step": 55905
    },
    {
      "entropy": 5.693823671340942,
      "epoch": 5.548828900357284,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023947682716484294,
      "loss": 4.9137,
      "mean_token_accuracy": 0.19227480739355088,
      "num_tokens": 122494594.0,
      "step": 55910
    },
    {
      "entropy": 5.655536890029907,
      "epoch": 5.5493251290194525,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023944184048468588,
      "loss": 4.9406,
      "mean_token_accuracy": 0.19522227495908737,
      "num_tokens": 122505087.0,
      "step": 55915
    },
    {
      "entropy": 5.68560676574707,
      "epoch": 5.54982135768162,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023940685468629448,
      "loss": 4.8451,
      "mean_token_accuracy": 0.206942555308342,
      "num_tokens": 122515668.0,
      "step": 55920
    },
    {
      "entropy": 5.626320648193359,
      "epoch": 5.5503175863437875,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023937186977053643,
      "loss": 4.8329,
      "mean_token_accuracy": 0.20725952386856078,
      "num_tokens": 122525613.0,
      "step": 55925
    },
    {
      "entropy": 5.687081289291382,
      "epoch": 5.550813815005955,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023933688573827923,
      "loss": 4.9273,
      "mean_token_accuracy": 0.20559446960687638,
      "num_tokens": 122535646.0,
      "step": 55930
    },
    {
      "entropy": 5.705054235458374,
      "epoch": 5.551310043668122,
      "grad_norm": 1.375,
      "learning_rate": 0.00023930190259039032,
      "loss": 4.9277,
      "mean_token_accuracy": 0.20379650890827178,
      "num_tokens": 122546114.0,
      "step": 55935
    },
    {
      "entropy": 5.6852775573730465,
      "epoch": 5.55180627233029,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023926692032773733,
      "loss": 4.8881,
      "mean_token_accuracy": 0.19680486172437667,
      "num_tokens": 122558046.0,
      "step": 55940
    },
    {
      "entropy": 5.710102558135986,
      "epoch": 5.552302500992457,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023923193895118778,
      "loss": 4.9606,
      "mean_token_accuracy": 0.2008986547589302,
      "num_tokens": 122568409.0,
      "step": 55945
    },
    {
      "entropy": 5.633537435531617,
      "epoch": 5.552798729654625,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002391969584616091,
      "loss": 4.9278,
      "mean_token_accuracy": 0.20387699007987975,
      "num_tokens": 122579707.0,
      "step": 55950
    },
    {
      "entropy": 5.695993423461914,
      "epoch": 5.553294958316792,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023916197885986862,
      "loss": 4.9745,
      "mean_token_accuracy": 0.18903883546590805,
      "num_tokens": 122590525.0,
      "step": 55955
    },
    {
      "entropy": 5.685845184326172,
      "epoch": 5.55379118697896,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002391270001468339,
      "loss": 4.8962,
      "mean_token_accuracy": 0.19497138261795044,
      "num_tokens": 122601696.0,
      "step": 55960
    },
    {
      "entropy": 5.684195804595947,
      "epoch": 5.554287415641127,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023909202232337228,
      "loss": 4.9006,
      "mean_token_accuracy": 0.19630715399980544,
      "num_tokens": 122613144.0,
      "step": 55965
    },
    {
      "entropy": 5.649554061889648,
      "epoch": 5.554783644303295,
      "grad_norm": 1.375,
      "learning_rate": 0.00023905704539035105,
      "loss": 4.888,
      "mean_token_accuracy": 0.20909304320812225,
      "num_tokens": 122623645.0,
      "step": 55970
    },
    {
      "entropy": 5.727469778060913,
      "epoch": 5.555279872965462,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023902206934863768,
      "loss": 4.9731,
      "mean_token_accuracy": 0.19359584003686905,
      "num_tokens": 122634271.0,
      "step": 55975
    },
    {
      "entropy": 5.752982044219971,
      "epoch": 5.55577610162763,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023898709419909952,
      "loss": 4.9617,
      "mean_token_accuracy": 0.1972220703959465,
      "num_tokens": 122645668.0,
      "step": 55980
    },
    {
      "entropy": 5.692802429199219,
      "epoch": 5.556272330289797,
      "grad_norm": 1.25,
      "learning_rate": 0.0002389521199426038,
      "loss": 4.9188,
      "mean_token_accuracy": 0.20023919492959977,
      "num_tokens": 122657121.0,
      "step": 55985
    },
    {
      "entropy": 5.6684845924377445,
      "epoch": 5.5567685589519655,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002389171465800178,
      "loss": 4.8967,
      "mean_token_accuracy": 0.19948478043079376,
      "num_tokens": 122668630.0,
      "step": 55990
    },
    {
      "entropy": 5.642437267303467,
      "epoch": 5.557264787614132,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023888217411220886,
      "loss": 4.8821,
      "mean_token_accuracy": 0.19940195381641387,
      "num_tokens": 122680326.0,
      "step": 55995
    },
    {
      "entropy": 5.6740340232849125,
      "epoch": 5.5577610162763005,
      "grad_norm": 1.25,
      "learning_rate": 0.00023884720254004405,
      "loss": 4.8953,
      "mean_token_accuracy": 0.20353467017412186,
      "num_tokens": 122692225.0,
      "step": 56000
    },
    {
      "entropy": 5.722947978973389,
      "epoch": 5.558257244938468,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023881223186439084,
      "loss": 5.0161,
      "mean_token_accuracy": 0.19516635239124297,
      "num_tokens": 122702502.0,
      "step": 56005
    },
    {
      "entropy": 5.650466346740723,
      "epoch": 5.558753473600635,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023877726208611629,
      "loss": 4.867,
      "mean_token_accuracy": 0.20802261233329772,
      "num_tokens": 122712904.0,
      "step": 56010
    },
    {
      "entropy": 5.681906223297119,
      "epoch": 5.559249702262803,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002387422932060876,
      "loss": 4.8829,
      "mean_token_accuracy": 0.19915490299463273,
      "num_tokens": 122723398.0,
      "step": 56015
    },
    {
      "entropy": 5.707686042785644,
      "epoch": 5.55974593092497,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002387073252251718,
      "loss": 4.9333,
      "mean_token_accuracy": 0.19930949658155442,
      "num_tokens": 122735088.0,
      "step": 56020
    },
    {
      "entropy": 5.7155204772949215,
      "epoch": 5.560242159587138,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023867235814423627,
      "loss": 4.9713,
      "mean_token_accuracy": 0.2027850091457367,
      "num_tokens": 122746357.0,
      "step": 56025
    },
    {
      "entropy": 5.713152599334717,
      "epoch": 5.560738388249305,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023863739196414786,
      "loss": 4.9121,
      "mean_token_accuracy": 0.20217020958662033,
      "num_tokens": 122759000.0,
      "step": 56030
    },
    {
      "entropy": 5.692658424377441,
      "epoch": 5.561234616911473,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023860242668577382,
      "loss": 4.8958,
      "mean_token_accuracy": 0.2047731325030327,
      "num_tokens": 122770627.0,
      "step": 56035
    },
    {
      "entropy": 5.659491395950317,
      "epoch": 5.56173084557364,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023856746230998117,
      "loss": 4.8998,
      "mean_token_accuracy": 0.20065607279539108,
      "num_tokens": 122782121.0,
      "step": 56040
    },
    {
      "entropy": 5.626701974868775,
      "epoch": 5.562227074235808,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000238532498837637,
      "loss": 4.8697,
      "mean_token_accuracy": 0.200515416264534,
      "num_tokens": 122791982.0,
      "step": 56045
    },
    {
      "entropy": 5.58896164894104,
      "epoch": 5.562723302897975,
      "grad_norm": 1.1875,
      "learning_rate": 0.00023849753626960818,
      "loss": 4.8153,
      "mean_token_accuracy": 0.21475036442279816,
      "num_tokens": 122803089.0,
      "step": 56050
    },
    {
      "entropy": 5.661938714981079,
      "epoch": 5.563219531560143,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023846257460676185,
      "loss": 4.9238,
      "mean_token_accuracy": 0.19726920425891875,
      "num_tokens": 122813985.0,
      "step": 56055
    },
    {
      "entropy": 5.698890542984008,
      "epoch": 5.56371576022231,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023842761384996499,
      "loss": 4.9203,
      "mean_token_accuracy": 0.1950390085577965,
      "num_tokens": 122824361.0,
      "step": 56060
    },
    {
      "entropy": 5.6360328674316404,
      "epoch": 5.564211988884478,
      "grad_norm": 1.5,
      "learning_rate": 0.00023839265400008448,
      "loss": 4.8795,
      "mean_token_accuracy": 0.20206502377986907,
      "num_tokens": 122834649.0,
      "step": 56065
    },
    {
      "entropy": 5.679733180999756,
      "epoch": 5.564708217546645,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002383576950579872,
      "loss": 4.9592,
      "mean_token_accuracy": 0.19460247904062272,
      "num_tokens": 122846373.0,
      "step": 56070
    },
    {
      "entropy": 5.649705934524536,
      "epoch": 5.5652044462088135,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002383227370245402,
      "loss": 4.9408,
      "mean_token_accuracy": 0.19915429204702378,
      "num_tokens": 122856383.0,
      "step": 56075
    },
    {
      "entropy": 5.785664749145508,
      "epoch": 5.56570067487098,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002382877799006103,
      "loss": 5.0892,
      "mean_token_accuracy": 0.19127354472875596,
      "num_tokens": 122867579.0,
      "step": 56080
    },
    {
      "entropy": 5.682699394226074,
      "epoch": 5.566196903533148,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023825282368706424,
      "loss": 4.8446,
      "mean_token_accuracy": 0.20545158237218858,
      "num_tokens": 122877269.0,
      "step": 56085
    },
    {
      "entropy": 5.716684198379516,
      "epoch": 5.566693132195316,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023821786838476905,
      "loss": 4.911,
      "mean_token_accuracy": 0.19950500577688218,
      "num_tokens": 122888308.0,
      "step": 56090
    },
    {
      "entropy": 5.709998655319214,
      "epoch": 5.567189360857483,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023818291399459147,
      "loss": 4.9462,
      "mean_token_accuracy": 0.19785359501838684,
      "num_tokens": 122900034.0,
      "step": 56095
    },
    {
      "entropy": 5.725315570831299,
      "epoch": 5.567685589519651,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023814796051739828,
      "loss": 4.9722,
      "mean_token_accuracy": 0.1969935342669487,
      "num_tokens": 122911023.0,
      "step": 56100
    },
    {
      "entropy": 5.650909757614135,
      "epoch": 5.568181818181818,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023811300795405628,
      "loss": 4.8554,
      "mean_token_accuracy": 0.20502639859914779,
      "num_tokens": 122922112.0,
      "step": 56105
    },
    {
      "entropy": 5.703216457366944,
      "epoch": 5.568678046843986,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023807805630543223,
      "loss": 4.9296,
      "mean_token_accuracy": 0.19730599522590636,
      "num_tokens": 122933231.0,
      "step": 56110
    },
    {
      "entropy": 5.632648563385009,
      "epoch": 5.569174275506153,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023804310557239266,
      "loss": 4.8804,
      "mean_token_accuracy": 0.20403602570295334,
      "num_tokens": 122945386.0,
      "step": 56115
    },
    {
      "entropy": 5.7417913436889645,
      "epoch": 5.569670504168321,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023800815575580452,
      "loss": 5.0252,
      "mean_token_accuracy": 0.19040438383817673,
      "num_tokens": 122955840.0,
      "step": 56120
    },
    {
      "entropy": 5.67536358833313,
      "epoch": 5.570166732830488,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023797320685653444,
      "loss": 4.9752,
      "mean_token_accuracy": 0.19663156867027282,
      "num_tokens": 122967012.0,
      "step": 56125
    },
    {
      "entropy": 5.713903522491455,
      "epoch": 5.570662961492656,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002379382588754491,
      "loss": 4.9478,
      "mean_token_accuracy": 0.20024412870407104,
      "num_tokens": 122977380.0,
      "step": 56130
    },
    {
      "entropy": 5.667921304702759,
      "epoch": 5.571159190154823,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023790331181341502,
      "loss": 4.9059,
      "mean_token_accuracy": 0.20721933841705323,
      "num_tokens": 122989300.0,
      "step": 56135
    },
    {
      "entropy": 5.731560277938843,
      "epoch": 5.571655418816991,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023786836567129887,
      "loss": 4.9569,
      "mean_token_accuracy": 0.1984405294060707,
      "num_tokens": 122999083.0,
      "step": 56140
    },
    {
      "entropy": 5.65902738571167,
      "epoch": 5.572151647479158,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002378334204499672,
      "loss": 4.9329,
      "mean_token_accuracy": 0.19853852391242982,
      "num_tokens": 123010328.0,
      "step": 56145
    },
    {
      "entropy": 5.759110832214356,
      "epoch": 5.572647876141326,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023779847615028665,
      "loss": 5.0416,
      "mean_token_accuracy": 0.19113052785396575,
      "num_tokens": 123021279.0,
      "step": 56150
    },
    {
      "entropy": 5.696023845672608,
      "epoch": 5.573144104803493,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023776353277312373,
      "loss": 4.9161,
      "mean_token_accuracy": 0.20739226937294006,
      "num_tokens": 123032692.0,
      "step": 56155
    },
    {
      "entropy": 5.7322288990020756,
      "epoch": 5.5736403334656615,
      "grad_norm": 1.125,
      "learning_rate": 0.000237728590319345,
      "loss": 4.9647,
      "mean_token_accuracy": 0.1886005401611328,
      "num_tokens": 123044018.0,
      "step": 56160
    },
    {
      "entropy": 5.772565841674805,
      "epoch": 5.574136562127828,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023769364878981688,
      "loss": 4.9846,
      "mean_token_accuracy": 0.20080512315034865,
      "num_tokens": 123056462.0,
      "step": 56165
    },
    {
      "entropy": 5.748279333114624,
      "epoch": 5.574632790789996,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00023765870818540593,
      "loss": 4.9973,
      "mean_token_accuracy": 0.190334652364254,
      "num_tokens": 123066686.0,
      "step": 56170
    },
    {
      "entropy": 5.703171014785767,
      "epoch": 5.575129019452164,
      "grad_norm": 1.375,
      "learning_rate": 0.00023762376850697842,
      "loss": 4.9943,
      "mean_token_accuracy": 0.19573455154895783,
      "num_tokens": 123077562.0,
      "step": 56175
    },
    {
      "entropy": 5.635232305526733,
      "epoch": 5.575625248114331,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023758882975540103,
      "loss": 4.8833,
      "mean_token_accuracy": 0.2086348354816437,
      "num_tokens": 123087823.0,
      "step": 56180
    },
    {
      "entropy": 5.732800817489624,
      "epoch": 5.576121476776499,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023755389193153999,
      "loss": 4.9979,
      "mean_token_accuracy": 0.19165003448724746,
      "num_tokens": 123098844.0,
      "step": 56185
    },
    {
      "entropy": 5.745448255538941,
      "epoch": 5.576617705438666,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002375189550362618,
      "loss": 4.9987,
      "mean_token_accuracy": 0.189316426217556,
      "num_tokens": 123110398.0,
      "step": 56190
    },
    {
      "entropy": 5.6820978164672855,
      "epoch": 5.577113934100834,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002374840190704327,
      "loss": 4.9376,
      "mean_token_accuracy": 0.1989198550581932,
      "num_tokens": 123121381.0,
      "step": 56195
    },
    {
      "entropy": 5.676547622680664,
      "epoch": 5.577610162763001,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023744908403491917,
      "loss": 4.8575,
      "mean_token_accuracy": 0.20462686568498611,
      "num_tokens": 123132598.0,
      "step": 56200
    },
    {
      "entropy": 5.762012767791748,
      "epoch": 5.578106391425169,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002374141499305873,
      "loss": 4.963,
      "mean_token_accuracy": 0.19425301402807235,
      "num_tokens": 123143797.0,
      "step": 56205
    },
    {
      "entropy": 5.685167694091797,
      "epoch": 5.578602620087336,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002373792167583036,
      "loss": 4.9688,
      "mean_token_accuracy": 0.1936524510383606,
      "num_tokens": 123154094.0,
      "step": 56210
    },
    {
      "entropy": 5.6648640632629395,
      "epoch": 5.579098848749504,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002373442845189343,
      "loss": 4.8669,
      "mean_token_accuracy": 0.1986162319779396,
      "num_tokens": 123165147.0,
      "step": 56215
    },
    {
      "entropy": 5.68911714553833,
      "epoch": 5.579595077411671,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023730935321334556,
      "loss": 4.9367,
      "mean_token_accuracy": 0.195066899061203,
      "num_tokens": 123175564.0,
      "step": 56220
    },
    {
      "entropy": 5.683941411972046,
      "epoch": 5.580091306073839,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002372744228424037,
      "loss": 4.9687,
      "mean_token_accuracy": 0.19003321081399918,
      "num_tokens": 123186859.0,
      "step": 56225
    },
    {
      "entropy": 5.603602600097656,
      "epoch": 5.580587534736006,
      "grad_norm": 1.1875,
      "learning_rate": 0.00023723949340697477,
      "loss": 4.8111,
      "mean_token_accuracy": 0.20927882194519043,
      "num_tokens": 123199018.0,
      "step": 56230
    },
    {
      "entropy": 5.682074546813965,
      "epoch": 5.581083763398174,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023720456490792502,
      "loss": 4.9055,
      "mean_token_accuracy": 0.20150502026081085,
      "num_tokens": 123209793.0,
      "step": 56235
    },
    {
      "entropy": 5.683266353607178,
      "epoch": 5.581579992060341,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023716963734612067,
      "loss": 4.8843,
      "mean_token_accuracy": 0.20896531343460084,
      "num_tokens": 123220708.0,
      "step": 56240
    },
    {
      "entropy": 5.6635438919067385,
      "epoch": 5.582076220722509,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023713471072242786,
      "loss": 4.8992,
      "mean_token_accuracy": 0.20272095501422882,
      "num_tokens": 123231337.0,
      "step": 56245
    },
    {
      "entropy": 5.642841243743897,
      "epoch": 5.582572449384676,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023709978503771257,
      "loss": 4.8559,
      "mean_token_accuracy": 0.20374771654605867,
      "num_tokens": 123241909.0,
      "step": 56250
    },
    {
      "entropy": 5.674492311477661,
      "epoch": 5.583068678046844,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023706486029284103,
      "loss": 4.9691,
      "mean_token_accuracy": 0.20018852204084397,
      "num_tokens": 123253230.0,
      "step": 56255
    },
    {
      "entropy": 5.6891600608825685,
      "epoch": 5.583564906709012,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023702993648867915,
      "loss": 4.936,
      "mean_token_accuracy": 0.19404903650283814,
      "num_tokens": 123263474.0,
      "step": 56260
    },
    {
      "entropy": 5.718634223937988,
      "epoch": 5.584061135371179,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023699501362609295,
      "loss": 4.9731,
      "mean_token_accuracy": 0.19519833326339722,
      "num_tokens": 123274453.0,
      "step": 56265
    },
    {
      "entropy": 5.7063844203948975,
      "epoch": 5.584557364033347,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023696009170594856,
      "loss": 4.986,
      "mean_token_accuracy": 0.1916609987616539,
      "num_tokens": 123286732.0,
      "step": 56270
    },
    {
      "entropy": 5.725224161148072,
      "epoch": 5.585053592695514,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023692517072911197,
      "loss": 4.925,
      "mean_token_accuracy": 0.19197659194469452,
      "num_tokens": 123296113.0,
      "step": 56275
    },
    {
      "entropy": 5.697150135040284,
      "epoch": 5.585549821357682,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023689025069644905,
      "loss": 4.9401,
      "mean_token_accuracy": 0.20074040591716766,
      "num_tokens": 123307140.0,
      "step": 56280
    },
    {
      "entropy": 5.734630489349366,
      "epoch": 5.586046050019849,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023685533160882583,
      "loss": 5.0018,
      "mean_token_accuracy": 0.1948416367173195,
      "num_tokens": 123318183.0,
      "step": 56285
    },
    {
      "entropy": 5.723876953125,
      "epoch": 5.586542278682017,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00023682041346710815,
      "loss": 4.9,
      "mean_token_accuracy": 0.20048570930957793,
      "num_tokens": 123328859.0,
      "step": 56290
    },
    {
      "entropy": 5.707445859909058,
      "epoch": 5.587038507344184,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023678549627216195,
      "loss": 4.9445,
      "mean_token_accuracy": 0.19408351331949233,
      "num_tokens": 123339863.0,
      "step": 56295
    },
    {
      "entropy": 5.668239307403565,
      "epoch": 5.587534736006352,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023675058002485316,
      "loss": 4.8939,
      "mean_token_accuracy": 0.20031320601701735,
      "num_tokens": 123350349.0,
      "step": 56300
    },
    {
      "entropy": 5.698450565338135,
      "epoch": 5.588030964668519,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023671566472604756,
      "loss": 4.9591,
      "mean_token_accuracy": 0.19375187903642654,
      "num_tokens": 123361382.0,
      "step": 56305
    },
    {
      "entropy": 5.634963417053223,
      "epoch": 5.588527193330687,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023668075037661096,
      "loss": 4.8825,
      "mean_token_accuracy": 0.195078806579113,
      "num_tokens": 123371784.0,
      "step": 56310
    },
    {
      "entropy": 5.7386922359466555,
      "epoch": 5.589023421992854,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023664583697740915,
      "loss": 5.0212,
      "mean_token_accuracy": 0.19051896929740905,
      "num_tokens": 123383429.0,
      "step": 56315
    },
    {
      "entropy": 5.74378809928894,
      "epoch": 5.5895196506550215,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023661092452930787,
      "loss": 5.0032,
      "mean_token_accuracy": 0.19437592923641206,
      "num_tokens": 123394453.0,
      "step": 56320
    },
    {
      "entropy": 5.740076065063477,
      "epoch": 5.590015879317189,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023657601303317304,
      "loss": 4.9453,
      "mean_token_accuracy": 0.18932686299085616,
      "num_tokens": 123405212.0,
      "step": 56325
    },
    {
      "entropy": 5.764393711090088,
      "epoch": 5.590512107979357,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023654110248987037,
      "loss": 5.0083,
      "mean_token_accuracy": 0.19264551997184753,
      "num_tokens": 123416554.0,
      "step": 56330
    },
    {
      "entropy": 5.689643383026123,
      "epoch": 5.591008336641524,
      "grad_norm": 1.125,
      "learning_rate": 0.00023650619290026538,
      "loss": 4.9335,
      "mean_token_accuracy": 0.20611998438835144,
      "num_tokens": 123428353.0,
      "step": 56335
    },
    {
      "entropy": 5.693751001358033,
      "epoch": 5.591504565303692,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023647128426522392,
      "loss": 4.8553,
      "mean_token_accuracy": 0.2115221753716469,
      "num_tokens": 123438722.0,
      "step": 56340
    },
    {
      "entropy": 5.657977676391601,
      "epoch": 5.59200079396586,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023643637658561151,
      "loss": 4.9086,
      "mean_token_accuracy": 0.1979384794831276,
      "num_tokens": 123450467.0,
      "step": 56345
    },
    {
      "entropy": 5.680176687240601,
      "epoch": 5.592497022628027,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023640146986229384,
      "loss": 4.9619,
      "mean_token_accuracy": 0.19127605706453324,
      "num_tokens": 123461340.0,
      "step": 56350
    },
    {
      "entropy": 5.685239553451538,
      "epoch": 5.592993251290195,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023636656409613662,
      "loss": 4.9208,
      "mean_token_accuracy": 0.20286624282598495,
      "num_tokens": 123471938.0,
      "step": 56355
    },
    {
      "entropy": 5.744185876846314,
      "epoch": 5.593489479952362,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023633165928800537,
      "loss": 4.9179,
      "mean_token_accuracy": 0.20339178740978242,
      "num_tokens": 123484890.0,
      "step": 56360
    },
    {
      "entropy": 5.75668158531189,
      "epoch": 5.59398570861453,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023629675543876562,
      "loss": 4.9842,
      "mean_token_accuracy": 0.19204585403203964,
      "num_tokens": 123497210.0,
      "step": 56365
    },
    {
      "entropy": 5.694683504104614,
      "epoch": 5.594481937276697,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023626185254928294,
      "loss": 4.9112,
      "mean_token_accuracy": 0.20412623435258864,
      "num_tokens": 123507093.0,
      "step": 56370
    },
    {
      "entropy": 5.679858636856079,
      "epoch": 5.594978165938865,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023622695062042287,
      "loss": 4.9808,
      "mean_token_accuracy": 0.19072602838277816,
      "num_tokens": 123517037.0,
      "step": 56375
    },
    {
      "entropy": 5.734122848510742,
      "epoch": 5.595474394601032,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002361920496530508,
      "loss": 4.9999,
      "mean_token_accuracy": 0.1909735307097435,
      "num_tokens": 123527401.0,
      "step": 56380
    },
    {
      "entropy": 5.6895318031311035,
      "epoch": 5.5959706232632,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002361571496480323,
      "loss": 4.8691,
      "mean_token_accuracy": 0.1956576958298683,
      "num_tokens": 123537581.0,
      "step": 56385
    },
    {
      "entropy": 5.632346296310425,
      "epoch": 5.596466851925367,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023612225060623283,
      "loss": 4.8164,
      "mean_token_accuracy": 0.21325492709875107,
      "num_tokens": 123547698.0,
      "step": 56390
    },
    {
      "entropy": 5.772349405288696,
      "epoch": 5.5969630805875346,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023608735252851778,
      "loss": 5.0735,
      "mean_token_accuracy": 0.1891405925154686,
      "num_tokens": 123559111.0,
      "step": 56395
    },
    {
      "entropy": 5.694631099700928,
      "epoch": 5.597459309249702,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002360524554157525,
      "loss": 4.8538,
      "mean_token_accuracy": 0.206472447514534,
      "num_tokens": 123570113.0,
      "step": 56400
    },
    {
      "entropy": 5.689432048797608,
      "epoch": 5.5979555379118695,
      "grad_norm": 1.59375,
      "learning_rate": 0.00023601755926880243,
      "loss": 4.9057,
      "mean_token_accuracy": 0.1952637627720833,
      "num_tokens": 123581018.0,
      "step": 56405
    },
    {
      "entropy": 5.6664008617401125,
      "epoch": 5.598451766574037,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002359826640885328,
      "loss": 4.9273,
      "mean_token_accuracy": 0.19457386285066605,
      "num_tokens": 123593129.0,
      "step": 56410
    },
    {
      "entropy": 5.6728006362915036,
      "epoch": 5.5989479952362045,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023594776987580912,
      "loss": 4.9381,
      "mean_token_accuracy": 0.19726924300193788,
      "num_tokens": 123603570.0,
      "step": 56415
    },
    {
      "entropy": 5.718664598464966,
      "epoch": 5.599444223898372,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023591287663149657,
      "loss": 4.9317,
      "mean_token_accuracy": 0.19676733165979385,
      "num_tokens": 123614425.0,
      "step": 56420
    },
    {
      "entropy": 5.697463464736939,
      "epoch": 5.59994045256054,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023587798435646052,
      "loss": 5.0066,
      "mean_token_accuracy": 0.19511964917182922,
      "num_tokens": 123626072.0,
      "step": 56425
    },
    {
      "entropy": 5.728346633911133,
      "epoch": 5.600436681222708,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023584309305156608,
      "loss": 4.889,
      "mean_token_accuracy": 0.20214928984642028,
      "num_tokens": 123636740.0,
      "step": 56430
    },
    {
      "entropy": 5.720387077331543,
      "epoch": 5.600932909884875,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002358082027176786,
      "loss": 4.9139,
      "mean_token_accuracy": 0.205728155374527,
      "num_tokens": 123647733.0,
      "step": 56435
    },
    {
      "entropy": 5.698518323898315,
      "epoch": 5.601429138547043,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023577331335566316,
      "loss": 4.9934,
      "mean_token_accuracy": 0.19340402632951736,
      "num_tokens": 123659905.0,
      "step": 56440
    },
    {
      "entropy": 5.756442737579346,
      "epoch": 5.60192536720921,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002357384249663851,
      "loss": 4.9706,
      "mean_token_accuracy": 0.19270990639925004,
      "num_tokens": 123673782.0,
      "step": 56445
    },
    {
      "entropy": 5.746847534179688,
      "epoch": 5.602421595871378,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023570353755070943,
      "loss": 4.978,
      "mean_token_accuracy": 0.19184593111276627,
      "num_tokens": 123684972.0,
      "step": 56450
    },
    {
      "entropy": 5.731316137313843,
      "epoch": 5.602917824533545,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002356686511095015,
      "loss": 4.9436,
      "mean_token_accuracy": 0.19096424132585527,
      "num_tokens": 123695481.0,
      "step": 56455
    },
    {
      "entropy": 5.728246116638184,
      "epoch": 5.603414053195713,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002356337656436261,
      "loss": 4.889,
      "mean_token_accuracy": 0.20621636956930162,
      "num_tokens": 123705578.0,
      "step": 56460
    },
    {
      "entropy": 5.689285945892334,
      "epoch": 5.60391028185788,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002355988811539486,
      "loss": 4.898,
      "mean_token_accuracy": 0.2014904648065567,
      "num_tokens": 123716490.0,
      "step": 56465
    },
    {
      "entropy": 5.686064386367798,
      "epoch": 5.604406510520048,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023556399764133385,
      "loss": 4.9132,
      "mean_token_accuracy": 0.19734452217817305,
      "num_tokens": 123726353.0,
      "step": 56470
    },
    {
      "entropy": 5.637900876998901,
      "epoch": 5.604902739182215,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023552911510664703,
      "loss": 4.8995,
      "mean_token_accuracy": 0.19589820355176926,
      "num_tokens": 123738078.0,
      "step": 56475
    },
    {
      "entropy": 5.720101547241211,
      "epoch": 5.6053989678443825,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002354942335507531,
      "loss": 4.9714,
      "mean_token_accuracy": 0.19484625309705733,
      "num_tokens": 123748445.0,
      "step": 56480
    },
    {
      "entropy": 5.722788286209107,
      "epoch": 5.60589519650655,
      "grad_norm": 1.1875,
      "learning_rate": 0.00023545935297451708,
      "loss": 4.9539,
      "mean_token_accuracy": 0.19338140785694122,
      "num_tokens": 123759636.0,
      "step": 56485
    },
    {
      "entropy": 5.696877384185791,
      "epoch": 5.6063914251687175,
      "grad_norm": 1.25,
      "learning_rate": 0.00023542447337880396,
      "loss": 4.912,
      "mean_token_accuracy": 0.19925432056188583,
      "num_tokens": 123771599.0,
      "step": 56490
    },
    {
      "entropy": 5.7348397731781,
      "epoch": 5.606887653830885,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023538959476447852,
      "loss": 5.0378,
      "mean_token_accuracy": 0.19337273240089417,
      "num_tokens": 123783171.0,
      "step": 56495
    },
    {
      "entropy": 5.760042953491211,
      "epoch": 5.607383882493052,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002353547171324057,
      "loss": 5.0134,
      "mean_token_accuracy": 0.1878489449620247,
      "num_tokens": 123794733.0,
      "step": 56500
    },
    {
      "entropy": 5.718564033508301,
      "epoch": 5.60788011115522,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023531984048345063,
      "loss": 4.9288,
      "mean_token_accuracy": 0.19490372389554977,
      "num_tokens": 123806050.0,
      "step": 56505
    },
    {
      "entropy": 5.664363861083984,
      "epoch": 5.608376339817388,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002352849648184779,
      "loss": 4.8548,
      "mean_token_accuracy": 0.2030266433954239,
      "num_tokens": 123816779.0,
      "step": 56510
    },
    {
      "entropy": 5.601879501342774,
      "epoch": 5.608872568479556,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023525009013835254,
      "loss": 4.8226,
      "mean_token_accuracy": 0.21159462183713912,
      "num_tokens": 123826835.0,
      "step": 56515
    },
    {
      "entropy": 5.739116764068603,
      "epoch": 5.609368797141723,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023521521644393929,
      "loss": 5.008,
      "mean_token_accuracy": 0.1963891565799713,
      "num_tokens": 123836733.0,
      "step": 56520
    },
    {
      "entropy": 5.609760761260986,
      "epoch": 5.609865025803891,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023518034373610293,
      "loss": 4.8178,
      "mean_token_accuracy": 0.20737264901399613,
      "num_tokens": 123847200.0,
      "step": 56525
    },
    {
      "entropy": 5.713777017593384,
      "epoch": 5.610361254466058,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023514547201570812,
      "loss": 4.9748,
      "mean_token_accuracy": 0.19664786756038666,
      "num_tokens": 123859570.0,
      "step": 56530
    },
    {
      "entropy": 5.741958236694336,
      "epoch": 5.610857483128226,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023511060128361994,
      "loss": 5.0459,
      "mean_token_accuracy": 0.1917707234621048,
      "num_tokens": 123871153.0,
      "step": 56535
    },
    {
      "entropy": 5.768474483489991,
      "epoch": 5.611353711790393,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023507573154070278,
      "loss": 5.0095,
      "mean_token_accuracy": 0.18921102285385133,
      "num_tokens": 123880799.0,
      "step": 56540
    },
    {
      "entropy": 5.712982225418091,
      "epoch": 5.611849940452561,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023504086278782144,
      "loss": 4.9318,
      "mean_token_accuracy": 0.19463254064321517,
      "num_tokens": 123891792.0,
      "step": 56545
    },
    {
      "entropy": 5.725600337982177,
      "epoch": 5.612346169114728,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023500599502584063,
      "loss": 5.021,
      "mean_token_accuracy": 0.1915903151035309,
      "num_tokens": 123903102.0,
      "step": 56550
    },
    {
      "entropy": 5.665148019790649,
      "epoch": 5.6128423977768955,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002349711282556249,
      "loss": 4.8459,
      "mean_token_accuracy": 0.20342055559158326,
      "num_tokens": 123913594.0,
      "step": 56555
    },
    {
      "entropy": 5.672199058532715,
      "epoch": 5.613338626439063,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023493626247803907,
      "loss": 4.9438,
      "mean_token_accuracy": 0.2010377749800682,
      "num_tokens": 123922762.0,
      "step": 56560
    },
    {
      "entropy": 5.724596405029297,
      "epoch": 5.6138348551012305,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023490139769394747,
      "loss": 4.9895,
      "mean_token_accuracy": 0.1957367479801178,
      "num_tokens": 123933731.0,
      "step": 56565
    },
    {
      "entropy": 5.72900424003601,
      "epoch": 5.614331083763398,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023486653390421486,
      "loss": 4.9007,
      "mean_token_accuracy": 0.20293912887573243,
      "num_tokens": 123944671.0,
      "step": 56570
    },
    {
      "entropy": 5.6556250095367435,
      "epoch": 5.614827312425565,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023483167110970572,
      "loss": 4.8796,
      "mean_token_accuracy": 0.2023915797472,
      "num_tokens": 123956426.0,
      "step": 56575
    },
    {
      "entropy": 5.661827087402344,
      "epoch": 5.615323541087733,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002347968093112846,
      "loss": 4.9068,
      "mean_token_accuracy": 0.1911024883389473,
      "num_tokens": 123968918.0,
      "step": 56580
    },
    {
      "entropy": 5.722555780410767,
      "epoch": 5.6158197697499,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023476194850981585,
      "loss": 4.8874,
      "mean_token_accuracy": 0.2085117965936661,
      "num_tokens": 123979812.0,
      "step": 56585
    },
    {
      "entropy": 5.669761276245117,
      "epoch": 5.616315998412068,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023472708870616427,
      "loss": 4.9517,
      "mean_token_accuracy": 0.201580910384655,
      "num_tokens": 123991153.0,
      "step": 56590
    },
    {
      "entropy": 5.614414548873901,
      "epoch": 5.616812227074236,
      "grad_norm": 1.1875,
      "learning_rate": 0.000234692229901194,
      "loss": 4.9429,
      "mean_token_accuracy": 0.19482723623514175,
      "num_tokens": 124002915.0,
      "step": 56595
    },
    {
      "entropy": 5.64342622756958,
      "epoch": 5.617308455736404,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002346573720957696,
      "loss": 4.9186,
      "mean_token_accuracy": 0.20164485424757003,
      "num_tokens": 124015361.0,
      "step": 56600
    },
    {
      "entropy": 5.652750062942505,
      "epoch": 5.617804684398571,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023462251529075546,
      "loss": 4.8704,
      "mean_token_accuracy": 0.20732984691858292,
      "num_tokens": 124025114.0,
      "step": 56605
    },
    {
      "entropy": 5.660867309570312,
      "epoch": 5.618300913060739,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023458765948701588,
      "loss": 4.9063,
      "mean_token_accuracy": 0.19751447290182114,
      "num_tokens": 124037208.0,
      "step": 56610
    },
    {
      "entropy": 5.730089998245239,
      "epoch": 5.618797141722906,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023455280468541524,
      "loss": 4.9507,
      "mean_token_accuracy": 0.19589560776948928,
      "num_tokens": 124048649.0,
      "step": 56615
    },
    {
      "entropy": 5.649482679367066,
      "epoch": 5.619293370385074,
      "grad_norm": 1.34375,
      "learning_rate": 0.000234517950886818,
      "loss": 4.8778,
      "mean_token_accuracy": 0.19775156527757645,
      "num_tokens": 124059768.0,
      "step": 56620
    },
    {
      "entropy": 5.654732894897461,
      "epoch": 5.619789599047241,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023448309809208834,
      "loss": 4.9226,
      "mean_token_accuracy": 0.19712636917829512,
      "num_tokens": 124072402.0,
      "step": 56625
    },
    {
      "entropy": 5.762871599197387,
      "epoch": 5.6202858277094085,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023444824630209043,
      "loss": 5.0933,
      "mean_token_accuracy": 0.18844958543777465,
      "num_tokens": 124084807.0,
      "step": 56630
    },
    {
      "entropy": 5.669821882247925,
      "epoch": 5.620782056371576,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00023441339551768874,
      "loss": 4.8834,
      "mean_token_accuracy": 0.2012685179710388,
      "num_tokens": 124096693.0,
      "step": 56635
    },
    {
      "entropy": 5.6258056640625,
      "epoch": 5.6212782850337435,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023437854573974732,
      "loss": 4.8551,
      "mean_token_accuracy": 0.20740684270858764,
      "num_tokens": 124107704.0,
      "step": 56640
    },
    {
      "entropy": 5.678643703460693,
      "epoch": 5.621774513695911,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002343436969691304,
      "loss": 4.9665,
      "mean_token_accuracy": 0.20099736452102662,
      "num_tokens": 124117746.0,
      "step": 56645
    },
    {
      "entropy": 5.643447160720825,
      "epoch": 5.622270742358078,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002343088492067022,
      "loss": 4.8991,
      "mean_token_accuracy": 0.1957450643181801,
      "num_tokens": 124128341.0,
      "step": 56650
    },
    {
      "entropy": 5.687854146957397,
      "epoch": 5.622766971020246,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023427400245332698,
      "loss": 4.9113,
      "mean_token_accuracy": 0.19928451180458068,
      "num_tokens": 124139244.0,
      "step": 56655
    },
    {
      "entropy": 5.714126539230347,
      "epoch": 5.623263199682413,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023423915670986862,
      "loss": 4.9887,
      "mean_token_accuracy": 0.1955009400844574,
      "num_tokens": 124149882.0,
      "step": 56660
    },
    {
      "entropy": 5.682484102249146,
      "epoch": 5.623759428344581,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00023420431197719143,
      "loss": 4.9107,
      "mean_token_accuracy": 0.20965574383735658,
      "num_tokens": 124160467.0,
      "step": 56665
    },
    {
      "entropy": 5.702582311630249,
      "epoch": 5.624255657006748,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023416946825615938,
      "loss": 4.9279,
      "mean_token_accuracy": 0.19957636296749115,
      "num_tokens": 124171046.0,
      "step": 56670
    },
    {
      "entropy": 5.703410530090332,
      "epoch": 5.624751885668916,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023413462554763641,
      "loss": 4.9504,
      "mean_token_accuracy": 0.19666056483983993,
      "num_tokens": 124183433.0,
      "step": 56675
    },
    {
      "entropy": 5.725261640548706,
      "epoch": 5.625248114331084,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023409978385248675,
      "loss": 5.008,
      "mean_token_accuracy": 0.19671859890222548,
      "num_tokens": 124194190.0,
      "step": 56680
    },
    {
      "entropy": 5.694408988952636,
      "epoch": 5.625744342993252,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023406494317157434,
      "loss": 4.9158,
      "mean_token_accuracy": 0.1905588462948799,
      "num_tokens": 124204994.0,
      "step": 56685
    },
    {
      "entropy": 5.715408086776733,
      "epoch": 5.626240571655419,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002340301035057631,
      "loss": 4.9502,
      "mean_token_accuracy": 0.20029383152723312,
      "num_tokens": 124216997.0,
      "step": 56690
    },
    {
      "entropy": 5.696012544631958,
      "epoch": 5.626736800317587,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023399526485591704,
      "loss": 4.9239,
      "mean_token_accuracy": 0.19062720090150834,
      "num_tokens": 124226718.0,
      "step": 56695
    },
    {
      "entropy": 5.759905672073364,
      "epoch": 5.627233028979754,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002339604272229001,
      "loss": 4.9281,
      "mean_token_accuracy": 0.1991986259818077,
      "num_tokens": 124236702.0,
      "step": 56700
    },
    {
      "entropy": 5.690348815917969,
      "epoch": 5.627729257641922,
      "grad_norm": 1.25,
      "learning_rate": 0.000233925590607576,
      "loss": 4.9771,
      "mean_token_accuracy": 0.19194547981023788,
      "num_tokens": 124249451.0,
      "step": 56705
    },
    {
      "entropy": 5.630345726013184,
      "epoch": 5.628225486304089,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002338907550108088,
      "loss": 4.8457,
      "mean_token_accuracy": 0.20746039748191833,
      "num_tokens": 124260278.0,
      "step": 56710
    },
    {
      "entropy": 5.692892551422119,
      "epoch": 5.6287217149662565,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023385592043346237,
      "loss": 4.9239,
      "mean_token_accuracy": 0.1961652785539627,
      "num_tokens": 124271094.0,
      "step": 56715
    },
    {
      "entropy": 5.632270336151123,
      "epoch": 5.629217943628424,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023382108687640047,
      "loss": 4.8799,
      "mean_token_accuracy": 0.20082227438688277,
      "num_tokens": 124281635.0,
      "step": 56720
    },
    {
      "entropy": 5.674724054336548,
      "epoch": 5.6297141722905915,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023378625434048678,
      "loss": 4.908,
      "mean_token_accuracy": 0.20886597782373428,
      "num_tokens": 124291401.0,
      "step": 56725
    },
    {
      "entropy": 5.602771520614624,
      "epoch": 5.630210400952759,
      "grad_norm": 1.375,
      "learning_rate": 0.00023375142282658525,
      "loss": 4.8215,
      "mean_token_accuracy": 0.2026708498597145,
      "num_tokens": 124301325.0,
      "step": 56730
    },
    {
      "entropy": 5.672991371154785,
      "epoch": 5.630706629614926,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023371659233555942,
      "loss": 4.881,
      "mean_token_accuracy": 0.2039599448442459,
      "num_tokens": 124311268.0,
      "step": 56735
    },
    {
      "entropy": 5.7466095924377445,
      "epoch": 5.631202858277094,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023368176286827326,
      "loss": 4.9827,
      "mean_token_accuracy": 0.19878959357738496,
      "num_tokens": 124324654.0,
      "step": 56740
    },
    {
      "entropy": 5.739438486099243,
      "epoch": 5.631699086939261,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023364693442559032,
      "loss": 4.9727,
      "mean_token_accuracy": 0.1940596356987953,
      "num_tokens": 124335639.0,
      "step": 56745
    },
    {
      "entropy": 5.703278541564941,
      "epoch": 5.632195315601429,
      "grad_norm": 1.5,
      "learning_rate": 0.00023361210700837431,
      "loss": 4.9449,
      "mean_token_accuracy": 0.19556598365306854,
      "num_tokens": 124346040.0,
      "step": 56750
    },
    {
      "entropy": 5.673750638961792,
      "epoch": 5.632691544263596,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023357728061748883,
      "loss": 4.847,
      "mean_token_accuracy": 0.21308604776859283,
      "num_tokens": 124356238.0,
      "step": 56755
    },
    {
      "entropy": 5.7129374027252195,
      "epoch": 5.633187772925764,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023354245525379757,
      "loss": 4.9526,
      "mean_token_accuracy": 0.1991087257862091,
      "num_tokens": 124368417.0,
      "step": 56760
    },
    {
      "entropy": 5.626130723953247,
      "epoch": 5.633684001587932,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023350763091816403,
      "loss": 4.8512,
      "mean_token_accuracy": 0.2044522926211357,
      "num_tokens": 124379090.0,
      "step": 56765
    },
    {
      "entropy": 5.718975591659546,
      "epoch": 5.6341802302501,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002334728076114519,
      "loss": 5.0248,
      "mean_token_accuracy": 0.19687004536390304,
      "num_tokens": 124389624.0,
      "step": 56770
    },
    {
      "entropy": 5.574037647247314,
      "epoch": 5.634676458912267,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023343798533452466,
      "loss": 4.8679,
      "mean_token_accuracy": 0.21246739178895951,
      "num_tokens": 124401486.0,
      "step": 56775
    },
    {
      "entropy": 5.71654314994812,
      "epoch": 5.635172687574435,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002334031640882458,
      "loss": 4.9726,
      "mean_token_accuracy": 0.2031576603651047,
      "num_tokens": 124411909.0,
      "step": 56780
    },
    {
      "entropy": 5.69323091506958,
      "epoch": 5.635668916236602,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002333683438734789,
      "loss": 4.8448,
      "mean_token_accuracy": 0.20674598962068558,
      "num_tokens": 124422968.0,
      "step": 56785
    },
    {
      "entropy": 5.633265161514283,
      "epoch": 5.6361651448987695,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023333352469108722,
      "loss": 4.8144,
      "mean_token_accuracy": 0.20548320561647415,
      "num_tokens": 124433323.0,
      "step": 56790
    },
    {
      "entropy": 5.720083379745484,
      "epoch": 5.636661373560937,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023329870654193436,
      "loss": 4.8809,
      "mean_token_accuracy": 0.19444368183612823,
      "num_tokens": 124444101.0,
      "step": 56795
    },
    {
      "entropy": 5.653717517852783,
      "epoch": 5.6371576022231045,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002332638894268838,
      "loss": 4.8327,
      "mean_token_accuracy": 0.208624367415905,
      "num_tokens": 124455611.0,
      "step": 56800
    },
    {
      "entropy": 5.692579698562622,
      "epoch": 5.637653830885272,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023322907334679882,
      "loss": 4.9075,
      "mean_token_accuracy": 0.20918919891119003,
      "num_tokens": 124467101.0,
      "step": 56805
    },
    {
      "entropy": 5.750669670104981,
      "epoch": 5.638150059547439,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023319425830254287,
      "loss": 4.9993,
      "mean_token_accuracy": 0.19273178279399872,
      "num_tokens": 124477274.0,
      "step": 56810
    },
    {
      "entropy": 5.602946853637695,
      "epoch": 5.638646288209607,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023315944429497926,
      "loss": 4.829,
      "mean_token_accuracy": 0.20700922161340712,
      "num_tokens": 124488005.0,
      "step": 56815
    },
    {
      "entropy": 5.658151721954345,
      "epoch": 5.639142516871774,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002331246313249711,
      "loss": 4.9122,
      "mean_token_accuracy": 0.19960331320762634,
      "num_tokens": 124498112.0,
      "step": 56820
    },
    {
      "entropy": 5.70918755531311,
      "epoch": 5.639638745533942,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002330898193933819,
      "loss": 4.9179,
      "mean_token_accuracy": 0.1998281419277191,
      "num_tokens": 124509439.0,
      "step": 56825
    },
    {
      "entropy": 5.625588750839233,
      "epoch": 5.640134974196109,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023305500850107498,
      "loss": 4.9292,
      "mean_token_accuracy": 0.19681322276592256,
      "num_tokens": 124520438.0,
      "step": 56830
    },
    {
      "entropy": 5.662208604812622,
      "epoch": 5.640631202858277,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023302019864891344,
      "loss": 4.9003,
      "mean_token_accuracy": 0.19977038353681564,
      "num_tokens": 124530268.0,
      "step": 56835
    },
    {
      "entropy": 5.669383859634399,
      "epoch": 5.641127431520444,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023298538983776047,
      "loss": 4.9179,
      "mean_token_accuracy": 0.200025674700737,
      "num_tokens": 124541753.0,
      "step": 56840
    },
    {
      "entropy": 5.694208335876465,
      "epoch": 5.641623660182612,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023295058206847935,
      "loss": 4.989,
      "mean_token_accuracy": 0.19368467181921006,
      "num_tokens": 124552514.0,
      "step": 56845
    },
    {
      "entropy": 5.67652907371521,
      "epoch": 5.64211988884478,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002329157753419331,
      "loss": 4.8534,
      "mean_token_accuracy": 0.205891714990139,
      "num_tokens": 124564722.0,
      "step": 56850
    },
    {
      "entropy": 5.662547397613525,
      "epoch": 5.642616117506948,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023288096965898503,
      "loss": 4.9046,
      "mean_token_accuracy": 0.20353569239377975,
      "num_tokens": 124576707.0,
      "step": 56855
    },
    {
      "entropy": 5.731645011901856,
      "epoch": 5.643112346169115,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023284616502049804,
      "loss": 4.9449,
      "mean_token_accuracy": 0.19801771640777588,
      "num_tokens": 124586765.0,
      "step": 56860
    },
    {
      "entropy": 5.741745042800903,
      "epoch": 5.6436085748312825,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00023281136142733544,
      "loss": 5.0069,
      "mean_token_accuracy": 0.2001892790198326,
      "num_tokens": 124597805.0,
      "step": 56865
    },
    {
      "entropy": 5.691254901885986,
      "epoch": 5.64410480349345,
      "grad_norm": 1.25,
      "learning_rate": 0.0002327765588803601,
      "loss": 4.8742,
      "mean_token_accuracy": 0.20221575498580932,
      "num_tokens": 124608631.0,
      "step": 56870
    },
    {
      "entropy": 5.6646044731140135,
      "epoch": 5.6446010321556175,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023274175738043518,
      "loss": 4.9587,
      "mean_token_accuracy": 0.19511809796094895,
      "num_tokens": 124620508.0,
      "step": 56875
    },
    {
      "entropy": 5.721599578857422,
      "epoch": 5.645097260817785,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00023270695692842352,
      "loss": 4.9724,
      "mean_token_accuracy": 0.19316985607147216,
      "num_tokens": 124630555.0,
      "step": 56880
    },
    {
      "entropy": 5.700437450408936,
      "epoch": 5.645593489479952,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023267215752518827,
      "loss": 4.9018,
      "mean_token_accuracy": 0.20512827336788178,
      "num_tokens": 124641970.0,
      "step": 56885
    },
    {
      "entropy": 5.766163873672485,
      "epoch": 5.64608971814212,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023263735917159224,
      "loss": 5.0289,
      "mean_token_accuracy": 0.18588302582502364,
      "num_tokens": 124652420.0,
      "step": 56890
    },
    {
      "entropy": 5.686111068725586,
      "epoch": 5.646585946804287,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002326025618684985,
      "loss": 4.9,
      "mean_token_accuracy": 0.1999327212572098,
      "num_tokens": 124663226.0,
      "step": 56895
    },
    {
      "entropy": 5.678968811035157,
      "epoch": 5.647082175466455,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002325677656167699,
      "loss": 4.9045,
      "mean_token_accuracy": 0.204301880300045,
      "num_tokens": 124674587.0,
      "step": 56900
    },
    {
      "entropy": 5.716163539886475,
      "epoch": 5.647578404128622,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023253297041726917,
      "loss": 4.9406,
      "mean_token_accuracy": 0.1944812908768654,
      "num_tokens": 124685965.0,
      "step": 56905
    },
    {
      "entropy": 5.657433748245239,
      "epoch": 5.64807463279079,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023249817627085922,
      "loss": 4.936,
      "mean_token_accuracy": 0.19724081009626387,
      "num_tokens": 124697846.0,
      "step": 56910
    },
    {
      "entropy": 5.6795097351074215,
      "epoch": 5.648570861452957,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023246338317840304,
      "loss": 4.924,
      "mean_token_accuracy": 0.20477309972047805,
      "num_tokens": 124708216.0,
      "step": 56915
    },
    {
      "entropy": 5.7323305130004885,
      "epoch": 5.649067090115125,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023242859114076326,
      "loss": 4.9773,
      "mean_token_accuracy": 0.19418517798185347,
      "num_tokens": 124719479.0,
      "step": 56920
    },
    {
      "entropy": 5.729579830169678,
      "epoch": 5.649563318777292,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023239380015880272,
      "loss": 4.9494,
      "mean_token_accuracy": 0.1906673640012741,
      "num_tokens": 124731572.0,
      "step": 56925
    },
    {
      "entropy": 5.7799583911895756,
      "epoch": 5.65005954743946,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023235901023338412,
      "loss": 5.0314,
      "mean_token_accuracy": 0.1866094723343849,
      "num_tokens": 124742264.0,
      "step": 56930
    },
    {
      "entropy": 5.715777921676636,
      "epoch": 5.650555776101628,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023232422136537013,
      "loss": 4.8861,
      "mean_token_accuracy": 0.19763533174991607,
      "num_tokens": 124753229.0,
      "step": 56935
    },
    {
      "entropy": 5.680811500549316,
      "epoch": 5.6510520047637955,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023228943355562342,
      "loss": 4.925,
      "mean_token_accuracy": 0.19893640130758286,
      "num_tokens": 124763318.0,
      "step": 56940
    },
    {
      "entropy": 5.704831314086914,
      "epoch": 5.651548233425963,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002322546468050069,
      "loss": 5.0096,
      "mean_token_accuracy": 0.19274074584245682,
      "num_tokens": 124774330.0,
      "step": 56945
    },
    {
      "entropy": 5.690048265457153,
      "epoch": 5.6520444620881305,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023221986111438299,
      "loss": 4.9561,
      "mean_token_accuracy": 0.1864188626408577,
      "num_tokens": 124784217.0,
      "step": 56950
    },
    {
      "entropy": 5.743573904037476,
      "epoch": 5.652540690750298,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023218507648461428,
      "loss": 4.98,
      "mean_token_accuracy": 0.19309429824352264,
      "num_tokens": 124794947.0,
      "step": 56955
    },
    {
      "entropy": 5.715927362442017,
      "epoch": 5.6530369194124654,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002321502929165635,
      "loss": 4.9334,
      "mean_token_accuracy": 0.19777640998363494,
      "num_tokens": 124805687.0,
      "step": 56960
    },
    {
      "entropy": 5.7127806663513185,
      "epoch": 5.653533148074633,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023211551041109303,
      "loss": 4.9643,
      "mean_token_accuracy": 0.1908029168844223,
      "num_tokens": 124816403.0,
      "step": 56965
    },
    {
      "entropy": 5.686462354660034,
      "epoch": 5.6540293767368,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023208072896906552,
      "loss": 4.9223,
      "mean_token_accuracy": 0.19446319937705994,
      "num_tokens": 124825880.0,
      "step": 56970
    },
    {
      "entropy": 5.629062080383301,
      "epoch": 5.654525605398968,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023204594859134342,
      "loss": 4.8467,
      "mean_token_accuracy": 0.1986384406685829,
      "num_tokens": 124837426.0,
      "step": 56975
    },
    {
      "entropy": 5.712105369567871,
      "epoch": 5.655021834061135,
      "grad_norm": 1.25,
      "learning_rate": 0.0002320111692787893,
      "loss": 4.9446,
      "mean_token_accuracy": 0.19689713269472123,
      "num_tokens": 124848250.0,
      "step": 56980
    },
    {
      "entropy": 5.737940835952759,
      "epoch": 5.655518062723303,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023197639103226554,
      "loss": 4.901,
      "mean_token_accuracy": 0.19223205149173736,
      "num_tokens": 124858577.0,
      "step": 56985
    },
    {
      "entropy": 5.738471746444702,
      "epoch": 5.65601429138547,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023194161385263457,
      "loss": 5.038,
      "mean_token_accuracy": 0.18625796288251878,
      "num_tokens": 124871175.0,
      "step": 56990
    },
    {
      "entropy": 5.673726272583008,
      "epoch": 5.656510520047638,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023190683774075883,
      "loss": 4.9702,
      "mean_token_accuracy": 0.19206226617097855,
      "num_tokens": 124881161.0,
      "step": 56995
    },
    {
      "entropy": 5.659063053131104,
      "epoch": 5.657006748709805,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002318720626975005,
      "loss": 4.8956,
      "mean_token_accuracy": 0.2049674302339554,
      "num_tokens": 124890982.0,
      "step": 57000
    },
    {
      "epoch": 5.657006748709805,
      "eval_entropy": 5.46751861579063,
      "eval_loss": 5.076785087585449,
      "eval_mean_token_accuracy": 0.19565201921569506,
      "eval_num_tokens": 124890982.0,
      "eval_runtime": 22.1296,
      "eval_samples_per_second": 1521.445,
      "eval_steps_per_second": 190.198,
      "step": 57000
    },
    {
      "entropy": 5.76356611251831,
      "epoch": 5.657502977371973,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023183728872372216,
      "loss": 4.9601,
      "mean_token_accuracy": 0.19077201187610626,
      "num_tokens": 124902102.0,
      "step": 57005
    },
    {
      "entropy": 5.733966827392578,
      "epoch": 5.65799920603414,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023180251582028617,
      "loss": 4.9333,
      "mean_token_accuracy": 0.20011114776134492,
      "num_tokens": 124913930.0,
      "step": 57010
    },
    {
      "entropy": 5.716236209869384,
      "epoch": 5.658495434696308,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002317677439880546,
      "loss": 4.9704,
      "mean_token_accuracy": 0.19733455628156663,
      "num_tokens": 124923990.0,
      "step": 57015
    },
    {
      "entropy": 5.718227672576904,
      "epoch": 5.658991663358476,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023173297322788984,
      "loss": 4.9085,
      "mean_token_accuracy": 0.1984020873904228,
      "num_tokens": 124934398.0,
      "step": 57020
    },
    {
      "entropy": 5.636017513275147,
      "epoch": 5.659487892020643,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023169820354065418,
      "loss": 4.8645,
      "mean_token_accuracy": 0.20315095633268357,
      "num_tokens": 124945167.0,
      "step": 57025
    },
    {
      "entropy": 5.661261701583863,
      "epoch": 5.659984120682811,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002316634349272096,
      "loss": 4.8722,
      "mean_token_accuracy": 0.20002501159906388,
      "num_tokens": 124955657.0,
      "step": 57030
    },
    {
      "entropy": 5.6237428188323975,
      "epoch": 5.6604803493449785,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023162866738841854,
      "loss": 4.8412,
      "mean_token_accuracy": 0.2064211905002594,
      "num_tokens": 124966831.0,
      "step": 57035
    },
    {
      "entropy": 5.691594457626342,
      "epoch": 5.660976578007146,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002315939009251431,
      "loss": 4.9324,
      "mean_token_accuracy": 0.20120691806077956,
      "num_tokens": 124976649.0,
      "step": 57040
    },
    {
      "entropy": 5.708933115005493,
      "epoch": 5.661472806669313,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023155913553824542,
      "loss": 4.9376,
      "mean_token_accuracy": 0.19474498629570008,
      "num_tokens": 124987193.0,
      "step": 57045
    },
    {
      "entropy": 5.701287794113159,
      "epoch": 5.661969035331481,
      "grad_norm": 1.25,
      "learning_rate": 0.0002315243712285875,
      "loss": 4.8968,
      "mean_token_accuracy": 0.198533096909523,
      "num_tokens": 124998695.0,
      "step": 57050
    },
    {
      "entropy": 5.69123911857605,
      "epoch": 5.662465263993648,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023148960799703144,
      "loss": 4.955,
      "mean_token_accuracy": 0.19972474426031112,
      "num_tokens": 125008772.0,
      "step": 57055
    },
    {
      "entropy": 5.652381706237793,
      "epoch": 5.662961492655816,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023145484584443944,
      "loss": 4.8691,
      "mean_token_accuracy": 0.20792421251535415,
      "num_tokens": 125018466.0,
      "step": 57060
    },
    {
      "entropy": 5.7171875,
      "epoch": 5.663457721317983,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023142008477167344,
      "loss": 4.9711,
      "mean_token_accuracy": 0.19715568274259568,
      "num_tokens": 125030119.0,
      "step": 57065
    },
    {
      "entropy": 5.634086227416992,
      "epoch": 5.663953949980151,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023138532477959538,
      "loss": 4.868,
      "mean_token_accuracy": 0.2010190635919571,
      "num_tokens": 125041337.0,
      "step": 57070
    },
    {
      "entropy": 5.746104955673218,
      "epoch": 5.664450178642318,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023135056586906733,
      "loss": 4.9371,
      "mean_token_accuracy": 0.20215841382741928,
      "num_tokens": 125051151.0,
      "step": 57075
    },
    {
      "entropy": 5.686360883712768,
      "epoch": 5.664946407304486,
      "grad_norm": 1.375,
      "learning_rate": 0.00023131580804095116,
      "loss": 4.8813,
      "mean_token_accuracy": 0.20245432406663894,
      "num_tokens": 125062198.0,
      "step": 57080
    },
    {
      "entropy": 5.660523891448975,
      "epoch": 5.665442635966653,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023128105129610876,
      "loss": 4.9173,
      "mean_token_accuracy": 0.19745589792728424,
      "num_tokens": 125073465.0,
      "step": 57085
    },
    {
      "entropy": 5.748007583618164,
      "epoch": 5.665938864628821,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023124629563540217,
      "loss": 5.0468,
      "mean_token_accuracy": 0.18812094926834105,
      "num_tokens": 125086839.0,
      "step": 57090
    },
    {
      "entropy": 5.7361767292022705,
      "epoch": 5.666435093290988,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023121154105969316,
      "loss": 4.9451,
      "mean_token_accuracy": 0.19357330948114396,
      "num_tokens": 125096893.0,
      "step": 57095
    },
    {
      "entropy": 5.763631057739258,
      "epoch": 5.666931321953156,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023117678756984355,
      "loss": 4.9846,
      "mean_token_accuracy": 0.19836968183517456,
      "num_tokens": 125108098.0,
      "step": 57100
    },
    {
      "entropy": 5.647469186782837,
      "epoch": 5.667427550615324,
      "grad_norm": 1.375,
      "learning_rate": 0.00023114203516671523,
      "loss": 4.874,
      "mean_token_accuracy": 0.19780821353197098,
      "num_tokens": 125118722.0,
      "step": 57105
    },
    {
      "entropy": 5.68454270362854,
      "epoch": 5.667923779277491,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023110728385116993,
      "loss": 4.9941,
      "mean_token_accuracy": 0.1887642428278923,
      "num_tokens": 125130066.0,
      "step": 57110
    },
    {
      "entropy": 5.804857349395752,
      "epoch": 5.668420007939659,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023107253362406928,
      "loss": 5.0506,
      "mean_token_accuracy": 0.19284340292215346,
      "num_tokens": 125140692.0,
      "step": 57115
    },
    {
      "entropy": 5.730945825576782,
      "epoch": 5.668916236601826,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023103778448627522,
      "loss": 4.9429,
      "mean_token_accuracy": 0.19680907279253007,
      "num_tokens": 125150610.0,
      "step": 57120
    },
    {
      "entropy": 5.7081602096557615,
      "epoch": 5.669412465263994,
      "grad_norm": 1.375,
      "learning_rate": 0.0002310030364386494,
      "loss": 4.9553,
      "mean_token_accuracy": 0.19933207035064698,
      "num_tokens": 125161017.0,
      "step": 57125
    },
    {
      "entropy": 5.689759922027588,
      "epoch": 5.669908693926161,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023096828948205344,
      "loss": 4.9936,
      "mean_token_accuracy": 0.1916545510292053,
      "num_tokens": 125171457.0,
      "step": 57130
    },
    {
      "entropy": 5.69994707107544,
      "epoch": 5.670404922588329,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023093354361734903,
      "loss": 4.9364,
      "mean_token_accuracy": 0.19831825345754622,
      "num_tokens": 125182158.0,
      "step": 57135
    },
    {
      "entropy": 5.73980860710144,
      "epoch": 5.670901151250496,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023089879884539781,
      "loss": 4.9864,
      "mean_token_accuracy": 0.19098710119724274,
      "num_tokens": 125192185.0,
      "step": 57140
    },
    {
      "entropy": 5.678405237197876,
      "epoch": 5.671397379912664,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023086405516706133,
      "loss": 4.8843,
      "mean_token_accuracy": 0.19481881111860275,
      "num_tokens": 125203430.0,
      "step": 57145
    },
    {
      "entropy": 5.752392530441284,
      "epoch": 5.671893608574831,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002308293125832011,
      "loss": 5.0183,
      "mean_token_accuracy": 0.19282903671264648,
      "num_tokens": 125213406.0,
      "step": 57150
    },
    {
      "entropy": 5.685533285140991,
      "epoch": 5.672389837236999,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023079457109467888,
      "loss": 4.9578,
      "mean_token_accuracy": 0.19545693695545197,
      "num_tokens": 125223633.0,
      "step": 57155
    },
    {
      "entropy": 5.693536758422852,
      "epoch": 5.672886065899166,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000230759830702356,
      "loss": 4.943,
      "mean_token_accuracy": 0.19911770820617675,
      "num_tokens": 125234482.0,
      "step": 57160
    },
    {
      "entropy": 5.70310230255127,
      "epoch": 5.673382294561334,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023072509140709395,
      "loss": 4.8423,
      "mean_token_accuracy": 0.20717442482709886,
      "num_tokens": 125245800.0,
      "step": 57165
    },
    {
      "entropy": 5.706425952911377,
      "epoch": 5.673878523223501,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023069035320975428,
      "loss": 4.8448,
      "mean_token_accuracy": 0.20623779296875,
      "num_tokens": 125256566.0,
      "step": 57170
    },
    {
      "entropy": 5.601538467407226,
      "epoch": 5.674374751885669,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023065561611119824,
      "loss": 4.845,
      "mean_token_accuracy": 0.2032925844192505,
      "num_tokens": 125267528.0,
      "step": 57175
    },
    {
      "entropy": 5.670350217819214,
      "epoch": 5.674870980547836,
      "grad_norm": 1.25,
      "learning_rate": 0.00023062088011228748,
      "loss": 5.007,
      "mean_token_accuracy": 0.1960061103105545,
      "num_tokens": 125278348.0,
      "step": 57180
    },
    {
      "entropy": 5.659159851074219,
      "epoch": 5.675367209210004,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023058614521388328,
      "loss": 4.9272,
      "mean_token_accuracy": 0.20029531568288803,
      "num_tokens": 125289931.0,
      "step": 57185
    },
    {
      "entropy": 5.7794280529022215,
      "epoch": 5.675863437872172,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023055141141684694,
      "loss": 5.0574,
      "mean_token_accuracy": 0.187374509871006,
      "num_tokens": 125302658.0,
      "step": 57190
    },
    {
      "entropy": 5.765229654312134,
      "epoch": 5.6763596665343385,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023051667872203986,
      "loss": 4.9858,
      "mean_token_accuracy": 0.19420638382434846,
      "num_tokens": 125314124.0,
      "step": 57195
    },
    {
      "entropy": 5.727114915847778,
      "epoch": 5.676855895196507,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023048194713032322,
      "loss": 4.8991,
      "mean_token_accuracy": 0.20287790894508362,
      "num_tokens": 125325027.0,
      "step": 57200
    },
    {
      "entropy": 5.739845228195191,
      "epoch": 5.677352123858674,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023044721664255836,
      "loss": 4.9215,
      "mean_token_accuracy": 0.19760802686214446,
      "num_tokens": 125334950.0,
      "step": 57205
    },
    {
      "entropy": 5.6872340679168705,
      "epoch": 5.677848352520842,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023041248725960657,
      "loss": 4.9638,
      "mean_token_accuracy": 0.19379936009645463,
      "num_tokens": 125345796.0,
      "step": 57210
    },
    {
      "entropy": 5.662620210647583,
      "epoch": 5.678344581183009,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023037775898232903,
      "loss": 4.9167,
      "mean_token_accuracy": 0.19847642183303832,
      "num_tokens": 125356408.0,
      "step": 57215
    },
    {
      "entropy": 5.663637399673462,
      "epoch": 5.678840809845177,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023034303181158688,
      "loss": 4.918,
      "mean_token_accuracy": 0.20064503699541092,
      "num_tokens": 125368705.0,
      "step": 57220
    },
    {
      "entropy": 5.706588983535767,
      "epoch": 5.679337038507344,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002303083057482414,
      "loss": 4.8887,
      "mean_token_accuracy": 0.20977792739868165,
      "num_tokens": 125380059.0,
      "step": 57225
    },
    {
      "entropy": 5.720794248580932,
      "epoch": 5.679833267169512,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023027358079315358,
      "loss": 4.9437,
      "mean_token_accuracy": 0.1989106073975563,
      "num_tokens": 125391156.0,
      "step": 57230
    },
    {
      "entropy": 5.7840698719024655,
      "epoch": 5.680329495831679,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002302388569471845,
      "loss": 5.0091,
      "mean_token_accuracy": 0.19616378247737884,
      "num_tokens": 125401962.0,
      "step": 57235
    },
    {
      "entropy": 5.6863525867462155,
      "epoch": 5.680825724493847,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023020413421119535,
      "loss": 4.9407,
      "mean_token_accuracy": 0.19795757681131362,
      "num_tokens": 125412807.0,
      "step": 57240
    },
    {
      "entropy": 5.798000955581665,
      "epoch": 5.681321953156014,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023016941258604723,
      "loss": 5.1218,
      "mean_token_accuracy": 0.18445071578025818,
      "num_tokens": 125425746.0,
      "step": 57245
    },
    {
      "entropy": 5.739651536941528,
      "epoch": 5.681818181818182,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023013469207260103,
      "loss": 4.9077,
      "mean_token_accuracy": 0.20165415853261948,
      "num_tokens": 125436982.0,
      "step": 57250
    },
    {
      "entropy": 5.640839004516602,
      "epoch": 5.682314410480349,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002300999726717178,
      "loss": 4.8216,
      "mean_token_accuracy": 0.21151793599128724,
      "num_tokens": 125449315.0,
      "step": 57255
    },
    {
      "entropy": 5.724557495117187,
      "epoch": 5.682810639142517,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023006525438425845,
      "loss": 4.9286,
      "mean_token_accuracy": 0.19684922993183135,
      "num_tokens": 125459514.0,
      "step": 57260
    },
    {
      "entropy": 5.724644470214844,
      "epoch": 5.683306867804684,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002300305372110839,
      "loss": 5.0295,
      "mean_token_accuracy": 0.19124844074249267,
      "num_tokens": 125470409.0,
      "step": 57265
    },
    {
      "entropy": 5.687508201599121,
      "epoch": 5.683803096466852,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022999582115305528,
      "loss": 4.9475,
      "mean_token_accuracy": 0.1950979486107826,
      "num_tokens": 125482621.0,
      "step": 57270
    },
    {
      "entropy": 5.712009572982788,
      "epoch": 5.68429932512902,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022996110621103334,
      "loss": 4.9444,
      "mean_token_accuracy": 0.1934907928109169,
      "num_tokens": 125494210.0,
      "step": 57275
    },
    {
      "entropy": 5.739159488677979,
      "epoch": 5.6847955537911865,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022992639238587881,
      "loss": 4.9336,
      "mean_token_accuracy": 0.1989642858505249,
      "num_tokens": 125505311.0,
      "step": 57280
    },
    {
      "entropy": 5.6635175228118895,
      "epoch": 5.685291782453355,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002298916796784527,
      "loss": 4.9322,
      "mean_token_accuracy": 0.19123883545398712,
      "num_tokens": 125516343.0,
      "step": 57285
    },
    {
      "entropy": 5.665826320648193,
      "epoch": 5.685788011115522,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002298569680896156,
      "loss": 4.8933,
      "mean_token_accuracy": 0.20477909445762635,
      "num_tokens": 125526651.0,
      "step": 57290
    },
    {
      "entropy": 5.629858160018921,
      "epoch": 5.68628423977769,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002298222576202285,
      "loss": 4.872,
      "mean_token_accuracy": 0.20876021236181258,
      "num_tokens": 125538314.0,
      "step": 57295
    },
    {
      "entropy": 5.7429036617279055,
      "epoch": 5.686780468439857,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022978754827115207,
      "loss": 4.992,
      "mean_token_accuracy": 0.18995350003242492,
      "num_tokens": 125547962.0,
      "step": 57300
    },
    {
      "entropy": 5.722551059722901,
      "epoch": 5.687276697102025,
      "grad_norm": 1.25,
      "learning_rate": 0.00022975284004324705,
      "loss": 4.9259,
      "mean_token_accuracy": 0.19360714107751847,
      "num_tokens": 125558831.0,
      "step": 57305
    },
    {
      "entropy": 5.704102373123169,
      "epoch": 5.687772925764192,
      "grad_norm": 1.125,
      "learning_rate": 0.0002297181329373741,
      "loss": 4.9178,
      "mean_token_accuracy": 0.2028651937842369,
      "num_tokens": 125571358.0,
      "step": 57310
    },
    {
      "entropy": 5.685427236557007,
      "epoch": 5.68826915442636,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022968342695439383,
      "loss": 4.8944,
      "mean_token_accuracy": 0.2011357605457306,
      "num_tokens": 125580945.0,
      "step": 57315
    },
    {
      "entropy": 5.673634386062622,
      "epoch": 5.688765383088527,
      "grad_norm": 1.53125,
      "learning_rate": 0.00022964872209516686,
      "loss": 4.9691,
      "mean_token_accuracy": 0.19571200907230377,
      "num_tokens": 125591411.0,
      "step": 57320
    },
    {
      "entropy": 5.69497652053833,
      "epoch": 5.689261611750695,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022961401836055396,
      "loss": 4.9805,
      "mean_token_accuracy": 0.19120020568370819,
      "num_tokens": 125602774.0,
      "step": 57325
    },
    {
      "entropy": 5.624991846084595,
      "epoch": 5.689757840412862,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022957931575141554,
      "loss": 4.8218,
      "mean_token_accuracy": 0.2051505208015442,
      "num_tokens": 125611946.0,
      "step": 57330
    },
    {
      "entropy": 5.714055490493775,
      "epoch": 5.69025406907503,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022954461426861228,
      "loss": 4.9471,
      "mean_token_accuracy": 0.1929927110671997,
      "num_tokens": 125622499.0,
      "step": 57335
    },
    {
      "entropy": 5.676737546920776,
      "epoch": 5.690750297737197,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022950991391300468,
      "loss": 4.9346,
      "mean_token_accuracy": 0.20317430049180984,
      "num_tokens": 125633192.0,
      "step": 57340
    },
    {
      "entropy": 5.711681890487671,
      "epoch": 5.691246526399365,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022947521468545307,
      "loss": 4.8972,
      "mean_token_accuracy": 0.20154905766248704,
      "num_tokens": 125644447.0,
      "step": 57345
    },
    {
      "entropy": 5.7471459865570065,
      "epoch": 5.691742755061532,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022944051658681803,
      "loss": 4.9125,
      "mean_token_accuracy": 0.20519231259822845,
      "num_tokens": 125655296.0,
      "step": 57350
    },
    {
      "entropy": 5.769979763031006,
      "epoch": 5.6922389837236995,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022940581961796013,
      "loss": 5.0143,
      "mean_token_accuracy": 0.18981271833181382,
      "num_tokens": 125665907.0,
      "step": 57355
    },
    {
      "entropy": 5.690736722946167,
      "epoch": 5.692735212385868,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022937112377973963,
      "loss": 4.9014,
      "mean_token_accuracy": 0.2008117139339447,
      "num_tokens": 125677403.0,
      "step": 57360
    },
    {
      "entropy": 5.750673866271972,
      "epoch": 5.6932314410480345,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022933642907301688,
      "loss": 5.0091,
      "mean_token_accuracy": 0.19569836556911469,
      "num_tokens": 125688030.0,
      "step": 57365
    },
    {
      "entropy": 5.736288785934448,
      "epoch": 5.693727669710203,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022930173549865236,
      "loss": 5.0271,
      "mean_token_accuracy": 0.19255225658416747,
      "num_tokens": 125699013.0,
      "step": 57370
    },
    {
      "entropy": 5.74921498298645,
      "epoch": 5.69422389837237,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022926704305750628,
      "loss": 4.9382,
      "mean_token_accuracy": 0.19459184855222703,
      "num_tokens": 125709533.0,
      "step": 57375
    },
    {
      "entropy": 5.746165800094604,
      "epoch": 5.694720127034538,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022923235175043895,
      "loss": 4.9749,
      "mean_token_accuracy": 0.19020717442035676,
      "num_tokens": 125720287.0,
      "step": 57380
    },
    {
      "entropy": 5.760943508148193,
      "epoch": 5.695216355696705,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022919766157831074,
      "loss": 5.0124,
      "mean_token_accuracy": 0.19429078996181487,
      "num_tokens": 125731653.0,
      "step": 57385
    },
    {
      "entropy": 5.7367900848388675,
      "epoch": 5.695712584358873,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022916297254198183,
      "loss": 4.9662,
      "mean_token_accuracy": 0.19637782275676727,
      "num_tokens": 125742000.0,
      "step": 57390
    },
    {
      "entropy": 5.7061059951782225,
      "epoch": 5.69620881302104,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002291282846423124,
      "loss": 4.9574,
      "mean_token_accuracy": 0.19266824424266815,
      "num_tokens": 125753540.0,
      "step": 57395
    },
    {
      "entropy": 5.727451324462891,
      "epoch": 5.696705041683208,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022909359788016265,
      "loss": 4.9068,
      "mean_token_accuracy": 0.19817356169223785,
      "num_tokens": 125764180.0,
      "step": 57400
    },
    {
      "entropy": 5.711997413635254,
      "epoch": 5.697201270345375,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022905891225639285,
      "loss": 4.9197,
      "mean_token_accuracy": 0.198482508957386,
      "num_tokens": 125775990.0,
      "step": 57405
    },
    {
      "entropy": 5.696449708938599,
      "epoch": 5.697697499007543,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022902422777186288,
      "loss": 4.8604,
      "mean_token_accuracy": 0.2041909545660019,
      "num_tokens": 125786055.0,
      "step": 57410
    },
    {
      "entropy": 5.684448289871216,
      "epoch": 5.69819372766971,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002289895444274331,
      "loss": 4.9319,
      "mean_token_accuracy": 0.19762658029794694,
      "num_tokens": 125797808.0,
      "step": 57415
    },
    {
      "entropy": 5.678409910202026,
      "epoch": 5.698689956331878,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022895486222396344,
      "loss": 4.8961,
      "mean_token_accuracy": 0.20181236714124678,
      "num_tokens": 125807729.0,
      "step": 57420
    },
    {
      "entropy": 5.63761305809021,
      "epoch": 5.699186184994045,
      "grad_norm": 1.375,
      "learning_rate": 0.000228920181162314,
      "loss": 4.8622,
      "mean_token_accuracy": 0.21342753171920775,
      "num_tokens": 125818899.0,
      "step": 57425
    },
    {
      "entropy": 5.696718215942383,
      "epoch": 5.6996824136562125,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022888550124334478,
      "loss": 4.8937,
      "mean_token_accuracy": 0.20426553189754487,
      "num_tokens": 125829792.0,
      "step": 57430
    },
    {
      "entropy": 5.716872453689575,
      "epoch": 5.70017864231838,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022885082246791577,
      "loss": 4.974,
      "mean_token_accuracy": 0.19957654625177385,
      "num_tokens": 125840345.0,
      "step": 57435
    },
    {
      "entropy": 5.740038156509399,
      "epoch": 5.7006748709805475,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022881614483688685,
      "loss": 4.9379,
      "mean_token_accuracy": 0.20322644263505935,
      "num_tokens": 125851824.0,
      "step": 57440
    },
    {
      "entropy": 5.641273832321167,
      "epoch": 5.701171099642716,
      "grad_norm": 1.40625,
      "learning_rate": 0.000228781468351118,
      "loss": 4.8695,
      "mean_token_accuracy": 0.20193453133106232,
      "num_tokens": 125863473.0,
      "step": 57445
    },
    {
      "entropy": 5.668656253814698,
      "epoch": 5.701667328304882,
      "grad_norm": 1.25,
      "learning_rate": 0.00022874679301146923,
      "loss": 4.9263,
      "mean_token_accuracy": 0.19468758255243301,
      "num_tokens": 125876223.0,
      "step": 57450
    },
    {
      "entropy": 5.695182228088379,
      "epoch": 5.702163556967051,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022871211881880034,
      "loss": 4.9238,
      "mean_token_accuracy": 0.19861430823802947,
      "num_tokens": 125887005.0,
      "step": 57455
    },
    {
      "entropy": 5.820603704452514,
      "epoch": 5.702659785629218,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002286774457739711,
      "loss": 5.0723,
      "mean_token_accuracy": 0.19242067337036134,
      "num_tokens": 125898173.0,
      "step": 57460
    },
    {
      "entropy": 5.715051794052124,
      "epoch": 5.703156014291386,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022864277387784143,
      "loss": 4.9453,
      "mean_token_accuracy": 0.1968793660402298,
      "num_tokens": 125909465.0,
      "step": 57465
    },
    {
      "entropy": 5.758805894851685,
      "epoch": 5.703652242953553,
      "grad_norm": 1.328125,
      "learning_rate": 0.000228608103131271,
      "loss": 5.0076,
      "mean_token_accuracy": 0.1978438213467598,
      "num_tokens": 125920289.0,
      "step": 57470
    },
    {
      "entropy": 5.731205129623413,
      "epoch": 5.704148471615721,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022857343353511973,
      "loss": 5.0042,
      "mean_token_accuracy": 0.18842843323945999,
      "num_tokens": 125931052.0,
      "step": 57475
    },
    {
      "entropy": 5.779666185379028,
      "epoch": 5.704644700277888,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022853876509024723,
      "loss": 5.1208,
      "mean_token_accuracy": 0.189353109896183,
      "num_tokens": 125942727.0,
      "step": 57480
    },
    {
      "entropy": 5.717427110671997,
      "epoch": 5.705140928940056,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002285040977975133,
      "loss": 4.9242,
      "mean_token_accuracy": 0.19382166415452956,
      "num_tokens": 125953216.0,
      "step": 57485
    },
    {
      "entropy": 5.741716146469116,
      "epoch": 5.705637157602223,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002284694316577775,
      "loss": 4.9622,
      "mean_token_accuracy": 0.2000552386045456,
      "num_tokens": 125963607.0,
      "step": 57490
    },
    {
      "entropy": 5.71840500831604,
      "epoch": 5.706133386264391,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022843476667189957,
      "loss": 4.9449,
      "mean_token_accuracy": 0.19996065348386766,
      "num_tokens": 125975207.0,
      "step": 57495
    },
    {
      "entropy": 5.637986040115356,
      "epoch": 5.706629614926558,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022840010284073899,
      "loss": 4.7515,
      "mean_token_accuracy": 0.22214930057525634,
      "num_tokens": 125986403.0,
      "step": 57500
    },
    {
      "entropy": 5.765388250350952,
      "epoch": 5.7071258435887255,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022836544016515553,
      "loss": 5.0442,
      "mean_token_accuracy": 0.1865478903055191,
      "num_tokens": 125996693.0,
      "step": 57505
    },
    {
      "entropy": 5.67388277053833,
      "epoch": 5.707622072250893,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022833077864600865,
      "loss": 4.9117,
      "mean_token_accuracy": 0.19648673087358476,
      "num_tokens": 126007229.0,
      "step": 57510
    },
    {
      "entropy": 5.7951390743255615,
      "epoch": 5.7081183009130605,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022829611828415792,
      "loss": 5.0406,
      "mean_token_accuracy": 0.19233272969722748,
      "num_tokens": 126018829.0,
      "step": 57515
    },
    {
      "entropy": 5.675030326843261,
      "epoch": 5.708614529575228,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002282614590804628,
      "loss": 4.8591,
      "mean_token_accuracy": 0.20888956338167192,
      "num_tokens": 126031037.0,
      "step": 57520
    },
    {
      "entropy": 5.638464450836182,
      "epoch": 5.7091107582373954,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022822680103578276,
      "loss": 4.8608,
      "mean_token_accuracy": 0.20804893225431442,
      "num_tokens": 126041485.0,
      "step": 57525
    },
    {
      "entropy": 5.715307664871216,
      "epoch": 5.709606986899564,
      "grad_norm": 1.1875,
      "learning_rate": 0.00022819214415097723,
      "loss": 4.9057,
      "mean_token_accuracy": 0.2066265895962715,
      "num_tokens": 126052899.0,
      "step": 57530
    },
    {
      "entropy": 5.660547971725464,
      "epoch": 5.71010321556173,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022815748842690576,
      "loss": 4.8998,
      "mean_token_accuracy": 0.20043423026800156,
      "num_tokens": 126063027.0,
      "step": 57535
    },
    {
      "entropy": 5.679384756088257,
      "epoch": 5.710599444223899,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002281228338644276,
      "loss": 4.9344,
      "mean_token_accuracy": 0.19367158710956572,
      "num_tokens": 126074149.0,
      "step": 57540
    },
    {
      "entropy": 5.695181798934937,
      "epoch": 5.711095672886066,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022808818046440212,
      "loss": 4.9385,
      "mean_token_accuracy": 0.19584265798330308,
      "num_tokens": 126084761.0,
      "step": 57545
    },
    {
      "entropy": 5.745791387557984,
      "epoch": 5.711591901548234,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022805352822768876,
      "loss": 4.9949,
      "mean_token_accuracy": 0.19454022198915483,
      "num_tokens": 126096030.0,
      "step": 57550
    },
    {
      "entropy": 5.701861381530762,
      "epoch": 5.712088130210401,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022801887715514658,
      "loss": 4.9777,
      "mean_token_accuracy": 0.1940966099500656,
      "num_tokens": 126106960.0,
      "step": 57555
    },
    {
      "entropy": 5.7067547798156735,
      "epoch": 5.712584358872569,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022798422724763513,
      "loss": 4.935,
      "mean_token_accuracy": 0.19602939635515212,
      "num_tokens": 126116223.0,
      "step": 57560
    },
    {
      "entropy": 5.7226142406463625,
      "epoch": 5.713080587534736,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002279495785060135,
      "loss": 5.0141,
      "mean_token_accuracy": 0.1893393099308014,
      "num_tokens": 126126065.0,
      "step": 57565
    },
    {
      "entropy": 5.74671573638916,
      "epoch": 5.713576816196904,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022791493093114103,
      "loss": 4.9617,
      "mean_token_accuracy": 0.19979931861162187,
      "num_tokens": 126136143.0,
      "step": 57570
    },
    {
      "entropy": 5.643641662597656,
      "epoch": 5.714073044859071,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002278802845238767,
      "loss": 4.9009,
      "mean_token_accuracy": 0.19928113371133804,
      "num_tokens": 126146511.0,
      "step": 57575
    },
    {
      "entropy": 5.66040005683899,
      "epoch": 5.714569273521239,
      "grad_norm": 1.375,
      "learning_rate": 0.00022784563928507985,
      "loss": 4.8862,
      "mean_token_accuracy": 0.20173007398843765,
      "num_tokens": 126156797.0,
      "step": 57580
    },
    {
      "entropy": 5.677392101287841,
      "epoch": 5.715065502183406,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022781099521560944,
      "loss": 4.8909,
      "mean_token_accuracy": 0.20380140990018844,
      "num_tokens": 126166725.0,
      "step": 57585
    },
    {
      "entropy": 5.650263929367066,
      "epoch": 5.7155617308455735,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022777635231632475,
      "loss": 4.957,
      "mean_token_accuracy": 0.1954507663846016,
      "num_tokens": 126178612.0,
      "step": 57590
    },
    {
      "entropy": 5.725354909896851,
      "epoch": 5.716057959507741,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022774171058808474,
      "loss": 5.0372,
      "mean_token_accuracy": 0.18571568727493287,
      "num_tokens": 126189148.0,
      "step": 57595
    },
    {
      "entropy": 5.74089241027832,
      "epoch": 5.7165541881699085,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002277070700317485,
      "loss": 4.9733,
      "mean_token_accuracy": 0.20340831130743026,
      "num_tokens": 126201489.0,
      "step": 57600
    },
    {
      "entropy": 5.730569219589233,
      "epoch": 5.717050416832076,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000227672430648175,
      "loss": 5.0077,
      "mean_token_accuracy": 0.19382553100585936,
      "num_tokens": 126211922.0,
      "step": 57605
    },
    {
      "entropy": 5.727821350097656,
      "epoch": 5.717546645494243,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022763779243822324,
      "loss": 4.9269,
      "mean_token_accuracy": 0.19256190061569214,
      "num_tokens": 126222473.0,
      "step": 57610
    },
    {
      "entropy": 5.693418264389038,
      "epoch": 5.718042874156412,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002276031554027521,
      "loss": 4.9423,
      "mean_token_accuracy": 0.19806684106588363,
      "num_tokens": 126233112.0,
      "step": 57615
    },
    {
      "entropy": 5.73820013999939,
      "epoch": 5.718539102818578,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002275685195426207,
      "loss": 5.0026,
      "mean_token_accuracy": 0.19973009079694748,
      "num_tokens": 126244332.0,
      "step": 57620
    },
    {
      "entropy": 5.701053285598755,
      "epoch": 5.719035331480747,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002275338848586877,
      "loss": 4.9516,
      "mean_token_accuracy": 0.19980217069387435,
      "num_tokens": 126254482.0,
      "step": 57625
    },
    {
      "entropy": 5.75920090675354,
      "epoch": 5.719531560142914,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002274992513518122,
      "loss": 4.9952,
      "mean_token_accuracy": 0.19895262122154236,
      "num_tokens": 126265110.0,
      "step": 57630
    },
    {
      "entropy": 5.7064135551452635,
      "epoch": 5.720027788805082,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022746461902285286,
      "loss": 4.9677,
      "mean_token_accuracy": 0.19536732733249665,
      "num_tokens": 126275518.0,
      "step": 57635
    },
    {
      "entropy": 5.654152202606201,
      "epoch": 5.720524017467249,
      "grad_norm": 1.5,
      "learning_rate": 0.0002274299878726685,
      "loss": 4.9154,
      "mean_token_accuracy": 0.2006849080324173,
      "num_tokens": 126285846.0,
      "step": 57640
    },
    {
      "entropy": 5.701455545425415,
      "epoch": 5.721020246129417,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022739535790211786,
      "loss": 4.8996,
      "mean_token_accuracy": 0.19831749498844148,
      "num_tokens": 126296326.0,
      "step": 57645
    },
    {
      "entropy": 5.6831011295318605,
      "epoch": 5.721516474791584,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022736072911205992,
      "loss": 4.9578,
      "mean_token_accuracy": 0.1960069105029106,
      "num_tokens": 126307916.0,
      "step": 57650
    },
    {
      "entropy": 5.7357861518859865,
      "epoch": 5.722012703453752,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002273261015033532,
      "loss": 4.9182,
      "mean_token_accuracy": 0.19575366079807283,
      "num_tokens": 126319027.0,
      "step": 57655
    },
    {
      "entropy": 5.639750957489014,
      "epoch": 5.722508932115919,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022729147507685637,
      "loss": 4.8296,
      "mean_token_accuracy": 0.20670978128910064,
      "num_tokens": 126328697.0,
      "step": 57660
    },
    {
      "entropy": 5.628925132751465,
      "epoch": 5.7230051607780865,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022725684983342825,
      "loss": 4.8684,
      "mean_token_accuracy": 0.19647005945444107,
      "num_tokens": 126340796.0,
      "step": 57665
    },
    {
      "entropy": 5.65014533996582,
      "epoch": 5.723501389440254,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022722222577392732,
      "loss": 4.873,
      "mean_token_accuracy": 0.2049919918179512,
      "num_tokens": 126350916.0,
      "step": 57670
    },
    {
      "entropy": 5.710920763015747,
      "epoch": 5.7239976181024215,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002271876028992122,
      "loss": 4.9473,
      "mean_token_accuracy": 0.1960403487086296,
      "num_tokens": 126361668.0,
      "step": 57675
    },
    {
      "entropy": 5.6810503005981445,
      "epoch": 5.724493846764589,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022715298121014156,
      "loss": 4.8246,
      "mean_token_accuracy": 0.2063462719321251,
      "num_tokens": 126372813.0,
      "step": 57680
    },
    {
      "entropy": 5.651573324203492,
      "epoch": 5.724990075426756,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022711836070757392,
      "loss": 4.8729,
      "mean_token_accuracy": 0.20621806681156157,
      "num_tokens": 126382901.0,
      "step": 57685
    },
    {
      "entropy": 5.668340635299683,
      "epoch": 5.725486304088924,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022708374139236766,
      "loss": 4.8992,
      "mean_token_accuracy": 0.19757632911205292,
      "num_tokens": 126393338.0,
      "step": 57690
    },
    {
      "entropy": 5.708284521102906,
      "epoch": 5.725982532751091,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022704912326538146,
      "loss": 4.9891,
      "mean_token_accuracy": 0.1948530152440071,
      "num_tokens": 126403402.0,
      "step": 57695
    },
    {
      "entropy": 5.702939701080322,
      "epoch": 5.726478761413259,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022701450632747367,
      "loss": 4.9134,
      "mean_token_accuracy": 0.20268339663743973,
      "num_tokens": 126413938.0,
      "step": 57700
    },
    {
      "entropy": 5.7198155403137205,
      "epoch": 5.726974990075426,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022697989057950262,
      "loss": 4.9629,
      "mean_token_accuracy": 0.1975981965661049,
      "num_tokens": 126425368.0,
      "step": 57705
    },
    {
      "entropy": 5.6690915584564205,
      "epoch": 5.727471218737595,
      "grad_norm": 1.25,
      "learning_rate": 0.00022694527602232684,
      "loss": 4.9026,
      "mean_token_accuracy": 0.20105647593736647,
      "num_tokens": 126437666.0,
      "step": 57710
    },
    {
      "entropy": 5.667279386520386,
      "epoch": 5.727967447399762,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022691066265680476,
      "loss": 4.9661,
      "mean_token_accuracy": 0.19934780299663543,
      "num_tokens": 126448902.0,
      "step": 57715
    },
    {
      "entropy": 5.713267993927002,
      "epoch": 5.72846367606193,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022687605048379457,
      "loss": 5.0335,
      "mean_token_accuracy": 0.1913342446088791,
      "num_tokens": 126459970.0,
      "step": 57720
    },
    {
      "entropy": 5.786056709289551,
      "epoch": 5.728959904724097,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002268414395041546,
      "loss": 4.9721,
      "mean_token_accuracy": 0.19518371373414994,
      "num_tokens": 126469599.0,
      "step": 57725
    },
    {
      "entropy": 5.767856597900391,
      "epoch": 5.729456133386265,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022680682971874323,
      "loss": 4.8944,
      "mean_token_accuracy": 0.2027645707130432,
      "num_tokens": 126480035.0,
      "step": 57730
    },
    {
      "entropy": 5.669315052032471,
      "epoch": 5.729952362048432,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002267722211284185,
      "loss": 4.9007,
      "mean_token_accuracy": 0.19610959738492967,
      "num_tokens": 126490549.0,
      "step": 57735
    },
    {
      "entropy": 5.68255467414856,
      "epoch": 5.7304485907105995,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022673761373403884,
      "loss": 4.9626,
      "mean_token_accuracy": 0.1946921706199646,
      "num_tokens": 126500893.0,
      "step": 57740
    },
    {
      "entropy": 5.719953203201294,
      "epoch": 5.730944819372767,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002267030075364624,
      "loss": 4.9444,
      "mean_token_accuracy": 0.19747137725353242,
      "num_tokens": 126511883.0,
      "step": 57745
    },
    {
      "entropy": 5.6570196628570555,
      "epoch": 5.7314410480349345,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002266684025365473,
      "loss": 4.8505,
      "mean_token_accuracy": 0.21143787652254104,
      "num_tokens": 126521860.0,
      "step": 57750
    },
    {
      "entropy": 5.707198572158814,
      "epoch": 5.731937276697102,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022663379873515165,
      "loss": 4.9404,
      "mean_token_accuracy": 0.1965254068374634,
      "num_tokens": 126533476.0,
      "step": 57755
    },
    {
      "entropy": 5.673159074783325,
      "epoch": 5.732433505359269,
      "grad_norm": 1.53125,
      "learning_rate": 0.00022659919613313362,
      "loss": 4.9395,
      "mean_token_accuracy": 0.20307771861553192,
      "num_tokens": 126543733.0,
      "step": 57760
    },
    {
      "entropy": 5.6324687004089355,
      "epoch": 5.732929734021437,
      "grad_norm": 1.375,
      "learning_rate": 0.00022656459473135112,
      "loss": 4.8474,
      "mean_token_accuracy": 0.20810934603214265,
      "num_tokens": 126555014.0,
      "step": 57765
    },
    {
      "entropy": 5.695126485824585,
      "epoch": 5.733425962683604,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022652999453066236,
      "loss": 4.9056,
      "mean_token_accuracy": 0.19705677330493926,
      "num_tokens": 126567740.0,
      "step": 57770
    },
    {
      "entropy": 5.695030593872071,
      "epoch": 5.733922191345772,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022649539553192533,
      "loss": 4.945,
      "mean_token_accuracy": 0.19517373144626618,
      "num_tokens": 126578462.0,
      "step": 57775
    },
    {
      "entropy": 5.740369987487793,
      "epoch": 5.734418420007939,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022646079773599792,
      "loss": 4.9709,
      "mean_token_accuracy": 0.19405906349420549,
      "num_tokens": 126588955.0,
      "step": 57780
    },
    {
      "entropy": 5.708536005020141,
      "epoch": 5.734914648670107,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022642620114373814,
      "loss": 4.9668,
      "mean_token_accuracy": 0.19249422699213029,
      "num_tokens": 126598641.0,
      "step": 57785
    },
    {
      "entropy": 5.704345750808716,
      "epoch": 5.735410877332274,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022639160575600386,
      "loss": 4.8876,
      "mean_token_accuracy": 0.20159151405096054,
      "num_tokens": 126610625.0,
      "step": 57790
    },
    {
      "entropy": 5.704319620132447,
      "epoch": 5.735907105994443,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022635701157365312,
      "loss": 4.8918,
      "mean_token_accuracy": 0.1992774248123169,
      "num_tokens": 126622603.0,
      "step": 57795
    },
    {
      "entropy": 5.703262042999268,
      "epoch": 5.73640333465661,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022632241859754366,
      "loss": 4.9376,
      "mean_token_accuracy": 0.20066009759902953,
      "num_tokens": 126633526.0,
      "step": 57800
    },
    {
      "entropy": 5.665335702896118,
      "epoch": 5.736899563318778,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022628782682853328,
      "loss": 4.8761,
      "mean_token_accuracy": 0.20261275470256807,
      "num_tokens": 126643386.0,
      "step": 57805
    },
    {
      "entropy": 5.646238803863525,
      "epoch": 5.737395791980945,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002262532362674799,
      "loss": 4.997,
      "mean_token_accuracy": 0.19494284242391585,
      "num_tokens": 126655394.0,
      "step": 57810
    },
    {
      "entropy": 5.692834186553955,
      "epoch": 5.7378920206431125,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002262186469152412,
      "loss": 4.9792,
      "mean_token_accuracy": 0.19951552301645278,
      "num_tokens": 126668358.0,
      "step": 57815
    },
    {
      "entropy": 5.704586553573608,
      "epoch": 5.73838824930528,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022618405877267478,
      "loss": 4.9637,
      "mean_token_accuracy": 0.19880110919475555,
      "num_tokens": 126679683.0,
      "step": 57820
    },
    {
      "entropy": 5.733295965194702,
      "epoch": 5.7388844779674475,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022614947184063867,
      "loss": 4.9027,
      "mean_token_accuracy": 0.19384318590164185,
      "num_tokens": 126690918.0,
      "step": 57825
    },
    {
      "entropy": 5.758746004104614,
      "epoch": 5.739380706629615,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022611488611999037,
      "loss": 4.964,
      "mean_token_accuracy": 0.19695304185152054,
      "num_tokens": 126700619.0,
      "step": 57830
    },
    {
      "entropy": 5.765311241149902,
      "epoch": 5.7398769352917824,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022608030161158754,
      "loss": 4.9964,
      "mean_token_accuracy": 0.18403900265693665,
      "num_tokens": 126711080.0,
      "step": 57835
    },
    {
      "entropy": 5.689316749572754,
      "epoch": 5.74037316395395,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022604571831628785,
      "loss": 4.9419,
      "mean_token_accuracy": 0.1978563815355301,
      "num_tokens": 126722564.0,
      "step": 57840
    },
    {
      "entropy": 5.65244197845459,
      "epoch": 5.740869392616117,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002260111362349489,
      "loss": 4.8982,
      "mean_token_accuracy": 0.20419661253690718,
      "num_tokens": 126733478.0,
      "step": 57845
    },
    {
      "entropy": 5.744994497299194,
      "epoch": 5.741365621278285,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002259765553684281,
      "loss": 4.9875,
      "mean_token_accuracy": 0.19256093949079514,
      "num_tokens": 126745215.0,
      "step": 57850
    },
    {
      "entropy": 5.71834077835083,
      "epoch": 5.741861849940452,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022594197571758307,
      "loss": 4.9608,
      "mean_token_accuracy": 0.1946859061717987,
      "num_tokens": 126755658.0,
      "step": 57855
    },
    {
      "entropy": 5.689161586761474,
      "epoch": 5.74235807860262,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022590739728327148,
      "loss": 4.8264,
      "mean_token_accuracy": 0.2043805882334709,
      "num_tokens": 126766393.0,
      "step": 57860
    },
    {
      "entropy": 5.69323501586914,
      "epoch": 5.742854307264787,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022587282006635058,
      "loss": 4.899,
      "mean_token_accuracy": 0.19569279700517656,
      "num_tokens": 126775874.0,
      "step": 57865
    },
    {
      "entropy": 5.6405376434326175,
      "epoch": 5.743350535926955,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022583824406767784,
      "loss": 4.8299,
      "mean_token_accuracy": 0.20819351375102996,
      "num_tokens": 126787120.0,
      "step": 57870
    },
    {
      "entropy": 5.742135381698608,
      "epoch": 5.743846764589122,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002258036692881108,
      "loss": 4.9225,
      "mean_token_accuracy": 0.1959949716925621,
      "num_tokens": 126797977.0,
      "step": 57875
    },
    {
      "entropy": 5.7039556980133055,
      "epoch": 5.744342993251291,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022576909572850662,
      "loss": 4.8888,
      "mean_token_accuracy": 0.20139039903879166,
      "num_tokens": 126810501.0,
      "step": 57880
    },
    {
      "entropy": 5.740424537658692,
      "epoch": 5.744839221913458,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002257345233897229,
      "loss": 5.0156,
      "mean_token_accuracy": 0.19600046426057816,
      "num_tokens": 126822318.0,
      "step": 57885
    },
    {
      "entropy": 5.719148588180542,
      "epoch": 5.745335450575626,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002256999522726168,
      "loss": 4.9161,
      "mean_token_accuracy": 0.2009805768728256,
      "num_tokens": 126833683.0,
      "step": 57890
    },
    {
      "entropy": 5.707897043228149,
      "epoch": 5.745831679237793,
      "grad_norm": 1.25,
      "learning_rate": 0.00022566538237804568,
      "loss": 4.9692,
      "mean_token_accuracy": 0.1936842456459999,
      "num_tokens": 126843838.0,
      "step": 57895
    },
    {
      "entropy": 5.724301052093506,
      "epoch": 5.7463279078999605,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022563081370686677,
      "loss": 4.9419,
      "mean_token_accuracy": 0.1969887137413025,
      "num_tokens": 126854358.0,
      "step": 57900
    },
    {
      "entropy": 5.663491296768188,
      "epoch": 5.746824136562128,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002255962462599373,
      "loss": 4.9566,
      "mean_token_accuracy": 0.19616785347461702,
      "num_tokens": 126865171.0,
      "step": 57905
    },
    {
      "entropy": 5.698957633972168,
      "epoch": 5.7473203652242955,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022556168003811433,
      "loss": 4.9161,
      "mean_token_accuracy": 0.20144039988517762,
      "num_tokens": 126875661.0,
      "step": 57910
    },
    {
      "entropy": 5.745759153366089,
      "epoch": 5.747816593886463,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022552711504225531,
      "loss": 5.002,
      "mean_token_accuracy": 0.19615411907434463,
      "num_tokens": 126887016.0,
      "step": 57915
    },
    {
      "entropy": 5.678473854064942,
      "epoch": 5.74831282254863,
      "grad_norm": 1.25,
      "learning_rate": 0.00022549255127321722,
      "loss": 4.8804,
      "mean_token_accuracy": 0.19569116830825806,
      "num_tokens": 126897775.0,
      "step": 57920
    },
    {
      "entropy": 5.720049047470093,
      "epoch": 5.748809051210798,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002254579887318572,
      "loss": 4.9949,
      "mean_token_accuracy": 0.19656566232442857,
      "num_tokens": 126909555.0,
      "step": 57925
    },
    {
      "entropy": 5.687486791610718,
      "epoch": 5.749305279872965,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022542342741903226,
      "loss": 4.9648,
      "mean_token_accuracy": 0.1905846878886223,
      "num_tokens": 126918832.0,
      "step": 57930
    },
    {
      "entropy": 5.749055480957031,
      "epoch": 5.749801508535133,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022538886733559948,
      "loss": 4.9534,
      "mean_token_accuracy": 0.1919687196612358,
      "num_tokens": 126929203.0,
      "step": 57935
    },
    {
      "entropy": 5.706974124908447,
      "epoch": 5.7502977371973,
      "grad_norm": 1.375,
      "learning_rate": 0.0002253543084824159,
      "loss": 5.0045,
      "mean_token_accuracy": 0.19079172760248184,
      "num_tokens": 126940524.0,
      "step": 57940
    },
    {
      "entropy": 5.6721944332122805,
      "epoch": 5.750793965859468,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002253197508603385,
      "loss": 4.8828,
      "mean_token_accuracy": 0.191769640147686,
      "num_tokens": 126951324.0,
      "step": 57945
    },
    {
      "entropy": 5.760726976394653,
      "epoch": 5.751290194521635,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022528519447022428,
      "loss": 4.9883,
      "mean_token_accuracy": 0.20074700862169265,
      "num_tokens": 126964166.0,
      "step": 57950
    },
    {
      "entropy": 5.799620771408081,
      "epoch": 5.751786423183803,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022525063931293004,
      "loss": 5.0078,
      "mean_token_accuracy": 0.19904833137989045,
      "num_tokens": 126975181.0,
      "step": 57955
    },
    {
      "entropy": 5.707713460922241,
      "epoch": 5.75228265184597,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022521608538931276,
      "loss": 4.8626,
      "mean_token_accuracy": 0.19945641607046127,
      "num_tokens": 126984858.0,
      "step": 57960
    },
    {
      "entropy": 5.6863181591033936,
      "epoch": 5.752778880508139,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022518153270022928,
      "loss": 5.0098,
      "mean_token_accuracy": 0.1938188835978508,
      "num_tokens": 126995750.0,
      "step": 57965
    },
    {
      "entropy": 5.694733715057373,
      "epoch": 5.753275109170306,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022514698124653638,
      "loss": 4.9088,
      "mean_token_accuracy": 0.19905686229467393,
      "num_tokens": 127005891.0,
      "step": 57970
    },
    {
      "entropy": 5.679846954345703,
      "epoch": 5.7537713378324735,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022511243102909108,
      "loss": 4.8848,
      "mean_token_accuracy": 0.19916158467531203,
      "num_tokens": 127017637.0,
      "step": 57975
    },
    {
      "entropy": 5.639434671401977,
      "epoch": 5.754267566494641,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022507788204874992,
      "loss": 4.8111,
      "mean_token_accuracy": 0.20661670416593553,
      "num_tokens": 127028392.0,
      "step": 57980
    },
    {
      "entropy": 5.716711044311523,
      "epoch": 5.7547637951568085,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022504333430636964,
      "loss": 4.9295,
      "mean_token_accuracy": 0.20197965502738952,
      "num_tokens": 127039556.0,
      "step": 57985
    },
    {
      "entropy": 5.665976667404175,
      "epoch": 5.755260023818976,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002250087878028071,
      "loss": 4.8561,
      "mean_token_accuracy": 0.20023768544197082,
      "num_tokens": 127050823.0,
      "step": 57990
    },
    {
      "entropy": 5.696610784530639,
      "epoch": 5.755756252481143,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002249742425389189,
      "loss": 4.9167,
      "mean_token_accuracy": 0.19707411676645278,
      "num_tokens": 127063101.0,
      "step": 57995
    },
    {
      "entropy": 5.746295547485351,
      "epoch": 5.756252481143311,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00022493969851556162,
      "loss": 4.9755,
      "mean_token_accuracy": 0.1960520327091217,
      "num_tokens": 127074030.0,
      "step": 58000
    },
    {
      "entropy": 5.705738973617554,
      "epoch": 5.756748709805478,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000224905155733592,
      "loss": 4.945,
      "mean_token_accuracy": 0.19648600071668626,
      "num_tokens": 127085087.0,
      "step": 58005
    },
    {
      "entropy": 5.708842611312866,
      "epoch": 5.757244938467646,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022487061419386656,
      "loss": 4.8513,
      "mean_token_accuracy": 0.19844525158405305,
      "num_tokens": 127096228.0,
      "step": 58010
    },
    {
      "entropy": 5.685176706314087,
      "epoch": 5.757741167129813,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022483607389724186,
      "loss": 4.9241,
      "mean_token_accuracy": 0.20383584648370742,
      "num_tokens": 127106641.0,
      "step": 58015
    },
    {
      "entropy": 5.656019020080566,
      "epoch": 5.758237395791981,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002248015348445745,
      "loss": 4.8875,
      "mean_token_accuracy": 0.1953232079744339,
      "num_tokens": 127116560.0,
      "step": 58020
    },
    {
      "entropy": 5.641917324066162,
      "epoch": 5.758733624454148,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00022476699703672076,
      "loss": 4.8308,
      "mean_token_accuracy": 0.20800233036279678,
      "num_tokens": 127128419.0,
      "step": 58025
    },
    {
      "entropy": 5.732458257675171,
      "epoch": 5.759229853116316,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022473246047453738,
      "loss": 5.0064,
      "mean_token_accuracy": 0.19336505234241486,
      "num_tokens": 127139465.0,
      "step": 58030
    },
    {
      "entropy": 5.776294898986817,
      "epoch": 5.759726081778483,
      "grad_norm": 1.5,
      "learning_rate": 0.00022469792515888065,
      "loss": 5.0203,
      "mean_token_accuracy": 0.18627913445234298,
      "num_tokens": 127149255.0,
      "step": 58035
    },
    {
      "entropy": 5.750331401824951,
      "epoch": 5.760222310440651,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022466339109060703,
      "loss": 4.9231,
      "mean_token_accuracy": 0.20102486461400987,
      "num_tokens": 127161281.0,
      "step": 58040
    },
    {
      "entropy": 5.689100837707519,
      "epoch": 5.760718539102818,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022462885827057277,
      "loss": 4.9613,
      "mean_token_accuracy": 0.19578615427017212,
      "num_tokens": 127173210.0,
      "step": 58045
    },
    {
      "entropy": 5.799692153930664,
      "epoch": 5.7612147677649865,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022459432669963425,
      "loss": 5.0738,
      "mean_token_accuracy": 0.19679509103298187,
      "num_tokens": 127184224.0,
      "step": 58050
    },
    {
      "entropy": 5.746367883682251,
      "epoch": 5.761710996427154,
      "grad_norm": 1.703125,
      "learning_rate": 0.00022455979637864783,
      "loss": 4.8821,
      "mean_token_accuracy": 0.20414696931838988,
      "num_tokens": 127195464.0,
      "step": 58055
    },
    {
      "entropy": 5.698834180831909,
      "epoch": 5.7622072250893215,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022452526730846984,
      "loss": 4.8511,
      "mean_token_accuracy": 0.20002758204936982,
      "num_tokens": 127205803.0,
      "step": 58060
    },
    {
      "entropy": 5.720197486877441,
      "epoch": 5.762703453751489,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022449073948995645,
      "loss": 4.9466,
      "mean_token_accuracy": 0.19648927450180054,
      "num_tokens": 127215554.0,
      "step": 58065
    },
    {
      "entropy": 5.693003082275391,
      "epoch": 5.763199682413656,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022445621292396385,
      "loss": 4.9838,
      "mean_token_accuracy": 0.19816411137580872,
      "num_tokens": 127225378.0,
      "step": 58070
    },
    {
      "entropy": 5.735278654098511,
      "epoch": 5.763695911075824,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022442168761134824,
      "loss": 5.0186,
      "mean_token_accuracy": 0.18821144700050355,
      "num_tokens": 127236765.0,
      "step": 58075
    },
    {
      "entropy": 5.716013240814209,
      "epoch": 5.764192139737991,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022438716355296579,
      "loss": 5.02,
      "mean_token_accuracy": 0.193942591547966,
      "num_tokens": 127248501.0,
      "step": 58080
    },
    {
      "entropy": 5.752778148651123,
      "epoch": 5.764688368400159,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002243526407496726,
      "loss": 4.9681,
      "mean_token_accuracy": 0.19488418400287627,
      "num_tokens": 127260503.0,
      "step": 58085
    },
    {
      "entropy": 5.730622720718384,
      "epoch": 5.765184597062326,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022431811920232487,
      "loss": 4.9442,
      "mean_token_accuracy": 0.20277346521615983,
      "num_tokens": 127271188.0,
      "step": 58090
    },
    {
      "entropy": 5.634637880325317,
      "epoch": 5.765680825724494,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00022428359891177858,
      "loss": 4.8766,
      "mean_token_accuracy": 0.20132311731576918,
      "num_tokens": 127281218.0,
      "step": 58095
    },
    {
      "entropy": 5.6919502258300785,
      "epoch": 5.766177054386661,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022424907987888965,
      "loss": 4.9529,
      "mean_token_accuracy": 0.19642252773046492,
      "num_tokens": 127291319.0,
      "step": 58100
    },
    {
      "entropy": 5.6537189960479735,
      "epoch": 5.766673283048829,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002242145621045143,
      "loss": 4.9024,
      "mean_token_accuracy": 0.1944386824965477,
      "num_tokens": 127302896.0,
      "step": 58105
    },
    {
      "entropy": 5.722942686080932,
      "epoch": 5.767169511710996,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022418004558950834,
      "loss": 4.9839,
      "mean_token_accuracy": 0.19759404957294463,
      "num_tokens": 127315551.0,
      "step": 58110
    },
    {
      "entropy": 5.647590970993042,
      "epoch": 5.767665740373164,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022414553033472763,
      "loss": 4.8383,
      "mean_token_accuracy": 0.20885000377893448,
      "num_tokens": 127326917.0,
      "step": 58115
    },
    {
      "entropy": 5.6961651802062985,
      "epoch": 5.768161969035331,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022411101634102826,
      "loss": 4.9577,
      "mean_token_accuracy": 0.19662870913743974,
      "num_tokens": 127337480.0,
      "step": 58120
    },
    {
      "entropy": 5.6134655475616455,
      "epoch": 5.768658197697499,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022407650360926608,
      "loss": 4.8314,
      "mean_token_accuracy": 0.2108493536710739,
      "num_tokens": 127347911.0,
      "step": 58125
    },
    {
      "entropy": 5.692152070999145,
      "epoch": 5.769154426359666,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002240419921402968,
      "loss": 4.958,
      "mean_token_accuracy": 0.19539682269096376,
      "num_tokens": 127358923.0,
      "step": 58130
    },
    {
      "entropy": 5.716971921920776,
      "epoch": 5.7696506550218345,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022400748193497633,
      "loss": 5.0021,
      "mean_token_accuracy": 0.19290910214185714,
      "num_tokens": 127369956.0,
      "step": 58135
    },
    {
      "entropy": 5.638066196441651,
      "epoch": 5.770146883684002,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022397297299416052,
      "loss": 4.8968,
      "mean_token_accuracy": 0.20387188494205474,
      "num_tokens": 127381328.0,
      "step": 58140
    },
    {
      "entropy": 5.619447851181031,
      "epoch": 5.7706431123461694,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022393846531870487,
      "loss": 4.8253,
      "mean_token_accuracy": 0.20720666497945786,
      "num_tokens": 127391410.0,
      "step": 58145
    },
    {
      "entropy": 5.676587247848511,
      "epoch": 5.771139341008337,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022390395890946536,
      "loss": 4.9693,
      "mean_token_accuracy": 0.19804570227861404,
      "num_tokens": 127401905.0,
      "step": 58150
    },
    {
      "entropy": 5.683741044998169,
      "epoch": 5.771635569670504,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022386945376729755,
      "loss": 4.9308,
      "mean_token_accuracy": 0.19798100739717484,
      "num_tokens": 127414003.0,
      "step": 58155
    },
    {
      "entropy": 5.751174688339233,
      "epoch": 5.772131798332672,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022383494989305707,
      "loss": 5.0461,
      "mean_token_accuracy": 0.19309426993131637,
      "num_tokens": 127425526.0,
      "step": 58160
    },
    {
      "entropy": 5.711386871337891,
      "epoch": 5.772628026994839,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022380044728759964,
      "loss": 4.8998,
      "mean_token_accuracy": 0.19898383766412736,
      "num_tokens": 127435592.0,
      "step": 58165
    },
    {
      "entropy": 5.762981939315796,
      "epoch": 5.773124255657007,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022376594595178078,
      "loss": 4.963,
      "mean_token_accuracy": 0.1984096050262451,
      "num_tokens": 127446995.0,
      "step": 58170
    },
    {
      "entropy": 5.774453067779541,
      "epoch": 5.773620484319174,
      "grad_norm": 1.171875,
      "learning_rate": 0.000223731445886456,
      "loss": 5.0209,
      "mean_token_accuracy": 0.19926493912935256,
      "num_tokens": 127458735.0,
      "step": 58175
    },
    {
      "entropy": 5.620956802368164,
      "epoch": 5.774116712981342,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000223696947092481,
      "loss": 4.8755,
      "mean_token_accuracy": 0.2059613823890686,
      "num_tokens": 127471557.0,
      "step": 58180
    },
    {
      "entropy": 5.679146862030029,
      "epoch": 5.774612941643509,
      "grad_norm": 1.25,
      "learning_rate": 0.00022366244957071102,
      "loss": 4.9345,
      "mean_token_accuracy": 0.201523357629776,
      "num_tokens": 127483020.0,
      "step": 58185
    },
    {
      "entropy": 5.724041080474853,
      "epoch": 5.775109170305677,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002236279533220018,
      "loss": 4.9737,
      "mean_token_accuracy": 0.1914254680275917,
      "num_tokens": 127493899.0,
      "step": 58190
    },
    {
      "entropy": 5.691632509231567,
      "epoch": 5.775605398967844,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022359345834720856,
      "loss": 4.8944,
      "mean_token_accuracy": 0.19949946999549867,
      "num_tokens": 127504725.0,
      "step": 58195
    },
    {
      "entropy": 5.704336023330688,
      "epoch": 5.776101627630012,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002235589646471869,
      "loss": 4.9267,
      "mean_token_accuracy": 0.20283062309026717,
      "num_tokens": 127514609.0,
      "step": 58200
    },
    {
      "entropy": 5.665882587432861,
      "epoch": 5.776597856292179,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022352447222279189,
      "loss": 4.8631,
      "mean_token_accuracy": 0.20058999210596085,
      "num_tokens": 127525283.0,
      "step": 58205
    },
    {
      "entropy": 5.625262832641601,
      "epoch": 5.777094084954347,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022348998107487916,
      "loss": 4.8805,
      "mean_token_accuracy": 0.20792245864868164,
      "num_tokens": 127536054.0,
      "step": 58210
    },
    {
      "entropy": 5.701074504852295,
      "epoch": 5.777590313616514,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022345549120430386,
      "loss": 4.9708,
      "mean_token_accuracy": 0.1942405194044113,
      "num_tokens": 127546851.0,
      "step": 58215
    },
    {
      "entropy": 5.7073148727417,
      "epoch": 5.7780865422786825,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022342100261192138,
      "loss": 4.8847,
      "mean_token_accuracy": 0.20212026834487914,
      "num_tokens": 127557817.0,
      "step": 58220
    },
    {
      "entropy": 5.69913969039917,
      "epoch": 5.778582770940849,
      "grad_norm": 1.375,
      "learning_rate": 0.00022338651529858685,
      "loss": 4.9621,
      "mean_token_accuracy": 0.19588710814714433,
      "num_tokens": 127568906.0,
      "step": 58225
    },
    {
      "entropy": 5.694211530685425,
      "epoch": 5.779078999603017,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022335202926515552,
      "loss": 4.9267,
      "mean_token_accuracy": 0.20085869878530502,
      "num_tokens": 127580557.0,
      "step": 58230
    },
    {
      "entropy": 5.703103065490723,
      "epoch": 5.779575228265185,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002233175445124825,
      "loss": 4.9477,
      "mean_token_accuracy": 0.19483615458011627,
      "num_tokens": 127590474.0,
      "step": 58235
    },
    {
      "entropy": 5.696908092498779,
      "epoch": 5.780071456927352,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002232830610414231,
      "loss": 4.914,
      "mean_token_accuracy": 0.19904749989509582,
      "num_tokens": 127601645.0,
      "step": 58240
    },
    {
      "entropy": 5.619973421096802,
      "epoch": 5.78056768558952,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022324857885283234,
      "loss": 4.894,
      "mean_token_accuracy": 0.20397546589374543,
      "num_tokens": 127612386.0,
      "step": 58245
    },
    {
      "entropy": 5.6731181144714355,
      "epoch": 5.781063914251687,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002232140979475653,
      "loss": 4.8455,
      "mean_token_accuracy": 0.21640425622463227,
      "num_tokens": 127622432.0,
      "step": 58250
    },
    {
      "entropy": 5.714103174209595,
      "epoch": 5.781560142913855,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002231796183264771,
      "loss": 4.9454,
      "mean_token_accuracy": 0.2010340303182602,
      "num_tokens": 127633244.0,
      "step": 58255
    },
    {
      "entropy": 5.682764959335327,
      "epoch": 5.782056371576022,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022314513999042263,
      "loss": 4.8966,
      "mean_token_accuracy": 0.1995510682463646,
      "num_tokens": 127643248.0,
      "step": 58260
    },
    {
      "entropy": 5.685223770141602,
      "epoch": 5.78255260023819,
      "grad_norm": 1.5,
      "learning_rate": 0.00022311066294025683,
      "loss": 4.8415,
      "mean_token_accuracy": 0.20304139256477355,
      "num_tokens": 127651931.0,
      "step": 58265
    },
    {
      "entropy": 5.687808895111084,
      "epoch": 5.783048828900357,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000223076187176835,
      "loss": 4.9691,
      "mean_token_accuracy": 0.19126427173614502,
      "num_tokens": 127663082.0,
      "step": 58270
    },
    {
      "entropy": 5.683142614364624,
      "epoch": 5.783545057562525,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022304171270101175,
      "loss": 4.9258,
      "mean_token_accuracy": 0.19314566403627395,
      "num_tokens": 127673823.0,
      "step": 58275
    },
    {
      "entropy": 5.676008987426758,
      "epoch": 5.784041286224692,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022300723951364205,
      "loss": 4.9223,
      "mean_token_accuracy": 0.19114744961261748,
      "num_tokens": 127684569.0,
      "step": 58280
    },
    {
      "entropy": 5.687517976760864,
      "epoch": 5.78453751488686,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022297276761558082,
      "loss": 4.9224,
      "mean_token_accuracy": 0.19643137603998184,
      "num_tokens": 127694881.0,
      "step": 58285
    },
    {
      "entropy": 5.687993907928467,
      "epoch": 5.785033743549027,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022293829700768274,
      "loss": 4.8406,
      "mean_token_accuracy": 0.20019372701644897,
      "num_tokens": 127705720.0,
      "step": 58290
    },
    {
      "entropy": 5.738160943984985,
      "epoch": 5.785529972211195,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022290382769080288,
      "loss": 5.0014,
      "mean_token_accuracy": 0.1946871519088745,
      "num_tokens": 127716463.0,
      "step": 58295
    },
    {
      "entropy": 5.762090158462525,
      "epoch": 5.786026200873362,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022286935966579568,
      "loss": 4.9816,
      "mean_token_accuracy": 0.19675974398851395,
      "num_tokens": 127728246.0,
      "step": 58300
    },
    {
      "entropy": 5.681997442245484,
      "epoch": 5.78652242953553,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022283489293351615,
      "loss": 4.8783,
      "mean_token_accuracy": 0.20438384264707565,
      "num_tokens": 127739426.0,
      "step": 58305
    },
    {
      "entropy": 5.704430961608887,
      "epoch": 5.787018658197697,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022280042749481878,
      "loss": 4.9403,
      "mean_token_accuracy": 0.1981241673231125,
      "num_tokens": 127750062.0,
      "step": 58310
    },
    {
      "entropy": 5.640171098709106,
      "epoch": 5.787514886859865,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022276596335055844,
      "loss": 4.8708,
      "mean_token_accuracy": 0.20164433121681213,
      "num_tokens": 127760761.0,
      "step": 58315
    },
    {
      "entropy": 5.662625789642334,
      "epoch": 5.788011115522033,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022273150050158947,
      "loss": 4.9851,
      "mean_token_accuracy": 0.19681290537118912,
      "num_tokens": 127772685.0,
      "step": 58320
    },
    {
      "entropy": 5.705163478851318,
      "epoch": 5.7885073441842,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022269703894876682,
      "loss": 4.9349,
      "mean_token_accuracy": 0.19709748178720474,
      "num_tokens": 127784376.0,
      "step": 58325
    },
    {
      "entropy": 5.625299882888794,
      "epoch": 5.789003572846368,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022266257869294482,
      "loss": 4.8537,
      "mean_token_accuracy": 0.20833001881837845,
      "num_tokens": 127796553.0,
      "step": 58330
    },
    {
      "entropy": 5.714344882965088,
      "epoch": 5.789499801508535,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002226281197349781,
      "loss": 4.9168,
      "mean_token_accuracy": 0.19295355230569838,
      "num_tokens": 127806194.0,
      "step": 58335
    },
    {
      "entropy": 5.607316732406616,
      "epoch": 5.789996030170703,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022259366207572125,
      "loss": 4.886,
      "mean_token_accuracy": 0.20088718980550765,
      "num_tokens": 127815557.0,
      "step": 58340
    },
    {
      "entropy": 5.662362098693848,
      "epoch": 5.79049225883287,
      "grad_norm": 1.515625,
      "learning_rate": 0.00022255920571602854,
      "loss": 4.8686,
      "mean_token_accuracy": 0.20342513173818588,
      "num_tokens": 127825298.0,
      "step": 58345
    },
    {
      "entropy": 5.672231578826905,
      "epoch": 5.790988487495038,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022252475065675454,
      "loss": 4.9651,
      "mean_token_accuracy": 0.19176231622695922,
      "num_tokens": 127835468.0,
      "step": 58350
    },
    {
      "entropy": 5.6874267578125,
      "epoch": 5.791484716157205,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022249029689875368,
      "loss": 4.9872,
      "mean_token_accuracy": 0.19087164252996444,
      "num_tokens": 127848457.0,
      "step": 58355
    },
    {
      "entropy": 5.698154258728027,
      "epoch": 5.791980944819373,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002224558444428803,
      "loss": 4.8857,
      "mean_token_accuracy": 0.2004486009478569,
      "num_tokens": 127859178.0,
      "step": 58360
    },
    {
      "entropy": 5.703315210342407,
      "epoch": 5.79247717348154,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022242139328998878,
      "loss": 4.9329,
      "mean_token_accuracy": 0.19525542557239534,
      "num_tokens": 127871848.0,
      "step": 58365
    },
    {
      "entropy": 5.653312635421753,
      "epoch": 5.792973402143708,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002223869434409334,
      "loss": 4.804,
      "mean_token_accuracy": 0.20834360718727113,
      "num_tokens": 127882458.0,
      "step": 58370
    },
    {
      "entropy": 5.6965100288391115,
      "epoch": 5.793469630805875,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00022235249489656846,
      "loss": 4.8784,
      "mean_token_accuracy": 0.2049626275897026,
      "num_tokens": 127894083.0,
      "step": 58375
    },
    {
      "entropy": 5.8535867691040036,
      "epoch": 5.7939658594680425,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002223180476577481,
      "loss": 5.1075,
      "mean_token_accuracy": 0.19343860000371932,
      "num_tokens": 127905730.0,
      "step": 58380
    },
    {
      "entropy": 5.6949817657470705,
      "epoch": 5.79446208813021,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002222836017253268,
      "loss": 4.9054,
      "mean_token_accuracy": 0.19632044434547424,
      "num_tokens": 127915865.0,
      "step": 58385
    },
    {
      "entropy": 5.680736255645752,
      "epoch": 5.794958316792378,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022224915710015853,
      "loss": 4.852,
      "mean_token_accuracy": 0.20393610298633574,
      "num_tokens": 127925650.0,
      "step": 58390
    },
    {
      "entropy": 5.704229688644409,
      "epoch": 5.795454545454545,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022221471378309744,
      "loss": 4.9558,
      "mean_token_accuracy": 0.19581321626901627,
      "num_tokens": 127936219.0,
      "step": 58395
    },
    {
      "entropy": 5.680030727386475,
      "epoch": 5.795950774116713,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002221802717749978,
      "loss": 4.9676,
      "mean_token_accuracy": 0.18934658020734788,
      "num_tokens": 127949682.0,
      "step": 58400
    },
    {
      "entropy": 5.684518432617187,
      "epoch": 5.796447002778881,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022214583107671348,
      "loss": 4.9191,
      "mean_token_accuracy": 0.2010464534163475,
      "num_tokens": 127959310.0,
      "step": 58405
    },
    {
      "entropy": 5.718071174621582,
      "epoch": 5.796943231441048,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002221113916890987,
      "loss": 4.9152,
      "mean_token_accuracy": 0.2004266768693924,
      "num_tokens": 127969826.0,
      "step": 58410
    },
    {
      "entropy": 5.604336977005005,
      "epoch": 5.797439460103216,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002220769536130075,
      "loss": 4.8314,
      "mean_token_accuracy": 0.2132965788245201,
      "num_tokens": 127980679.0,
      "step": 58415
    },
    {
      "entropy": 5.735607719421386,
      "epoch": 5.797935688765383,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022204251684929376,
      "loss": 5.0441,
      "mean_token_accuracy": 0.1932264819741249,
      "num_tokens": 127992967.0,
      "step": 58420
    },
    {
      "entropy": 5.718303537368774,
      "epoch": 5.798431917427551,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002220080813988115,
      "loss": 4.8978,
      "mean_token_accuracy": 0.1958673968911171,
      "num_tokens": 128002264.0,
      "step": 58425
    },
    {
      "entropy": 5.751611518859863,
      "epoch": 5.798928146089718,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022197364726241466,
      "loss": 4.9457,
      "mean_token_accuracy": 0.19854889512062074,
      "num_tokens": 128012648.0,
      "step": 58430
    },
    {
      "entropy": 5.673133420944214,
      "epoch": 5.799424374751886,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022193921444095712,
      "loss": 4.9195,
      "mean_token_accuracy": 0.19920603781938553,
      "num_tokens": 128022806.0,
      "step": 58435
    },
    {
      "entropy": 5.659834003448486,
      "epoch": 5.799920603414053,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022190478293529264,
      "loss": 4.9158,
      "mean_token_accuracy": 0.19309787899255754,
      "num_tokens": 128033135.0,
      "step": 58440
    },
    {
      "entropy": 5.696817398071289,
      "epoch": 5.800416832076221,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022187035274627516,
      "loss": 4.9377,
      "mean_token_accuracy": 0.2011595770716667,
      "num_tokens": 128043568.0,
      "step": 58445
    },
    {
      "entropy": 5.690385007858277,
      "epoch": 5.800913060738388,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002218359238747585,
      "loss": 4.8808,
      "mean_token_accuracy": 0.20826277881860733,
      "num_tokens": 128054505.0,
      "step": 58450
    },
    {
      "entropy": 5.718676614761352,
      "epoch": 5.801409289400556,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022180149632159636,
      "loss": 4.8994,
      "mean_token_accuracy": 0.1947806164622307,
      "num_tokens": 128064785.0,
      "step": 58455
    },
    {
      "entropy": 5.662133693695068,
      "epoch": 5.801905518062723,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022176707008764242,
      "loss": 4.8783,
      "mean_token_accuracy": 0.2033776581287384,
      "num_tokens": 128076158.0,
      "step": 58460
    },
    {
      "entropy": 5.742129278182984,
      "epoch": 5.8024017467248905,
      "grad_norm": 1.5625,
      "learning_rate": 0.00022173264517375054,
      "loss": 4.9878,
      "mean_token_accuracy": 0.199283929169178,
      "num_tokens": 128087274.0,
      "step": 58465
    },
    {
      "entropy": 5.6559689998626705,
      "epoch": 5.802897975387058,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022169822158077412,
      "loss": 4.914,
      "mean_token_accuracy": 0.19756375551223754,
      "num_tokens": 128097057.0,
      "step": 58470
    },
    {
      "entropy": 5.678902816772461,
      "epoch": 5.803394204049226,
      "grad_norm": 1.5,
      "learning_rate": 0.00022166379930956703,
      "loss": 4.8745,
      "mean_token_accuracy": 0.20314205884933473,
      "num_tokens": 128106671.0,
      "step": 58475
    },
    {
      "entropy": 5.712132120132447,
      "epoch": 5.803890432711393,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022162937836098286,
      "loss": 4.9464,
      "mean_token_accuracy": 0.20054898411035538,
      "num_tokens": 128117851.0,
      "step": 58480
    },
    {
      "entropy": 5.74404239654541,
      "epoch": 5.804386661373561,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022159495873587504,
      "loss": 5.0243,
      "mean_token_accuracy": 0.1911612033843994,
      "num_tokens": 128128557.0,
      "step": 58485
    },
    {
      "entropy": 5.705015087127686,
      "epoch": 5.804882890035729,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022156054043509716,
      "loss": 4.8931,
      "mean_token_accuracy": 0.20134054720401764,
      "num_tokens": 128138815.0,
      "step": 58490
    },
    {
      "entropy": 5.752691316604614,
      "epoch": 5.805379118697896,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022152612345950273,
      "loss": 4.9668,
      "mean_token_accuracy": 0.1994567781686783,
      "num_tokens": 128150665.0,
      "step": 58495
    },
    {
      "entropy": 5.7024938583374025,
      "epoch": 5.805875347360064,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022149170780994521,
      "loss": 4.9698,
      "mean_token_accuracy": 0.19583300203084947,
      "num_tokens": 128162594.0,
      "step": 58500
    },
    {
      "entropy": 5.705687046051025,
      "epoch": 5.806371576022231,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022145729348727806,
      "loss": 4.9716,
      "mean_token_accuracy": 0.19848543256521226,
      "num_tokens": 128175149.0,
      "step": 58505
    },
    {
      "entropy": 5.694423627853394,
      "epoch": 5.806867804684399,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002214228804923546,
      "loss": 4.8969,
      "mean_token_accuracy": 0.19436553865671158,
      "num_tokens": 128185446.0,
      "step": 58510
    },
    {
      "entropy": 5.673777675628662,
      "epoch": 5.807364033346566,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002213884688260283,
      "loss": 4.9251,
      "mean_token_accuracy": 0.19366333782672882,
      "num_tokens": 128196193.0,
      "step": 58515
    },
    {
      "entropy": 5.726281261444091,
      "epoch": 5.807860262008734,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002213540584891525,
      "loss": 4.9257,
      "mean_token_accuracy": 0.19844977110624312,
      "num_tokens": 128205888.0,
      "step": 58520
    },
    {
      "entropy": 5.710624504089355,
      "epoch": 5.808356490670901,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022131964948258033,
      "loss": 4.922,
      "mean_token_accuracy": 0.20141729563474656,
      "num_tokens": 128215582.0,
      "step": 58525
    },
    {
      "entropy": 5.682279300689697,
      "epoch": 5.808852719333069,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022128524180716524,
      "loss": 4.944,
      "mean_token_accuracy": 0.19966210871934892,
      "num_tokens": 128227075.0,
      "step": 58530
    },
    {
      "entropy": 5.673195743560791,
      "epoch": 5.809348947995236,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022125083546376047,
      "loss": 4.9457,
      "mean_token_accuracy": 0.20107483267784118,
      "num_tokens": 128237870.0,
      "step": 58535
    },
    {
      "entropy": 5.683201265335083,
      "epoch": 5.8098451766574035,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002212164304532191,
      "loss": 4.9359,
      "mean_token_accuracy": 0.19631602764129638,
      "num_tokens": 128248964.0,
      "step": 58540
    },
    {
      "entropy": 5.62153639793396,
      "epoch": 5.810341405319571,
      "grad_norm": 1.078125,
      "learning_rate": 0.00022118202677639439,
      "loss": 4.8495,
      "mean_token_accuracy": 0.2010572224855423,
      "num_tokens": 128259926.0,
      "step": 58545
    },
    {
      "entropy": 5.656550312042237,
      "epoch": 5.8108376339817385,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022114762443413945,
      "loss": 4.8382,
      "mean_token_accuracy": 0.20520540773868562,
      "num_tokens": 128269751.0,
      "step": 58550
    },
    {
      "entropy": 5.714500188827515,
      "epoch": 5.811333862643906,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022111322342730733,
      "loss": 4.9434,
      "mean_token_accuracy": 0.20051014870405198,
      "num_tokens": 128280824.0,
      "step": 58555
    },
    {
      "entropy": 5.73269305229187,
      "epoch": 5.811830091306074,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022107882375675114,
      "loss": 5.0355,
      "mean_token_accuracy": 0.19524383395910264,
      "num_tokens": 128293383.0,
      "step": 58560
    },
    {
      "entropy": 5.717476463317871,
      "epoch": 5.812326319968241,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022104442542332398,
      "loss": 5.0212,
      "mean_token_accuracy": 0.19679967015981675,
      "num_tokens": 128303304.0,
      "step": 58565
    },
    {
      "entropy": 5.679752254486084,
      "epoch": 5.812822548630409,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022101002842787887,
      "loss": 4.9004,
      "mean_token_accuracy": 0.19419539868831634,
      "num_tokens": 128314180.0,
      "step": 58570
    },
    {
      "entropy": 5.725925779342651,
      "epoch": 5.813318777292577,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002209756327712686,
      "loss": 4.9555,
      "mean_token_accuracy": 0.19320940226316452,
      "num_tokens": 128325692.0,
      "step": 58575
    },
    {
      "entropy": 5.677733516693115,
      "epoch": 5.813815005954744,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022094123845434633,
      "loss": 4.9606,
      "mean_token_accuracy": 0.19623627066612243,
      "num_tokens": 128336832.0,
      "step": 58580
    },
    {
      "entropy": 5.677932071685791,
      "epoch": 5.814311234616912,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022090684547796475,
      "loss": 4.8957,
      "mean_token_accuracy": 0.20283156931400298,
      "num_tokens": 128348630.0,
      "step": 58585
    },
    {
      "entropy": 5.748557662963867,
      "epoch": 5.814807463279079,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022087245384297684,
      "loss": 4.9545,
      "mean_token_accuracy": 0.1890639156103134,
      "num_tokens": 128359766.0,
      "step": 58590
    },
    {
      "entropy": 5.782574462890625,
      "epoch": 5.815303691941247,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002208380635502355,
      "loss": 4.9466,
      "mean_token_accuracy": 0.20069824904203415,
      "num_tokens": 128370903.0,
      "step": 58595
    },
    {
      "entropy": 5.769389009475708,
      "epoch": 5.815799920603414,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022080367460059353,
      "loss": 5.0047,
      "mean_token_accuracy": 0.19616011679172515,
      "num_tokens": 128381495.0,
      "step": 58600
    },
    {
      "entropy": 5.628420162200928,
      "epoch": 5.816296149265582,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022076928699490362,
      "loss": 4.8974,
      "mean_token_accuracy": 0.1987290635704994,
      "num_tokens": 128392166.0,
      "step": 58605
    },
    {
      "entropy": 5.649743890762329,
      "epoch": 5.816792377927749,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022073490073401848,
      "loss": 4.912,
      "mean_token_accuracy": 0.198316690325737,
      "num_tokens": 128402338.0,
      "step": 58610
    },
    {
      "entropy": 5.704355144500733,
      "epoch": 5.8172886065899165,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022070051581879086,
      "loss": 4.997,
      "mean_token_accuracy": 0.18962029963731766,
      "num_tokens": 128414268.0,
      "step": 58615
    },
    {
      "entropy": 5.72223973274231,
      "epoch": 5.817784835252084,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022066613225007347,
      "loss": 4.9652,
      "mean_token_accuracy": 0.18469696044921874,
      "num_tokens": 128425671.0,
      "step": 58620
    },
    {
      "entropy": 5.697499704360962,
      "epoch": 5.8182810639142515,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022063175002871892,
      "loss": 4.9145,
      "mean_token_accuracy": 0.20312091261148452,
      "num_tokens": 128436100.0,
      "step": 58625
    },
    {
      "entropy": 5.633808612823486,
      "epoch": 5.818777292576419,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022059736915557989,
      "loss": 4.8488,
      "mean_token_accuracy": 0.2054091736674309,
      "num_tokens": 128446781.0,
      "step": 58630
    },
    {
      "entropy": 5.668025827407837,
      "epoch": 5.819273521238586,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022056298963150884,
      "loss": 4.8955,
      "mean_token_accuracy": 0.20501996874809264,
      "num_tokens": 128458371.0,
      "step": 58635
    },
    {
      "entropy": 5.653288888931274,
      "epoch": 5.819769749900754,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002205286114573583,
      "loss": 4.8855,
      "mean_token_accuracy": 0.2025964692234993,
      "num_tokens": 128469331.0,
      "step": 58640
    },
    {
      "entropy": 5.646843385696411,
      "epoch": 5.820265978562922,
      "grad_norm": 1.375,
      "learning_rate": 0.0002204942346339807,
      "loss": 4.875,
      "mean_token_accuracy": 0.20392417013645173,
      "num_tokens": 128479899.0,
      "step": 58645
    },
    {
      "entropy": 5.695883846282959,
      "epoch": 5.820762207225089,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022045985916222883,
      "loss": 4.9313,
      "mean_token_accuracy": 0.198910391330719,
      "num_tokens": 128489153.0,
      "step": 58650
    },
    {
      "entropy": 5.719696855545044,
      "epoch": 5.821258435887257,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022042548504295483,
      "loss": 4.9653,
      "mean_token_accuracy": 0.1974313348531723,
      "num_tokens": 128500798.0,
      "step": 58655
    },
    {
      "entropy": 5.731538009643555,
      "epoch": 5.821754664549425,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002203911122770113,
      "loss": 4.9728,
      "mean_token_accuracy": 0.19152195900678634,
      "num_tokens": 128511300.0,
      "step": 58660
    },
    {
      "entropy": 5.733255624771118,
      "epoch": 5.822250893211592,
      "grad_norm": 1.546875,
      "learning_rate": 0.00022035674086525053,
      "loss": 4.891,
      "mean_token_accuracy": 0.19573720246553422,
      "num_tokens": 128522953.0,
      "step": 58665
    },
    {
      "entropy": 5.721945905685425,
      "epoch": 5.82274712187376,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022032237080852474,
      "loss": 4.9408,
      "mean_token_accuracy": 0.2006119132041931,
      "num_tokens": 128534680.0,
      "step": 58670
    },
    {
      "entropy": 5.600015544891358,
      "epoch": 5.823243350535927,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002202880021076863,
      "loss": 4.8125,
      "mean_token_accuracy": 0.21075097620487213,
      "num_tokens": 128544961.0,
      "step": 58675
    },
    {
      "entropy": 5.660052394866943,
      "epoch": 5.823739579198095,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022025363476358757,
      "loss": 4.9373,
      "mean_token_accuracy": 0.19271739572286606,
      "num_tokens": 128555505.0,
      "step": 58680
    },
    {
      "entropy": 5.671162557601929,
      "epoch": 5.824235807860262,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022021926877708077,
      "loss": 4.9308,
      "mean_token_accuracy": 0.20039832592010498,
      "num_tokens": 128565521.0,
      "step": 58685
    },
    {
      "entropy": 5.6701477527618405,
      "epoch": 5.8247320365224295,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022018490414901803,
      "loss": 4.917,
      "mean_token_accuracy": 0.1963099092245102,
      "num_tokens": 128577143.0,
      "step": 58690
    },
    {
      "entropy": 5.741491842269897,
      "epoch": 5.825228265184597,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002201505408802516,
      "loss": 4.9498,
      "mean_token_accuracy": 0.19870574921369552,
      "num_tokens": 128588773.0,
      "step": 58695
    },
    {
      "entropy": 5.716389751434326,
      "epoch": 5.8257244938467645,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022011617897163348,
      "loss": 4.9523,
      "mean_token_accuracy": 0.1996202915906906,
      "num_tokens": 128600144.0,
      "step": 58700
    },
    {
      "entropy": 5.716932868957519,
      "epoch": 5.826220722508932,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022008181842401588,
      "loss": 4.9568,
      "mean_token_accuracy": 0.19559662491083146,
      "num_tokens": 128611269.0,
      "step": 58705
    },
    {
      "entropy": 5.678540992736816,
      "epoch": 5.8267169511710994,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022004745923825082,
      "loss": 4.9331,
      "mean_token_accuracy": 0.1961693599820137,
      "num_tokens": 128622255.0,
      "step": 58710
    },
    {
      "entropy": 5.7413405418396,
      "epoch": 5.827213179833267,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022001310141519042,
      "loss": 4.9191,
      "mean_token_accuracy": 0.19987040609121323,
      "num_tokens": 128631897.0,
      "step": 58715
    },
    {
      "entropy": 5.721428108215332,
      "epoch": 5.827709408495434,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002199787449556866,
      "loss": 4.9563,
      "mean_token_accuracy": 0.2024021863937378,
      "num_tokens": 128642792.0,
      "step": 58720
    },
    {
      "entropy": 5.7013928413391115,
      "epoch": 5.828205637157602,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021994438986059134,
      "loss": 4.9274,
      "mean_token_accuracy": 0.20046203285455705,
      "num_tokens": 128653555.0,
      "step": 58725
    },
    {
      "entropy": 5.717069244384765,
      "epoch": 5.82870186581977,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021991003613075662,
      "loss": 5.0032,
      "mean_token_accuracy": 0.19196523278951644,
      "num_tokens": 128663815.0,
      "step": 58730
    },
    {
      "entropy": 5.692243909835815,
      "epoch": 5.829198094481937,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002198756837670342,
      "loss": 4.9524,
      "mean_token_accuracy": 0.18955641984939575,
      "num_tokens": 128676289.0,
      "step": 58735
    },
    {
      "entropy": 5.668915557861328,
      "epoch": 5.829694323144105,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021984133277027607,
      "loss": 4.8675,
      "mean_token_accuracy": 0.2000662699341774,
      "num_tokens": 128686632.0,
      "step": 58740
    },
    {
      "entropy": 5.665944337844849,
      "epoch": 5.830190551806273,
      "grad_norm": 1.25,
      "learning_rate": 0.00021980698314133407,
      "loss": 4.9717,
      "mean_token_accuracy": 0.1839038610458374,
      "num_tokens": 128697964.0,
      "step": 58745
    },
    {
      "entropy": 5.662736511230468,
      "epoch": 5.83068678046844,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021977263488105997,
      "loss": 4.891,
      "mean_token_accuracy": 0.18997611701488495,
      "num_tokens": 128707986.0,
      "step": 58750
    },
    {
      "entropy": 5.714537334442139,
      "epoch": 5.831183009130608,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002197382879903055,
      "loss": 4.9109,
      "mean_token_accuracy": 0.1999625489115715,
      "num_tokens": 128718314.0,
      "step": 58755
    },
    {
      "entropy": 5.6288834571838375,
      "epoch": 5.831679237792775,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002197039424699224,
      "loss": 4.8617,
      "mean_token_accuracy": 0.20075143724679947,
      "num_tokens": 128728411.0,
      "step": 58760
    },
    {
      "entropy": 5.67671389579773,
      "epoch": 5.832175466454943,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021966959832076232,
      "loss": 4.9261,
      "mean_token_accuracy": 0.1953698590397835,
      "num_tokens": 128739216.0,
      "step": 58765
    },
    {
      "entropy": 5.602358770370484,
      "epoch": 5.83267169511711,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021963525554367702,
      "loss": 4.8684,
      "mean_token_accuracy": 0.2044105887413025,
      "num_tokens": 128749655.0,
      "step": 58770
    },
    {
      "entropy": 5.672017002105713,
      "epoch": 5.8331679237792775,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002196009141395181,
      "loss": 4.873,
      "mean_token_accuracy": 0.2053074434399605,
      "num_tokens": 128760814.0,
      "step": 58775
    },
    {
      "entropy": 5.706921911239624,
      "epoch": 5.833664152441445,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021956657410913716,
      "loss": 4.9731,
      "mean_token_accuracy": 0.19904617071151734,
      "num_tokens": 128771721.0,
      "step": 58780
    },
    {
      "entropy": 5.671377182006836,
      "epoch": 5.8341603811036125,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002195322354533857,
      "loss": 4.9134,
      "mean_token_accuracy": 0.20132191181182862,
      "num_tokens": 128782729.0,
      "step": 58785
    },
    {
      "entropy": 5.658155107498169,
      "epoch": 5.83465660976578,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021949789817311523,
      "loss": 4.8322,
      "mean_token_accuracy": 0.20702512115240096,
      "num_tokens": 128793858.0,
      "step": 58790
    },
    {
      "entropy": 5.66515736579895,
      "epoch": 5.835152838427947,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021946356226917735,
      "loss": 4.919,
      "mean_token_accuracy": 0.20139754712581634,
      "num_tokens": 128804520.0,
      "step": 58795
    },
    {
      "entropy": 5.658835124969483,
      "epoch": 5.835649067090115,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021942922774242346,
      "loss": 4.9024,
      "mean_token_accuracy": 0.196315298974514,
      "num_tokens": 128816882.0,
      "step": 58800
    },
    {
      "entropy": 5.70069146156311,
      "epoch": 5.836145295752282,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00021939489459370492,
      "loss": 4.9328,
      "mean_token_accuracy": 0.19767501205205917,
      "num_tokens": 128828192.0,
      "step": 58805
    },
    {
      "entropy": 5.671513366699219,
      "epoch": 5.83664152441445,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002193605628238733,
      "loss": 4.8641,
      "mean_token_accuracy": 0.1990642786026001,
      "num_tokens": 128838249.0,
      "step": 58810
    },
    {
      "entropy": 5.728222894668579,
      "epoch": 5.837137753076618,
      "grad_norm": 1.140625,
      "learning_rate": 0.00021932623243377968,
      "loss": 4.9524,
      "mean_token_accuracy": 0.19592468142509462,
      "num_tokens": 128850334.0,
      "step": 58815
    },
    {
      "entropy": 5.685283422470093,
      "epoch": 5.837633981738785,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021929190342427559,
      "loss": 4.9092,
      "mean_token_accuracy": 0.1978769838809967,
      "num_tokens": 128862389.0,
      "step": 58820
    },
    {
      "entropy": 5.694844532012939,
      "epoch": 5.838130210400953,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021925757579621224,
      "loss": 4.9221,
      "mean_token_accuracy": 0.19652153700590133,
      "num_tokens": 128872497.0,
      "step": 58825
    },
    {
      "entropy": 5.7210118770599365,
      "epoch": 5.838626439063121,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000219223249550441,
      "loss": 4.9058,
      "mean_token_accuracy": 0.19645897895097733,
      "num_tokens": 128883338.0,
      "step": 58830
    },
    {
      "entropy": 5.695846462249756,
      "epoch": 5.839122667725288,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021918892468781288,
      "loss": 4.9375,
      "mean_token_accuracy": 0.19589650481939316,
      "num_tokens": 128894564.0,
      "step": 58835
    },
    {
      "entropy": 5.709429597854614,
      "epoch": 5.839618896387456,
      "grad_norm": 1.375,
      "learning_rate": 0.00021915460120917917,
      "loss": 4.9512,
      "mean_token_accuracy": 0.19808417558670044,
      "num_tokens": 128905809.0,
      "step": 58840
    },
    {
      "entropy": 5.750549983978272,
      "epoch": 5.840115125049623,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00021912027911539107,
      "loss": 5.043,
      "mean_token_accuracy": 0.18473837822675704,
      "num_tokens": 128917668.0,
      "step": 58845
    },
    {
      "entropy": 5.729784679412842,
      "epoch": 5.8406113537117905,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021908595840729956,
      "loss": 4.9824,
      "mean_token_accuracy": 0.1940373256802559,
      "num_tokens": 128928766.0,
      "step": 58850
    },
    {
      "entropy": 5.690691089630127,
      "epoch": 5.841107582373958,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002190516390857558,
      "loss": 4.9622,
      "mean_token_accuracy": 0.1891911581158638,
      "num_tokens": 128939473.0,
      "step": 58855
    },
    {
      "entropy": 5.660176849365234,
      "epoch": 5.8416038110361255,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021901732115161087,
      "loss": 4.8718,
      "mean_token_accuracy": 0.203238607943058,
      "num_tokens": 128950019.0,
      "step": 58860
    },
    {
      "entropy": 5.731686067581177,
      "epoch": 5.842100039698293,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002189830046057158,
      "loss": 4.9972,
      "mean_token_accuracy": 0.19627575874328612,
      "num_tokens": 128962556.0,
      "step": 58865
    },
    {
      "entropy": 5.716835975646973,
      "epoch": 5.84259626836046,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021894868944892143,
      "loss": 4.9399,
      "mean_token_accuracy": 0.19608516991138458,
      "num_tokens": 128972626.0,
      "step": 58870
    },
    {
      "entropy": 5.719924449920654,
      "epoch": 5.843092497022628,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021891437568207883,
      "loss": 4.9204,
      "mean_token_accuracy": 0.1936043158173561,
      "num_tokens": 128983089.0,
      "step": 58875
    },
    {
      "entropy": 5.734201669692993,
      "epoch": 5.843588725684795,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021888006330603876,
      "loss": 5.0163,
      "mean_token_accuracy": 0.19429763406515121,
      "num_tokens": 128994395.0,
      "step": 58880
    },
    {
      "entropy": 5.753650140762329,
      "epoch": 5.844084954346963,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021884575232165222,
      "loss": 4.9773,
      "mean_token_accuracy": 0.20833645462989808,
      "num_tokens": 129005985.0,
      "step": 58885
    },
    {
      "entropy": 5.698477363586425,
      "epoch": 5.84458118300913,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002188114427297701,
      "loss": 4.985,
      "mean_token_accuracy": 0.19062856435775757,
      "num_tokens": 129017662.0,
      "step": 58890
    },
    {
      "entropy": 5.695486545562744,
      "epoch": 5.845077411671298,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021877713453124316,
      "loss": 4.8738,
      "mean_token_accuracy": 0.20056526660919188,
      "num_tokens": 129028545.0,
      "step": 58895
    },
    {
      "entropy": 5.707680749893188,
      "epoch": 5.845573640333466,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000218742827726922,
      "loss": 4.9688,
      "mean_token_accuracy": 0.19793115109205245,
      "num_tokens": 129040056.0,
      "step": 58900
    },
    {
      "entropy": 5.685453319549561,
      "epoch": 5.846069868995633,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002187085223176576,
      "loss": 4.9602,
      "mean_token_accuracy": 0.19470561146736146,
      "num_tokens": 129049925.0,
      "step": 58905
    },
    {
      "entropy": 5.798084306716919,
      "epoch": 5.846566097657801,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021867421830430045,
      "loss": 4.964,
      "mean_token_accuracy": 0.19840373545885087,
      "num_tokens": 129062347.0,
      "step": 58910
    },
    {
      "entropy": 5.719425344467163,
      "epoch": 5.847062326319969,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002186399156877014,
      "loss": 4.8971,
      "mean_token_accuracy": 0.2096584692597389,
      "num_tokens": 129074036.0,
      "step": 58915
    },
    {
      "entropy": 5.726748895645142,
      "epoch": 5.847558554982136,
      "grad_norm": 1.125,
      "learning_rate": 0.0002186056144687109,
      "loss": 4.9504,
      "mean_token_accuracy": 0.20135612189769744,
      "num_tokens": 129086622.0,
      "step": 58920
    },
    {
      "entropy": 5.686344718933105,
      "epoch": 5.8480547836443035,
      "grad_norm": 1.25,
      "learning_rate": 0.00021857131464817975,
      "loss": 4.9019,
      "mean_token_accuracy": 0.19580177813768387,
      "num_tokens": 129097552.0,
      "step": 58925
    },
    {
      "entropy": 5.744256591796875,
      "epoch": 5.848551012306471,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021853701622695832,
      "loss": 4.9972,
      "mean_token_accuracy": 0.19344762116670608,
      "num_tokens": 129109292.0,
      "step": 58930
    },
    {
      "entropy": 5.740137386322021,
      "epoch": 5.8490472409686385,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021850271920589731,
      "loss": 4.9348,
      "mean_token_accuracy": 0.19321485459804535,
      "num_tokens": 129120228.0,
      "step": 58935
    },
    {
      "entropy": 5.688745641708374,
      "epoch": 5.849543469630806,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021846842358584702,
      "loss": 4.9356,
      "mean_token_accuracy": 0.19906515181064605,
      "num_tokens": 129130956.0,
      "step": 58940
    },
    {
      "entropy": 5.722800779342651,
      "epoch": 5.850039698292973,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021843412936765804,
      "loss": 4.943,
      "mean_token_accuracy": 0.19556229412555695,
      "num_tokens": 129141991.0,
      "step": 58945
    },
    {
      "entropy": 5.704760646820068,
      "epoch": 5.850535926955141,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002183998365521808,
      "loss": 5.0094,
      "mean_token_accuracy": 0.18917498737573624,
      "num_tokens": 129154101.0,
      "step": 58950
    },
    {
      "entropy": 5.668781328201294,
      "epoch": 5.851032155617308,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021836554514026568,
      "loss": 4.8771,
      "mean_token_accuracy": 0.1999330997467041,
      "num_tokens": 129164668.0,
      "step": 58955
    },
    {
      "entropy": 5.7489231586456295,
      "epoch": 5.851528384279476,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021833125513276297,
      "loss": 5.0382,
      "mean_token_accuracy": 0.19254676550626754,
      "num_tokens": 129175125.0,
      "step": 58960
    },
    {
      "entropy": 5.770584297180176,
      "epoch": 5.852024612941643,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000218296966530523,
      "loss": 4.9506,
      "mean_token_accuracy": 0.19021288752555848,
      "num_tokens": 129185859.0,
      "step": 58965
    },
    {
      "entropy": 5.7195188999176025,
      "epoch": 5.852520841603811,
      "grad_norm": 1.25,
      "learning_rate": 0.00021826267933439602,
      "loss": 4.8954,
      "mean_token_accuracy": 0.20246095657348634,
      "num_tokens": 129196087.0,
      "step": 58970
    },
    {
      "entropy": 5.734885549545288,
      "epoch": 5.853017070265978,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021822839354523245,
      "loss": 4.9585,
      "mean_token_accuracy": 0.2021318107843399,
      "num_tokens": 129206967.0,
      "step": 58975
    },
    {
      "entropy": 5.701860618591309,
      "epoch": 5.853513298928146,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021819410916388234,
      "loss": 4.9575,
      "mean_token_accuracy": 0.20133223831653596,
      "num_tokens": 129217812.0,
      "step": 58980
    },
    {
      "entropy": 5.672459316253662,
      "epoch": 5.854009527590313,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002181598261911959,
      "loss": 4.9413,
      "mean_token_accuracy": 0.19351304620504378,
      "num_tokens": 129227321.0,
      "step": 58985
    },
    {
      "entropy": 5.689752292633057,
      "epoch": 5.854505756252481,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021812554462802337,
      "loss": 4.9277,
      "mean_token_accuracy": 0.20081242620944978,
      "num_tokens": 129238601.0,
      "step": 58990
    },
    {
      "entropy": 5.7211004257202145,
      "epoch": 5.855001984914649,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021809126447521466,
      "loss": 4.9551,
      "mean_token_accuracy": 0.19819144606590272,
      "num_tokens": 129250879.0,
      "step": 58995
    },
    {
      "entropy": 5.656569099426269,
      "epoch": 5.8554982135768165,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021805698573361997,
      "loss": 4.8099,
      "mean_token_accuracy": 0.2081671193242073,
      "num_tokens": 129261642.0,
      "step": 59000
    },
    {
      "entropy": 5.706584072113037,
      "epoch": 5.855994442238984,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021802270840408944,
      "loss": 4.9257,
      "mean_token_accuracy": 0.1953839525580406,
      "num_tokens": 129272493.0,
      "step": 59005
    },
    {
      "entropy": 5.707567310333252,
      "epoch": 5.8564906709011515,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002179884324874729,
      "loss": 4.9154,
      "mean_token_accuracy": 0.1998950645327568,
      "num_tokens": 129282855.0,
      "step": 59010
    },
    {
      "entropy": 5.721368503570557,
      "epoch": 5.856986899563319,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021795415798462042,
      "loss": 4.95,
      "mean_token_accuracy": 0.20079316943883896,
      "num_tokens": 129294684.0,
      "step": 59015
    },
    {
      "entropy": 5.680648756027222,
      "epoch": 5.8574831282254864,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021791988489638188,
      "loss": 4.8722,
      "mean_token_accuracy": 0.1991426467895508,
      "num_tokens": 129304468.0,
      "step": 59020
    },
    {
      "entropy": 5.68699893951416,
      "epoch": 5.857979356887654,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002178856132236071,
      "loss": 4.9061,
      "mean_token_accuracy": 0.2000993847846985,
      "num_tokens": 129315526.0,
      "step": 59025
    },
    {
      "entropy": 5.690967559814453,
      "epoch": 5.858475585549821,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021785134296714622,
      "loss": 4.8922,
      "mean_token_accuracy": 0.19988665729761124,
      "num_tokens": 129327120.0,
      "step": 59030
    },
    {
      "entropy": 5.716082382202148,
      "epoch": 5.858971814211989,
      "grad_norm": 1.375,
      "learning_rate": 0.00021781707412784875,
      "loss": 4.8638,
      "mean_token_accuracy": 0.20910332798957826,
      "num_tokens": 129337473.0,
      "step": 59035
    },
    {
      "entropy": 5.647953748703003,
      "epoch": 5.859468042874156,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002177828067065647,
      "loss": 4.869,
      "mean_token_accuracy": 0.200468747317791,
      "num_tokens": 129348563.0,
      "step": 59040
    },
    {
      "entropy": 5.680612945556641,
      "epoch": 5.859964271536324,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021774854070414373,
      "loss": 4.9509,
      "mean_token_accuracy": 0.1966561034321785,
      "num_tokens": 129358970.0,
      "step": 59045
    },
    {
      "entropy": 5.678194665908814,
      "epoch": 5.860460500198491,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021771427612143562,
      "loss": 4.9283,
      "mean_token_accuracy": 0.19772442430257797,
      "num_tokens": 129369998.0,
      "step": 59050
    },
    {
      "entropy": 5.708745098114013,
      "epoch": 5.860956728860659,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002176800129592899,
      "loss": 5.0057,
      "mean_token_accuracy": 0.19181694090366364,
      "num_tokens": 129381833.0,
      "step": 59055
    },
    {
      "entropy": 5.72815318107605,
      "epoch": 5.861452957522826,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021764575121855644,
      "loss": 4.86,
      "mean_token_accuracy": 0.20808594524860383,
      "num_tokens": 129393032.0,
      "step": 59060
    },
    {
      "entropy": 5.75811014175415,
      "epoch": 5.861949186184994,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021761149090008476,
      "loss": 4.9758,
      "mean_token_accuracy": 0.18131479918956755,
      "num_tokens": 129405347.0,
      "step": 59065
    },
    {
      "entropy": 5.695041036605835,
      "epoch": 5.862445414847161,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021757723200472445,
      "loss": 4.8353,
      "mean_token_accuracy": 0.20103423297405243,
      "num_tokens": 129417576.0,
      "step": 59070
    },
    {
      "entropy": 5.764144802093506,
      "epoch": 5.862941643509329,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021754297453332506,
      "loss": 4.9658,
      "mean_token_accuracy": 0.1965098038315773,
      "num_tokens": 129428843.0,
      "step": 59075
    },
    {
      "entropy": 5.703233146667481,
      "epoch": 5.863437872171497,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021750871848673608,
      "loss": 5.0047,
      "mean_token_accuracy": 0.19730053842067719,
      "num_tokens": 129440141.0,
      "step": 59080
    },
    {
      "entropy": 5.664038944244385,
      "epoch": 5.8639341008336645,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021747446386580693,
      "loss": 4.8672,
      "mean_token_accuracy": 0.20201416164636612,
      "num_tokens": 129451159.0,
      "step": 59085
    },
    {
      "entropy": 5.701893711090088,
      "epoch": 5.864430329495832,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002174402106713872,
      "loss": 4.9266,
      "mean_token_accuracy": 0.20324042737483977,
      "num_tokens": 129462000.0,
      "step": 59090
    },
    {
      "entropy": 5.721996736526489,
      "epoch": 5.8649265581579995,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021740595890432625,
      "loss": 4.9316,
      "mean_token_accuracy": 0.19706047028303147,
      "num_tokens": 129474271.0,
      "step": 59095
    },
    {
      "entropy": 5.736874389648437,
      "epoch": 5.865422786820167,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021737170856547333,
      "loss": 4.8842,
      "mean_token_accuracy": 0.1945027306675911,
      "num_tokens": 129484104.0,
      "step": 59100
    },
    {
      "entropy": 5.720763874053955,
      "epoch": 5.865919015482334,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021733745965567792,
      "loss": 4.9687,
      "mean_token_accuracy": 0.19560340344905852,
      "num_tokens": 129494082.0,
      "step": 59105
    },
    {
      "entropy": 5.646195793151856,
      "epoch": 5.866415244144502,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002173032121757892,
      "loss": 4.9223,
      "mean_token_accuracy": 0.19626785963773727,
      "num_tokens": 129505332.0,
      "step": 59110
    },
    {
      "entropy": 5.68068118095398,
      "epoch": 5.866911472806669,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021726896612665647,
      "loss": 4.8777,
      "mean_token_accuracy": 0.20247475653886796,
      "num_tokens": 129517032.0,
      "step": 59115
    },
    {
      "entropy": 5.705804252624512,
      "epoch": 5.867407701468837,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021723472150912904,
      "loss": 4.9415,
      "mean_token_accuracy": 0.1988467678427696,
      "num_tokens": 129527930.0,
      "step": 59120
    },
    {
      "entropy": 5.709426021575927,
      "epoch": 5.867903930131004,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021720047832405603,
      "loss": 4.9982,
      "mean_token_accuracy": 0.19739652127027513,
      "num_tokens": 129539986.0,
      "step": 59125
    },
    {
      "entropy": 5.727176094055176,
      "epoch": 5.868400158793172,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021716623657228664,
      "loss": 4.909,
      "mean_token_accuracy": 0.20172260701656342,
      "num_tokens": 129550158.0,
      "step": 59130
    },
    {
      "entropy": 5.677068281173706,
      "epoch": 5.868896387455339,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021713199625466995,
      "loss": 4.9081,
      "mean_token_accuracy": 0.20270522832870483,
      "num_tokens": 129561616.0,
      "step": 59135
    },
    {
      "entropy": 5.671291446685791,
      "epoch": 5.869392616117507,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021709775737205513,
      "loss": 4.8851,
      "mean_token_accuracy": 0.2042611926794052,
      "num_tokens": 129572410.0,
      "step": 59140
    },
    {
      "entropy": 5.740066766738892,
      "epoch": 5.869888844779674,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000217063519925291,
      "loss": 4.9708,
      "mean_token_accuracy": 0.1979313999414444,
      "num_tokens": 129582990.0,
      "step": 59145
    },
    {
      "entropy": 5.763862562179566,
      "epoch": 5.870385073441842,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00021702928391522676,
      "loss": 5.0353,
      "mean_token_accuracy": 0.198440845310688,
      "num_tokens": 129594071.0,
      "step": 59150
    },
    {
      "entropy": 5.67559027671814,
      "epoch": 5.870881302104009,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021699504934271147,
      "loss": 4.9441,
      "mean_token_accuracy": 0.1927391842007637,
      "num_tokens": 129603983.0,
      "step": 59155
    },
    {
      "entropy": 5.769874620437622,
      "epoch": 5.871377530766177,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021696081620859386,
      "loss": 4.9906,
      "mean_token_accuracy": 0.19573559761047363,
      "num_tokens": 129615652.0,
      "step": 59160
    },
    {
      "entropy": 5.688311338424683,
      "epoch": 5.871873759428345,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021692658451372306,
      "loss": 4.886,
      "mean_token_accuracy": 0.20241330116987227,
      "num_tokens": 129627305.0,
      "step": 59165
    },
    {
      "entropy": 5.659388160705566,
      "epoch": 5.8723699880905125,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021689235425894778,
      "loss": 4.8631,
      "mean_token_accuracy": 0.2076686829328537,
      "num_tokens": 129637770.0,
      "step": 59170
    },
    {
      "entropy": 5.679310989379883,
      "epoch": 5.87286621675268,
      "grad_norm": 1.25,
      "learning_rate": 0.00021685812544511678,
      "loss": 4.8987,
      "mean_token_accuracy": 0.19746416062116623,
      "num_tokens": 129648852.0,
      "step": 59175
    },
    {
      "entropy": 5.709510278701782,
      "epoch": 5.873362445414847,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021682389807307906,
      "loss": 4.961,
      "mean_token_accuracy": 0.1995352327823639,
      "num_tokens": 129660596.0,
      "step": 59180
    },
    {
      "entropy": 5.743139219284058,
      "epoch": 5.873858674077015,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002167896721436834,
      "loss": 4.9671,
      "mean_token_accuracy": 0.19801629334688187,
      "num_tokens": 129671274.0,
      "step": 59185
    },
    {
      "entropy": 5.755550527572632,
      "epoch": 5.874354902739182,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002167554476577784,
      "loss": 4.9782,
      "mean_token_accuracy": 0.19862630367279052,
      "num_tokens": 129682419.0,
      "step": 59190
    },
    {
      "entropy": 5.693540143966675,
      "epoch": 5.87485113140135,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021672122461621274,
      "loss": 4.9168,
      "mean_token_accuracy": 0.1995275065302849,
      "num_tokens": 129693322.0,
      "step": 59195
    },
    {
      "entropy": 5.682809352874756,
      "epoch": 5.875347360063517,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002166870030198352,
      "loss": 4.886,
      "mean_token_accuracy": 0.2019612729549408,
      "num_tokens": 129704844.0,
      "step": 59200
    },
    {
      "entropy": 5.735593509674072,
      "epoch": 5.875843588725685,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021665278286949418,
      "loss": 4.9488,
      "mean_token_accuracy": 0.1957168012857437,
      "num_tokens": 129715420.0,
      "step": 59205
    },
    {
      "entropy": 5.69718828201294,
      "epoch": 5.876339817387852,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021661856416603858,
      "loss": 4.8969,
      "mean_token_accuracy": 0.20077372193336487,
      "num_tokens": 129726622.0,
      "step": 59210
    },
    {
      "entropy": 5.712718200683594,
      "epoch": 5.87683604605002,
      "grad_norm": 1.375,
      "learning_rate": 0.00021658434691031674,
      "loss": 4.9587,
      "mean_token_accuracy": 0.19979117810726166,
      "num_tokens": 129737960.0,
      "step": 59215
    },
    {
      "entropy": 5.67739577293396,
      "epoch": 5.877332274712187,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021655013110317723,
      "loss": 4.95,
      "mean_token_accuracy": 0.1970570296049118,
      "num_tokens": 129749110.0,
      "step": 59220
    },
    {
      "entropy": 5.676516199111939,
      "epoch": 5.877828503374355,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002165159167454685,
      "loss": 4.9347,
      "mean_token_accuracy": 0.20476323664188384,
      "num_tokens": 129758764.0,
      "step": 59225
    },
    {
      "entropy": 5.650768899917603,
      "epoch": 5.878324732036522,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021648170383803907,
      "loss": 4.8653,
      "mean_token_accuracy": 0.20158150047063828,
      "num_tokens": 129769795.0,
      "step": 59230
    },
    {
      "entropy": 5.724016809463501,
      "epoch": 5.87882096069869,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021644749238173717,
      "loss": 4.9357,
      "mean_token_accuracy": 0.20389060974121093,
      "num_tokens": 129781437.0,
      "step": 59235
    },
    {
      "entropy": 5.705437898635864,
      "epoch": 5.879317189360857,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021641328237741137,
      "loss": 4.8962,
      "mean_token_accuracy": 0.20086769610643387,
      "num_tokens": 129791759.0,
      "step": 59240
    },
    {
      "entropy": 5.75815715789795,
      "epoch": 5.879813418023025,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021637907382590987,
      "loss": 4.942,
      "mean_token_accuracy": 0.20084756165742873,
      "num_tokens": 129803752.0,
      "step": 59245
    },
    {
      "entropy": 5.68695912361145,
      "epoch": 5.880309646685193,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002163448667280811,
      "loss": 4.8993,
      "mean_token_accuracy": 0.1963734969496727,
      "num_tokens": 129813800.0,
      "step": 59250
    },
    {
      "entropy": 5.6843505859375,
      "epoch": 5.88080587534736,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021631066108477326,
      "loss": 4.9711,
      "mean_token_accuracy": 0.1957295447587967,
      "num_tokens": 129825859.0,
      "step": 59255
    },
    {
      "entropy": 5.6881519794464115,
      "epoch": 5.881302104009528,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002162764568968345,
      "loss": 4.9144,
      "mean_token_accuracy": 0.19724965691566468,
      "num_tokens": 129836858.0,
      "step": 59260
    },
    {
      "entropy": 5.6855425357818605,
      "epoch": 5.881798332671695,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000216242254165113,
      "loss": 4.9068,
      "mean_token_accuracy": 0.2047008067369461,
      "num_tokens": 129849061.0,
      "step": 59265
    },
    {
      "entropy": 5.759973239898682,
      "epoch": 5.882294561333863,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021620805289045708,
      "loss": 4.9313,
      "mean_token_accuracy": 0.2026597887277603,
      "num_tokens": 129859939.0,
      "step": 59270
    },
    {
      "entropy": 5.695826387405395,
      "epoch": 5.88279078999603,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021617385307371478,
      "loss": 4.9012,
      "mean_token_accuracy": 0.2013796165585518,
      "num_tokens": 129871004.0,
      "step": 59275
    },
    {
      "entropy": 5.680336284637451,
      "epoch": 5.883287018658198,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021613965471573417,
      "loss": 4.9086,
      "mean_token_accuracy": 0.19978829175233842,
      "num_tokens": 129882645.0,
      "step": 59280
    },
    {
      "entropy": 5.694992637634277,
      "epoch": 5.883783247320365,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021610545781736325,
      "loss": 4.8762,
      "mean_token_accuracy": 0.20389530658721924,
      "num_tokens": 129893405.0,
      "step": 59285
    },
    {
      "entropy": 5.655923175811767,
      "epoch": 5.884279475982533,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021607126237945,
      "loss": 4.9514,
      "mean_token_accuracy": 0.19148722290992737,
      "num_tokens": 129905148.0,
      "step": 59290
    },
    {
      "entropy": 5.666068887710571,
      "epoch": 5.8847757046447,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021603706840284253,
      "loss": 4.8606,
      "mean_token_accuracy": 0.2038567766547203,
      "num_tokens": 129916880.0,
      "step": 59295
    },
    {
      "entropy": 5.7569169998168945,
      "epoch": 5.885271933306868,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021600287588838875,
      "loss": 5.0,
      "mean_token_accuracy": 0.1906591013073921,
      "num_tokens": 129927233.0,
      "step": 59300
    },
    {
      "entropy": 5.671768474578857,
      "epoch": 5.885768161969035,
      "grad_norm": 1.046875,
      "learning_rate": 0.00021596868483693654,
      "loss": 4.9197,
      "mean_token_accuracy": 0.19911676943302153,
      "num_tokens": 129939688.0,
      "step": 59305
    },
    {
      "entropy": 5.731566524505615,
      "epoch": 5.886264390631203,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002159344952493336,
      "loss": 4.9413,
      "mean_token_accuracy": 0.19871903955936432,
      "num_tokens": 129950053.0,
      "step": 59310
    },
    {
      "entropy": 5.72462420463562,
      "epoch": 5.88676061929337,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021590030712642807,
      "loss": 4.9851,
      "mean_token_accuracy": 0.19389821141958236,
      "num_tokens": 129961060.0,
      "step": 59315
    },
    {
      "entropy": 5.71266918182373,
      "epoch": 5.887256847955538,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021586612046906744,
      "loss": 4.9163,
      "mean_token_accuracy": 0.20258814543485643,
      "num_tokens": 129971260.0,
      "step": 59320
    },
    {
      "entropy": 5.6778217315673825,
      "epoch": 5.887753076617705,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002158319352780997,
      "loss": 4.929,
      "mean_token_accuracy": 0.19892264753580094,
      "num_tokens": 129981093.0,
      "step": 59325
    },
    {
      "entropy": 5.690774250030517,
      "epoch": 5.888249305279873,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021579775155437236,
      "loss": 4.9114,
      "mean_token_accuracy": 0.20072823166847228,
      "num_tokens": 129990983.0,
      "step": 59330
    },
    {
      "entropy": 5.699438381195068,
      "epoch": 5.888745533942041,
      "grad_norm": 1.5,
      "learning_rate": 0.00021576356929873332,
      "loss": 4.9084,
      "mean_token_accuracy": 0.19663426727056504,
      "num_tokens": 130001492.0,
      "step": 59335
    },
    {
      "entropy": 5.698934841156006,
      "epoch": 5.889241762604208,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021572938851203012,
      "loss": 4.8887,
      "mean_token_accuracy": 0.19852905869483947,
      "num_tokens": 130012171.0,
      "step": 59340
    },
    {
      "entropy": 5.647183227539062,
      "epoch": 5.889737991266376,
      "grad_norm": 1.140625,
      "learning_rate": 0.00021569520919511033,
      "loss": 4.8657,
      "mean_token_accuracy": 0.20931769162416458,
      "num_tokens": 130025232.0,
      "step": 59345
    },
    {
      "entropy": 5.658660793304444,
      "epoch": 5.890234219928543,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021566103134882148,
      "loss": 4.9199,
      "mean_token_accuracy": 0.20060172975063323,
      "num_tokens": 130036447.0,
      "step": 59350
    },
    {
      "entropy": 5.742452478408813,
      "epoch": 5.890730448590711,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002156268549740113,
      "loss": 5.0334,
      "mean_token_accuracy": 0.19518833756446838,
      "num_tokens": 130047105.0,
      "step": 59355
    },
    {
      "entropy": 5.69251389503479,
      "epoch": 5.891226677252878,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021559268007152716,
      "loss": 4.9002,
      "mean_token_accuracy": 0.19999437481164933,
      "num_tokens": 130058090.0,
      "step": 59360
    },
    {
      "entropy": 5.691939210891723,
      "epoch": 5.891722905915046,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002155585066422165,
      "loss": 4.9528,
      "mean_token_accuracy": 0.1910791739821434,
      "num_tokens": 130068961.0,
      "step": 59365
    },
    {
      "entropy": 5.706874799728394,
      "epoch": 5.892219134577213,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002155243346869269,
      "loss": 4.9306,
      "mean_token_accuracy": 0.19788529425859452,
      "num_tokens": 130079220.0,
      "step": 59370
    },
    {
      "entropy": 5.712114000320435,
      "epoch": 5.892715363239381,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021549016420650553,
      "loss": 4.9562,
      "mean_token_accuracy": 0.20040934681892394,
      "num_tokens": 130090638.0,
      "step": 59375
    },
    {
      "entropy": 5.7096199035644535,
      "epoch": 5.893211591901548,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021545599520179982,
      "loss": 4.8727,
      "mean_token_accuracy": 0.19825115352869033,
      "num_tokens": 130100543.0,
      "step": 59380
    },
    {
      "entropy": 5.748258209228515,
      "epoch": 5.893707820563716,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002154218276736572,
      "loss": 4.912,
      "mean_token_accuracy": 0.19680766463279725,
      "num_tokens": 130109800.0,
      "step": 59385
    },
    {
      "entropy": 5.760344696044922,
      "epoch": 5.894204049225883,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002153876616229249,
      "loss": 5.0285,
      "mean_token_accuracy": 0.1948051080107689,
      "num_tokens": 130119696.0,
      "step": 59390
    },
    {
      "entropy": 5.697346496582031,
      "epoch": 5.894700277888051,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021535349705045008,
      "loss": 4.9359,
      "mean_token_accuracy": 0.19749054908752442,
      "num_tokens": 130129500.0,
      "step": 59395
    },
    {
      "entropy": 5.769998836517334,
      "epoch": 5.895196506550218,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021531933395708004,
      "loss": 4.9788,
      "mean_token_accuracy": 0.20253632664680482,
      "num_tokens": 130140946.0,
      "step": 59400
    },
    {
      "entropy": 5.656508541107177,
      "epoch": 5.895692735212386,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002152851723436619,
      "loss": 4.8655,
      "mean_token_accuracy": 0.19832761734724044,
      "num_tokens": 130150259.0,
      "step": 59405
    },
    {
      "entropy": 5.6773268699646,
      "epoch": 5.896188963874553,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021525101221104277,
      "loss": 4.8689,
      "mean_token_accuracy": 0.197517092525959,
      "num_tokens": 130162614.0,
      "step": 59410
    },
    {
      "entropy": 5.635092258453369,
      "epoch": 5.8966851925367205,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021521685356006988,
      "loss": 4.9068,
      "mean_token_accuracy": 0.2002578407526016,
      "num_tokens": 130174124.0,
      "step": 59415
    },
    {
      "entropy": 5.697138023376465,
      "epoch": 5.897181421198889,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021518269639159016,
      "loss": 4.892,
      "mean_token_accuracy": 0.19884766042232513,
      "num_tokens": 130184898.0,
      "step": 59420
    },
    {
      "entropy": 5.707520961761475,
      "epoch": 5.897677649861056,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002151485407064507,
      "loss": 4.9058,
      "mean_token_accuracy": 0.19693934470415114,
      "num_tokens": 130194477.0,
      "step": 59425
    },
    {
      "entropy": 5.718619728088379,
      "epoch": 5.898173878523224,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00021511438650549842,
      "loss": 4.927,
      "mean_token_accuracy": 0.2035319373011589,
      "num_tokens": 130205808.0,
      "step": 59430
    },
    {
      "entropy": 5.69550256729126,
      "epoch": 5.898670107185391,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021508023378958037,
      "loss": 4.9099,
      "mean_token_accuracy": 0.19746458679437637,
      "num_tokens": 130217295.0,
      "step": 59435
    },
    {
      "entropy": 5.673517417907715,
      "epoch": 5.899166335847559,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021504608255954328,
      "loss": 4.9164,
      "mean_token_accuracy": 0.2021649405360222,
      "num_tokens": 130228938.0,
      "step": 59440
    },
    {
      "entropy": 5.667645025253296,
      "epoch": 5.899662564509726,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002150119328162342,
      "loss": 4.9137,
      "mean_token_accuracy": 0.19702220410108567,
      "num_tokens": 130239417.0,
      "step": 59445
    },
    {
      "entropy": 5.73737850189209,
      "epoch": 5.900158793171894,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002149777845605,
      "loss": 5.0139,
      "mean_token_accuracy": 0.1940672814846039,
      "num_tokens": 130250221.0,
      "step": 59450
    },
    {
      "entropy": 5.686158275604248,
      "epoch": 5.900655021834061,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021494363779318737,
      "loss": 4.8882,
      "mean_token_accuracy": 0.20017136186361312,
      "num_tokens": 130262304.0,
      "step": 59455
    },
    {
      "entropy": 5.711582517623901,
      "epoch": 5.901151250496229,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021490949251514314,
      "loss": 4.9926,
      "mean_token_accuracy": 0.19203971922397614,
      "num_tokens": 130273815.0,
      "step": 59460
    },
    {
      "entropy": 5.672995662689209,
      "epoch": 5.901647479158396,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021487534872721405,
      "loss": 4.9418,
      "mean_token_accuracy": 0.19942341446876527,
      "num_tokens": 130285623.0,
      "step": 59465
    },
    {
      "entropy": 5.731771993637085,
      "epoch": 5.902143707820564,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021484120643024658,
      "loss": 4.9223,
      "mean_token_accuracy": 0.19257931858301164,
      "num_tokens": 130295641.0,
      "step": 59470
    },
    {
      "entropy": 5.606039810180664,
      "epoch": 5.902639936482731,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002148070656250876,
      "loss": 4.8395,
      "mean_token_accuracy": 0.2114996075630188,
      "num_tokens": 130306589.0,
      "step": 59475
    },
    {
      "entropy": 5.65090503692627,
      "epoch": 5.903136165144899,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021477292631258383,
      "loss": 4.8674,
      "mean_token_accuracy": 0.2060799404978752,
      "num_tokens": 130318170.0,
      "step": 59480
    },
    {
      "entropy": 5.692888498306274,
      "epoch": 5.903632393807066,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021473878849358164,
      "loss": 4.9227,
      "mean_token_accuracy": 0.202184097468853,
      "num_tokens": 130329727.0,
      "step": 59485
    },
    {
      "entropy": 5.770032072067261,
      "epoch": 5.9041286224692335,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021470465216892758,
      "loss": 5.0343,
      "mean_token_accuracy": 0.19610034674406052,
      "num_tokens": 130339984.0,
      "step": 59490
    },
    {
      "entropy": 5.689475202560425,
      "epoch": 5.904624851131401,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021467051733946836,
      "loss": 4.8894,
      "mean_token_accuracy": 0.19986572861671448,
      "num_tokens": 130351247.0,
      "step": 59495
    },
    {
      "entropy": 5.66401777267456,
      "epoch": 5.9051210797935685,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002146363840060501,
      "loss": 4.8727,
      "mean_token_accuracy": 0.2028010979294777,
      "num_tokens": 130362717.0,
      "step": 59500
    },
    {
      "entropy": 5.6965248584747314,
      "epoch": 5.905617308455737,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021460225216951968,
      "loss": 4.9601,
      "mean_token_accuracy": 0.201768858730793,
      "num_tokens": 130373503.0,
      "step": 59505
    },
    {
      "entropy": 5.67793755531311,
      "epoch": 5.906113537117903,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002145681218307231,
      "loss": 4.8787,
      "mean_token_accuracy": 0.2071689173579216,
      "num_tokens": 130384227.0,
      "step": 59510
    },
    {
      "entropy": 5.7021064281463625,
      "epoch": 5.906609765780072,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021453399299050697,
      "loss": 4.9253,
      "mean_token_accuracy": 0.1969575047492981,
      "num_tokens": 130393532.0,
      "step": 59515
    },
    {
      "entropy": 5.6174102306365965,
      "epoch": 5.907105994442239,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021449986564971746,
      "loss": 4.8012,
      "mean_token_accuracy": 0.20233005285263062,
      "num_tokens": 130404151.0,
      "step": 59520
    },
    {
      "entropy": 5.663174486160278,
      "epoch": 5.907602223104407,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021446573980920085,
      "loss": 4.904,
      "mean_token_accuracy": 0.1953785553574562,
      "num_tokens": 130414193.0,
      "step": 59525
    },
    {
      "entropy": 5.695481491088867,
      "epoch": 5.908098451766574,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002144316154698036,
      "loss": 4.9862,
      "mean_token_accuracy": 0.18857215195894242,
      "num_tokens": 130425739.0,
      "step": 59530
    },
    {
      "entropy": 5.757262516021728,
      "epoch": 5.908594680428742,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021439749263237179,
      "loss": 5.0153,
      "mean_token_accuracy": 0.1908609002828598,
      "num_tokens": 130436422.0,
      "step": 59535
    },
    {
      "entropy": 5.70394196510315,
      "epoch": 5.909090909090909,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021436337129775146,
      "loss": 4.9144,
      "mean_token_accuracy": 0.2028014600276947,
      "num_tokens": 130445902.0,
      "step": 59540
    },
    {
      "entropy": 5.746118688583374,
      "epoch": 5.909587137753077,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021432925146678893,
      "loss": 5.0,
      "mean_token_accuracy": 0.1921256422996521,
      "num_tokens": 130458284.0,
      "step": 59545
    },
    {
      "entropy": 5.785829401016235,
      "epoch": 5.910083366415244,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021429513314033023,
      "loss": 5.0309,
      "mean_token_accuracy": 0.1857544645667076,
      "num_tokens": 130469128.0,
      "step": 59550
    },
    {
      "entropy": 5.7030257225036625,
      "epoch": 5.910579595077412,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021426101631922135,
      "loss": 4.9294,
      "mean_token_accuracy": 0.20411998331546782,
      "num_tokens": 130480371.0,
      "step": 59555
    },
    {
      "entropy": 5.750036954879761,
      "epoch": 5.911075823739579,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021422690100430836,
      "loss": 5.0167,
      "mean_token_accuracy": 0.1974821850657463,
      "num_tokens": 130492308.0,
      "step": 59560
    },
    {
      "entropy": 5.7220072746276855,
      "epoch": 5.9115720524017465,
      "grad_norm": 1.640625,
      "learning_rate": 0.00021419278719643741,
      "loss": 4.919,
      "mean_token_accuracy": 0.196364326775074,
      "num_tokens": 130503358.0,
      "step": 59565
    },
    {
      "entropy": 5.663523006439209,
      "epoch": 5.912068281063914,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002141586748964542,
      "loss": 4.8159,
      "mean_token_accuracy": 0.20791595429182053,
      "num_tokens": 130513152.0,
      "step": 59570
    },
    {
      "entropy": 5.6745781898498535,
      "epoch": 5.9125645097260815,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021412456410520482,
      "loss": 4.9476,
      "mean_token_accuracy": 0.19767093360424043,
      "num_tokens": 130523919.0,
      "step": 59575
    },
    {
      "entropy": 5.6593420028686525,
      "epoch": 5.913060738388249,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00021409045482353512,
      "loss": 4.946,
      "mean_token_accuracy": 0.20613580346107482,
      "num_tokens": 130535902.0,
      "step": 59580
    },
    {
      "entropy": 5.70627007484436,
      "epoch": 5.9135569670504164,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002140563470522907,
      "loss": 5.0171,
      "mean_token_accuracy": 0.197878697514534,
      "num_tokens": 130547562.0,
      "step": 59585
    },
    {
      "entropy": 5.666899299621582,
      "epoch": 5.914053195712585,
      "grad_norm": 1.375,
      "learning_rate": 0.00021402224079231764,
      "loss": 4.8853,
      "mean_token_accuracy": 0.20109389424324037,
      "num_tokens": 130558604.0,
      "step": 59590
    },
    {
      "entropy": 5.691468954086304,
      "epoch": 5.914549424374751,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021398813604446172,
      "loss": 4.8837,
      "mean_token_accuracy": 0.2108568876981735,
      "num_tokens": 130569443.0,
      "step": 59595
    },
    {
      "entropy": 5.701705980300903,
      "epoch": 5.91504565303692,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002139540328095685,
      "loss": 4.9427,
      "mean_token_accuracy": 0.20385424494743348,
      "num_tokens": 130578944.0,
      "step": 59600
    },
    {
      "entropy": 5.663276290893554,
      "epoch": 5.915541881699087,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021391993108848368,
      "loss": 4.8678,
      "mean_token_accuracy": 0.2035858914256096,
      "num_tokens": 130589312.0,
      "step": 59605
    },
    {
      "entropy": 5.706214380264282,
      "epoch": 5.916038110361255,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000213885830882053,
      "loss": 4.9707,
      "mean_token_accuracy": 0.1898348495364189,
      "num_tokens": 130600358.0,
      "step": 59610
    },
    {
      "entropy": 5.780005168914795,
      "epoch": 5.916534339023422,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021385173219112192,
      "loss": 4.9778,
      "mean_token_accuracy": 0.19470855891704558,
      "num_tokens": 130610101.0,
      "step": 59615
    },
    {
      "entropy": 5.714764881134033,
      "epoch": 5.91703056768559,
      "grad_norm": 1.25,
      "learning_rate": 0.00021381763501653622,
      "loss": 4.9778,
      "mean_token_accuracy": 0.19340762197971345,
      "num_tokens": 130621258.0,
      "step": 59620
    },
    {
      "entropy": 5.700492238998413,
      "epoch": 5.917526796347757,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002137835393591413,
      "loss": 4.9206,
      "mean_token_accuracy": 0.19595878422260285,
      "num_tokens": 130631916.0,
      "step": 59625
    },
    {
      "entropy": 5.776569414138794,
      "epoch": 5.918023025009925,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021374944521978278,
      "loss": 4.943,
      "mean_token_accuracy": 0.19911076724529267,
      "num_tokens": 130642436.0,
      "step": 59630
    },
    {
      "entropy": 5.731266021728516,
      "epoch": 5.918519253672092,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002137153525993059,
      "loss": 4.9494,
      "mean_token_accuracy": 0.19632771611213684,
      "num_tokens": 130653772.0,
      "step": 59635
    },
    {
      "entropy": 5.743726396560669,
      "epoch": 5.91901548233426,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021368126149855633,
      "loss": 5.0021,
      "mean_token_accuracy": 0.19417546391487123,
      "num_tokens": 130664689.0,
      "step": 59640
    },
    {
      "entropy": 5.714429330825806,
      "epoch": 5.919511710996427,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021364717191837924,
      "loss": 4.9749,
      "mean_token_accuracy": 0.1926955133676529,
      "num_tokens": 130675351.0,
      "step": 59645
    },
    {
      "entropy": 5.687149095535278,
      "epoch": 5.9200079396585945,
      "grad_norm": 1.25,
      "learning_rate": 0.0002136130838596202,
      "loss": 4.9857,
      "mean_token_accuracy": 0.19728726595640184,
      "num_tokens": 130686725.0,
      "step": 59650
    },
    {
      "entropy": 5.702374219894409,
      "epoch": 5.920504168320762,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002135789973231243,
      "loss": 4.8748,
      "mean_token_accuracy": 0.20333820283412934,
      "num_tokens": 130697524.0,
      "step": 59655
    },
    {
      "entropy": 5.731326770782471,
      "epoch": 5.9210003969829295,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000213544912309737,
      "loss": 4.9362,
      "mean_token_accuracy": 0.19659093767404556,
      "num_tokens": 130708349.0,
      "step": 59660
    },
    {
      "entropy": 5.660531759262085,
      "epoch": 5.921496625645097,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021351082882030344,
      "loss": 4.8703,
      "mean_token_accuracy": 0.20527153909206391,
      "num_tokens": 130719695.0,
      "step": 59665
    },
    {
      "entropy": 5.624723863601685,
      "epoch": 5.921992854307264,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021347674685566877,
      "loss": 4.8226,
      "mean_token_accuracy": 0.2069172739982605,
      "num_tokens": 130729690.0,
      "step": 59670
    },
    {
      "entropy": 5.736315488815308,
      "epoch": 5.922489082969433,
      "grad_norm": 1.375,
      "learning_rate": 0.00021344266641667816,
      "loss": 4.9936,
      "mean_token_accuracy": 0.1915529415011406,
      "num_tokens": 130741487.0,
      "step": 59675
    },
    {
      "entropy": 5.654846048355102,
      "epoch": 5.922985311631599,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002134085875041769,
      "loss": 4.9584,
      "mean_token_accuracy": 0.20126839578151703,
      "num_tokens": 130752588.0,
      "step": 59680
    },
    {
      "entropy": 5.655264806747437,
      "epoch": 5.923481540293768,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021337451011900987,
      "loss": 4.8397,
      "mean_token_accuracy": 0.21851756423711777,
      "num_tokens": 130763023.0,
      "step": 59685
    },
    {
      "entropy": 5.685493326187133,
      "epoch": 5.923977768955935,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021334043426202227,
      "loss": 4.9126,
      "mean_token_accuracy": 0.19707223922014236,
      "num_tokens": 130774996.0,
      "step": 59690
    },
    {
      "entropy": 5.706793785095215,
      "epoch": 5.924473997618103,
      "grad_norm": 1.421875,
      "learning_rate": 0.000213306359934059,
      "loss": 4.9428,
      "mean_token_accuracy": 0.19580747336149215,
      "num_tokens": 130786637.0,
      "step": 59695
    },
    {
      "entropy": 5.723227834701538,
      "epoch": 5.92497022628027,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021327228713596507,
      "loss": 4.9297,
      "mean_token_accuracy": 0.1979196086525917,
      "num_tokens": 130796927.0,
      "step": 59700
    },
    {
      "entropy": 5.70212345123291,
      "epoch": 5.925466454942438,
      "grad_norm": 1.25,
      "learning_rate": 0.00021323821586858533,
      "loss": 4.9294,
      "mean_token_accuracy": 0.1941842868924141,
      "num_tokens": 130807114.0,
      "step": 59705
    },
    {
      "entropy": 5.769134140014648,
      "epoch": 5.925962683604605,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021320414613276484,
      "loss": 5.0832,
      "mean_token_accuracy": 0.18915882855653762,
      "num_tokens": 130819290.0,
      "step": 59710
    },
    {
      "entropy": 5.667085599899292,
      "epoch": 5.926458912266773,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002131700779293484,
      "loss": 4.8773,
      "mean_token_accuracy": 0.19686415940523147,
      "num_tokens": 130829974.0,
      "step": 59715
    },
    {
      "entropy": 5.699860763549805,
      "epoch": 5.92695514092894,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021313601125918076,
      "loss": 4.9007,
      "mean_token_accuracy": 0.2053038775920868,
      "num_tokens": 130842052.0,
      "step": 59720
    },
    {
      "entropy": 5.680440139770508,
      "epoch": 5.9274513695911075,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021310194612310674,
      "loss": 4.9719,
      "mean_token_accuracy": 0.19649318158626555,
      "num_tokens": 130852829.0,
      "step": 59725
    },
    {
      "entropy": 5.677450561523438,
      "epoch": 5.927947598253275,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021306788252197108,
      "loss": 4.8631,
      "mean_token_accuracy": 0.2026061162352562,
      "num_tokens": 130864271.0,
      "step": 59730
    },
    {
      "entropy": 5.687388467788696,
      "epoch": 5.9284438269154425,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002130338204566184,
      "loss": 4.9445,
      "mean_token_accuracy": 0.1976570874452591,
      "num_tokens": 130875277.0,
      "step": 59735
    },
    {
      "entropy": 5.67320065498352,
      "epoch": 5.92894005557761,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002129997599278935,
      "loss": 4.9242,
      "mean_token_accuracy": 0.19737193137407302,
      "num_tokens": 130887888.0,
      "step": 59740
    },
    {
      "entropy": 5.682087707519531,
      "epoch": 5.929436284239777,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021296570093664097,
      "loss": 4.868,
      "mean_token_accuracy": 0.20088112652301787,
      "num_tokens": 130898647.0,
      "step": 59745
    },
    {
      "entropy": 5.776618957519531,
      "epoch": 5.929932512901945,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021293164348370536,
      "loss": 5.0513,
      "mean_token_accuracy": 0.1892718181014061,
      "num_tokens": 130910277.0,
      "step": 59750
    },
    {
      "entropy": 5.733076429367065,
      "epoch": 5.930428741564112,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021289758756993127,
      "loss": 4.9641,
      "mean_token_accuracy": 0.20132143646478654,
      "num_tokens": 130920571.0,
      "step": 59755
    },
    {
      "entropy": 5.6928845882415775,
      "epoch": 5.930924970226281,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021286353319616315,
      "loss": 4.9086,
      "mean_token_accuracy": 0.20159224718809127,
      "num_tokens": 130930141.0,
      "step": 59760
    },
    {
      "entropy": 5.713991498947143,
      "epoch": 5.931421198888447,
      "grad_norm": 1.25,
      "learning_rate": 0.00021282948036324562,
      "loss": 4.9401,
      "mean_token_accuracy": 0.2022542729973793,
      "num_tokens": 130942019.0,
      "step": 59765
    },
    {
      "entropy": 5.695647239685059,
      "epoch": 5.931917427550616,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021279542907202286,
      "loss": 4.9311,
      "mean_token_accuracy": 0.20099413394927979,
      "num_tokens": 130951238.0,
      "step": 59770
    },
    {
      "entropy": 5.664012718200683,
      "epoch": 5.932413656212783,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021276137932333962,
      "loss": 4.9602,
      "mean_token_accuracy": 0.19220511466264725,
      "num_tokens": 130961207.0,
      "step": 59775
    },
    {
      "entropy": 5.730967664718628,
      "epoch": 5.932909884874951,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021272733111803993,
      "loss": 4.9324,
      "mean_token_accuracy": 0.2031039848923683,
      "num_tokens": 130971899.0,
      "step": 59780
    },
    {
      "entropy": 5.6808452129364015,
      "epoch": 5.933406113537118,
      "grad_norm": 1.375,
      "learning_rate": 0.00021269328445696823,
      "loss": 4.9156,
      "mean_token_accuracy": 0.1987678214907646,
      "num_tokens": 130982306.0,
      "step": 59785
    },
    {
      "entropy": 5.696409702301025,
      "epoch": 5.933902342199286,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021265923934096875,
      "loss": 4.948,
      "mean_token_accuracy": 0.1964745879173279,
      "num_tokens": 130993350.0,
      "step": 59790
    },
    {
      "entropy": 5.7129223346710205,
      "epoch": 5.934398570861453,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021262519577088597,
      "loss": 4.994,
      "mean_token_accuracy": 0.1912258043885231,
      "num_tokens": 131004584.0,
      "step": 59795
    },
    {
      "entropy": 5.672406244277954,
      "epoch": 5.9348947995236205,
      "grad_norm": 1.25,
      "learning_rate": 0.00021259115374756382,
      "loss": 4.9107,
      "mean_token_accuracy": 0.20233932435512542,
      "num_tokens": 131016628.0,
      "step": 59800
    },
    {
      "entropy": 5.680035066604614,
      "epoch": 5.935391028185788,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021255711327184662,
      "loss": 4.8953,
      "mean_token_accuracy": 0.20171072632074355,
      "num_tokens": 131027188.0,
      "step": 59805
    },
    {
      "entropy": 5.711127185821534,
      "epoch": 5.9358872568479555,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021252307434457845,
      "loss": 4.9066,
      "mean_token_accuracy": 0.19993361234664916,
      "num_tokens": 131038105.0,
      "step": 59810
    },
    {
      "entropy": 5.741501188278198,
      "epoch": 5.936383485510123,
      "grad_norm": 1.125,
      "learning_rate": 0.00021248903696660342,
      "loss": 5.0027,
      "mean_token_accuracy": 0.197239451110363,
      "num_tokens": 131050951.0,
      "step": 59815
    },
    {
      "entropy": 5.690712976455688,
      "epoch": 5.93687971417229,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021245500113876553,
      "loss": 4.8849,
      "mean_token_accuracy": 0.20781055092811584,
      "num_tokens": 131062314.0,
      "step": 59820
    },
    {
      "entropy": 5.649306726455689,
      "epoch": 5.937375942834458,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021242096686190887,
      "loss": 4.9218,
      "mean_token_accuracy": 0.1961693659424782,
      "num_tokens": 131074564.0,
      "step": 59825
    },
    {
      "entropy": 5.644203948974609,
      "epoch": 5.937872171496625,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021238693413687742,
      "loss": 4.8599,
      "mean_token_accuracy": 0.20192952156066896,
      "num_tokens": 131085260.0,
      "step": 59830
    },
    {
      "entropy": 5.633214521408081,
      "epoch": 5.938368400158793,
      "grad_norm": 1.375,
      "learning_rate": 0.00021235290296451502,
      "loss": 4.8545,
      "mean_token_accuracy": 0.20630219280719758,
      "num_tokens": 131095294.0,
      "step": 59835
    },
    {
      "entropy": 5.6938074111938475,
      "epoch": 5.93886462882096,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021231887334566575,
      "loss": 4.8917,
      "mean_token_accuracy": 0.19786432087421418,
      "num_tokens": 131105561.0,
      "step": 59840
    },
    {
      "entropy": 5.65969877243042,
      "epoch": 5.939360857483129,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021228484528117327,
      "loss": 4.9942,
      "mean_token_accuracy": 0.1918088361620903,
      "num_tokens": 131117078.0,
      "step": 59845
    },
    {
      "entropy": 5.74864239692688,
      "epoch": 5.939857086145295,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002122508187718814,
      "loss": 4.936,
      "mean_token_accuracy": 0.2013899564743042,
      "num_tokens": 131127024.0,
      "step": 59850
    },
    {
      "entropy": 5.726706886291504,
      "epoch": 5.940353314807464,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021221679381863413,
      "loss": 4.9258,
      "mean_token_accuracy": 0.2006937563419342,
      "num_tokens": 131137977.0,
      "step": 59855
    },
    {
      "entropy": 5.703578996658325,
      "epoch": 5.940849543469631,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021218277042227507,
      "loss": 4.8583,
      "mean_token_accuracy": 0.20262287855148314,
      "num_tokens": 131148125.0,
      "step": 59860
    },
    {
      "entropy": 5.715030193328857,
      "epoch": 5.941345772131799,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002121487485836479,
      "loss": 4.9534,
      "mean_token_accuracy": 0.19365907609462737,
      "num_tokens": 131159024.0,
      "step": 59865
    },
    {
      "entropy": 5.66949315071106,
      "epoch": 5.941842000793966,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002121147283035964,
      "loss": 4.8919,
      "mean_token_accuracy": 0.20222659707069396,
      "num_tokens": 131169023.0,
      "step": 59870
    },
    {
      "entropy": 5.670153331756592,
      "epoch": 5.9423382294561335,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002120807095829641,
      "loss": 4.8857,
      "mean_token_accuracy": 0.19534919112920762,
      "num_tokens": 131179733.0,
      "step": 59875
    },
    {
      "entropy": 5.713242721557617,
      "epoch": 5.942834458118301,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021204669242259454,
      "loss": 4.9538,
      "mean_token_accuracy": 0.19672275632619857,
      "num_tokens": 131191274.0,
      "step": 59880
    },
    {
      "entropy": 5.726413583755493,
      "epoch": 5.9433306867804685,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021201267682333149,
      "loss": 4.9182,
      "mean_token_accuracy": 0.2046283081173897,
      "num_tokens": 131203260.0,
      "step": 59885
    },
    {
      "entropy": 5.723476219177246,
      "epoch": 5.943826915442636,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002119786627860183,
      "loss": 4.93,
      "mean_token_accuracy": 0.20281623005867006,
      "num_tokens": 131214549.0,
      "step": 59890
    },
    {
      "entropy": 5.6283070087432865,
      "epoch": 5.9443231441048034,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021194465031149846,
      "loss": 4.8389,
      "mean_token_accuracy": 0.2060016632080078,
      "num_tokens": 131225170.0,
      "step": 59895
    },
    {
      "entropy": 5.650482225418091,
      "epoch": 5.944819372766971,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002119106394006154,
      "loss": 4.9053,
      "mean_token_accuracy": 0.205111300945282,
      "num_tokens": 131236107.0,
      "step": 59900
    },
    {
      "entropy": 5.687500381469727,
      "epoch": 5.945315601429138,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021187663005421258,
      "loss": 4.8578,
      "mean_token_accuracy": 0.20919426381587983,
      "num_tokens": 131246661.0,
      "step": 59905
    },
    {
      "entropy": 5.779783248901367,
      "epoch": 5.945811830091306,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021184262227313318,
      "loss": 5.0355,
      "mean_token_accuracy": 0.19762625247240068,
      "num_tokens": 131257678.0,
      "step": 59910
    },
    {
      "entropy": 5.715792560577393,
      "epoch": 5.946308058753473,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021180861605822078,
      "loss": 4.8913,
      "mean_token_accuracy": 0.20146759897470473,
      "num_tokens": 131268363.0,
      "step": 59915
    },
    {
      "entropy": 5.690916585922241,
      "epoch": 5.946804287415641,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021177461141031856,
      "loss": 4.9484,
      "mean_token_accuracy": 0.19770626574754716,
      "num_tokens": 131279480.0,
      "step": 59920
    },
    {
      "entropy": 5.785494756698609,
      "epoch": 5.947300516077808,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002117406083302697,
      "loss": 4.9959,
      "mean_token_accuracy": 0.19224423170089722,
      "num_tokens": 131291157.0,
      "step": 59925
    },
    {
      "entropy": 5.786647319793701,
      "epoch": 5.947796744739977,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002117066068189174,
      "loss": 5.0736,
      "mean_token_accuracy": 0.1856045424938202,
      "num_tokens": 131302993.0,
      "step": 59930
    },
    {
      "entropy": 5.751722240447998,
      "epoch": 5.948292973402143,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021167260687710493,
      "loss": 5.0164,
      "mean_token_accuracy": 0.19283689558506012,
      "num_tokens": 131315044.0,
      "step": 59935
    },
    {
      "entropy": 5.755949020385742,
      "epoch": 5.948789202064312,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002116386085056753,
      "loss": 4.9606,
      "mean_token_accuracy": 0.19546984285116195,
      "num_tokens": 131327437.0,
      "step": 59940
    },
    {
      "entropy": 5.716287279129029,
      "epoch": 5.949285430726479,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021160461170547168,
      "loss": 4.9067,
      "mean_token_accuracy": 0.20245832651853563,
      "num_tokens": 131337655.0,
      "step": 59945
    },
    {
      "entropy": 5.595172882080078,
      "epoch": 5.949781659388647,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021157061647733704,
      "loss": 4.8549,
      "mean_token_accuracy": 0.2011598154902458,
      "num_tokens": 131348633.0,
      "step": 59950
    },
    {
      "entropy": 5.641309976577759,
      "epoch": 5.950277888050814,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002115366228221145,
      "loss": 4.9248,
      "mean_token_accuracy": 0.19732559472322464,
      "num_tokens": 131358569.0,
      "step": 59955
    },
    {
      "entropy": 5.728995656967163,
      "epoch": 5.9507741167129815,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021150263074064696,
      "loss": 5.0288,
      "mean_token_accuracy": 0.19671018570661544,
      "num_tokens": 131369203.0,
      "step": 59960
    },
    {
      "entropy": 5.640692234039307,
      "epoch": 5.951270345375149,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021146864023377732,
      "loss": 4.8367,
      "mean_token_accuracy": 0.20640223622322082,
      "num_tokens": 131380921.0,
      "step": 59965
    },
    {
      "entropy": 5.772744131088257,
      "epoch": 5.9517665740373165,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002114346513023484,
      "loss": 5.0315,
      "mean_token_accuracy": 0.1908811643719673,
      "num_tokens": 131391925.0,
      "step": 59970
    },
    {
      "entropy": 5.735947370529175,
      "epoch": 5.952262802699484,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021140066394720336,
      "loss": 4.9474,
      "mean_token_accuracy": 0.1969047635793686,
      "num_tokens": 131403875.0,
      "step": 59975
    },
    {
      "entropy": 5.725822877883911,
      "epoch": 5.952759031361651,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021136667816918465,
      "loss": 4.9876,
      "mean_token_accuracy": 0.1981656149029732,
      "num_tokens": 131415355.0,
      "step": 59980
    },
    {
      "entropy": 5.747513294219971,
      "epoch": 5.953255260023819,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002113326939691353,
      "loss": 4.9811,
      "mean_token_accuracy": 0.20096552670001983,
      "num_tokens": 131426303.0,
      "step": 59985
    },
    {
      "entropy": 5.663601922988891,
      "epoch": 5.953751488685986,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021129871134789797,
      "loss": 4.8705,
      "mean_token_accuracy": 0.20007025301456452,
      "num_tokens": 131436243.0,
      "step": 59990
    },
    {
      "entropy": 5.7361006259918215,
      "epoch": 5.954247717348154,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021126473030631527,
      "loss": 5.0014,
      "mean_token_accuracy": 0.19447623938322067,
      "num_tokens": 131447264.0,
      "step": 59995
    },
    {
      "entropy": 5.705611991882324,
      "epoch": 5.954743946010321,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021123075084522984,
      "loss": 4.9273,
      "mean_token_accuracy": 0.19812578558921815,
      "num_tokens": 131457833.0,
      "step": 60000
    },
    {
      "epoch": 5.954743946010321,
      "eval_entropy": 5.4406968461388105,
      "eval_loss": 5.07008695602417,
      "eval_mean_token_accuracy": 0.1970447106988307,
      "eval_num_tokens": 131457833.0,
      "eval_runtime": 20.4048,
      "eval_samples_per_second": 1650.055,
      "eval_steps_per_second": 206.275,
      "step": 60000
    },
    {
      "entropy": 5.721195316314697,
      "epoch": 5.955240174672489,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002111967729654845,
      "loss": 4.9262,
      "mean_token_accuracy": 0.202276711165905,
      "num_tokens": 131469436.0,
      "step": 60005
    },
    {
      "entropy": 5.692776727676391,
      "epoch": 5.955736403334656,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021116279666792172,
      "loss": 4.9552,
      "mean_token_accuracy": 0.19921912103891373,
      "num_tokens": 131481239.0,
      "step": 60010
    },
    {
      "entropy": 5.663550186157226,
      "epoch": 5.956232631996825,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021112882195338395,
      "loss": 4.858,
      "mean_token_accuracy": 0.2042541339993477,
      "num_tokens": 131492090.0,
      "step": 60015
    },
    {
      "entropy": 5.674162483215332,
      "epoch": 5.956728860658991,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021109484882271385,
      "loss": 4.9635,
      "mean_token_accuracy": 0.19202346950769425,
      "num_tokens": 131503645.0,
      "step": 60020
    },
    {
      "entropy": 5.705339813232422,
      "epoch": 5.95722508932116,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021106087727675366,
      "loss": 4.9046,
      "mean_token_accuracy": 0.2016379028558731,
      "num_tokens": 131515352.0,
      "step": 60025
    },
    {
      "entropy": 5.627721452713013,
      "epoch": 5.957721317983327,
      "grad_norm": 1.375,
      "learning_rate": 0.00021102690731634606,
      "loss": 4.8303,
      "mean_token_accuracy": 0.21024305522441863,
      "num_tokens": 131526076.0,
      "step": 60030
    },
    {
      "entropy": 5.748900365829468,
      "epoch": 5.9582175466454945,
      "grad_norm": 1.1875,
      "learning_rate": 0.00021099293894233324,
      "loss": 5.0064,
      "mean_token_accuracy": 0.1866586536169052,
      "num_tokens": 131535724.0,
      "step": 60035
    },
    {
      "entropy": 5.671171474456787,
      "epoch": 5.958713775307662,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002109589721555577,
      "loss": 4.9177,
      "mean_token_accuracy": 0.20005494803190232,
      "num_tokens": 131546814.0,
      "step": 60040
    },
    {
      "entropy": 5.676276111602784,
      "epoch": 5.9592100039698295,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021092500695686156,
      "loss": 4.9015,
      "mean_token_accuracy": 0.2051983192563057,
      "num_tokens": 131558635.0,
      "step": 60045
    },
    {
      "entropy": 5.7036031723022464,
      "epoch": 5.959706232631997,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021089104334708726,
      "loss": 4.9741,
      "mean_token_accuracy": 0.20561902225017548,
      "num_tokens": 131570009.0,
      "step": 60050
    },
    {
      "entropy": 5.6976762294769285,
      "epoch": 5.960202461294164,
      "grad_norm": 1.25,
      "learning_rate": 0.00021085708132707675,
      "loss": 4.9812,
      "mean_token_accuracy": 0.1961926117539406,
      "num_tokens": 131581776.0,
      "step": 60055
    },
    {
      "entropy": 5.6808069229125975,
      "epoch": 5.960698689956332,
      "grad_norm": 1.5,
      "learning_rate": 0.00021082312089767263,
      "loss": 4.9762,
      "mean_token_accuracy": 0.19024735391139985,
      "num_tokens": 131593488.0,
      "step": 60060
    },
    {
      "entropy": 5.697676801681519,
      "epoch": 5.961194918618499,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002107891620597167,
      "loss": 4.9319,
      "mean_token_accuracy": 0.19837160259485245,
      "num_tokens": 131606501.0,
      "step": 60065
    },
    {
      "entropy": 5.708368301391602,
      "epoch": 5.961691147280667,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002107552048140513,
      "loss": 4.9173,
      "mean_token_accuracy": 0.1981634870171547,
      "num_tokens": 131616310.0,
      "step": 60070
    },
    {
      "entropy": 5.773905229568482,
      "epoch": 5.962187375942834,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021072124916151835,
      "loss": 5.0078,
      "mean_token_accuracy": 0.19510003924369812,
      "num_tokens": 131627462.0,
      "step": 60075
    },
    {
      "entropy": 5.676878452301025,
      "epoch": 5.962683604605002,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021068729510295987,
      "loss": 4.909,
      "mean_token_accuracy": 0.2003096431493759,
      "num_tokens": 131638750.0,
      "step": 60080
    },
    {
      "entropy": 5.696421480178833,
      "epoch": 5.963179833267169,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021065334263921782,
      "loss": 4.9454,
      "mean_token_accuracy": 0.19988838136196135,
      "num_tokens": 131648975.0,
      "step": 60085
    },
    {
      "entropy": 5.7112212657928465,
      "epoch": 5.963676061929337,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021061939177113433,
      "loss": 4.9634,
      "mean_token_accuracy": 0.1968934565782547,
      "num_tokens": 131660034.0,
      "step": 60090
    },
    {
      "entropy": 5.688348913192749,
      "epoch": 5.964172290591504,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021058544249955114,
      "loss": 4.9642,
      "mean_token_accuracy": 0.19511135518550873,
      "num_tokens": 131672479.0,
      "step": 60095
    },
    {
      "entropy": 5.666338729858398,
      "epoch": 5.964668519253673,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002105514948253102,
      "loss": 4.8951,
      "mean_token_accuracy": 0.19890083968639374,
      "num_tokens": 131683785.0,
      "step": 60100
    },
    {
      "entropy": 5.694564485549927,
      "epoch": 5.965164747915839,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021051754874925338,
      "loss": 4.9405,
      "mean_token_accuracy": 0.20126907974481584,
      "num_tokens": 131694108.0,
      "step": 60105
    },
    {
      "entropy": 5.697823810577392,
      "epoch": 5.9656609765780075,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002104836042722223,
      "loss": 4.8651,
      "mean_token_accuracy": 0.199151211977005,
      "num_tokens": 131704930.0,
      "step": 60110
    },
    {
      "entropy": 5.717173337936401,
      "epoch": 5.966157205240175,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021044966139505884,
      "loss": 4.9222,
      "mean_token_accuracy": 0.19592248201370238,
      "num_tokens": 131715993.0,
      "step": 60115
    },
    {
      "entropy": 5.645786952972412,
      "epoch": 5.9666534339023425,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021041572011860465,
      "loss": 4.8313,
      "mean_token_accuracy": 0.20798166394233703,
      "num_tokens": 131726386.0,
      "step": 60120
    },
    {
      "entropy": 5.72010350227356,
      "epoch": 5.96714966256451,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021038178044370154,
      "loss": 5.0294,
      "mean_token_accuracy": 0.19708416163921355,
      "num_tokens": 131738209.0,
      "step": 60125
    },
    {
      "entropy": 5.704982948303223,
      "epoch": 5.967645891226677,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021034784237119097,
      "loss": 4.9409,
      "mean_token_accuracy": 0.19704508632421494,
      "num_tokens": 131750583.0,
      "step": 60130
    },
    {
      "entropy": 5.731986999511719,
      "epoch": 5.968142119888845,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021031390590191463,
      "loss": 4.9969,
      "mean_token_accuracy": 0.19277465641498565,
      "num_tokens": 131760829.0,
      "step": 60135
    },
    {
      "entropy": 5.709158039093017,
      "epoch": 5.968638348551012,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021027997103671405,
      "loss": 4.937,
      "mean_token_accuracy": 0.198141385614872,
      "num_tokens": 131771703.0,
      "step": 60140
    },
    {
      "entropy": 5.6945994853973385,
      "epoch": 5.96913457721318,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002102460377764306,
      "loss": 4.9246,
      "mean_token_accuracy": 0.19579919278621674,
      "num_tokens": 131782419.0,
      "step": 60145
    },
    {
      "entropy": 5.70576491355896,
      "epoch": 5.969630805875347,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021021210612190595,
      "loss": 4.919,
      "mean_token_accuracy": 0.19429901838302613,
      "num_tokens": 131793971.0,
      "step": 60150
    },
    {
      "entropy": 5.714389419555664,
      "epoch": 5.970127034537515,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021017817607398148,
      "loss": 4.9403,
      "mean_token_accuracy": 0.19785120338201523,
      "num_tokens": 131803309.0,
      "step": 60155
    },
    {
      "entropy": 5.67614164352417,
      "epoch": 5.970623263199682,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021014424763349859,
      "loss": 4.9091,
      "mean_token_accuracy": 0.20580487102270126,
      "num_tokens": 131814164.0,
      "step": 60160
    },
    {
      "entropy": 5.659655284881592,
      "epoch": 5.97111949186185,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021011032080129862,
      "loss": 4.921,
      "mean_token_accuracy": 0.20516272336244584,
      "num_tokens": 131824242.0,
      "step": 60165
    },
    {
      "entropy": 5.694208335876465,
      "epoch": 5.971615720524017,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002100763955782229,
      "loss": 4.9114,
      "mean_token_accuracy": 0.2023157760500908,
      "num_tokens": 131834085.0,
      "step": 60170
    },
    {
      "entropy": 5.684908056259156,
      "epoch": 5.972111949186185,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002100424719651125,
      "loss": 4.9348,
      "mean_token_accuracy": 0.19860821813344956,
      "num_tokens": 131845546.0,
      "step": 60175
    },
    {
      "entropy": 5.65731520652771,
      "epoch": 5.972608177848352,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021000854996280888,
      "loss": 4.9081,
      "mean_token_accuracy": 0.20171902030706407,
      "num_tokens": 131856042.0,
      "step": 60180
    },
    {
      "entropy": 5.668509197235108,
      "epoch": 5.97310440651052,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020997462957215329,
      "loss": 4.9258,
      "mean_token_accuracy": 0.19464553147554398,
      "num_tokens": 131867770.0,
      "step": 60185
    },
    {
      "entropy": 5.684914207458496,
      "epoch": 5.973600635172687,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020994071079398675,
      "loss": 4.8421,
      "mean_token_accuracy": 0.1991121217608452,
      "num_tokens": 131878253.0,
      "step": 60190
    },
    {
      "entropy": 5.68364839553833,
      "epoch": 5.9740968638348555,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020990679362915026,
      "loss": 4.9228,
      "mean_token_accuracy": 0.19351789057254792,
      "num_tokens": 131888597.0,
      "step": 60195
    },
    {
      "entropy": 5.667377376556397,
      "epoch": 5.974593092497023,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020987287807848514,
      "loss": 4.9469,
      "mean_token_accuracy": 0.19407424032688142,
      "num_tokens": 131900548.0,
      "step": 60200
    },
    {
      "entropy": 5.7123040676116945,
      "epoch": 5.9750893211591904,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020983896414283225,
      "loss": 4.9429,
      "mean_token_accuracy": 0.1949862778186798,
      "num_tokens": 131911843.0,
      "step": 60205
    },
    {
      "entropy": 5.695088529586792,
      "epoch": 5.975585549821358,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020980505182303266,
      "loss": 4.9083,
      "mean_token_accuracy": 0.20068131238222123,
      "num_tokens": 131923012.0,
      "step": 60210
    },
    {
      "entropy": 5.692037057876587,
      "epoch": 5.976081778483525,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020977114111992736,
      "loss": 4.9554,
      "mean_token_accuracy": 0.1955655112862587,
      "num_tokens": 131933459.0,
      "step": 60215
    },
    {
      "entropy": 5.726960945129394,
      "epoch": 5.976578007145693,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020973723203435713,
      "loss": 4.909,
      "mean_token_accuracy": 0.19351866841316223,
      "num_tokens": 131943495.0,
      "step": 60220
    },
    {
      "entropy": 5.708671951293946,
      "epoch": 5.97707423580786,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020970332456716296,
      "loss": 4.9269,
      "mean_token_accuracy": 0.19889624565839767,
      "num_tokens": 131953915.0,
      "step": 60225
    },
    {
      "entropy": 5.6367748260498045,
      "epoch": 5.977570464470028,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020966941871918565,
      "loss": 4.8484,
      "mean_token_accuracy": 0.20059937685728074,
      "num_tokens": 131965128.0,
      "step": 60230
    },
    {
      "entropy": 5.713674402236938,
      "epoch": 5.978066693132195,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002096355144912659,
      "loss": 4.932,
      "mean_token_accuracy": 0.20759959816932677,
      "num_tokens": 131975560.0,
      "step": 60235
    },
    {
      "entropy": 5.747639274597168,
      "epoch": 5.978562921794363,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002096016118842446,
      "loss": 4.9769,
      "mean_token_accuracy": 0.19528599679470063,
      "num_tokens": 131986884.0,
      "step": 60240
    },
    {
      "entropy": 5.755369043350219,
      "epoch": 5.97905915045653,
      "grad_norm": 1.375,
      "learning_rate": 0.00020956771089896236,
      "loss": 4.9526,
      "mean_token_accuracy": 0.19464517682790755,
      "num_tokens": 131997745.0,
      "step": 60245
    },
    {
      "entropy": 5.682821750640869,
      "epoch": 5.979555379118698,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020953381153626,
      "loss": 4.8537,
      "mean_token_accuracy": 0.20248212665319443,
      "num_tokens": 132008676.0,
      "step": 60250
    },
    {
      "entropy": 5.71793942451477,
      "epoch": 5.980051607780865,
      "grad_norm": 1.375,
      "learning_rate": 0.00020949991379697798,
      "loss": 4.9907,
      "mean_token_accuracy": 0.1930861160159111,
      "num_tokens": 132019765.0,
      "step": 60255
    },
    {
      "entropy": 5.685354948043823,
      "epoch": 5.980547836443033,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002094660176819569,
      "loss": 4.9253,
      "mean_token_accuracy": 0.1971399486064911,
      "num_tokens": 132030128.0,
      "step": 60260
    },
    {
      "entropy": 5.698445892333984,
      "epoch": 5.9810440651052,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020943212319203742,
      "loss": 4.984,
      "mean_token_accuracy": 0.1938847541809082,
      "num_tokens": 132041173.0,
      "step": 60265
    },
    {
      "entropy": 5.688604402542114,
      "epoch": 5.981540293767368,
      "grad_norm": 1.21875,
      "learning_rate": 0.00020939823032806005,
      "loss": 4.8848,
      "mean_token_accuracy": 0.20151294767856598,
      "num_tokens": 132051681.0,
      "step": 60270
    },
    {
      "entropy": 5.6721889019012455,
      "epoch": 5.982036522429535,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002093643390908651,
      "loss": 4.8688,
      "mean_token_accuracy": 0.20209864526987076,
      "num_tokens": 132062196.0,
      "step": 60275
    },
    {
      "entropy": 5.720602798461914,
      "epoch": 5.9825327510917035,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020933044948129325,
      "loss": 4.9599,
      "mean_token_accuracy": 0.20196379274129866,
      "num_tokens": 132073538.0,
      "step": 60280
    },
    {
      "entropy": 5.710158967971802,
      "epoch": 5.983028979753871,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020929656150018466,
      "loss": 4.9245,
      "mean_token_accuracy": 0.197139148414135,
      "num_tokens": 132084822.0,
      "step": 60285
    },
    {
      "entropy": 5.720504093170166,
      "epoch": 5.983525208416038,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002092626751483797,
      "loss": 4.9674,
      "mean_token_accuracy": 0.199613818526268,
      "num_tokens": 132095950.0,
      "step": 60290
    },
    {
      "entropy": 5.632220649719239,
      "epoch": 5.984021437078206,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020922879042671884,
      "loss": 4.8975,
      "mean_token_accuracy": 0.19687164425849915,
      "num_tokens": 132106996.0,
      "step": 60295
    },
    {
      "entropy": 5.661283540725708,
      "epoch": 5.984517665740373,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00020919490733604224,
      "loss": 4.9122,
      "mean_token_accuracy": 0.19750812649726868,
      "num_tokens": 132116805.0,
      "step": 60300
    },
    {
      "entropy": 5.709272623062134,
      "epoch": 5.985013894402541,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020916102587719015,
      "loss": 4.9066,
      "mean_token_accuracy": 0.20639797747135163,
      "num_tokens": 132126798.0,
      "step": 60305
    },
    {
      "entropy": 5.771388196945191,
      "epoch": 5.985510123064708,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020912714605100276,
      "loss": 5.0303,
      "mean_token_accuracy": 0.19343677014112473,
      "num_tokens": 132137645.0,
      "step": 60310
    },
    {
      "entropy": 5.677811527252198,
      "epoch": 5.986006351726876,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00020909326785832017,
      "loss": 4.8152,
      "mean_token_accuracy": 0.21176477521657944,
      "num_tokens": 132149029.0,
      "step": 60315
    },
    {
      "entropy": 5.738540363311768,
      "epoch": 5.986502580389043,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020905939129998248,
      "loss": 4.9744,
      "mean_token_accuracy": 0.19794583916664124,
      "num_tokens": 132160468.0,
      "step": 60320
    },
    {
      "entropy": 5.654541063308716,
      "epoch": 5.986998809051211,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002090255163768298,
      "loss": 4.9292,
      "mean_token_accuracy": 0.20305422842502593,
      "num_tokens": 132171578.0,
      "step": 60325
    },
    {
      "entropy": 5.687994909286499,
      "epoch": 5.987495037713378,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020899164308970226,
      "loss": 4.933,
      "mean_token_accuracy": 0.19524179250001908,
      "num_tokens": 132183589.0,
      "step": 60330
    },
    {
      "entropy": 5.75799036026001,
      "epoch": 5.987991266375546,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020895777143943966,
      "loss": 4.9394,
      "mean_token_accuracy": 0.19928989559412003,
      "num_tokens": 132194064.0,
      "step": 60335
    },
    {
      "entropy": 5.717281007766724,
      "epoch": 5.988487495037713,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000208923901426882,
      "loss": 4.9685,
      "mean_token_accuracy": 0.19937365204095842,
      "num_tokens": 132205900.0,
      "step": 60340
    },
    {
      "entropy": 5.730182266235351,
      "epoch": 5.988983723699881,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002088900330528693,
      "loss": 4.9644,
      "mean_token_accuracy": 0.197673399746418,
      "num_tokens": 132216728.0,
      "step": 60345
    },
    {
      "entropy": 5.673335075378418,
      "epoch": 5.989479952362048,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020885616631824118,
      "loss": 4.9144,
      "mean_token_accuracy": 0.20256800651550294,
      "num_tokens": 132228198.0,
      "step": 60350
    },
    {
      "entropy": 5.740612602233886,
      "epoch": 5.989976181024216,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020882230122383773,
      "loss": 4.9663,
      "mean_token_accuracy": 0.1999475359916687,
      "num_tokens": 132238127.0,
      "step": 60355
    },
    {
      "entropy": 5.73858232498169,
      "epoch": 5.990472409686383,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020878843777049854,
      "loss": 4.9227,
      "mean_token_accuracy": 0.20428859442472458,
      "num_tokens": 132249219.0,
      "step": 60360
    },
    {
      "entropy": 5.693262910842895,
      "epoch": 5.990968638348551,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002087545759590635,
      "loss": 4.9221,
      "mean_token_accuracy": 0.19167297333478928,
      "num_tokens": 132260315.0,
      "step": 60365
    },
    {
      "entropy": 5.649129390716553,
      "epoch": 5.991464867010719,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020872071579037217,
      "loss": 4.9334,
      "mean_token_accuracy": 0.19359766244888305,
      "num_tokens": 132272951.0,
      "step": 60370
    },
    {
      "entropy": 5.700398063659668,
      "epoch": 5.991961095672886,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020868685726526426,
      "loss": 4.9725,
      "mean_token_accuracy": 0.18868068903684615,
      "num_tokens": 132285152.0,
      "step": 60375
    },
    {
      "entropy": 5.758655691146851,
      "epoch": 5.992457324335054,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002086530003845794,
      "loss": 4.9838,
      "mean_token_accuracy": 0.19261329025030136,
      "num_tokens": 132296522.0,
      "step": 60380
    },
    {
      "entropy": 5.670643663406372,
      "epoch": 5.992953552997221,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002086191451491572,
      "loss": 4.8605,
      "mean_token_accuracy": 0.2036848321557045,
      "num_tokens": 132307222.0,
      "step": 60385
    },
    {
      "entropy": 5.672957706451416,
      "epoch": 5.993449781659389,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020858529155983707,
      "loss": 4.9231,
      "mean_token_accuracy": 0.20489185601472854,
      "num_tokens": 132318390.0,
      "step": 60390
    },
    {
      "entropy": 5.650257205963134,
      "epoch": 5.993946010321556,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020855143961745872,
      "loss": 4.8874,
      "mean_token_accuracy": 0.20062429010868071,
      "num_tokens": 132327988.0,
      "step": 60395
    },
    {
      "entropy": 5.715911054611206,
      "epoch": 5.994442238983724,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020851758932286148,
      "loss": 4.9507,
      "mean_token_accuracy": 0.19850225895643234,
      "num_tokens": 132339018.0,
      "step": 60400
    },
    {
      "entropy": 5.72442216873169,
      "epoch": 5.994938467645891,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020848374067688474,
      "loss": 4.9657,
      "mean_token_accuracy": 0.19631759375333785,
      "num_tokens": 132349731.0,
      "step": 60405
    },
    {
      "entropy": 5.723477363586426,
      "epoch": 5.995434696308059,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020844989368036776,
      "loss": 4.9206,
      "mean_token_accuracy": 0.19383301138877868,
      "num_tokens": 132360308.0,
      "step": 60410
    },
    {
      "entropy": 5.677591228485108,
      "epoch": 5.995930924970226,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020841604833415016,
      "loss": 4.9143,
      "mean_token_accuracy": 0.19906774014234543,
      "num_tokens": 132370623.0,
      "step": 60415
    },
    {
      "entropy": 5.695804929733276,
      "epoch": 5.996427153632394,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020838220463907116,
      "loss": 4.9054,
      "mean_token_accuracy": 0.20670846253633499,
      "num_tokens": 132382718.0,
      "step": 60420
    },
    {
      "entropy": 5.661651372909546,
      "epoch": 5.996923382294561,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002083483625959698,
      "loss": 4.8523,
      "mean_token_accuracy": 0.20610806494951248,
      "num_tokens": 132392523.0,
      "step": 60425
    },
    {
      "entropy": 5.715108013153076,
      "epoch": 5.997419610956729,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020831452220568548,
      "loss": 4.9439,
      "mean_token_accuracy": 0.2052432745695114,
      "num_tokens": 132403306.0,
      "step": 60430
    },
    {
      "entropy": 5.752877187728882,
      "epoch": 5.997915839618896,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00020828068346905726,
      "loss": 4.8882,
      "mean_token_accuracy": 0.2029338076710701,
      "num_tokens": 132412723.0,
      "step": 60435
    },
    {
      "entropy": 5.723170232772827,
      "epoch": 5.9984120682810635,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020824684638692432,
      "loss": 4.9393,
      "mean_token_accuracy": 0.20586216151714326,
      "num_tokens": 132423049.0,
      "step": 60440
    },
    {
      "entropy": 5.7092857837677,
      "epoch": 5.998908296943231,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020821301096012585,
      "loss": 4.9534,
      "mean_token_accuracy": 0.19354789704084396,
      "num_tokens": 132433468.0,
      "step": 60445
    },
    {
      "entropy": 5.59419527053833,
      "epoch": 5.999404525605399,
      "grad_norm": 1.1875,
      "learning_rate": 0.00020817917718950074,
      "loss": 4.7785,
      "mean_token_accuracy": 0.20847920924425126,
      "num_tokens": 132443625.0,
      "step": 60450
    },
    {
      "entropy": 5.621834993362427,
      "epoch": 5.999900754267567,
      "grad_norm": 1.390625,
      "learning_rate": 0.000208145345075888,
      "loss": 4.8591,
      "mean_token_accuracy": 0.20232568979263305,
      "num_tokens": 132453460.0,
      "step": 60455
    },
    {
      "entropy": 5.7301243305206295,
      "epoch": 6.000396982929734,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020811151462012673,
      "loss": 4.9778,
      "mean_token_accuracy": 0.19598259329795836,
      "num_tokens": 132463207.0,
      "step": 60460
    },
    {
      "entropy": 5.7238397121429445,
      "epoch": 6.000893211591902,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002080776858230558,
      "loss": 4.9781,
      "mean_token_accuracy": 0.19489801675081253,
      "num_tokens": 132474290.0,
      "step": 60465
    },
    {
      "entropy": 5.72144718170166,
      "epoch": 6.001389440254069,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002080438586855139,
      "loss": 4.9595,
      "mean_token_accuracy": 0.19021906107664108,
      "num_tokens": 132484736.0,
      "step": 60470
    },
    {
      "entropy": 5.686188888549805,
      "epoch": 6.001885668916237,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020801003320834006,
      "loss": 4.8966,
      "mean_token_accuracy": 0.1981736734509468,
      "num_tokens": 132495949.0,
      "step": 60475
    },
    {
      "entropy": 5.697795724868774,
      "epoch": 6.002381897578404,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020797620939237317,
      "loss": 4.939,
      "mean_token_accuracy": 0.19885202944278718,
      "num_tokens": 132505885.0,
      "step": 60480
    },
    {
      "entropy": 5.633116912841797,
      "epoch": 6.002878126240572,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020794238723845183,
      "loss": 4.8156,
      "mean_token_accuracy": 0.20433103293180466,
      "num_tokens": 132518152.0,
      "step": 60485
    },
    {
      "entropy": 5.684381628036499,
      "epoch": 6.003374354902739,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020790856674741465,
      "loss": 4.8728,
      "mean_token_accuracy": 0.21060343980789184,
      "num_tokens": 132530036.0,
      "step": 60490
    },
    {
      "entropy": 5.760218286514283,
      "epoch": 6.003870583564907,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002078747479201006,
      "loss": 4.9617,
      "mean_token_accuracy": 0.20148265808820726,
      "num_tokens": 132541392.0,
      "step": 60495
    },
    {
      "entropy": 5.734358453750611,
      "epoch": 6.004366812227074,
      "grad_norm": 1.203125,
      "learning_rate": 0.00020784093075734805,
      "loss": 4.919,
      "mean_token_accuracy": 0.20147812217473984,
      "num_tokens": 132552768.0,
      "step": 60500
    },
    {
      "entropy": 5.738463115692139,
      "epoch": 6.004863040889242,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020780711525999573,
      "loss": 4.8849,
      "mean_token_accuracy": 0.20558219999074936,
      "num_tokens": 132563063.0,
      "step": 60505
    },
    {
      "entropy": 5.761860752105713,
      "epoch": 6.005359269551409,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020777330142888223,
      "loss": 4.9521,
      "mean_token_accuracy": 0.1953127071261406,
      "num_tokens": 132572736.0,
      "step": 60510
    },
    {
      "entropy": 5.692454195022583,
      "epoch": 6.005855498213577,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020773948926484597,
      "loss": 4.7785,
      "mean_token_accuracy": 0.21287340223789214,
      "num_tokens": 132582810.0,
      "step": 60515
    },
    {
      "entropy": 5.689834785461426,
      "epoch": 6.006351726875744,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020770567876872542,
      "loss": 4.9093,
      "mean_token_accuracy": 0.20375317931175232,
      "num_tokens": 132593149.0,
      "step": 60520
    },
    {
      "entropy": 5.7256043434143065,
      "epoch": 6.0068479555379115,
      "grad_norm": 1.375,
      "learning_rate": 0.00020767186994135895,
      "loss": 4.9946,
      "mean_token_accuracy": 0.19690203219652175,
      "num_tokens": 132604191.0,
      "step": 60525
    },
    {
      "entropy": 5.701676893234253,
      "epoch": 6.007344184200079,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00020763806278358517,
      "loss": 4.9283,
      "mean_token_accuracy": 0.19455475509166717,
      "num_tokens": 132615331.0,
      "step": 60530
    },
    {
      "entropy": 5.842006397247315,
      "epoch": 6.007840412862247,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020760425729624227,
      "loss": 5.114,
      "mean_token_accuracy": 0.19275864064693451,
      "num_tokens": 132626680.0,
      "step": 60535
    },
    {
      "entropy": 5.755704116821289,
      "epoch": 6.008336641524415,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020757045348016845,
      "loss": 4.9279,
      "mean_token_accuracy": 0.1962530344724655,
      "num_tokens": 132637934.0,
      "step": 60540
    },
    {
      "entropy": 5.769914436340332,
      "epoch": 6.008832870186582,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002075366513362022,
      "loss": 4.9843,
      "mean_token_accuracy": 0.1969414159655571,
      "num_tokens": 132648661.0,
      "step": 60545
    },
    {
      "entropy": 5.676542043685913,
      "epoch": 6.00932909884875,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020750285086518157,
      "loss": 4.7957,
      "mean_token_accuracy": 0.20715073943138124,
      "num_tokens": 132658595.0,
      "step": 60550
    },
    {
      "entropy": 5.707382249832153,
      "epoch": 6.009825327510917,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020746905206794477,
      "loss": 4.9074,
      "mean_token_accuracy": 0.20900894999504088,
      "num_tokens": 132669805.0,
      "step": 60555
    },
    {
      "entropy": 5.669692659378052,
      "epoch": 6.010321556173085,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020743525494533,
      "loss": 4.9172,
      "mean_token_accuracy": 0.19534459412097932,
      "num_tokens": 132680969.0,
      "step": 60560
    },
    {
      "entropy": 5.6155266761779785,
      "epoch": 6.010817784835252,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020740145949817534,
      "loss": 4.8646,
      "mean_token_accuracy": 0.20838411003351212,
      "num_tokens": 132691763.0,
      "step": 60565
    },
    {
      "entropy": 5.74844126701355,
      "epoch": 6.01131401349742,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002073676657273188,
      "loss": 4.9529,
      "mean_token_accuracy": 0.20046593397855758,
      "num_tokens": 132702741.0,
      "step": 60570
    },
    {
      "entropy": 5.735857105255127,
      "epoch": 6.011810242159587,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002073338736335984,
      "loss": 4.9665,
      "mean_token_accuracy": 0.20306348353624343,
      "num_tokens": 132713884.0,
      "step": 60575
    },
    {
      "entropy": 5.7821718692779545,
      "epoch": 6.012306470821755,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002073000832178522,
      "loss": 4.997,
      "mean_token_accuracy": 0.19439663290977477,
      "num_tokens": 132727107.0,
      "step": 60580
    },
    {
      "entropy": 5.706995058059692,
      "epoch": 6.012802699483922,
      "grad_norm": 1.203125,
      "learning_rate": 0.00020726629448091793,
      "loss": 4.9097,
      "mean_token_accuracy": 0.20099641978740693,
      "num_tokens": 132739300.0,
      "step": 60585
    },
    {
      "entropy": 5.719777917861938,
      "epoch": 6.01329892814609,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020723250742363364,
      "loss": 4.9296,
      "mean_token_accuracy": 0.2008029565215111,
      "num_tokens": 132750175.0,
      "step": 60590
    },
    {
      "entropy": 5.645334625244141,
      "epoch": 6.013795156808257,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020719872204683721,
      "loss": 4.8556,
      "mean_token_accuracy": 0.2002517834305763,
      "num_tokens": 132762067.0,
      "step": 60595
    },
    {
      "entropy": 5.682585763931274,
      "epoch": 6.0142913854704245,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020716493835136636,
      "loss": 4.8614,
      "mean_token_accuracy": 0.21109971404075623,
      "num_tokens": 132772646.0,
      "step": 60600
    },
    {
      "entropy": 5.720117092132568,
      "epoch": 6.014787614132592,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020713115633805878,
      "loss": 4.9585,
      "mean_token_accuracy": 0.20246980339288712,
      "num_tokens": 132783023.0,
      "step": 60605
    },
    {
      "entropy": 5.722769021987915,
      "epoch": 6.0152838427947595,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002070973760077524,
      "loss": 4.9641,
      "mean_token_accuracy": 0.19717566668987274,
      "num_tokens": 132794136.0,
      "step": 60610
    },
    {
      "entropy": 5.728279161453247,
      "epoch": 6.015780071456927,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020706359736128466,
      "loss": 4.9078,
      "mean_token_accuracy": 0.19721226543188095,
      "num_tokens": 132806164.0,
      "step": 60615
    },
    {
      "entropy": 5.677603530883789,
      "epoch": 6.016276300119095,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020702982039949336,
      "loss": 4.848,
      "mean_token_accuracy": 0.20116776525974273,
      "num_tokens": 132816612.0,
      "step": 60620
    },
    {
      "entropy": 5.705096578598022,
      "epoch": 6.016772528781263,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020699604512321607,
      "loss": 4.9431,
      "mean_token_accuracy": 0.19059587270021439,
      "num_tokens": 132825857.0,
      "step": 60625
    },
    {
      "entropy": 5.7406209945678714,
      "epoch": 6.01726875744343,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020696227153329033,
      "loss": 5.0066,
      "mean_token_accuracy": 0.19545146375894545,
      "num_tokens": 132837253.0,
      "step": 60630
    },
    {
      "entropy": 5.679847478866577,
      "epoch": 6.017764986105598,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020692849963055367,
      "loss": 4.9182,
      "mean_token_accuracy": 0.196126589179039,
      "num_tokens": 132848031.0,
      "step": 60635
    },
    {
      "entropy": 5.689463424682617,
      "epoch": 6.018261214767765,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020689472941584354,
      "loss": 4.9181,
      "mean_token_accuracy": 0.2014735758304596,
      "num_tokens": 132859051.0,
      "step": 60640
    },
    {
      "entropy": 5.742237186431884,
      "epoch": 6.018757443429933,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020686096088999722,
      "loss": 4.9506,
      "mean_token_accuracy": 0.19890383929014205,
      "num_tokens": 132869874.0,
      "step": 60645
    },
    {
      "entropy": 5.696318244934082,
      "epoch": 6.0192536720921,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020682719405385242,
      "loss": 4.8809,
      "mean_token_accuracy": 0.1956501841545105,
      "num_tokens": 132880241.0,
      "step": 60650
    },
    {
      "entropy": 5.735658407211304,
      "epoch": 6.019749900754268,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002067934289082462,
      "loss": 4.9544,
      "mean_token_accuracy": 0.200473652780056,
      "num_tokens": 132891369.0,
      "step": 60655
    },
    {
      "entropy": 5.691991758346558,
      "epoch": 6.020246129416435,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020675966545401605,
      "loss": 4.8491,
      "mean_token_accuracy": 0.2022157222032547,
      "num_tokens": 132901339.0,
      "step": 60660
    },
    {
      "entropy": 5.712312030792236,
      "epoch": 6.020742358078603,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002067259036919991,
      "loss": 4.95,
      "mean_token_accuracy": 0.19877487868070604,
      "num_tokens": 132911901.0,
      "step": 60665
    },
    {
      "entropy": 5.7095356464385985,
      "epoch": 6.02123858674077,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020669214362303268,
      "loss": 4.9367,
      "mean_token_accuracy": 0.20491761863231658,
      "num_tokens": 132921851.0,
      "step": 60670
    },
    {
      "entropy": 5.712263393402099,
      "epoch": 6.0217348154029375,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020665838524795376,
      "loss": 4.9369,
      "mean_token_accuracy": 0.19879864752292634,
      "num_tokens": 132933169.0,
      "step": 60675
    },
    {
      "entropy": 5.689131879806519,
      "epoch": 6.022231044065105,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020662462856759978,
      "loss": 4.8909,
      "mean_token_accuracy": 0.2099997490644455,
      "num_tokens": 132944039.0,
      "step": 60680
    },
    {
      "entropy": 5.7317675113677975,
      "epoch": 6.0227272727272725,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020659087358280759,
      "loss": 4.9611,
      "mean_token_accuracy": 0.19480008482933045,
      "num_tokens": 132955542.0,
      "step": 60685
    },
    {
      "entropy": 5.666561651229858,
      "epoch": 6.02322350138944,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002065571202944144,
      "loss": 4.8853,
      "mean_token_accuracy": 0.19600412994623184,
      "num_tokens": 132967383.0,
      "step": 60690
    },
    {
      "entropy": 5.76980447769165,
      "epoch": 6.023719730051607,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020652336870325712,
      "loss": 4.9689,
      "mean_token_accuracy": 0.19382482916116714,
      "num_tokens": 132977796.0,
      "step": 60695
    },
    {
      "entropy": 5.75766224861145,
      "epoch": 6.024215958713775,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020648961881017275,
      "loss": 4.9691,
      "mean_token_accuracy": 0.19402454644441605,
      "num_tokens": 132988109.0,
      "step": 60700
    },
    {
      "entropy": 5.682295227050782,
      "epoch": 6.024712187375943,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020645587061599808,
      "loss": 4.8727,
      "mean_token_accuracy": 0.19930859804153442,
      "num_tokens": 132997621.0,
      "step": 60705
    },
    {
      "entropy": 5.679393768310547,
      "epoch": 6.025208416038111,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020642212412157025,
      "loss": 4.821,
      "mean_token_accuracy": 0.20812648683786392,
      "num_tokens": 133008761.0,
      "step": 60710
    },
    {
      "entropy": 5.673998117446899,
      "epoch": 6.025704644700278,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020638837932772597,
      "loss": 4.9014,
      "mean_token_accuracy": 0.19782688170671464,
      "num_tokens": 133020471.0,
      "step": 60715
    },
    {
      "entropy": 5.715488719940185,
      "epoch": 6.026200873362446,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020635463623530197,
      "loss": 4.9875,
      "mean_token_accuracy": 0.19645845741033555,
      "num_tokens": 133032007.0,
      "step": 60720
    },
    {
      "entropy": 5.6711669921875,
      "epoch": 6.026697102024613,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020632089484513516,
      "loss": 4.9528,
      "mean_token_accuracy": 0.1980686828494072,
      "num_tokens": 133043373.0,
      "step": 60725
    },
    {
      "entropy": 5.719118452072143,
      "epoch": 6.027193330686781,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002062871551580621,
      "loss": 4.8819,
      "mean_token_accuracy": 0.206955386698246,
      "num_tokens": 133054655.0,
      "step": 60730
    },
    {
      "entropy": 5.680014133453369,
      "epoch": 6.027689559348948,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020625341717491952,
      "loss": 4.8493,
      "mean_token_accuracy": 0.20661866217851638,
      "num_tokens": 133064803.0,
      "step": 60735
    },
    {
      "entropy": 5.7195577144622805,
      "epoch": 6.028185788011116,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002062196808965441,
      "loss": 4.9579,
      "mean_token_accuracy": 0.20381987243890762,
      "num_tokens": 133076933.0,
      "step": 60740
    },
    {
      "entropy": 5.705780696868897,
      "epoch": 6.028682016673283,
      "grad_norm": 1.375,
      "learning_rate": 0.00020618594632377241,
      "loss": 4.8961,
      "mean_token_accuracy": 0.2035196378827095,
      "num_tokens": 133088101.0,
      "step": 60745
    },
    {
      "entropy": 5.696098947525025,
      "epoch": 6.0291782453354505,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002061522134574409,
      "loss": 4.8963,
      "mean_token_accuracy": 0.19842665493488312,
      "num_tokens": 133097559.0,
      "step": 60750
    },
    {
      "entropy": 5.757685565948487,
      "epoch": 6.029674473997618,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002061184822983862,
      "loss": 4.9361,
      "mean_token_accuracy": 0.19926866739988328,
      "num_tokens": 133108106.0,
      "step": 60755
    },
    {
      "entropy": 5.657991504669189,
      "epoch": 6.0301707026597855,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020608475284744466,
      "loss": 4.8525,
      "mean_token_accuracy": 0.20944846719503402,
      "num_tokens": 133118013.0,
      "step": 60760
    },
    {
      "entropy": 5.744143772125244,
      "epoch": 6.030666931321953,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020605102510545283,
      "loss": 4.9503,
      "mean_token_accuracy": 0.20265233963727952,
      "num_tokens": 133128303.0,
      "step": 60765
    },
    {
      "entropy": 5.736759901046753,
      "epoch": 6.0311631599841204,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002060172990732469,
      "loss": 4.9005,
      "mean_token_accuracy": 0.19847930669784547,
      "num_tokens": 133138154.0,
      "step": 60770
    },
    {
      "entropy": 5.726166582107544,
      "epoch": 6.031659388646288,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020598357475166341,
      "loss": 4.97,
      "mean_token_accuracy": 0.1971089571714401,
      "num_tokens": 133149458.0,
      "step": 60775
    },
    {
      "entropy": 5.6949316501617435,
      "epoch": 6.032155617308455,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020594985214153843,
      "loss": 4.9532,
      "mean_token_accuracy": 0.20029391646385192,
      "num_tokens": 133161579.0,
      "step": 60780
    },
    {
      "entropy": 5.688586568832397,
      "epoch": 6.032651845970623,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020591613124370844,
      "loss": 4.8547,
      "mean_token_accuracy": 0.20570995807647705,
      "num_tokens": 133172457.0,
      "step": 60785
    },
    {
      "entropy": 5.687713766098023,
      "epoch": 6.033148074632791,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00020588241205900948,
      "loss": 4.8528,
      "mean_token_accuracy": 0.20905074775218963,
      "num_tokens": 133183352.0,
      "step": 60790
    },
    {
      "entropy": 5.726465892791748,
      "epoch": 6.033644303294959,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020584869458827778,
      "loss": 4.8626,
      "mean_token_accuracy": 0.20314246267080308,
      "num_tokens": 133194616.0,
      "step": 60795
    },
    {
      "entropy": 5.711715364456177,
      "epoch": 6.034140531957126,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002058149788323494,
      "loss": 4.9492,
      "mean_token_accuracy": 0.19272368103265763,
      "num_tokens": 133203968.0,
      "step": 60800
    },
    {
      "entropy": 5.608884382247925,
      "epoch": 6.034636760619294,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020578126479206055,
      "loss": 4.8831,
      "mean_token_accuracy": 0.19692312479019164,
      "num_tokens": 133214663.0,
      "step": 60805
    },
    {
      "entropy": 5.667923259735107,
      "epoch": 6.035132989281461,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020574755246824717,
      "loss": 4.8938,
      "mean_token_accuracy": 0.2057239979505539,
      "num_tokens": 133225704.0,
      "step": 60810
    },
    {
      "entropy": 5.678434896469116,
      "epoch": 6.035629217943629,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020571384186174525,
      "loss": 4.9121,
      "mean_token_accuracy": 0.1958496555685997,
      "num_tokens": 133236798.0,
      "step": 60815
    },
    {
      "entropy": 5.734053134918213,
      "epoch": 6.036125446605796,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020568013297339068,
      "loss": 5.0068,
      "mean_token_accuracy": 0.1937143698334694,
      "num_tokens": 133248032.0,
      "step": 60820
    },
    {
      "entropy": 5.70851697921753,
      "epoch": 6.036621675267964,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002056464258040195,
      "loss": 4.9442,
      "mean_token_accuracy": 0.1984277918934822,
      "num_tokens": 133258041.0,
      "step": 60825
    },
    {
      "entropy": 5.721043252944947,
      "epoch": 6.037117903930131,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020561272035446755,
      "loss": 4.9663,
      "mean_token_accuracy": 0.1895644932985306,
      "num_tokens": 133268508.0,
      "step": 60830
    },
    {
      "entropy": 5.72245078086853,
      "epoch": 6.0376141325922985,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020557901662557055,
      "loss": 4.9382,
      "mean_token_accuracy": 0.19861883223056792,
      "num_tokens": 133278531.0,
      "step": 60835
    },
    {
      "entropy": 5.7121823787689205,
      "epoch": 6.038110361254466,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002055453146181644,
      "loss": 4.9334,
      "mean_token_accuracy": 0.1975862756371498,
      "num_tokens": 133290494.0,
      "step": 60840
    },
    {
      "entropy": 5.670694017410279,
      "epoch": 6.0386065899166335,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020551161433308475,
      "loss": 4.9234,
      "mean_token_accuracy": 0.20275644958019257,
      "num_tokens": 133300215.0,
      "step": 60845
    },
    {
      "entropy": 5.6663675785064695,
      "epoch": 6.039102818578801,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002054779157711673,
      "loss": 4.8538,
      "mean_token_accuracy": 0.20245365500450135,
      "num_tokens": 133310606.0,
      "step": 60850
    },
    {
      "entropy": 5.678756999969482,
      "epoch": 6.039599047240968,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020544421893324783,
      "loss": 4.847,
      "mean_token_accuracy": 0.20021541714668273,
      "num_tokens": 133320673.0,
      "step": 60855
    },
    {
      "entropy": 5.66982250213623,
      "epoch": 6.040095275903136,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020541052382016178,
      "loss": 4.8583,
      "mean_token_accuracy": 0.20271312147378923,
      "num_tokens": 133331927.0,
      "step": 60860
    },
    {
      "entropy": 5.6523055076599125,
      "epoch": 6.040591504565303,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002053768304327448,
      "loss": 4.938,
      "mean_token_accuracy": 0.19758803099393846,
      "num_tokens": 133342900.0,
      "step": 60865
    },
    {
      "entropy": 5.696521949768067,
      "epoch": 6.041087733227471,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002053431387718324,
      "loss": 4.9288,
      "mean_token_accuracy": 0.19741827696561814,
      "num_tokens": 133352704.0,
      "step": 60870
    },
    {
      "entropy": 5.721332979202271,
      "epoch": 6.041583961889638,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020530944883826013,
      "loss": 4.9377,
      "mean_token_accuracy": 0.19506451785564421,
      "num_tokens": 133362666.0,
      "step": 60875
    },
    {
      "entropy": 5.667854070663452,
      "epoch": 6.042080190551807,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020527576063286324,
      "loss": 4.9076,
      "mean_token_accuracy": 0.2039100632071495,
      "num_tokens": 133374091.0,
      "step": 60880
    },
    {
      "entropy": 5.688100433349609,
      "epoch": 6.042576419213974,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020524207415647728,
      "loss": 4.8547,
      "mean_token_accuracy": 0.2062878578901291,
      "num_tokens": 133386266.0,
      "step": 60885
    },
    {
      "entropy": 5.751254653930664,
      "epoch": 6.043072647876142,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020520838940993764,
      "loss": 4.959,
      "mean_token_accuracy": 0.1998226076364517,
      "num_tokens": 133398261.0,
      "step": 60890
    },
    {
      "entropy": 5.714268493652344,
      "epoch": 6.043568876538309,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020517470639407949,
      "loss": 4.8819,
      "mean_token_accuracy": 0.20736767053604127,
      "num_tokens": 133407989.0,
      "step": 60895
    },
    {
      "entropy": 5.668926239013672,
      "epoch": 6.044065105200477,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002051410251097382,
      "loss": 4.8559,
      "mean_token_accuracy": 0.2062785357236862,
      "num_tokens": 133418252.0,
      "step": 60900
    },
    {
      "entropy": 5.692953824996948,
      "epoch": 6.044561333862644,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002051073455577489,
      "loss": 4.9557,
      "mean_token_accuracy": 0.19779417365789415,
      "num_tokens": 133429403.0,
      "step": 60905
    },
    {
      "entropy": 5.746861600875855,
      "epoch": 6.0450575625248115,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020507366773894682,
      "loss": 5.0423,
      "mean_token_accuracy": 0.1898661121726036,
      "num_tokens": 133441493.0,
      "step": 60910
    },
    {
      "entropy": 5.73219313621521,
      "epoch": 6.045553791186979,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020503999165416708,
      "loss": 4.9585,
      "mean_token_accuracy": 0.19183394312858582,
      "num_tokens": 133452446.0,
      "step": 60915
    },
    {
      "entropy": 5.7256214141845705,
      "epoch": 6.0460500198491465,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002050063173042449,
      "loss": 4.9116,
      "mean_token_accuracy": 0.2023862361907959,
      "num_tokens": 133464594.0,
      "step": 60920
    },
    {
      "entropy": 5.72002420425415,
      "epoch": 6.046546248511314,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002049726446900152,
      "loss": 4.9962,
      "mean_token_accuracy": 0.19606832414865494,
      "num_tokens": 133475264.0,
      "step": 60925
    },
    {
      "entropy": 5.67519736289978,
      "epoch": 6.047042477173481,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020493897381231296,
      "loss": 4.894,
      "mean_token_accuracy": 0.19877471923828124,
      "num_tokens": 133486537.0,
      "step": 60930
    },
    {
      "entropy": 5.721724176406861,
      "epoch": 6.047538705835649,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020490530467197327,
      "loss": 4.9381,
      "mean_token_accuracy": 0.19952379465103148,
      "num_tokens": 133497123.0,
      "step": 60935
    },
    {
      "entropy": 5.656545877456665,
      "epoch": 6.048034934497816,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020487163726983092,
      "loss": 4.8596,
      "mean_token_accuracy": 0.2022467538714409,
      "num_tokens": 133507329.0,
      "step": 60940
    },
    {
      "entropy": 5.722591209411621,
      "epoch": 6.048531163159984,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020483797160672092,
      "loss": 4.9286,
      "mean_token_accuracy": 0.19654358476400374,
      "num_tokens": 133517482.0,
      "step": 60945
    },
    {
      "entropy": 5.6418524265289305,
      "epoch": 6.049027391822151,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020480430768347795,
      "loss": 4.831,
      "mean_token_accuracy": 0.2112932398915291,
      "num_tokens": 133528867.0,
      "step": 60950
    },
    {
      "entropy": 5.637978553771973,
      "epoch": 6.049523620484319,
      "grad_norm": 1.359375,
      "learning_rate": 0.000204770645500937,
      "loss": 4.8883,
      "mean_token_accuracy": 0.20437922477722167,
      "num_tokens": 133538631.0,
      "step": 60955
    },
    {
      "entropy": 5.708505487442016,
      "epoch": 6.050019849146486,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002047369850599326,
      "loss": 4.922,
      "mean_token_accuracy": 0.19993330240249635,
      "num_tokens": 133550498.0,
      "step": 60960
    },
    {
      "entropy": 5.75303692817688,
      "epoch": 6.050516077808655,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020470332636129967,
      "loss": 4.9685,
      "mean_token_accuracy": 0.1914299100637436,
      "num_tokens": 133561219.0,
      "step": 60965
    },
    {
      "entropy": 5.771357154846191,
      "epoch": 6.051012306470822,
      "grad_norm": 1.5625,
      "learning_rate": 0.00020466966940587268,
      "loss": 5.0267,
      "mean_token_accuracy": 0.19275304228067397,
      "num_tokens": 133571307.0,
      "step": 60970
    },
    {
      "entropy": 5.682817506790161,
      "epoch": 6.05150853513299,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020463601419448642,
      "loss": 4.8931,
      "mean_token_accuracy": 0.204560749232769,
      "num_tokens": 133582839.0,
      "step": 60975
    },
    {
      "entropy": 5.671124267578125,
      "epoch": 6.052004763795157,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020460236072797529,
      "loss": 4.8907,
      "mean_token_accuracy": 0.20098349153995515,
      "num_tokens": 133592647.0,
      "step": 60980
    },
    {
      "entropy": 5.725154542922974,
      "epoch": 6.0525009924573245,
      "grad_norm": 1.296875,
      "learning_rate": 0.000204568709007174,
      "loss": 4.9109,
      "mean_token_accuracy": 0.19693800956010818,
      "num_tokens": 133604414.0,
      "step": 60985
    },
    {
      "entropy": 5.72282919883728,
      "epoch": 6.052997221119492,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020453505903291697,
      "loss": 4.8659,
      "mean_token_accuracy": 0.20618866533041,
      "num_tokens": 133615366.0,
      "step": 60990
    },
    {
      "entropy": 5.672766399383545,
      "epoch": 6.0534934497816595,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020450141080603863,
      "loss": 4.8971,
      "mean_token_accuracy": 0.19813231229782105,
      "num_tokens": 133626575.0,
      "step": 60995
    },
    {
      "entropy": 5.680014371871948,
      "epoch": 6.053989678443827,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002044677643273733,
      "loss": 4.8844,
      "mean_token_accuracy": 0.19907656013965608,
      "num_tokens": 133637805.0,
      "step": 61000
    },
    {
      "entropy": 5.722097158432007,
      "epoch": 6.054485907105994,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020443411959775547,
      "loss": 4.9661,
      "mean_token_accuracy": 0.19648875892162324,
      "num_tokens": 133648743.0,
      "step": 61005
    },
    {
      "entropy": 5.670767068862915,
      "epoch": 6.054982135768162,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020440047661801947,
      "loss": 4.8413,
      "mean_token_accuracy": 0.21146931797266005,
      "num_tokens": 133661592.0,
      "step": 61010
    },
    {
      "entropy": 5.667594575881958,
      "epoch": 6.055478364430329,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020436683538899953,
      "loss": 4.8312,
      "mean_token_accuracy": 0.20704407542943953,
      "num_tokens": 133671389.0,
      "step": 61015
    },
    {
      "entropy": 5.7206565856933596,
      "epoch": 6.055974593092497,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020433319591152987,
      "loss": 4.9574,
      "mean_token_accuracy": 0.19903567284345627,
      "num_tokens": 133681883.0,
      "step": 61020
    },
    {
      "entropy": 5.7500756740570065,
      "epoch": 6.056470821754664,
      "grad_norm": 1.53125,
      "learning_rate": 0.00020429955818644457,
      "loss": 4.9402,
      "mean_token_accuracy": 0.19503648430109025,
      "num_tokens": 133692568.0,
      "step": 61025
    },
    {
      "entropy": 5.760653734207153,
      "epoch": 6.056967050416832,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020426592221457795,
      "loss": 4.995,
      "mean_token_accuracy": 0.19968628138303757,
      "num_tokens": 133703582.0,
      "step": 61030
    },
    {
      "entropy": 5.746571159362793,
      "epoch": 6.057463279078999,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020423228799676407,
      "loss": 4.885,
      "mean_token_accuracy": 0.20640651434659957,
      "num_tokens": 133713098.0,
      "step": 61035
    },
    {
      "entropy": 5.668308877944947,
      "epoch": 6.057959507741167,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020419865553383698,
      "loss": 4.8407,
      "mean_token_accuracy": 0.20533135235309602,
      "num_tokens": 133724532.0,
      "step": 61040
    },
    {
      "entropy": 5.666226720809936,
      "epoch": 6.058455736403334,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020416502482663057,
      "loss": 4.9291,
      "mean_token_accuracy": 0.20077853351831437,
      "num_tokens": 133735198.0,
      "step": 61045
    },
    {
      "entropy": 5.750625371932983,
      "epoch": 6.058951965065503,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000204131395875979,
      "loss": 4.9839,
      "mean_token_accuracy": 0.19868098497390746,
      "num_tokens": 133746409.0,
      "step": 61050
    },
    {
      "entropy": 5.674253940582275,
      "epoch": 6.05944819372767,
      "grad_norm": 1.296875,
      "learning_rate": 0.000204097768682716,
      "loss": 4.822,
      "mean_token_accuracy": 0.21475311517715454,
      "num_tokens": 133757447.0,
      "step": 61055
    },
    {
      "entropy": 5.75428991317749,
      "epoch": 6.0599444223898375,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002040641432476757,
      "loss": 4.9251,
      "mean_token_accuracy": 0.20481323599815368,
      "num_tokens": 133768705.0,
      "step": 61060
    },
    {
      "entropy": 5.708058071136475,
      "epoch": 6.060440651052005,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020403051957169167,
      "loss": 4.8646,
      "mean_token_accuracy": 0.1963426038622856,
      "num_tokens": 133779022.0,
      "step": 61065
    },
    {
      "entropy": 5.690629148483277,
      "epoch": 6.0609368797141725,
      "grad_norm": 1.21875,
      "learning_rate": 0.00020399689765559793,
      "loss": 4.9213,
      "mean_token_accuracy": 0.19455012679100037,
      "num_tokens": 133790144.0,
      "step": 61070
    },
    {
      "entropy": 5.671297073364258,
      "epoch": 6.06143310837634,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020396327750022808,
      "loss": 4.8918,
      "mean_token_accuracy": 0.2006544515490532,
      "num_tokens": 133801098.0,
      "step": 61075
    },
    {
      "entropy": 5.670603847503662,
      "epoch": 6.0619293370385074,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00020392965910641587,
      "loss": 4.8861,
      "mean_token_accuracy": 0.20341193228960036,
      "num_tokens": 133811890.0,
      "step": 61080
    },
    {
      "entropy": 5.716427850723266,
      "epoch": 6.062425565700675,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002038960424749949,
      "loss": 4.8679,
      "mean_token_accuracy": 0.21034106016159057,
      "num_tokens": 133821672.0,
      "step": 61085
    },
    {
      "entropy": 5.7600115776062015,
      "epoch": 6.062921794362842,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020386242760679895,
      "loss": 4.9463,
      "mean_token_accuracy": 0.19321951866149903,
      "num_tokens": 133832019.0,
      "step": 61090
    },
    {
      "entropy": 5.699896001815796,
      "epoch": 6.06341802302501,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002038288145026615,
      "loss": 4.9078,
      "mean_token_accuracy": 0.204019895195961,
      "num_tokens": 133843115.0,
      "step": 61095
    },
    {
      "entropy": 5.734546661376953,
      "epoch": 6.063914251687177,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002037952031634161,
      "loss": 4.9542,
      "mean_token_accuracy": 0.19392789155244827,
      "num_tokens": 133855065.0,
      "step": 61100
    },
    {
      "entropy": 5.664192867279053,
      "epoch": 6.064410480349345,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020376159358989626,
      "loss": 4.8585,
      "mean_token_accuracy": 0.20869521349668502,
      "num_tokens": 133864893.0,
      "step": 61105
    },
    {
      "entropy": 5.6864821434021,
      "epoch": 6.064906709011512,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020372798578293533,
      "loss": 4.8852,
      "mean_token_accuracy": 0.19671149849891661,
      "num_tokens": 133875874.0,
      "step": 61110
    },
    {
      "entropy": 5.639093065261841,
      "epoch": 6.06540293767368,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020369437974336675,
      "loss": 4.8586,
      "mean_token_accuracy": 0.19970354437828064,
      "num_tokens": 133887699.0,
      "step": 61115
    },
    {
      "entropy": 5.619020605087281,
      "epoch": 6.065899166335847,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020366077547202393,
      "loss": 4.791,
      "mean_token_accuracy": 0.2076675131917,
      "num_tokens": 133899860.0,
      "step": 61120
    },
    {
      "entropy": 5.671938037872314,
      "epoch": 6.066395394998015,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020362717296974015,
      "loss": 4.8858,
      "mean_token_accuracy": 0.20266369730234146,
      "num_tokens": 133911573.0,
      "step": 61125
    },
    {
      "entropy": 5.714977359771728,
      "epoch": 6.066891623660182,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020359357223734875,
      "loss": 4.9532,
      "mean_token_accuracy": 0.2041508898139,
      "num_tokens": 133921339.0,
      "step": 61130
    },
    {
      "entropy": 5.624254131317139,
      "epoch": 6.067387852322351,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002035599732756829,
      "loss": 4.7618,
      "mean_token_accuracy": 0.2147146314382553,
      "num_tokens": 133933301.0,
      "step": 61135
    },
    {
      "entropy": 5.735362195968628,
      "epoch": 6.067884080984518,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020352637608557568,
      "loss": 4.9966,
      "mean_token_accuracy": 0.1987687036395073,
      "num_tokens": 133944014.0,
      "step": 61140
    },
    {
      "entropy": 5.661330795288086,
      "epoch": 6.0683803096466855,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020349278066786032,
      "loss": 4.9299,
      "mean_token_accuracy": 0.19604319483041763,
      "num_tokens": 133954466.0,
      "step": 61145
    },
    {
      "entropy": 5.686342811584472,
      "epoch": 6.068876538308853,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020345918702336995,
      "loss": 4.872,
      "mean_token_accuracy": 0.19835812449455262,
      "num_tokens": 133965249.0,
      "step": 61150
    },
    {
      "entropy": 5.75054817199707,
      "epoch": 6.0693727669710205,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020342559515293762,
      "loss": 4.9891,
      "mean_token_accuracy": 0.1987570345401764,
      "num_tokens": 133976051.0,
      "step": 61155
    },
    {
      "entropy": 5.708262729644775,
      "epoch": 6.069868995633188,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020339200505739625,
      "loss": 4.8415,
      "mean_token_accuracy": 0.20472889244556428,
      "num_tokens": 133986512.0,
      "step": 61160
    },
    {
      "entropy": 5.721229600906372,
      "epoch": 6.070365224295355,
      "grad_norm": 1.5,
      "learning_rate": 0.00020335841673757893,
      "loss": 4.8877,
      "mean_token_accuracy": 0.20339870303869248,
      "num_tokens": 133996034.0,
      "step": 61165
    },
    {
      "entropy": 5.708962678909302,
      "epoch": 6.070861452957523,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020332483019431846,
      "loss": 4.9221,
      "mean_token_accuracy": 0.20565017759799958,
      "num_tokens": 134006905.0,
      "step": 61170
    },
    {
      "entropy": 5.753000211715698,
      "epoch": 6.07135768161969,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020329124542844767,
      "loss": 4.9663,
      "mean_token_accuracy": 0.20176504254341127,
      "num_tokens": 134017689.0,
      "step": 61175
    },
    {
      "entropy": 5.760462808609009,
      "epoch": 6.071853910281858,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020325766244079957,
      "loss": 4.9773,
      "mean_token_accuracy": 0.1905999019742012,
      "num_tokens": 134028110.0,
      "step": 61180
    },
    {
      "entropy": 5.728041362762451,
      "epoch": 6.072350138944025,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002032240812322068,
      "loss": 4.976,
      "mean_token_accuracy": 0.19481221586465836,
      "num_tokens": 134039810.0,
      "step": 61185
    },
    {
      "entropy": 5.653920269012451,
      "epoch": 6.072846367606193,
      "grad_norm": 1.25,
      "learning_rate": 0.0002031905018035022,
      "loss": 4.9341,
      "mean_token_accuracy": 0.1978396937251091,
      "num_tokens": 134050976.0,
      "step": 61190
    },
    {
      "entropy": 5.65382571220398,
      "epoch": 6.07334259626836,
      "grad_norm": 1.625,
      "learning_rate": 0.00020315692415551846,
      "loss": 4.8392,
      "mean_token_accuracy": 0.21383485198020935,
      "num_tokens": 134060781.0,
      "step": 61195
    },
    {
      "entropy": 5.616952514648437,
      "epoch": 6.073838824930528,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020312334828908812,
      "loss": 4.7631,
      "mean_token_accuracy": 0.20998858511447907,
      "num_tokens": 134069629.0,
      "step": 61200
    },
    {
      "entropy": 5.779649782180786,
      "epoch": 6.074335053592695,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020308977420504382,
      "loss": 5.0344,
      "mean_token_accuracy": 0.19059999883174897,
      "num_tokens": 134080472.0,
      "step": 61205
    },
    {
      "entropy": 5.599898147583008,
      "epoch": 6.074831282254863,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020305620190421819,
      "loss": 4.7944,
      "mean_token_accuracy": 0.2010295256972313,
      "num_tokens": 134092039.0,
      "step": 61210
    },
    {
      "entropy": 5.691368150711059,
      "epoch": 6.07532751091703,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020302263138744376,
      "loss": 4.8844,
      "mean_token_accuracy": 0.2061114341020584,
      "num_tokens": 134103552.0,
      "step": 61215
    },
    {
      "entropy": 5.687214469909668,
      "epoch": 6.0758237395791985,
      "grad_norm": 1.359375,
      "learning_rate": 0.000202989062655553,
      "loss": 4.847,
      "mean_token_accuracy": 0.20520375967025756,
      "num_tokens": 134113627.0,
      "step": 61220
    },
    {
      "entropy": 5.655078268051147,
      "epoch": 6.076319968241366,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020295549570937833,
      "loss": 4.8915,
      "mean_token_accuracy": 0.20420713424682618,
      "num_tokens": 134123962.0,
      "step": 61225
    },
    {
      "entropy": 5.658208131790161,
      "epoch": 6.0768161969035335,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002029219305497521,
      "loss": 4.8404,
      "mean_token_accuracy": 0.20228863060474395,
      "num_tokens": 134135353.0,
      "step": 61230
    },
    {
      "entropy": 5.6877823829650875,
      "epoch": 6.077312425565701,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020288836717750658,
      "loss": 4.8954,
      "mean_token_accuracy": 0.1964315116405487,
      "num_tokens": 134145663.0,
      "step": 61235
    },
    {
      "entropy": 5.740699386596679,
      "epoch": 6.077808654227868,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020285480559347423,
      "loss": 4.925,
      "mean_token_accuracy": 0.1983701303601265,
      "num_tokens": 134157754.0,
      "step": 61240
    },
    {
      "entropy": 5.6259825229644775,
      "epoch": 6.078304882890036,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020282124579848725,
      "loss": 4.8843,
      "mean_token_accuracy": 0.19824183285236358,
      "num_tokens": 134168706.0,
      "step": 61245
    },
    {
      "entropy": 5.675060558319092,
      "epoch": 6.078801111552203,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020278768779337792,
      "loss": 4.8415,
      "mean_token_accuracy": 0.21111956089735032,
      "num_tokens": 134179987.0,
      "step": 61250
    },
    {
      "entropy": 5.751262617111206,
      "epoch": 6.079297340214371,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020275413157897827,
      "loss": 4.9956,
      "mean_token_accuracy": 0.1940122976899147,
      "num_tokens": 134190435.0,
      "step": 61255
    },
    {
      "entropy": 5.700652837753296,
      "epoch": 6.079793568876538,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020272057715612035,
      "loss": 4.8794,
      "mean_token_accuracy": 0.20503900051116944,
      "num_tokens": 134201337.0,
      "step": 61260
    },
    {
      "entropy": 5.719839572906494,
      "epoch": 6.080289797538706,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020268702452563652,
      "loss": 4.917,
      "mean_token_accuracy": 0.2061299279332161,
      "num_tokens": 134213304.0,
      "step": 61265
    },
    {
      "entropy": 5.729849767684937,
      "epoch": 6.080786026200873,
      "grad_norm": 1.25,
      "learning_rate": 0.00020265347368835868,
      "loss": 4.9584,
      "mean_token_accuracy": 0.19228070676326753,
      "num_tokens": 134225274.0,
      "step": 61270
    },
    {
      "entropy": 5.71643385887146,
      "epoch": 6.081282254863041,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020261992464511874,
      "loss": 4.9189,
      "mean_token_accuracy": 0.20104919523000717,
      "num_tokens": 134236550.0,
      "step": 61275
    },
    {
      "entropy": 5.708967781066894,
      "epoch": 6.081778483525208,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020258637739674874,
      "loss": 4.8752,
      "mean_token_accuracy": 0.20852095186710357,
      "num_tokens": 134246672.0,
      "step": 61280
    },
    {
      "entropy": 5.673945474624634,
      "epoch": 6.082274712187376,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020255283194408054,
      "loss": 4.9057,
      "mean_token_accuracy": 0.19697480499744416,
      "num_tokens": 134258600.0,
      "step": 61285
    },
    {
      "entropy": 5.672286128997802,
      "epoch": 6.082770940849543,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020251928828794591,
      "loss": 4.9276,
      "mean_token_accuracy": 0.19764361530542374,
      "num_tokens": 134269560.0,
      "step": 61290
    },
    {
      "entropy": 5.673965072631836,
      "epoch": 6.083267169511711,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020248574642917682,
      "loss": 4.8994,
      "mean_token_accuracy": 0.20749470293521882,
      "num_tokens": 134280144.0,
      "step": 61295
    },
    {
      "entropy": 5.740665864944458,
      "epoch": 6.083763398173878,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020245220636860495,
      "loss": 4.9342,
      "mean_token_accuracy": 0.19990265220403672,
      "num_tokens": 134291234.0,
      "step": 61300
    },
    {
      "entropy": 5.685626220703125,
      "epoch": 6.0842596268360465,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020241866810706205,
      "loss": 4.8905,
      "mean_token_accuracy": 0.20103127509355545,
      "num_tokens": 134302397.0,
      "step": 61305
    },
    {
      "entropy": 5.687787771224976,
      "epoch": 6.084755855498214,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002023851316453798,
      "loss": 4.941,
      "mean_token_accuracy": 0.19564807713031768,
      "num_tokens": 134313059.0,
      "step": 61310
    },
    {
      "entropy": 5.708653545379638,
      "epoch": 6.085252084160381,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020235159698438982,
      "loss": 4.8993,
      "mean_token_accuracy": 0.19618826657533645,
      "num_tokens": 134324060.0,
      "step": 61315
    },
    {
      "entropy": 5.665898847579956,
      "epoch": 6.085748312822549,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002023180641249236,
      "loss": 4.8627,
      "mean_token_accuracy": 0.20345341861248017,
      "num_tokens": 134333943.0,
      "step": 61320
    },
    {
      "entropy": 5.666442346572876,
      "epoch": 6.086244541484716,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002022845330678128,
      "loss": 4.8954,
      "mean_token_accuracy": 0.2008093535900116,
      "num_tokens": 134345736.0,
      "step": 61325
    },
    {
      "entropy": 5.593930864334107,
      "epoch": 6.086740770146884,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000202251003813889,
      "loss": 4.8161,
      "mean_token_accuracy": 0.20260337591171265,
      "num_tokens": 134356511.0,
      "step": 61330
    },
    {
      "entropy": 5.646573686599732,
      "epoch": 6.087236998809051,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020221747636398348,
      "loss": 4.8142,
      "mean_token_accuracy": 0.2106114536523819,
      "num_tokens": 134367785.0,
      "step": 61335
    },
    {
      "entropy": 5.641175746917725,
      "epoch": 6.087733227471219,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002021839507189277,
      "loss": 4.847,
      "mean_token_accuracy": 0.20779570192098618,
      "num_tokens": 134378165.0,
      "step": 61340
    },
    {
      "entropy": 5.6330132484436035,
      "epoch": 6.088229456133386,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020215042687955304,
      "loss": 4.888,
      "mean_token_accuracy": 0.19769428968429564,
      "num_tokens": 134389008.0,
      "step": 61345
    },
    {
      "entropy": 5.668115186691284,
      "epoch": 6.088725684795554,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020211690484669075,
      "loss": 4.8776,
      "mean_token_accuracy": 0.20585260391235352,
      "num_tokens": 134399108.0,
      "step": 61350
    },
    {
      "entropy": 5.726064157485962,
      "epoch": 6.089221913457721,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020208338462117232,
      "loss": 4.976,
      "mean_token_accuracy": 0.1996076762676239,
      "num_tokens": 134409926.0,
      "step": 61355
    },
    {
      "entropy": 5.768187999725342,
      "epoch": 6.089718142119889,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002020498662038287,
      "loss": 5.0526,
      "mean_token_accuracy": 0.19419893026351928,
      "num_tokens": 134421485.0,
      "step": 61360
    },
    {
      "entropy": 5.685212659835815,
      "epoch": 6.090214370782056,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020201634959549132,
      "loss": 4.8961,
      "mean_token_accuracy": 0.19521175920963288,
      "num_tokens": 134431345.0,
      "step": 61365
    },
    {
      "entropy": 5.6966503143310545,
      "epoch": 6.090710599444224,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002019828347969911,
      "loss": 4.9132,
      "mean_token_accuracy": 0.19636917412281035,
      "num_tokens": 134442744.0,
      "step": 61370
    },
    {
      "entropy": 5.640894937515259,
      "epoch": 6.091206828106391,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002019493218091593,
      "loss": 4.8658,
      "mean_token_accuracy": 0.19855407774448394,
      "num_tokens": 134453194.0,
      "step": 61375
    },
    {
      "entropy": 5.767516231536865,
      "epoch": 6.091703056768559,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020191581063282684,
      "loss": 5.0252,
      "mean_token_accuracy": 0.19597758501768112,
      "num_tokens": 134465049.0,
      "step": 61380
    },
    {
      "entropy": 5.727264356613159,
      "epoch": 6.092199285430726,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020188230126882484,
      "loss": 4.9588,
      "mean_token_accuracy": 0.1972113311290741,
      "num_tokens": 134476806.0,
      "step": 61385
    },
    {
      "entropy": 5.688585376739502,
      "epoch": 6.092695514092894,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002018487937179842,
      "loss": 4.8826,
      "mean_token_accuracy": 0.2014603316783905,
      "num_tokens": 134487387.0,
      "step": 61390
    },
    {
      "entropy": 5.713282823562622,
      "epoch": 6.093191742755062,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020181528798113592,
      "loss": 4.8964,
      "mean_token_accuracy": 0.20357882380485534,
      "num_tokens": 134497913.0,
      "step": 61395
    },
    {
      "entropy": 5.71408519744873,
      "epoch": 6.093687971417229,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020178178405911074,
      "loss": 4.9538,
      "mean_token_accuracy": 0.20329698622226716,
      "num_tokens": 134510429.0,
      "step": 61400
    },
    {
      "entropy": 5.764868927001953,
      "epoch": 6.094184200079397,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020174828195273952,
      "loss": 5.0741,
      "mean_token_accuracy": 0.19585650712251662,
      "num_tokens": 134521938.0,
      "step": 61405
    },
    {
      "entropy": 5.653223133087158,
      "epoch": 6.094680428741564,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020171478166285307,
      "loss": 4.8173,
      "mean_token_accuracy": 0.21169168800115584,
      "num_tokens": 134532707.0,
      "step": 61410
    },
    {
      "entropy": 5.676598024368286,
      "epoch": 6.095176657403732,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020168128319028217,
      "loss": 4.8688,
      "mean_token_accuracy": 0.19956940412521362,
      "num_tokens": 134542583.0,
      "step": 61415
    },
    {
      "entropy": 5.754965877532959,
      "epoch": 6.095672886065899,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00020164778653585742,
      "loss": 5.0805,
      "mean_token_accuracy": 0.19338318407535554,
      "num_tokens": 134553894.0,
      "step": 61420
    },
    {
      "entropy": 5.713241529464722,
      "epoch": 6.096169114728067,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002016142917004095,
      "loss": 4.8821,
      "mean_token_accuracy": 0.19965171366930007,
      "num_tokens": 134565173.0,
      "step": 61425
    },
    {
      "entropy": 5.694644641876221,
      "epoch": 6.096665343390234,
      "grad_norm": 1.25,
      "learning_rate": 0.0002015807986847691,
      "loss": 4.8735,
      "mean_token_accuracy": 0.1996569886803627,
      "num_tokens": 134576061.0,
      "step": 61430
    },
    {
      "entropy": 5.708933591842651,
      "epoch": 6.097161572052402,
      "grad_norm": 1.375,
      "learning_rate": 0.00020154730748976659,
      "loss": 4.8881,
      "mean_token_accuracy": 0.2070642188191414,
      "num_tokens": 134587341.0,
      "step": 61435
    },
    {
      "entropy": 5.730370998382568,
      "epoch": 6.097657800714569,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020151381811623248,
      "loss": 4.8863,
      "mean_token_accuracy": 0.20291005223989486,
      "num_tokens": 134597950.0,
      "step": 61440
    },
    {
      "entropy": 5.669116973876953,
      "epoch": 6.098154029376737,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020148033056499747,
      "loss": 4.8646,
      "mean_token_accuracy": 0.20821046829223633,
      "num_tokens": 134609320.0,
      "step": 61445
    },
    {
      "entropy": 5.679932880401611,
      "epoch": 6.098650258038904,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00020144684483689191,
      "loss": 4.9036,
      "mean_token_accuracy": 0.20523741096258163,
      "num_tokens": 134618743.0,
      "step": 61450
    },
    {
      "entropy": 5.568005609512329,
      "epoch": 6.099146486701072,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000201413360932746,
      "loss": 4.7528,
      "mean_token_accuracy": 0.21259793192148208,
      "num_tokens": 134629796.0,
      "step": 61455
    },
    {
      "entropy": 5.667682123184204,
      "epoch": 6.099642715363239,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020137987885339022,
      "loss": 4.8488,
      "mean_token_accuracy": 0.19870001524686814,
      "num_tokens": 134640620.0,
      "step": 61460
    },
    {
      "entropy": 5.688909435272217,
      "epoch": 6.100138944025407,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002013463985996548,
      "loss": 4.8932,
      "mean_token_accuracy": 0.20168917924165725,
      "num_tokens": 134650528.0,
      "step": 61465
    },
    {
      "entropy": 5.681227302551269,
      "epoch": 6.100635172687574,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002013129201723699,
      "loss": 4.9317,
      "mean_token_accuracy": 0.192239248752594,
      "num_tokens": 134662562.0,
      "step": 61470
    },
    {
      "entropy": 5.74027533531189,
      "epoch": 6.1011314013497415,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020127944357236583,
      "loss": 4.9761,
      "mean_token_accuracy": 0.19383792728185653,
      "num_tokens": 134673860.0,
      "step": 61475
    },
    {
      "entropy": 5.717449951171875,
      "epoch": 6.10162763001191,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002012459688004728,
      "loss": 4.9322,
      "mean_token_accuracy": 0.19957848340272905,
      "num_tokens": 134686255.0,
      "step": 61480
    },
    {
      "entropy": 5.69516887664795,
      "epoch": 6.102123858674077,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020121249585752077,
      "loss": 4.7836,
      "mean_token_accuracy": 0.21076490581035615,
      "num_tokens": 134695700.0,
      "step": 61485
    },
    {
      "entropy": 5.665520286560058,
      "epoch": 6.102620087336245,
      "grad_norm": 1.375,
      "learning_rate": 0.00020117902474433986,
      "loss": 4.813,
      "mean_token_accuracy": 0.2061945155262947,
      "num_tokens": 134705811.0,
      "step": 61490
    },
    {
      "entropy": 5.646066761016845,
      "epoch": 6.103116315998412,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020114555546176004,
      "loss": 4.9226,
      "mean_token_accuracy": 0.20151073783636092,
      "num_tokens": 134718725.0,
      "step": 61495
    },
    {
      "entropy": 5.78639178276062,
      "epoch": 6.10361254466058,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020111208801061126,
      "loss": 4.9812,
      "mean_token_accuracy": 0.1966495618224144,
      "num_tokens": 134728815.0,
      "step": 61500
    },
    {
      "entropy": 5.759202861785889,
      "epoch": 6.104108773322747,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020107862239172354,
      "loss": 4.8918,
      "mean_token_accuracy": 0.2054001972079277,
      "num_tokens": 134738207.0,
      "step": 61505
    },
    {
      "entropy": 5.677643346786499,
      "epoch": 6.104605001984915,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020104515860592666,
      "loss": 4.8604,
      "mean_token_accuracy": 0.20150002092123032,
      "num_tokens": 134748453.0,
      "step": 61510
    },
    {
      "entropy": 5.6377379417419435,
      "epoch": 6.105101230647082,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020101169665405055,
      "loss": 4.8626,
      "mean_token_accuracy": 0.20896118581295015,
      "num_tokens": 134759781.0,
      "step": 61515
    },
    {
      "entropy": 5.737674617767334,
      "epoch": 6.10559745930925,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020097823653692488,
      "loss": 4.9258,
      "mean_token_accuracy": 0.19507086277008057,
      "num_tokens": 134769521.0,
      "step": 61520
    },
    {
      "entropy": 5.725995397567749,
      "epoch": 6.106093687971417,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020094477825537943,
      "loss": 4.942,
      "mean_token_accuracy": 0.19230195879936218,
      "num_tokens": 134780200.0,
      "step": 61525
    },
    {
      "entropy": 5.65110445022583,
      "epoch": 6.106589916633585,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020091132181024391,
      "loss": 4.863,
      "mean_token_accuracy": 0.2070044159889221,
      "num_tokens": 134791692.0,
      "step": 61530
    },
    {
      "entropy": 5.699625444412232,
      "epoch": 6.107086145295752,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020087786720234796,
      "loss": 4.8946,
      "mean_token_accuracy": 0.195994533598423,
      "num_tokens": 134803015.0,
      "step": 61535
    },
    {
      "entropy": 5.679690504074097,
      "epoch": 6.10758237395792,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00020084441443252128,
      "loss": 4.8856,
      "mean_token_accuracy": 0.20446095019578933,
      "num_tokens": 134813850.0,
      "step": 61540
    },
    {
      "entropy": 5.666565418243408,
      "epoch": 6.108078602620087,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020081096350159323,
      "loss": 4.8613,
      "mean_token_accuracy": 0.2000594362616539,
      "num_tokens": 134824410.0,
      "step": 61545
    },
    {
      "entropy": 5.698553466796875,
      "epoch": 6.1085748312822545,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020077751441039342,
      "loss": 4.9708,
      "mean_token_accuracy": 0.19859795570373534,
      "num_tokens": 134835729.0,
      "step": 61550
    },
    {
      "entropy": 5.7286717891693115,
      "epoch": 6.109071059944422,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002007440671597513,
      "loss": 4.922,
      "mean_token_accuracy": 0.1909013047814369,
      "num_tokens": 134845494.0,
      "step": 61555
    },
    {
      "entropy": 5.712095594406128,
      "epoch": 6.1095672886065895,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020071062175049643,
      "loss": 4.9189,
      "mean_token_accuracy": 0.19357375353574752,
      "num_tokens": 134857290.0,
      "step": 61560
    },
    {
      "entropy": 5.7048036575317385,
      "epoch": 6.110063517268758,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000200677178183458,
      "loss": 4.9557,
      "mean_token_accuracy": 0.19579256176948548,
      "num_tokens": 134867208.0,
      "step": 61565
    },
    {
      "entropy": 5.733241224288941,
      "epoch": 6.110559745930925,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020064373645946538,
      "loss": 4.9621,
      "mean_token_accuracy": 0.19904810935258865,
      "num_tokens": 134877340.0,
      "step": 61570
    },
    {
      "entropy": 5.696691703796387,
      "epoch": 6.111055974593093,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020061029657934788,
      "loss": 4.9118,
      "mean_token_accuracy": 0.2028820499777794,
      "num_tokens": 134887033.0,
      "step": 61575
    },
    {
      "entropy": 5.704538822174072,
      "epoch": 6.11155220325526,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002005768585439347,
      "loss": 4.8542,
      "mean_token_accuracy": 0.205798003077507,
      "num_tokens": 134897455.0,
      "step": 61580
    },
    {
      "entropy": 5.729047966003418,
      "epoch": 6.112048431917428,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002005434223540551,
      "loss": 4.9905,
      "mean_token_accuracy": 0.20358819961547853,
      "num_tokens": 134908648.0,
      "step": 61585
    },
    {
      "entropy": 5.707803583145141,
      "epoch": 6.112544660579595,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020050998801053816,
      "loss": 4.8997,
      "mean_token_accuracy": 0.20129021257162094,
      "num_tokens": 134919541.0,
      "step": 61590
    },
    {
      "entropy": 5.714483404159546,
      "epoch": 6.113040889241763,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020047655551421302,
      "loss": 4.9796,
      "mean_token_accuracy": 0.20109852105379106,
      "num_tokens": 134930391.0,
      "step": 61595
    },
    {
      "entropy": 5.80821385383606,
      "epoch": 6.11353711790393,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002004431248659087,
      "loss": 4.9721,
      "mean_token_accuracy": 0.199986532330513,
      "num_tokens": 134941317.0,
      "step": 61600
    },
    {
      "entropy": 5.678833723068237,
      "epoch": 6.114033346566098,
      "grad_norm": 1.25,
      "learning_rate": 0.00020040969606645426,
      "loss": 4.8377,
      "mean_token_accuracy": 0.19794143438339235,
      "num_tokens": 134953003.0,
      "step": 61605
    },
    {
      "entropy": 5.721447610855103,
      "epoch": 6.114529575228265,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020037626911667867,
      "loss": 4.93,
      "mean_token_accuracy": 0.19317311346530913,
      "num_tokens": 134963673.0,
      "step": 61610
    },
    {
      "entropy": 5.696203327178955,
      "epoch": 6.115025803890433,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002003428440174107,
      "loss": 5.0038,
      "mean_token_accuracy": 0.19681335985660553,
      "num_tokens": 134975576.0,
      "step": 61615
    },
    {
      "entropy": 5.75191297531128,
      "epoch": 6.1155220325526,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020030942076947934,
      "loss": 4.9565,
      "mean_token_accuracy": 0.1999325856566429,
      "num_tokens": 134987636.0,
      "step": 61620
    },
    {
      "entropy": 5.686937713623047,
      "epoch": 6.1160182612147675,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020027599937371348,
      "loss": 4.8053,
      "mean_token_accuracy": 0.2073462724685669,
      "num_tokens": 134999570.0,
      "step": 61625
    },
    {
      "entropy": 5.734977436065674,
      "epoch": 6.116514489876935,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002002425798309418,
      "loss": 4.8061,
      "mean_token_accuracy": 0.20544063448905944,
      "num_tokens": 135009613.0,
      "step": 61630
    },
    {
      "entropy": 5.6174479007720945,
      "epoch": 6.1170107185391025,
      "grad_norm": 1.375,
      "learning_rate": 0.00020020916214199297,
      "loss": 4.8627,
      "mean_token_accuracy": 0.19494717717170715,
      "num_tokens": 135019611.0,
      "step": 61635
    },
    {
      "entropy": 5.628169584274292,
      "epoch": 6.11750694720127,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020017574630769587,
      "loss": 4.8042,
      "mean_token_accuracy": 0.21486461609601976,
      "num_tokens": 135030881.0,
      "step": 61640
    },
    {
      "entropy": 5.708952951431274,
      "epoch": 6.1180031758634374,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002001423323288789,
      "loss": 4.9936,
      "mean_token_accuracy": 0.19125272929668427,
      "num_tokens": 135042831.0,
      "step": 61645
    },
    {
      "entropy": 5.680934715270996,
      "epoch": 6.118499404525606,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020010892020637085,
      "loss": 4.8857,
      "mean_token_accuracy": 0.2061346337199211,
      "num_tokens": 135053597.0,
      "step": 61650
    },
    {
      "entropy": 5.673884344100952,
      "epoch": 6.118995633187773,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00020007550994100026,
      "loss": 4.9423,
      "mean_token_accuracy": 0.20210806876420975,
      "num_tokens": 135065397.0,
      "step": 61655
    },
    {
      "entropy": 5.727202129364014,
      "epoch": 6.119491861849941,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020004210153359554,
      "loss": 4.9292,
      "mean_token_accuracy": 0.19458996802568435,
      "num_tokens": 135076224.0,
      "step": 61660
    },
    {
      "entropy": 5.709954929351807,
      "epoch": 6.119988090512108,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002000086949849852,
      "loss": 4.9287,
      "mean_token_accuracy": 0.20638663321733475,
      "num_tokens": 135087966.0,
      "step": 61665
    },
    {
      "entropy": 5.6882246971130375,
      "epoch": 6.120484319174276,
      "grad_norm": 1.25,
      "learning_rate": 0.0001999752902959976,
      "loss": 4.9124,
      "mean_token_accuracy": 0.19896892756223677,
      "num_tokens": 135099233.0,
      "step": 61670
    },
    {
      "entropy": 5.667942953109741,
      "epoch": 6.120980547836443,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019994188746746107,
      "loss": 4.8432,
      "mean_token_accuracy": 0.21045730262994766,
      "num_tokens": 135110602.0,
      "step": 61675
    },
    {
      "entropy": 5.657522487640381,
      "epoch": 6.121476776498611,
      "grad_norm": 1.125,
      "learning_rate": 0.0001999084865002041,
      "loss": 4.8484,
      "mean_token_accuracy": 0.2052591472864151,
      "num_tokens": 135121916.0,
      "step": 61680
    },
    {
      "entropy": 5.690977478027344,
      "epoch": 6.121973005160778,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019987508739505484,
      "loss": 4.8565,
      "mean_token_accuracy": 0.20417088717222215,
      "num_tokens": 135132043.0,
      "step": 61685
    },
    {
      "entropy": 5.698845100402832,
      "epoch": 6.122469233822946,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019984169015284153,
      "loss": 4.9581,
      "mean_token_accuracy": 0.20420969873666764,
      "num_tokens": 135143564.0,
      "step": 61690
    },
    {
      "entropy": 5.695373487472534,
      "epoch": 6.122965462485113,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019980829477439234,
      "loss": 4.8755,
      "mean_token_accuracy": 0.20156623870134355,
      "num_tokens": 135154699.0,
      "step": 61695
    },
    {
      "entropy": 5.749026536941528,
      "epoch": 6.123461691147281,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019977490126053544,
      "loss": 4.988,
      "mean_token_accuracy": 0.19928258806467056,
      "num_tokens": 135166112.0,
      "step": 61700
    },
    {
      "entropy": 5.702793407440185,
      "epoch": 6.123957919809448,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001997415096120988,
      "loss": 4.8633,
      "mean_token_accuracy": 0.20674585103988646,
      "num_tokens": 135177050.0,
      "step": 61705
    },
    {
      "entropy": 5.75167236328125,
      "epoch": 6.1244541484716155,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019970811982991066,
      "loss": 4.9318,
      "mean_token_accuracy": 0.19861089289188386,
      "num_tokens": 135186860.0,
      "step": 61710
    },
    {
      "entropy": 5.699536371231079,
      "epoch": 6.124950377133783,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019967473191479884,
      "loss": 4.8732,
      "mean_token_accuracy": 0.20130464881658555,
      "num_tokens": 135197447.0,
      "step": 61715
    },
    {
      "entropy": 5.707622385025024,
      "epoch": 6.1254466057959505,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019964134586759142,
      "loss": 4.9666,
      "mean_token_accuracy": 0.20399663746356964,
      "num_tokens": 135208049.0,
      "step": 61720
    },
    {
      "entropy": 5.705779933929444,
      "epoch": 6.125942834458118,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019960796168911626,
      "loss": 4.9291,
      "mean_token_accuracy": 0.20437075346708297,
      "num_tokens": 135220600.0,
      "step": 61725
    },
    {
      "entropy": 5.6659900665283205,
      "epoch": 6.126439063120285,
      "grad_norm": 1.109375,
      "learning_rate": 0.00019957457938020112,
      "loss": 4.9139,
      "mean_token_accuracy": 0.20044122785329818,
      "num_tokens": 135232965.0,
      "step": 61730
    },
    {
      "entropy": 5.689108419418335,
      "epoch": 6.126935291782454,
      "grad_norm": 1.375,
      "learning_rate": 0.00019954119894167384,
      "loss": 4.9168,
      "mean_token_accuracy": 0.198274102807045,
      "num_tokens": 135244065.0,
      "step": 61735
    },
    {
      "entropy": 5.728894853591919,
      "epoch": 6.127431520444621,
      "grad_norm": 1.234375,
      "learning_rate": 0.00019950782037436227,
      "loss": 4.911,
      "mean_token_accuracy": 0.19891097992658616,
      "num_tokens": 135255315.0,
      "step": 61740
    },
    {
      "entropy": 5.659724187850952,
      "epoch": 6.127927749106789,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001994744436790941,
      "loss": 4.8441,
      "mean_token_accuracy": 0.19826024174690246,
      "num_tokens": 135266773.0,
      "step": 61745
    },
    {
      "entropy": 5.7046104907989506,
      "epoch": 6.128423977768956,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019944106885669705,
      "loss": 4.9088,
      "mean_token_accuracy": 0.20155390053987504,
      "num_tokens": 135278200.0,
      "step": 61750
    },
    {
      "entropy": 5.729331159591675,
      "epoch": 6.128920206431124,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019940769590799862,
      "loss": 4.9624,
      "mean_token_accuracy": 0.19782997518777848,
      "num_tokens": 135290361.0,
      "step": 61755
    },
    {
      "entropy": 5.720707893371582,
      "epoch": 6.129416435093291,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001993743248338264,
      "loss": 5.0029,
      "mean_token_accuracy": 0.19267125576734542,
      "num_tokens": 135301137.0,
      "step": 61760
    },
    {
      "entropy": 5.776051616668701,
      "epoch": 6.129912663755459,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019934095563500797,
      "loss": 4.9598,
      "mean_token_accuracy": 0.1951628163456917,
      "num_tokens": 135312687.0,
      "step": 61765
    },
    {
      "entropy": 5.675012302398682,
      "epoch": 6.130408892417626,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001993075883123709,
      "loss": 4.8701,
      "mean_token_accuracy": 0.19346311092376708,
      "num_tokens": 135322008.0,
      "step": 61770
    },
    {
      "entropy": 5.715427732467651,
      "epoch": 6.130905121079794,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019927422286674247,
      "loss": 4.9182,
      "mean_token_accuracy": 0.20004615932703018,
      "num_tokens": 135333744.0,
      "step": 61775
    },
    {
      "entropy": 5.669026279449463,
      "epoch": 6.131401349741961,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019924085929895013,
      "loss": 4.9298,
      "mean_token_accuracy": 0.20201186686754227,
      "num_tokens": 135344279.0,
      "step": 61780
    },
    {
      "entropy": 5.695479249954223,
      "epoch": 6.1318975784041285,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019920749760982131,
      "loss": 4.8774,
      "mean_token_accuracy": 0.20165987610816954,
      "num_tokens": 135354805.0,
      "step": 61785
    },
    {
      "entropy": 5.632945728302002,
      "epoch": 6.132393807066296,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019917413780018312,
      "loss": 4.7871,
      "mean_token_accuracy": 0.20562542974948883,
      "num_tokens": 135365605.0,
      "step": 61790
    },
    {
      "entropy": 5.692930746078491,
      "epoch": 6.1328900357284635,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019914077987086298,
      "loss": 4.8782,
      "mean_token_accuracy": 0.20484456270933152,
      "num_tokens": 135375646.0,
      "step": 61795
    },
    {
      "entropy": 5.668422555923462,
      "epoch": 6.133386264390631,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019910742382268803,
      "loss": 4.9152,
      "mean_token_accuracy": 0.19270734935998918,
      "num_tokens": 135386290.0,
      "step": 61800
    },
    {
      "entropy": 5.68654088973999,
      "epoch": 6.133882493052798,
      "grad_norm": 1.375,
      "learning_rate": 0.0001990740696564855,
      "loss": 4.9183,
      "mean_token_accuracy": 0.20776932537555695,
      "num_tokens": 135397222.0,
      "step": 61805
    },
    {
      "entropy": 5.707780599594116,
      "epoch": 6.134378721714966,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019904071737308237,
      "loss": 4.9179,
      "mean_token_accuracy": 0.20192853659391402,
      "num_tokens": 135409138.0,
      "step": 61810
    },
    {
      "entropy": 5.678353500366211,
      "epoch": 6.134874950377133,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019900736697330573,
      "loss": 4.8797,
      "mean_token_accuracy": 0.20364612638950347,
      "num_tokens": 135420803.0,
      "step": 61815
    },
    {
      "entropy": 5.74007830619812,
      "epoch": 6.135371179039302,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019897401845798263,
      "loss": 4.9232,
      "mean_token_accuracy": 0.20238386541604997,
      "num_tokens": 135431136.0,
      "step": 61820
    },
    {
      "entropy": 5.687038946151733,
      "epoch": 6.135867407701469,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019894067182794017,
      "loss": 4.8898,
      "mean_token_accuracy": 0.1975399821996689,
      "num_tokens": 135442195.0,
      "step": 61825
    },
    {
      "entropy": 5.664111852645874,
      "epoch": 6.136363636363637,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001989073270840051,
      "loss": 4.8866,
      "mean_token_accuracy": 0.19768983721733094,
      "num_tokens": 135453174.0,
      "step": 61830
    },
    {
      "entropy": 5.703228855133057,
      "epoch": 6.136859865025804,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001988739842270043,
      "loss": 4.9843,
      "mean_token_accuracy": 0.19983448833227158,
      "num_tokens": 135464383.0,
      "step": 61835
    },
    {
      "entropy": 5.7296020030975345,
      "epoch": 6.137356093687972,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001988406432577648,
      "loss": 4.9816,
      "mean_token_accuracy": 0.19170401245355606,
      "num_tokens": 135475003.0,
      "step": 61840
    },
    {
      "entropy": 5.720136451721191,
      "epoch": 6.137852322350139,
      "grad_norm": 1.109375,
      "learning_rate": 0.0001988073041771131,
      "loss": 4.9807,
      "mean_token_accuracy": 0.19601045846939086,
      "num_tokens": 135486716.0,
      "step": 61845
    },
    {
      "entropy": 5.676627826690674,
      "epoch": 6.138348551012307,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019877396698587607,
      "loss": 4.9194,
      "mean_token_accuracy": 0.19971925914287567,
      "num_tokens": 135497926.0,
      "step": 61850
    },
    {
      "entropy": 5.692255640029908,
      "epoch": 6.138844779674474,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001987406316848805,
      "loss": 4.921,
      "mean_token_accuracy": 0.1972098767757416,
      "num_tokens": 135508840.0,
      "step": 61855
    },
    {
      "entropy": 5.626956653594971,
      "epoch": 6.1393410083366415,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019870729827495287,
      "loss": 4.832,
      "mean_token_accuracy": 0.20206634998321532,
      "num_tokens": 135519860.0,
      "step": 61860
    },
    {
      "entropy": 5.691202926635742,
      "epoch": 6.139837236998809,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019867396675691984,
      "loss": 4.9127,
      "mean_token_accuracy": 0.194980888068676,
      "num_tokens": 135529989.0,
      "step": 61865
    },
    {
      "entropy": 5.699759292602539,
      "epoch": 6.1403334656609765,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000198640637131608,
      "loss": 4.9845,
      "mean_token_accuracy": 0.19574409872293472,
      "num_tokens": 135542173.0,
      "step": 61870
    },
    {
      "entropy": 5.665677118301391,
      "epoch": 6.140829694323144,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019860730939984383,
      "loss": 4.8935,
      "mean_token_accuracy": 0.2036796674132347,
      "num_tokens": 135554036.0,
      "step": 61875
    },
    {
      "entropy": 5.691773843765259,
      "epoch": 6.141325922985311,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019857398356245372,
      "loss": 4.9306,
      "mean_token_accuracy": 0.1944731816649437,
      "num_tokens": 135564258.0,
      "step": 61880
    },
    {
      "entropy": 5.666269540786743,
      "epoch": 6.141822151647479,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019854065962026418,
      "loss": 4.8351,
      "mean_token_accuracy": 0.20187530666589737,
      "num_tokens": 135574460.0,
      "step": 61885
    },
    {
      "entropy": 5.764513826370239,
      "epoch": 6.142318380309646,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019850733757410154,
      "loss": 4.9486,
      "mean_token_accuracy": 0.20102235674858093,
      "num_tokens": 135585550.0,
      "step": 61890
    },
    {
      "entropy": 5.813395595550537,
      "epoch": 6.142814608971814,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001984740174247921,
      "loss": 5.012,
      "mean_token_accuracy": 0.19695772230625153,
      "num_tokens": 135596953.0,
      "step": 61895
    },
    {
      "entropy": 5.657052516937256,
      "epoch": 6.143310837633981,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001984406991731621,
      "loss": 4.9261,
      "mean_token_accuracy": 0.20407765805721284,
      "num_tokens": 135609234.0,
      "step": 61900
    },
    {
      "entropy": 5.67870831489563,
      "epoch": 6.14380706629615,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019840738282003785,
      "loss": 4.9588,
      "mean_token_accuracy": 0.20012464970350266,
      "num_tokens": 135620302.0,
      "step": 61905
    },
    {
      "entropy": 5.674736166000367,
      "epoch": 6.144303294958317,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019837406836624533,
      "loss": 4.8721,
      "mean_token_accuracy": 0.20489384233951569,
      "num_tokens": 135630252.0,
      "step": 61910
    },
    {
      "entropy": 5.690127372741699,
      "epoch": 6.144799523620485,
      "grad_norm": 1.25,
      "learning_rate": 0.0001983407558126109,
      "loss": 4.891,
      "mean_token_accuracy": 0.19972956031560898,
      "num_tokens": 135641939.0,
      "step": 61915
    },
    {
      "entropy": 5.748289489746094,
      "epoch": 6.145295752282652,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001983074451599605,
      "loss": 4.9782,
      "mean_token_accuracy": 0.19091855883598327,
      "num_tokens": 135652409.0,
      "step": 61920
    },
    {
      "entropy": 5.655802774429321,
      "epoch": 6.14579198094482,
      "grad_norm": 1.25,
      "learning_rate": 0.0001982741364091203,
      "loss": 4.8693,
      "mean_token_accuracy": 0.2027869015932083,
      "num_tokens": 135663722.0,
      "step": 61925
    },
    {
      "entropy": 5.678165912628174,
      "epoch": 6.146288209606987,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001982408295609161,
      "loss": 4.9721,
      "mean_token_accuracy": 0.19541243314743043,
      "num_tokens": 135676051.0,
      "step": 61930
    },
    {
      "entropy": 5.766265058517456,
      "epoch": 6.1467844382691545,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019820752461617398,
      "loss": 4.9965,
      "mean_token_accuracy": 0.19391796886920928,
      "num_tokens": 135686915.0,
      "step": 61935
    },
    {
      "entropy": 5.64467568397522,
      "epoch": 6.147280666931322,
      "grad_norm": 1.265625,
      "learning_rate": 0.00019817422157571962,
      "loss": 4.8323,
      "mean_token_accuracy": 0.2094885915517807,
      "num_tokens": 135697496.0,
      "step": 61940
    },
    {
      "entropy": 5.733235454559326,
      "epoch": 6.1477768955934895,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00019814092044037913,
      "loss": 5.0309,
      "mean_token_accuracy": 0.19335747212171556,
      "num_tokens": 135709752.0,
      "step": 61945
    },
    {
      "entropy": 5.70768027305603,
      "epoch": 6.148273124255657,
      "grad_norm": 1.25,
      "learning_rate": 0.00019810762121097824,
      "loss": 4.9003,
      "mean_token_accuracy": 0.20239977687597274,
      "num_tokens": 135721434.0,
      "step": 61950
    },
    {
      "entropy": 5.72360463142395,
      "epoch": 6.1487693529178244,
      "grad_norm": 1.21875,
      "learning_rate": 0.00019807432388834262,
      "loss": 4.9199,
      "mean_token_accuracy": 0.2023131161928177,
      "num_tokens": 135732880.0,
      "step": 61955
    },
    {
      "entropy": 5.775210618972778,
      "epoch": 6.149265581579992,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019804102847329803,
      "loss": 4.9654,
      "mean_token_accuracy": 0.19589459002017975,
      "num_tokens": 135743666.0,
      "step": 61960
    },
    {
      "entropy": 5.734547901153564,
      "epoch": 6.149761810242159,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019800773496667006,
      "loss": 4.9381,
      "mean_token_accuracy": 0.19929332584142684,
      "num_tokens": 135756472.0,
      "step": 61965
    },
    {
      "entropy": 5.702520561218262,
      "epoch": 6.150258038904327,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019797444336928433,
      "loss": 4.9321,
      "mean_token_accuracy": 0.19999436289072037,
      "num_tokens": 135767054.0,
      "step": 61970
    },
    {
      "entropy": 5.687344360351562,
      "epoch": 6.150754267566494,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00019794115368196652,
      "loss": 4.9987,
      "mean_token_accuracy": 0.18987831026315688,
      "num_tokens": 135779148.0,
      "step": 61975
    },
    {
      "entropy": 5.7484406471252445,
      "epoch": 6.151250496228662,
      "grad_norm": 1.40625,
      "learning_rate": 0.000197907865905542,
      "loss": 4.9966,
      "mean_token_accuracy": 0.19776706993579865,
      "num_tokens": 135788779.0,
      "step": 61980
    },
    {
      "entropy": 5.656371307373047,
      "epoch": 6.151746724890829,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001978745800408363,
      "loss": 4.8197,
      "mean_token_accuracy": 0.20135354101657868,
      "num_tokens": 135800828.0,
      "step": 61985
    },
    {
      "entropy": 5.715967607498169,
      "epoch": 6.152242953552998,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001978412960886748,
      "loss": 4.8731,
      "mean_token_accuracy": 0.20102430284023284,
      "num_tokens": 135811139.0,
      "step": 61990
    },
    {
      "entropy": 5.6763664245605465,
      "epoch": 6.152739182215165,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019780801404988292,
      "loss": 4.9095,
      "mean_token_accuracy": 0.2058582365512848,
      "num_tokens": 135821739.0,
      "step": 61995
    },
    {
      "entropy": 5.699926805496216,
      "epoch": 6.153235410877333,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001977747339252859,
      "loss": 4.9358,
      "mean_token_accuracy": 0.19937419444322585,
      "num_tokens": 135832962.0,
      "step": 62000
    },
    {
      "entropy": 5.727347135543823,
      "epoch": 6.1537316395395,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019774145571570918,
      "loss": 4.9382,
      "mean_token_accuracy": 0.19582561254501343,
      "num_tokens": 135842928.0,
      "step": 62005
    },
    {
      "entropy": 5.635860586166382,
      "epoch": 6.154227868201668,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019770817942197776,
      "loss": 4.8524,
      "mean_token_accuracy": 0.2034474268555641,
      "num_tokens": 135853877.0,
      "step": 62010
    },
    {
      "entropy": 5.720334577560425,
      "epoch": 6.154724096863835,
      "grad_norm": 1.234375,
      "learning_rate": 0.00019767490504491703,
      "loss": 4.9216,
      "mean_token_accuracy": 0.19122948497533798,
      "num_tokens": 135864381.0,
      "step": 62015
    },
    {
      "entropy": 5.725993299484253,
      "epoch": 6.1552203255260025,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019764163258535205,
      "loss": 4.9459,
      "mean_token_accuracy": 0.20763470381498336,
      "num_tokens": 135874657.0,
      "step": 62020
    },
    {
      "entropy": 5.746837329864502,
      "epoch": 6.15571655418817,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019760836204410782,
      "loss": 4.9972,
      "mean_token_accuracy": 0.1984763041138649,
      "num_tokens": 135885676.0,
      "step": 62025
    },
    {
      "entropy": 5.704109477996826,
      "epoch": 6.1562127828503375,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001975750934220094,
      "loss": 4.8764,
      "mean_token_accuracy": 0.20261053144931793,
      "num_tokens": 135895890.0,
      "step": 62030
    },
    {
      "entropy": 5.709080743789673,
      "epoch": 6.156709011512505,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00019754182671988198,
      "loss": 4.9305,
      "mean_token_accuracy": 0.19971081614494324,
      "num_tokens": 135906836.0,
      "step": 62035
    },
    {
      "entropy": 5.670199728012085,
      "epoch": 6.157205240174672,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019750856193855026,
      "loss": 4.8347,
      "mean_token_accuracy": 0.20547869354486464,
      "num_tokens": 135917765.0,
      "step": 62040
    },
    {
      "entropy": 5.73497896194458,
      "epoch": 6.15770146883684,
      "grad_norm": 1.265625,
      "learning_rate": 0.00019747529907883927,
      "loss": 4.9626,
      "mean_token_accuracy": 0.19133630394935608,
      "num_tokens": 135929582.0,
      "step": 62045
    },
    {
      "entropy": 5.674385023117066,
      "epoch": 6.158197697499007,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001974420381415738,
      "loss": 4.93,
      "mean_token_accuracy": 0.20016085356473923,
      "num_tokens": 135941716.0,
      "step": 62050
    },
    {
      "entropy": 5.672333812713623,
      "epoch": 6.158693926161175,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019740877912757855,
      "loss": 4.8704,
      "mean_token_accuracy": 0.20643150955438613,
      "num_tokens": 135952656.0,
      "step": 62055
    },
    {
      "entropy": 5.740928792953492,
      "epoch": 6.159190154823342,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019737552203767845,
      "loss": 4.8975,
      "mean_token_accuracy": 0.20348322093486787,
      "num_tokens": 135963598.0,
      "step": 62060
    },
    {
      "entropy": 5.70317964553833,
      "epoch": 6.15968638348551,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019734226687269818,
      "loss": 4.9259,
      "mean_token_accuracy": 0.20575055330991746,
      "num_tokens": 135973055.0,
      "step": 62065
    },
    {
      "entropy": 5.618092107772827,
      "epoch": 6.160182612147677,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019730901363346236,
      "loss": 4.7531,
      "mean_token_accuracy": 0.21140186190605165,
      "num_tokens": 135983150.0,
      "step": 62070
    },
    {
      "entropy": 5.732272815704346,
      "epoch": 6.160678840809846,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019727576232079558,
      "loss": 4.9276,
      "mean_token_accuracy": 0.2028466746211052,
      "num_tokens": 135995286.0,
      "step": 62075
    },
    {
      "entropy": 5.7331945419311525,
      "epoch": 6.161175069472013,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019724251293552243,
      "loss": 4.9583,
      "mean_token_accuracy": 0.2025888130068779,
      "num_tokens": 136006216.0,
      "step": 62080
    },
    {
      "entropy": 5.719434928894043,
      "epoch": 6.161671298134181,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019720926547846728,
      "loss": 4.9459,
      "mean_token_accuracy": 0.19759924262762069,
      "num_tokens": 136017961.0,
      "step": 62085
    },
    {
      "entropy": 5.684509515762329,
      "epoch": 6.162167526796348,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019717601995045491,
      "loss": 4.8832,
      "mean_token_accuracy": 0.20439072102308273,
      "num_tokens": 136028007.0,
      "step": 62090
    },
    {
      "entropy": 5.666795539855957,
      "epoch": 6.1626637554585155,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019714277635230943,
      "loss": 4.9018,
      "mean_token_accuracy": 0.2005745142698288,
      "num_tokens": 136039157.0,
      "step": 62095
    },
    {
      "entropy": 5.664177703857422,
      "epoch": 6.163159984120683,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019710953468485536,
      "loss": 4.9395,
      "mean_token_accuracy": 0.1937172755599022,
      "num_tokens": 136050360.0,
      "step": 62100
    },
    {
      "entropy": 5.702025699615478,
      "epoch": 6.1636562127828505,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019707629494891698,
      "loss": 4.9247,
      "mean_token_accuracy": 0.19343501478433608,
      "num_tokens": 136062151.0,
      "step": 62105
    },
    {
      "entropy": 5.679749202728272,
      "epoch": 6.164152441445018,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001970430571453186,
      "loss": 4.8639,
      "mean_token_accuracy": 0.2061292916536331,
      "num_tokens": 136073595.0,
      "step": 62110
    },
    {
      "entropy": 5.700574827194214,
      "epoch": 6.164648670107185,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019700982127488436,
      "loss": 4.8996,
      "mean_token_accuracy": 0.20162872225046158,
      "num_tokens": 136083870.0,
      "step": 62115
    },
    {
      "entropy": 5.672293996810913,
      "epoch": 6.165144898769353,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001969765873384386,
      "loss": 4.9376,
      "mean_token_accuracy": 0.19316633194684982,
      "num_tokens": 136094950.0,
      "step": 62120
    },
    {
      "entropy": 5.696435594558716,
      "epoch": 6.16564112743152,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019694335533680524,
      "loss": 4.9062,
      "mean_token_accuracy": 0.20016980171203613,
      "num_tokens": 136105849.0,
      "step": 62125
    },
    {
      "entropy": 5.726386260986328,
      "epoch": 6.166137356093688,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019691012527080855,
      "loss": 4.9058,
      "mean_token_accuracy": 0.19685453772544861,
      "num_tokens": 136117041.0,
      "step": 62130
    },
    {
      "entropy": 5.725692224502564,
      "epoch": 6.166633584755855,
      "grad_norm": 1.375,
      "learning_rate": 0.0001968768971412725,
      "loss": 4.9309,
      "mean_token_accuracy": 0.19549549669027327,
      "num_tokens": 136128653.0,
      "step": 62135
    },
    {
      "entropy": 5.755386781692505,
      "epoch": 6.167129813418023,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019684367094902094,
      "loss": 4.9868,
      "mean_token_accuracy": 0.19604647606611253,
      "num_tokens": 136139398.0,
      "step": 62140
    },
    {
      "entropy": 5.6982237815856935,
      "epoch": 6.16762604208019,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000196810446694878,
      "loss": 4.9061,
      "mean_token_accuracy": 0.19432743340730668,
      "num_tokens": 136150810.0,
      "step": 62145
    },
    {
      "entropy": 5.726276397705078,
      "epoch": 6.168122270742358,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001967772243796675,
      "loss": 5.0055,
      "mean_token_accuracy": 0.18415856659412383,
      "num_tokens": 136161470.0,
      "step": 62150
    },
    {
      "entropy": 5.735334491729736,
      "epoch": 6.168618499404525,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00019674400400421333,
      "loss": 4.9214,
      "mean_token_accuracy": 0.19249696433544158,
      "num_tokens": 136172049.0,
      "step": 62155
    },
    {
      "entropy": 5.705492639541626,
      "epoch": 6.169114728066694,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019671078556933919,
      "loss": 4.8795,
      "mean_token_accuracy": 0.20043254196643828,
      "num_tokens": 136183043.0,
      "step": 62160
    },
    {
      "entropy": 5.641473865509033,
      "epoch": 6.169610956728861,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019667756907586887,
      "loss": 4.8567,
      "mean_token_accuracy": 0.20572547763586044,
      "num_tokens": 136195112.0,
      "step": 62165
    },
    {
      "entropy": 5.749824094772339,
      "epoch": 6.1701071853910285,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019664435452462604,
      "loss": 4.8921,
      "mean_token_accuracy": 0.200393408536911,
      "num_tokens": 136205537.0,
      "step": 62170
    },
    {
      "entropy": 5.709978103637695,
      "epoch": 6.170603414053196,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019661114191643438,
      "loss": 4.9672,
      "mean_token_accuracy": 0.19765092581510543,
      "num_tokens": 136215158.0,
      "step": 62175
    },
    {
      "entropy": 5.7079285144805905,
      "epoch": 6.1710996427153635,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001965779312521175,
      "loss": 4.8823,
      "mean_token_accuracy": 0.19943766593933104,
      "num_tokens": 136226825.0,
      "step": 62180
    },
    {
      "entropy": 5.725777339935303,
      "epoch": 6.171595871377531,
      "grad_norm": 1.203125,
      "learning_rate": 0.000196544722532499,
      "loss": 4.9904,
      "mean_token_accuracy": 0.1959374502301216,
      "num_tokens": 136238071.0,
      "step": 62185
    },
    {
      "entropy": 5.773301124572754,
      "epoch": 6.172092100039698,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019651151575840226,
      "loss": 4.9753,
      "mean_token_accuracy": 0.19401497393846512,
      "num_tokens": 136249115.0,
      "step": 62190
    },
    {
      "entropy": 5.813975763320923,
      "epoch": 6.172588328701866,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019647831093065084,
      "loss": 4.9879,
      "mean_token_accuracy": 0.20210545808076857,
      "num_tokens": 136260547.0,
      "step": 62195
    },
    {
      "entropy": 5.728480958938599,
      "epoch": 6.173084557364033,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001964451080500681,
      "loss": 4.9768,
      "mean_token_accuracy": 0.19607169479131697,
      "num_tokens": 136271739.0,
      "step": 62200
    },
    {
      "entropy": 5.744627809524536,
      "epoch": 6.173580786026201,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001964119071174773,
      "loss": 4.9412,
      "mean_token_accuracy": 0.19625693559646606,
      "num_tokens": 136283259.0,
      "step": 62205
    },
    {
      "entropy": 5.735250854492188,
      "epoch": 6.174077014688368,
      "grad_norm": 1.265625,
      "learning_rate": 0.00019637870813370197,
      "loss": 5.0523,
      "mean_token_accuracy": 0.19400109201669694,
      "num_tokens": 136295414.0,
      "step": 62210
    },
    {
      "entropy": 5.643333053588867,
      "epoch": 6.174573243350536,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001963455110995653,
      "loss": 4.8133,
      "mean_token_accuracy": 0.20240146070718765,
      "num_tokens": 136306531.0,
      "step": 62215
    },
    {
      "entropy": 5.6981206893920895,
      "epoch": 6.175069472012703,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001963123160158904,
      "loss": 4.9089,
      "mean_token_accuracy": 0.19936011284589766,
      "num_tokens": 136317109.0,
      "step": 62220
    },
    {
      "entropy": 5.844471549987793,
      "epoch": 6.175565700674871,
      "grad_norm": 1.375,
      "learning_rate": 0.00019627912288350057,
      "loss": 5.037,
      "mean_token_accuracy": 0.19407104700803757,
      "num_tokens": 136327147.0,
      "step": 62225
    },
    {
      "entropy": 5.684536695480347,
      "epoch": 6.176061929337038,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019624593170321886,
      "loss": 4.8215,
      "mean_token_accuracy": 0.21280240267515182,
      "num_tokens": 136339203.0,
      "step": 62230
    },
    {
      "entropy": 5.643567895889282,
      "epoch": 6.176558157999206,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019621274247586817,
      "loss": 4.8512,
      "mean_token_accuracy": 0.2041942745447159,
      "num_tokens": 136349787.0,
      "step": 62235
    },
    {
      "entropy": 5.649439764022827,
      "epoch": 6.177054386661373,
      "grad_norm": 1.65625,
      "learning_rate": 0.00019617955520227182,
      "loss": 4.9251,
      "mean_token_accuracy": 0.1967655226588249,
      "num_tokens": 136359658.0,
      "step": 62240
    },
    {
      "entropy": 5.644493055343628,
      "epoch": 6.177550615323541,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019614636988325268,
      "loss": 4.9108,
      "mean_token_accuracy": 0.20170880556106568,
      "num_tokens": 136370128.0,
      "step": 62245
    },
    {
      "entropy": 5.739590978622436,
      "epoch": 6.178046843985709,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019611318651963373,
      "loss": 5.0056,
      "mean_token_accuracy": 0.19929995387792587,
      "num_tokens": 136381301.0,
      "step": 62250
    },
    {
      "entropy": 5.72540807723999,
      "epoch": 6.1785430726478765,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019608000511223763,
      "loss": 4.8952,
      "mean_token_accuracy": 0.19877939373254777,
      "num_tokens": 136392147.0,
      "step": 62255
    },
    {
      "entropy": 5.649861526489258,
      "epoch": 6.179039301310044,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019604682566188732,
      "loss": 4.8144,
      "mean_token_accuracy": 0.2076709732413292,
      "num_tokens": 136404205.0,
      "step": 62260
    },
    {
      "entropy": 5.771981287002563,
      "epoch": 6.1795355299722114,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019601364816940574,
      "loss": 4.9166,
      "mean_token_accuracy": 0.2083067551255226,
      "num_tokens": 136415510.0,
      "step": 62265
    },
    {
      "entropy": 5.697856473922729,
      "epoch": 6.180031758634379,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001959804726356154,
      "loss": 4.9015,
      "mean_token_accuracy": 0.19897760897874833,
      "num_tokens": 136426162.0,
      "step": 62270
    },
    {
      "entropy": 5.6751899242401125,
      "epoch": 6.180527987296546,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019594729906133911,
      "loss": 4.8792,
      "mean_token_accuracy": 0.2050700843334198,
      "num_tokens": 136438443.0,
      "step": 62275
    },
    {
      "entropy": 5.672528171539307,
      "epoch": 6.181024215958714,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019591412744739944,
      "loss": 4.857,
      "mean_token_accuracy": 0.19699142426252364,
      "num_tokens": 136450210.0,
      "step": 62280
    },
    {
      "entropy": 5.6693434715271,
      "epoch": 6.181520444620881,
      "grad_norm": 1.40625,
      "learning_rate": 0.000195880957794619,
      "loss": 4.887,
      "mean_token_accuracy": 0.1995425969362259,
      "num_tokens": 136460608.0,
      "step": 62285
    },
    {
      "entropy": 5.699929571151733,
      "epoch": 6.182016673283049,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019584779010382026,
      "loss": 4.9337,
      "mean_token_accuracy": 0.20313896238803864,
      "num_tokens": 136472790.0,
      "step": 62290
    },
    {
      "entropy": 5.687392854690552,
      "epoch": 6.182512901945216,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001958146243758258,
      "loss": 4.8926,
      "mean_token_accuracy": 0.20422655791044236,
      "num_tokens": 136483453.0,
      "step": 62295
    },
    {
      "entropy": 5.696000576019287,
      "epoch": 6.183009130607384,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019578146061145812,
      "loss": 4.8756,
      "mean_token_accuracy": 0.20637920498847961,
      "num_tokens": 136493804.0,
      "step": 62300
    },
    {
      "entropy": 5.6411292552948,
      "epoch": 6.183505359269551,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019574829881153943,
      "loss": 4.868,
      "mean_token_accuracy": 0.20667274445295333,
      "num_tokens": 136504623.0,
      "step": 62305
    },
    {
      "entropy": 5.662030363082886,
      "epoch": 6.184001587931719,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001957151389768922,
      "loss": 4.7951,
      "mean_token_accuracy": 0.2107531726360321,
      "num_tokens": 136515889.0,
      "step": 62310
    },
    {
      "entropy": 5.701466655731201,
      "epoch": 6.184497816593886,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019568198110833868,
      "loss": 4.938,
      "mean_token_accuracy": 0.202451890707016,
      "num_tokens": 136527091.0,
      "step": 62315
    },
    {
      "entropy": 5.621879434585571,
      "epoch": 6.184994045256054,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019564882520670102,
      "loss": 4.8738,
      "mean_token_accuracy": 0.2054089352488518,
      "num_tokens": 136537354.0,
      "step": 62320
    },
    {
      "entropy": 5.705887794494629,
      "epoch": 6.185490273918221,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001956156712728015,
      "loss": 4.9299,
      "mean_token_accuracy": 0.20281882733106613,
      "num_tokens": 136548002.0,
      "step": 62325
    },
    {
      "entropy": 5.7186564922332765,
      "epoch": 6.185986502580389,
      "grad_norm": 1.375,
      "learning_rate": 0.0001955825193074624,
      "loss": 4.9016,
      "mean_token_accuracy": 0.19410171806812287,
      "num_tokens": 136558185.0,
      "step": 62330
    },
    {
      "entropy": 5.685239601135254,
      "epoch": 6.186482731242557,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019554936931150563,
      "loss": 4.8738,
      "mean_token_accuracy": 0.20919760763645173,
      "num_tokens": 136569866.0,
      "step": 62335
    },
    {
      "entropy": 5.726355838775635,
      "epoch": 6.1869789599047245,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019551622128575336,
      "loss": 4.9718,
      "mean_token_accuracy": 0.19238912761211396,
      "num_tokens": 136580176.0,
      "step": 62340
    },
    {
      "entropy": 5.6417442798614506,
      "epoch": 6.187475188566892,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019548307523102747,
      "loss": 4.8622,
      "mean_token_accuracy": 0.20284813046455383,
      "num_tokens": 136591309.0,
      "step": 62345
    },
    {
      "entropy": 5.713158655166626,
      "epoch": 6.187971417229059,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001954499311481499,
      "loss": 4.9355,
      "mean_token_accuracy": 0.19770543575286864,
      "num_tokens": 136603420.0,
      "step": 62350
    },
    {
      "entropy": 5.634771299362183,
      "epoch": 6.188467645891227,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001954167890379427,
      "loss": 4.8576,
      "mean_token_accuracy": 0.2041895344853401,
      "num_tokens": 136614892.0,
      "step": 62355
    },
    {
      "entropy": 5.690328550338745,
      "epoch": 6.188963874553394,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019538364890122774,
      "loss": 4.8965,
      "mean_token_accuracy": 0.20183644592761993,
      "num_tokens": 136626908.0,
      "step": 62360
    },
    {
      "entropy": 5.704567384719849,
      "epoch": 6.189460103215562,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019535051073882664,
      "loss": 4.8768,
      "mean_token_accuracy": 0.20160287320613862,
      "num_tokens": 136638313.0,
      "step": 62365
    },
    {
      "entropy": 5.72185640335083,
      "epoch": 6.189956331877729,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019531737455156118,
      "loss": 4.9574,
      "mean_token_accuracy": 0.20289555937051773,
      "num_tokens": 136650214.0,
      "step": 62370
    },
    {
      "entropy": 5.696155405044555,
      "epoch": 6.190452560539897,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019528424034025323,
      "loss": 4.8995,
      "mean_token_accuracy": 0.20305124223232268,
      "num_tokens": 136661534.0,
      "step": 62375
    },
    {
      "entropy": 5.661320447921753,
      "epoch": 6.190948789202064,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019525110810572426,
      "loss": 4.8389,
      "mean_token_accuracy": 0.19751085788011552,
      "num_tokens": 136671866.0,
      "step": 62380
    },
    {
      "entropy": 5.731723308563232,
      "epoch": 6.191445017864232,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019521797784879602,
      "loss": 4.8762,
      "mean_token_accuracy": 0.20565202534198762,
      "num_tokens": 136683219.0,
      "step": 62385
    },
    {
      "entropy": 5.704385280609131,
      "epoch": 6.191941246526399,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019518484957028998,
      "loss": 4.942,
      "mean_token_accuracy": 0.20516000241041182,
      "num_tokens": 136694167.0,
      "step": 62390
    },
    {
      "entropy": 5.680079078674316,
      "epoch": 6.192437475188567,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019515172327102775,
      "loss": 4.8463,
      "mean_token_accuracy": 0.20322991609573365,
      "num_tokens": 136705061.0,
      "step": 62395
    },
    {
      "entropy": 5.655365085601806,
      "epoch": 6.192933703850734,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019511859895183066,
      "loss": 4.8849,
      "mean_token_accuracy": 0.20622476786375046,
      "num_tokens": 136714995.0,
      "step": 62400
    },
    {
      "entropy": 5.579283666610718,
      "epoch": 6.193429932512902,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019508547661352028,
      "loss": 4.8647,
      "mean_token_accuracy": 0.20989277362823486,
      "num_tokens": 136725853.0,
      "step": 62405
    },
    {
      "entropy": 5.656535911560058,
      "epoch": 6.193926161175069,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00019505235625691774,
      "loss": 4.9142,
      "mean_token_accuracy": 0.2021881803870201,
      "num_tokens": 136738043.0,
      "step": 62410
    },
    {
      "entropy": 5.731587076187134,
      "epoch": 6.194422389837237,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019501923788284458,
      "loss": 5.0348,
      "mean_token_accuracy": 0.19050649255514146,
      "num_tokens": 136748149.0,
      "step": 62415
    },
    {
      "entropy": 5.6834397315979,
      "epoch": 6.194918618499405,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001949861214921219,
      "loss": 4.9173,
      "mean_token_accuracy": 0.20163773596286774,
      "num_tokens": 136759708.0,
      "step": 62420
    },
    {
      "entropy": 5.755815124511718,
      "epoch": 6.195414847161572,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019495300708557113,
      "loss": 4.9479,
      "mean_token_accuracy": 0.19565377831459047,
      "num_tokens": 136771110.0,
      "step": 62425
    },
    {
      "entropy": 5.695738649368286,
      "epoch": 6.19591107582374,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019491989466401322,
      "loss": 4.9541,
      "mean_token_accuracy": 0.19642702043056487,
      "num_tokens": 136781646.0,
      "step": 62430
    },
    {
      "entropy": 5.7026050090789795,
      "epoch": 6.196407304485907,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001948867842282694,
      "loss": 4.906,
      "mean_token_accuracy": 0.19729816168546677,
      "num_tokens": 136790988.0,
      "step": 62435
    },
    {
      "entropy": 5.697010660171509,
      "epoch": 6.196903533148075,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001948536757791606,
      "loss": 4.8689,
      "mean_token_accuracy": 0.2046201005578041,
      "num_tokens": 136802713.0,
      "step": 62440
    },
    {
      "entropy": 5.717421770095825,
      "epoch": 6.197399761810242,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019482056931750802,
      "loss": 4.9119,
      "mean_token_accuracy": 0.20178499519824983,
      "num_tokens": 136814079.0,
      "step": 62445
    },
    {
      "entropy": 5.7398497581481935,
      "epoch": 6.19789599047241,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001947874648441325,
      "loss": 4.9016,
      "mean_token_accuracy": 0.2010807365179062,
      "num_tokens": 136824677.0,
      "step": 62450
    },
    {
      "entropy": 5.7158064365386965,
      "epoch": 6.198392219134577,
      "grad_norm": 1.25,
      "learning_rate": 0.00019475436235985512,
      "loss": 4.8967,
      "mean_token_accuracy": 0.20428072214126586,
      "num_tokens": 136835369.0,
      "step": 62455
    },
    {
      "entropy": 5.6602918148040775,
      "epoch": 6.198888447796745,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001947212618654966,
      "loss": 4.8159,
      "mean_token_accuracy": 0.20499587506055833,
      "num_tokens": 136846158.0,
      "step": 62460
    },
    {
      "entropy": 5.709001874923706,
      "epoch": 6.199384676458912,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019468816336187773,
      "loss": 5.0028,
      "mean_token_accuracy": 0.19787591248750686,
      "num_tokens": 136857829.0,
      "step": 62465
    },
    {
      "entropy": 5.711225461959839,
      "epoch": 6.19988090512108,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001946550668498193,
      "loss": 4.9512,
      "mean_token_accuracy": 0.19878966361284256,
      "num_tokens": 136870429.0,
      "step": 62470
    },
    {
      "entropy": 5.64646544456482,
      "epoch": 6.200377133783247,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019462197233014228,
      "loss": 4.8422,
      "mean_token_accuracy": 0.2019138365983963,
      "num_tokens": 136881180.0,
      "step": 62475
    },
    {
      "entropy": 5.682736873626709,
      "epoch": 6.200873362445415,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019458887980366703,
      "loss": 4.8833,
      "mean_token_accuracy": 0.20568422675132753,
      "num_tokens": 136892506.0,
      "step": 62480
    },
    {
      "entropy": 5.739467668533325,
      "epoch": 6.201369591107582,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001945557892712143,
      "loss": 4.9656,
      "mean_token_accuracy": 0.20051826983690263,
      "num_tokens": 136903708.0,
      "step": 62485
    },
    {
      "entropy": 5.65866174697876,
      "epoch": 6.20186581976975,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019452270073360466,
      "loss": 4.9554,
      "mean_token_accuracy": 0.20066216439008713,
      "num_tokens": 136914353.0,
      "step": 62490
    },
    {
      "entropy": 5.691237831115723,
      "epoch": 6.202362048431917,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019448961419165855,
      "loss": 4.906,
      "mean_token_accuracy": 0.1970837637782097,
      "num_tokens": 136926537.0,
      "step": 62495
    },
    {
      "entropy": 5.701826000213623,
      "epoch": 6.2028582770940845,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019445652964619663,
      "loss": 4.8317,
      "mean_token_accuracy": 0.19982543885707854,
      "num_tokens": 136936279.0,
      "step": 62500
    },
    {
      "entropy": 5.709072208404541,
      "epoch": 6.203354505756252,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019442344709803922,
      "loss": 4.863,
      "mean_token_accuracy": 0.2107880234718323,
      "num_tokens": 136946407.0,
      "step": 62505
    },
    {
      "entropy": 5.630603361129761,
      "epoch": 6.20385073441842,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019439036654800667,
      "loss": 4.8193,
      "mean_token_accuracy": 0.20390178710222245,
      "num_tokens": 136957362.0,
      "step": 62510
    },
    {
      "entropy": 5.639616918563843,
      "epoch": 6.204346963080588,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019435728799691932,
      "loss": 4.827,
      "mean_token_accuracy": 0.21155322343111038,
      "num_tokens": 136968243.0,
      "step": 62515
    },
    {
      "entropy": 5.6337573528289795,
      "epoch": 6.204843191742755,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019432421144559747,
      "loss": 4.8703,
      "mean_token_accuracy": 0.19718268066644667,
      "num_tokens": 136979665.0,
      "step": 62520
    },
    {
      "entropy": 5.7034741878509525,
      "epoch": 6.205339420404923,
      "grad_norm": 1.375,
      "learning_rate": 0.00019429113689486128,
      "loss": 4.979,
      "mean_token_accuracy": 0.18818101584911345,
      "num_tokens": 136990345.0,
      "step": 62525
    },
    {
      "entropy": 5.727999210357666,
      "epoch": 6.20583564906709,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019425806434553107,
      "loss": 4.8956,
      "mean_token_accuracy": 0.19978911578655242,
      "num_tokens": 137001531.0,
      "step": 62530
    },
    {
      "entropy": 5.669075202941895,
      "epoch": 6.206331877729258,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019422499379842677,
      "loss": 4.8614,
      "mean_token_accuracy": 0.20559021979570388,
      "num_tokens": 137011878.0,
      "step": 62535
    },
    {
      "entropy": 5.707058477401733,
      "epoch": 6.206828106391425,
      "grad_norm": 1.609375,
      "learning_rate": 0.00019419192525436875,
      "loss": 4.9875,
      "mean_token_accuracy": 0.2051149696111679,
      "num_tokens": 137022733.0,
      "step": 62540
    },
    {
      "entropy": 5.656683158874512,
      "epoch": 6.207324335053593,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019415885871417677,
      "loss": 4.9322,
      "mean_token_accuracy": 0.19632971584796904,
      "num_tokens": 137034049.0,
      "step": 62545
    },
    {
      "entropy": 5.79048638343811,
      "epoch": 6.20782056371576,
      "grad_norm": 1.5,
      "learning_rate": 0.0001941257941786708,
      "loss": 5.0701,
      "mean_token_accuracy": 0.19508604258298873,
      "num_tokens": 137045838.0,
      "step": 62550
    },
    {
      "entropy": 5.726972770690918,
      "epoch": 6.208316792377928,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001940927316486709,
      "loss": 4.913,
      "mean_token_accuracy": 0.2044660732150078,
      "num_tokens": 137056697.0,
      "step": 62555
    },
    {
      "entropy": 5.732287216186523,
      "epoch": 6.208813021040095,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000194059671124997,
      "loss": 4.8647,
      "mean_token_accuracy": 0.20572065114974974,
      "num_tokens": 137066900.0,
      "step": 62560
    },
    {
      "entropy": 5.697673416137695,
      "epoch": 6.209309249702263,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001940266126084688,
      "loss": 4.9102,
      "mean_token_accuracy": 0.20169590264558793,
      "num_tokens": 137078797.0,
      "step": 62565
    },
    {
      "entropy": 5.6629761219024655,
      "epoch": 6.20980547836443,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001939935560999061,
      "loss": 4.857,
      "mean_token_accuracy": 0.20939727872610092,
      "num_tokens": 137091573.0,
      "step": 62570
    },
    {
      "entropy": 5.681825399398804,
      "epoch": 6.210301707026598,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019396050160012875,
      "loss": 4.9022,
      "mean_token_accuracy": 0.21060024201869965,
      "num_tokens": 137102786.0,
      "step": 62575
    },
    {
      "entropy": 5.761125564575195,
      "epoch": 6.210797935688765,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00019392744910995624,
      "loss": 5.0078,
      "mean_token_accuracy": 0.19316853880882262,
      "num_tokens": 137114636.0,
      "step": 62580
    },
    {
      "entropy": 5.7033477306365965,
      "epoch": 6.2112941643509325,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00019389439863020825,
      "loss": 4.959,
      "mean_token_accuracy": 0.19527039527893067,
      "num_tokens": 137126443.0,
      "step": 62585
    },
    {
      "entropy": 5.761886024475098,
      "epoch": 6.2117903930131,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019386135016170454,
      "loss": 4.99,
      "mean_token_accuracy": 0.18888258785009385,
      "num_tokens": 137136413.0,
      "step": 62590
    },
    {
      "entropy": 5.741451978683472,
      "epoch": 6.212286621675268,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019382830370526444,
      "loss": 4.9588,
      "mean_token_accuracy": 0.19538117349147796,
      "num_tokens": 137146411.0,
      "step": 62595
    },
    {
      "entropy": 5.709322357177735,
      "epoch": 6.212782850337436,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019379525926170748,
      "loss": 4.9398,
      "mean_token_accuracy": 0.20293351262807846,
      "num_tokens": 137158179.0,
      "step": 62600
    },
    {
      "entropy": 5.655697917938232,
      "epoch": 6.213279078999603,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001937622168318532,
      "loss": 4.9818,
      "mean_token_accuracy": 0.1974427133798599,
      "num_tokens": 137169411.0,
      "step": 62605
    },
    {
      "entropy": 5.688670444488525,
      "epoch": 6.213775307661771,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001937291764165209,
      "loss": 4.866,
      "mean_token_accuracy": 0.20744286775588988,
      "num_tokens": 137178948.0,
      "step": 62610
    },
    {
      "entropy": 5.638241004943848,
      "epoch": 6.214271536323938,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001936961380165297,
      "loss": 4.837,
      "mean_token_accuracy": 0.211682690680027,
      "num_tokens": 137190085.0,
      "step": 62615
    },
    {
      "entropy": 5.670615148544312,
      "epoch": 6.214767764986106,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019366310163269922,
      "loss": 4.8599,
      "mean_token_accuracy": 0.20003947466611863,
      "num_tokens": 137200709.0,
      "step": 62620
    },
    {
      "entropy": 5.681040096282959,
      "epoch": 6.215263993648273,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019363006726584852,
      "loss": 4.8806,
      "mean_token_accuracy": 0.19762954115867615,
      "num_tokens": 137212197.0,
      "step": 62625
    },
    {
      "entropy": 5.720522069931031,
      "epoch": 6.215760222310441,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019359703491679685,
      "loss": 4.9201,
      "mean_token_accuracy": 0.19425335228443147,
      "num_tokens": 137222453.0,
      "step": 62630
    },
    {
      "entropy": 5.6615917682647705,
      "epoch": 6.216256450972608,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019356400458636337,
      "loss": 4.8847,
      "mean_token_accuracy": 0.20505349040031434,
      "num_tokens": 137232242.0,
      "step": 62635
    },
    {
      "entropy": 5.6422148704528805,
      "epoch": 6.216752679634776,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019353097627536702,
      "loss": 4.8742,
      "mean_token_accuracy": 0.20629259198904037,
      "num_tokens": 137242205.0,
      "step": 62640
    },
    {
      "entropy": 5.673815774917602,
      "epoch": 6.217248908296943,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019349794998462682,
      "loss": 4.9043,
      "mean_token_accuracy": 0.20107100903987885,
      "num_tokens": 137252295.0,
      "step": 62645
    },
    {
      "entropy": 5.731435298919678,
      "epoch": 6.217745136959111,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019346492571496192,
      "loss": 4.8932,
      "mean_token_accuracy": 0.20344675779342652,
      "num_tokens": 137263562.0,
      "step": 62650
    },
    {
      "entropy": 5.729282665252685,
      "epoch": 6.218241365621278,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019343190346719126,
      "loss": 4.9184,
      "mean_token_accuracy": 0.19909863471984862,
      "num_tokens": 137274086.0,
      "step": 62655
    },
    {
      "entropy": 5.712268686294555,
      "epoch": 6.2187375942834455,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019339888324213357,
      "loss": 4.8998,
      "mean_token_accuracy": 0.1987889215350151,
      "num_tokens": 137285882.0,
      "step": 62660
    },
    {
      "entropy": 5.681636810302734,
      "epoch": 6.219233822945613,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019336586504060772,
      "loss": 4.9392,
      "mean_token_accuracy": 0.19523107409477233,
      "num_tokens": 137295212.0,
      "step": 62665
    },
    {
      "entropy": 5.737975263595581,
      "epoch": 6.2197300516077805,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019333284886343256,
      "loss": 4.9657,
      "mean_token_accuracy": 0.20107192695140838,
      "num_tokens": 137306522.0,
      "step": 62670
    },
    {
      "entropy": 5.728535461425781,
      "epoch": 6.220226280269948,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019329983471142665,
      "loss": 4.9022,
      "mean_token_accuracy": 0.20259182304143905,
      "num_tokens": 137318187.0,
      "step": 62675
    },
    {
      "entropy": 5.709632539749146,
      "epoch": 6.220722508932116,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019326682258540885,
      "loss": 4.8258,
      "mean_token_accuracy": 0.2076668530702591,
      "num_tokens": 137328562.0,
      "step": 62680
    },
    {
      "entropy": 5.712702131271362,
      "epoch": 6.221218737594284,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019323381248619774,
      "loss": 4.9439,
      "mean_token_accuracy": 0.19732068181037904,
      "num_tokens": 137339000.0,
      "step": 62685
    },
    {
      "entropy": 5.639112520217895,
      "epoch": 6.221714966256451,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019320080441461197,
      "loss": 4.8629,
      "mean_token_accuracy": 0.20918402820825577,
      "num_tokens": 137350110.0,
      "step": 62690
    },
    {
      "entropy": 5.76972017288208,
      "epoch": 6.222211194918619,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019316779837146985,
      "loss": 5.0636,
      "mean_token_accuracy": 0.191653373837471,
      "num_tokens": 137360392.0,
      "step": 62695
    },
    {
      "entropy": 5.722178888320923,
      "epoch": 6.222707423580786,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019313479435759012,
      "loss": 4.9246,
      "mean_token_accuracy": 0.2092644080519676,
      "num_tokens": 137373356.0,
      "step": 62700
    },
    {
      "entropy": 5.70400710105896,
      "epoch": 6.223203652242954,
      "grad_norm": 1.125,
      "learning_rate": 0.00019310179237379095,
      "loss": 4.9383,
      "mean_token_accuracy": 0.19909267127513885,
      "num_tokens": 137385664.0,
      "step": 62705
    },
    {
      "entropy": 5.69390377998352,
      "epoch": 6.223699880905121,
      "grad_norm": 1.296875,
      "learning_rate": 0.000193068792420891,
      "loss": 4.8682,
      "mean_token_accuracy": 0.20250563323497772,
      "num_tokens": 137396151.0,
      "step": 62710
    },
    {
      "entropy": 5.639114141464233,
      "epoch": 6.224196109567289,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019303579449970833,
      "loss": 4.7962,
      "mean_token_accuracy": 0.21196747422218323,
      "num_tokens": 137407134.0,
      "step": 62715
    },
    {
      "entropy": 5.754920816421508,
      "epoch": 6.224692338229456,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001930027986110614,
      "loss": 5.0022,
      "mean_token_accuracy": 0.19138982892036438,
      "num_tokens": 137417994.0,
      "step": 62720
    },
    {
      "entropy": 5.7256481647491455,
      "epoch": 6.225188566891624,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019296980475576843,
      "loss": 4.9665,
      "mean_token_accuracy": 0.19528300315141678,
      "num_tokens": 137428493.0,
      "step": 62725
    },
    {
      "entropy": 5.671507835388184,
      "epoch": 6.225684795553791,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001929368129346475,
      "loss": 4.8967,
      "mean_token_accuracy": 0.20224788337945937,
      "num_tokens": 137440617.0,
      "step": 62730
    },
    {
      "entropy": 5.681618356704712,
      "epoch": 6.2261810242159585,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019290382314851663,
      "loss": 4.8651,
      "mean_token_accuracy": 0.20649808198213576,
      "num_tokens": 137450861.0,
      "step": 62735
    },
    {
      "entropy": 5.677626895904541,
      "epoch": 6.226677252878126,
      "grad_norm": 1.375,
      "learning_rate": 0.0001928708353981942,
      "loss": 4.854,
      "mean_token_accuracy": 0.20351926982402802,
      "num_tokens": 137462496.0,
      "step": 62740
    },
    {
      "entropy": 5.687057542800903,
      "epoch": 6.2271734815402935,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001928378496844981,
      "loss": 4.8926,
      "mean_token_accuracy": 0.19714183062314988,
      "num_tokens": 137472240.0,
      "step": 62745
    },
    {
      "entropy": 5.6723682403564455,
      "epoch": 6.227669710202461,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019280486600824627,
      "loss": 4.8853,
      "mean_token_accuracy": 0.20061819851398469,
      "num_tokens": 137483104.0,
      "step": 62750
    },
    {
      "entropy": 5.690190839767456,
      "epoch": 6.228165938864628,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001927718843702566,
      "loss": 4.88,
      "mean_token_accuracy": 0.20736555904150009,
      "num_tokens": 137493234.0,
      "step": 62755
    },
    {
      "entropy": 5.714555549621582,
      "epoch": 6.228662167526796,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019273890477134698,
      "loss": 4.8913,
      "mean_token_accuracy": 0.2019403249025345,
      "num_tokens": 137504990.0,
      "step": 62760
    },
    {
      "entropy": 5.676735830307007,
      "epoch": 6.229158396188964,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019270592721233533,
      "loss": 4.9356,
      "mean_token_accuracy": 0.19902696460485458,
      "num_tokens": 137515939.0,
      "step": 62765
    },
    {
      "entropy": 5.649839115142822,
      "epoch": 6.229654624851132,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019267295169403937,
      "loss": 4.883,
      "mean_token_accuracy": 0.20406474620103837,
      "num_tokens": 137525843.0,
      "step": 62770
    },
    {
      "entropy": 5.654269886016846,
      "epoch": 6.230150853513299,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019263997821727675,
      "loss": 4.9065,
      "mean_token_accuracy": 0.19881955683231353,
      "num_tokens": 137536407.0,
      "step": 62775
    },
    {
      "entropy": 5.708767366409302,
      "epoch": 6.230647082175467,
      "grad_norm": 1.671875,
      "learning_rate": 0.00019260700678286525,
      "loss": 4.9214,
      "mean_token_accuracy": 0.1952758476138115,
      "num_tokens": 137545595.0,
      "step": 62780
    },
    {
      "entropy": 5.667197227478027,
      "epoch": 6.231143310837634,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019257403739162245,
      "loss": 4.9027,
      "mean_token_accuracy": 0.19897820204496383,
      "num_tokens": 137556739.0,
      "step": 62785
    },
    {
      "entropy": 5.6965131759643555,
      "epoch": 6.231639539499802,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019254107004436583,
      "loss": 4.9406,
      "mean_token_accuracy": 0.19852906614542007,
      "num_tokens": 137568497.0,
      "step": 62790
    },
    {
      "entropy": 5.712039804458618,
      "epoch": 6.232135768161969,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019250810474191304,
      "loss": 4.9019,
      "mean_token_accuracy": 0.19817545264959335,
      "num_tokens": 137578746.0,
      "step": 62795
    },
    {
      "entropy": 5.726601839065552,
      "epoch": 6.232631996824137,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001924751414850815,
      "loss": 4.9057,
      "mean_token_accuracy": 0.20305635780096054,
      "num_tokens": 137590059.0,
      "step": 62800
    },
    {
      "entropy": 5.682858276367187,
      "epoch": 6.233128225486304,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019244218027468862,
      "loss": 4.8631,
      "mean_token_accuracy": 0.203788922727108,
      "num_tokens": 137601291.0,
      "step": 62805
    },
    {
      "entropy": 5.680079555511474,
      "epoch": 6.2336244541484715,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019240922111155178,
      "loss": 4.8855,
      "mean_token_accuracy": 0.20207413136959076,
      "num_tokens": 137613306.0,
      "step": 62810
    },
    {
      "entropy": 5.704284143447876,
      "epoch": 6.234120682810639,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019237626399648834,
      "loss": 4.9069,
      "mean_token_accuracy": 0.20035467743873597,
      "num_tokens": 137624006.0,
      "step": 62815
    },
    {
      "entropy": 5.780747365951538,
      "epoch": 6.2346169114728065,
      "grad_norm": 1.375,
      "learning_rate": 0.00019234330893031537,
      "loss": 4.9865,
      "mean_token_accuracy": 0.19898549914360047,
      "num_tokens": 137634309.0,
      "step": 62820
    },
    {
      "entropy": 5.690260648727417,
      "epoch": 6.235113140134974,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019231035591385037,
      "loss": 4.9262,
      "mean_token_accuracy": 0.21024973839521408,
      "num_tokens": 137645810.0,
      "step": 62825
    },
    {
      "entropy": 5.707040023803711,
      "epoch": 6.2356093687971414,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001922774049479103,
      "loss": 4.8857,
      "mean_token_accuracy": 0.2003293126821518,
      "num_tokens": 137656623.0,
      "step": 62830
    },
    {
      "entropy": 5.679150390625,
      "epoch": 6.236105597459309,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019224445603331237,
      "loss": 4.8805,
      "mean_token_accuracy": 0.20495698601007462,
      "num_tokens": 137667358.0,
      "step": 62835
    },
    {
      "entropy": 5.610935115814209,
      "epoch": 6.236601826121476,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001922115091708736,
      "loss": 4.8084,
      "mean_token_accuracy": 0.20641179531812667,
      "num_tokens": 137678353.0,
      "step": 62840
    },
    {
      "entropy": 5.670429039001465,
      "epoch": 6.237098054783644,
      "grad_norm": 1.296875,
      "learning_rate": 0.000192178564361411,
      "loss": 4.8951,
      "mean_token_accuracy": 0.20163182616233827,
      "num_tokens": 137688684.0,
      "step": 62845
    },
    {
      "entropy": 5.639734077453613,
      "epoch": 6.237594283445812,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019214562160574145,
      "loss": 4.8765,
      "mean_token_accuracy": 0.20732534527778626,
      "num_tokens": 137699420.0,
      "step": 62850
    },
    {
      "entropy": 5.69889988899231,
      "epoch": 6.23809051210798,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001921126809046821,
      "loss": 4.9288,
      "mean_token_accuracy": 0.20328554064035415,
      "num_tokens": 137711088.0,
      "step": 62855
    },
    {
      "entropy": 5.665953874588013,
      "epoch": 6.238586740770147,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019207974225904955,
      "loss": 4.8631,
      "mean_token_accuracy": 0.20365960001945496,
      "num_tokens": 137722038.0,
      "step": 62860
    },
    {
      "entropy": 5.66584210395813,
      "epoch": 6.239082969432315,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019204680566966076,
      "loss": 4.8304,
      "mean_token_accuracy": 0.21268026679754257,
      "num_tokens": 137733387.0,
      "step": 62865
    },
    {
      "entropy": 5.7363965034484865,
      "epoch": 6.239579198094482,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019201387113733248,
      "loss": 4.8785,
      "mean_token_accuracy": 0.20527564436197282,
      "num_tokens": 137744439.0,
      "step": 62870
    },
    {
      "entropy": 5.665463304519653,
      "epoch": 6.24007542675665,
      "grad_norm": 1.375,
      "learning_rate": 0.00019198093866288135,
      "loss": 4.8206,
      "mean_token_accuracy": 0.2101377546787262,
      "num_tokens": 137754270.0,
      "step": 62875
    },
    {
      "entropy": 5.607282590866089,
      "epoch": 6.240571655418817,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019194800824712398,
      "loss": 4.8473,
      "mean_token_accuracy": 0.2074152112007141,
      "num_tokens": 137764665.0,
      "step": 62880
    },
    {
      "entropy": 5.683204269409179,
      "epoch": 6.241067884080985,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019191507989087708,
      "loss": 4.9307,
      "mean_token_accuracy": 0.1930494487285614,
      "num_tokens": 137774855.0,
      "step": 62885
    },
    {
      "entropy": 5.767612266540527,
      "epoch": 6.241564112743152,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019188215359495725,
      "loss": 4.9311,
      "mean_token_accuracy": 0.20347122997045516,
      "num_tokens": 137784917.0,
      "step": 62890
    },
    {
      "entropy": 5.724046611785889,
      "epoch": 6.2420603414053195,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019184922936018084,
      "loss": 4.8803,
      "mean_token_accuracy": 0.20211491286754607,
      "num_tokens": 137795029.0,
      "step": 62895
    },
    {
      "entropy": 5.684124374389649,
      "epoch": 6.242556570067487,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019181630718736437,
      "loss": 4.8635,
      "mean_token_accuracy": 0.20063506066799164,
      "num_tokens": 137805865.0,
      "step": 62900
    },
    {
      "entropy": 5.654435157775879,
      "epoch": 6.2430527987296545,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019178338707732424,
      "loss": 4.9251,
      "mean_token_accuracy": 0.190389247238636,
      "num_tokens": 137816738.0,
      "step": 62905
    },
    {
      "entropy": 5.729090881347656,
      "epoch": 6.243549027391822,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019175046903087677,
      "loss": 4.9932,
      "mean_token_accuracy": 0.18965789079666137,
      "num_tokens": 137827692.0,
      "step": 62910
    },
    {
      "entropy": 5.733955574035645,
      "epoch": 6.244045256053989,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001917175530488383,
      "loss": 4.9414,
      "mean_token_accuracy": 0.20121438652276993,
      "num_tokens": 137838143.0,
      "step": 62915
    },
    {
      "entropy": 5.81390700340271,
      "epoch": 6.244541484716157,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001916846391320251,
      "loss": 4.9915,
      "mean_token_accuracy": 0.1915248915553093,
      "num_tokens": 137850641.0,
      "step": 62920
    },
    {
      "entropy": 5.695261430740357,
      "epoch": 6.245037713378324,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019165172728125325,
      "loss": 4.8444,
      "mean_token_accuracy": 0.21281609386205674,
      "num_tokens": 137861133.0,
      "step": 62925
    },
    {
      "entropy": 5.659260892868042,
      "epoch": 6.245533942040492,
      "grad_norm": 1.328125,
      "learning_rate": 0.000191618817497339,
      "loss": 4.9167,
      "mean_token_accuracy": 0.20094117820262908,
      "num_tokens": 137871522.0,
      "step": 62930
    },
    {
      "entropy": 5.6552611827850345,
      "epoch": 6.24603017070266,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019158590978109847,
      "loss": 4.8468,
      "mean_token_accuracy": 0.19930476695299149,
      "num_tokens": 137882157.0,
      "step": 62935
    },
    {
      "entropy": 5.679531383514404,
      "epoch": 6.246526399364828,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001915530041333475,
      "loss": 4.9229,
      "mean_token_accuracy": 0.20182065665721893,
      "num_tokens": 137894894.0,
      "step": 62940
    },
    {
      "entropy": 5.711005163192749,
      "epoch": 6.247022628026995,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019152010055490222,
      "loss": 4.8959,
      "mean_token_accuracy": 0.20434650778770447,
      "num_tokens": 137905277.0,
      "step": 62945
    },
    {
      "entropy": 5.688394165039062,
      "epoch": 6.247518856689163,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019148719904657868,
      "loss": 4.8739,
      "mean_token_accuracy": 0.20820943266153336,
      "num_tokens": 137916269.0,
      "step": 62950
    },
    {
      "entropy": 5.671181011199951,
      "epoch": 6.24801508535133,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019145429960919263,
      "loss": 4.8969,
      "mean_token_accuracy": 0.20016213804483413,
      "num_tokens": 137925993.0,
      "step": 62955
    },
    {
      "entropy": 5.675709772109985,
      "epoch": 6.248511314013498,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001914214022435599,
      "loss": 4.8918,
      "mean_token_accuracy": 0.19978659152984618,
      "num_tokens": 137936382.0,
      "step": 62960
    },
    {
      "entropy": 5.615249919891357,
      "epoch": 6.249007542675665,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001913885069504963,
      "loss": 4.8544,
      "mean_token_accuracy": 0.20574960559606553,
      "num_tokens": 137948865.0,
      "step": 62965
    },
    {
      "entropy": 5.657465648651123,
      "epoch": 6.2495037713378325,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019135561373081746,
      "loss": 4.837,
      "mean_token_accuracy": 0.21316928565502166,
      "num_tokens": 137960073.0,
      "step": 62970
    },
    {
      "entropy": 5.641042232513428,
      "epoch": 6.25,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019132272258533923,
      "loss": 4.9065,
      "mean_token_accuracy": 0.2006911039352417,
      "num_tokens": 137971082.0,
      "step": 62975
    },
    {
      "entropy": 5.746663808822632,
      "epoch": 6.2504962286621675,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001912898335148772,
      "loss": 4.9926,
      "mean_token_accuracy": 0.19163846373558044,
      "num_tokens": 137981475.0,
      "step": 62980
    },
    {
      "entropy": 5.708488845825196,
      "epoch": 6.250992457324335,
      "grad_norm": 1.390625,
      "learning_rate": 0.000191256946520247,
      "loss": 4.8576,
      "mean_token_accuracy": 0.2020548775792122,
      "num_tokens": 137992622.0,
      "step": 62985
    },
    {
      "entropy": 5.668320178985596,
      "epoch": 6.251488685986502,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019122406160226396,
      "loss": 4.8566,
      "mean_token_accuracy": 0.20765987187623977,
      "num_tokens": 138003956.0,
      "step": 62990
    },
    {
      "entropy": 5.696124124526977,
      "epoch": 6.25198491464867,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019119117876174364,
      "loss": 4.8759,
      "mean_token_accuracy": 0.20401666462421417,
      "num_tokens": 138014759.0,
      "step": 62995
    },
    {
      "entropy": 5.730753707885742,
      "epoch": 6.252481143310837,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019115829799950157,
      "loss": 4.9188,
      "mean_token_accuracy": 0.19788380116224288,
      "num_tokens": 138025669.0,
      "step": 63000
    },
    {
      "epoch": 6.252481143310837,
      "eval_entropy": 5.457750020442934,
      "eval_loss": 5.070143222808838,
      "eval_mean_token_accuracy": 0.19760292278676622,
      "eval_num_tokens": 138025669.0,
      "eval_runtime": 20.4858,
      "eval_samples_per_second": 1643.532,
      "eval_steps_per_second": 205.46,
      "step": 63000
    },
    {
      "entropy": 5.667622470855713,
      "epoch": 6.252977371973005,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019112541931635307,
      "loss": 4.8815,
      "mean_token_accuracy": 0.20519034266471864,
      "num_tokens": 138038470.0,
      "step": 63005
    },
    {
      "entropy": 5.760633897781372,
      "epoch": 6.253473600635172,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019109254271311337,
      "loss": 4.9649,
      "mean_token_accuracy": 0.19742030054330825,
      "num_tokens": 138048486.0,
      "step": 63010
    },
    {
      "entropy": 5.766771841049194,
      "epoch": 6.25396982929734,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019105966819059784,
      "loss": 5.0348,
      "mean_token_accuracy": 0.19691794365644455,
      "num_tokens": 138060032.0,
      "step": 63015
    },
    {
      "entropy": 5.686627817153931,
      "epoch": 6.254466057959508,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019102679574962167,
      "loss": 4.9089,
      "mean_token_accuracy": 0.20560417622327803,
      "num_tokens": 138070583.0,
      "step": 63020
    },
    {
      "entropy": 5.753935194015503,
      "epoch": 6.254962286621676,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019099392539100003,
      "loss": 5.0235,
      "mean_token_accuracy": 0.1948859065771103,
      "num_tokens": 138080834.0,
      "step": 63025
    },
    {
      "entropy": 5.76933798789978,
      "epoch": 6.255458515283843,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019096105711554796,
      "loss": 4.9127,
      "mean_token_accuracy": 0.19788138717412948,
      "num_tokens": 138091706.0,
      "step": 63030
    },
    {
      "entropy": 5.6490137577056885,
      "epoch": 6.255954743946011,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019092819092408065,
      "loss": 4.8738,
      "mean_token_accuracy": 0.20126689225435257,
      "num_tokens": 138102499.0,
      "step": 63035
    },
    {
      "entropy": 5.69964861869812,
      "epoch": 6.256450972608178,
      "grad_norm": 1.203125,
      "learning_rate": 0.000190895326817413,
      "loss": 4.9915,
      "mean_token_accuracy": 0.19642421007156372,
      "num_tokens": 138115257.0,
      "step": 63040
    },
    {
      "entropy": 5.704549264907837,
      "epoch": 6.2569472012703455,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019086246479636004,
      "loss": 4.9363,
      "mean_token_accuracy": 0.19941162168979645,
      "num_tokens": 138126292.0,
      "step": 63045
    },
    {
      "entropy": 5.674832677841186,
      "epoch": 6.257443429932513,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019082960486173668,
      "loss": 4.8812,
      "mean_token_accuracy": 0.20211059004068374,
      "num_tokens": 138137921.0,
      "step": 63050
    },
    {
      "entropy": 5.712151288986206,
      "epoch": 6.2579396585946805,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019079674701435767,
      "loss": 4.984,
      "mean_token_accuracy": 0.19808993935585023,
      "num_tokens": 138149469.0,
      "step": 63055
    },
    {
      "entropy": 5.742129945755005,
      "epoch": 6.258435887256848,
      "grad_norm": 1.265625,
      "learning_rate": 0.00019076389125503797,
      "loss": 5.0142,
      "mean_token_accuracy": 0.19004351645708084,
      "num_tokens": 138161687.0,
      "step": 63060
    },
    {
      "entropy": 5.79317717552185,
      "epoch": 6.258932115919015,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001907310375845922,
      "loss": 5.0653,
      "mean_token_accuracy": 0.19056570082902907,
      "num_tokens": 138174732.0,
      "step": 63065
    },
    {
      "entropy": 5.749254989624023,
      "epoch": 6.259428344581183,
      "grad_norm": 1.53125,
      "learning_rate": 0.00019069818600383516,
      "loss": 5.0199,
      "mean_token_accuracy": 0.19520462453365325,
      "num_tokens": 138187117.0,
      "step": 63070
    },
    {
      "entropy": 5.6381019115447994,
      "epoch": 6.25992457324335,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019066533651358142,
      "loss": 4.7586,
      "mean_token_accuracy": 0.22069377303123475,
      "num_tokens": 138197975.0,
      "step": 63075
    },
    {
      "entropy": 5.724153614044189,
      "epoch": 6.260420801905518,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019063248911464564,
      "loss": 4.9182,
      "mean_token_accuracy": 0.20295796990394593,
      "num_tokens": 138208788.0,
      "step": 63080
    },
    {
      "entropy": 5.714772891998291,
      "epoch": 6.260917030567685,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00019059964380784228,
      "loss": 4.9041,
      "mean_token_accuracy": 0.2052178293466568,
      "num_tokens": 138220892.0,
      "step": 63085
    },
    {
      "entropy": 5.741413116455078,
      "epoch": 6.261413259229853,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001905668005939859,
      "loss": 5.0104,
      "mean_token_accuracy": 0.1881946951150894,
      "num_tokens": 138233065.0,
      "step": 63090
    },
    {
      "entropy": 5.678645133972168,
      "epoch": 6.26190948789202,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019053395947389102,
      "loss": 4.8404,
      "mean_token_accuracy": 0.2022772639989853,
      "num_tokens": 138243779.0,
      "step": 63095
    },
    {
      "entropy": 5.680841541290283,
      "epoch": 6.262405716554188,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019050112044837186,
      "loss": 4.8864,
      "mean_token_accuracy": 0.19933129847049713,
      "num_tokens": 138254128.0,
      "step": 63100
    },
    {
      "entropy": 5.633296251296997,
      "epoch": 6.262901945216356,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001904682835182428,
      "loss": 4.7713,
      "mean_token_accuracy": 0.21030991077423095,
      "num_tokens": 138264431.0,
      "step": 63105
    },
    {
      "entropy": 5.652439260482788,
      "epoch": 6.263398173878524,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019043544868431827,
      "loss": 4.8458,
      "mean_token_accuracy": 0.19924398958683015,
      "num_tokens": 138274910.0,
      "step": 63110
    },
    {
      "entropy": 5.675093746185302,
      "epoch": 6.263894402540691,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001904026159474122,
      "loss": 4.8649,
      "mean_token_accuracy": 0.20893218964338303,
      "num_tokens": 138284852.0,
      "step": 63115
    },
    {
      "entropy": 5.620331573486328,
      "epoch": 6.2643906312028586,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019036978530833916,
      "loss": 4.8505,
      "mean_token_accuracy": 0.20100757479667664,
      "num_tokens": 138294797.0,
      "step": 63120
    },
    {
      "entropy": 5.668729829788208,
      "epoch": 6.264886859865026,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019033695676791296,
      "loss": 4.8782,
      "mean_token_accuracy": 0.20781537890434265,
      "num_tokens": 138305982.0,
      "step": 63125
    },
    {
      "entropy": 5.675786018371582,
      "epoch": 6.2653830885271935,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019030413032694788,
      "loss": 4.8591,
      "mean_token_accuracy": 0.20820656716823577,
      "num_tokens": 138316553.0,
      "step": 63130
    },
    {
      "entropy": 5.651567554473877,
      "epoch": 6.265879317189361,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019027130598625786,
      "loss": 4.8811,
      "mean_token_accuracy": 0.20179155617952346,
      "num_tokens": 138327911.0,
      "step": 63135
    },
    {
      "entropy": 5.719608974456787,
      "epoch": 6.2663755458515285,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019023848374665687,
      "loss": 4.9058,
      "mean_token_accuracy": 0.20034659951925277,
      "num_tokens": 138337950.0,
      "step": 63140
    },
    {
      "entropy": 5.6682013988494875,
      "epoch": 6.266871774513696,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00019020566360895876,
      "loss": 4.8914,
      "mean_token_accuracy": 0.20050593167543412,
      "num_tokens": 138347434.0,
      "step": 63145
    },
    {
      "entropy": 5.692667531967163,
      "epoch": 6.267368003175863,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019017284557397752,
      "loss": 4.8923,
      "mean_token_accuracy": 0.20619749873876572,
      "num_tokens": 138357824.0,
      "step": 63150
    },
    {
      "entropy": 5.669407892227173,
      "epoch": 6.267864231838031,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019014002964252698,
      "loss": 4.843,
      "mean_token_accuracy": 0.1994778037071228,
      "num_tokens": 138369283.0,
      "step": 63155
    },
    {
      "entropy": 5.713550186157226,
      "epoch": 6.268360460500198,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019010721581542078,
      "loss": 4.9214,
      "mean_token_accuracy": 0.20001482516527175,
      "num_tokens": 138380796.0,
      "step": 63160
    },
    {
      "entropy": 5.68416657447815,
      "epoch": 6.268856689162366,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019007440409347275,
      "loss": 4.8821,
      "mean_token_accuracy": 0.20572631508111955,
      "num_tokens": 138391733.0,
      "step": 63165
    },
    {
      "entropy": 5.727869081497192,
      "epoch": 6.269352917824533,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019004159447749648,
      "loss": 4.9348,
      "mean_token_accuracy": 0.19829776436090468,
      "num_tokens": 138403694.0,
      "step": 63170
    },
    {
      "entropy": 5.7062084674835205,
      "epoch": 6.269849146486701,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019000878696830555,
      "loss": 4.899,
      "mean_token_accuracy": 0.2048858806490898,
      "num_tokens": 138414659.0,
      "step": 63175
    },
    {
      "entropy": 5.67007417678833,
      "epoch": 6.270345375148868,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001899759815667136,
      "loss": 4.9073,
      "mean_token_accuracy": 0.19722485095262526,
      "num_tokens": 138426346.0,
      "step": 63180
    },
    {
      "entropy": 5.71921181678772,
      "epoch": 6.270841603811036,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018994317827353418,
      "loss": 4.9704,
      "mean_token_accuracy": 0.19650205820798874,
      "num_tokens": 138436867.0,
      "step": 63185
    },
    {
      "entropy": 5.758481168746949,
      "epoch": 6.271337832473204,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018991037708958054,
      "loss": 5.008,
      "mean_token_accuracy": 0.19590024650096893,
      "num_tokens": 138448537.0,
      "step": 63190
    },
    {
      "entropy": 5.68446798324585,
      "epoch": 6.271834061135372,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018987757801566625,
      "loss": 4.8535,
      "mean_token_accuracy": 0.20463480949401855,
      "num_tokens": 138459461.0,
      "step": 63195
    },
    {
      "entropy": 5.730010366439819,
      "epoch": 6.272330289797539,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001898447810526046,
      "loss": 4.9077,
      "mean_token_accuracy": 0.20648758560419084,
      "num_tokens": 138470353.0,
      "step": 63200
    },
    {
      "entropy": 5.630303239822387,
      "epoch": 6.2728265184597065,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018981198620120882,
      "loss": 4.8661,
      "mean_token_accuracy": 0.20743145942687988,
      "num_tokens": 138480777.0,
      "step": 63205
    },
    {
      "entropy": 5.720454311370849,
      "epoch": 6.273322747121874,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018977919346229227,
      "loss": 4.9557,
      "mean_token_accuracy": 0.19140737354755402,
      "num_tokens": 138492615.0,
      "step": 63210
    },
    {
      "entropy": 5.723819065093994,
      "epoch": 6.2738189757840415,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018974640283666815,
      "loss": 4.9058,
      "mean_token_accuracy": 0.19961764812469482,
      "num_tokens": 138505552.0,
      "step": 63215
    },
    {
      "entropy": 5.661168098449707,
      "epoch": 6.274315204446209,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018971361432514944,
      "loss": 4.8102,
      "mean_token_accuracy": 0.21493198573589326,
      "num_tokens": 138515478.0,
      "step": 63220
    },
    {
      "entropy": 5.721432495117187,
      "epoch": 6.274811433108376,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001896808279285494,
      "loss": 4.9063,
      "mean_token_accuracy": 0.1964477702975273,
      "num_tokens": 138525795.0,
      "step": 63225
    },
    {
      "entropy": 5.635205411911011,
      "epoch": 6.275307661770544,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018964804364768095,
      "loss": 4.842,
      "mean_token_accuracy": 0.19973498582839966,
      "num_tokens": 138537786.0,
      "step": 63230
    },
    {
      "entropy": 5.6574461460113525,
      "epoch": 6.275803890432711,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018961526148335707,
      "loss": 4.8818,
      "mean_token_accuracy": 0.2047334298491478,
      "num_tokens": 138548727.0,
      "step": 63235
    },
    {
      "entropy": 5.720238780975341,
      "epoch": 6.276300119094879,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001895824814363907,
      "loss": 4.9035,
      "mean_token_accuracy": 0.20587449818849562,
      "num_tokens": 138559423.0,
      "step": 63240
    },
    {
      "entropy": 5.712627840042114,
      "epoch": 6.276796347757046,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018954970350759477,
      "loss": 4.9514,
      "mean_token_accuracy": 0.19572920054197313,
      "num_tokens": 138570319.0,
      "step": 63245
    },
    {
      "entropy": 5.720590209960937,
      "epoch": 6.277292576419214,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018951692769778213,
      "loss": 4.9567,
      "mean_token_accuracy": 0.1952232077717781,
      "num_tokens": 138582554.0,
      "step": 63250
    },
    {
      "entropy": 5.727817726135254,
      "epoch": 6.277788805081381,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018948415400776535,
      "loss": 4.9085,
      "mean_token_accuracy": 0.1962484449148178,
      "num_tokens": 138594024.0,
      "step": 63255
    },
    {
      "entropy": 5.648039197921753,
      "epoch": 6.278285033743549,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001894513824383573,
      "loss": 4.8284,
      "mean_token_accuracy": 0.2065550521016121,
      "num_tokens": 138604442.0,
      "step": 63260
    },
    {
      "entropy": 5.691067314147949,
      "epoch": 6.278781262405716,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001894186129903707,
      "loss": 4.9888,
      "mean_token_accuracy": 0.20006561875343323,
      "num_tokens": 138615580.0,
      "step": 63265
    },
    {
      "entropy": 5.784974384307861,
      "epoch": 6.279277491067884,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00018938584566461803,
      "loss": 4.967,
      "mean_token_accuracy": 0.20294901877641677,
      "num_tokens": 138627075.0,
      "step": 63270
    },
    {
      "entropy": 5.739080476760864,
      "epoch": 6.279773719730052,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00018935308046191197,
      "loss": 4.9038,
      "mean_token_accuracy": 0.19952611178159713,
      "num_tokens": 138637191.0,
      "step": 63275
    },
    {
      "entropy": 5.678196668624878,
      "epoch": 6.2802699483922195,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018932031738306494,
      "loss": 4.9024,
      "mean_token_accuracy": 0.19680542051792144,
      "num_tokens": 138648000.0,
      "step": 63280
    },
    {
      "entropy": 5.690667247772216,
      "epoch": 6.280766177054387,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001892875564288894,
      "loss": 4.9194,
      "mean_token_accuracy": 0.19927702993154525,
      "num_tokens": 138660017.0,
      "step": 63285
    },
    {
      "entropy": 5.710535860061645,
      "epoch": 6.2812624057165545,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001892547976001977,
      "loss": 4.9237,
      "mean_token_accuracy": 0.1969743087887764,
      "num_tokens": 138669701.0,
      "step": 63290
    },
    {
      "entropy": 5.678999137878418,
      "epoch": 6.281758634378722,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018922204089780238,
      "loss": 4.8132,
      "mean_token_accuracy": 0.20090134739875792,
      "num_tokens": 138680833.0,
      "step": 63295
    },
    {
      "entropy": 5.734553861618042,
      "epoch": 6.282254863040889,
      "grad_norm": 1.5,
      "learning_rate": 0.00018918928632251554,
      "loss": 5.0106,
      "mean_token_accuracy": 0.19259785264730453,
      "num_tokens": 138691289.0,
      "step": 63300
    },
    {
      "entropy": 5.668897390365601,
      "epoch": 6.282751091703057,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018915653387514947,
      "loss": 4.8171,
      "mean_token_accuracy": 0.20953751355409622,
      "num_tokens": 138702273.0,
      "step": 63305
    },
    {
      "entropy": 5.680537939071655,
      "epoch": 6.283247320365224,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001891237835565165,
      "loss": 4.8643,
      "mean_token_accuracy": 0.20234772115945815,
      "num_tokens": 138714183.0,
      "step": 63310
    },
    {
      "entropy": 5.701854991912842,
      "epoch": 6.283743549027392,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001890910353674285,
      "loss": 4.8633,
      "mean_token_accuracy": 0.20504888147115707,
      "num_tokens": 138726060.0,
      "step": 63315
    },
    {
      "entropy": 5.767755365371704,
      "epoch": 6.284239777689559,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018905828930869777,
      "loss": 5.0166,
      "mean_token_accuracy": 0.19140901267528534,
      "num_tokens": 138738531.0,
      "step": 63320
    },
    {
      "entropy": 5.728489637374878,
      "epoch": 6.284736006351727,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018902554538113625,
      "loss": 4.9487,
      "mean_token_accuracy": 0.19405455589294435,
      "num_tokens": 138748391.0,
      "step": 63325
    },
    {
      "entropy": 5.680075454711914,
      "epoch": 6.285232235013894,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018899280358555603,
      "loss": 4.9061,
      "mean_token_accuracy": 0.2027137905359268,
      "num_tokens": 138758507.0,
      "step": 63330
    },
    {
      "entropy": 5.687555742263794,
      "epoch": 6.285728463676062,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018896006392276889,
      "loss": 4.9268,
      "mean_token_accuracy": 0.19340310841798783,
      "num_tokens": 138769489.0,
      "step": 63335
    },
    {
      "entropy": 5.7078100681304935,
      "epoch": 6.286224692338229,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018892732639358678,
      "loss": 4.8757,
      "mean_token_accuracy": 0.20087929517030717,
      "num_tokens": 138779742.0,
      "step": 63340
    },
    {
      "entropy": 5.686503314971924,
      "epoch": 6.286720921000397,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018889459099882155,
      "loss": 4.9229,
      "mean_token_accuracy": 0.20334941595792771,
      "num_tokens": 138790275.0,
      "step": 63345
    },
    {
      "entropy": 5.662028217315674,
      "epoch": 6.287217149662564,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018886185773928478,
      "loss": 4.9423,
      "mean_token_accuracy": 0.19629822820425033,
      "num_tokens": 138802292.0,
      "step": 63350
    },
    {
      "entropy": 5.70114369392395,
      "epoch": 6.287713378324732,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018882912661578844,
      "loss": 4.854,
      "mean_token_accuracy": 0.20420707762241364,
      "num_tokens": 138813943.0,
      "step": 63355
    },
    {
      "entropy": 5.7114574909210205,
      "epoch": 6.2882096069869,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018879639762914407,
      "loss": 4.9024,
      "mean_token_accuracy": 0.1943308487534523,
      "num_tokens": 138824293.0,
      "step": 63360
    },
    {
      "entropy": 5.744084358215332,
      "epoch": 6.2887058356490675,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018876367078016332,
      "loss": 5.0046,
      "mean_token_accuracy": 0.19751785546541215,
      "num_tokens": 138833908.0,
      "step": 63365
    },
    {
      "entropy": 5.755186653137207,
      "epoch": 6.289202064311235,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018873094606965764,
      "loss": 5.0293,
      "mean_token_accuracy": 0.19087895154953002,
      "num_tokens": 138845874.0,
      "step": 63370
    },
    {
      "entropy": 5.661325931549072,
      "epoch": 6.289698292973402,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018869822349843867,
      "loss": 4.8647,
      "mean_token_accuracy": 0.203884257376194,
      "num_tokens": 138856072.0,
      "step": 63375
    },
    {
      "entropy": 5.715287923812866,
      "epoch": 6.29019452163557,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018866550306731767,
      "loss": 4.8986,
      "mean_token_accuracy": 0.1996566191315651,
      "num_tokens": 138865938.0,
      "step": 63380
    },
    {
      "entropy": 5.670785522460937,
      "epoch": 6.290690750297737,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018863278477710627,
      "loss": 4.8608,
      "mean_token_accuracy": 0.20237826257944108,
      "num_tokens": 138876621.0,
      "step": 63385
    },
    {
      "entropy": 5.655376720428467,
      "epoch": 6.291186978959905,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018860006862861568,
      "loss": 4.8672,
      "mean_token_accuracy": 0.2013505980372429,
      "num_tokens": 138887250.0,
      "step": 63390
    },
    {
      "entropy": 5.68577356338501,
      "epoch": 6.291683207622072,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018856735462265722,
      "loss": 4.9133,
      "mean_token_accuracy": 0.19431473463773727,
      "num_tokens": 138897552.0,
      "step": 63395
    },
    {
      "entropy": 5.671465015411377,
      "epoch": 6.29217943628424,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018853464276004213,
      "loss": 4.9223,
      "mean_token_accuracy": 0.19511008858680726,
      "num_tokens": 138908320.0,
      "step": 63400
    },
    {
      "entropy": 5.698071908950806,
      "epoch": 6.292675664946407,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018850193304158154,
      "loss": 4.8643,
      "mean_token_accuracy": 0.2071971982717514,
      "num_tokens": 138919467.0,
      "step": 63405
    },
    {
      "entropy": 5.695884132385254,
      "epoch": 6.293171893608575,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018846922546808658,
      "loss": 4.8481,
      "mean_token_accuracy": 0.20828070640563964,
      "num_tokens": 138929312.0,
      "step": 63410
    },
    {
      "entropy": 5.7663041114807125,
      "epoch": 6.293668122270742,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018843652004036848,
      "loss": 5.0179,
      "mean_token_accuracy": 0.1960947260260582,
      "num_tokens": 138940236.0,
      "step": 63415
    },
    {
      "entropy": 5.704818058013916,
      "epoch": 6.29416435093291,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018840381675923806,
      "loss": 4.907,
      "mean_token_accuracy": 0.1976462721824646,
      "num_tokens": 138951109.0,
      "step": 63420
    },
    {
      "entropy": 5.675628757476806,
      "epoch": 6.294660579595077,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018837111562550651,
      "loss": 4.8695,
      "mean_token_accuracy": 0.1986882820725441,
      "num_tokens": 138961919.0,
      "step": 63425
    },
    {
      "entropy": 5.655388879776001,
      "epoch": 6.295156808257245,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018833841663998453,
      "loss": 4.885,
      "mean_token_accuracy": 0.20264869481325148,
      "num_tokens": 138972861.0,
      "step": 63430
    },
    {
      "entropy": 5.660034990310669,
      "epoch": 6.295653036919412,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001883057198034831,
      "loss": 4.8386,
      "mean_token_accuracy": 0.20210316479206086,
      "num_tokens": 138983947.0,
      "step": 63435
    },
    {
      "entropy": 5.656430578231811,
      "epoch": 6.29614926558158,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018827302511681299,
      "loss": 4.9327,
      "mean_token_accuracy": 0.19573203176259996,
      "num_tokens": 138995303.0,
      "step": 63440
    },
    {
      "entropy": 5.80503568649292,
      "epoch": 6.296645494243748,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000188240332580785,
      "loss": 4.9974,
      "mean_token_accuracy": 0.19671648293733596,
      "num_tokens": 139006684.0,
      "step": 63445
    },
    {
      "entropy": 5.7089379787445065,
      "epoch": 6.2971417229059155,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018820764219620978,
      "loss": 4.9271,
      "mean_token_accuracy": 0.19731345921754836,
      "num_tokens": 139017168.0,
      "step": 63450
    },
    {
      "entropy": 5.67939305305481,
      "epoch": 6.297637951568083,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018817495396389805,
      "loss": 4.9447,
      "mean_token_accuracy": 0.19785936623811723,
      "num_tokens": 139028397.0,
      "step": 63455
    },
    {
      "entropy": 5.678179693222046,
      "epoch": 6.29813418023025,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018814226788466045,
      "loss": 4.8122,
      "mean_token_accuracy": 0.21128551214933394,
      "num_tokens": 139038903.0,
      "step": 63460
    },
    {
      "entropy": 5.739499044418335,
      "epoch": 6.298630408892418,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018810958395930734,
      "loss": 4.8629,
      "mean_token_accuracy": 0.20612484812736512,
      "num_tokens": 139050017.0,
      "step": 63465
    },
    {
      "entropy": 5.742258977890015,
      "epoch": 6.299126637554585,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018807690218864925,
      "loss": 4.9838,
      "mean_token_accuracy": 0.20239002704620362,
      "num_tokens": 139060818.0,
      "step": 63470
    },
    {
      "entropy": 5.6122761249542235,
      "epoch": 6.299622866216753,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001880442225734968,
      "loss": 4.8087,
      "mean_token_accuracy": 0.20936422795057297,
      "num_tokens": 139072682.0,
      "step": 63475
    },
    {
      "entropy": 5.787866926193237,
      "epoch": 6.30011909487892,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00018801154511466022,
      "loss": 4.9782,
      "mean_token_accuracy": 0.19751675575971603,
      "num_tokens": 139082788.0,
      "step": 63480
    },
    {
      "entropy": 5.7177238941192625,
      "epoch": 6.300615323541088,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018797886981294988,
      "loss": 4.8831,
      "mean_token_accuracy": 0.2030871406197548,
      "num_tokens": 139093847.0,
      "step": 63485
    },
    {
      "entropy": 5.6939826011657715,
      "epoch": 6.301111552203255,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018794619666917607,
      "loss": 4.9052,
      "mean_token_accuracy": 0.2068771779537201,
      "num_tokens": 139104826.0,
      "step": 63490
    },
    {
      "entropy": 5.663870143890381,
      "epoch": 6.301607780865423,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018791352568414894,
      "loss": 4.8865,
      "mean_token_accuracy": 0.198517045378685,
      "num_tokens": 139116124.0,
      "step": 63495
    },
    {
      "entropy": 5.667790460586548,
      "epoch": 6.30210400952759,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018788085685867873,
      "loss": 4.8807,
      "mean_token_accuracy": 0.2032017305493355,
      "num_tokens": 139127184.0,
      "step": 63500
    },
    {
      "entropy": 5.755379104614258,
      "epoch": 6.302600238189758,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018784819019357558,
      "loss": 4.9342,
      "mean_token_accuracy": 0.1969210609793663,
      "num_tokens": 139138625.0,
      "step": 63505
    },
    {
      "entropy": 5.706150865554809,
      "epoch": 6.303096466851925,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018781552568964956,
      "loss": 4.994,
      "mean_token_accuracy": 0.20019921511411667,
      "num_tokens": 139150219.0,
      "step": 63510
    },
    {
      "entropy": 5.662858963012695,
      "epoch": 6.303592695514093,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018778286334771054,
      "loss": 4.9355,
      "mean_token_accuracy": 0.19391525089740752,
      "num_tokens": 139161103.0,
      "step": 63515
    },
    {
      "entropy": 5.73513822555542,
      "epoch": 6.30408892417626,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018775020316856868,
      "loss": 4.9401,
      "mean_token_accuracy": 0.19524553120136262,
      "num_tokens": 139171252.0,
      "step": 63520
    },
    {
      "entropy": 5.7015996932983395,
      "epoch": 6.304585152838428,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001877175451530337,
      "loss": 4.8898,
      "mean_token_accuracy": 0.2010420873761177,
      "num_tokens": 139183406.0,
      "step": 63525
    },
    {
      "entropy": 5.727590131759643,
      "epoch": 6.305081381500596,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018768488930191553,
      "loss": 4.9482,
      "mean_token_accuracy": 0.1899822548031807,
      "num_tokens": 139194269.0,
      "step": 63530
    },
    {
      "entropy": 5.744100093841553,
      "epoch": 6.305577610162763,
      "grad_norm": 1.375,
      "learning_rate": 0.00018765223561602396,
      "loss": 4.9691,
      "mean_token_accuracy": 0.1926095202565193,
      "num_tokens": 139205310.0,
      "step": 63535
    },
    {
      "entropy": 5.746565198898315,
      "epoch": 6.306073838824931,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018761958409616874,
      "loss": 4.9551,
      "mean_token_accuracy": 0.19612492769956588,
      "num_tokens": 139216205.0,
      "step": 63540
    },
    {
      "entropy": 5.677095794677735,
      "epoch": 6.306570067487098,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018758693474315953,
      "loss": 4.8621,
      "mean_token_accuracy": 0.20598286241292954,
      "num_tokens": 139226194.0,
      "step": 63545
    },
    {
      "entropy": 5.692973852157593,
      "epoch": 6.307066296149266,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018755428755780606,
      "loss": 4.9049,
      "mean_token_accuracy": 0.2052670881152153,
      "num_tokens": 139237513.0,
      "step": 63550
    },
    {
      "entropy": 5.740522956848144,
      "epoch": 6.307562524811433,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001875216425409177,
      "loss": 4.9699,
      "mean_token_accuracy": 0.19907222539186478,
      "num_tokens": 139249403.0,
      "step": 63555
    },
    {
      "entropy": 5.706694173812866,
      "epoch": 6.308058753473601,
      "grad_norm": 1.375,
      "learning_rate": 0.00018748899969330414,
      "loss": 4.9041,
      "mean_token_accuracy": 0.20208242088556289,
      "num_tokens": 139260043.0,
      "step": 63560
    },
    {
      "entropy": 5.724907684326172,
      "epoch": 6.308554982135768,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018745635901577483,
      "loss": 4.9432,
      "mean_token_accuracy": 0.19800515621900558,
      "num_tokens": 139269566.0,
      "step": 63565
    },
    {
      "entropy": 5.6797261238098145,
      "epoch": 6.309051210797936,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018742372050913926,
      "loss": 4.8961,
      "mean_token_accuracy": 0.19794438630342484,
      "num_tokens": 139280260.0,
      "step": 63570
    },
    {
      "entropy": 5.760859489440918,
      "epoch": 6.309547439460103,
      "grad_norm": 1.375,
      "learning_rate": 0.00018739108417420664,
      "loss": 5.0061,
      "mean_token_accuracy": 0.19309030920267106,
      "num_tokens": 139291139.0,
      "step": 63575
    },
    {
      "entropy": 5.694052648544312,
      "epoch": 6.310043668122271,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018735845001178637,
      "loss": 4.8436,
      "mean_token_accuracy": 0.2041521579027176,
      "num_tokens": 139301368.0,
      "step": 63580
    },
    {
      "entropy": 5.731304025650024,
      "epoch": 6.310539896784438,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001873258180226876,
      "loss": 4.9214,
      "mean_token_accuracy": 0.198323492705822,
      "num_tokens": 139312406.0,
      "step": 63585
    },
    {
      "entropy": 5.718233919143676,
      "epoch": 6.311036125446606,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018729318820771972,
      "loss": 4.9971,
      "mean_token_accuracy": 0.19264405965805054,
      "num_tokens": 139323343.0,
      "step": 63590
    },
    {
      "entropy": 5.683839750289917,
      "epoch": 6.311532354108773,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001872605605676918,
      "loss": 4.9025,
      "mean_token_accuracy": 0.20155311971902848,
      "num_tokens": 139332775.0,
      "step": 63595
    },
    {
      "entropy": 5.752187490463257,
      "epoch": 6.312028582770941,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018722793510341285,
      "loss": 4.9914,
      "mean_token_accuracy": 0.19664286375045775,
      "num_tokens": 139343437.0,
      "step": 63600
    },
    {
      "entropy": 5.683420038223266,
      "epoch": 6.312524811433108,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018719531181569205,
      "loss": 4.824,
      "mean_token_accuracy": 0.21300261169672013,
      "num_tokens": 139354583.0,
      "step": 63605
    },
    {
      "entropy": 5.674527359008789,
      "epoch": 6.3130210400952755,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018716269070533827,
      "loss": 4.8784,
      "mean_token_accuracy": 0.20133473724126816,
      "num_tokens": 139365464.0,
      "step": 63610
    },
    {
      "entropy": 5.708689498901367,
      "epoch": 6.313517268757444,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001871300717731604,
      "loss": 4.9381,
      "mean_token_accuracy": 0.206102354824543,
      "num_tokens": 139375439.0,
      "step": 63615
    },
    {
      "entropy": 5.690601968765259,
      "epoch": 6.3140134974196105,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001870974550199675,
      "loss": 4.985,
      "mean_token_accuracy": 0.19152650684118272,
      "num_tokens": 139386295.0,
      "step": 63620
    },
    {
      "entropy": 5.699298763275147,
      "epoch": 6.314509726081779,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018706484044656835,
      "loss": 4.9529,
      "mean_token_accuracy": 0.1999931290745735,
      "num_tokens": 139397161.0,
      "step": 63625
    },
    {
      "entropy": 5.674933433532715,
      "epoch": 6.315005954743946,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018703222805377157,
      "loss": 4.8643,
      "mean_token_accuracy": 0.2065305307507515,
      "num_tokens": 139407964.0,
      "step": 63630
    },
    {
      "entropy": 5.765839576721191,
      "epoch": 6.315502183406114,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018699961784238606,
      "loss": 4.9997,
      "mean_token_accuracy": 0.20073607414960862,
      "num_tokens": 139419963.0,
      "step": 63635
    },
    {
      "entropy": 5.712760257720947,
      "epoch": 6.315998412068281,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001869670098132204,
      "loss": 4.8948,
      "mean_token_accuracy": 0.20156621783971787,
      "num_tokens": 139430289.0,
      "step": 63640
    },
    {
      "entropy": 5.693074941635132,
      "epoch": 6.316494640730449,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001869344039670831,
      "loss": 4.955,
      "mean_token_accuracy": 0.19460188895463942,
      "num_tokens": 139442529.0,
      "step": 63645
    },
    {
      "entropy": 5.6980202198028564,
      "epoch": 6.316990869392616,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018690180030478287,
      "loss": 4.8887,
      "mean_token_accuracy": 0.20049253553152085,
      "num_tokens": 139452991.0,
      "step": 63650
    },
    {
      "entropy": 5.698981189727784,
      "epoch": 6.317487098054784,
      "grad_norm": 1.1875,
      "learning_rate": 0.00018686919882712823,
      "loss": 4.9091,
      "mean_token_accuracy": 0.20208034068346023,
      "num_tokens": 139466796.0,
      "step": 63655
    },
    {
      "entropy": 5.646934556961059,
      "epoch": 6.317983326716951,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001868365995349275,
      "loss": 4.871,
      "mean_token_accuracy": 0.20363843739032744,
      "num_tokens": 139477206.0,
      "step": 63660
    },
    {
      "entropy": 5.7029565334320065,
      "epoch": 6.318479555379119,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018680400242898916,
      "loss": 4.8892,
      "mean_token_accuracy": 0.20407021790742874,
      "num_tokens": 139488668.0,
      "step": 63665
    },
    {
      "entropy": 5.754603147506714,
      "epoch": 6.318975784041286,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001867714075101215,
      "loss": 4.9366,
      "mean_token_accuracy": 0.20101200342178344,
      "num_tokens": 139499514.0,
      "step": 63670
    },
    {
      "entropy": 5.7017356872558596,
      "epoch": 6.319472012703454,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018673881477913274,
      "loss": 4.9014,
      "mean_token_accuracy": 0.202612966299057,
      "num_tokens": 139509944.0,
      "step": 63675
    },
    {
      "entropy": 5.743167400360107,
      "epoch": 6.319968241365621,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001867062242368312,
      "loss": 5.0255,
      "mean_token_accuracy": 0.19529372751712798,
      "num_tokens": 139522361.0,
      "step": 63680
    },
    {
      "entropy": 5.737201976776123,
      "epoch": 6.3204644700277886,
      "grad_norm": 1.5,
      "learning_rate": 0.00018667363588402514,
      "loss": 4.9315,
      "mean_token_accuracy": 0.19821523278951644,
      "num_tokens": 139531700.0,
      "step": 63685
    },
    {
      "entropy": 5.801406145095825,
      "epoch": 6.320960698689956,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018664104972152256,
      "loss": 4.9252,
      "mean_token_accuracy": 0.20291664600372314,
      "num_tokens": 139542275.0,
      "step": 63690
    },
    {
      "entropy": 5.691563320159912,
      "epoch": 6.3214569273521235,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018660846575013152,
      "loss": 4.8966,
      "mean_token_accuracy": 0.2002711459994316,
      "num_tokens": 139553718.0,
      "step": 63695
    },
    {
      "entropy": 5.654160356521606,
      "epoch": 6.321953156014291,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018657588397066007,
      "loss": 4.8725,
      "mean_token_accuracy": 0.2034850537776947,
      "num_tokens": 139564660.0,
      "step": 63700
    },
    {
      "entropy": 5.741305541992188,
      "epoch": 6.3224493846764585,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018654330438391614,
      "loss": 4.9529,
      "mean_token_accuracy": 0.19632067233324052,
      "num_tokens": 139575595.0,
      "step": 63705
    },
    {
      "entropy": 5.6681328296661375,
      "epoch": 6.322945613338627,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018651072699070765,
      "loss": 4.9046,
      "mean_token_accuracy": 0.19842773973941802,
      "num_tokens": 139587085.0,
      "step": 63710
    },
    {
      "entropy": 5.646006345748901,
      "epoch": 6.323441842000794,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018647815179184245,
      "loss": 4.8302,
      "mean_token_accuracy": 0.21023142635822295,
      "num_tokens": 139596920.0,
      "step": 63715
    },
    {
      "entropy": 5.74117522239685,
      "epoch": 6.323938070662962,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018644557878812846,
      "loss": 4.9904,
      "mean_token_accuracy": 0.19452545046806335,
      "num_tokens": 139609168.0,
      "step": 63720
    },
    {
      "entropy": 5.683355474472046,
      "epoch": 6.324434299325129,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018641300798037314,
      "loss": 4.8672,
      "mean_token_accuracy": 0.20802152454853057,
      "num_tokens": 139620275.0,
      "step": 63725
    },
    {
      "entropy": 5.710424757003784,
      "epoch": 6.324930527987297,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018638043936938437,
      "loss": 4.9309,
      "mean_token_accuracy": 0.19510431438684464,
      "num_tokens": 139631381.0,
      "step": 63730
    },
    {
      "entropy": 5.671069002151489,
      "epoch": 6.325426756649464,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018634787295596978,
      "loss": 4.9459,
      "mean_token_accuracy": 0.19921431988477706,
      "num_tokens": 139642960.0,
      "step": 63735
    },
    {
      "entropy": 5.669099998474121,
      "epoch": 6.325922985311632,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018631530874093694,
      "loss": 4.9159,
      "mean_token_accuracy": 0.19760051667690276,
      "num_tokens": 139653512.0,
      "step": 63740
    },
    {
      "entropy": 5.684909152984619,
      "epoch": 6.326419213973799,
      "grad_norm": 1.375,
      "learning_rate": 0.00018628274672509332,
      "loss": 4.8635,
      "mean_token_accuracy": 0.2051510989665985,
      "num_tokens": 139664163.0,
      "step": 63745
    },
    {
      "entropy": 5.715574836730957,
      "epoch": 6.326915442635967,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018625018690924638,
      "loss": 4.984,
      "mean_token_accuracy": 0.1968182936310768,
      "num_tokens": 139675096.0,
      "step": 63750
    },
    {
      "entropy": 5.682506990432739,
      "epoch": 6.327411671298134,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001862176292942036,
      "loss": 4.8465,
      "mean_token_accuracy": 0.200956030189991,
      "num_tokens": 139686799.0,
      "step": 63755
    },
    {
      "entropy": 5.713012170791626,
      "epoch": 6.327907899960302,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018618507388077228,
      "loss": 4.9164,
      "mean_token_accuracy": 0.2008747011423111,
      "num_tokens": 139697262.0,
      "step": 63760
    },
    {
      "entropy": 5.678997373580932,
      "epoch": 6.328404128622469,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018615252066975975,
      "loss": 4.9457,
      "mean_token_accuracy": 0.20331032127141951,
      "num_tokens": 139708964.0,
      "step": 63765
    },
    {
      "entropy": 5.685187911987304,
      "epoch": 6.3289003572846365,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018611996966197325,
      "loss": 4.8796,
      "mean_token_accuracy": 0.20372643768787385,
      "num_tokens": 139720091.0,
      "step": 63770
    },
    {
      "entropy": 5.661456298828125,
      "epoch": 6.329396585946804,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018608742085822,
      "loss": 4.8977,
      "mean_token_accuracy": 0.20737744867801666,
      "num_tokens": 139732149.0,
      "step": 63775
    },
    {
      "entropy": 5.651502895355224,
      "epoch": 6.3298928146089715,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018605487425930712,
      "loss": 4.8433,
      "mean_token_accuracy": 0.20527878105640412,
      "num_tokens": 139743355.0,
      "step": 63780
    },
    {
      "entropy": 5.6742137432098385,
      "epoch": 6.330389043271139,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018602232986604178,
      "loss": 4.8994,
      "mean_token_accuracy": 0.20095515102148057,
      "num_tokens": 139754867.0,
      "step": 63785
    },
    {
      "entropy": 5.69367151260376,
      "epoch": 6.330885271933306,
      "grad_norm": 1.546875,
      "learning_rate": 0.00018598978767923074,
      "loss": 4.8346,
      "mean_token_accuracy": 0.21239873319864272,
      "num_tokens": 139766218.0,
      "step": 63790
    },
    {
      "entropy": 5.716232538223267,
      "epoch": 6.331381500595475,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018595724769968126,
      "loss": 4.8865,
      "mean_token_accuracy": 0.20156137496232987,
      "num_tokens": 139776814.0,
      "step": 63795
    },
    {
      "entropy": 5.623367595672607,
      "epoch": 6.331877729257642,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018592470992820026,
      "loss": 4.7567,
      "mean_token_accuracy": 0.21372735053300856,
      "num_tokens": 139788015.0,
      "step": 63800
    },
    {
      "entropy": 5.6641092777252195,
      "epoch": 6.33237395791981,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018589217436559447,
      "loss": 4.8574,
      "mean_token_accuracy": 0.20342317521572112,
      "num_tokens": 139798118.0,
      "step": 63805
    },
    {
      "entropy": 5.686186981201172,
      "epoch": 6.332870186581977,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018585964101267067,
      "loss": 4.8834,
      "mean_token_accuracy": 0.20609972923994063,
      "num_tokens": 139809561.0,
      "step": 63810
    },
    {
      "entropy": 5.689540147781372,
      "epoch": 6.333366415244145,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018582710987023577,
      "loss": 4.8916,
      "mean_token_accuracy": 0.20300279706716537,
      "num_tokens": 139820387.0,
      "step": 63815
    },
    {
      "entropy": 5.581750822067261,
      "epoch": 6.333862643906312,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018579458093909628,
      "loss": 4.8152,
      "mean_token_accuracy": 0.20711336135864258,
      "num_tokens": 139832037.0,
      "step": 63820
    },
    {
      "entropy": 5.719182014465332,
      "epoch": 6.33435887256848,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001857620542200591,
      "loss": 4.982,
      "mean_token_accuracy": 0.1943846881389618,
      "num_tokens": 139844183.0,
      "step": 63825
    },
    {
      "entropy": 5.658280515670777,
      "epoch": 6.334855101230647,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018572952971393058,
      "loss": 4.8863,
      "mean_token_accuracy": 0.2055346980690956,
      "num_tokens": 139855737.0,
      "step": 63830
    },
    {
      "entropy": 5.672159719467163,
      "epoch": 6.335351329892815,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018569700742151747,
      "loss": 4.844,
      "mean_token_accuracy": 0.20869184881448746,
      "num_tokens": 139866478.0,
      "step": 63835
    },
    {
      "entropy": 5.707066774368286,
      "epoch": 6.335847558554982,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018566448734362606,
      "loss": 4.8435,
      "mean_token_accuracy": 0.20260179489851,
      "num_tokens": 139875775.0,
      "step": 63840
    },
    {
      "entropy": 5.673334455490112,
      "epoch": 6.3363437872171495,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018563196948106293,
      "loss": 4.912,
      "mean_token_accuracy": 0.20212336182594298,
      "num_tokens": 139887160.0,
      "step": 63845
    },
    {
      "entropy": 5.662720012664795,
      "epoch": 6.336840015879317,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018559945383463428,
      "loss": 4.8897,
      "mean_token_accuracy": 0.19684447348117828,
      "num_tokens": 139899281.0,
      "step": 63850
    },
    {
      "entropy": 5.700632858276367,
      "epoch": 6.3373362445414845,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001855669404051466,
      "loss": 4.8958,
      "mean_token_accuracy": 0.20275094360113144,
      "num_tokens": 139911031.0,
      "step": 63855
    },
    {
      "entropy": 5.727288246154785,
      "epoch": 6.337832473203652,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018553442919340608,
      "loss": 4.8901,
      "mean_token_accuracy": 0.20248036086559296,
      "num_tokens": 139920505.0,
      "step": 63860
    },
    {
      "entropy": 5.7653850555419925,
      "epoch": 6.338328701865819,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018550192020021895,
      "loss": 4.9578,
      "mean_token_accuracy": 0.1964816689491272,
      "num_tokens": 139930306.0,
      "step": 63865
    },
    {
      "entropy": 5.772605752944946,
      "epoch": 6.338824930527987,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001854694134263914,
      "loss": 5.0258,
      "mean_token_accuracy": 0.19304924011230468,
      "num_tokens": 139940678.0,
      "step": 63870
    },
    {
      "entropy": 5.755818653106689,
      "epoch": 6.339321159190154,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00018543690887272936,
      "loss": 4.9898,
      "mean_token_accuracy": 0.199086332321167,
      "num_tokens": 139953196.0,
      "step": 63875
    },
    {
      "entropy": 5.769414901733398,
      "epoch": 6.339817387852323,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018540440654003898,
      "loss": 4.9868,
      "mean_token_accuracy": 0.1969156250357628,
      "num_tokens": 139963998.0,
      "step": 63880
    },
    {
      "entropy": 5.6830915927886965,
      "epoch": 6.34031361651449,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018537190642912632,
      "loss": 4.8867,
      "mean_token_accuracy": 0.19897379726171494,
      "num_tokens": 139974229.0,
      "step": 63885
    },
    {
      "entropy": 5.683215856552124,
      "epoch": 6.340809845176658,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018533940854079727,
      "loss": 4.8643,
      "mean_token_accuracy": 0.20725573748350143,
      "num_tokens": 139985246.0,
      "step": 63890
    },
    {
      "entropy": 5.680768251419067,
      "epoch": 6.341306073838825,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018530691287585771,
      "loss": 4.9306,
      "mean_token_accuracy": 0.20319516360759735,
      "num_tokens": 139996625.0,
      "step": 63895
    },
    {
      "entropy": 5.712121057510376,
      "epoch": 6.341802302500993,
      "grad_norm": 1.375,
      "learning_rate": 0.0001852744194351134,
      "loss": 4.8787,
      "mean_token_accuracy": 0.20164939016103745,
      "num_tokens": 140007659.0,
      "step": 63900
    },
    {
      "entropy": 5.760205221176148,
      "epoch": 6.34229853116316,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00018524192821937012,
      "loss": 4.8963,
      "mean_token_accuracy": 0.20365228056907653,
      "num_tokens": 140017991.0,
      "step": 63905
    },
    {
      "entropy": 5.706704092025757,
      "epoch": 6.342794759825328,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018520943922943357,
      "loss": 4.8825,
      "mean_token_accuracy": 0.20650040954351426,
      "num_tokens": 140028805.0,
      "step": 63910
    },
    {
      "entropy": 5.739732980728149,
      "epoch": 6.343290988487495,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018517695246610949,
      "loss": 4.8678,
      "mean_token_accuracy": 0.1999776467680931,
      "num_tokens": 140038919.0,
      "step": 63915
    },
    {
      "entropy": 5.7255699157714846,
      "epoch": 6.3437872171496625,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001851444679302035,
      "loss": 4.9612,
      "mean_token_accuracy": 0.1968400478363037,
      "num_tokens": 140049393.0,
      "step": 63920
    },
    {
      "entropy": 5.705273628234863,
      "epoch": 6.34428344581183,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018511198562252103,
      "loss": 4.9383,
      "mean_token_accuracy": 0.20058574974536897,
      "num_tokens": 140058612.0,
      "step": 63925
    },
    {
      "entropy": 5.766872119903565,
      "epoch": 6.3447796744739975,
      "grad_norm": 1.375,
      "learning_rate": 0.00018507950554386767,
      "loss": 5.0444,
      "mean_token_accuracy": 0.1939220428466797,
      "num_tokens": 140071050.0,
      "step": 63930
    },
    {
      "entropy": 5.79569444656372,
      "epoch": 6.345275903136165,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001850470276950488,
      "loss": 5.0303,
      "mean_token_accuracy": 0.19319384098052977,
      "num_tokens": 140083038.0,
      "step": 63935
    },
    {
      "entropy": 5.636788940429687,
      "epoch": 6.345772131798332,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018501455207686968,
      "loss": 4.7672,
      "mean_token_accuracy": 0.21683691442012787,
      "num_tokens": 140093314.0,
      "step": 63940
    },
    {
      "entropy": 5.69094877243042,
      "epoch": 6.3462683604605,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018498207869013588,
      "loss": 4.8837,
      "mean_token_accuracy": 0.20578389316797258,
      "num_tokens": 140103885.0,
      "step": 63945
    },
    {
      "entropy": 5.7391890525817875,
      "epoch": 6.346764589122667,
      "grad_norm": 1.5,
      "learning_rate": 0.0001849496075356526,
      "loss": 4.9665,
      "mean_token_accuracy": 0.19914692640304565,
      "num_tokens": 140114670.0,
      "step": 63950
    },
    {
      "entropy": 5.744200706481934,
      "epoch": 6.347260817784835,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018491713861422493,
      "loss": 4.9318,
      "mean_token_accuracy": 0.20158901512622834,
      "num_tokens": 140125784.0,
      "step": 63955
    },
    {
      "entropy": 5.671752548217773,
      "epoch": 6.347757046447002,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018488467192665815,
      "loss": 4.8192,
      "mean_token_accuracy": 0.2039359599351883,
      "num_tokens": 140136120.0,
      "step": 63960
    },
    {
      "entropy": 5.735361194610595,
      "epoch": 6.348253275109171,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018485220747375736,
      "loss": 4.9424,
      "mean_token_accuracy": 0.1953176125884056,
      "num_tokens": 140146205.0,
      "step": 63965
    },
    {
      "entropy": 5.675499010086059,
      "epoch": 6.348749503771338,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018481974525632746,
      "loss": 4.8748,
      "mean_token_accuracy": 0.19935758113861085,
      "num_tokens": 140156222.0,
      "step": 63970
    },
    {
      "entropy": 5.6985095024108885,
      "epoch": 6.349245732433506,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018478728527517358,
      "loss": 4.9493,
      "mean_token_accuracy": 0.19893973767757417,
      "num_tokens": 140167232.0,
      "step": 63975
    },
    {
      "entropy": 5.701881742477417,
      "epoch": 6.349741961095673,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018475482753110075,
      "loss": 4.9345,
      "mean_token_accuracy": 0.19583040624856948,
      "num_tokens": 140177121.0,
      "step": 63980
    },
    {
      "entropy": 5.662257862091065,
      "epoch": 6.350238189757841,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018472237202491365,
      "loss": 4.8872,
      "mean_token_accuracy": 0.1957710787653923,
      "num_tokens": 140188770.0,
      "step": 63985
    },
    {
      "entropy": 5.771463775634766,
      "epoch": 6.350734418420008,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018468991875741718,
      "loss": 4.9005,
      "mean_token_accuracy": 0.195416821539402,
      "num_tokens": 140199852.0,
      "step": 63990
    },
    {
      "entropy": 5.710716915130615,
      "epoch": 6.3512306470821756,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018465746772941612,
      "loss": 4.8993,
      "mean_token_accuracy": 0.2074705421924591,
      "num_tokens": 140211640.0,
      "step": 63995
    },
    {
      "entropy": 5.676698064804077,
      "epoch": 6.351726875744343,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00018462501894171525,
      "loss": 4.9464,
      "mean_token_accuracy": 0.19741385281085969,
      "num_tokens": 140222428.0,
      "step": 64000
    },
    {
      "entropy": 5.705914258956909,
      "epoch": 6.3522231044065105,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018459257239511917,
      "loss": 4.921,
      "mean_token_accuracy": 0.19600037187337876,
      "num_tokens": 140233813.0,
      "step": 64005
    },
    {
      "entropy": 5.713460254669189,
      "epoch": 6.352719333068678,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018456012809043242,
      "loss": 4.9099,
      "mean_token_accuracy": 0.20016555190086366,
      "num_tokens": 140244365.0,
      "step": 64010
    },
    {
      "entropy": 5.693756866455078,
      "epoch": 6.3532155617308455,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018452768602845966,
      "loss": 4.8595,
      "mean_token_accuracy": 0.21315380483865737,
      "num_tokens": 140256334.0,
      "step": 64015
    },
    {
      "entropy": 5.699941730499267,
      "epoch": 6.353711790393013,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001844952462100053,
      "loss": 4.9628,
      "mean_token_accuracy": 0.1948478564620018,
      "num_tokens": 140266497.0,
      "step": 64020
    },
    {
      "entropy": 5.789487552642822,
      "epoch": 6.35420801905518,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001844628086358738,
      "loss": 5.0164,
      "mean_token_accuracy": 0.19286593645811081,
      "num_tokens": 140278288.0,
      "step": 64025
    },
    {
      "entropy": 5.723340320587158,
      "epoch": 6.354704247717348,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018443037330686962,
      "loss": 4.9307,
      "mean_token_accuracy": 0.20340829640626906,
      "num_tokens": 140289799.0,
      "step": 64030
    },
    {
      "entropy": 5.650664043426514,
      "epoch": 6.355200476379515,
      "grad_norm": 1.375,
      "learning_rate": 0.00018439794022379703,
      "loss": 4.8148,
      "mean_token_accuracy": 0.20886829644441604,
      "num_tokens": 140300111.0,
      "step": 64035
    },
    {
      "entropy": 5.712608814239502,
      "epoch": 6.355696705041683,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018436550938746022,
      "loss": 4.9246,
      "mean_token_accuracy": 0.20182544589042664,
      "num_tokens": 140311409.0,
      "step": 64040
    },
    {
      "entropy": 5.6855062484741214,
      "epoch": 6.35619293370385,
      "grad_norm": 1.515625,
      "learning_rate": 0.00018433308079866356,
      "loss": 4.9524,
      "mean_token_accuracy": 0.19744889438152313,
      "num_tokens": 140321906.0,
      "step": 64045
    },
    {
      "entropy": 5.668853092193603,
      "epoch": 6.356689162366019,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001843006544582111,
      "loss": 4.9951,
      "mean_token_accuracy": 0.1972769096493721,
      "num_tokens": 140333578.0,
      "step": 64050
    },
    {
      "entropy": 5.748107004165649,
      "epoch": 6.357185391028186,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018426823036690688,
      "loss": 4.9816,
      "mean_token_accuracy": 0.19096195101737976,
      "num_tokens": 140345660.0,
      "step": 64055
    },
    {
      "entropy": 5.74438099861145,
      "epoch": 6.357681619690354,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018423580852555506,
      "loss": 4.8968,
      "mean_token_accuracy": 0.1982138678431511,
      "num_tokens": 140356624.0,
      "step": 64060
    },
    {
      "entropy": 5.709512424468994,
      "epoch": 6.358177848352521,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001842033889349597,
      "loss": 4.8966,
      "mean_token_accuracy": 0.19278143346309662,
      "num_tokens": 140367163.0,
      "step": 64065
    },
    {
      "entropy": 5.66106104850769,
      "epoch": 6.358674077014689,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001841709715959246,
      "loss": 4.8959,
      "mean_token_accuracy": 0.19781821966171265,
      "num_tokens": 140378179.0,
      "step": 64070
    },
    {
      "entropy": 5.688974237442016,
      "epoch": 6.359170305676856,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001841385565092537,
      "loss": 4.9456,
      "mean_token_accuracy": 0.1947566494345665,
      "num_tokens": 140389638.0,
      "step": 64075
    },
    {
      "entropy": 5.631637001037598,
      "epoch": 6.3596665343390235,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018410614367575078,
      "loss": 4.8397,
      "mean_token_accuracy": 0.20961589068174363,
      "num_tokens": 140400025.0,
      "step": 64080
    },
    {
      "entropy": 5.761458063125611,
      "epoch": 6.360162763001191,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00018407373309621962,
      "loss": 4.9836,
      "mean_token_accuracy": 0.1980668991804123,
      "num_tokens": 140411105.0,
      "step": 64085
    },
    {
      "entropy": 5.630042219161988,
      "epoch": 6.3606589916633585,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018404132477146395,
      "loss": 4.8217,
      "mean_token_accuracy": 0.2065185785293579,
      "num_tokens": 140422425.0,
      "step": 64090
    },
    {
      "entropy": 5.748852729797363,
      "epoch": 6.361155220325526,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018400891870228752,
      "loss": 4.9396,
      "mean_token_accuracy": 0.20030049234628677,
      "num_tokens": 140433319.0,
      "step": 64095
    },
    {
      "entropy": 5.665716600418091,
      "epoch": 6.361651448987693,
      "grad_norm": 1.5,
      "learning_rate": 0.0001839765148894938,
      "loss": 4.887,
      "mean_token_accuracy": 0.20294148921966554,
      "num_tokens": 140443147.0,
      "step": 64100
    },
    {
      "entropy": 5.681781387329101,
      "epoch": 6.362147677649861,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018394411333388634,
      "loss": 4.8706,
      "mean_token_accuracy": 0.20473772436380386,
      "num_tokens": 140454006.0,
      "step": 64105
    },
    {
      "entropy": 5.675600337982178,
      "epoch": 6.362643906312028,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001839117140362687,
      "loss": 4.8531,
      "mean_token_accuracy": 0.21166103184223176,
      "num_tokens": 140465150.0,
      "step": 64110
    },
    {
      "entropy": 5.686463356018066,
      "epoch": 6.363140134974196,
      "grad_norm": 1.375,
      "learning_rate": 0.0001838793169974442,
      "loss": 4.9267,
      "mean_token_accuracy": 0.20267184674739838,
      "num_tokens": 140476476.0,
      "step": 64115
    },
    {
      "entropy": 5.705564498901367,
      "epoch": 6.363636363636363,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018384692221821636,
      "loss": 4.941,
      "mean_token_accuracy": 0.19652683287858963,
      "num_tokens": 140487755.0,
      "step": 64120
    },
    {
      "entropy": 5.707587385177613,
      "epoch": 6.364132592298531,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018381452969938833,
      "loss": 4.9118,
      "mean_token_accuracy": 0.19360441267490386,
      "num_tokens": 140499739.0,
      "step": 64125
    },
    {
      "entropy": 5.68523268699646,
      "epoch": 6.364628820960698,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018378213944176365,
      "loss": 4.8843,
      "mean_token_accuracy": 0.2050814300775528,
      "num_tokens": 140511080.0,
      "step": 64130
    },
    {
      "entropy": 5.69313793182373,
      "epoch": 6.365125049622867,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018374975144614525,
      "loss": 4.8562,
      "mean_token_accuracy": 0.19958158135414122,
      "num_tokens": 140521660.0,
      "step": 64135
    },
    {
      "entropy": 5.749138593673706,
      "epoch": 6.365621278285034,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018371736571333637,
      "loss": 4.9351,
      "mean_token_accuracy": 0.20079925507307053,
      "num_tokens": 140532725.0,
      "step": 64140
    },
    {
      "entropy": 5.710740852355957,
      "epoch": 6.366117506947202,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018368498224414015,
      "loss": 4.915,
      "mean_token_accuracy": 0.20626451224088668,
      "num_tokens": 140543055.0,
      "step": 64145
    },
    {
      "entropy": 5.706451845169068,
      "epoch": 6.366613735609369,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001836526010393596,
      "loss": 4.9123,
      "mean_token_accuracy": 0.1986142799258232,
      "num_tokens": 140554862.0,
      "step": 64150
    },
    {
      "entropy": 5.704891920089722,
      "epoch": 6.3671099642715365,
      "grad_norm": 1.5,
      "learning_rate": 0.00018362022209979768,
      "loss": 4.9562,
      "mean_token_accuracy": 0.1947559967637062,
      "num_tokens": 140565347.0,
      "step": 64155
    },
    {
      "entropy": 5.8250635147094725,
      "epoch": 6.367606192933704,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001835878454262574,
      "loss": 5.0769,
      "mean_token_accuracy": 0.19521547555923463,
      "num_tokens": 140575469.0,
      "step": 64160
    },
    {
      "entropy": 5.68509521484375,
      "epoch": 6.3681024215958715,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001835554710195416,
      "loss": 4.827,
      "mean_token_accuracy": 0.20413412749767304,
      "num_tokens": 140586639.0,
      "step": 64165
    },
    {
      "entropy": 5.74748215675354,
      "epoch": 6.368598650258039,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018352309888045303,
      "loss": 4.9757,
      "mean_token_accuracy": 0.19876070618629454,
      "num_tokens": 140597981.0,
      "step": 64170
    },
    {
      "entropy": 5.757516479492187,
      "epoch": 6.369094878920206,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018349072900979442,
      "loss": 4.9535,
      "mean_token_accuracy": 0.2004552885890007,
      "num_tokens": 140607112.0,
      "step": 64175
    },
    {
      "entropy": 5.690462923049926,
      "epoch": 6.369591107582374,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018345836140836862,
      "loss": 4.8919,
      "mean_token_accuracy": 0.20171308666467666,
      "num_tokens": 140617791.0,
      "step": 64180
    },
    {
      "entropy": 5.718970489501953,
      "epoch": 6.370087336244541,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001834259960769782,
      "loss": 4.901,
      "mean_token_accuracy": 0.1935281202197075,
      "num_tokens": 140629162.0,
      "step": 64185
    },
    {
      "entropy": 5.76987133026123,
      "epoch": 6.370583564906709,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018339363301642575,
      "loss": 4.9366,
      "mean_token_accuracy": 0.19758159667253494,
      "num_tokens": 140639791.0,
      "step": 64190
    },
    {
      "entropy": 5.745288562774658,
      "epoch": 6.371079793568876,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00018336127222751386,
      "loss": 5.0036,
      "mean_token_accuracy": 0.19861911684274675,
      "num_tokens": 140652312.0,
      "step": 64195
    },
    {
      "entropy": 5.657134580612182,
      "epoch": 6.371576022231044,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018332891371104489,
      "loss": 4.8959,
      "mean_token_accuracy": 0.1997937321662903,
      "num_tokens": 140663332.0,
      "step": 64200
    },
    {
      "entropy": 5.687296724319458,
      "epoch": 6.372072250893211,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018329655746782126,
      "loss": 4.8648,
      "mean_token_accuracy": 0.2058036893606186,
      "num_tokens": 140673844.0,
      "step": 64205
    },
    {
      "entropy": 5.695055627822876,
      "epoch": 6.372568479555379,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00018326420349864552,
      "loss": 4.8697,
      "mean_token_accuracy": 0.2007608160376549,
      "num_tokens": 140684281.0,
      "step": 64210
    },
    {
      "entropy": 5.706142616271973,
      "epoch": 6.373064708217546,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018323185180431985,
      "loss": 4.9764,
      "mean_token_accuracy": 0.1870686963200569,
      "num_tokens": 140694610.0,
      "step": 64215
    },
    {
      "entropy": 5.680764389038086,
      "epoch": 6.373560936879715,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018319950238564648,
      "loss": 4.8021,
      "mean_token_accuracy": 0.20920652896165848,
      "num_tokens": 140704718.0,
      "step": 64220
    },
    {
      "entropy": 5.604216766357422,
      "epoch": 6.374057165541882,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001831671552434277,
      "loss": 4.8359,
      "mean_token_accuracy": 0.20434404909610748,
      "num_tokens": 140715313.0,
      "step": 64225
    },
    {
      "entropy": 5.72027702331543,
      "epoch": 6.3745533942040495,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018313481037846542,
      "loss": 4.9319,
      "mean_token_accuracy": 0.2007973462343216,
      "num_tokens": 140725785.0,
      "step": 64230
    },
    {
      "entropy": 5.780927276611328,
      "epoch": 6.375049622866217,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018310246779156203,
      "loss": 4.9932,
      "mean_token_accuracy": 0.1945149526000023,
      "num_tokens": 140736983.0,
      "step": 64235
    },
    {
      "entropy": 5.606533622741699,
      "epoch": 6.3755458515283845,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00018307012748351937,
      "loss": 4.7617,
      "mean_token_accuracy": 0.20996496975421905,
      "num_tokens": 140747170.0,
      "step": 64240
    },
    {
      "entropy": 5.664201450347901,
      "epoch": 6.376042080190552,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018303778945513954,
      "loss": 4.9255,
      "mean_token_accuracy": 0.1953069657087326,
      "num_tokens": 140757150.0,
      "step": 64245
    },
    {
      "entropy": 5.700435304641724,
      "epoch": 6.376538308852719,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018300545370722427,
      "loss": 4.9269,
      "mean_token_accuracy": 0.20029883831739426,
      "num_tokens": 140768341.0,
      "step": 64250
    },
    {
      "entropy": 5.729792928695678,
      "epoch": 6.377034537514887,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018297312024057562,
      "loss": 4.9176,
      "mean_token_accuracy": 0.1966695323586464,
      "num_tokens": 140779123.0,
      "step": 64255
    },
    {
      "entropy": 5.702646207809448,
      "epoch": 6.377530766177054,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018294078905599508,
      "loss": 4.8741,
      "mean_token_accuracy": 0.20185130834579468,
      "num_tokens": 140790286.0,
      "step": 64260
    },
    {
      "entropy": 5.679545879364014,
      "epoch": 6.378026994839222,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001829084601542848,
      "loss": 4.8742,
      "mean_token_accuracy": 0.20442072600126265,
      "num_tokens": 140801272.0,
      "step": 64265
    },
    {
      "entropy": 5.7119481563568115,
      "epoch": 6.378523223501389,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018287613353624622,
      "loss": 4.9212,
      "mean_token_accuracy": 0.20169055163860322,
      "num_tokens": 140811728.0,
      "step": 64270
    },
    {
      "entropy": 5.719863653182983,
      "epoch": 6.379019452163557,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018284380920268102,
      "loss": 4.9682,
      "mean_token_accuracy": 0.18525414168834686,
      "num_tokens": 140823601.0,
      "step": 64275
    },
    {
      "entropy": 5.692059326171875,
      "epoch": 6.379515680825724,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018281148715439084,
      "loss": 4.8616,
      "mean_token_accuracy": 0.20418046712875365,
      "num_tokens": 140834232.0,
      "step": 64280
    },
    {
      "entropy": 5.7054359912872314,
      "epoch": 6.380011909487892,
      "grad_norm": 1.375,
      "learning_rate": 0.000182779167392177,
      "loss": 4.9123,
      "mean_token_accuracy": 0.20214020162820817,
      "num_tokens": 140844733.0,
      "step": 64285
    },
    {
      "entropy": 5.662629079818726,
      "epoch": 6.380508138150059,
      "grad_norm": 1.515625,
      "learning_rate": 0.00018274684991684105,
      "loss": 4.9396,
      "mean_token_accuracy": 0.19447498619556428,
      "num_tokens": 140854754.0,
      "step": 64290
    },
    {
      "entropy": 5.71527214050293,
      "epoch": 6.381004366812227,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001827145347291846,
      "loss": 4.9098,
      "mean_token_accuracy": 0.19839398115873336,
      "num_tokens": 140865455.0,
      "step": 64295
    },
    {
      "entropy": 5.764943075180054,
      "epoch": 6.381500595474394,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018268222183000872,
      "loss": 5.0587,
      "mean_token_accuracy": 0.18970625400543212,
      "num_tokens": 140877524.0,
      "step": 64300
    },
    {
      "entropy": 5.748681354522705,
      "epoch": 6.3819968241365626,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018264991122011489,
      "loss": 4.9495,
      "mean_token_accuracy": 0.1968034878373146,
      "num_tokens": 140888388.0,
      "step": 64305
    },
    {
      "entropy": 5.747801971435547,
      "epoch": 6.38249305279873,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00018261760290030421,
      "loss": 4.9673,
      "mean_token_accuracy": 0.19710808098316193,
      "num_tokens": 140898618.0,
      "step": 64310
    },
    {
      "entropy": 5.756382179260254,
      "epoch": 6.3829892814608975,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018258529687137793,
      "loss": 4.9674,
      "mean_token_accuracy": 0.19972231090068818,
      "num_tokens": 140909374.0,
      "step": 64315
    },
    {
      "entropy": 5.764496326446533,
      "epoch": 6.383485510123065,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001825529931341371,
      "loss": 4.9758,
      "mean_token_accuracy": 0.20178018659353256,
      "num_tokens": 140920458.0,
      "step": 64320
    },
    {
      "entropy": 5.690861320495605,
      "epoch": 6.3839817387852325,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001825206916893829,
      "loss": 4.9239,
      "mean_token_accuracy": 0.20509617030620575,
      "num_tokens": 140930727.0,
      "step": 64325
    },
    {
      "entropy": 5.7519035816192625,
      "epoch": 6.3844779674474,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018248839253791628,
      "loss": 4.9275,
      "mean_token_accuracy": 0.2015499010682106,
      "num_tokens": 140940815.0,
      "step": 64330
    },
    {
      "entropy": 5.68404188156128,
      "epoch": 6.384974196109567,
      "grad_norm": 1.375,
      "learning_rate": 0.00018245609568053818,
      "loss": 4.8534,
      "mean_token_accuracy": 0.20537610799074174,
      "num_tokens": 140952141.0,
      "step": 64335
    },
    {
      "entropy": 5.644695043563843,
      "epoch": 6.385470424771735,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018242380111804945,
      "loss": 4.8844,
      "mean_token_accuracy": 0.20149648785591126,
      "num_tokens": 140962687.0,
      "step": 64340
    },
    {
      "entropy": 5.6990679740905765,
      "epoch": 6.385966653433902,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018239150885125094,
      "loss": 4.9177,
      "mean_token_accuracy": 0.19936433881521226,
      "num_tokens": 140973084.0,
      "step": 64345
    },
    {
      "entropy": 5.711680507659912,
      "epoch": 6.38646288209607,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018235921888094353,
      "loss": 4.9174,
      "mean_token_accuracy": 0.19963138699531555,
      "num_tokens": 140983921.0,
      "step": 64350
    },
    {
      "entropy": 5.72605881690979,
      "epoch": 6.386959110758237,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018232693120792781,
      "loss": 4.9782,
      "mean_token_accuracy": 0.19519952982664107,
      "num_tokens": 140996789.0,
      "step": 64355
    },
    {
      "entropy": 5.734042739868164,
      "epoch": 6.387455339420405,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001822946458330046,
      "loss": 4.8781,
      "mean_token_accuracy": 0.20500514507293702,
      "num_tokens": 141006618.0,
      "step": 64360
    },
    {
      "entropy": 5.745874977111816,
      "epoch": 6.387951568082572,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018226236275697426,
      "loss": 4.9231,
      "mean_token_accuracy": 0.20227544009685516,
      "num_tokens": 141016262.0,
      "step": 64365
    },
    {
      "entropy": 5.722727727890015,
      "epoch": 6.38844779674474,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001822300819806377,
      "loss": 4.9274,
      "mean_token_accuracy": 0.2014557734131813,
      "num_tokens": 141026973.0,
      "step": 64370
    },
    {
      "entropy": 5.633992338180542,
      "epoch": 6.388944025406907,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018219780350479502,
      "loss": 4.8888,
      "mean_token_accuracy": 0.19568799436092377,
      "num_tokens": 141038427.0,
      "step": 64375
    },
    {
      "entropy": 5.6528177738189695,
      "epoch": 6.389440254069075,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018216552733024688,
      "loss": 4.9104,
      "mean_token_accuracy": 0.2013856589794159,
      "num_tokens": 141049828.0,
      "step": 64380
    },
    {
      "entropy": 5.687179374694824,
      "epoch": 6.389936482731242,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018213325345779364,
      "loss": 4.9042,
      "mean_token_accuracy": 0.203017982840538,
      "num_tokens": 141060771.0,
      "step": 64385
    },
    {
      "entropy": 5.693623208999634,
      "epoch": 6.3904327113934105,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001821009818882356,
      "loss": 4.8587,
      "mean_token_accuracy": 0.20018745809793473,
      "num_tokens": 141070283.0,
      "step": 64390
    },
    {
      "entropy": 5.66389307975769,
      "epoch": 6.390928940055578,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018206871262237308,
      "loss": 4.7647,
      "mean_token_accuracy": 0.2124108076095581,
      "num_tokens": 141080568.0,
      "step": 64395
    },
    {
      "entropy": 5.708030652999878,
      "epoch": 6.3914251687177455,
      "grad_norm": 1.1875,
      "learning_rate": 0.00018203644566100618,
      "loss": 4.9141,
      "mean_token_accuracy": 0.19718639999628068,
      "num_tokens": 141091402.0,
      "step": 64400
    },
    {
      "entropy": 5.671696043014526,
      "epoch": 6.391921397379913,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018200418100493515,
      "loss": 4.9419,
      "mean_token_accuracy": 0.1977217122912407,
      "num_tokens": 141101356.0,
      "step": 64405
    },
    {
      "entropy": 5.645659589767456,
      "epoch": 6.39241762604208,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018197191865495995,
      "loss": 4.871,
      "mean_token_accuracy": 0.20693568587303163,
      "num_tokens": 141112642.0,
      "step": 64410
    },
    {
      "entropy": 5.681612777709961,
      "epoch": 6.392913854704248,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018193965861188077,
      "loss": 4.8775,
      "mean_token_accuracy": 0.19974787086248397,
      "num_tokens": 141123338.0,
      "step": 64415
    },
    {
      "entropy": 5.720340633392334,
      "epoch": 6.393410083366415,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018190740087649753,
      "loss": 4.9528,
      "mean_token_accuracy": 0.20705569237470628,
      "num_tokens": 141135655.0,
      "step": 64420
    },
    {
      "entropy": 5.706362056732178,
      "epoch": 6.393906312028583,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001818751454496102,
      "loss": 4.8841,
      "mean_token_accuracy": 0.2045709192752838,
      "num_tokens": 141145401.0,
      "step": 64425
    },
    {
      "entropy": 5.6981299877166744,
      "epoch": 6.39440254069075,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018184289233201856,
      "loss": 4.8701,
      "mean_token_accuracy": 0.20559332519769669,
      "num_tokens": 141155936.0,
      "step": 64430
    },
    {
      "entropy": 5.722616481781006,
      "epoch": 6.394898769352918,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001818106415245225,
      "loss": 4.9552,
      "mean_token_accuracy": 0.19626902043819427,
      "num_tokens": 141168141.0,
      "step": 64435
    },
    {
      "entropy": 5.743749046325684,
      "epoch": 6.395394998015085,
      "grad_norm": 1.25,
      "learning_rate": 0.0001817783930279216,
      "loss": 4.9303,
      "mean_token_accuracy": 0.20364166051149368,
      "num_tokens": 141179317.0,
      "step": 64440
    },
    {
      "entropy": 5.736152839660645,
      "epoch": 6.395891226677253,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018174614684301577,
      "loss": 4.9035,
      "mean_token_accuracy": 0.19890190213918685,
      "num_tokens": 141190429.0,
      "step": 64445
    },
    {
      "entropy": 5.713809299468994,
      "epoch": 6.39638745533942,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018171390297060452,
      "loss": 4.9574,
      "mean_token_accuracy": 0.20613517314195634,
      "num_tokens": 141201782.0,
      "step": 64450
    },
    {
      "entropy": 5.69929609298706,
      "epoch": 6.396883684001588,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001816816614114875,
      "loss": 4.9357,
      "mean_token_accuracy": 0.20240603536367416,
      "num_tokens": 141213367.0,
      "step": 64455
    },
    {
      "entropy": 5.736020565032959,
      "epoch": 6.397379912663755,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018164942216646418,
      "loss": 4.9189,
      "mean_token_accuracy": 0.195868681371212,
      "num_tokens": 141224634.0,
      "step": 64460
    },
    {
      "entropy": 5.706407928466797,
      "epoch": 6.397876141325923,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018161718523633402,
      "loss": 4.9861,
      "mean_token_accuracy": 0.19385963380336763,
      "num_tokens": 141235131.0,
      "step": 64465
    },
    {
      "entropy": 5.691521453857422,
      "epoch": 6.39837236998809,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001815849506218964,
      "loss": 4.8707,
      "mean_token_accuracy": 0.2066229060292244,
      "num_tokens": 141245420.0,
      "step": 64470
    },
    {
      "entropy": 5.71909818649292,
      "epoch": 6.3988685986502585,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001815527183239508,
      "loss": 4.9048,
      "mean_token_accuracy": 0.20675126314163209,
      "num_tokens": 141256134.0,
      "step": 64475
    },
    {
      "entropy": 5.740906667709351,
      "epoch": 6.399364827312426,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018152048834329637,
      "loss": 4.9666,
      "mean_token_accuracy": 0.19746782183647155,
      "num_tokens": 141266619.0,
      "step": 64480
    },
    {
      "entropy": 5.683387422561646,
      "epoch": 6.399861055974593,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018148826068073242,
      "loss": 4.9072,
      "mean_token_accuracy": 0.20599694550037384,
      "num_tokens": 141279142.0,
      "step": 64485
    },
    {
      "entropy": 5.646283960342407,
      "epoch": 6.400357284636761,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018145603533705818,
      "loss": 4.8804,
      "mean_token_accuracy": 0.20320666581392288,
      "num_tokens": 141290135.0,
      "step": 64490
    },
    {
      "entropy": 5.680440235137939,
      "epoch": 6.400853513298928,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018142381231307252,
      "loss": 4.9601,
      "mean_token_accuracy": 0.1960691511631012,
      "num_tokens": 141301927.0,
      "step": 64495
    },
    {
      "entropy": 5.717540264129639,
      "epoch": 6.401349741961096,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001813915916095748,
      "loss": 4.9637,
      "mean_token_accuracy": 0.19815190732479096,
      "num_tokens": 141312783.0,
      "step": 64500
    },
    {
      "entropy": 5.74260950088501,
      "epoch": 6.401845970623263,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001813593732273639,
      "loss": 4.9326,
      "mean_token_accuracy": 0.19425464570522308,
      "num_tokens": 141323494.0,
      "step": 64505
    },
    {
      "entropy": 5.7382577419281,
      "epoch": 6.402342199285431,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001813271571672388,
      "loss": 4.9213,
      "mean_token_accuracy": 0.20124197900295257,
      "num_tokens": 141334776.0,
      "step": 64510
    },
    {
      "entropy": 5.7226433753967285,
      "epoch": 6.402838427947598,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001812949434299983,
      "loss": 4.9828,
      "mean_token_accuracy": 0.19242235124111176,
      "num_tokens": 141345161.0,
      "step": 64515
    },
    {
      "entropy": 5.683582592010498,
      "epoch": 6.403334656609766,
      "grad_norm": 1.375,
      "learning_rate": 0.0001812627320164412,
      "loss": 4.9715,
      "mean_token_accuracy": 0.20172838717699051,
      "num_tokens": 141357104.0,
      "step": 64520
    },
    {
      "entropy": 5.739828109741211,
      "epoch": 6.403830885271933,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018123052292736642,
      "loss": 4.8697,
      "mean_token_accuracy": 0.2034697115421295,
      "num_tokens": 141367852.0,
      "step": 64525
    },
    {
      "entropy": 5.767837953567505,
      "epoch": 6.404327113934101,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00018119831616357263,
      "loss": 4.894,
      "mean_token_accuracy": 0.2033145785331726,
      "num_tokens": 141378387.0,
      "step": 64530
    },
    {
      "entropy": 5.687147760391236,
      "epoch": 6.404823342596268,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001811661117258585,
      "loss": 4.896,
      "mean_token_accuracy": 0.20775501877069474,
      "num_tokens": 141387980.0,
      "step": 64535
    },
    {
      "entropy": 5.67966775894165,
      "epoch": 6.405319571258436,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018113390961502264,
      "loss": 4.845,
      "mean_token_accuracy": 0.21030091792345046,
      "num_tokens": 141398897.0,
      "step": 64540
    },
    {
      "entropy": 5.66959547996521,
      "epoch": 6.405815799920603,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001811017098318636,
      "loss": 4.9087,
      "mean_token_accuracy": 0.19914028346538543,
      "num_tokens": 141410780.0,
      "step": 64545
    },
    {
      "entropy": 5.69325852394104,
      "epoch": 6.406312028582771,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018106951237717962,
      "loss": 4.9248,
      "mean_token_accuracy": 0.19681723415851593,
      "num_tokens": 141422305.0,
      "step": 64550
    },
    {
      "entropy": 5.646556234359741,
      "epoch": 6.406808257244938,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018103731725176947,
      "loss": 4.8616,
      "mean_token_accuracy": 0.198015758395195,
      "num_tokens": 141434201.0,
      "step": 64555
    },
    {
      "entropy": 5.632434320449829,
      "epoch": 6.407304485907106,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001810051244564313,
      "loss": 4.8236,
      "mean_token_accuracy": 0.20954911559820175,
      "num_tokens": 141444879.0,
      "step": 64560
    },
    {
      "entropy": 5.743421459197998,
      "epoch": 6.407800714569274,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018097293399196358,
      "loss": 4.9408,
      "mean_token_accuracy": 0.19533207565546035,
      "num_tokens": 141454645.0,
      "step": 64565
    },
    {
      "entropy": 5.6790965557098385,
      "epoch": 6.408296943231441,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018094074585916448,
      "loss": 4.8752,
      "mean_token_accuracy": 0.1976602479815483,
      "num_tokens": 141464446.0,
      "step": 64570
    },
    {
      "entropy": 5.756632328033447,
      "epoch": 6.408793171893609,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001809085600588322,
      "loss": 5.0087,
      "mean_token_accuracy": 0.19206585884094238,
      "num_tokens": 141474417.0,
      "step": 64575
    },
    {
      "entropy": 5.694465255737304,
      "epoch": 6.409289400555776,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018087637659176491,
      "loss": 4.8964,
      "mean_token_accuracy": 0.19806654751300812,
      "num_tokens": 141485140.0,
      "step": 64580
    },
    {
      "entropy": 5.707103586196899,
      "epoch": 6.409785629217944,
      "grad_norm": 1.375,
      "learning_rate": 0.00018084419545876058,
      "loss": 4.8931,
      "mean_token_accuracy": 0.20312961041927338,
      "num_tokens": 141496737.0,
      "step": 64585
    },
    {
      "entropy": 5.62021484375,
      "epoch": 6.410281857880111,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018081201666061725,
      "loss": 4.8041,
      "mean_token_accuracy": 0.20162529796361922,
      "num_tokens": 141507772.0,
      "step": 64590
    },
    {
      "entropy": 5.638330936431885,
      "epoch": 6.410778086542279,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001807798401981331,
      "loss": 4.8893,
      "mean_token_accuracy": 0.20082218796014786,
      "num_tokens": 141518516.0,
      "step": 64595
    },
    {
      "entropy": 5.721786642074585,
      "epoch": 6.411274315204446,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018074766607210585,
      "loss": 4.9557,
      "mean_token_accuracy": 0.2028407171368599,
      "num_tokens": 141530194.0,
      "step": 64600
    },
    {
      "entropy": 5.7094779968261715,
      "epoch": 6.411770543866614,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018071549428333344,
      "loss": 4.92,
      "mean_token_accuracy": 0.19696082621812822,
      "num_tokens": 141540616.0,
      "step": 64605
    },
    {
      "entropy": 5.6441394805908205,
      "epoch": 6.412266772528781,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018068332483261358,
      "loss": 4.8401,
      "mean_token_accuracy": 0.20791476517915725,
      "num_tokens": 141551508.0,
      "step": 64610
    },
    {
      "entropy": 5.654355430603028,
      "epoch": 6.412763001190949,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018065115772074392,
      "loss": 4.8569,
      "mean_token_accuracy": 0.20007718056440355,
      "num_tokens": 141562341.0,
      "step": 64615
    },
    {
      "entropy": 5.773579072952271,
      "epoch": 6.413259229853116,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018061899294852224,
      "loss": 5.0192,
      "mean_token_accuracy": 0.1925663948059082,
      "num_tokens": 141574157.0,
      "step": 64620
    },
    {
      "entropy": 5.648352813720703,
      "epoch": 6.413755458515284,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001805868305167463,
      "loss": 4.8574,
      "mean_token_accuracy": 0.207219859957695,
      "num_tokens": 141584474.0,
      "step": 64625
    },
    {
      "entropy": 5.677443122863769,
      "epoch": 6.414251687177451,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018055467042621344,
      "loss": 4.8636,
      "mean_token_accuracy": 0.2008536532521248,
      "num_tokens": 141596365.0,
      "step": 64630
    },
    {
      "entropy": 5.749535846710205,
      "epoch": 6.414747915839619,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018052251267772128,
      "loss": 4.9969,
      "mean_token_accuracy": 0.19792632460594178,
      "num_tokens": 141608859.0,
      "step": 64635
    },
    {
      "entropy": 5.713398551940918,
      "epoch": 6.415244144501786,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018049035727206724,
      "loss": 4.96,
      "mean_token_accuracy": 0.1992316871881485,
      "num_tokens": 141619746.0,
      "step": 64640
    },
    {
      "entropy": 5.703238248825073,
      "epoch": 6.415740373163954,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001804582042100486,
      "loss": 4.8377,
      "mean_token_accuracy": 0.19802510142326354,
      "num_tokens": 141630094.0,
      "step": 64645
    },
    {
      "entropy": 5.627750396728516,
      "epoch": 6.416236601826122,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018042605349246287,
      "loss": 4.7912,
      "mean_token_accuracy": 0.20589603930711747,
      "num_tokens": 141641359.0,
      "step": 64650
    },
    {
      "entropy": 5.641937685012818,
      "epoch": 6.416732830488289,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001803939051201071,
      "loss": 4.8666,
      "mean_token_accuracy": 0.20033324211835862,
      "num_tokens": 141651540.0,
      "step": 64655
    },
    {
      "entropy": 5.678428745269775,
      "epoch": 6.417229059150457,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018036175909377878,
      "loss": 4.9648,
      "mean_token_accuracy": 0.19354701787233353,
      "num_tokens": 141663638.0,
      "step": 64660
    },
    {
      "entropy": 5.711199331283569,
      "epoch": 6.417725287812624,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018032961541427484,
      "loss": 4.9122,
      "mean_token_accuracy": 0.1919893890619278,
      "num_tokens": 141675232.0,
      "step": 64665
    },
    {
      "entropy": 5.6319504737854,
      "epoch": 6.418221516474792,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001802974740823925,
      "loss": 4.8865,
      "mean_token_accuracy": 0.2018788531422615,
      "num_tokens": 141686490.0,
      "step": 64670
    },
    {
      "entropy": 5.589042997360229,
      "epoch": 6.418717745136959,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018026533509892862,
      "loss": 4.8125,
      "mean_token_accuracy": 0.21140512079000473,
      "num_tokens": 141697397.0,
      "step": 64675
    },
    {
      "entropy": 5.751994037628174,
      "epoch": 6.419213973799127,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001802331984646804,
      "loss": 4.9803,
      "mean_token_accuracy": 0.1973927989602089,
      "num_tokens": 141708076.0,
      "step": 64680
    },
    {
      "entropy": 5.63619966506958,
      "epoch": 6.419710202461294,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018020106418044447,
      "loss": 4.8176,
      "mean_token_accuracy": 0.21100780814886094,
      "num_tokens": 141719309.0,
      "step": 64685
    },
    {
      "entropy": 5.718036079406739,
      "epoch": 6.420206431123462,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018016893224701808,
      "loss": 4.9106,
      "mean_token_accuracy": 0.2022289514541626,
      "num_tokens": 141730706.0,
      "step": 64690
    },
    {
      "entropy": 5.74221715927124,
      "epoch": 6.420702659785629,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018013680266519777,
      "loss": 4.9317,
      "mean_token_accuracy": 0.19470663666725158,
      "num_tokens": 141741517.0,
      "step": 64695
    },
    {
      "entropy": 5.714552879333496,
      "epoch": 6.421198888447797,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018010467543578035,
      "loss": 4.8834,
      "mean_token_accuracy": 0.19754279255867005,
      "num_tokens": 141752771.0,
      "step": 64700
    },
    {
      "entropy": 5.731656408309936,
      "epoch": 6.421695117109964,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001800725505595624,
      "loss": 4.9972,
      "mean_token_accuracy": 0.19577919393777848,
      "num_tokens": 141763447.0,
      "step": 64705
    },
    {
      "entropy": 5.659796714782715,
      "epoch": 6.422191345772132,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018004042803734078,
      "loss": 4.8379,
      "mean_token_accuracy": 0.2061975359916687,
      "num_tokens": 141774059.0,
      "step": 64710
    },
    {
      "entropy": 5.6774498462677006,
      "epoch": 6.422687574434299,
      "grad_norm": 1.203125,
      "learning_rate": 0.00018000830786991186,
      "loss": 4.8602,
      "mean_token_accuracy": 0.21200665682554246,
      "num_tokens": 141784449.0,
      "step": 64715
    },
    {
      "entropy": 5.727631711959839,
      "epoch": 6.4231838030964665,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017997619005807218,
      "loss": 5.0,
      "mean_token_accuracy": 0.20136308372020723,
      "num_tokens": 141794852.0,
      "step": 64720
    },
    {
      "entropy": 5.744229364395141,
      "epoch": 6.423680031758634,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001799440746026183,
      "loss": 4.9366,
      "mean_token_accuracy": 0.2046593263745308,
      "num_tokens": 141806514.0,
      "step": 64725
    },
    {
      "entropy": 5.706175661087036,
      "epoch": 6.424176260420802,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017991196150434646,
      "loss": 4.8968,
      "mean_token_accuracy": 0.19785246104002,
      "num_tokens": 141817077.0,
      "step": 64730
    },
    {
      "entropy": 5.698449993133545,
      "epoch": 6.42467248908297,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00017987985076405318,
      "loss": 4.8557,
      "mean_token_accuracy": 0.20461099445819855,
      "num_tokens": 141829346.0,
      "step": 64735
    },
    {
      "entropy": 5.684509134292602,
      "epoch": 6.425168717745137,
      "grad_norm": 1.375,
      "learning_rate": 0.00017984774238253458,
      "loss": 4.9031,
      "mean_token_accuracy": 0.20040879398584366,
      "num_tokens": 141840419.0,
      "step": 64740
    },
    {
      "entropy": 5.656356525421143,
      "epoch": 6.425664946407305,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017981563636058695,
      "loss": 4.8508,
      "mean_token_accuracy": 0.21055281907320023,
      "num_tokens": 141851295.0,
      "step": 64745
    },
    {
      "entropy": 5.63439531326294,
      "epoch": 6.426161175069472,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001797835326990064,
      "loss": 4.8817,
      "mean_token_accuracy": 0.20894320011138917,
      "num_tokens": 141861011.0,
      "step": 64750
    },
    {
      "entropy": 5.748449468612671,
      "epoch": 6.42665740373164,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017975143139858913,
      "loss": 4.9986,
      "mean_token_accuracy": 0.1931667670607567,
      "num_tokens": 141871216.0,
      "step": 64755
    },
    {
      "entropy": 5.7061896324157715,
      "epoch": 6.427153632393807,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000179719332460131,
      "loss": 4.9149,
      "mean_token_accuracy": 0.2014117196202278,
      "num_tokens": 141880719.0,
      "step": 64760
    },
    {
      "entropy": 5.715649080276489,
      "epoch": 6.427649861055975,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017968723588442824,
      "loss": 4.9042,
      "mean_token_accuracy": 0.20424285680055618,
      "num_tokens": 141891558.0,
      "step": 64765
    },
    {
      "entropy": 5.666348791122436,
      "epoch": 6.428146089718142,
      "grad_norm": 1.375,
      "learning_rate": 0.00017965514167227659,
      "loss": 4.8684,
      "mean_token_accuracy": 0.2038439095020294,
      "num_tokens": 141901365.0,
      "step": 64770
    },
    {
      "entropy": 5.7100080966949465,
      "epoch": 6.42864231838031,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00017962304982447203,
      "loss": 4.9722,
      "mean_token_accuracy": 0.20017272233963013,
      "num_tokens": 141913202.0,
      "step": 64775
    },
    {
      "entropy": 5.698890733718872,
      "epoch": 6.429138547042477,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001795909603418102,
      "loss": 4.9983,
      "mean_token_accuracy": 0.1909737169742584,
      "num_tokens": 141925389.0,
      "step": 64780
    },
    {
      "entropy": 5.76895923614502,
      "epoch": 6.429634775704645,
      "grad_norm": 1.375,
      "learning_rate": 0.00017955887322508707,
      "loss": 5.0046,
      "mean_token_accuracy": 0.20245164334774018,
      "num_tokens": 141936328.0,
      "step": 64785
    },
    {
      "entropy": 5.674752235412598,
      "epoch": 6.430131004366812,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017952678847509818,
      "loss": 4.8619,
      "mean_token_accuracy": 0.20569155216217042,
      "num_tokens": 141947895.0,
      "step": 64790
    },
    {
      "entropy": 5.702035760879516,
      "epoch": 6.4306272330289795,
      "grad_norm": 1.46875,
      "learning_rate": 0.00017949470609263923,
      "loss": 4.8877,
      "mean_token_accuracy": 0.2037683129310608,
      "num_tokens": 141957799.0,
      "step": 64795
    },
    {
      "entropy": 5.693123388290405,
      "epoch": 6.431123461691147,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00017946262607850582,
      "loss": 4.9503,
      "mean_token_accuracy": 0.2026366338133812,
      "num_tokens": 141968756.0,
      "step": 64800
    },
    {
      "entropy": 5.675360488891601,
      "epoch": 6.4316196903533145,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017943054843349337,
      "loss": 4.8844,
      "mean_token_accuracy": 0.2001672640442848,
      "num_tokens": 141979529.0,
      "step": 64805
    },
    {
      "entropy": 5.70154299736023,
      "epoch": 6.432115919015482,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001793984731583974,
      "loss": 4.9644,
      "mean_token_accuracy": 0.19816241413354874,
      "num_tokens": 141990570.0,
      "step": 64810
    },
    {
      "entropy": 5.676688385009766,
      "epoch": 6.43261214767765,
      "grad_norm": 1.546875,
      "learning_rate": 0.00017936640025401323,
      "loss": 4.8935,
      "mean_token_accuracy": 0.20528977662324904,
      "num_tokens": 141999716.0,
      "step": 64815
    },
    {
      "entropy": 5.718646383285522,
      "epoch": 6.433108376339818,
      "grad_norm": 1.5625,
      "learning_rate": 0.00017933432972113623,
      "loss": 4.9024,
      "mean_token_accuracy": 0.1977451115846634,
      "num_tokens": 142010166.0,
      "step": 64820
    },
    {
      "entropy": 5.731052875518799,
      "epoch": 6.433604605001985,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017930226156056184,
      "loss": 4.9297,
      "mean_token_accuracy": 0.19846864342689513,
      "num_tokens": 142022388.0,
      "step": 64825
    },
    {
      "entropy": 5.685515546798706,
      "epoch": 6.434100833664153,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017927019577308512,
      "loss": 4.904,
      "mean_token_accuracy": 0.20039243400096893,
      "num_tokens": 142032611.0,
      "step": 64830
    },
    {
      "entropy": 5.595603084564209,
      "epoch": 6.43459706232632,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017923813235950126,
      "loss": 4.7683,
      "mean_token_accuracy": 0.20980293154716492,
      "num_tokens": 142043723.0,
      "step": 64835
    },
    {
      "entropy": 5.656597995758057,
      "epoch": 6.435093290988488,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001792060713206054,
      "loss": 4.8584,
      "mean_token_accuracy": 0.19924768507480622,
      "num_tokens": 142053535.0,
      "step": 64840
    },
    {
      "entropy": 5.628881359100342,
      "epoch": 6.435589519650655,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017917401265719242,
      "loss": 4.846,
      "mean_token_accuracy": 0.20215778052806854,
      "num_tokens": 142063583.0,
      "step": 64845
    },
    {
      "entropy": 5.728031396865845,
      "epoch": 6.436085748312823,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001791419563700575,
      "loss": 4.966,
      "mean_token_accuracy": 0.19613841772079468,
      "num_tokens": 142073848.0,
      "step": 64850
    },
    {
      "entropy": 5.7276201248168945,
      "epoch": 6.43658197697499,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017910990245999554,
      "loss": 4.9619,
      "mean_token_accuracy": 0.19635746031999587,
      "num_tokens": 142082880.0,
      "step": 64855
    },
    {
      "entropy": 5.713687515258789,
      "epoch": 6.437078205637158,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017907785092780143,
      "loss": 4.9292,
      "mean_token_accuracy": 0.20140100866556168,
      "num_tokens": 142093905.0,
      "step": 64860
    },
    {
      "entropy": 5.671404314041138,
      "epoch": 6.437574434299325,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017904580177426988,
      "loss": 4.8532,
      "mean_token_accuracy": 0.2059663012623787,
      "num_tokens": 142105415.0,
      "step": 64865
    },
    {
      "entropy": 5.679664421081543,
      "epoch": 6.4380706629614926,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017901375500019567,
      "loss": 4.8407,
      "mean_token_accuracy": 0.20345701426267623,
      "num_tokens": 142115607.0,
      "step": 64870
    },
    {
      "entropy": 5.64707727432251,
      "epoch": 6.43856689162366,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017898171060637353,
      "loss": 4.9267,
      "mean_token_accuracy": 0.2007965162396431,
      "num_tokens": 142126008.0,
      "step": 64875
    },
    {
      "entropy": 5.694258213043213,
      "epoch": 6.4390631202858275,
      "grad_norm": 1.203125,
      "learning_rate": 0.00017894966859359802,
      "loss": 5.0127,
      "mean_token_accuracy": 0.1958266317844391,
      "num_tokens": 142138126.0,
      "step": 64880
    },
    {
      "entropy": 5.8206085681915285,
      "epoch": 6.439559348947995,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017891762896266367,
      "loss": 5.0256,
      "mean_token_accuracy": 0.1936203956604004,
      "num_tokens": 142150552.0,
      "step": 64885
    },
    {
      "entropy": 5.730115795135498,
      "epoch": 6.4400555776101625,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001788855917143652,
      "loss": 4.8971,
      "mean_token_accuracy": 0.20095257014036177,
      "num_tokens": 142160922.0,
      "step": 64890
    },
    {
      "entropy": 5.746193218231201,
      "epoch": 6.44055180627233,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000178853556849497,
      "loss": 4.9027,
      "mean_token_accuracy": 0.20239651799201966,
      "num_tokens": 142172026.0,
      "step": 64895
    },
    {
      "entropy": 5.663785839080811,
      "epoch": 6.441048034934497,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017882152436885334,
      "loss": 4.897,
      "mean_token_accuracy": 0.20041343867778777,
      "num_tokens": 142183867.0,
      "step": 64900
    },
    {
      "entropy": 5.65174880027771,
      "epoch": 6.441544263596665,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00017878949427322865,
      "loss": 4.8796,
      "mean_token_accuracy": 0.20786049365997314,
      "num_tokens": 142195199.0,
      "step": 64905
    },
    {
      "entropy": 5.692864656448364,
      "epoch": 6.442040492258833,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017875746656341712,
      "loss": 4.8622,
      "mean_token_accuracy": 0.20413007885217666,
      "num_tokens": 142205651.0,
      "step": 64910
    },
    {
      "entropy": 5.684857988357544,
      "epoch": 6.442536720921001,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017872544124021306,
      "loss": 4.9474,
      "mean_token_accuracy": 0.1947985976934433,
      "num_tokens": 142218347.0,
      "step": 64915
    },
    {
      "entropy": 5.688266897201538,
      "epoch": 6.443032949583168,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017869341830441067,
      "loss": 4.8756,
      "mean_token_accuracy": 0.20077381283044815,
      "num_tokens": 142228839.0,
      "step": 64920
    },
    {
      "entropy": 5.774335241317749,
      "epoch": 6.443529178245336,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017866139775680396,
      "loss": 5.0239,
      "mean_token_accuracy": 0.19262947291135787,
      "num_tokens": 142239477.0,
      "step": 64925
    },
    {
      "entropy": 5.735717678070069,
      "epoch": 6.444025406907503,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017862937959818698,
      "loss": 4.9311,
      "mean_token_accuracy": 0.20049121528863906,
      "num_tokens": 142249874.0,
      "step": 64930
    },
    {
      "entropy": 5.748757410049438,
      "epoch": 6.444521635569671,
      "grad_norm": 1.375,
      "learning_rate": 0.00017859736382935375,
      "loss": 4.943,
      "mean_token_accuracy": 0.20273570865392684,
      "num_tokens": 142260799.0,
      "step": 64935
    },
    {
      "entropy": 5.70927677154541,
      "epoch": 6.445017864231838,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017856535045109807,
      "loss": 4.9516,
      "mean_token_accuracy": 0.19624215811491014,
      "num_tokens": 142271612.0,
      "step": 64940
    },
    {
      "entropy": 5.656954669952393,
      "epoch": 6.445514092894006,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017853333946421402,
      "loss": 4.8193,
      "mean_token_accuracy": 0.21144130676984788,
      "num_tokens": 142282395.0,
      "step": 64945
    },
    {
      "entropy": 5.7406717300415036,
      "epoch": 6.446010321556173,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00017850133086949515,
      "loss": 4.9769,
      "mean_token_accuracy": 0.1986526444554329,
      "num_tokens": 142294200.0,
      "step": 64950
    },
    {
      "entropy": 5.695454692840576,
      "epoch": 6.4465065502183405,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017846932466773551,
      "loss": 4.8491,
      "mean_token_accuracy": 0.20139432102441787,
      "num_tokens": 142304386.0,
      "step": 64955
    },
    {
      "entropy": 5.685807466506958,
      "epoch": 6.447002778880508,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017843732085972853,
      "loss": 4.8249,
      "mean_token_accuracy": 0.2064025655388832,
      "num_tokens": 142315252.0,
      "step": 64960
    },
    {
      "entropy": 5.686699295043946,
      "epoch": 6.4474990075426755,
      "grad_norm": 1.203125,
      "learning_rate": 0.00017840531944626796,
      "loss": 4.8815,
      "mean_token_accuracy": 0.20666262358427048,
      "num_tokens": 142326168.0,
      "step": 64965
    },
    {
      "entropy": 5.654635763168335,
      "epoch": 6.447995236204843,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017837332042814719,
      "loss": 4.8397,
      "mean_token_accuracy": 0.20501139312982558,
      "num_tokens": 142336978.0,
      "step": 64970
    },
    {
      "entropy": 5.726064157485962,
      "epoch": 6.44849146486701,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017834132380616002,
      "loss": 4.9812,
      "mean_token_accuracy": 0.19499383568763734,
      "num_tokens": 142347812.0,
      "step": 64975
    },
    {
      "entropy": 5.694739913940429,
      "epoch": 6.448987693529178,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017830932958109958,
      "loss": 4.8366,
      "mean_token_accuracy": 0.201439169049263,
      "num_tokens": 142358289.0,
      "step": 64980
    },
    {
      "entropy": 5.756964445114136,
      "epoch": 6.449483922191345,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017827733775375955,
      "loss": 4.9397,
      "mean_token_accuracy": 0.20026437044143677,
      "num_tokens": 142368394.0,
      "step": 64985
    },
    {
      "entropy": 5.666964292526245,
      "epoch": 6.449980150853513,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001782453483249331,
      "loss": 4.9069,
      "mean_token_accuracy": 0.20300507843494414,
      "num_tokens": 142380226.0,
      "step": 64990
    },
    {
      "entropy": 5.7178730964660645,
      "epoch": 6.450476379515681,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001782133612954135,
      "loss": 4.9476,
      "mean_token_accuracy": 0.1953831911087036,
      "num_tokens": 142390730.0,
      "step": 64995
    },
    {
      "entropy": 5.7228288650512695,
      "epoch": 6.450972608177849,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000178181376665994,
      "loss": 4.9608,
      "mean_token_accuracy": 0.19699962586164474,
      "num_tokens": 142403026.0,
      "step": 65000
    },
    {
      "entropy": 5.727644968032837,
      "epoch": 6.451468836840016,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017814939443746775,
      "loss": 4.8906,
      "mean_token_accuracy": 0.20079285502433777,
      "num_tokens": 142414111.0,
      "step": 65005
    },
    {
      "entropy": 5.7052857875823975,
      "epoch": 6.451965065502184,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001781174146106278,
      "loss": 4.9051,
      "mean_token_accuracy": 0.20298154652118683,
      "num_tokens": 142426877.0,
      "step": 65010
    },
    {
      "entropy": 5.6691858768463135,
      "epoch": 6.452461294164351,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017808543718626714,
      "loss": 4.9091,
      "mean_token_accuracy": 0.19930787831544877,
      "num_tokens": 142437853.0,
      "step": 65015
    },
    {
      "entropy": 5.6902250289917,
      "epoch": 6.452957522826519,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017805346216517891,
      "loss": 4.9117,
      "mean_token_accuracy": 0.19731711596250534,
      "num_tokens": 142448843.0,
      "step": 65020
    },
    {
      "entropy": 5.686580991744995,
      "epoch": 6.453453751488686,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017802148954815578,
      "loss": 4.8658,
      "mean_token_accuracy": 0.2061750441789627,
      "num_tokens": 142459099.0,
      "step": 65025
    },
    {
      "entropy": 5.648319959640503,
      "epoch": 6.4539499801508535,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017798951933599088,
      "loss": 4.9043,
      "mean_token_accuracy": 0.2058124542236328,
      "num_tokens": 142470373.0,
      "step": 65030
    },
    {
      "entropy": 5.671259498596191,
      "epoch": 6.454446208813021,
      "grad_norm": 1.203125,
      "learning_rate": 0.00017795755152947684,
      "loss": 4.8811,
      "mean_token_accuracy": 0.20576899647712707,
      "num_tokens": 142480204.0,
      "step": 65035
    },
    {
      "entropy": 5.710053730010986,
      "epoch": 6.4549424374751885,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001779255861294064,
      "loss": 4.9416,
      "mean_token_accuracy": 0.20199519395828247,
      "num_tokens": 142491110.0,
      "step": 65040
    },
    {
      "entropy": 5.654209089279175,
      "epoch": 6.455438666137356,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001778936231365722,
      "loss": 4.8275,
      "mean_token_accuracy": 0.20937645733356475,
      "num_tokens": 142502194.0,
      "step": 65045
    },
    {
      "entropy": 5.663603639602661,
      "epoch": 6.455934894799523,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000177861662551767,
      "loss": 4.9215,
      "mean_token_accuracy": 0.2040819138288498,
      "num_tokens": 142514251.0,
      "step": 65050
    },
    {
      "entropy": 5.711012458801269,
      "epoch": 6.456431123461691,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001778297043757831,
      "loss": 4.8992,
      "mean_token_accuracy": 0.1995910108089447,
      "num_tokens": 142524208.0,
      "step": 65055
    },
    {
      "entropy": 5.718306207656861,
      "epoch": 6.456927352123858,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017779774860941327,
      "loss": 4.8843,
      "mean_token_accuracy": 0.20167796313762665,
      "num_tokens": 142535622.0,
      "step": 65060
    },
    {
      "entropy": 5.689274597167969,
      "epoch": 6.457423580786026,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017776579525344982,
      "loss": 4.8815,
      "mean_token_accuracy": 0.20334179401397706,
      "num_tokens": 142548277.0,
      "step": 65065
    },
    {
      "entropy": 5.6643595695495605,
      "epoch": 6.457919809448193,
      "grad_norm": 1.375,
      "learning_rate": 0.0001777338443086851,
      "loss": 4.9101,
      "mean_token_accuracy": 0.2012820288538933,
      "num_tokens": 142559349.0,
      "step": 65070
    },
    {
      "entropy": 5.645092439651489,
      "epoch": 6.458416038110361,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001777018957759114,
      "loss": 4.8566,
      "mean_token_accuracy": 0.21039385050535203,
      "num_tokens": 142570515.0,
      "step": 65075
    },
    {
      "entropy": 5.7125977039337155,
      "epoch": 6.458912266772529,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001776699496559211,
      "loss": 4.9316,
      "mean_token_accuracy": 0.2004665106534958,
      "num_tokens": 142583104.0,
      "step": 65080
    },
    {
      "entropy": 5.638589906692505,
      "epoch": 6.459408495434697,
      "grad_norm": 1.25,
      "learning_rate": 0.00017763800594950625,
      "loss": 4.8718,
      "mean_token_accuracy": 0.20503858029842376,
      "num_tokens": 142593672.0,
      "step": 65085
    },
    {
      "entropy": 5.751476907730103,
      "epoch": 6.459904724096864,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001776060646574591,
      "loss": 4.9623,
      "mean_token_accuracy": 0.1897532030940056,
      "num_tokens": 142604913.0,
      "step": 65090
    },
    {
      "entropy": 5.682224988937378,
      "epoch": 6.460400952759032,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017757412578057175,
      "loss": 4.8993,
      "mean_token_accuracy": 0.20244086682796478,
      "num_tokens": 142617573.0,
      "step": 65095
    },
    {
      "entropy": 5.672818946838379,
      "epoch": 6.460897181421199,
      "grad_norm": 1.46875,
      "learning_rate": 0.00017754218931963613,
      "loss": 4.8029,
      "mean_token_accuracy": 0.20845697969198226,
      "num_tokens": 142627406.0,
      "step": 65100
    },
    {
      "entropy": 5.628749513626099,
      "epoch": 6.4613934100833665,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017751025527544417,
      "loss": 4.8807,
      "mean_token_accuracy": 0.20436931401491165,
      "num_tokens": 142637373.0,
      "step": 65105
    },
    {
      "entropy": 5.701884603500366,
      "epoch": 6.461889638745534,
      "grad_norm": 1.375,
      "learning_rate": 0.00017747832364878769,
      "loss": 4.9376,
      "mean_token_accuracy": 0.19857622385025026,
      "num_tokens": 142648187.0,
      "step": 65110
    },
    {
      "entropy": 5.678127288818359,
      "epoch": 6.4623858674077015,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017744639444045867,
      "loss": 4.8862,
      "mean_token_accuracy": 0.20200430750846862,
      "num_tokens": 142658855.0,
      "step": 65115
    },
    {
      "entropy": 5.671934604644775,
      "epoch": 6.462882096069869,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017741446765124903,
      "loss": 4.9355,
      "mean_token_accuracy": 0.19842273741960526,
      "num_tokens": 142669338.0,
      "step": 65120
    },
    {
      "entropy": 5.713703203201294,
      "epoch": 6.463378324732036,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001773825432819502,
      "loss": 4.942,
      "mean_token_accuracy": 0.19856634438037873,
      "num_tokens": 142679959.0,
      "step": 65125
    },
    {
      "entropy": 5.764170837402344,
      "epoch": 6.463874553394204,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017735062133335404,
      "loss": 4.9954,
      "mean_token_accuracy": 0.1930038183927536,
      "num_tokens": 142691766.0,
      "step": 65130
    },
    {
      "entropy": 5.662719964981079,
      "epoch": 6.464370782056371,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000177318701806252,
      "loss": 4.8724,
      "mean_token_accuracy": 0.1972167819738388,
      "num_tokens": 142703189.0,
      "step": 65135
    },
    {
      "entropy": 5.717951059341431,
      "epoch": 6.464867010718539,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001772867847014356,
      "loss": 4.921,
      "mean_token_accuracy": 0.2108253315091133,
      "num_tokens": 142717051.0,
      "step": 65140
    },
    {
      "entropy": 5.729372882843018,
      "epoch": 6.465363239380706,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017725487001969638,
      "loss": 4.9795,
      "mean_token_accuracy": 0.1978456676006317,
      "num_tokens": 142728051.0,
      "step": 65145
    },
    {
      "entropy": 5.73871078491211,
      "epoch": 6.465859468042874,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017722295776182584,
      "loss": 4.9123,
      "mean_token_accuracy": 0.20038407891988755,
      "num_tokens": 142739198.0,
      "step": 65150
    },
    {
      "entropy": 5.728751945495605,
      "epoch": 6.466355696705041,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017719104792861528,
      "loss": 4.8791,
      "mean_token_accuracy": 0.1976843774318695,
      "num_tokens": 142750303.0,
      "step": 65155
    },
    {
      "entropy": 5.692975854873657,
      "epoch": 6.466851925367209,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017715914052085597,
      "loss": 4.94,
      "mean_token_accuracy": 0.2022380918264389,
      "num_tokens": 142760435.0,
      "step": 65160
    },
    {
      "entropy": 5.66571741104126,
      "epoch": 6.467348154029377,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017712723553933913,
      "loss": 4.9143,
      "mean_token_accuracy": 0.194686421751976,
      "num_tokens": 142771202.0,
      "step": 65165
    },
    {
      "entropy": 5.6285419940948485,
      "epoch": 6.467844382691545,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017709533298485597,
      "loss": 4.8106,
      "mean_token_accuracy": 0.20649287849664688,
      "num_tokens": 142782041.0,
      "step": 65170
    },
    {
      "entropy": 5.687617015838623,
      "epoch": 6.468340611353712,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017706343285819748,
      "loss": 4.9064,
      "mean_token_accuracy": 0.19530563801527023,
      "num_tokens": 142792204.0,
      "step": 65175
    },
    {
      "entropy": 5.737946319580078,
      "epoch": 6.4688368400158796,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017703153516015484,
      "loss": 4.9533,
      "mean_token_accuracy": 0.1918202206492424,
      "num_tokens": 142802835.0,
      "step": 65180
    },
    {
      "entropy": 5.742277145385742,
      "epoch": 6.469333068678047,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017699963989151906,
      "loss": 4.9423,
      "mean_token_accuracy": 0.19377157390117644,
      "num_tokens": 142813728.0,
      "step": 65185
    },
    {
      "entropy": 5.684210586547851,
      "epoch": 6.4698292973402145,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017696774705308111,
      "loss": 4.886,
      "mean_token_accuracy": 0.20794615894556046,
      "num_tokens": 142824163.0,
      "step": 65190
    },
    {
      "entropy": 5.698226356506348,
      "epoch": 6.470325526002382,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001769358566456318,
      "loss": 4.9729,
      "mean_token_accuracy": 0.19105523377656936,
      "num_tokens": 142835172.0,
      "step": 65195
    },
    {
      "entropy": 5.725917816162109,
      "epoch": 6.4708217546645495,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017690396866996188,
      "loss": 4.9439,
      "mean_token_accuracy": 0.1991724357008934,
      "num_tokens": 142846349.0,
      "step": 65200
    },
    {
      "entropy": 5.670822286605835,
      "epoch": 6.471317983326717,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0001768720831268621,
      "loss": 4.8934,
      "mean_token_accuracy": 0.20826905816793442,
      "num_tokens": 142856478.0,
      "step": 65205
    },
    {
      "entropy": 5.690026664733887,
      "epoch": 6.471814211988884,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001768402000171232,
      "loss": 4.9155,
      "mean_token_accuracy": 0.19747214615345002,
      "num_tokens": 142867115.0,
      "step": 65210
    },
    {
      "entropy": 5.719462299346924,
      "epoch": 6.472310440651052,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001768083193415359,
      "loss": 4.8917,
      "mean_token_accuracy": 0.1999397397041321,
      "num_tokens": 142878223.0,
      "step": 65215
    },
    {
      "entropy": 5.6455662727355955,
      "epoch": 6.472806669313219,
      "grad_norm": 1.1875,
      "learning_rate": 0.00017677644110089075,
      "loss": 4.8675,
      "mean_token_accuracy": 0.19877422004938125,
      "num_tokens": 142891218.0,
      "step": 65220
    },
    {
      "entropy": 5.689729118347168,
      "epoch": 6.473302897975387,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017674456529597816,
      "loss": 4.8571,
      "mean_token_accuracy": 0.20413469523191452,
      "num_tokens": 142902557.0,
      "step": 65225
    },
    {
      "entropy": 5.661953163146973,
      "epoch": 6.473799126637554,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017671269192758853,
      "loss": 4.8722,
      "mean_token_accuracy": 0.20938224494457244,
      "num_tokens": 142913352.0,
      "step": 65230
    },
    {
      "entropy": 5.681027603149414,
      "epoch": 6.474295355299722,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017668082099651244,
      "loss": 4.869,
      "mean_token_accuracy": 0.1973078280687332,
      "num_tokens": 142924546.0,
      "step": 65235
    },
    {
      "entropy": 5.67193922996521,
      "epoch": 6.474791583961889,
      "grad_norm": 1.375,
      "learning_rate": 0.0001766489525035402,
      "loss": 4.8331,
      "mean_token_accuracy": 0.20683986097574233,
      "num_tokens": 142935395.0,
      "step": 65240
    },
    {
      "entropy": 5.6700139999389645,
      "epoch": 6.475287812624057,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017661708644946177,
      "loss": 4.9053,
      "mean_token_accuracy": 0.19814651906490327,
      "num_tokens": 142946598.0,
      "step": 65245
    },
    {
      "entropy": 5.694777965545654,
      "epoch": 6.475784041286225,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017658522283506783,
      "loss": 4.8529,
      "mean_token_accuracy": 0.2029457226395607,
      "num_tokens": 142957784.0,
      "step": 65250
    },
    {
      "entropy": 5.752114343643188,
      "epoch": 6.476280269948393,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017655336166114827,
      "loss": 4.9895,
      "mean_token_accuracy": 0.20414441525936128,
      "num_tokens": 142969267.0,
      "step": 65255
    },
    {
      "entropy": 5.738491344451904,
      "epoch": 6.47677649861056,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001765215029284931,
      "loss": 4.9951,
      "mean_token_accuracy": 0.19799454808235167,
      "num_tokens": 142980044.0,
      "step": 65260
    },
    {
      "entropy": 5.800307846069336,
      "epoch": 6.4772727272727275,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001764896466378926,
      "loss": 5.0018,
      "mean_token_accuracy": 0.18894015699625016,
      "num_tokens": 142990962.0,
      "step": 65265
    },
    {
      "entropy": 5.752678298950196,
      "epoch": 6.477768955934895,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017645779279013647,
      "loss": 4.9334,
      "mean_token_accuracy": 0.19206928610801696,
      "num_tokens": 143001381.0,
      "step": 65270
    },
    {
      "entropy": 5.759385585784912,
      "epoch": 6.4782651845970625,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017642594138601476,
      "loss": 4.9245,
      "mean_token_accuracy": 0.1986774370074272,
      "num_tokens": 143011961.0,
      "step": 65275
    },
    {
      "entropy": 5.6927131652832035,
      "epoch": 6.47876141325923,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017639409242631732,
      "loss": 4.9446,
      "mean_token_accuracy": 0.19775439202785491,
      "num_tokens": 143022482.0,
      "step": 65280
    },
    {
      "entropy": 5.660953044891357,
      "epoch": 6.479257641921397,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017636224591183392,
      "loss": 4.8964,
      "mean_token_accuracy": 0.20120954811573027,
      "num_tokens": 143034405.0,
      "step": 65285
    },
    {
      "entropy": 5.744264221191406,
      "epoch": 6.479753870583565,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017633040184335426,
      "loss": 5.0226,
      "mean_token_accuracy": 0.19606141448020936,
      "num_tokens": 143044078.0,
      "step": 65290
    },
    {
      "entropy": 5.760510540008545,
      "epoch": 6.480250099245732,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017629856022166802,
      "loss": 4.9807,
      "mean_token_accuracy": 0.19797332882881163,
      "num_tokens": 143055740.0,
      "step": 65295
    },
    {
      "entropy": 5.706749439239502,
      "epoch": 6.4807463279079,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001762667210475648,
      "loss": 4.9073,
      "mean_token_accuracy": 0.1983901694417,
      "num_tokens": 143065760.0,
      "step": 65300
    },
    {
      "entropy": 5.749209690093994,
      "epoch": 6.481242556570067,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001762348843218341,
      "loss": 4.9677,
      "mean_token_accuracy": 0.19542246907949448,
      "num_tokens": 143076480.0,
      "step": 65305
    },
    {
      "entropy": 5.706157970428467,
      "epoch": 6.481738785232235,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017620305004526553,
      "loss": 4.8552,
      "mean_token_accuracy": 0.20234586149454117,
      "num_tokens": 143086549.0,
      "step": 65310
    },
    {
      "entropy": 5.661033058166504,
      "epoch": 6.482235013894402,
      "grad_norm": 1.375,
      "learning_rate": 0.00017617121821864845,
      "loss": 4.9271,
      "mean_token_accuracy": 0.19881673008203507,
      "num_tokens": 143097486.0,
      "step": 65315
    },
    {
      "entropy": 5.640725231170654,
      "epoch": 6.48273124255657,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001761393888427721,
      "loss": 4.8599,
      "mean_token_accuracy": 0.21207499206066133,
      "num_tokens": 143108996.0,
      "step": 65320
    },
    {
      "entropy": 5.670123815536499,
      "epoch": 6.483227471218737,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017610756191842596,
      "loss": 4.829,
      "mean_token_accuracy": 0.21136242896318436,
      "num_tokens": 143121287.0,
      "step": 65325
    },
    {
      "entropy": 5.727341222763061,
      "epoch": 6.483723699880905,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001760757374463992,
      "loss": 4.9997,
      "mean_token_accuracy": 0.19671145230531692,
      "num_tokens": 143132998.0,
      "step": 65330
    },
    {
      "entropy": 5.798129558563232,
      "epoch": 6.484219928543073,
      "grad_norm": 1.21875,
      "learning_rate": 0.000176043915427481,
      "loss": 4.9658,
      "mean_token_accuracy": 0.1998977169394493,
      "num_tokens": 143145952.0,
      "step": 65335
    },
    {
      "entropy": 5.678918790817261,
      "epoch": 6.4847161572052405,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001760120958624603,
      "loss": 4.9046,
      "mean_token_accuracy": 0.20312696993350982,
      "num_tokens": 143156925.0,
      "step": 65340
    },
    {
      "entropy": 5.692148113250733,
      "epoch": 6.485212385867408,
      "grad_norm": 1.546875,
      "learning_rate": 0.00017598027875212653,
      "loss": 4.9002,
      "mean_token_accuracy": 0.2028700143098831,
      "num_tokens": 143168327.0,
      "step": 65345
    },
    {
      "entropy": 5.69256477355957,
      "epoch": 6.4857086145295755,
      "grad_norm": 1.65625,
      "learning_rate": 0.00017594846409726828,
      "loss": 4.8716,
      "mean_token_accuracy": 0.2047620251774788,
      "num_tokens": 143180349.0,
      "step": 65350
    },
    {
      "entropy": 5.670135641098023,
      "epoch": 6.486204843191743,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017591665189867483,
      "loss": 4.8983,
      "mean_token_accuracy": 0.20135068148374557,
      "num_tokens": 143191773.0,
      "step": 65355
    },
    {
      "entropy": 5.662018918991089,
      "epoch": 6.48670107185391,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001758848421571349,
      "loss": 4.8553,
      "mean_token_accuracy": 0.19753126055002213,
      "num_tokens": 143203228.0,
      "step": 65360
    },
    {
      "entropy": 5.699320220947266,
      "epoch": 6.487197300516078,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001758530348734373,
      "loss": 4.9048,
      "mean_token_accuracy": 0.2050596535205841,
      "num_tokens": 143215308.0,
      "step": 65365
    },
    {
      "entropy": 5.65110502243042,
      "epoch": 6.487693529178245,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017582123004837076,
      "loss": 4.803,
      "mean_token_accuracy": 0.2134815752506256,
      "num_tokens": 143226170.0,
      "step": 65370
    },
    {
      "entropy": 5.678685140609741,
      "epoch": 6.488189757840413,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00017578942768272404,
      "loss": 4.9497,
      "mean_token_accuracy": 0.2010998398065567,
      "num_tokens": 143237615.0,
      "step": 65375
    },
    {
      "entropy": 5.686753654479981,
      "epoch": 6.48868598650258,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017575762777728566,
      "loss": 4.8752,
      "mean_token_accuracy": 0.19824335277080535,
      "num_tokens": 143247332.0,
      "step": 65380
    },
    {
      "entropy": 5.728474044799805,
      "epoch": 6.489182215164748,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017572583033284429,
      "loss": 4.9569,
      "mean_token_accuracy": 0.1918475806713104,
      "num_tokens": 143259035.0,
      "step": 65385
    },
    {
      "entropy": 5.745803594589233,
      "epoch": 6.489678443826915,
      "grad_norm": 1.25,
      "learning_rate": 0.00017569403535018852,
      "loss": 4.98,
      "mean_token_accuracy": 0.18894582092761994,
      "num_tokens": 143270596.0,
      "step": 65390
    },
    {
      "entropy": 5.718194675445557,
      "epoch": 6.490174672489083,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017566224283010657,
      "loss": 4.9421,
      "mean_token_accuracy": 0.20159298032522202,
      "num_tokens": 143281506.0,
      "step": 65395
    },
    {
      "entropy": 5.67809100151062,
      "epoch": 6.49067090115125,
      "grad_norm": 1.203125,
      "learning_rate": 0.000175630452773387,
      "loss": 4.8703,
      "mean_token_accuracy": 0.2078109636902809,
      "num_tokens": 143292554.0,
      "step": 65400
    },
    {
      "entropy": 5.674228811264038,
      "epoch": 6.491167129813418,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017559866518081801,
      "loss": 4.925,
      "mean_token_accuracy": 0.19961044639348985,
      "num_tokens": 143303015.0,
      "step": 65405
    },
    {
      "entropy": 5.690738296508789,
      "epoch": 6.491663358475585,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017556688005318795,
      "loss": 4.8568,
      "mean_token_accuracy": 0.1977105125784874,
      "num_tokens": 143312561.0,
      "step": 65410
    },
    {
      "entropy": 5.736988210678101,
      "epoch": 6.492159587137753,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017553509739128502,
      "loss": 4.9519,
      "mean_token_accuracy": 0.20400580912828445,
      "num_tokens": 143324003.0,
      "step": 65415
    },
    {
      "entropy": 5.6925163745880125,
      "epoch": 6.492655815799921,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001755033171958974,
      "loss": 4.8953,
      "mean_token_accuracy": 0.2029072031378746,
      "num_tokens": 143335881.0,
      "step": 65420
    },
    {
      "entropy": 5.679710483551025,
      "epoch": 6.4931520444620885,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017547153946781307,
      "loss": 4.9169,
      "mean_token_accuracy": 0.2100569099187851,
      "num_tokens": 143348271.0,
      "step": 65425
    },
    {
      "entropy": 5.725186443328857,
      "epoch": 6.493648273124256,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017543976420782015,
      "loss": 4.8792,
      "mean_token_accuracy": 0.20509220212697982,
      "num_tokens": 143357375.0,
      "step": 65430
    },
    {
      "entropy": 5.670387125015258,
      "epoch": 6.494144501786423,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001754079914167065,
      "loss": 4.9329,
      "mean_token_accuracy": 0.20465234220027922,
      "num_tokens": 143367436.0,
      "step": 65435
    },
    {
      "entropy": 5.683390808105469,
      "epoch": 6.494640730448591,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017537622109525998,
      "loss": 4.842,
      "mean_token_accuracy": 0.2075305014848709,
      "num_tokens": 143378129.0,
      "step": 65440
    },
    {
      "entropy": 5.6416261196136475,
      "epoch": 6.495136959110758,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017534445324426855,
      "loss": 4.8359,
      "mean_token_accuracy": 0.2060224860906601,
      "num_tokens": 143389632.0,
      "step": 65445
    },
    {
      "entropy": 5.778876113891601,
      "epoch": 6.495633187772926,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017531268786452004,
      "loss": 5.0315,
      "mean_token_accuracy": 0.1879217118024826,
      "num_tokens": 143400268.0,
      "step": 65450
    },
    {
      "entropy": 5.7258140563964846,
      "epoch": 6.496129416435093,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00017528092495680203,
      "loss": 4.959,
      "mean_token_accuracy": 0.19945481568574905,
      "num_tokens": 143411754.0,
      "step": 65455
    },
    {
      "entropy": 5.728122043609619,
      "epoch": 6.496625645097261,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0001752491645219022,
      "loss": 4.9397,
      "mean_token_accuracy": 0.19683486819267274,
      "num_tokens": 143421337.0,
      "step": 65460
    },
    {
      "entropy": 5.655675458908081,
      "epoch": 6.497121873759428,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001752174065606082,
      "loss": 4.9142,
      "mean_token_accuracy": 0.1959928721189499,
      "num_tokens": 143432985.0,
      "step": 65465
    },
    {
      "entropy": 5.613988065719605,
      "epoch": 6.497618102421596,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00017518565107370733,
      "loss": 4.788,
      "mean_token_accuracy": 0.20676187574863433,
      "num_tokens": 143442728.0,
      "step": 65470
    },
    {
      "entropy": 5.7194681644439695,
      "epoch": 6.498114331083763,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017515389806198724,
      "loss": 4.9286,
      "mean_token_accuracy": 0.19517738819122316,
      "num_tokens": 143453741.0,
      "step": 65475
    },
    {
      "entropy": 5.784008979797363,
      "epoch": 6.498610559745931,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017512214752623544,
      "loss": 4.923,
      "mean_token_accuracy": 0.20077454298734665,
      "num_tokens": 143465031.0,
      "step": 65480
    },
    {
      "entropy": 5.713120985031128,
      "epoch": 6.499106788408098,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017509039946723921,
      "loss": 4.9591,
      "mean_token_accuracy": 0.2019290432333946,
      "num_tokens": 143476120.0,
      "step": 65485
    },
    {
      "entropy": 5.697488689422608,
      "epoch": 6.499603017070266,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017505865388578573,
      "loss": 4.9085,
      "mean_token_accuracy": 0.20133262276649475,
      "num_tokens": 143487506.0,
      "step": 65490
    },
    {
      "entropy": 5.734708261489868,
      "epoch": 6.500099245732433,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001750269107826622,
      "loss": 4.949,
      "mean_token_accuracy": 0.19359723776578902,
      "num_tokens": 143499185.0,
      "step": 65495
    },
    {
      "entropy": 5.7191774368286135,
      "epoch": 6.500595474394601,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00017499517015865597,
      "loss": 4.9505,
      "mean_token_accuracy": 0.19419045001268387,
      "num_tokens": 143511037.0,
      "step": 65500
    },
    {
      "entropy": 5.762527751922607,
      "epoch": 6.501091703056769,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001749634320145539,
      "loss": 4.9446,
      "mean_token_accuracy": 0.20546620786190034,
      "num_tokens": 143522644.0,
      "step": 65505
    },
    {
      "entropy": 5.7401895999908445,
      "epoch": 6.5015879317189365,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00017493169635114327,
      "loss": 4.9239,
      "mean_token_accuracy": 0.1969723805785179,
      "num_tokens": 143533474.0,
      "step": 65510
    },
    {
      "entropy": 5.662156963348389,
      "epoch": 6.502084160381104,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017489996316921098,
      "loss": 4.9348,
      "mean_token_accuracy": 0.19550943225622178,
      "num_tokens": 143543964.0,
      "step": 65515
    },
    {
      "entropy": 5.7464323997497555,
      "epoch": 6.502580389043271,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017486823246954387,
      "loss": 5.0152,
      "mean_token_accuracy": 0.19348429888486862,
      "num_tokens": 143555377.0,
      "step": 65520
    },
    {
      "entropy": 5.715032958984375,
      "epoch": 6.503076617705439,
      "grad_norm": 1.5,
      "learning_rate": 0.00017483650425292874,
      "loss": 4.9256,
      "mean_token_accuracy": 0.2017512947320938,
      "num_tokens": 143565934.0,
      "step": 65525
    },
    {
      "entropy": 5.601769876480103,
      "epoch": 6.503572846367606,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017480477852015254,
      "loss": 4.8382,
      "mean_token_accuracy": 0.20631899684667587,
      "num_tokens": 143577571.0,
      "step": 65530
    },
    {
      "entropy": 5.688141202926635,
      "epoch": 6.504069075029774,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00017477305527200185,
      "loss": 4.8782,
      "mean_token_accuracy": 0.19873571246862412,
      "num_tokens": 143586678.0,
      "step": 65535
    },
    {
      "entropy": 5.716694974899292,
      "epoch": 6.504565303691941,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017474133450926354,
      "loss": 4.9387,
      "mean_token_accuracy": 0.20272014886140824,
      "num_tokens": 143597139.0,
      "step": 65540
    },
    {
      "entropy": 5.6867616176605225,
      "epoch": 6.505061532354109,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017470961623272408,
      "loss": 4.8847,
      "mean_token_accuracy": 0.2022603929042816,
      "num_tokens": 143607296.0,
      "step": 65545
    },
    {
      "entropy": 5.655364465713501,
      "epoch": 6.505557761016276,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017467790044317004,
      "loss": 4.8269,
      "mean_token_accuracy": 0.20883614718914031,
      "num_tokens": 143618283.0,
      "step": 65550
    },
    {
      "entropy": 5.655049276351929,
      "epoch": 6.506053989678444,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017464618714138776,
      "loss": 4.8567,
      "mean_token_accuracy": 0.206848706305027,
      "num_tokens": 143630056.0,
      "step": 65555
    },
    {
      "entropy": 5.688804292678833,
      "epoch": 6.506550218340611,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017461447632816397,
      "loss": 4.9117,
      "mean_token_accuracy": 0.20407965779304504,
      "num_tokens": 143641801.0,
      "step": 65560
    },
    {
      "entropy": 5.658902359008789,
      "epoch": 6.507046447002779,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017458276800428476,
      "loss": 4.9036,
      "mean_token_accuracy": 0.20793278366327286,
      "num_tokens": 143654013.0,
      "step": 65565
    },
    {
      "entropy": 5.6503016471862795,
      "epoch": 6.507542675664946,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001745510621705365,
      "loss": 4.7976,
      "mean_token_accuracy": 0.21165806353092192,
      "num_tokens": 143664379.0,
      "step": 65570
    },
    {
      "entropy": 5.738427019119262,
      "epoch": 6.508038904327114,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0001745193588277055,
      "loss": 4.9472,
      "mean_token_accuracy": 0.20223244577646254,
      "num_tokens": 143674808.0,
      "step": 65575
    },
    {
      "entropy": 5.722258567810059,
      "epoch": 6.508535132989281,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017448765797657792,
      "loss": 4.988,
      "mean_token_accuracy": 0.20044735074043274,
      "num_tokens": 143684160.0,
      "step": 65580
    },
    {
      "entropy": 5.696242809295654,
      "epoch": 6.509031361651449,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001744559596179397,
      "loss": 4.8809,
      "mean_token_accuracy": 0.20188807249069213,
      "num_tokens": 143695177.0,
      "step": 65585
    },
    {
      "entropy": 5.681606006622315,
      "epoch": 6.509527590313617,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017442426375257716,
      "loss": 4.8867,
      "mean_token_accuracy": 0.20322927087545395,
      "num_tokens": 143705318.0,
      "step": 65590
    },
    {
      "entropy": 5.676342487335205,
      "epoch": 6.510023818975784,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017439257038127614,
      "loss": 4.9108,
      "mean_token_accuracy": 0.20188543647527696,
      "num_tokens": 143716625.0,
      "step": 65595
    },
    {
      "entropy": 5.692744636535645,
      "epoch": 6.510520047637952,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017436087950482252,
      "loss": 4.8603,
      "mean_token_accuracy": 0.20483953952789308,
      "num_tokens": 143726885.0,
      "step": 65600
    },
    {
      "entropy": 5.731878757476807,
      "epoch": 6.511016276300119,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017432919112400232,
      "loss": 4.902,
      "mean_token_accuracy": 0.20074465721845627,
      "num_tokens": 143737678.0,
      "step": 65605
    },
    {
      "entropy": 5.7311302661895756,
      "epoch": 6.511512504962287,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001742975052396012,
      "loss": 4.8957,
      "mean_token_accuracy": 0.2036069005727768,
      "num_tokens": 143747557.0,
      "step": 65610
    },
    {
      "entropy": 5.663173866271973,
      "epoch": 6.512008733624454,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00017426582185240492,
      "loss": 4.8744,
      "mean_token_accuracy": 0.19664829671382905,
      "num_tokens": 143759678.0,
      "step": 65615
    },
    {
      "entropy": 5.71770920753479,
      "epoch": 6.512504962286622,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017423414096319926,
      "loss": 4.9425,
      "mean_token_accuracy": 0.19736763834953308,
      "num_tokens": 143770652.0,
      "step": 65620
    },
    {
      "entropy": 5.724849510192871,
      "epoch": 6.513001190948789,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017420246257276978,
      "loss": 4.9049,
      "mean_token_accuracy": 0.20624592453241347,
      "num_tokens": 143781746.0,
      "step": 65625
    },
    {
      "entropy": 5.678242492675781,
      "epoch": 6.513497419610957,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017417078668190202,
      "loss": 4.9241,
      "mean_token_accuracy": 0.1969040334224701,
      "num_tokens": 143792336.0,
      "step": 65630
    },
    {
      "entropy": 5.686086368560791,
      "epoch": 6.513993648273124,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017413911329138137,
      "loss": 4.9103,
      "mean_token_accuracy": 0.20096641033887863,
      "num_tokens": 143802533.0,
      "step": 65635
    },
    {
      "entropy": 5.68936014175415,
      "epoch": 6.514489876935292,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001741074424019935,
      "loss": 4.9212,
      "mean_token_accuracy": 0.20225869864225388,
      "num_tokens": 143813992.0,
      "step": 65640
    },
    {
      "entropy": 5.718305778503418,
      "epoch": 6.514986105597459,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017407577401452358,
      "loss": 4.9321,
      "mean_token_accuracy": 0.20067155361175537,
      "num_tokens": 143824499.0,
      "step": 65645
    },
    {
      "entropy": 5.675519323348999,
      "epoch": 6.515482334259627,
      "grad_norm": 1.1875,
      "learning_rate": 0.00017404410812975704,
      "loss": 4.8974,
      "mean_token_accuracy": 0.2032473385334015,
      "num_tokens": 143836149.0,
      "step": 65650
    },
    {
      "entropy": 5.6486317157745365,
      "epoch": 6.515978562921794,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001740124447484791,
      "loss": 4.8718,
      "mean_token_accuracy": 0.20546513050794601,
      "num_tokens": 143846152.0,
      "step": 65655
    },
    {
      "entropy": 5.654922294616699,
      "epoch": 6.516474791583962,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017398078387147493,
      "loss": 4.8828,
      "mean_token_accuracy": 0.20098646432161332,
      "num_tokens": 143856708.0,
      "step": 65660
    },
    {
      "entropy": 5.702030944824219,
      "epoch": 6.516971020246129,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00017394912549952954,
      "loss": 4.9438,
      "mean_token_accuracy": 0.20161276012659074,
      "num_tokens": 143868355.0,
      "step": 65665
    },
    {
      "entropy": 5.7337593078613285,
      "epoch": 6.5174672489082965,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001739174696334282,
      "loss": 4.9808,
      "mean_token_accuracy": 0.19756002277135848,
      "num_tokens": 143880784.0,
      "step": 65670
    },
    {
      "entropy": 5.664275789260865,
      "epoch": 6.517963477570465,
      "grad_norm": 1.375,
      "learning_rate": 0.00017388581627395566,
      "loss": 4.8584,
      "mean_token_accuracy": 0.20436666011810303,
      "num_tokens": 143891542.0,
      "step": 65675
    },
    {
      "entropy": 5.7025110721588135,
      "epoch": 6.518459706232632,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017385416542189713,
      "loss": 4.9156,
      "mean_token_accuracy": 0.2035646066069603,
      "num_tokens": 143901515.0,
      "step": 65680
    },
    {
      "entropy": 5.660878038406372,
      "epoch": 6.5189559348948,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001738225170780373,
      "loss": 4.8655,
      "mean_token_accuracy": 0.19908531606197358,
      "num_tokens": 143913444.0,
      "step": 65685
    },
    {
      "entropy": 5.774013328552246,
      "epoch": 6.519452163556967,
      "grad_norm": 1.546875,
      "learning_rate": 0.00017379087124316109,
      "loss": 4.9128,
      "mean_token_accuracy": 0.20317060351371766,
      "num_tokens": 143925084.0,
      "step": 65690
    },
    {
      "entropy": 5.691645526885987,
      "epoch": 6.519948392219135,
      "grad_norm": 1.65625,
      "learning_rate": 0.00017375922791805317,
      "loss": 4.8462,
      "mean_token_accuracy": 0.2138863444328308,
      "num_tokens": 143934659.0,
      "step": 65695
    },
    {
      "entropy": 5.813983488082886,
      "epoch": 6.520444620881302,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017372758710349813,
      "loss": 5.058,
      "mean_token_accuracy": 0.19174070805311202,
      "num_tokens": 143945591.0,
      "step": 65700
    },
    {
      "entropy": 5.693696308135986,
      "epoch": 6.52094084954347,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017369594880028074,
      "loss": 4.9235,
      "mean_token_accuracy": 0.1961329847574234,
      "num_tokens": 143955795.0,
      "step": 65705
    },
    {
      "entropy": 5.725585508346557,
      "epoch": 6.521437078205637,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017366431300918562,
      "loss": 4.9019,
      "mean_token_accuracy": 0.19922980964183806,
      "num_tokens": 143967563.0,
      "step": 65710
    },
    {
      "entropy": 5.638078117370606,
      "epoch": 6.521933306867805,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017363267973099714,
      "loss": 4.8662,
      "mean_token_accuracy": 0.20235104560852052,
      "num_tokens": 143979098.0,
      "step": 65715
    },
    {
      "entropy": 5.78922848701477,
      "epoch": 6.522429535529972,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001736010489664998,
      "loss": 5.0213,
      "mean_token_accuracy": 0.19647902250289917,
      "num_tokens": 143990210.0,
      "step": 65720
    },
    {
      "entropy": 5.745431184768677,
      "epoch": 6.52292576419214,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00017356942071647795,
      "loss": 4.9313,
      "mean_token_accuracy": 0.20280096679925919,
      "num_tokens": 144001881.0,
      "step": 65725
    },
    {
      "entropy": 5.743996667861938,
      "epoch": 6.523421992854307,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001735377949817158,
      "loss": 4.9742,
      "mean_token_accuracy": 0.19743096828460693,
      "num_tokens": 144012856.0,
      "step": 65730
    },
    {
      "entropy": 5.660165643692016,
      "epoch": 6.523918221516475,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001735061717629977,
      "loss": 4.7956,
      "mean_token_accuracy": 0.22060542404651642,
      "num_tokens": 144023960.0,
      "step": 65735
    },
    {
      "entropy": 5.756851291656494,
      "epoch": 6.524414450178642,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017347455106110797,
      "loss": 4.9562,
      "mean_token_accuracy": 0.1980862319469452,
      "num_tokens": 144035152.0,
      "step": 65740
    },
    {
      "entropy": 5.683874225616455,
      "epoch": 6.5249106788408096,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017344293287683055,
      "loss": 4.9477,
      "mean_token_accuracy": 0.2016838550567627,
      "num_tokens": 144045096.0,
      "step": 65745
    },
    {
      "entropy": 5.705682611465454,
      "epoch": 6.525406907502977,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001734113172109496,
      "loss": 4.8916,
      "mean_token_accuracy": 0.20464368015527726,
      "num_tokens": 144056722.0,
      "step": 65750
    },
    {
      "entropy": 5.676158285140991,
      "epoch": 6.5259031361651445,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017337970406424902,
      "loss": 4.8891,
      "mean_token_accuracy": 0.20391820818185807,
      "num_tokens": 144067790.0,
      "step": 65755
    },
    {
      "entropy": 5.750707960128784,
      "epoch": 6.526399364827313,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017334809343751278,
      "loss": 4.9355,
      "mean_token_accuracy": 0.1955208346247673,
      "num_tokens": 144078521.0,
      "step": 65760
    },
    {
      "entropy": 5.717554092407227,
      "epoch": 6.52689559348948,
      "grad_norm": 1.25,
      "learning_rate": 0.00017331648533152483,
      "loss": 4.9098,
      "mean_token_accuracy": 0.2036807805299759,
      "num_tokens": 144090219.0,
      "step": 65765
    },
    {
      "entropy": 5.783486843109131,
      "epoch": 6.527391822151648,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017328487974706893,
      "loss": 5.0209,
      "mean_token_accuracy": 0.19341582357883452,
      "num_tokens": 144100514.0,
      "step": 65770
    },
    {
      "entropy": 5.7143199920654295,
      "epoch": 6.527888050813815,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017325327668492882,
      "loss": 4.9302,
      "mean_token_accuracy": 0.1999264642596245,
      "num_tokens": 144111566.0,
      "step": 65775
    },
    {
      "entropy": 5.717830562591553,
      "epoch": 6.528384279475983,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017322167614588823,
      "loss": 4.9105,
      "mean_token_accuracy": 0.19976987391710282,
      "num_tokens": 144122092.0,
      "step": 65780
    },
    {
      "entropy": 5.696316242218018,
      "epoch": 6.52888050813815,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001731900781307308,
      "loss": 4.8744,
      "mean_token_accuracy": 0.1993413046002388,
      "num_tokens": 144134335.0,
      "step": 65785
    },
    {
      "entropy": 5.746307134628296,
      "epoch": 6.529376736800318,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017315848264023992,
      "loss": 5.0007,
      "mean_token_accuracy": 0.19605872482061387,
      "num_tokens": 144145811.0,
      "step": 65790
    },
    {
      "entropy": 5.646976613998413,
      "epoch": 6.529872965462485,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017312688967519935,
      "loss": 4.7842,
      "mean_token_accuracy": 0.20753265917301178,
      "num_tokens": 144156699.0,
      "step": 65795
    },
    {
      "entropy": 5.63734188079834,
      "epoch": 6.530369194124653,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017309529923639227,
      "loss": 4.8858,
      "mean_token_accuracy": 0.2041487827897072,
      "num_tokens": 144166716.0,
      "step": 65800
    },
    {
      "entropy": 5.655330944061279,
      "epoch": 6.53086542278682,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001730637113246023,
      "loss": 4.8727,
      "mean_token_accuracy": 0.2085216000676155,
      "num_tokens": 144177762.0,
      "step": 65805
    },
    {
      "entropy": 5.652601861953736,
      "epoch": 6.531361651448988,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017303212594061267,
      "loss": 4.8669,
      "mean_token_accuracy": 0.20610871613025666,
      "num_tokens": 144189082.0,
      "step": 65810
    },
    {
      "entropy": 5.756561183929444,
      "epoch": 6.531857880111155,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017300054308520653,
      "loss": 4.8897,
      "mean_token_accuracy": 0.20403087884187698,
      "num_tokens": 144198854.0,
      "step": 65815
    },
    {
      "entropy": 5.72381100654602,
      "epoch": 6.532354108773323,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017296896275916708,
      "loss": 4.9745,
      "mean_token_accuracy": 0.1939168319106102,
      "num_tokens": 144208909.0,
      "step": 65820
    },
    {
      "entropy": 5.68629469871521,
      "epoch": 6.53285033743549,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00017293738496327755,
      "loss": 4.8675,
      "mean_token_accuracy": 0.20375407189130784,
      "num_tokens": 144219527.0,
      "step": 65825
    },
    {
      "entropy": 5.695686769485474,
      "epoch": 6.5333465660976575,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017290580969832087,
      "loss": 4.9356,
      "mean_token_accuracy": 0.20853277146816254,
      "num_tokens": 144230634.0,
      "step": 65830
    },
    {
      "entropy": 5.68896517753601,
      "epoch": 6.533842794759825,
      "grad_norm": 1.375,
      "learning_rate": 0.00017287423696508023,
      "loss": 4.8215,
      "mean_token_accuracy": 0.2152531310915947,
      "num_tokens": 144243179.0,
      "step": 65835
    },
    {
      "entropy": 5.725126934051514,
      "epoch": 6.5343390234219925,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001728426667643384,
      "loss": 4.9136,
      "mean_token_accuracy": 0.19885604083538055,
      "num_tokens": 144253832.0,
      "step": 65840
    },
    {
      "entropy": 5.686166381835937,
      "epoch": 6.534835252084161,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017281109909687836,
      "loss": 4.8439,
      "mean_token_accuracy": 0.20940425097942353,
      "num_tokens": 144264261.0,
      "step": 65845
    },
    {
      "entropy": 5.690149450302124,
      "epoch": 6.535331480746327,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017277953396348274,
      "loss": 4.9258,
      "mean_token_accuracy": 0.19715060889720917,
      "num_tokens": 144275941.0,
      "step": 65850
    },
    {
      "entropy": 5.7195580959320065,
      "epoch": 6.535827709408496,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00017274797136493454,
      "loss": 4.9616,
      "mean_token_accuracy": 0.1966053754091263,
      "num_tokens": 144286843.0,
      "step": 65855
    },
    {
      "entropy": 5.664500761032104,
      "epoch": 6.536323938070663,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001727164113020163,
      "loss": 4.8614,
      "mean_token_accuracy": 0.20658601075410843,
      "num_tokens": 144297128.0,
      "step": 65860
    },
    {
      "entropy": 5.692560291290283,
      "epoch": 6.536820166732831,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017268485377551056,
      "loss": 4.9461,
      "mean_token_accuracy": 0.1990199416875839,
      "num_tokens": 144308991.0,
      "step": 65865
    },
    {
      "entropy": 5.689681768417358,
      "epoch": 6.537316395394998,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017265329878620006,
      "loss": 4.8717,
      "mean_token_accuracy": 0.20351507812738417,
      "num_tokens": 144319143.0,
      "step": 65870
    },
    {
      "entropy": 5.689689826965332,
      "epoch": 6.537812624057166,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017262174633486723,
      "loss": 4.8878,
      "mean_token_accuracy": 0.20395260453224182,
      "num_tokens": 144329711.0,
      "step": 65875
    },
    {
      "entropy": 5.689277791976929,
      "epoch": 6.538308852719333,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001725901964222944,
      "loss": 4.9178,
      "mean_token_accuracy": 0.2026051864027977,
      "num_tokens": 144340656.0,
      "step": 65880
    },
    {
      "entropy": 5.687834787368774,
      "epoch": 6.538805081381501,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001725586490492641,
      "loss": 4.8837,
      "mean_token_accuracy": 0.20640761852264405,
      "num_tokens": 144351360.0,
      "step": 65885
    },
    {
      "entropy": 5.7099748134613035,
      "epoch": 6.539301310043668,
      "grad_norm": 1.59375,
      "learning_rate": 0.00017252710421655854,
      "loss": 4.9523,
      "mean_token_accuracy": 0.2035541981458664,
      "num_tokens": 144361579.0,
      "step": 65890
    },
    {
      "entropy": 5.7056052684783936,
      "epoch": 6.539797538705836,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001724955619249599,
      "loss": 4.8831,
      "mean_token_accuracy": 0.20523497611284255,
      "num_tokens": 144372485.0,
      "step": 65895
    },
    {
      "entropy": 5.717576932907105,
      "epoch": 6.540293767368003,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001724640221752506,
      "loss": 4.909,
      "mean_token_accuracy": 0.2006325751543045,
      "num_tokens": 144383536.0,
      "step": 65900
    },
    {
      "entropy": 5.725314569473267,
      "epoch": 6.5407899960301705,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017243248496821257,
      "loss": 4.9234,
      "mean_token_accuracy": 0.20409359335899352,
      "num_tokens": 144394099.0,
      "step": 65905
    },
    {
      "entropy": 5.669673681259155,
      "epoch": 6.541286224692338,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017240095030462776,
      "loss": 4.8091,
      "mean_token_accuracy": 0.20373433232307434,
      "num_tokens": 144404915.0,
      "step": 65910
    },
    {
      "entropy": 5.7041410446167,
      "epoch": 6.5417824533545055,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017236941818527847,
      "loss": 4.8983,
      "mean_token_accuracy": 0.2023361176252365,
      "num_tokens": 144416018.0,
      "step": 65915
    },
    {
      "entropy": 5.66958909034729,
      "epoch": 6.542278682016673,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001723378886109464,
      "loss": 4.9287,
      "mean_token_accuracy": 0.19730671495199203,
      "num_tokens": 144426253.0,
      "step": 65920
    },
    {
      "entropy": 5.681567764282226,
      "epoch": 6.54277491067884,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017230636158241353,
      "loss": 4.8923,
      "mean_token_accuracy": 0.20098339915275573,
      "num_tokens": 144437430.0,
      "step": 65925
    },
    {
      "entropy": 5.711220932006836,
      "epoch": 6.543271139341009,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017227483710046144,
      "loss": 4.9292,
      "mean_token_accuracy": 0.20676267594099046,
      "num_tokens": 144448429.0,
      "step": 65930
    },
    {
      "entropy": 5.68306884765625,
      "epoch": 6.543767368003175,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001722433151658722,
      "loss": 4.8572,
      "mean_token_accuracy": 0.20211687982082366,
      "num_tokens": 144458454.0,
      "step": 65935
    },
    {
      "entropy": 5.707964515686035,
      "epoch": 6.544263596665344,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017221179577942719,
      "loss": 4.9088,
      "mean_token_accuracy": 0.20348822474479675,
      "num_tokens": 144468593.0,
      "step": 65940
    },
    {
      "entropy": 5.775660181045533,
      "epoch": 6.544759825327511,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017218027894190825,
      "loss": 4.9865,
      "mean_token_accuracy": 0.19315119981765747,
      "num_tokens": 144479728.0,
      "step": 65945
    },
    {
      "entropy": 5.720261192321777,
      "epoch": 6.545256053989679,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017214876465409686,
      "loss": 4.8307,
      "mean_token_accuracy": 0.20592222660779952,
      "num_tokens": 144490972.0,
      "step": 65950
    },
    {
      "entropy": 5.713695335388183,
      "epoch": 6.545752282651846,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001721172529167745,
      "loss": 4.9004,
      "mean_token_accuracy": 0.21573924273252487,
      "num_tokens": 144502405.0,
      "step": 65955
    },
    {
      "entropy": 5.637930774688721,
      "epoch": 6.546248511314014,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017208574373072242,
      "loss": 4.899,
      "mean_token_accuracy": 0.20303386002779006,
      "num_tokens": 144513527.0,
      "step": 65960
    },
    {
      "entropy": 5.72286286354065,
      "epoch": 6.546744739976181,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017205423709672215,
      "loss": 4.8753,
      "mean_token_accuracy": 0.20356908291578293,
      "num_tokens": 144524316.0,
      "step": 65965
    },
    {
      "entropy": 5.743804597854615,
      "epoch": 6.547240968638349,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001720227330155551,
      "loss": 4.9629,
      "mean_token_accuracy": 0.196466563642025,
      "num_tokens": 144534287.0,
      "step": 65970
    },
    {
      "entropy": 5.703456783294678,
      "epoch": 6.547737197300516,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017199123148800235,
      "loss": 4.9092,
      "mean_token_accuracy": 0.209930320084095,
      "num_tokens": 144546318.0,
      "step": 65975
    },
    {
      "entropy": 5.694854497909546,
      "epoch": 6.5482334259626835,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001719597325148451,
      "loss": 4.9161,
      "mean_token_accuracy": 0.20321817696094513,
      "num_tokens": 144558580.0,
      "step": 65980
    },
    {
      "entropy": 5.742320919036866,
      "epoch": 6.548729654624851,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017192823609686448,
      "loss": 4.9929,
      "mean_token_accuracy": 0.19851973205804824,
      "num_tokens": 144569789.0,
      "step": 65985
    },
    {
      "entropy": 5.744552087783814,
      "epoch": 6.5492258832870185,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017189674223484137,
      "loss": 4.9502,
      "mean_token_accuracy": 0.20266568958759307,
      "num_tokens": 144580532.0,
      "step": 65990
    },
    {
      "entropy": 5.735367918014527,
      "epoch": 6.549722111949186,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000171865250929557,
      "loss": 4.8662,
      "mean_token_accuracy": 0.20935247987508773,
      "num_tokens": 144591124.0,
      "step": 65995
    },
    {
      "entropy": 5.755359029769897,
      "epoch": 6.550218340611353,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017183376218179205,
      "loss": 4.9609,
      "mean_token_accuracy": 0.19961828142404556,
      "num_tokens": 144603013.0,
      "step": 66000
    },
    {
      "epoch": 6.550218340611353,
      "eval_entropy": 5.441474796276133,
      "eval_loss": 5.066817760467529,
      "eval_mean_token_accuracy": 0.1983317474740139,
      "eval_num_tokens": 144603013.0,
      "eval_runtime": 20.5327,
      "eval_samples_per_second": 1639.778,
      "eval_steps_per_second": 204.991,
      "step": 66000
    },
    {
      "entropy": 5.75438494682312,
      "epoch": 6.550714569273521,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017180227599232758,
      "loss": 4.9844,
      "mean_token_accuracy": 0.1953607529401779,
      "num_tokens": 144614422.0,
      "step": 66005
    },
    {
      "entropy": 5.646512889862061,
      "epoch": 6.551210797935688,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017177079236194436,
      "loss": 4.8369,
      "mean_token_accuracy": 0.2035844147205353,
      "num_tokens": 144625443.0,
      "step": 66010
    },
    {
      "entropy": 5.599533987045288,
      "epoch": 6.551707026597857,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000171739311291423,
      "loss": 4.8447,
      "mean_token_accuracy": 0.2019587978720665,
      "num_tokens": 144636406.0,
      "step": 66015
    },
    {
      "entropy": 5.65777063369751,
      "epoch": 6.552203255260023,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001717078327815442,
      "loss": 4.8427,
      "mean_token_accuracy": 0.20664166808128356,
      "num_tokens": 144645613.0,
      "step": 66020
    },
    {
      "entropy": 5.725731277465821,
      "epoch": 6.552699483922192,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017167635683308852,
      "loss": 4.9883,
      "mean_token_accuracy": 0.1922306537628174,
      "num_tokens": 144657581.0,
      "step": 66025
    },
    {
      "entropy": 5.720325088500976,
      "epoch": 6.553195712584359,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017164488344683653,
      "loss": 4.9041,
      "mean_token_accuracy": 0.19879868477582932,
      "num_tokens": 144668080.0,
      "step": 66030
    },
    {
      "entropy": 5.6734332084655765,
      "epoch": 6.553691941246527,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017161341262356884,
      "loss": 4.8555,
      "mean_token_accuracy": 0.2110210657119751,
      "num_tokens": 144678764.0,
      "step": 66035
    },
    {
      "entropy": 5.640888643264771,
      "epoch": 6.554188169908694,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001715819443640657,
      "loss": 4.8475,
      "mean_token_accuracy": 0.2022179216146469,
      "num_tokens": 144689990.0,
      "step": 66040
    },
    {
      "entropy": 5.750077152252198,
      "epoch": 6.554684398570862,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001715504786691075,
      "loss": 4.964,
      "mean_token_accuracy": 0.19685679972171782,
      "num_tokens": 144701292.0,
      "step": 66045
    },
    {
      "entropy": 5.756901741027832,
      "epoch": 6.555180627233029,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001715190155394745,
      "loss": 4.9775,
      "mean_token_accuracy": 0.19266862869262696,
      "num_tokens": 144713089.0,
      "step": 66050
    },
    {
      "entropy": 5.722842836380005,
      "epoch": 6.5556768558951966,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017148755497594688,
      "loss": 4.9405,
      "mean_token_accuracy": 0.19452158659696578,
      "num_tokens": 144725435.0,
      "step": 66055
    },
    {
      "entropy": 5.683868885040283,
      "epoch": 6.556173084557364,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001714560969793048,
      "loss": 4.8727,
      "mean_token_accuracy": 0.20709591656923293,
      "num_tokens": 144736720.0,
      "step": 66060
    },
    {
      "entropy": 5.726371574401855,
      "epoch": 6.5566693132195315,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017142464155032853,
      "loss": 4.8754,
      "mean_token_accuracy": 0.20344270020723343,
      "num_tokens": 144747896.0,
      "step": 66065
    },
    {
      "entropy": 5.731780576705932,
      "epoch": 6.557165541881699,
      "grad_norm": 1.53125,
      "learning_rate": 0.00017139318868979793,
      "loss": 4.9616,
      "mean_token_accuracy": 0.20052599906921387,
      "num_tokens": 144758374.0,
      "step": 66070
    },
    {
      "entropy": 5.702274322509766,
      "epoch": 6.5576617705438665,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017136173839849307,
      "loss": 4.892,
      "mean_token_accuracy": 0.20095353126525878,
      "num_tokens": 144768342.0,
      "step": 66075
    },
    {
      "entropy": 5.673625469207764,
      "epoch": 6.558157999206034,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017133029067719371,
      "loss": 4.8585,
      "mean_token_accuracy": 0.20478913933038712,
      "num_tokens": 144781331.0,
      "step": 66080
    },
    {
      "entropy": 5.692926931381225,
      "epoch": 6.558654227868201,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001712988455266797,
      "loss": 4.8819,
      "mean_token_accuracy": 0.20427312403917314,
      "num_tokens": 144792098.0,
      "step": 66085
    },
    {
      "entropy": 5.715472936630249,
      "epoch": 6.559150456530369,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001712674029477309,
      "loss": 4.9575,
      "mean_token_accuracy": 0.20276398807764054,
      "num_tokens": 144803207.0,
      "step": 66090
    },
    {
      "entropy": 5.675462961196899,
      "epoch": 6.559646685192536,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017123596294112692,
      "loss": 4.806,
      "mean_token_accuracy": 0.20830383747816086,
      "num_tokens": 144813216.0,
      "step": 66095
    },
    {
      "entropy": 5.683100700378418,
      "epoch": 6.560142913854705,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001712045255076476,
      "loss": 4.8719,
      "mean_token_accuracy": 0.20459727495908736,
      "num_tokens": 144824870.0,
      "step": 66100
    },
    {
      "entropy": 5.755134248733521,
      "epoch": 6.560639142516871,
      "grad_norm": 1.140625,
      "learning_rate": 0.00017117309064807235,
      "loss": 5.0095,
      "mean_token_accuracy": 0.20042819827795028,
      "num_tokens": 144836586.0,
      "step": 66105
    },
    {
      "entropy": 5.763587808609008,
      "epoch": 6.56113537117904,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017114165836318074,
      "loss": 4.9577,
      "mean_token_accuracy": 0.19101309776306152,
      "num_tokens": 144846568.0,
      "step": 66110
    },
    {
      "entropy": 5.7541313648223875,
      "epoch": 6.561631599841207,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017111022865375208,
      "loss": 4.9588,
      "mean_token_accuracy": 0.19221652597188948,
      "num_tokens": 144857079.0,
      "step": 66115
    },
    {
      "entropy": 5.705224561691284,
      "epoch": 6.562127828503375,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000171078801520566,
      "loss": 4.8649,
      "mean_token_accuracy": 0.20074879974126816,
      "num_tokens": 144868389.0,
      "step": 66120
    },
    {
      "entropy": 5.676906681060791,
      "epoch": 6.562624057165542,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017104737696440163,
      "loss": 4.8496,
      "mean_token_accuracy": 0.20277070999145508,
      "num_tokens": 144881171.0,
      "step": 66125
    },
    {
      "entropy": 5.820451736450195,
      "epoch": 6.56312028582771,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017101595498603838,
      "loss": 4.9786,
      "mean_token_accuracy": 0.1981346294283867,
      "num_tokens": 144891730.0,
      "step": 66130
    },
    {
      "entropy": 5.752694034576416,
      "epoch": 6.563616514489877,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017098453558625544,
      "loss": 4.9218,
      "mean_token_accuracy": 0.20241603553295134,
      "num_tokens": 144903406.0,
      "step": 66135
    },
    {
      "entropy": 5.684077358245849,
      "epoch": 6.5641127431520445,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001709531187658318,
      "loss": 4.8931,
      "mean_token_accuracy": 0.19693462550640106,
      "num_tokens": 144914485.0,
      "step": 66140
    },
    {
      "entropy": 5.7288947105407715,
      "epoch": 6.564608971814212,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017092170452554656,
      "loss": 4.9936,
      "mean_token_accuracy": 0.19614556282758713,
      "num_tokens": 144924097.0,
      "step": 66145
    },
    {
      "entropy": 5.705931091308594,
      "epoch": 6.5651052004763795,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001708902928661789,
      "loss": 4.9366,
      "mean_token_accuracy": 0.20523421466350555,
      "num_tokens": 144935615.0,
      "step": 66150
    },
    {
      "entropy": 5.760741949081421,
      "epoch": 6.565601429138547,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017085888378850762,
      "loss": 5.081,
      "mean_token_accuracy": 0.19330208599567414,
      "num_tokens": 144947733.0,
      "step": 66155
    },
    {
      "entropy": 5.708574819564819,
      "epoch": 6.566097657800714,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017082747729331153,
      "loss": 4.8899,
      "mean_token_accuracy": 0.1986674815416336,
      "num_tokens": 144958011.0,
      "step": 66160
    },
    {
      "entropy": 5.725519704818725,
      "epoch": 6.566593886462882,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001707960733813696,
      "loss": 4.8882,
      "mean_token_accuracy": 0.2025570183992386,
      "num_tokens": 144969342.0,
      "step": 66165
    },
    {
      "entropy": 5.7332212924957275,
      "epoch": 6.567090115125049,
      "grad_norm": 1.1875,
      "learning_rate": 0.00017076467205346052,
      "loss": 4.9437,
      "mean_token_accuracy": 0.2011904999613762,
      "num_tokens": 144981798.0,
      "step": 66170
    },
    {
      "entropy": 5.732491779327392,
      "epoch": 6.567586343787217,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017073327331036287,
      "loss": 4.9628,
      "mean_token_accuracy": 0.20157224237918853,
      "num_tokens": 144991989.0,
      "step": 66175
    },
    {
      "entropy": 5.689449644088745,
      "epoch": 6.568082572449384,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001707018771528555,
      "loss": 4.8768,
      "mean_token_accuracy": 0.2044170081615448,
      "num_tokens": 145002233.0,
      "step": 66180
    },
    {
      "entropy": 5.672785043716431,
      "epoch": 6.568578801111553,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001706704835817168,
      "loss": 4.8454,
      "mean_token_accuracy": 0.20395491570234298,
      "num_tokens": 145013152.0,
      "step": 66185
    },
    {
      "entropy": 5.6986246585845945,
      "epoch": 6.569075029773719,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017063909259772519,
      "loss": 4.8991,
      "mean_token_accuracy": 0.20587679594755173,
      "num_tokens": 145023892.0,
      "step": 66190
    },
    {
      "entropy": 5.721050262451172,
      "epoch": 6.569571258435888,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017060770420165933,
      "loss": 5.0111,
      "mean_token_accuracy": 0.20003390461206436,
      "num_tokens": 145035039.0,
      "step": 66195
    },
    {
      "entropy": 5.707358646392822,
      "epoch": 6.570067487098055,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017057631839429745,
      "loss": 4.9542,
      "mean_token_accuracy": 0.19790956228971482,
      "num_tokens": 145046979.0,
      "step": 66200
    },
    {
      "entropy": 5.68123426437378,
      "epoch": 6.570563715760223,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017054493517641776,
      "loss": 4.8766,
      "mean_token_accuracy": 0.19628486335277556,
      "num_tokens": 145057649.0,
      "step": 66205
    },
    {
      "entropy": 5.693785810470581,
      "epoch": 6.57105994442239,
      "grad_norm": 1.203125,
      "learning_rate": 0.00017051355454879869,
      "loss": 4.9186,
      "mean_token_accuracy": 0.20602132230997086,
      "num_tokens": 145070160.0,
      "step": 66210
    },
    {
      "entropy": 5.63556137084961,
      "epoch": 6.5715561730845575,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017048217651221832,
      "loss": 4.8045,
      "mean_token_accuracy": 0.2124409034848213,
      "num_tokens": 145080891.0,
      "step": 66215
    },
    {
      "entropy": 5.6797154426574705,
      "epoch": 6.572052401746725,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017045080106745474,
      "loss": 4.9101,
      "mean_token_accuracy": 0.19896160066127777,
      "num_tokens": 145091835.0,
      "step": 66220
    },
    {
      "entropy": 5.704905557632446,
      "epoch": 6.5725486304088925,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017041942821528595,
      "loss": 4.8485,
      "mean_token_accuracy": 0.20576452761888503,
      "num_tokens": 145102285.0,
      "step": 66225
    },
    {
      "entropy": 5.754557037353516,
      "epoch": 6.57304485907106,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017038805795648997,
      "loss": 4.9478,
      "mean_token_accuracy": 0.20309972912073135,
      "num_tokens": 145112966.0,
      "step": 66230
    },
    {
      "entropy": 5.605068683624268,
      "epoch": 6.573541087733227,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017035669029184485,
      "loss": 4.8158,
      "mean_token_accuracy": 0.20704911649227142,
      "num_tokens": 145125250.0,
      "step": 66235
    },
    {
      "entropy": 5.753292989730835,
      "epoch": 6.574037316395395,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001703253252221283,
      "loss": 4.9811,
      "mean_token_accuracy": 0.20122705399990082,
      "num_tokens": 145135446.0,
      "step": 66240
    },
    {
      "entropy": 5.672989225387573,
      "epoch": 6.574533545057562,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017029396274811814,
      "loss": 4.8463,
      "mean_token_accuracy": 0.20301771014928818,
      "num_tokens": 145144838.0,
      "step": 66245
    },
    {
      "entropy": 5.596189641952515,
      "epoch": 6.57502977371973,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00017026260287059203,
      "loss": 4.7842,
      "mean_token_accuracy": 0.20775424242019652,
      "num_tokens": 145155311.0,
      "step": 66250
    },
    {
      "entropy": 5.681007814407349,
      "epoch": 6.575526002381897,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017023124559032766,
      "loss": 4.939,
      "mean_token_accuracy": 0.19625871479511262,
      "num_tokens": 145166072.0,
      "step": 66255
    },
    {
      "entropy": 5.698826789855957,
      "epoch": 6.576022231044065,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001701998909081026,
      "loss": 4.9132,
      "mean_token_accuracy": 0.19630457758903502,
      "num_tokens": 145177027.0,
      "step": 66260
    },
    {
      "entropy": 5.737509441375733,
      "epoch": 6.576518459706232,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017016853882469455,
      "loss": 4.9349,
      "mean_token_accuracy": 0.198694409430027,
      "num_tokens": 145188425.0,
      "step": 66265
    },
    {
      "entropy": 5.731070041656494,
      "epoch": 6.577014688368401,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017013718934088084,
      "loss": 4.9625,
      "mean_token_accuracy": 0.19591965824365615,
      "num_tokens": 145200649.0,
      "step": 66270
    },
    {
      "entropy": 5.711991596221924,
      "epoch": 6.577510917030567,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017010584245743886,
      "loss": 4.889,
      "mean_token_accuracy": 0.2000196561217308,
      "num_tokens": 145210272.0,
      "step": 66275
    },
    {
      "entropy": 5.7840025424957275,
      "epoch": 6.578007145692736,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017007449817514595,
      "loss": 4.9719,
      "mean_token_accuracy": 0.1987984761595726,
      "num_tokens": 145221236.0,
      "step": 66280
    },
    {
      "entropy": 5.6579972267150875,
      "epoch": 6.578503374354903,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001700431564947793,
      "loss": 4.8725,
      "mean_token_accuracy": 0.204373499751091,
      "num_tokens": 145231383.0,
      "step": 66285
    },
    {
      "entropy": 5.693653774261475,
      "epoch": 6.5789996030170705,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017001181741711625,
      "loss": 4.9646,
      "mean_token_accuracy": 0.1962625801563263,
      "num_tokens": 145243871.0,
      "step": 66290
    },
    {
      "entropy": 5.706801414489746,
      "epoch": 6.579495831679238,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016998048094293394,
      "loss": 4.9427,
      "mean_token_accuracy": 0.2011193111538887,
      "num_tokens": 145256003.0,
      "step": 66295
    },
    {
      "entropy": 5.7068627834320065,
      "epoch": 6.5799920603414055,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016994914707300945,
      "loss": 4.8527,
      "mean_token_accuracy": 0.20814907252788545,
      "num_tokens": 145267245.0,
      "step": 66300
    },
    {
      "entropy": 5.718771028518677,
      "epoch": 6.580488289003573,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016991781580811971,
      "loss": 4.8838,
      "mean_token_accuracy": 0.20633617639541627,
      "num_tokens": 145277261.0,
      "step": 66305
    },
    {
      "entropy": 5.6484355449676515,
      "epoch": 6.58098451766574,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016988648714904176,
      "loss": 4.8814,
      "mean_token_accuracy": 0.20403511822223663,
      "num_tokens": 145287439.0,
      "step": 66310
    },
    {
      "entropy": 5.65683388710022,
      "epoch": 6.581480746327908,
      "grad_norm": 1.5625,
      "learning_rate": 0.00016985516109655236,
      "loss": 4.907,
      "mean_token_accuracy": 0.1980672597885132,
      "num_tokens": 145298581.0,
      "step": 66315
    },
    {
      "entropy": 5.673052167892456,
      "epoch": 6.581976974990075,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016982383765142833,
      "loss": 4.8221,
      "mean_token_accuracy": 0.20860812664031983,
      "num_tokens": 145309739.0,
      "step": 66320
    },
    {
      "entropy": 5.679898738861084,
      "epoch": 6.582473203652243,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001697925168144465,
      "loss": 4.8913,
      "mean_token_accuracy": 0.200460684299469,
      "num_tokens": 145321244.0,
      "step": 66325
    },
    {
      "entropy": 5.748720693588257,
      "epoch": 6.58296943231441,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016976119858638368,
      "loss": 4.9564,
      "mean_token_accuracy": 0.20454743802547454,
      "num_tokens": 145331992.0,
      "step": 66330
    },
    {
      "entropy": 5.740364503860474,
      "epoch": 6.583465660976578,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00016972988296801628,
      "loss": 5.0148,
      "mean_token_accuracy": 0.19635576754808426,
      "num_tokens": 145341491.0,
      "step": 66335
    },
    {
      "entropy": 5.67385311126709,
      "epoch": 6.583961889638745,
      "grad_norm": 1.3125,
      "learning_rate": 0.000169698569960121,
      "loss": 4.8153,
      "mean_token_accuracy": 0.20961445569992065,
      "num_tokens": 145351625.0,
      "step": 66340
    },
    {
      "entropy": 5.709109354019165,
      "epoch": 6.584458118300913,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016966725956347426,
      "loss": 4.812,
      "mean_token_accuracy": 0.20799775421619415,
      "num_tokens": 145363144.0,
      "step": 66345
    },
    {
      "entropy": 5.728043842315674,
      "epoch": 6.58495434696308,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016963595177885244,
      "loss": 4.9704,
      "mean_token_accuracy": 0.19508325755596162,
      "num_tokens": 145374787.0,
      "step": 66350
    },
    {
      "entropy": 5.6714582443237305,
      "epoch": 6.585450575625249,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016960464660703195,
      "loss": 4.8503,
      "mean_token_accuracy": 0.2024632751941681,
      "num_tokens": 145384959.0,
      "step": 66355
    },
    {
      "entropy": 5.7229540824890135,
      "epoch": 6.585946804287415,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016957334404878925,
      "loss": 4.9168,
      "mean_token_accuracy": 0.20151139944791793,
      "num_tokens": 145396276.0,
      "step": 66360
    },
    {
      "entropy": 5.605371570587158,
      "epoch": 6.5864430329495836,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016954204410490043,
      "loss": 4.8388,
      "mean_token_accuracy": 0.2173761785030365,
      "num_tokens": 145409168.0,
      "step": 66365
    },
    {
      "entropy": 5.703322029113769,
      "epoch": 6.586939261611751,
      "grad_norm": 1.53125,
      "learning_rate": 0.00016951074677614165,
      "loss": 4.9314,
      "mean_token_accuracy": 0.2014580711722374,
      "num_tokens": 145419401.0,
      "step": 66370
    },
    {
      "entropy": 5.729886150360107,
      "epoch": 6.5874354902739185,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016947945206328903,
      "loss": 4.9415,
      "mean_token_accuracy": 0.20370731949806214,
      "num_tokens": 145429352.0,
      "step": 66375
    },
    {
      "entropy": 5.71525559425354,
      "epoch": 6.587931718936086,
      "grad_norm": 1.578125,
      "learning_rate": 0.00016944815996711857,
      "loss": 4.8704,
      "mean_token_accuracy": 0.2054670661687851,
      "num_tokens": 145439390.0,
      "step": 66380
    },
    {
      "entropy": 5.759871912002564,
      "epoch": 6.5884279475982535,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016941687048840638,
      "loss": 4.9565,
      "mean_token_accuracy": 0.20063073188066483,
      "num_tokens": 145449635.0,
      "step": 66385
    },
    {
      "entropy": 5.741946792602539,
      "epoch": 6.588924176260421,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001693855836279282,
      "loss": 4.9774,
      "mean_token_accuracy": 0.19592114686965942,
      "num_tokens": 145460389.0,
      "step": 66390
    },
    {
      "entropy": 5.747984409332275,
      "epoch": 6.589420404922588,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016935429938646002,
      "loss": 4.886,
      "mean_token_accuracy": 0.1991167575120926,
      "num_tokens": 145470742.0,
      "step": 66395
    },
    {
      "entropy": 5.68766827583313,
      "epoch": 6.589916633584756,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001693230177647776,
      "loss": 4.8689,
      "mean_token_accuracy": 0.20175960063934326,
      "num_tokens": 145479979.0,
      "step": 66400
    },
    {
      "entropy": 5.700883913040161,
      "epoch": 6.590412862246923,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016929173876365663,
      "loss": 4.885,
      "mean_token_accuracy": 0.19993037581443787,
      "num_tokens": 145490248.0,
      "step": 66405
    },
    {
      "entropy": 5.701032209396362,
      "epoch": 6.590909090909091,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016926046238387266,
      "loss": 4.8989,
      "mean_token_accuracy": 0.2040208175778389,
      "num_tokens": 145500743.0,
      "step": 66410
    },
    {
      "entropy": 5.628908348083496,
      "epoch": 6.591405319571258,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016922918862620148,
      "loss": 4.8082,
      "mean_token_accuracy": 0.21674221605062485,
      "num_tokens": 145511131.0,
      "step": 66415
    },
    {
      "entropy": 5.7106348991394045,
      "epoch": 6.591901548233426,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016919791749141842,
      "loss": 4.9587,
      "mean_token_accuracy": 0.19153624027967453,
      "num_tokens": 145521355.0,
      "step": 66420
    },
    {
      "entropy": 5.7438445568084715,
      "epoch": 6.592397776895593,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016916664898029908,
      "loss": 4.9272,
      "mean_token_accuracy": 0.2067641630768776,
      "num_tokens": 145531145.0,
      "step": 66425
    },
    {
      "entropy": 5.680214309692383,
      "epoch": 6.592894005557761,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001691353830936188,
      "loss": 4.8153,
      "mean_token_accuracy": 0.2100769430398941,
      "num_tokens": 145541038.0,
      "step": 66430
    },
    {
      "entropy": 5.72214560508728,
      "epoch": 6.593390234219928,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001691041198321529,
      "loss": 4.9109,
      "mean_token_accuracy": 0.20041720420122147,
      "num_tokens": 145551104.0,
      "step": 66435
    },
    {
      "entropy": 5.690158462524414,
      "epoch": 6.593886462882097,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001690728591966766,
      "loss": 4.9465,
      "mean_token_accuracy": 0.20252465307712555,
      "num_tokens": 145561377.0,
      "step": 66440
    },
    {
      "entropy": 5.65726523399353,
      "epoch": 6.594382691544263,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001690416011879652,
      "loss": 4.8989,
      "mean_token_accuracy": 0.19794320464134216,
      "num_tokens": 145572088.0,
      "step": 66445
    },
    {
      "entropy": 5.807177495956421,
      "epoch": 6.5948789202064315,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016901034580679375,
      "loss": 4.9968,
      "mean_token_accuracy": 0.19341244399547577,
      "num_tokens": 145582207.0,
      "step": 66450
    },
    {
      "entropy": 5.714050912857056,
      "epoch": 6.595375148868599,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016897909305393726,
      "loss": 4.912,
      "mean_token_accuracy": 0.20297021716833114,
      "num_tokens": 145592690.0,
      "step": 66455
    },
    {
      "entropy": 5.739378118515015,
      "epoch": 6.5958713775307665,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00016894784293017092,
      "loss": 4.9798,
      "mean_token_accuracy": 0.19954613000154495,
      "num_tokens": 145602737.0,
      "step": 66460
    },
    {
      "entropy": 5.775196504592896,
      "epoch": 6.596367606192934,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001689165954362694,
      "loss": 4.949,
      "mean_token_accuracy": 0.20014283657073975,
      "num_tokens": 145612590.0,
      "step": 66465
    },
    {
      "entropy": 5.731106185913086,
      "epoch": 6.596863834855101,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001688853505730079,
      "loss": 4.9638,
      "mean_token_accuracy": 0.1985527127981186,
      "num_tokens": 145624502.0,
      "step": 66470
    },
    {
      "entropy": 5.757048273086548,
      "epoch": 6.597360063517269,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000168854108341161,
      "loss": 4.9827,
      "mean_token_accuracy": 0.19600838273763657,
      "num_tokens": 145635664.0,
      "step": 66475
    },
    {
      "entropy": 5.727694034576416,
      "epoch": 6.597856292179436,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001688228687415035,
      "loss": 4.9387,
      "mean_token_accuracy": 0.20498693138360977,
      "num_tokens": 145648208.0,
      "step": 66480
    },
    {
      "entropy": 5.660043478012085,
      "epoch": 6.598352520841604,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016879163177480993,
      "loss": 4.89,
      "mean_token_accuracy": 0.20946349650621415,
      "num_tokens": 145660265.0,
      "step": 66485
    },
    {
      "entropy": 5.634526205062866,
      "epoch": 6.598848749503771,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016876039744185517,
      "loss": 4.8794,
      "mean_token_accuracy": 0.19531392604112624,
      "num_tokens": 145671549.0,
      "step": 66490
    },
    {
      "entropy": 5.668732500076294,
      "epoch": 6.599344978165939,
      "grad_norm": 1.609375,
      "learning_rate": 0.00016872916574341354,
      "loss": 4.8855,
      "mean_token_accuracy": 0.2062298074364662,
      "num_tokens": 145682753.0,
      "step": 66495
    },
    {
      "entropy": 5.6957714557647705,
      "epoch": 6.599841206828106,
      "grad_norm": 1.65625,
      "learning_rate": 0.00016869793668025968,
      "loss": 4.8403,
      "mean_token_accuracy": 0.20910882502794265,
      "num_tokens": 145692517.0,
      "step": 66500
    },
    {
      "entropy": 5.71058669090271,
      "epoch": 6.600337435490274,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016866671025316793,
      "loss": 4.8853,
      "mean_token_accuracy": 0.20641054660081865,
      "num_tokens": 145702852.0,
      "step": 66505
    },
    {
      "entropy": 5.665703201293946,
      "epoch": 6.600833664152441,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016863548646291266,
      "loss": 4.8115,
      "mean_token_accuracy": 0.21641784459352492,
      "num_tokens": 145712719.0,
      "step": 66510
    },
    {
      "entropy": 5.6682535171508786,
      "epoch": 6.601329892814609,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00016860426531026805,
      "loss": 4.872,
      "mean_token_accuracy": 0.1969972938299179,
      "num_tokens": 145723074.0,
      "step": 66515
    },
    {
      "entropy": 5.684383678436279,
      "epoch": 6.601826121476776,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016857304679600843,
      "loss": 4.874,
      "mean_token_accuracy": 0.20581323355436326,
      "num_tokens": 145733969.0,
      "step": 66520
    },
    {
      "entropy": 5.658956670761109,
      "epoch": 6.602322350138944,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016854183092090787,
      "loss": 4.8818,
      "mean_token_accuracy": 0.20639892369508744,
      "num_tokens": 145744229.0,
      "step": 66525
    },
    {
      "entropy": 5.735975503921509,
      "epoch": 6.602818578801111,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016851061768574055,
      "loss": 4.9412,
      "mean_token_accuracy": 0.19836848676204683,
      "num_tokens": 145755811.0,
      "step": 66530
    },
    {
      "entropy": 5.669772005081176,
      "epoch": 6.6033148074632795,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016847940709128052,
      "loss": 4.8934,
      "mean_token_accuracy": 0.1995086833834648,
      "num_tokens": 145767027.0,
      "step": 66535
    },
    {
      "entropy": 5.697030544281006,
      "epoch": 6.603811036125447,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001684481991383016,
      "loss": 4.9099,
      "mean_token_accuracy": 0.20661768615245818,
      "num_tokens": 145776685.0,
      "step": 66540
    },
    {
      "entropy": 5.66457839012146,
      "epoch": 6.604307264787614,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016841699382757778,
      "loss": 4.9047,
      "mean_token_accuracy": 0.2039113685488701,
      "num_tokens": 145786505.0,
      "step": 66545
    },
    {
      "entropy": 5.649172115325928,
      "epoch": 6.604803493449782,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016838579115988276,
      "loss": 4.8528,
      "mean_token_accuracy": 0.20505717396736145,
      "num_tokens": 145796702.0,
      "step": 66550
    },
    {
      "entropy": 5.674807119369507,
      "epoch": 6.605299722111949,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001683545911359904,
      "loss": 4.9067,
      "mean_token_accuracy": 0.2006167307496071,
      "num_tokens": 145806364.0,
      "step": 66555
    },
    {
      "entropy": 5.658866310119629,
      "epoch": 6.605795950774117,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001683233937566744,
      "loss": 4.8875,
      "mean_token_accuracy": 0.20292219072580336,
      "num_tokens": 145817982.0,
      "step": 66560
    },
    {
      "entropy": 5.71027135848999,
      "epoch": 6.606292179436284,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00016829219902270842,
      "loss": 4.946,
      "mean_token_accuracy": 0.2030674159526825,
      "num_tokens": 145828167.0,
      "step": 66565
    },
    {
      "entropy": 5.640036964416504,
      "epoch": 6.606788408098452,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016826100693486596,
      "loss": 4.7969,
      "mean_token_accuracy": 0.20578701794147491,
      "num_tokens": 145839302.0,
      "step": 66570
    },
    {
      "entropy": 5.713031768798828,
      "epoch": 6.607284636760619,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016822981749392053,
      "loss": 4.9266,
      "mean_token_accuracy": 0.20007849484682083,
      "num_tokens": 145850121.0,
      "step": 66575
    },
    {
      "entropy": 5.667903423309326,
      "epoch": 6.607780865422787,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016819863070064546,
      "loss": 4.8868,
      "mean_token_accuracy": 0.20639984756708146,
      "num_tokens": 145860783.0,
      "step": 66580
    },
    {
      "entropy": 5.6592981815338135,
      "epoch": 6.608277094084954,
      "grad_norm": 1.375,
      "learning_rate": 0.00016816744655581418,
      "loss": 4.8826,
      "mean_token_accuracy": 0.20722000300884247,
      "num_tokens": 145870824.0,
      "step": 66585
    },
    {
      "entropy": 5.696536493301392,
      "epoch": 6.608773322747122,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016813626506020018,
      "loss": 4.9271,
      "mean_token_accuracy": 0.19963008314371108,
      "num_tokens": 145881780.0,
      "step": 66590
    },
    {
      "entropy": 5.806149625778199,
      "epoch": 6.609269551409289,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016810508621457648,
      "loss": 5.0244,
      "mean_token_accuracy": 0.19231298118829726,
      "num_tokens": 145891660.0,
      "step": 66595
    },
    {
      "entropy": 5.720508337020874,
      "epoch": 6.609765780071457,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016807391001971635,
      "loss": 4.9163,
      "mean_token_accuracy": 0.20290490239858627,
      "num_tokens": 145903051.0,
      "step": 66600
    },
    {
      "entropy": 5.680889701843261,
      "epoch": 6.610262008733624,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001680427364763928,
      "loss": 4.91,
      "mean_token_accuracy": 0.20298989862203598,
      "num_tokens": 145913737.0,
      "step": 66605
    },
    {
      "entropy": 5.677901268005371,
      "epoch": 6.610758237395792,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001680115655853789,
      "loss": 4.8796,
      "mean_token_accuracy": 0.19938537925481797,
      "num_tokens": 145924500.0,
      "step": 66610
    },
    {
      "entropy": 5.607950639724732,
      "epoch": 6.611254466057959,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016798039734744764,
      "loss": 4.8128,
      "mean_token_accuracy": 0.21406224519014358,
      "num_tokens": 145935312.0,
      "step": 66615
    },
    {
      "entropy": 5.703776168823242,
      "epoch": 6.611750694720127,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016794923176337184,
      "loss": 4.9171,
      "mean_token_accuracy": 0.1971818909049034,
      "num_tokens": 145946070.0,
      "step": 66620
    },
    {
      "entropy": 5.701095914840698,
      "epoch": 6.612246923382295,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001679180688339245,
      "loss": 4.8696,
      "mean_token_accuracy": 0.19914990663528442,
      "num_tokens": 145957603.0,
      "step": 66625
    },
    {
      "entropy": 5.72453088760376,
      "epoch": 6.612743152044462,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00016788690855987836,
      "loss": 4.9029,
      "mean_token_accuracy": 0.19493196159601212,
      "num_tokens": 145968862.0,
      "step": 66630
    },
    {
      "entropy": 5.743849563598633,
      "epoch": 6.61323938070663,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016785575094200605,
      "loss": 4.9405,
      "mean_token_accuracy": 0.19591363221406938,
      "num_tokens": 145979047.0,
      "step": 66635
    },
    {
      "entropy": 5.734225940704346,
      "epoch": 6.613735609368797,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016782459598108025,
      "loss": 4.9837,
      "mean_token_accuracy": 0.20326274186372756,
      "num_tokens": 145990049.0,
      "step": 66640
    },
    {
      "entropy": 5.733135509490967,
      "epoch": 6.614231838030965,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00016779344367787342,
      "loss": 4.9469,
      "mean_token_accuracy": 0.19225206822156907,
      "num_tokens": 146001036.0,
      "step": 66645
    },
    {
      "entropy": 5.750565814971924,
      "epoch": 6.614728066693132,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001677622940331582,
      "loss": 4.8955,
      "mean_token_accuracy": 0.2013338327407837,
      "num_tokens": 146010613.0,
      "step": 66650
    },
    {
      "entropy": 5.726624822616577,
      "epoch": 6.6152242953553,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016773114704770708,
      "loss": 4.9517,
      "mean_token_accuracy": 0.19518731534481049,
      "num_tokens": 146021118.0,
      "step": 66655
    },
    {
      "entropy": 5.706900501251221,
      "epoch": 6.615720524017467,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001677000027222924,
      "loss": 4.8476,
      "mean_token_accuracy": 0.20806994438171386,
      "num_tokens": 146032532.0,
      "step": 66660
    },
    {
      "entropy": 5.690373659133911,
      "epoch": 6.616216752679635,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001676688610576864,
      "loss": 4.9423,
      "mean_token_accuracy": 0.1993782639503479,
      "num_tokens": 146043137.0,
      "step": 66665
    },
    {
      "entropy": 5.6484208583831785,
      "epoch": 6.616712981341802,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016763772205466142,
      "loss": 4.8453,
      "mean_token_accuracy": 0.20513207465410233,
      "num_tokens": 146054563.0,
      "step": 66670
    },
    {
      "entropy": 5.702189111709595,
      "epoch": 6.61720921000397,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016760658571398946,
      "loss": 4.91,
      "mean_token_accuracy": 0.20982744991779329,
      "num_tokens": 146065764.0,
      "step": 66675
    },
    {
      "entropy": 5.692409038543701,
      "epoch": 6.617705438666137,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016757545203644286,
      "loss": 4.8671,
      "mean_token_accuracy": 0.20735528469085693,
      "num_tokens": 146076943.0,
      "step": 66680
    },
    {
      "entropy": 5.729983329772949,
      "epoch": 6.618201667328305,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001675443210227935,
      "loss": 4.8823,
      "mean_token_accuracy": 0.2111161842942238,
      "num_tokens": 146088538.0,
      "step": 66685
    },
    {
      "entropy": 5.750519704818726,
      "epoch": 6.618697895990472,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016751319267381348,
      "loss": 4.9843,
      "mean_token_accuracy": 0.19418792128562928,
      "num_tokens": 146102133.0,
      "step": 66690
    },
    {
      "entropy": 5.711822605133056,
      "epoch": 6.61919412465264,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00016748206699027468,
      "loss": 4.9108,
      "mean_token_accuracy": 0.2058954730629921,
      "num_tokens": 146113145.0,
      "step": 66695
    },
    {
      "entropy": 5.694185400009156,
      "epoch": 6.619690353314807,
      "grad_norm": 1.5625,
      "learning_rate": 0.00016745094397294897,
      "loss": 4.8597,
      "mean_token_accuracy": 0.20686090290546416,
      "num_tokens": 146122330.0,
      "step": 66700
    },
    {
      "entropy": 5.672105550765991,
      "epoch": 6.620186581976975,
      "grad_norm": 1.28125,
      "learning_rate": 0.000167419823622608,
      "loss": 4.8293,
      "mean_token_accuracy": 0.19885418713092803,
      "num_tokens": 146133213.0,
      "step": 66705
    },
    {
      "entropy": 5.662668514251709,
      "epoch": 6.620682810639143,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00016738870594002365,
      "loss": 4.8551,
      "mean_token_accuracy": 0.20334564447402953,
      "num_tokens": 146143711.0,
      "step": 66710
    },
    {
      "entropy": 5.675401210784912,
      "epoch": 6.62117903930131,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016735759092596748,
      "loss": 4.9335,
      "mean_token_accuracy": 0.20496983230113983,
      "num_tokens": 146156001.0,
      "step": 66715
    },
    {
      "entropy": 5.723723745346069,
      "epoch": 6.621675267963478,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016732647858121112,
      "loss": 5.0025,
      "mean_token_accuracy": 0.19251253008842467,
      "num_tokens": 146166537.0,
      "step": 66720
    },
    {
      "entropy": 5.599199438095093,
      "epoch": 6.622171496625645,
      "grad_norm": 1.375,
      "learning_rate": 0.00016729536890652613,
      "loss": 4.7785,
      "mean_token_accuracy": 0.2143157258629799,
      "num_tokens": 146176854.0,
      "step": 66725
    },
    {
      "entropy": 5.662532377243042,
      "epoch": 6.622667725287813,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001672642619026838,
      "loss": 4.9131,
      "mean_token_accuracy": 0.19502490162849426,
      "num_tokens": 146188103.0,
      "step": 66730
    },
    {
      "entropy": 5.646632719039917,
      "epoch": 6.62316395394998,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001672331575704557,
      "loss": 4.8453,
      "mean_token_accuracy": 0.20572880059480667,
      "num_tokens": 146198726.0,
      "step": 66735
    },
    {
      "entropy": 5.704725742340088,
      "epoch": 6.623660182612148,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001672020559106131,
      "loss": 4.9581,
      "mean_token_accuracy": 0.20314017981290816,
      "num_tokens": 146209926.0,
      "step": 66740
    },
    {
      "entropy": 5.636924076080322,
      "epoch": 6.624156411274315,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016717095692392713,
      "loss": 4.8177,
      "mean_token_accuracy": 0.20681734532117843,
      "num_tokens": 146220573.0,
      "step": 66745
    },
    {
      "entropy": 5.669756460189819,
      "epoch": 6.624652639936483,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00016713986061116915,
      "loss": 4.8695,
      "mean_token_accuracy": 0.20283394008874894,
      "num_tokens": 146230001.0,
      "step": 66750
    },
    {
      "entropy": 5.6971272945404055,
      "epoch": 6.62514886859865,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016710876697311023,
      "loss": 4.9826,
      "mean_token_accuracy": 0.1924174338579178,
      "num_tokens": 146242054.0,
      "step": 66755
    },
    {
      "entropy": 5.773870277404785,
      "epoch": 6.625645097260818,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001670776760105213,
      "loss": 5.0273,
      "mean_token_accuracy": 0.20107542276382445,
      "num_tokens": 146254310.0,
      "step": 66760
    },
    {
      "entropy": 5.736856985092163,
      "epoch": 6.626141325922985,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001670465877241736,
      "loss": 4.9347,
      "mean_token_accuracy": 0.19235622137784958,
      "num_tokens": 146265109.0,
      "step": 66765
    },
    {
      "entropy": 5.695629692077636,
      "epoch": 6.626637554585153,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016701550211483784,
      "loss": 4.9215,
      "mean_token_accuracy": 0.19271002262830733,
      "num_tokens": 146276593.0,
      "step": 66770
    },
    {
      "entropy": 5.7011395454406735,
      "epoch": 6.62713378324732,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016698441918328497,
      "loss": 4.946,
      "mean_token_accuracy": 0.20017705708742142,
      "num_tokens": 146285632.0,
      "step": 66775
    },
    {
      "entropy": 5.76882119178772,
      "epoch": 6.6276300119094875,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016695333893028562,
      "loss": 4.9925,
      "mean_token_accuracy": 0.20346392393112184,
      "num_tokens": 146297287.0,
      "step": 66780
    },
    {
      "entropy": 5.7015931606292725,
      "epoch": 6.628126240571655,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016692226135661082,
      "loss": 4.8734,
      "mean_token_accuracy": 0.20269855558872224,
      "num_tokens": 146307114.0,
      "step": 66785
    },
    {
      "entropy": 5.688730335235595,
      "epoch": 6.628622469233823,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001668911864630309,
      "loss": 4.8458,
      "mean_token_accuracy": 0.2084992527961731,
      "num_tokens": 146318879.0,
      "step": 66790
    },
    {
      "entropy": 5.720897054672241,
      "epoch": 6.629118697895991,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001668601142503167,
      "loss": 4.914,
      "mean_token_accuracy": 0.20264948457479476,
      "num_tokens": 146329510.0,
      "step": 66795
    },
    {
      "entropy": 5.700384998321534,
      "epoch": 6.629614926558158,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016682904471923863,
      "loss": 4.9486,
      "mean_token_accuracy": 0.20469415783882142,
      "num_tokens": 146341018.0,
      "step": 66800
    },
    {
      "entropy": 5.6853760242462155,
      "epoch": 6.630111155220326,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016679797787056718,
      "loss": 4.9463,
      "mean_token_accuracy": 0.20224428027868271,
      "num_tokens": 146352423.0,
      "step": 66805
    },
    {
      "entropy": 5.70307035446167,
      "epoch": 6.630607383882493,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016676691370507264,
      "loss": 4.9124,
      "mean_token_accuracy": 0.19760935008525848,
      "num_tokens": 146363665.0,
      "step": 66810
    },
    {
      "entropy": 5.717015600204467,
      "epoch": 6.631103612544661,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001667358522235255,
      "loss": 4.9242,
      "mean_token_accuracy": 0.20282432138919831,
      "num_tokens": 146374544.0,
      "step": 66815
    },
    {
      "entropy": 5.846549367904663,
      "epoch": 6.631599841206828,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016670479342669584,
      "loss": 5.0717,
      "mean_token_accuracy": 0.19005496501922609,
      "num_tokens": 146385405.0,
      "step": 66820
    },
    {
      "entropy": 5.695884037017822,
      "epoch": 6.632096069868996,
      "grad_norm": 1.25,
      "learning_rate": 0.00016667373731535403,
      "loss": 4.9047,
      "mean_token_accuracy": 0.19398027509450913,
      "num_tokens": 146395850.0,
      "step": 66825
    },
    {
      "entropy": 5.698381233215332,
      "epoch": 6.632592298531163,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016664268389027015,
      "loss": 4.9599,
      "mean_token_accuracy": 0.2007714256644249,
      "num_tokens": 146406976.0,
      "step": 66830
    },
    {
      "entropy": 5.714422798156738,
      "epoch": 6.633088527193331,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001666116331522142,
      "loss": 4.9583,
      "mean_token_accuracy": 0.19816772192716597,
      "num_tokens": 146419206.0,
      "step": 66835
    },
    {
      "entropy": 5.6869580268859865,
      "epoch": 6.633584755855498,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001665805851019562,
      "loss": 4.8768,
      "mean_token_accuracy": 0.2002001166343689,
      "num_tokens": 146430461.0,
      "step": 66840
    },
    {
      "entropy": 5.657197380065918,
      "epoch": 6.634080984517666,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016654953974026596,
      "loss": 4.8972,
      "mean_token_accuracy": 0.19994949102401732,
      "num_tokens": 146441676.0,
      "step": 66845
    },
    {
      "entropy": 5.667426490783692,
      "epoch": 6.634577213179833,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016651849706791343,
      "loss": 4.8497,
      "mean_token_accuracy": 0.2035098060965538,
      "num_tokens": 146452066.0,
      "step": 66850
    },
    {
      "entropy": 5.677710390090942,
      "epoch": 6.6350734418420005,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016648745708566857,
      "loss": 4.893,
      "mean_token_accuracy": 0.20717411041259765,
      "num_tokens": 146464243.0,
      "step": 66855
    },
    {
      "entropy": 5.705164194107056,
      "epoch": 6.635569670504168,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016645641979430088,
      "loss": 4.8843,
      "mean_token_accuracy": 0.19985323995351792,
      "num_tokens": 146475812.0,
      "step": 66860
    },
    {
      "entropy": 5.716047286987305,
      "epoch": 6.6360658991663355,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001664253851945801,
      "loss": 4.9103,
      "mean_token_accuracy": 0.19775169789791108,
      "num_tokens": 146486153.0,
      "step": 66865
    },
    {
      "entropy": 5.698606157302857,
      "epoch": 6.636562127828503,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00016639435328727586,
      "loss": 4.8885,
      "mean_token_accuracy": 0.20548801273107528,
      "num_tokens": 146497375.0,
      "step": 66870
    },
    {
      "entropy": 5.720222425460816,
      "epoch": 6.637058356490671,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001663633240731575,
      "loss": 4.9422,
      "mean_token_accuracy": 0.1995338410139084,
      "num_tokens": 146508490.0,
      "step": 66875
    },
    {
      "entropy": 5.637593364715576,
      "epoch": 6.637554585152839,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001663322975529946,
      "loss": 4.8354,
      "mean_token_accuracy": 0.2056897461414337,
      "num_tokens": 146519344.0,
      "step": 66880
    },
    {
      "entropy": 5.725462865829468,
      "epoch": 6.638050813815006,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001663012737275567,
      "loss": 4.9198,
      "mean_token_accuracy": 0.20160120129585266,
      "num_tokens": 146529640.0,
      "step": 66885
    },
    {
      "entropy": 5.709135580062866,
      "epoch": 6.638547042477174,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016627025259761296,
      "loss": 4.952,
      "mean_token_accuracy": 0.19844433218240737,
      "num_tokens": 146541810.0,
      "step": 66890
    },
    {
      "entropy": 5.767633104324341,
      "epoch": 6.639043271139341,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016623923416393267,
      "loss": 5.0152,
      "mean_token_accuracy": 0.1887466862797737,
      "num_tokens": 146552548.0,
      "step": 66895
    },
    {
      "entropy": 5.736945343017578,
      "epoch": 6.639539499801509,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016620821842728499,
      "loss": 4.9294,
      "mean_token_accuracy": 0.1982081115245819,
      "num_tokens": 146563783.0,
      "step": 66900
    },
    {
      "entropy": 5.730089521408081,
      "epoch": 6.640035728463676,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016617720538843905,
      "loss": 4.9535,
      "mean_token_accuracy": 0.1992978885769844,
      "num_tokens": 146574935.0,
      "step": 66905
    },
    {
      "entropy": 5.761652088165283,
      "epoch": 6.640531957125844,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00016614619504816392,
      "loss": 4.984,
      "mean_token_accuracy": 0.1963501587510109,
      "num_tokens": 146586761.0,
      "step": 66910
    },
    {
      "entropy": 5.708215188980103,
      "epoch": 6.641028185788011,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016611518740722853,
      "loss": 4.9064,
      "mean_token_accuracy": 0.20442621409893036,
      "num_tokens": 146598017.0,
      "step": 66915
    },
    {
      "entropy": 5.6889513492584225,
      "epoch": 6.641524414450179,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016608418246640198,
      "loss": 4.9168,
      "mean_token_accuracy": 0.19908225238323213,
      "num_tokens": 146609327.0,
      "step": 66920
    },
    {
      "entropy": 5.665620470046997,
      "epoch": 6.642020643112346,
      "grad_norm": 1.484375,
      "learning_rate": 0.000166053180226453,
      "loss": 4.8563,
      "mean_token_accuracy": 0.20213147699832917,
      "num_tokens": 146618830.0,
      "step": 66925
    },
    {
      "entropy": 5.603551816940308,
      "epoch": 6.6425168717745136,
      "grad_norm": 1.15625,
      "learning_rate": 0.00016602218068815039,
      "loss": 4.8752,
      "mean_token_accuracy": 0.2068727880716324,
      "num_tokens": 146632336.0,
      "step": 66930
    },
    {
      "entropy": 5.711255502700806,
      "epoch": 6.643013100436681,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016599118385226286,
      "loss": 4.9981,
      "mean_token_accuracy": 0.19473231583833694,
      "num_tokens": 146644048.0,
      "step": 66935
    },
    {
      "entropy": 5.728542995452881,
      "epoch": 6.6435093290988485,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000165960189719559,
      "loss": 4.8468,
      "mean_token_accuracy": 0.20729511678218843,
      "num_tokens": 146654245.0,
      "step": 66940
    },
    {
      "entropy": 5.6430134773254395,
      "epoch": 6.644005557761016,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016592919829080748,
      "loss": 4.7993,
      "mean_token_accuracy": 0.20714931190013885,
      "num_tokens": 146664647.0,
      "step": 66945
    },
    {
      "entropy": 5.676347255706787,
      "epoch": 6.6445017864231835,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016589820956677693,
      "loss": 4.8914,
      "mean_token_accuracy": 0.2007976785302162,
      "num_tokens": 146675235.0,
      "step": 66950
    },
    {
      "entropy": 5.719195747375489,
      "epoch": 6.644998015085351,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00016586722354823566,
      "loss": 4.8416,
      "mean_token_accuracy": 0.21198468506336213,
      "num_tokens": 146686142.0,
      "step": 66955
    },
    {
      "entropy": 5.689234209060669,
      "epoch": 6.645494243747519,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016583624023595205,
      "loss": 4.9334,
      "mean_token_accuracy": 0.19758798629045488,
      "num_tokens": 146697157.0,
      "step": 66960
    },
    {
      "entropy": 5.677664184570313,
      "epoch": 6.645990472409687,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016580525963069443,
      "loss": 4.9072,
      "mean_token_accuracy": 0.19549735486507416,
      "num_tokens": 146707252.0,
      "step": 66965
    },
    {
      "entropy": 5.675231885910034,
      "epoch": 6.646486701071854,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016577428173323113,
      "loss": 4.903,
      "mean_token_accuracy": 0.2003978580236435,
      "num_tokens": 146718557.0,
      "step": 66970
    },
    {
      "entropy": 5.774390125274659,
      "epoch": 6.646982929734022,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00016574330654433027,
      "loss": 4.9711,
      "mean_token_accuracy": 0.2005539506673813,
      "num_tokens": 146729278.0,
      "step": 66975
    },
    {
      "entropy": 5.694638681411743,
      "epoch": 6.647479158396189,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016571233406475995,
      "loss": 4.915,
      "mean_token_accuracy": 0.19937595576047898,
      "num_tokens": 146739910.0,
      "step": 66980
    },
    {
      "entropy": 5.7075684547424315,
      "epoch": 6.647975387058357,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001656813642952883,
      "loss": 4.899,
      "mean_token_accuracy": 0.20524150431156157,
      "num_tokens": 146750944.0,
      "step": 66985
    },
    {
      "entropy": 5.652526378631592,
      "epoch": 6.648471615720524,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016565039723668326,
      "loss": 4.7936,
      "mean_token_accuracy": 0.20772742480039597,
      "num_tokens": 146761663.0,
      "step": 66990
    },
    {
      "entropy": 5.660124397277832,
      "epoch": 6.648967844382692,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016561943288971264,
      "loss": 4.8215,
      "mean_token_accuracy": 0.20351164788007736,
      "num_tokens": 146772340.0,
      "step": 66995
    },
    {
      "entropy": 5.7150791645050045,
      "epoch": 6.649464073044859,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016558847125514451,
      "loss": 4.932,
      "mean_token_accuracy": 0.19314448684453964,
      "num_tokens": 146782698.0,
      "step": 67000
    },
    {
      "entropy": 5.677914142608643,
      "epoch": 6.649960301707027,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001655575123337465,
      "loss": 4.8804,
      "mean_token_accuracy": 0.20324520021677017,
      "num_tokens": 146793710.0,
      "step": 67005
    },
    {
      "entropy": 5.668262147903443,
      "epoch": 6.650456530369194,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016552655612628626,
      "loss": 4.8854,
      "mean_token_accuracy": 0.201912584900856,
      "num_tokens": 146804247.0,
      "step": 67010
    },
    {
      "entropy": 5.65361909866333,
      "epoch": 6.6509527590313615,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016549560263353166,
      "loss": 4.8875,
      "mean_token_accuracy": 0.20035144090652465,
      "num_tokens": 146815972.0,
      "step": 67015
    },
    {
      "entropy": 5.7102419376373295,
      "epoch": 6.651448987693529,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016546465185625014,
      "loss": 4.9522,
      "mean_token_accuracy": 0.19922931492328644,
      "num_tokens": 146827963.0,
      "step": 67020
    },
    {
      "entropy": 5.786422872543335,
      "epoch": 6.6519452163556965,
      "grad_norm": 1.25,
      "learning_rate": 0.00016543370379520916,
      "loss": 4.9702,
      "mean_token_accuracy": 0.19813242554664612,
      "num_tokens": 146838908.0,
      "step": 67025
    },
    {
      "entropy": 5.814096355438233,
      "epoch": 6.652441445017864,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016540275845117628,
      "loss": 4.993,
      "mean_token_accuracy": 0.19954592734575272,
      "num_tokens": 146850524.0,
      "step": 67030
    },
    {
      "entropy": 5.7186627864837645,
      "epoch": 6.652937673680031,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016537181582491884,
      "loss": 4.8863,
      "mean_token_accuracy": 0.20051683932542802,
      "num_tokens": 146860591.0,
      "step": 67035
    },
    {
      "entropy": 5.7714954853057865,
      "epoch": 6.653433902342199,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016534087591720404,
      "loss": 5.0048,
      "mean_token_accuracy": 0.19497784227132797,
      "num_tokens": 146871959.0,
      "step": 67040
    },
    {
      "entropy": 5.7371715068817135,
      "epoch": 6.653930131004367,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016530993872879932,
      "loss": 4.9094,
      "mean_token_accuracy": 0.2016497313976288,
      "num_tokens": 146882731.0,
      "step": 67045
    },
    {
      "entropy": 5.673428010940552,
      "epoch": 6.654426359666534,
      "grad_norm": 1.21875,
      "learning_rate": 0.00016527900426047178,
      "loss": 4.7981,
      "mean_token_accuracy": 0.2153930351138115,
      "num_tokens": 146894454.0,
      "step": 67050
    },
    {
      "entropy": 5.745143604278565,
      "epoch": 6.654922588328702,
      "grad_norm": 1.5,
      "learning_rate": 0.0001652480725129884,
      "loss": 4.9496,
      "mean_token_accuracy": 0.19412653744220734,
      "num_tokens": 146905841.0,
      "step": 67055
    },
    {
      "entropy": 5.649278211593628,
      "epoch": 6.65541881699087,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016521714348711642,
      "loss": 4.921,
      "mean_token_accuracy": 0.19494407325983049,
      "num_tokens": 146917021.0,
      "step": 67060
    },
    {
      "entropy": 5.69220666885376,
      "epoch": 6.655915045653037,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001651862171836227,
      "loss": 4.9557,
      "mean_token_accuracy": 0.19447592049837112,
      "num_tokens": 146928217.0,
      "step": 67065
    },
    {
      "entropy": 5.7052699565887455,
      "epoch": 6.656411274315205,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016515529360327418,
      "loss": 4.893,
      "mean_token_accuracy": 0.19997561126947402,
      "num_tokens": 146938783.0,
      "step": 67070
    },
    {
      "entropy": 5.681835699081421,
      "epoch": 6.656907502977372,
      "grad_norm": 1.375,
      "learning_rate": 0.00016512437274683763,
      "loss": 4.8996,
      "mean_token_accuracy": 0.19725261628627777,
      "num_tokens": 146949406.0,
      "step": 67075
    },
    {
      "entropy": 5.65516242980957,
      "epoch": 6.65740373163954,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001650934546150799,
      "loss": 4.8664,
      "mean_token_accuracy": 0.20463763326406478,
      "num_tokens": 146960500.0,
      "step": 67080
    },
    {
      "entropy": 5.684510135650635,
      "epoch": 6.657899960301707,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016506253920876761,
      "loss": 4.8833,
      "mean_token_accuracy": 0.20540997087955476,
      "num_tokens": 146970621.0,
      "step": 67085
    },
    {
      "entropy": 5.655406761169433,
      "epoch": 6.6583961889638745,
      "grad_norm": 1.375,
      "learning_rate": 0.00016503162652866757,
      "loss": 4.8526,
      "mean_token_accuracy": 0.20431250035762788,
      "num_tokens": 146982020.0,
      "step": 67090
    },
    {
      "entropy": 5.648063707351684,
      "epoch": 6.658892417626042,
      "grad_norm": 1.375,
      "learning_rate": 0.00016500071657554621,
      "loss": 4.8694,
      "mean_token_accuracy": 0.20801501274108886,
      "num_tokens": 146991626.0,
      "step": 67095
    },
    {
      "entropy": 5.7026113986969,
      "epoch": 6.6593886462882095,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016496980935017003,
      "loss": 4.8954,
      "mean_token_accuracy": 0.19873828142881395,
      "num_tokens": 147002834.0,
      "step": 67100
    },
    {
      "entropy": 5.769192504882812,
      "epoch": 6.659884874950377,
      "grad_norm": 1.53125,
      "learning_rate": 0.00016493890485330543,
      "loss": 5.0018,
      "mean_token_accuracy": 0.19289848655462266,
      "num_tokens": 147014395.0,
      "step": 67105
    },
    {
      "entropy": 5.707742071151733,
      "epoch": 6.660381103612544,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001649080030857189,
      "loss": 4.9114,
      "mean_token_accuracy": 0.20463947653770448,
      "num_tokens": 147025923.0,
      "step": 67110
    },
    {
      "entropy": 5.7040526390075685,
      "epoch": 6.660877332274712,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016487710404817658,
      "loss": 4.8882,
      "mean_token_accuracy": 0.19614317864179612,
      "num_tokens": 147037208.0,
      "step": 67115
    },
    {
      "entropy": 5.736116123199463,
      "epoch": 6.661373560936879,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016484620774144488,
      "loss": 5.0203,
      "mean_token_accuracy": 0.1946963533759117,
      "num_tokens": 147049798.0,
      "step": 67120
    },
    {
      "entropy": 5.781946897506714,
      "epoch": 6.661869789599047,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016481531416628987,
      "loss": 5.019,
      "mean_token_accuracy": 0.1931630253791809,
      "num_tokens": 147061973.0,
      "step": 67125
    },
    {
      "entropy": 5.683085155487061,
      "epoch": 6.662366018261215,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001647844233234776,
      "loss": 4.8804,
      "mean_token_accuracy": 0.20183707177639007,
      "num_tokens": 147074118.0,
      "step": 67130
    },
    {
      "entropy": 5.676279306411743,
      "epoch": 6.662862246923382,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016475353521377416,
      "loss": 4.8527,
      "mean_token_accuracy": 0.2021913707256317,
      "num_tokens": 147084971.0,
      "step": 67135
    },
    {
      "entropy": 5.655103635787964,
      "epoch": 6.66335847558555,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016472264983794543,
      "loss": 4.8394,
      "mean_token_accuracy": 0.21081102192401885,
      "num_tokens": 147095650.0,
      "step": 67140
    },
    {
      "entropy": 5.757729196548462,
      "epoch": 6.663854704247718,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016469176719675736,
      "loss": 5.0173,
      "mean_token_accuracy": 0.19866426885128022,
      "num_tokens": 147107242.0,
      "step": 67145
    },
    {
      "entropy": 5.64719557762146,
      "epoch": 6.664350932909885,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001646608872909758,
      "loss": 4.8566,
      "mean_token_accuracy": 0.20481985956430435,
      "num_tokens": 147116732.0,
      "step": 67150
    },
    {
      "entropy": 5.750892925262451,
      "epoch": 6.664847161572053,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016463001012136653,
      "loss": 4.944,
      "mean_token_accuracy": 0.19720419049263,
      "num_tokens": 147128848.0,
      "step": 67155
    },
    {
      "entropy": 5.697476053237915,
      "epoch": 6.66534339023422,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016459913568869517,
      "loss": 4.8909,
      "mean_token_accuracy": 0.20273361057043077,
      "num_tokens": 147139419.0,
      "step": 67160
    },
    {
      "entropy": 5.703161668777466,
      "epoch": 6.6658396188963875,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016456826399372735,
      "loss": 4.9163,
      "mean_token_accuracy": 0.20394345223903657,
      "num_tokens": 147150806.0,
      "step": 67165
    },
    {
      "entropy": 5.711654758453369,
      "epoch": 6.666335847558555,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001645373950372285,
      "loss": 4.9185,
      "mean_token_accuracy": 0.20189241766929628,
      "num_tokens": 147161939.0,
      "step": 67170
    },
    {
      "entropy": 5.679683637619019,
      "epoch": 6.6668320762207225,
      "grad_norm": 1.25,
      "learning_rate": 0.00016450652881996427,
      "loss": 4.8954,
      "mean_token_accuracy": 0.20316213220357895,
      "num_tokens": 147173241.0,
      "step": 67175
    },
    {
      "entropy": 5.657502603530884,
      "epoch": 6.66732830488289,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001644756653427001,
      "loss": 4.839,
      "mean_token_accuracy": 0.20898933708667755,
      "num_tokens": 147182370.0,
      "step": 67180
    },
    {
      "entropy": 5.67007303237915,
      "epoch": 6.667824533545057,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016444480460620127,
      "loss": 4.8471,
      "mean_token_accuracy": 0.21057102382183074,
      "num_tokens": 147192872.0,
      "step": 67185
    },
    {
      "entropy": 5.73383150100708,
      "epoch": 6.668320762207225,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001644139466112331,
      "loss": 4.9652,
      "mean_token_accuracy": 0.18798897862434388,
      "num_tokens": 147202741.0,
      "step": 67190
    },
    {
      "entropy": 5.6973536014556885,
      "epoch": 6.668816990869392,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016438309135856073,
      "loss": 4.9702,
      "mean_token_accuracy": 0.19760435223579406,
      "num_tokens": 147214232.0,
      "step": 67195
    },
    {
      "entropy": 5.7164287090301515,
      "epoch": 6.66931321953156,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016435223884894923,
      "loss": 4.9411,
      "mean_token_accuracy": 0.19116836041212082,
      "num_tokens": 147224391.0,
      "step": 67200
    },
    {
      "entropy": 5.624178171157837,
      "epoch": 6.669809448193727,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016432138908316387,
      "loss": 4.8315,
      "mean_token_accuracy": 0.2062985360622406,
      "num_tokens": 147236054.0,
      "step": 67205
    },
    {
      "entropy": 5.731834602355957,
      "epoch": 6.670305676855895,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016429054206196952,
      "loss": 4.9478,
      "mean_token_accuracy": 0.199845090508461,
      "num_tokens": 147246864.0,
      "step": 67210
    },
    {
      "entropy": 5.678431224822998,
      "epoch": 6.670801905518063,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016425969778613125,
      "loss": 4.8628,
      "mean_token_accuracy": 0.20341575741767884,
      "num_tokens": 147257313.0,
      "step": 67215
    },
    {
      "entropy": 5.811432409286499,
      "epoch": 6.67129813418023,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016422885625641382,
      "loss": 5.0469,
      "mean_token_accuracy": 0.19382265359163284,
      "num_tokens": 147267834.0,
      "step": 67220
    },
    {
      "entropy": 5.722662115097046,
      "epoch": 6.671794362842398,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001641980174735821,
      "loss": 4.9488,
      "mean_token_accuracy": 0.20294727087020875,
      "num_tokens": 147280363.0,
      "step": 67225
    },
    {
      "entropy": 5.707086420059204,
      "epoch": 6.672290591504566,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001641671814384007,
      "loss": 4.9309,
      "mean_token_accuracy": 0.19891926646232605,
      "num_tokens": 147292337.0,
      "step": 67230
    },
    {
      "entropy": 5.734258222579956,
      "epoch": 6.672786820166733,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016413634815163442,
      "loss": 4.9529,
      "mean_token_accuracy": 0.19385023564100265,
      "num_tokens": 147303069.0,
      "step": 67235
    },
    {
      "entropy": 5.700078058242798,
      "epoch": 6.6732830488289006,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001641055176140478,
      "loss": 4.9296,
      "mean_token_accuracy": 0.19683742374181748,
      "num_tokens": 147313843.0,
      "step": 67240
    },
    {
      "entropy": 5.684733867645264,
      "epoch": 6.673779277491068,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001640746898264054,
      "loss": 4.8887,
      "mean_token_accuracy": 0.19434906989336015,
      "num_tokens": 147325064.0,
      "step": 67245
    },
    {
      "entropy": 5.701662206649781,
      "epoch": 6.6742755061532355,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001640438647894717,
      "loss": 4.9003,
      "mean_token_accuracy": 0.20297095626592637,
      "num_tokens": 147335283.0,
      "step": 67250
    },
    {
      "entropy": 5.680263710021973,
      "epoch": 6.674771734815403,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001640130425040111,
      "loss": 4.8486,
      "mean_token_accuracy": 0.21018856912851333,
      "num_tokens": 147346204.0,
      "step": 67255
    },
    {
      "entropy": 5.713793420791626,
      "epoch": 6.6752679634775705,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016398222297078774,
      "loss": 4.927,
      "mean_token_accuracy": 0.20182880610227585,
      "num_tokens": 147357102.0,
      "step": 67260
    },
    {
      "entropy": 5.684169340133667,
      "epoch": 6.675764192139738,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016395140619056614,
      "loss": 4.8725,
      "mean_token_accuracy": 0.19808996915817262,
      "num_tokens": 147367669.0,
      "step": 67265
    },
    {
      "entropy": 5.767709445953369,
      "epoch": 6.676260420801905,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00016392059216411032,
      "loss": 4.9964,
      "mean_token_accuracy": 0.19730380326509475,
      "num_tokens": 147378562.0,
      "step": 67270
    },
    {
      "entropy": 5.746854591369629,
      "epoch": 6.676756649464073,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001638897808921845,
      "loss": 4.9423,
      "mean_token_accuracy": 0.19320568144321443,
      "num_tokens": 147388462.0,
      "step": 67275
    },
    {
      "entropy": 5.704023647308349,
      "epoch": 6.67725287812624,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001638589723755527,
      "loss": 4.9087,
      "mean_token_accuracy": 0.19894755780696868,
      "num_tokens": 147400815.0,
      "step": 67280
    },
    {
      "entropy": 5.643922519683838,
      "epoch": 6.677749106788408,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016382816661497893,
      "loss": 4.8365,
      "mean_token_accuracy": 0.20621782392263413,
      "num_tokens": 147411342.0,
      "step": 67285
    },
    {
      "entropy": 5.677250671386719,
      "epoch": 6.678245335450575,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016379736361122696,
      "loss": 4.8512,
      "mean_token_accuracy": 0.2022796243429184,
      "num_tokens": 147422055.0,
      "step": 67290
    },
    {
      "entropy": 5.798306608200074,
      "epoch": 6.678741564112743,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016376656336506085,
      "loss": 4.9963,
      "mean_token_accuracy": 0.19295727759599685,
      "num_tokens": 147432829.0,
      "step": 67295
    },
    {
      "entropy": 5.752338933944702,
      "epoch": 6.679237792774911,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016373576587724428,
      "loss": 4.9247,
      "mean_token_accuracy": 0.2008796140551567,
      "num_tokens": 147443943.0,
      "step": 67300
    },
    {
      "entropy": 5.690648126602173,
      "epoch": 6.679734021437078,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016370497114854087,
      "loss": 4.8537,
      "mean_token_accuracy": 0.20467304140329362,
      "num_tokens": 147455197.0,
      "step": 67305
    },
    {
      "entropy": 5.6710014820098875,
      "epoch": 6.680230250099246,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016367417917971446,
      "loss": 4.9108,
      "mean_token_accuracy": 0.20497473180294037,
      "num_tokens": 147465503.0,
      "step": 67310
    },
    {
      "entropy": 5.74067006111145,
      "epoch": 6.680726478761414,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016364338997152853,
      "loss": 4.9292,
      "mean_token_accuracy": 0.20086637735366822,
      "num_tokens": 147475452.0,
      "step": 67315
    },
    {
      "entropy": 5.655896949768066,
      "epoch": 6.681222707423581,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016361260352474645,
      "loss": 4.9101,
      "mean_token_accuracy": 0.19929031133651734,
      "num_tokens": 147486401.0,
      "step": 67320
    },
    {
      "entropy": 5.710185432434082,
      "epoch": 6.6817189360857485,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016358181984013186,
      "loss": 4.9418,
      "mean_token_accuracy": 0.19722288399934768,
      "num_tokens": 147497315.0,
      "step": 67325
    },
    {
      "entropy": 5.678990840911865,
      "epoch": 6.682215164747916,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016355103891844812,
      "loss": 4.8664,
      "mean_token_accuracy": 0.20195318311452864,
      "num_tokens": 147508153.0,
      "step": 67330
    },
    {
      "entropy": 5.631090259552002,
      "epoch": 6.6827113934100835,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00016352026076045834,
      "loss": 4.7998,
      "mean_token_accuracy": 0.21554650962352753,
      "num_tokens": 147518367.0,
      "step": 67335
    },
    {
      "entropy": 5.675663471221924,
      "epoch": 6.683207622072251,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016348948536692594,
      "loss": 4.9387,
      "mean_token_accuracy": 0.19943732619285584,
      "num_tokens": 147528314.0,
      "step": 67340
    },
    {
      "entropy": 5.719103002548218,
      "epoch": 6.683703850734418,
      "grad_norm": 1.65625,
      "learning_rate": 0.00016345871273861404,
      "loss": 4.8842,
      "mean_token_accuracy": 0.2072012111544609,
      "num_tokens": 147538025.0,
      "step": 67345
    },
    {
      "entropy": 5.751867628097534,
      "epoch": 6.684200079396586,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016342794287628565,
      "loss": 4.9654,
      "mean_token_accuracy": 0.20436645895242692,
      "num_tokens": 147548731.0,
      "step": 67350
    },
    {
      "entropy": 5.697307682037353,
      "epoch": 6.684696308058753,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001633971757807039,
      "loss": 4.9029,
      "mean_token_accuracy": 0.20823503881692887,
      "num_tokens": 147559327.0,
      "step": 67355
    },
    {
      "entropy": 5.7175905227661135,
      "epoch": 6.685192536720921,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016336641145263175,
      "loss": 4.9252,
      "mean_token_accuracy": 0.20122965723276137,
      "num_tokens": 147570267.0,
      "step": 67360
    },
    {
      "entropy": 5.705065774917602,
      "epoch": 6.685688765383088,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016333564989283203,
      "loss": 4.8887,
      "mean_token_accuracy": 0.20371826440095903,
      "num_tokens": 147580930.0,
      "step": 67365
    },
    {
      "entropy": 5.721197938919067,
      "epoch": 6.686184994045256,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016330489110206748,
      "loss": 4.9326,
      "mean_token_accuracy": 0.19345504343509673,
      "num_tokens": 147591986.0,
      "step": 67370
    },
    {
      "entropy": 5.754781866073609,
      "epoch": 6.686681222707423,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016327413508110104,
      "loss": 4.9642,
      "mean_token_accuracy": 0.1910769000649452,
      "num_tokens": 147602991.0,
      "step": 67375
    },
    {
      "entropy": 5.6886039733886715,
      "epoch": 6.687177451369591,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00016324338183069527,
      "loss": 4.9096,
      "mean_token_accuracy": 0.2029825747013092,
      "num_tokens": 147614618.0,
      "step": 67380
    },
    {
      "entropy": 5.6847916603088375,
      "epoch": 6.687673680031759,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016321263135161282,
      "loss": 4.8816,
      "mean_token_accuracy": 0.20555948913097383,
      "num_tokens": 147625509.0,
      "step": 67385
    },
    {
      "entropy": 5.729335021972656,
      "epoch": 6.688169908693926,
      "grad_norm": 1.25,
      "learning_rate": 0.00016318188364461625,
      "loss": 4.9279,
      "mean_token_accuracy": 0.19581844806671142,
      "num_tokens": 147636706.0,
      "step": 67390
    },
    {
      "entropy": 5.649440288543701,
      "epoch": 6.688666137356094,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016315113871046805,
      "loss": 4.8213,
      "mean_token_accuracy": 0.20429671108722686,
      "num_tokens": 147647609.0,
      "step": 67395
    },
    {
      "entropy": 5.706158924102783,
      "epoch": 6.6891623660182615,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001631203965499305,
      "loss": 4.8698,
      "mean_token_accuracy": 0.20161640495061875,
      "num_tokens": 147658532.0,
      "step": 67400
    },
    {
      "entropy": 5.735040283203125,
      "epoch": 6.689658594680429,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001630896571637661,
      "loss": 4.9838,
      "mean_token_accuracy": 0.20176281183958053,
      "num_tokens": 147671159.0,
      "step": 67405
    },
    {
      "entropy": 5.6886646270751955,
      "epoch": 6.6901548233425965,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016305892055273703,
      "loss": 4.8965,
      "mean_token_accuracy": 0.2002809002995491,
      "num_tokens": 147682230.0,
      "step": 67410
    },
    {
      "entropy": 5.678173637390136,
      "epoch": 6.690651052004764,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016302818671760555,
      "loss": 4.9087,
      "mean_token_accuracy": 0.19748791456222534,
      "num_tokens": 147694021.0,
      "step": 67415
    },
    {
      "entropy": 5.693254709243774,
      "epoch": 6.691147280666931,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016299745565913375,
      "loss": 4.9401,
      "mean_token_accuracy": 0.19874971657991408,
      "num_tokens": 147706470.0,
      "step": 67420
    },
    {
      "entropy": 5.7387927055358885,
      "epoch": 6.691643509329099,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016296672737808372,
      "loss": 4.9771,
      "mean_token_accuracy": 0.20663937479257583,
      "num_tokens": 147717787.0,
      "step": 67425
    },
    {
      "entropy": 5.656655883789062,
      "epoch": 6.692139737991266,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016293600187521746,
      "loss": 4.8753,
      "mean_token_accuracy": 0.20424562841653823,
      "num_tokens": 147727759.0,
      "step": 67430
    },
    {
      "entropy": 5.73601975440979,
      "epoch": 6.692635966653434,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001629052791512968,
      "loss": 4.9962,
      "mean_token_accuracy": 0.20185858607292176,
      "num_tokens": 147739719.0,
      "step": 67435
    },
    {
      "entropy": 5.740939664840698,
      "epoch": 6.693132195315601,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016287455920708366,
      "loss": 4.8614,
      "mean_token_accuracy": 0.2061205416917801,
      "num_tokens": 147751947.0,
      "step": 67440
    },
    {
      "entropy": 5.7162261486053465,
      "epoch": 6.693628423977769,
      "grad_norm": 1.375,
      "learning_rate": 0.00016284384204333992,
      "loss": 4.9743,
      "mean_token_accuracy": 0.19442973732948304,
      "num_tokens": 147763195.0,
      "step": 67445
    },
    {
      "entropy": 5.647413063049316,
      "epoch": 6.694124652639936,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016281312766082724,
      "loss": 4.8703,
      "mean_token_accuracy": 0.20280933529138565,
      "num_tokens": 147774526.0,
      "step": 67450
    },
    {
      "entropy": 5.76348385810852,
      "epoch": 6.694620881302104,
      "grad_norm": 1.109375,
      "learning_rate": 0.00016278241606030726,
      "loss": 5.0198,
      "mean_token_accuracy": 0.19570470750331878,
      "num_tokens": 147788014.0,
      "step": 67455
    },
    {
      "entropy": 5.712098836898804,
      "epoch": 6.695117109964271,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001627517072425415,
      "loss": 4.8793,
      "mean_token_accuracy": 0.20563858151435851,
      "num_tokens": 147797700.0,
      "step": 67460
    },
    {
      "entropy": 5.693419933319092,
      "epoch": 6.695613338626439,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00016272100120829147,
      "loss": 4.8861,
      "mean_token_accuracy": 0.20279688835144044,
      "num_tokens": 147807770.0,
      "step": 67465
    },
    {
      "entropy": 5.529650497436523,
      "epoch": 6.696109567288607,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001626902979583187,
      "loss": 4.7653,
      "mean_token_accuracy": 0.2222749650478363,
      "num_tokens": 147820284.0,
      "step": 67470
    },
    {
      "entropy": 5.711925983428955,
      "epoch": 6.696605795950774,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001626595974933846,
      "loss": 4.9342,
      "mean_token_accuracy": 0.20291167348623276,
      "num_tokens": 147831863.0,
      "step": 67475
    },
    {
      "entropy": 5.711427831649781,
      "epoch": 6.697102024612942,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001626288998142504,
      "loss": 4.9275,
      "mean_token_accuracy": 0.2083769917488098,
      "num_tokens": 147843338.0,
      "step": 67480
    },
    {
      "entropy": 5.689284133911133,
      "epoch": 6.6975982532751095,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016259820492167738,
      "loss": 4.8959,
      "mean_token_accuracy": 0.19921900182962418,
      "num_tokens": 147853178.0,
      "step": 67485
    },
    {
      "entropy": 5.618690729141235,
      "epoch": 6.698094481937277,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016256751281642663,
      "loss": 4.8262,
      "mean_token_accuracy": 0.2172811135649681,
      "num_tokens": 147864145.0,
      "step": 67490
    },
    {
      "entropy": 5.6525531768798825,
      "epoch": 6.698590710599444,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016253682349925925,
      "loss": 4.8375,
      "mean_token_accuracy": 0.20244384407997132,
      "num_tokens": 147874302.0,
      "step": 67495
    },
    {
      "entropy": 5.712622022628784,
      "epoch": 6.699086939261612,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00016250613697093625,
      "loss": 4.9429,
      "mean_token_accuracy": 0.20056877732276918,
      "num_tokens": 147885110.0,
      "step": 67500
    },
    {
      "entropy": 5.7104674816131595,
      "epoch": 6.699583167923779,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001624754532322188,
      "loss": 4.9086,
      "mean_token_accuracy": 0.1952034741640091,
      "num_tokens": 147895681.0,
      "step": 67505
    },
    {
      "entropy": 5.7753458499908445,
      "epoch": 6.700079396585947,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00016244477228386756,
      "loss": 4.9885,
      "mean_token_accuracy": 0.19692305773496627,
      "num_tokens": 147906314.0,
      "step": 67510
    },
    {
      "entropy": 5.671217584609986,
      "epoch": 6.700575625248114,
      "grad_norm": 1.140625,
      "learning_rate": 0.00016241409412664348,
      "loss": 4.8477,
      "mean_token_accuracy": 0.21045420318841934,
      "num_tokens": 147918705.0,
      "step": 67515
    },
    {
      "entropy": 5.627885723114014,
      "epoch": 6.701071853910282,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001623834187613073,
      "loss": 4.865,
      "mean_token_accuracy": 0.19593089818954468,
      "num_tokens": 147928583.0,
      "step": 67520
    },
    {
      "entropy": 5.676254177093506,
      "epoch": 6.701568082572449,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016235274618861954,
      "loss": 4.9286,
      "mean_token_accuracy": 0.20530960112810134,
      "num_tokens": 147939683.0,
      "step": 67525
    },
    {
      "entropy": 5.625769186019897,
      "epoch": 6.702064311234617,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016232207640934102,
      "loss": 4.8292,
      "mean_token_accuracy": 0.20326988697052,
      "num_tokens": 147950642.0,
      "step": 67530
    },
    {
      "entropy": 5.655528402328491,
      "epoch": 6.702560539896784,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016229140942423216,
      "loss": 4.8392,
      "mean_token_accuracy": 0.20127130895853043,
      "num_tokens": 147961543.0,
      "step": 67535
    },
    {
      "entropy": 5.636840438842773,
      "epoch": 6.703056768558952,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016226074523405354,
      "loss": 4.7934,
      "mean_token_accuracy": 0.21375545710325242,
      "num_tokens": 147971769.0,
      "step": 67540
    },
    {
      "entropy": 5.6046875476837155,
      "epoch": 6.703552997221119,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001622300838395655,
      "loss": 4.8017,
      "mean_token_accuracy": 0.21020233780145645,
      "num_tokens": 147982199.0,
      "step": 67545
    },
    {
      "entropy": 5.664495420455933,
      "epoch": 6.704049225883287,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00016219942524152843,
      "loss": 4.9366,
      "mean_token_accuracy": 0.19776924699544907,
      "num_tokens": 147992864.0,
      "step": 67550
    },
    {
      "entropy": 5.7185674667358395,
      "epoch": 6.704545454545455,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001621687694407024,
      "loss": 4.9268,
      "mean_token_accuracy": 0.197156223654747,
      "num_tokens": 148003084.0,
      "step": 67555
    },
    {
      "entropy": 5.684404993057251,
      "epoch": 6.705041683207622,
      "grad_norm": 1.578125,
      "learning_rate": 0.00016213811643784787,
      "loss": 4.9011,
      "mean_token_accuracy": 0.20119091421365737,
      "num_tokens": 148013760.0,
      "step": 67560
    },
    {
      "entropy": 5.725073385238647,
      "epoch": 6.70553791186979,
      "grad_norm": 1.46875,
      "learning_rate": 0.00016210746623372473,
      "loss": 4.9183,
      "mean_token_accuracy": 0.19879712164402008,
      "num_tokens": 148024076.0,
      "step": 67565
    },
    {
      "entropy": 5.686135578155517,
      "epoch": 6.7060341405319575,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016207681882909333,
      "loss": 4.9287,
      "mean_token_accuracy": 0.20115216672420502,
      "num_tokens": 148034935.0,
      "step": 67570
    },
    {
      "entropy": 5.744079542160034,
      "epoch": 6.706530369194125,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016204617422471346,
      "loss": 4.938,
      "mean_token_accuracy": 0.20097264051437377,
      "num_tokens": 148045916.0,
      "step": 67575
    },
    {
      "entropy": 5.661976861953735,
      "epoch": 6.707026597856292,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016201553242134508,
      "loss": 4.921,
      "mean_token_accuracy": 0.20124263912439347,
      "num_tokens": 148055932.0,
      "step": 67580
    },
    {
      "entropy": 5.7111608505249025,
      "epoch": 6.70752282651846,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016198489341974793,
      "loss": 4.9177,
      "mean_token_accuracy": 0.20311158448457717,
      "num_tokens": 148067092.0,
      "step": 67585
    },
    {
      "entropy": 5.697144985198975,
      "epoch": 6.708019055180627,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016195425722068202,
      "loss": 4.9451,
      "mean_token_accuracy": 0.19667662233114241,
      "num_tokens": 148079499.0,
      "step": 67590
    },
    {
      "entropy": 5.670868158340454,
      "epoch": 6.708515283842795,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016192362382490692,
      "loss": 4.8631,
      "mean_token_accuracy": 0.20703083872795106,
      "num_tokens": 148090326.0,
      "step": 67595
    },
    {
      "entropy": 5.742550802230835,
      "epoch": 6.709011512504962,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00016189299323318218,
      "loss": 4.9782,
      "mean_token_accuracy": 0.19752060770988464,
      "num_tokens": 148101605.0,
      "step": 67600
    },
    {
      "entropy": 5.6586301803588865,
      "epoch": 6.70950774116713,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001618623654462676,
      "loss": 4.8567,
      "mean_token_accuracy": 0.20401072204113008,
      "num_tokens": 148112649.0,
      "step": 67605
    },
    {
      "entropy": 5.697345829010009,
      "epoch": 6.710003969829297,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00016183174046492256,
      "loss": 4.92,
      "mean_token_accuracy": 0.20062589794397354,
      "num_tokens": 148123211.0,
      "step": 67610
    },
    {
      "entropy": 5.688342905044555,
      "epoch": 6.710500198491465,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001618011182899064,
      "loss": 4.8912,
      "mean_token_accuracy": 0.20319671481847762,
      "num_tokens": 148133487.0,
      "step": 67615
    },
    {
      "entropy": 5.706612205505371,
      "epoch": 6.710996427153632,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016177049892197867,
      "loss": 4.9533,
      "mean_token_accuracy": 0.1990778550505638,
      "num_tokens": 148143437.0,
      "step": 67620
    },
    {
      "entropy": 5.668582916259766,
      "epoch": 6.7114926558158,
      "grad_norm": 1.171875,
      "learning_rate": 0.00016173988236189863,
      "loss": 4.8923,
      "mean_token_accuracy": 0.20442293882369994,
      "num_tokens": 148155793.0,
      "step": 67625
    },
    {
      "entropy": 5.682475471496582,
      "epoch": 6.711988884477967,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016170926861042528,
      "loss": 4.9078,
      "mean_token_accuracy": 0.20112231075763704,
      "num_tokens": 148166386.0,
      "step": 67630
    },
    {
      "entropy": 5.702221441268921,
      "epoch": 6.712485113140135,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001616786576683181,
      "loss": 4.95,
      "mean_token_accuracy": 0.20209379643201827,
      "num_tokens": 148177219.0,
      "step": 67635
    },
    {
      "entropy": 5.645877552032471,
      "epoch": 6.712981341802303,
      "grad_norm": 1.390625,
      "learning_rate": 0.000161648049536336,
      "loss": 4.9443,
      "mean_token_accuracy": 0.19891990721225739,
      "num_tokens": 148188629.0,
      "step": 67640
    },
    {
      "entropy": 5.664719533920288,
      "epoch": 6.71347757046447,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016161744421523797,
      "loss": 4.8974,
      "mean_token_accuracy": 0.20921050906181335,
      "num_tokens": 148199471.0,
      "step": 67645
    },
    {
      "entropy": 5.6936439037322994,
      "epoch": 6.713973799126638,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016158684170578308,
      "loss": 4.8804,
      "mean_token_accuracy": 0.20093452036380768,
      "num_tokens": 148209976.0,
      "step": 67650
    },
    {
      "entropy": 5.627262115478516,
      "epoch": 6.714470027788805,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001615562420087301,
      "loss": 4.8502,
      "mean_token_accuracy": 0.2078617125749588,
      "num_tokens": 148220711.0,
      "step": 67655
    },
    {
      "entropy": 5.6497495651245115,
      "epoch": 6.714966256450973,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016152564512483786,
      "loss": 4.8439,
      "mean_token_accuracy": 0.19784566164016723,
      "num_tokens": 148230436.0,
      "step": 67660
    },
    {
      "entropy": 5.639988708496094,
      "epoch": 6.71546248511314,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016149505105486506,
      "loss": 4.9263,
      "mean_token_accuracy": 0.1958985760807991,
      "num_tokens": 148242445.0,
      "step": 67665
    },
    {
      "entropy": 5.716557836532592,
      "epoch": 6.715958713775308,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016146445979957049,
      "loss": 4.9298,
      "mean_token_accuracy": 0.19506614357233049,
      "num_tokens": 148253436.0,
      "step": 67670
    },
    {
      "entropy": 5.749986886978149,
      "epoch": 6.716454942437475,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016143387135971257,
      "loss": 4.9644,
      "mean_token_accuracy": 0.19662521332502364,
      "num_tokens": 148264011.0,
      "step": 67675
    },
    {
      "entropy": 5.68887767791748,
      "epoch": 6.716951171099643,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016140328573605001,
      "loss": 4.8979,
      "mean_token_accuracy": 0.2099531650543213,
      "num_tokens": 148275327.0,
      "step": 67680
    },
    {
      "entropy": 5.698068904876709,
      "epoch": 6.71744739976181,
      "grad_norm": 1.375,
      "learning_rate": 0.00016137270292934118,
      "loss": 4.9445,
      "mean_token_accuracy": 0.19434602558612823,
      "num_tokens": 148287580.0,
      "step": 67685
    },
    {
      "entropy": 5.775600528717041,
      "epoch": 6.717943628423978,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001613421229403445,
      "loss": 4.9944,
      "mean_token_accuracy": 0.19864246994256973,
      "num_tokens": 148298087.0,
      "step": 67690
    },
    {
      "entropy": 5.724307441711426,
      "epoch": 6.718439857086145,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016131154576981815,
      "loss": 4.9419,
      "mean_token_accuracy": 0.20582115948200225,
      "num_tokens": 148308494.0,
      "step": 67695
    },
    {
      "entropy": 5.755790853500367,
      "epoch": 6.718936085748313,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001612809714185205,
      "loss": 5.0836,
      "mean_token_accuracy": 0.19199766367673873,
      "num_tokens": 148319298.0,
      "step": 67700
    },
    {
      "entropy": 5.704002714157104,
      "epoch": 6.71943231441048,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001612503998872098,
      "loss": 4.8794,
      "mean_token_accuracy": 0.20194277167320251,
      "num_tokens": 148331265.0,
      "step": 67705
    },
    {
      "entropy": 5.752448177337646,
      "epoch": 6.719928543072648,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016121983117664408,
      "loss": 4.9505,
      "mean_token_accuracy": 0.20166206061840058,
      "num_tokens": 148343252.0,
      "step": 67710
    },
    {
      "entropy": 5.760645198822021,
      "epoch": 6.720424771734815,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016118926528758138,
      "loss": 4.966,
      "mean_token_accuracy": 0.20104589462280273,
      "num_tokens": 148353628.0,
      "step": 67715
    },
    {
      "entropy": 5.721792697906494,
      "epoch": 6.720921000396983,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016115870222077963,
      "loss": 4.9404,
      "mean_token_accuracy": 0.19998917132616043,
      "num_tokens": 148364612.0,
      "step": 67720
    },
    {
      "entropy": 5.667470455169678,
      "epoch": 6.721417229059151,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016112814197699672,
      "loss": 4.8537,
      "mean_token_accuracy": 0.20233023017644883,
      "num_tokens": 148374376.0,
      "step": 67725
    },
    {
      "entropy": 5.6937634468078615,
      "epoch": 6.7219134577213175,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016109758455699053,
      "loss": 4.9362,
      "mean_token_accuracy": 0.2056390479207039,
      "num_tokens": 148385800.0,
      "step": 67730
    },
    {
      "entropy": 5.6728630542755125,
      "epoch": 6.722409686383486,
      "grad_norm": 1.375,
      "learning_rate": 0.0001610670299615188,
      "loss": 4.8773,
      "mean_token_accuracy": 0.20713665336370468,
      "num_tokens": 148396799.0,
      "step": 67735
    },
    {
      "entropy": 5.712523603439331,
      "epoch": 6.722905915045653,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016103647819133924,
      "loss": 4.9041,
      "mean_token_accuracy": 0.19840124994516373,
      "num_tokens": 148407789.0,
      "step": 67740
    },
    {
      "entropy": 5.672275066375732,
      "epoch": 6.723402143707821,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016100592924720946,
      "loss": 4.8932,
      "mean_token_accuracy": 0.20237582921981812,
      "num_tokens": 148418638.0,
      "step": 67745
    },
    {
      "entropy": 5.600052165985107,
      "epoch": 6.723898372369988,
      "grad_norm": 1.375,
      "learning_rate": 0.00016097538312988705,
      "loss": 4.806,
      "mean_token_accuracy": 0.20268456935882567,
      "num_tokens": 148428776.0,
      "step": 67750
    },
    {
      "entropy": 5.714335012435913,
      "epoch": 6.724394601032156,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001609448398401294,
      "loss": 4.9488,
      "mean_token_accuracy": 0.196196511387825,
      "num_tokens": 148438601.0,
      "step": 67755
    },
    {
      "entropy": 5.695247459411621,
      "epoch": 6.724890829694323,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016091429937869386,
      "loss": 4.8657,
      "mean_token_accuracy": 0.21229604333639146,
      "num_tokens": 148450233.0,
      "step": 67760
    },
    {
      "entropy": 5.712052917480468,
      "epoch": 6.725387058356491,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001608837617463379,
      "loss": 4.8891,
      "mean_token_accuracy": 0.20671789944171906,
      "num_tokens": 148461834.0,
      "step": 67765
    },
    {
      "entropy": 5.666047239303589,
      "epoch": 6.725883287018658,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016085322694381883,
      "loss": 4.8845,
      "mean_token_accuracy": 0.19562235623598098,
      "num_tokens": 148472612.0,
      "step": 67770
    },
    {
      "entropy": 5.746413087844848,
      "epoch": 6.726379515680826,
      "grad_norm": 1.5,
      "learning_rate": 0.00016082269497189373,
      "loss": 4.9882,
      "mean_token_accuracy": 0.19776758998632432,
      "num_tokens": 148482986.0,
      "step": 67775
    },
    {
      "entropy": 5.7287109851837155,
      "epoch": 6.726875744342993,
      "grad_norm": 1.25,
      "learning_rate": 0.00016079216583131977,
      "loss": 4.9506,
      "mean_token_accuracy": 0.20383272022008897,
      "num_tokens": 148495373.0,
      "step": 67780
    },
    {
      "entropy": 5.724574089050293,
      "epoch": 6.727371973005161,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000160761639522854,
      "loss": 4.9256,
      "mean_token_accuracy": 0.1960126653313637,
      "num_tokens": 148506218.0,
      "step": 67785
    },
    {
      "entropy": 5.696036720275879,
      "epoch": 6.727868201667328,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016073111604725338,
      "loss": 4.8799,
      "mean_token_accuracy": 0.200698646903038,
      "num_tokens": 148516424.0,
      "step": 67790
    },
    {
      "entropy": 5.719060468673706,
      "epoch": 6.728364430329496,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001607005954052748,
      "loss": 4.9526,
      "mean_token_accuracy": 0.20679884403944016,
      "num_tokens": 148527357.0,
      "step": 67795
    },
    {
      "entropy": 5.689233636856079,
      "epoch": 6.728860658991663,
      "grad_norm": 1.53125,
      "learning_rate": 0.00016067007759767527,
      "loss": 4.8409,
      "mean_token_accuracy": 0.206057608127594,
      "num_tokens": 148536989.0,
      "step": 67800
    },
    {
      "entropy": 5.720527410507202,
      "epoch": 6.7293568876538306,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016063956262521147,
      "loss": 4.9532,
      "mean_token_accuracy": 0.19522683024406434,
      "num_tokens": 148547382.0,
      "step": 67805
    },
    {
      "entropy": 5.672994709014892,
      "epoch": 6.729853116315998,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016060905048864005,
      "loss": 4.8914,
      "mean_token_accuracy": 0.211470590531826,
      "num_tokens": 148557639.0,
      "step": 67810
    },
    {
      "entropy": 5.710834074020386,
      "epoch": 6.7303493449781655,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001605785411887177,
      "loss": 4.9105,
      "mean_token_accuracy": 0.2008835643529892,
      "num_tokens": 148569728.0,
      "step": 67815
    },
    {
      "entropy": 5.796010112762451,
      "epoch": 6.730845573640334,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001605480347262009,
      "loss": 4.9847,
      "mean_token_accuracy": 0.19830338954925536,
      "num_tokens": 148582953.0,
      "step": 67820
    },
    {
      "entropy": 5.709751129150391,
      "epoch": 6.731341802302501,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001605175311018463,
      "loss": 4.8802,
      "mean_token_accuracy": 0.20716699361801147,
      "num_tokens": 148593132.0,
      "step": 67825
    },
    {
      "entropy": 5.688250207901001,
      "epoch": 6.731838030964669,
      "grad_norm": 1.546875,
      "learning_rate": 0.00016048703031641015,
      "loss": 4.8206,
      "mean_token_accuracy": 0.21418639570474624,
      "num_tokens": 148604597.0,
      "step": 67830
    },
    {
      "entropy": 5.74671483039856,
      "epoch": 6.732334259626836,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016045653237064899,
      "loss": 4.898,
      "mean_token_accuracy": 0.20655460208654403,
      "num_tokens": 148616238.0,
      "step": 67835
    },
    {
      "entropy": 5.732238340377807,
      "epoch": 6.732830488289004,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00016042603726531898,
      "loss": 4.8962,
      "mean_token_accuracy": 0.20057521015405655,
      "num_tokens": 148626487.0,
      "step": 67840
    },
    {
      "entropy": 5.695968437194824,
      "epoch": 6.733326716951171,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001603955450011764,
      "loss": 4.9666,
      "mean_token_accuracy": 0.19892832338809968,
      "num_tokens": 148636976.0,
      "step": 67845
    },
    {
      "entropy": 5.690607690811158,
      "epoch": 6.733822945613339,
      "grad_norm": 1.484375,
      "learning_rate": 0.00016036505557897723,
      "loss": 4.9652,
      "mean_token_accuracy": 0.20061546862125396,
      "num_tokens": 148648521.0,
      "step": 67850
    },
    {
      "entropy": 5.700825929641724,
      "epoch": 6.734319174275506,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00016033456899947776,
      "loss": 4.9291,
      "mean_token_accuracy": 0.19907562881708146,
      "num_tokens": 148660468.0,
      "step": 67855
    },
    {
      "entropy": 5.752671241760254,
      "epoch": 6.734815402937674,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016030408526343382,
      "loss": 5.0146,
      "mean_token_accuracy": 0.19087569713592528,
      "num_tokens": 148671861.0,
      "step": 67860
    },
    {
      "entropy": 5.747940397262573,
      "epoch": 6.735311631599841,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001602736043716015,
      "loss": 4.9509,
      "mean_token_accuracy": 0.19626833796501159,
      "num_tokens": 148682511.0,
      "step": 67865
    },
    {
      "entropy": 5.727700090408325,
      "epoch": 6.735807860262009,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016024312632473658,
      "loss": 4.8955,
      "mean_token_accuracy": 0.20303464382886888,
      "num_tokens": 148693131.0,
      "step": 67870
    },
    {
      "entropy": 5.699342679977417,
      "epoch": 6.736304088924176,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016021265112359486,
      "loss": 4.8962,
      "mean_token_accuracy": 0.20476410537958145,
      "num_tokens": 148703926.0,
      "step": 67875
    },
    {
      "entropy": 5.587106323242187,
      "epoch": 6.736800317586344,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016018217876893193,
      "loss": 4.8445,
      "mean_token_accuracy": 0.2107957124710083,
      "num_tokens": 148715224.0,
      "step": 67880
    },
    {
      "entropy": 5.676707744598389,
      "epoch": 6.737296546248511,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016015170926150367,
      "loss": 4.926,
      "mean_token_accuracy": 0.19657054990530015,
      "num_tokens": 148726110.0,
      "step": 67885
    },
    {
      "entropy": 5.662082195281982,
      "epoch": 6.7377927749106785,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016012124260206557,
      "loss": 4.8769,
      "mean_token_accuracy": 0.20394660234451295,
      "num_tokens": 148737462.0,
      "step": 67890
    },
    {
      "entropy": 5.700879144668579,
      "epoch": 6.738289003572846,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016009077879137296,
      "loss": 4.8606,
      "mean_token_accuracy": 0.20310295671224593,
      "num_tokens": 148748157.0,
      "step": 67895
    },
    {
      "entropy": 5.673614597320556,
      "epoch": 6.7387852322350135,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016006031783018153,
      "loss": 4.8415,
      "mean_token_accuracy": 0.19838490933179856,
      "num_tokens": 148759376.0,
      "step": 67900
    },
    {
      "entropy": 5.763548994064331,
      "epoch": 6.739281460897182,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001600298597192466,
      "loss": 4.9681,
      "mean_token_accuracy": 0.20544423013925553,
      "num_tokens": 148771552.0,
      "step": 67905
    },
    {
      "entropy": 5.6035614013671875,
      "epoch": 6.739777689559349,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015999940445932326,
      "loss": 4.7724,
      "mean_token_accuracy": 0.21699840128421782,
      "num_tokens": 148783131.0,
      "step": 67910
    },
    {
      "entropy": 5.695402526855469,
      "epoch": 6.740273918221517,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000159968952051167,
      "loss": 4.9325,
      "mean_token_accuracy": 0.19829317629337312,
      "num_tokens": 148793597.0,
      "step": 67915
    },
    {
      "entropy": 5.651641035079956,
      "epoch": 6.740770146883684,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015993850249553288,
      "loss": 4.8625,
      "mean_token_accuracy": 0.1993645802140236,
      "num_tokens": 148804456.0,
      "step": 67920
    },
    {
      "entropy": 5.671404838562012,
      "epoch": 6.741266375545852,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015990805579317585,
      "loss": 4.9038,
      "mean_token_accuracy": 0.19651215970516206,
      "num_tokens": 148815396.0,
      "step": 67925
    },
    {
      "entropy": 5.7349132061004635,
      "epoch": 6.741762604208019,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015987761194485113,
      "loss": 5.0217,
      "mean_token_accuracy": 0.1884700283408165,
      "num_tokens": 148827227.0,
      "step": 67930
    },
    {
      "entropy": 5.648659324645996,
      "epoch": 6.742258832870187,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00015984717095131353,
      "loss": 4.8437,
      "mean_token_accuracy": 0.20207817405462264,
      "num_tokens": 148838572.0,
      "step": 67935
    },
    {
      "entropy": 5.708712196350097,
      "epoch": 6.742755061532354,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015981673281331794,
      "loss": 4.9466,
      "mean_token_accuracy": 0.19922202527523042,
      "num_tokens": 148849843.0,
      "step": 67940
    },
    {
      "entropy": 5.685940599441528,
      "epoch": 6.743251290194522,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015978629753161922,
      "loss": 4.9194,
      "mean_token_accuracy": 0.2019764244556427,
      "num_tokens": 148860519.0,
      "step": 67945
    },
    {
      "entropy": 5.7670331478118895,
      "epoch": 6.743747518856689,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015975586510697207,
      "loss": 4.9696,
      "mean_token_accuracy": 0.1902385801076889,
      "num_tokens": 148870101.0,
      "step": 67950
    },
    {
      "entropy": 5.672207880020141,
      "epoch": 6.744243747518857,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015972543554013103,
      "loss": 4.8604,
      "mean_token_accuracy": 0.20599315613508223,
      "num_tokens": 148881349.0,
      "step": 67955
    },
    {
      "entropy": 5.721887159347534,
      "epoch": 6.744739976181024,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001596950088318509,
      "loss": 4.9099,
      "mean_token_accuracy": 0.19943522810935974,
      "num_tokens": 148894103.0,
      "step": 67960
    },
    {
      "entropy": 5.729144144058227,
      "epoch": 6.7452362048431915,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000159664584982886,
      "loss": 4.9381,
      "mean_token_accuracy": 0.2010687068104744,
      "num_tokens": 148905107.0,
      "step": 67965
    },
    {
      "entropy": 5.664491653442383,
      "epoch": 6.745732433505359,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015963416399399094,
      "loss": 4.8849,
      "mean_token_accuracy": 0.2036600738763809,
      "num_tokens": 148915403.0,
      "step": 67970
    },
    {
      "entropy": 5.714465665817261,
      "epoch": 6.7462286621675265,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00015960374586592002,
      "loss": 4.9502,
      "mean_token_accuracy": 0.1985227271914482,
      "num_tokens": 148925215.0,
      "step": 67975
    },
    {
      "entropy": 5.646738767623901,
      "epoch": 6.746724890829694,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001595733305994276,
      "loss": 4.8428,
      "mean_token_accuracy": 0.20352879166603088,
      "num_tokens": 148935064.0,
      "step": 67980
    },
    {
      "entropy": 5.695181083679199,
      "epoch": 6.747221119491861,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015954291819526782,
      "loss": 4.9009,
      "mean_token_accuracy": 0.2007752999663353,
      "num_tokens": 148945025.0,
      "step": 67985
    },
    {
      "entropy": 5.688970327377319,
      "epoch": 6.74771734815403,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015951250865419477,
      "loss": 4.9368,
      "mean_token_accuracy": 0.19473817348480224,
      "num_tokens": 148955299.0,
      "step": 67990
    },
    {
      "entropy": 5.663405799865723,
      "epoch": 6.748213576816197,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001594821019769627,
      "loss": 4.8906,
      "mean_token_accuracy": 0.20058441311120986,
      "num_tokens": 148965906.0,
      "step": 67995
    },
    {
      "entropy": 5.686687755584717,
      "epoch": 6.748709805478365,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001594516981643256,
      "loss": 4.9617,
      "mean_token_accuracy": 0.19658115953207017,
      "num_tokens": 148976307.0,
      "step": 68000
    },
    {
      "entropy": 5.649754333496094,
      "epoch": 6.749206034140532,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015942129721703747,
      "loss": 4.8788,
      "mean_token_accuracy": 0.20055259317159652,
      "num_tokens": 148986046.0,
      "step": 68005
    },
    {
      "entropy": 5.747590065002441,
      "epoch": 6.7497022628027,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001593908991358521,
      "loss": 4.8777,
      "mean_token_accuracy": 0.20033055990934373,
      "num_tokens": 148996403.0,
      "step": 68010
    },
    {
      "entropy": 5.685548496246338,
      "epoch": 6.750198491464867,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001593605039215233,
      "loss": 4.8801,
      "mean_token_accuracy": 0.20172174125909806,
      "num_tokens": 149006964.0,
      "step": 68015
    },
    {
      "entropy": 5.7460168361663815,
      "epoch": 6.750694720127035,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015933011157480477,
      "loss": 5.0931,
      "mean_token_accuracy": 0.19552399218082428,
      "num_tokens": 149017434.0,
      "step": 68020
    },
    {
      "entropy": 5.639495086669922,
      "epoch": 6.751190948789202,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015929972209645016,
      "loss": 4.9055,
      "mean_token_accuracy": 0.19698176085948943,
      "num_tokens": 149028697.0,
      "step": 68025
    },
    {
      "entropy": 5.680973100662231,
      "epoch": 6.75168717745137,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015926933548721328,
      "loss": 4.9015,
      "mean_token_accuracy": 0.1986715763807297,
      "num_tokens": 149039616.0,
      "step": 68030
    },
    {
      "entropy": 5.737221050262451,
      "epoch": 6.752183406113537,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00015923895174784747,
      "loss": 4.983,
      "mean_token_accuracy": 0.19761675000190734,
      "num_tokens": 149049803.0,
      "step": 68035
    },
    {
      "entropy": 5.668868684768677,
      "epoch": 6.7526796347757045,
      "grad_norm": 1.375,
      "learning_rate": 0.00015920857087910623,
      "loss": 4.8822,
      "mean_token_accuracy": 0.2041816905140877,
      "num_tokens": 149061558.0,
      "step": 68040
    },
    {
      "entropy": 5.68905348777771,
      "epoch": 6.753175863437872,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015917819288174294,
      "loss": 4.9271,
      "mean_token_accuracy": 0.19993999749422073,
      "num_tokens": 149072009.0,
      "step": 68045
    },
    {
      "entropy": 5.682847023010254,
      "epoch": 6.7536720921000395,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015914781775651086,
      "loss": 4.8699,
      "mean_token_accuracy": 0.20461785644292832,
      "num_tokens": 149084054.0,
      "step": 68050
    },
    {
      "entropy": 5.708584022521973,
      "epoch": 6.754168320762207,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015911744550416317,
      "loss": 4.8852,
      "mean_token_accuracy": 0.2012624502182007,
      "num_tokens": 149093907.0,
      "step": 68055
    },
    {
      "entropy": 5.677396011352539,
      "epoch": 6.754664549424374,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015908707612545314,
      "loss": 4.8639,
      "mean_token_accuracy": 0.2073346793651581,
      "num_tokens": 149107351.0,
      "step": 68060
    },
    {
      "entropy": 5.72088475227356,
      "epoch": 6.755160778086542,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00015905670962113394,
      "loss": 4.9494,
      "mean_token_accuracy": 0.1951567217707634,
      "num_tokens": 149119460.0,
      "step": 68065
    },
    {
      "entropy": 5.687888860702515,
      "epoch": 6.755657006748709,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015902634599195855,
      "loss": 4.8155,
      "mean_token_accuracy": 0.21067144572734833,
      "num_tokens": 149130485.0,
      "step": 68070
    },
    {
      "entropy": 5.724873733520508,
      "epoch": 6.756153235410878,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001589959852386798,
      "loss": 4.9074,
      "mean_token_accuracy": 0.19889366626739502,
      "num_tokens": 149141244.0,
      "step": 68075
    },
    {
      "entropy": 5.6856283187866214,
      "epoch": 6.756649464073045,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001589656273620507,
      "loss": 4.8827,
      "mean_token_accuracy": 0.2087772771716118,
      "num_tokens": 149152056.0,
      "step": 68080
    },
    {
      "entropy": 5.695381355285645,
      "epoch": 6.757145692735213,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015893527236282389,
      "loss": 4.9427,
      "mean_token_accuracy": 0.19824261963367462,
      "num_tokens": 149163019.0,
      "step": 68085
    },
    {
      "entropy": 5.700993299484253,
      "epoch": 6.75764192139738,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015890492024175223,
      "loss": 4.9378,
      "mean_token_accuracy": 0.2001590073108673,
      "num_tokens": 149174015.0,
      "step": 68090
    },
    {
      "entropy": 5.655029153823852,
      "epoch": 6.758138150059548,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001588745709995885,
      "loss": 4.9128,
      "mean_token_accuracy": 0.2128644272685051,
      "num_tokens": 149185545.0,
      "step": 68095
    },
    {
      "entropy": 5.718153953552246,
      "epoch": 6.758634378721715,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015884422463708516,
      "loss": 4.8996,
      "mean_token_accuracy": 0.20410332679748536,
      "num_tokens": 149195444.0,
      "step": 68100
    },
    {
      "entropy": 5.736576938629151,
      "epoch": 6.759130607383883,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015881388115499475,
      "loss": 4.9194,
      "mean_token_accuracy": 0.20929653346538543,
      "num_tokens": 149207135.0,
      "step": 68105
    },
    {
      "entropy": 5.70635838508606,
      "epoch": 6.75962683604605,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00015878354055406973,
      "loss": 4.8488,
      "mean_token_accuracy": 0.20621096342802048,
      "num_tokens": 149216862.0,
      "step": 68110
    },
    {
      "entropy": 5.674902725219726,
      "epoch": 6.7601230647082176,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00015875320283506244,
      "loss": 4.8894,
      "mean_token_accuracy": 0.2052169680595398,
      "num_tokens": 149227477.0,
      "step": 68115
    },
    {
      "entropy": 5.565043878555298,
      "epoch": 6.760619293370385,
      "grad_norm": 1.203125,
      "learning_rate": 0.00015872286799872528,
      "loss": 4.8033,
      "mean_token_accuracy": 0.22175778746604918,
      "num_tokens": 149239705.0,
      "step": 68120
    },
    {
      "entropy": 5.7044227600097654,
      "epoch": 6.7611155220325525,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015869253604581035,
      "loss": 4.9246,
      "mean_token_accuracy": 0.19814592748880386,
      "num_tokens": 149250807.0,
      "step": 68125
    },
    {
      "entropy": 5.684468412399292,
      "epoch": 6.76161175069472,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015866220697707,
      "loss": 4.9017,
      "mean_token_accuracy": 0.1984074518084526,
      "num_tokens": 149261498.0,
      "step": 68130
    },
    {
      "entropy": 5.707872772216797,
      "epoch": 6.7621079793568875,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001586318807932562,
      "loss": 4.9003,
      "mean_token_accuracy": 0.20302305221557618,
      "num_tokens": 149271561.0,
      "step": 68135
    },
    {
      "entropy": 5.672442674636841,
      "epoch": 6.762604208019055,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015860155749512105,
      "loss": 4.9295,
      "mean_token_accuracy": 0.2030763477087021,
      "num_tokens": 149282562.0,
      "step": 68140
    },
    {
      "entropy": 5.661020040512085,
      "epoch": 6.763100436681222,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00015857123708341636,
      "loss": 4.8759,
      "mean_token_accuracy": 0.20814913362264634,
      "num_tokens": 149293535.0,
      "step": 68145
    },
    {
      "entropy": 5.735895395278931,
      "epoch": 6.76359666534339,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015854091955889416,
      "loss": 4.9153,
      "mean_token_accuracy": 0.19911184906959534,
      "num_tokens": 149303934.0,
      "step": 68150
    },
    {
      "entropy": 5.704564380645752,
      "epoch": 6.764092894005557,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015851060492230617,
      "loss": 4.8686,
      "mean_token_accuracy": 0.20234696567058563,
      "num_tokens": 149315121.0,
      "step": 68155
    },
    {
      "entropy": 5.646079635620117,
      "epoch": 6.764589122667726,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001584802931744042,
      "loss": 4.871,
      "mean_token_accuracy": 0.20150844603776932,
      "num_tokens": 149326950.0,
      "step": 68160
    },
    {
      "entropy": 5.741142225265503,
      "epoch": 6.765085351329893,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015844998431593997,
      "loss": 4.9975,
      "mean_token_accuracy": 0.19596098959445954,
      "num_tokens": 149337559.0,
      "step": 68165
    },
    {
      "entropy": 5.609616422653199,
      "epoch": 6.765581579992061,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001584196783476649,
      "loss": 4.8294,
      "mean_token_accuracy": 0.20665157437324524,
      "num_tokens": 149349560.0,
      "step": 68170
    },
    {
      "entropy": 5.7004491806030275,
      "epoch": 6.766077808654228,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015838937527033053,
      "loss": 4.9119,
      "mean_token_accuracy": 0.20275452435016633,
      "num_tokens": 149360780.0,
      "step": 68175
    },
    {
      "entropy": 5.641674900054932,
      "epoch": 6.766574037316396,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001583590750846885,
      "loss": 4.8208,
      "mean_token_accuracy": 0.20620876997709275,
      "num_tokens": 149371369.0,
      "step": 68180
    },
    {
      "entropy": 5.672190761566162,
      "epoch": 6.767070265978563,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015832877779149005,
      "loss": 4.9024,
      "mean_token_accuracy": 0.20064588040113449,
      "num_tokens": 149382773.0,
      "step": 68185
    },
    {
      "entropy": 5.63791184425354,
      "epoch": 6.767566494640731,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015829848339148642,
      "loss": 4.8279,
      "mean_token_accuracy": 0.20007691085338591,
      "num_tokens": 149393237.0,
      "step": 68190
    },
    {
      "entropy": 5.690739250183105,
      "epoch": 6.768062723302898,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015826819188542902,
      "loss": 4.8835,
      "mean_token_accuracy": 0.20448748767375946,
      "num_tokens": 149404345.0,
      "step": 68195
    },
    {
      "entropy": 5.656376457214355,
      "epoch": 6.7685589519650655,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015823790327406883,
      "loss": 4.8569,
      "mean_token_accuracy": 0.20309916138648987,
      "num_tokens": 149415789.0,
      "step": 68200
    },
    {
      "entropy": 5.66817798614502,
      "epoch": 6.769055180627233,
      "grad_norm": 1.46875,
      "learning_rate": 0.00015820761755815712,
      "loss": 4.8948,
      "mean_token_accuracy": 0.2057142361998558,
      "num_tokens": 149426493.0,
      "step": 68205
    },
    {
      "entropy": 5.694105529785157,
      "epoch": 6.7695514092894005,
      "grad_norm": 1.375,
      "learning_rate": 0.00015817733473844488,
      "loss": 4.9164,
      "mean_token_accuracy": 0.20288937538862228,
      "num_tokens": 149437199.0,
      "step": 68210
    },
    {
      "entropy": 5.640878534317016,
      "epoch": 6.770047637951568,
      "grad_norm": 1.46875,
      "learning_rate": 0.00015814705481568292,
      "loss": 4.8692,
      "mean_token_accuracy": 0.19855777472257613,
      "num_tokens": 149448079.0,
      "step": 68215
    },
    {
      "entropy": 5.661366033554077,
      "epoch": 6.770543866613735,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015811677779062217,
      "loss": 4.9193,
      "mean_token_accuracy": 0.19531808495521547,
      "num_tokens": 149458293.0,
      "step": 68220
    },
    {
      "entropy": 5.716139221191407,
      "epoch": 6.771040095275903,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015808650366401356,
      "loss": 4.8874,
      "mean_token_accuracy": 0.208372363448143,
      "num_tokens": 149468818.0,
      "step": 68225
    },
    {
      "entropy": 5.630776596069336,
      "epoch": 6.77153632393807,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015805623243660765,
      "loss": 4.8016,
      "mean_token_accuracy": 0.20278911739587785,
      "num_tokens": 149480719.0,
      "step": 68230
    },
    {
      "entropy": 5.729862785339355,
      "epoch": 6.772032552600238,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015802596410915525,
      "loss": 4.9648,
      "mean_token_accuracy": 0.19472062438726426,
      "num_tokens": 149491168.0,
      "step": 68235
    },
    {
      "entropy": 5.6450670719146725,
      "epoch": 6.772528781262405,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001579956986824069,
      "loss": 4.8586,
      "mean_token_accuracy": 0.20732536017894745,
      "num_tokens": 149501432.0,
      "step": 68240
    },
    {
      "entropy": 5.66961784362793,
      "epoch": 6.773025009924574,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015796543615711312,
      "loss": 4.9152,
      "mean_token_accuracy": 0.1964913412928581,
      "num_tokens": 149512629.0,
      "step": 68245
    },
    {
      "entropy": 5.628403806686402,
      "epoch": 6.773521238586741,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015793517653402419,
      "loss": 4.7493,
      "mean_token_accuracy": 0.2109095886349678,
      "num_tokens": 149522840.0,
      "step": 68250
    },
    {
      "entropy": 5.6621424674987795,
      "epoch": 6.774017467248909,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015790491981389077,
      "loss": 4.9098,
      "mean_token_accuracy": 0.201895272731781,
      "num_tokens": 149533659.0,
      "step": 68255
    },
    {
      "entropy": 5.71020359992981,
      "epoch": 6.774513695911076,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015787466599746292,
      "loss": 4.9611,
      "mean_token_accuracy": 0.1944917231798172,
      "num_tokens": 149542944.0,
      "step": 68260
    },
    {
      "entropy": 5.723931550979614,
      "epoch": 6.775009924573244,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015784441508549102,
      "loss": 4.937,
      "mean_token_accuracy": 0.20333362817764283,
      "num_tokens": 149554141.0,
      "step": 68265
    },
    {
      "entropy": 5.737019395828247,
      "epoch": 6.775506153235411,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001578141670787252,
      "loss": 4.9779,
      "mean_token_accuracy": 0.19109453409910201,
      "num_tokens": 149565515.0,
      "step": 68270
    },
    {
      "entropy": 5.698125314712525,
      "epoch": 6.7760023818975785,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015778392197791555,
      "loss": 4.8842,
      "mean_token_accuracy": 0.20520382672548293,
      "num_tokens": 149575791.0,
      "step": 68275
    },
    {
      "entropy": 5.652368593215942,
      "epoch": 6.776498610559746,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015775367978381206,
      "loss": 4.9245,
      "mean_token_accuracy": 0.20303902328014373,
      "num_tokens": 149586782.0,
      "step": 68280
    },
    {
      "entropy": 5.695388126373291,
      "epoch": 6.7769948392219135,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015772344049716454,
      "loss": 4.8926,
      "mean_token_accuracy": 0.19842275232076645,
      "num_tokens": 149596648.0,
      "step": 68285
    },
    {
      "entropy": 5.714103746414184,
      "epoch": 6.777491067884081,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00015769320411872302,
      "loss": 4.995,
      "mean_token_accuracy": 0.19677828401327133,
      "num_tokens": 149608625.0,
      "step": 68290
    },
    {
      "entropy": 5.7047099590301515,
      "epoch": 6.777987296546248,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015766297064923732,
      "loss": 4.9201,
      "mean_token_accuracy": 0.1971409037709236,
      "num_tokens": 149620122.0,
      "step": 68295
    },
    {
      "entropy": 5.6794288635253904,
      "epoch": 6.778483525208416,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015763274008945714,
      "loss": 4.8535,
      "mean_token_accuracy": 0.21458702385425568,
      "num_tokens": 149630557.0,
      "step": 68300
    },
    {
      "entropy": 5.765902137756347,
      "epoch": 6.778979753870583,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001576025124401321,
      "loss": 5.006,
      "mean_token_accuracy": 0.1980995625257492,
      "num_tokens": 149642471.0,
      "step": 68305
    },
    {
      "entropy": 5.738450908660889,
      "epoch": 6.779475982532751,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015757228770201177,
      "loss": 4.9097,
      "mean_token_accuracy": 0.20314210802316665,
      "num_tokens": 149652823.0,
      "step": 68310
    },
    {
      "entropy": 5.687532186508179,
      "epoch": 6.779972211194918,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015754206587584557,
      "loss": 4.8653,
      "mean_token_accuracy": 0.2042321428656578,
      "num_tokens": 149664167.0,
      "step": 68315
    },
    {
      "entropy": 5.73581371307373,
      "epoch": 6.780468439857086,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015751184696238306,
      "loss": 5.0075,
      "mean_token_accuracy": 0.1916787475347519,
      "num_tokens": 149675520.0,
      "step": 68320
    },
    {
      "entropy": 5.686846160888672,
      "epoch": 6.780964668519253,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015748163096237366,
      "loss": 4.9422,
      "mean_token_accuracy": 0.2023798331618309,
      "num_tokens": 149685410.0,
      "step": 68325
    },
    {
      "entropy": 5.663173770904541,
      "epoch": 6.781460897181422,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001574514178765666,
      "loss": 4.8632,
      "mean_token_accuracy": 0.20339182019233704,
      "num_tokens": 149695806.0,
      "step": 68330
    },
    {
      "entropy": 5.7241028308868405,
      "epoch": 6.781957125843588,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015742120770571107,
      "loss": 4.9501,
      "mean_token_accuracy": 0.1941471979022026,
      "num_tokens": 149707156.0,
      "step": 68335
    },
    {
      "entropy": 5.725420570373535,
      "epoch": 6.782453354505757,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001573910004505562,
      "loss": 4.9224,
      "mean_token_accuracy": 0.1987036019563675,
      "num_tokens": 149718511.0,
      "step": 68340
    },
    {
      "entropy": 5.712145566940308,
      "epoch": 6.782949583167924,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001573607961118511,
      "loss": 4.9377,
      "mean_token_accuracy": 0.20468338429927826,
      "num_tokens": 149728942.0,
      "step": 68345
    },
    {
      "entropy": 5.775754594802857,
      "epoch": 6.7834458118300915,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015733059469034472,
      "loss": 4.9911,
      "mean_token_accuracy": 0.19836159348487853,
      "num_tokens": 149739053.0,
      "step": 68350
    },
    {
      "entropy": 5.70702600479126,
      "epoch": 6.783942040492259,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015730039618678602,
      "loss": 4.8885,
      "mean_token_accuracy": 0.20893944799900055,
      "num_tokens": 149750394.0,
      "step": 68355
    },
    {
      "entropy": 5.694977283477783,
      "epoch": 6.7844382691544265,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015727020060192392,
      "loss": 4.9129,
      "mean_token_accuracy": 0.20130905956029893,
      "num_tokens": 149761673.0,
      "step": 68360
    },
    {
      "entropy": 5.694761037826538,
      "epoch": 6.784934497816594,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015724000793650722,
      "loss": 4.9828,
      "mean_token_accuracy": 0.19451065063476564,
      "num_tokens": 149773464.0,
      "step": 68365
    },
    {
      "entropy": 5.704509830474853,
      "epoch": 6.785430726478761,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015720981819128456,
      "loss": 4.9164,
      "mean_token_accuracy": 0.19692452251911163,
      "num_tokens": 149784211.0,
      "step": 68370
    },
    {
      "entropy": 5.702070140838623,
      "epoch": 6.785926955140929,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015717963136700457,
      "loss": 4.8811,
      "mean_token_accuracy": 0.20154855102300645,
      "num_tokens": 149795587.0,
      "step": 68375
    },
    {
      "entropy": 5.7484557151794435,
      "epoch": 6.786423183803096,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001571494474644158,
      "loss": 4.9274,
      "mean_token_accuracy": 0.19988210797309874,
      "num_tokens": 149806748.0,
      "step": 68380
    },
    {
      "entropy": 5.72670259475708,
      "epoch": 6.786919412465264,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015711926648426673,
      "loss": 4.9671,
      "mean_token_accuracy": 0.19387253224849701,
      "num_tokens": 149817445.0,
      "step": 68385
    },
    {
      "entropy": 5.723784017562866,
      "epoch": 6.787415641127431,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015708908842730596,
      "loss": 4.937,
      "mean_token_accuracy": 0.1942323312163353,
      "num_tokens": 149828452.0,
      "step": 68390
    },
    {
      "entropy": 5.642859077453613,
      "epoch": 6.787911869789599,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001570589132942817,
      "loss": 4.8528,
      "mean_token_accuracy": 0.20356917530298232,
      "num_tokens": 149841024.0,
      "step": 68395
    },
    {
      "entropy": 5.734978818893433,
      "epoch": 6.788408098451766,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001570287410859423,
      "loss": 4.9539,
      "mean_token_accuracy": 0.19584614485502244,
      "num_tokens": 149851326.0,
      "step": 68400
    },
    {
      "entropy": 5.66922287940979,
      "epoch": 6.788904327113934,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00015699857180303585,
      "loss": 4.9075,
      "mean_token_accuracy": 0.19927024990320205,
      "num_tokens": 149861444.0,
      "step": 68405
    },
    {
      "entropy": 5.6941358089447025,
      "epoch": 6.789400555776101,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015696840544631048,
      "loss": 4.8473,
      "mean_token_accuracy": 0.20721921026706697,
      "num_tokens": 149871637.0,
      "step": 68410
    },
    {
      "entropy": 5.6180847644805905,
      "epoch": 6.78989678443827,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001569382420165144,
      "loss": 4.7348,
      "mean_token_accuracy": 0.21942515373229982,
      "num_tokens": 149882927.0,
      "step": 68415
    },
    {
      "entropy": 5.650504064559937,
      "epoch": 6.790393013100436,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015690808151439548,
      "loss": 4.8738,
      "mean_token_accuracy": 0.20603155046701432,
      "num_tokens": 149893596.0,
      "step": 68420
    },
    {
      "entropy": 5.732588815689087,
      "epoch": 6.7908892417626046,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001568779239407017,
      "loss": 4.9621,
      "mean_token_accuracy": 0.19683683663606644,
      "num_tokens": 149905872.0,
      "step": 68425
    },
    {
      "entropy": 5.684322261810303,
      "epoch": 6.791385470424772,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015684776929618095,
      "loss": 4.9185,
      "mean_token_accuracy": 0.19478714615106582,
      "num_tokens": 149917660.0,
      "step": 68430
    },
    {
      "entropy": 5.671824169158936,
      "epoch": 6.7918816990869395,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00015681761758158086,
      "loss": 4.8897,
      "mean_token_accuracy": 0.2063358470797539,
      "num_tokens": 149929962.0,
      "step": 68435
    },
    {
      "entropy": 5.7046473026275635,
      "epoch": 6.792377927749107,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001567874687976491,
      "loss": 4.9032,
      "mean_token_accuracy": 0.20670912861824037,
      "num_tokens": 149939506.0,
      "step": 68440
    },
    {
      "entropy": 5.688534784317016,
      "epoch": 6.7928741564112745,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015675732294513352,
      "loss": 4.9096,
      "mean_token_accuracy": 0.20684953778982162,
      "num_tokens": 149949548.0,
      "step": 68445
    },
    {
      "entropy": 5.732886552810669,
      "epoch": 6.793370385073442,
      "grad_norm": 1.5,
      "learning_rate": 0.0001567271800247814,
      "loss": 4.8841,
      "mean_token_accuracy": 0.19605180323123933,
      "num_tokens": 149958655.0,
      "step": 68450
    },
    {
      "entropy": 5.642630767822266,
      "epoch": 6.793866613735609,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015669704003734052,
      "loss": 4.8453,
      "mean_token_accuracy": 0.20334277004003526,
      "num_tokens": 149970100.0,
      "step": 68455
    },
    {
      "entropy": 5.680525398254394,
      "epoch": 6.794362842397777,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015666690298355806,
      "loss": 4.85,
      "mean_token_accuracy": 0.2044537290930748,
      "num_tokens": 149979968.0,
      "step": 68460
    },
    {
      "entropy": 5.617535734176636,
      "epoch": 6.794859071059944,
      "grad_norm": 1.46875,
      "learning_rate": 0.00015663676886418137,
      "loss": 4.8598,
      "mean_token_accuracy": 0.2073962613940239,
      "num_tokens": 149991064.0,
      "step": 68465
    },
    {
      "entropy": 5.652126693725586,
      "epoch": 6.795355299722112,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00015660663767995785,
      "loss": 4.8766,
      "mean_token_accuracy": 0.2008587747812271,
      "num_tokens": 150002621.0,
      "step": 68470
    },
    {
      "entropy": 5.676389837265015,
      "epoch": 6.795851528384279,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00015657650943163459,
      "loss": 4.8814,
      "mean_token_accuracy": 0.20312952548265456,
      "num_tokens": 150012778.0,
      "step": 68475
    },
    {
      "entropy": 5.636977624893189,
      "epoch": 6.796347757046447,
      "grad_norm": 1.25,
      "learning_rate": 0.00015654638411995863,
      "loss": 4.8087,
      "mean_token_accuracy": 0.2024806931614876,
      "num_tokens": 150024208.0,
      "step": 68480
    },
    {
      "entropy": 5.707226657867432,
      "epoch": 6.796843985708614,
      "grad_norm": 1.65625,
      "learning_rate": 0.00015651626174567718,
      "loss": 4.9314,
      "mean_token_accuracy": 0.2014932632446289,
      "num_tokens": 150034711.0,
      "step": 68485
    },
    {
      "entropy": 5.7402478694915775,
      "epoch": 6.797340214370782,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015648614230953718,
      "loss": 4.9648,
      "mean_token_accuracy": 0.20083801746368407,
      "num_tokens": 150044955.0,
      "step": 68490
    },
    {
      "entropy": 5.719609355926513,
      "epoch": 6.797836443032949,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001564560258122853,
      "loss": 4.9313,
      "mean_token_accuracy": 0.19663993865251542,
      "num_tokens": 150056453.0,
      "step": 68495
    },
    {
      "entropy": 5.763877010345459,
      "epoch": 6.798332671695118,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001564259122546687,
      "loss": 4.9488,
      "mean_token_accuracy": 0.20136189460754395,
      "num_tokens": 150067903.0,
      "step": 68500
    },
    {
      "entropy": 5.653004455566406,
      "epoch": 6.798828900357284,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001563958016374339,
      "loss": 4.874,
      "mean_token_accuracy": 0.2026468500494957,
      "num_tokens": 150078311.0,
      "step": 68505
    },
    {
      "entropy": 5.660408353805542,
      "epoch": 6.7993251290194525,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015636569396132772,
      "loss": 4.9173,
      "mean_token_accuracy": 0.20027070194482804,
      "num_tokens": 150090877.0,
      "step": 68510
    },
    {
      "entropy": 5.75298285484314,
      "epoch": 6.79982135768162,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015633558922709657,
      "loss": 4.9488,
      "mean_token_accuracy": 0.19714037477970123,
      "num_tokens": 150102874.0,
      "step": 68515
    },
    {
      "entropy": 5.671145343780518,
      "epoch": 6.8003175863437875,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015630548743548724,
      "loss": 4.8599,
      "mean_token_accuracy": 0.20131770819425582,
      "num_tokens": 150113890.0,
      "step": 68520
    },
    {
      "entropy": 5.7281084060668945,
      "epoch": 6.800813815005955,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015627538858724595,
      "loss": 4.9457,
      "mean_token_accuracy": 0.20190160870552062,
      "num_tokens": 150124486.0,
      "step": 68525
    },
    {
      "entropy": 5.734810733795166,
      "epoch": 6.801310043668122,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00015624529268311923,
      "loss": 4.9243,
      "mean_token_accuracy": 0.19522153884172438,
      "num_tokens": 150135813.0,
      "step": 68530
    },
    {
      "entropy": 5.684515714645386,
      "epoch": 6.80180627233029,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001562151997238534,
      "loss": 4.8488,
      "mean_token_accuracy": 0.20025141388177872,
      "num_tokens": 150145631.0,
      "step": 68535
    },
    {
      "entropy": 5.641709804534912,
      "epoch": 6.802302500992457,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015618510971019466,
      "loss": 4.8731,
      "mean_token_accuracy": 0.21733921021223068,
      "num_tokens": 150156990.0,
      "step": 68540
    },
    {
      "entropy": 5.704302740097046,
      "epoch": 6.802798729654625,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015615502264288905,
      "loss": 4.9567,
      "mean_token_accuracy": 0.1974819928407669,
      "num_tokens": 150168564.0,
      "step": 68545
    },
    {
      "entropy": 5.651795434951782,
      "epoch": 6.803294958316792,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015612493852268293,
      "loss": 4.8296,
      "mean_token_accuracy": 0.20414791256189346,
      "num_tokens": 150179985.0,
      "step": 68550
    },
    {
      "entropy": 5.701302862167358,
      "epoch": 6.80379118697896,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015609485735032207,
      "loss": 4.9672,
      "mean_token_accuracy": 0.19558335840702057,
      "num_tokens": 150191220.0,
      "step": 68555
    },
    {
      "entropy": 5.6821918964385985,
      "epoch": 6.804287415641127,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015606477912655264,
      "loss": 4.903,
      "mean_token_accuracy": 0.19404688328504563,
      "num_tokens": 150202589.0,
      "step": 68560
    },
    {
      "entropy": 5.733524417877197,
      "epoch": 6.804783644303295,
      "grad_norm": 1.640625,
      "learning_rate": 0.00015603470385212038,
      "loss": 4.926,
      "mean_token_accuracy": 0.19860607087612153,
      "num_tokens": 150212809.0,
      "step": 68565
    },
    {
      "entropy": 5.6551446437835695,
      "epoch": 6.805279872965462,
      "grad_norm": 1.375,
      "learning_rate": 0.0001560046315277711,
      "loss": 4.8993,
      "mean_token_accuracy": 0.19657449573278427,
      "num_tokens": 150223034.0,
      "step": 68570
    },
    {
      "entropy": 5.651805448532104,
      "epoch": 6.80577610162763,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001559745621542506,
      "loss": 4.9123,
      "mean_token_accuracy": 0.1974350020289421,
      "num_tokens": 150234703.0,
      "step": 68575
    },
    {
      "entropy": 5.6552777767181395,
      "epoch": 6.806272330289797,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015594449573230436,
      "loss": 4.8892,
      "mean_token_accuracy": 0.19783450663089752,
      "num_tokens": 150245632.0,
      "step": 68580
    },
    {
      "entropy": 5.693780851364136,
      "epoch": 6.8067685589519655,
      "grad_norm": 1.25,
      "learning_rate": 0.0001559144322626781,
      "loss": 4.8808,
      "mean_token_accuracy": 0.20911235362291336,
      "num_tokens": 150258794.0,
      "step": 68585
    },
    {
      "entropy": 5.609170579910279,
      "epoch": 6.807264787614132,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001558843717461174,
      "loss": 4.7989,
      "mean_token_accuracy": 0.21776664406061172,
      "num_tokens": 150269560.0,
      "step": 68590
    },
    {
      "entropy": 5.719814395904541,
      "epoch": 6.8077610162763005,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015585431418336758,
      "loss": 4.9344,
      "mean_token_accuracy": 0.2021138995885849,
      "num_tokens": 150280640.0,
      "step": 68595
    },
    {
      "entropy": 5.677964305877685,
      "epoch": 6.808257244938468,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015582425957517404,
      "loss": 4.9524,
      "mean_token_accuracy": 0.19738211035728453,
      "num_tokens": 150291490.0,
      "step": 68600
    },
    {
      "entropy": 5.71344256401062,
      "epoch": 6.808753473600635,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015579420792228207,
      "loss": 4.969,
      "mean_token_accuracy": 0.20067476928234101,
      "num_tokens": 150303463.0,
      "step": 68605
    },
    {
      "entropy": 5.697098684310913,
      "epoch": 6.809249702262803,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015576415922543679,
      "loss": 4.9095,
      "mean_token_accuracy": 0.1979170098900795,
      "num_tokens": 150313413.0,
      "step": 68610
    },
    {
      "entropy": 5.642182493209839,
      "epoch": 6.80974593092497,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015573411348538345,
      "loss": 4.8437,
      "mean_token_accuracy": 0.20265941619873046,
      "num_tokens": 150324706.0,
      "step": 68615
    },
    {
      "entropy": 5.6252100467681885,
      "epoch": 6.810242159587138,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001557040707028672,
      "loss": 4.9052,
      "mean_token_accuracy": 0.20968620032072066,
      "num_tokens": 150336718.0,
      "step": 68620
    },
    {
      "entropy": 5.66585636138916,
      "epoch": 6.810738388249305,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001556740308786329,
      "loss": 4.8444,
      "mean_token_accuracy": 0.2061208590865135,
      "num_tokens": 150347785.0,
      "step": 68625
    },
    {
      "entropy": 5.702195549011231,
      "epoch": 6.811234616911473,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015564399401342549,
      "loss": 4.9107,
      "mean_token_accuracy": 0.1985623747110367,
      "num_tokens": 150357810.0,
      "step": 68630
    },
    {
      "entropy": 5.720114421844483,
      "epoch": 6.81173084557364,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015561396010798986,
      "loss": 4.8912,
      "mean_token_accuracy": 0.20643081068992614,
      "num_tokens": 150369103.0,
      "step": 68635
    },
    {
      "entropy": 5.776167726516723,
      "epoch": 6.812227074235808,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001555839291630707,
      "loss": 5.0478,
      "mean_token_accuracy": 0.19315432459115983,
      "num_tokens": 150380687.0,
      "step": 68640
    },
    {
      "entropy": 5.708545398712158,
      "epoch": 6.812723302897975,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015555390117941287,
      "loss": 4.913,
      "mean_token_accuracy": 0.19107935428619385,
      "num_tokens": 150391492.0,
      "step": 68645
    },
    {
      "entropy": 5.764432334899903,
      "epoch": 6.813219531560143,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015552387615776079,
      "loss": 5.0264,
      "mean_token_accuracy": 0.19692951440811157,
      "num_tokens": 150402677.0,
      "step": 68650
    },
    {
      "entropy": 5.698663425445557,
      "epoch": 6.81371576022231,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015549385409885925,
      "loss": 4.949,
      "mean_token_accuracy": 0.19397050887346268,
      "num_tokens": 150415738.0,
      "step": 68655
    },
    {
      "entropy": 5.731507968902588,
      "epoch": 6.814211988884478,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015546383500345256,
      "loss": 4.9504,
      "mean_token_accuracy": 0.19313507676124572,
      "num_tokens": 150427119.0,
      "step": 68660
    },
    {
      "entropy": 5.700031757354736,
      "epoch": 6.814708217546645,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015543381887228526,
      "loss": 4.8453,
      "mean_token_accuracy": 0.21159439086914061,
      "num_tokens": 150438650.0,
      "step": 68665
    },
    {
      "entropy": 5.683742952346802,
      "epoch": 6.8152044462088135,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015540380570610153,
      "loss": 4.8495,
      "mean_token_accuracy": 0.2062753826379776,
      "num_tokens": 150449286.0,
      "step": 68670
    },
    {
      "entropy": 5.658876657485962,
      "epoch": 6.81570067487098,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00015537379550564567,
      "loss": 4.8281,
      "mean_token_accuracy": 0.2043117552995682,
      "num_tokens": 150459534.0,
      "step": 68675
    },
    {
      "entropy": 5.719955968856811,
      "epoch": 6.816196903533148,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001553437882716619,
      "loss": 4.9499,
      "mean_token_accuracy": 0.19512264877557756,
      "num_tokens": 150470120.0,
      "step": 68680
    },
    {
      "entropy": 5.681433534622192,
      "epoch": 6.816693132195316,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00015531378400489442,
      "loss": 4.9093,
      "mean_token_accuracy": 0.20081153512001038,
      "num_tokens": 150481685.0,
      "step": 68685
    },
    {
      "entropy": 5.80794587135315,
      "epoch": 6.817189360857483,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015528378270608724,
      "loss": 5.0213,
      "mean_token_accuracy": 0.19045692235231398,
      "num_tokens": 150490726.0,
      "step": 68690
    },
    {
      "entropy": 5.714056348800659,
      "epoch": 6.817685589519651,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015525378437598425,
      "loss": 4.8885,
      "mean_token_accuracy": 0.20863743275403976,
      "num_tokens": 150501681.0,
      "step": 68695
    },
    {
      "entropy": 5.612164402008057,
      "epoch": 6.818181818181818,
      "grad_norm": 1.625,
      "learning_rate": 0.0001552237890153293,
      "loss": 4.8731,
      "mean_token_accuracy": 0.20190970599651337,
      "num_tokens": 150511792.0,
      "step": 68700
    },
    {
      "entropy": 5.609955787658691,
      "epoch": 6.818678046843986,
      "grad_norm": 1.375,
      "learning_rate": 0.00015519379662486636,
      "loss": 4.8036,
      "mean_token_accuracy": 0.20992761999368667,
      "num_tokens": 150523077.0,
      "step": 68705
    },
    {
      "entropy": 5.627978992462158,
      "epoch": 6.819174275506153,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015516380720533902,
      "loss": 4.8816,
      "mean_token_accuracy": 0.19857651740312576,
      "num_tokens": 150533939.0,
      "step": 68710
    },
    {
      "entropy": 5.713487386703491,
      "epoch": 6.819670504168321,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015513382075749114,
      "loss": 4.9112,
      "mean_token_accuracy": 0.19793895334005357,
      "num_tokens": 150544855.0,
      "step": 68715
    },
    {
      "entropy": 5.712920999526977,
      "epoch": 6.820166732830488,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015510383728206618,
      "loss": 4.9595,
      "mean_token_accuracy": 0.20682377517223358,
      "num_tokens": 150555154.0,
      "step": 68720
    },
    {
      "entropy": 5.676968288421631,
      "epoch": 6.820662961492656,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015507385677980773,
      "loss": 4.9215,
      "mean_token_accuracy": 0.2026427999138832,
      "num_tokens": 150566517.0,
      "step": 68725
    },
    {
      "entropy": 5.73049578666687,
      "epoch": 6.821159190154823,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015504387925145916,
      "loss": 5.0008,
      "mean_token_accuracy": 0.19778937548398973,
      "num_tokens": 150578593.0,
      "step": 68730
    },
    {
      "entropy": 5.68637866973877,
      "epoch": 6.821655418816991,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015501390469776393,
      "loss": 4.8888,
      "mean_token_accuracy": 0.20048887729644777,
      "num_tokens": 150589687.0,
      "step": 68735
    },
    {
      "entropy": 5.738058185577392,
      "epoch": 6.822151647479158,
      "grad_norm": 1.46875,
      "learning_rate": 0.00015498393311946535,
      "loss": 4.9106,
      "mean_token_accuracy": 0.19780651181936265,
      "num_tokens": 150600635.0,
      "step": 68740
    },
    {
      "entropy": 5.7556883811950685,
      "epoch": 6.822647876141326,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00015495396451730654,
      "loss": 4.9538,
      "mean_token_accuracy": 0.1929629221558571,
      "num_tokens": 150612078.0,
      "step": 68745
    },
    {
      "entropy": 5.71554708480835,
      "epoch": 6.823144104803493,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015492399889203078,
      "loss": 4.8764,
      "mean_token_accuracy": 0.19800323098897935,
      "num_tokens": 150622538.0,
      "step": 68750
    },
    {
      "entropy": 5.739840698242188,
      "epoch": 6.8236403334656615,
      "grad_norm": 1.5625,
      "learning_rate": 0.00015489403624438115,
      "loss": 4.9452,
      "mean_token_accuracy": 0.19822938591241837,
      "num_tokens": 150634168.0,
      "step": 68755
    },
    {
      "entropy": 5.7144838809967045,
      "epoch": 6.824136562127828,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015486407657510048,
      "loss": 4.8608,
      "mean_token_accuracy": 0.20581091940402985,
      "num_tokens": 150644384.0,
      "step": 68760
    },
    {
      "entropy": 5.652218770980835,
      "epoch": 6.824632790789996,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000154834119884932,
      "loss": 4.8096,
      "mean_token_accuracy": 0.20910502970218658,
      "num_tokens": 150655681.0,
      "step": 68765
    },
    {
      "entropy": 5.610623455047607,
      "epoch": 6.825129019452164,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015480416617461834,
      "loss": 4.8256,
      "mean_token_accuracy": 0.20828530341386794,
      "num_tokens": 150665956.0,
      "step": 68770
    },
    {
      "entropy": 5.638995599746704,
      "epoch": 6.825625248114331,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015477421544490228,
      "loss": 4.8916,
      "mean_token_accuracy": 0.2075223445892334,
      "num_tokens": 150677390.0,
      "step": 68775
    },
    {
      "entropy": 5.7156662940979,
      "epoch": 6.826121476776499,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015474426769652674,
      "loss": 4.9657,
      "mean_token_accuracy": 0.20294613838195802,
      "num_tokens": 150689452.0,
      "step": 68780
    },
    {
      "entropy": 5.710358810424805,
      "epoch": 6.826617705438666,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015471432293023423,
      "loss": 4.8721,
      "mean_token_accuracy": 0.2041141301393509,
      "num_tokens": 150701143.0,
      "step": 68785
    },
    {
      "entropy": 5.709344434738159,
      "epoch": 6.827113934100834,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015468438114676718,
      "loss": 4.9123,
      "mean_token_accuracy": 0.19861846268177033,
      "num_tokens": 150711263.0,
      "step": 68790
    },
    {
      "entropy": 5.716709232330322,
      "epoch": 6.827610162763001,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015465444234686836,
      "loss": 4.9537,
      "mean_token_accuracy": 0.19742266833782196,
      "num_tokens": 150722099.0,
      "step": 68795
    },
    {
      "entropy": 5.688414621353149,
      "epoch": 6.828106391425169,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015462450653128004,
      "loss": 4.9279,
      "mean_token_accuracy": 0.20121295005083084,
      "num_tokens": 150734029.0,
      "step": 68800
    },
    {
      "entropy": 5.702828884124756,
      "epoch": 6.828602620087336,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00015459457370074458,
      "loss": 4.9599,
      "mean_token_accuracy": 0.1998475521802902,
      "num_tokens": 150744881.0,
      "step": 68805
    },
    {
      "entropy": 5.748701286315918,
      "epoch": 6.829098848749504,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001545646438560041,
      "loss": 4.9643,
      "mean_token_accuracy": 0.19436809420585632,
      "num_tokens": 150755464.0,
      "step": 68810
    },
    {
      "entropy": 5.725692939758301,
      "epoch": 6.829595077411671,
      "grad_norm": 1.375,
      "learning_rate": 0.0001545347169978011,
      "loss": 4.957,
      "mean_token_accuracy": 0.19665095210075378,
      "num_tokens": 150765436.0,
      "step": 68815
    },
    {
      "entropy": 5.711906576156617,
      "epoch": 6.830091306073839,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001545047931268774,
      "loss": 4.927,
      "mean_token_accuracy": 0.2009672701358795,
      "num_tokens": 150775377.0,
      "step": 68820
    },
    {
      "entropy": 5.704653739929199,
      "epoch": 6.830587534736006,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015447487224397527,
      "loss": 4.9235,
      "mean_token_accuracy": 0.20411166548728943,
      "num_tokens": 150785995.0,
      "step": 68825
    },
    {
      "entropy": 5.651759004592895,
      "epoch": 6.831083763398174,
      "grad_norm": 1.25,
      "learning_rate": 0.00015444495434983663,
      "loss": 4.8977,
      "mean_token_accuracy": 0.20507954359054564,
      "num_tokens": 150796147.0,
      "step": 68830
    },
    {
      "entropy": 5.762526893615723,
      "epoch": 6.831579992060341,
      "grad_norm": 1.375,
      "learning_rate": 0.00015441503944520332,
      "loss": 4.9381,
      "mean_token_accuracy": 0.20304495543241502,
      "num_tokens": 150806039.0,
      "step": 68835
    },
    {
      "entropy": 5.632604455947876,
      "epoch": 6.832076220722509,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00015438512753081712,
      "loss": 4.8289,
      "mean_token_accuracy": 0.21563850790262223,
      "num_tokens": 150816686.0,
      "step": 68840
    },
    {
      "entropy": 5.726649761199951,
      "epoch": 6.832572449384676,
      "grad_norm": 1.46875,
      "learning_rate": 0.00015435521860741996,
      "loss": 4.9626,
      "mean_token_accuracy": 0.201476027071476,
      "num_tokens": 150828729.0,
      "step": 68845
    },
    {
      "entropy": 5.649485301971436,
      "epoch": 6.833068678046844,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015432531267575334,
      "loss": 4.8215,
      "mean_token_accuracy": 0.20905750542879104,
      "num_tokens": 150838716.0,
      "step": 68850
    },
    {
      "entropy": 5.677534914016723,
      "epoch": 6.833564906709012,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015429540973655898,
      "loss": 4.8598,
      "mean_token_accuracy": 0.20440250337123872,
      "num_tokens": 150848926.0,
      "step": 68855
    },
    {
      "entropy": 5.661960458755493,
      "epoch": 6.834061135371179,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015426550979057835,
      "loss": 4.8592,
      "mean_token_accuracy": 0.20200374722480774,
      "num_tokens": 150859644.0,
      "step": 68860
    },
    {
      "entropy": 5.712015581130982,
      "epoch": 6.834557364033347,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015423561283855295,
      "loss": 4.8596,
      "mean_token_accuracy": 0.20495944768190383,
      "num_tokens": 150871422.0,
      "step": 68865
    },
    {
      "entropy": 5.778095436096192,
      "epoch": 6.835053592695514,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001542057188812241,
      "loss": 5.0235,
      "mean_token_accuracy": 0.1922045648097992,
      "num_tokens": 150881220.0,
      "step": 68870
    },
    {
      "entropy": 5.65389723777771,
      "epoch": 6.835549821357682,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001541758279193331,
      "loss": 4.8788,
      "mean_token_accuracy": 0.20691534429788588,
      "num_tokens": 150891242.0,
      "step": 68875
    },
    {
      "entropy": 5.691067409515381,
      "epoch": 6.836046050019849,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00015414593995362116,
      "loss": 4.8988,
      "mean_token_accuracy": 0.20123401582241057,
      "num_tokens": 150901487.0,
      "step": 68880
    },
    {
      "entropy": 5.73117094039917,
      "epoch": 6.836542278682017,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001541160549848296,
      "loss": 4.9898,
      "mean_token_accuracy": 0.19304679781198503,
      "num_tokens": 150913914.0,
      "step": 68885
    },
    {
      "entropy": 5.706584501266479,
      "epoch": 6.837038507344184,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015408617301369942,
      "loss": 4.944,
      "mean_token_accuracy": 0.19658534675836564,
      "num_tokens": 150925299.0,
      "step": 68890
    },
    {
      "entropy": 5.624533367156983,
      "epoch": 6.837534736006352,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015405629404097157,
      "loss": 4.7833,
      "mean_token_accuracy": 0.2158450350165367,
      "num_tokens": 150937652.0,
      "step": 68895
    },
    {
      "entropy": 5.719785356521607,
      "epoch": 6.838030964668519,
      "grad_norm": 1.390625,
      "learning_rate": 0.000154026418067387,
      "loss": 4.9229,
      "mean_token_accuracy": 0.20128607451915742,
      "num_tokens": 150947972.0,
      "step": 68900
    },
    {
      "entropy": 5.728865146636963,
      "epoch": 6.838527193330687,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015399654509368654,
      "loss": 4.9787,
      "mean_token_accuracy": 0.19776504784822463,
      "num_tokens": 150958229.0,
      "step": 68905
    },
    {
      "entropy": 5.728024816513061,
      "epoch": 6.839023421992854,
      "grad_norm": 1.359375,
      "learning_rate": 0.000153966675120611,
      "loss": 4.9007,
      "mean_token_accuracy": 0.20407365411520004,
      "num_tokens": 150969906.0,
      "step": 68910
    },
    {
      "entropy": 5.662119293212891,
      "epoch": 6.8395196506550215,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00015393680814890123,
      "loss": 4.832,
      "mean_token_accuracy": 0.2013708919286728,
      "num_tokens": 150981791.0,
      "step": 68915
    },
    {
      "entropy": 5.743438577651977,
      "epoch": 6.840015879317189,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015390694417929772,
      "loss": 4.9575,
      "mean_token_accuracy": 0.1953026071190834,
      "num_tokens": 150992098.0,
      "step": 68920
    },
    {
      "entropy": 5.632068347930908,
      "epoch": 6.840512107979357,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015387708321254107,
      "loss": 4.838,
      "mean_token_accuracy": 0.21177971512079238,
      "num_tokens": 151003600.0,
      "step": 68925
    },
    {
      "entropy": 5.692697381973266,
      "epoch": 6.841008336641524,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00015384722524937176,
      "loss": 4.8361,
      "mean_token_accuracy": 0.20240271985530853,
      "num_tokens": 151014501.0,
      "step": 68930
    },
    {
      "entropy": 5.653683853149414,
      "epoch": 6.841504565303692,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001538173702905301,
      "loss": 4.8926,
      "mean_token_accuracy": 0.2072292074561119,
      "num_tokens": 151025212.0,
      "step": 68935
    },
    {
      "entropy": 5.740349292755127,
      "epoch": 6.84200079396586,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00015378751833675664,
      "loss": 4.9764,
      "mean_token_accuracy": 0.19822409600019456,
      "num_tokens": 151036683.0,
      "step": 68940
    },
    {
      "entropy": 5.6669024467468265,
      "epoch": 6.842497022628027,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015375766938879147,
      "loss": 4.8216,
      "mean_token_accuracy": 0.20680440664291383,
      "num_tokens": 151048102.0,
      "step": 68945
    },
    {
      "entropy": 5.735555219650268,
      "epoch": 6.842993251290195,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015372782344737486,
      "loss": 4.9782,
      "mean_token_accuracy": 0.19037696123123168,
      "num_tokens": 151058765.0,
      "step": 68950
    },
    {
      "entropy": 5.7329038143157955,
      "epoch": 6.843489479952362,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015369798051324696,
      "loss": 4.9481,
      "mean_token_accuracy": 0.1965814098715782,
      "num_tokens": 151069770.0,
      "step": 68955
    },
    {
      "entropy": 5.769258499145508,
      "epoch": 6.84398570861453,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015366814058714778,
      "loss": 5.0248,
      "mean_token_accuracy": 0.19570285230875015,
      "num_tokens": 151081902.0,
      "step": 68960
    },
    {
      "entropy": 5.699860715866089,
      "epoch": 6.844481937276697,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015363830366981715,
      "loss": 4.9183,
      "mean_token_accuracy": 0.20268917679786683,
      "num_tokens": 151092847.0,
      "step": 68965
    },
    {
      "entropy": 5.751021432876587,
      "epoch": 6.844978165938865,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015360846976199516,
      "loss": 4.9283,
      "mean_token_accuracy": 0.20373705476522447,
      "num_tokens": 151104590.0,
      "step": 68970
    },
    {
      "entropy": 5.777047348022461,
      "epoch": 6.845474394601032,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015357863886442146,
      "loss": 4.9706,
      "mean_token_accuracy": 0.1973254233598709,
      "num_tokens": 151117041.0,
      "step": 68975
    },
    {
      "entropy": 5.698089218139648,
      "epoch": 6.8459706232632,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015354881097783595,
      "loss": 4.8859,
      "mean_token_accuracy": 0.21052294969558716,
      "num_tokens": 151128835.0,
      "step": 68980
    },
    {
      "entropy": 5.682157039642334,
      "epoch": 6.846466851925367,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015351898610297827,
      "loss": 4.9405,
      "mean_token_accuracy": 0.20415019094944,
      "num_tokens": 151141034.0,
      "step": 68985
    },
    {
      "entropy": 5.68270206451416,
      "epoch": 6.8469630805875346,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00015348916424058796,
      "loss": 4.8897,
      "mean_token_accuracy": 0.20410769879817964,
      "num_tokens": 151152338.0,
      "step": 68990
    },
    {
      "entropy": 5.686428833007812,
      "epoch": 6.847459309249702,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001534593453914044,
      "loss": 4.8208,
      "mean_token_accuracy": 0.2068308800458908,
      "num_tokens": 151163332.0,
      "step": 68995
    },
    {
      "entropy": 5.662587594985962,
      "epoch": 6.8479555379118695,
      "grad_norm": 1.4375,
      "learning_rate": 0.00015342952955616735,
      "loss": 4.881,
      "mean_token_accuracy": 0.20284761488437653,
      "num_tokens": 151173621.0,
      "step": 69000
    },
    {
      "epoch": 6.8479555379118695,
      "eval_entropy": 5.424501273716451,
      "eval_loss": 5.062732696533203,
      "eval_mean_token_accuracy": 0.1986024705084635,
      "eval_num_tokens": 151173621.0,
      "eval_runtime": 20.4379,
      "eval_samples_per_second": 1647.379,
      "eval_steps_per_second": 205.941,
      "step": 69000
    },
    {
      "entropy": 5.715849256515503,
      "epoch": 6.848451766574037,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015339971673561584,
      "loss": 4.9407,
      "mean_token_accuracy": 0.20432147681713103,
      "num_tokens": 151184268.0,
      "step": 69005
    },
    {
      "entropy": 5.663975620269776,
      "epoch": 6.8489479952362045,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015336990693048946,
      "loss": 4.8556,
      "mean_token_accuracy": 0.20938023179769516,
      "num_tokens": 151194648.0,
      "step": 69010
    },
    {
      "entropy": 5.667794466018677,
      "epoch": 6.849444223898372,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015334010014152728,
      "loss": 4.9039,
      "mean_token_accuracy": 0.2040925458073616,
      "num_tokens": 151205097.0,
      "step": 69015
    },
    {
      "entropy": 5.713924694061279,
      "epoch": 6.84994045256054,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015331029636946844,
      "loss": 4.9257,
      "mean_token_accuracy": 0.19468609988689423,
      "num_tokens": 151216231.0,
      "step": 69020
    },
    {
      "entropy": 5.74437780380249,
      "epoch": 6.850436681222708,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015328049561505198,
      "loss": 4.9436,
      "mean_token_accuracy": 0.19461339414119722,
      "num_tokens": 151227474.0,
      "step": 69025
    },
    {
      "entropy": 5.7561047077178955,
      "epoch": 6.850932909884875,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000153250697879017,
      "loss": 5.0101,
      "mean_token_accuracy": 0.1958902359008789,
      "num_tokens": 151239118.0,
      "step": 69030
    },
    {
      "entropy": 5.744930601119995,
      "epoch": 6.851429138547043,
      "grad_norm": 1.5,
      "learning_rate": 0.0001532209031621024,
      "loss": 4.9391,
      "mean_token_accuracy": 0.19530216306447984,
      "num_tokens": 151248718.0,
      "step": 69035
    },
    {
      "entropy": 5.729693126678467,
      "epoch": 6.85192536720921,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015319111146504691,
      "loss": 4.9187,
      "mean_token_accuracy": 0.19761050939559938,
      "num_tokens": 151259835.0,
      "step": 69040
    },
    {
      "entropy": 5.6972705841064455,
      "epoch": 6.852421595871378,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015316132278858947,
      "loss": 4.8552,
      "mean_token_accuracy": 0.20191811621189118,
      "num_tokens": 151271936.0,
      "step": 69045
    },
    {
      "entropy": 5.665978050231933,
      "epoch": 6.852917824533545,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015313153713346868,
      "loss": 4.8267,
      "mean_token_accuracy": 0.21360026448965072,
      "num_tokens": 151282733.0,
      "step": 69050
    },
    {
      "entropy": 5.613889694213867,
      "epoch": 6.853414053195713,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015310175450042305,
      "loss": 4.8444,
      "mean_token_accuracy": 0.21153218150138856,
      "num_tokens": 151292322.0,
      "step": 69055
    },
    {
      "entropy": 5.678687381744385,
      "epoch": 6.85391028185788,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00015307197489019137,
      "loss": 4.9219,
      "mean_token_accuracy": 0.20219200551509858,
      "num_tokens": 151305021.0,
      "step": 69060
    },
    {
      "entropy": 5.7394458770751955,
      "epoch": 6.854406510520048,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015304219830351192,
      "loss": 4.9213,
      "mean_token_accuracy": 0.19579849392175674,
      "num_tokens": 151314822.0,
      "step": 69065
    },
    {
      "entropy": 5.752188587188721,
      "epoch": 6.854902739182215,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001530124247411231,
      "loss": 4.9711,
      "mean_token_accuracy": 0.19785282909870147,
      "num_tokens": 151326764.0,
      "step": 69070
    },
    {
      "entropy": 5.685807132720948,
      "epoch": 6.8553989678443825,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015298265420376342,
      "loss": 4.8288,
      "mean_token_accuracy": 0.20436523258686065,
      "num_tokens": 151336647.0,
      "step": 69075
    },
    {
      "entropy": 5.7389280796051025,
      "epoch": 6.85589519650655,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00015295288669217095,
      "loss": 4.9478,
      "mean_token_accuracy": 0.2068041369318962,
      "num_tokens": 151348283.0,
      "step": 69080
    },
    {
      "entropy": 5.769308042526245,
      "epoch": 6.8563914251687175,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015292312220708382,
      "loss": 5.0208,
      "mean_token_accuracy": 0.19606920778751374,
      "num_tokens": 151361324.0,
      "step": 69085
    },
    {
      "entropy": 5.704714965820313,
      "epoch": 6.856887653830885,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00015289336074924027,
      "loss": 4.8724,
      "mean_token_accuracy": 0.20417087972164155,
      "num_tokens": 151373330.0,
      "step": 69090
    },
    {
      "entropy": 5.729242467880249,
      "epoch": 6.857383882493052,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00015286360231937828,
      "loss": 4.9236,
      "mean_token_accuracy": 0.19876490384340287,
      "num_tokens": 151383991.0,
      "step": 69095
    },
    {
      "entropy": 5.674645042419433,
      "epoch": 6.85788011115522,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015283384691823577,
      "loss": 4.9606,
      "mean_token_accuracy": 0.20901334285736084,
      "num_tokens": 151395478.0,
      "step": 69100
    },
    {
      "entropy": 5.665039205551148,
      "epoch": 6.858376339817388,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015280409454655053,
      "loss": 4.8094,
      "mean_token_accuracy": 0.2109810844063759,
      "num_tokens": 151405777.0,
      "step": 69105
    },
    {
      "entropy": 5.665099048614502,
      "epoch": 6.858872568479556,
      "grad_norm": 1.4375,
      "learning_rate": 0.00015277434520506053,
      "loss": 4.803,
      "mean_token_accuracy": 0.20990019142627717,
      "num_tokens": 151415957.0,
      "step": 69110
    },
    {
      "entropy": 5.702636337280273,
      "epoch": 6.859368797141723,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015274459889450328,
      "loss": 4.9126,
      "mean_token_accuracy": 0.19841643720865249,
      "num_tokens": 151425705.0,
      "step": 69115
    },
    {
      "entropy": 5.601934766769409,
      "epoch": 6.859865025803891,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015271485561561666,
      "loss": 4.7862,
      "mean_token_accuracy": 0.21553343236446382,
      "num_tokens": 151436942.0,
      "step": 69120
    },
    {
      "entropy": 5.6669210433959964,
      "epoch": 6.860361254466058,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001526851153691381,
      "loss": 4.898,
      "mean_token_accuracy": 0.2006716623902321,
      "num_tokens": 151447681.0,
      "step": 69125
    },
    {
      "entropy": 5.697145271301269,
      "epoch": 6.860857483128226,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001526553781558051,
      "loss": 4.9577,
      "mean_token_accuracy": 0.192792084813118,
      "num_tokens": 151458894.0,
      "step": 69130
    },
    {
      "entropy": 5.673519229888916,
      "epoch": 6.861353711790393,
      "grad_norm": 1.5625,
      "learning_rate": 0.00015262564397635503,
      "loss": 4.9331,
      "mean_token_accuracy": 0.19497084319591523,
      "num_tokens": 151469087.0,
      "step": 69135
    },
    {
      "entropy": 5.720129013061523,
      "epoch": 6.861849940452561,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015259591283152537,
      "loss": 4.9442,
      "mean_token_accuracy": 0.20113009959459305,
      "num_tokens": 151480872.0,
      "step": 69140
    },
    {
      "entropy": 5.7021143436431885,
      "epoch": 6.862346169114728,
      "grad_norm": 1.25,
      "learning_rate": 0.00015256618472205325,
      "loss": 4.923,
      "mean_token_accuracy": 0.20248174667358398,
      "num_tokens": 151492177.0,
      "step": 69145
    },
    {
      "entropy": 5.6955193996429445,
      "epoch": 6.8628423977768955,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015253645964867602,
      "loss": 4.9265,
      "mean_token_accuracy": 0.1905639499425888,
      "num_tokens": 151502571.0,
      "step": 69150
    },
    {
      "entropy": 5.703076410293579,
      "epoch": 6.863338626439063,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015250673761213067,
      "loss": 4.9432,
      "mean_token_accuracy": 0.19885947853326796,
      "num_tokens": 151512985.0,
      "step": 69155
    },
    {
      "entropy": 5.723704481124878,
      "epoch": 6.8638348551012305,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015247701861315427,
      "loss": 4.8927,
      "mean_token_accuracy": 0.20639290958642958,
      "num_tokens": 151523323.0,
      "step": 69160
    },
    {
      "entropy": 5.671932792663574,
      "epoch": 6.864331083763398,
      "grad_norm": 1.546875,
      "learning_rate": 0.00015244730265248375,
      "loss": 4.8478,
      "mean_token_accuracy": 0.20101711601018907,
      "num_tokens": 151533040.0,
      "step": 69165
    },
    {
      "entropy": 5.703259992599487,
      "epoch": 6.864827312425565,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001524175897308561,
      "loss": 4.9205,
      "mean_token_accuracy": 0.19853001683950425,
      "num_tokens": 151543593.0,
      "step": 69170
    },
    {
      "entropy": 5.775791311264038,
      "epoch": 6.865323541087733,
      "grad_norm": 1.34375,
      "learning_rate": 0.000152387879849008,
      "loss": 5.0045,
      "mean_token_accuracy": 0.1964867651462555,
      "num_tokens": 151554415.0,
      "step": 69175
    },
    {
      "entropy": 5.725537252426148,
      "epoch": 6.8658197697499,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001523581730076764,
      "loss": 4.8654,
      "mean_token_accuracy": 0.2023081049323082,
      "num_tokens": 151564213.0,
      "step": 69180
    },
    {
      "entropy": 5.6329680442810055,
      "epoch": 6.866315998412068,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001523284692075978,
      "loss": 4.7757,
      "mean_token_accuracy": 0.21410948038101196,
      "num_tokens": 151574513.0,
      "step": 69185
    },
    {
      "entropy": 5.646521520614624,
      "epoch": 6.866812227074236,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015229876844950882,
      "loss": 4.8297,
      "mean_token_accuracy": 0.21044315546751022,
      "num_tokens": 151585853.0,
      "step": 69190
    },
    {
      "entropy": 5.673357629776001,
      "epoch": 6.867308455736404,
      "grad_norm": 1.46875,
      "learning_rate": 0.000152269070734146,
      "loss": 4.9154,
      "mean_token_accuracy": 0.2035004898905754,
      "num_tokens": 151597636.0,
      "step": 69195
    },
    {
      "entropy": 5.686870431900024,
      "epoch": 6.867804684398571,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001522393760622457,
      "loss": 4.9092,
      "mean_token_accuracy": 0.21208426505327224,
      "num_tokens": 151608821.0,
      "step": 69200
    },
    {
      "entropy": 5.6845118522644045,
      "epoch": 6.868300913060739,
      "grad_norm": 1.515625,
      "learning_rate": 0.00015220968443454432,
      "loss": 4.9058,
      "mean_token_accuracy": 0.20016888678073883,
      "num_tokens": 151619091.0,
      "step": 69205
    },
    {
      "entropy": 5.6778953075408936,
      "epoch": 6.868797141722906,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00015217999585177827,
      "loss": 4.9038,
      "mean_token_accuracy": 0.1959816962480545,
      "num_tokens": 151629716.0,
      "step": 69210
    },
    {
      "entropy": 5.699000215530395,
      "epoch": 6.869293370385074,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015215031031468368,
      "loss": 4.8999,
      "mean_token_accuracy": 0.20877779871225358,
      "num_tokens": 151641482.0,
      "step": 69215
    },
    {
      "entropy": 5.687633323669433,
      "epoch": 6.869789599047241,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015212062782399665,
      "loss": 4.8985,
      "mean_token_accuracy": 0.20806677341461183,
      "num_tokens": 151651448.0,
      "step": 69220
    },
    {
      "entropy": 5.679198598861694,
      "epoch": 6.8702858277094085,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001520909483804533,
      "loss": 4.9221,
      "mean_token_accuracy": 0.2043036088347435,
      "num_tokens": 151661950.0,
      "step": 69225
    },
    {
      "entropy": 5.660561752319336,
      "epoch": 6.870782056371576,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015206127198478947,
      "loss": 4.881,
      "mean_token_accuracy": 0.20515020489692687,
      "num_tokens": 151674192.0,
      "step": 69230
    },
    {
      "entropy": 5.712720823287964,
      "epoch": 6.8712782850337435,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001520315986377412,
      "loss": 4.8538,
      "mean_token_accuracy": 0.202618870139122,
      "num_tokens": 151683811.0,
      "step": 69235
    },
    {
      "entropy": 5.662455224990845,
      "epoch": 6.871774513695911,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001520019283400444,
      "loss": 4.8797,
      "mean_token_accuracy": 0.20995544046163558,
      "num_tokens": 151695953.0,
      "step": 69240
    },
    {
      "entropy": 5.726804113388061,
      "epoch": 6.872270742358078,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00015197226109243473,
      "loss": 4.931,
      "mean_token_accuracy": 0.20437301695346832,
      "num_tokens": 151708167.0,
      "step": 69245
    },
    {
      "entropy": 5.730953168869019,
      "epoch": 6.872766971020246,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001519425968956479,
      "loss": 4.9373,
      "mean_token_accuracy": 0.19815810769796371,
      "num_tokens": 151718188.0,
      "step": 69250
    },
    {
      "entropy": 5.659829568862915,
      "epoch": 6.873263199682413,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015191293575041948,
      "loss": 4.896,
      "mean_token_accuracy": 0.20950496941804886,
      "num_tokens": 151727787.0,
      "step": 69255
    },
    {
      "entropy": 5.6765556812286375,
      "epoch": 6.873759428344581,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015188327765748493,
      "loss": 4.8776,
      "mean_token_accuracy": 0.20600256621837615,
      "num_tokens": 151739101.0,
      "step": 69260
    },
    {
      "entropy": 5.684288120269775,
      "epoch": 6.874255657006748,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015185362261757997,
      "loss": 4.8623,
      "mean_token_accuracy": 0.2074646919965744,
      "num_tokens": 151749955.0,
      "step": 69265
    },
    {
      "entropy": 5.805069923400879,
      "epoch": 6.874751885668916,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015182397063143963,
      "loss": 5.0426,
      "mean_token_accuracy": 0.19028931707143784,
      "num_tokens": 151760814.0,
      "step": 69270
    },
    {
      "entropy": 5.646119165420532,
      "epoch": 6.875248114331084,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00015179432169979952,
      "loss": 4.8616,
      "mean_token_accuracy": 0.20379243791103363,
      "num_tokens": 151772676.0,
      "step": 69275
    },
    {
      "entropy": 5.640384292602539,
      "epoch": 6.875744342993252,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015176467582339476,
      "loss": 4.8439,
      "mean_token_accuracy": 0.20956088304519654,
      "num_tokens": 151784311.0,
      "step": 69280
    },
    {
      "entropy": 5.674380540847778,
      "epoch": 6.876240571655419,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015173503300296048,
      "loss": 4.9116,
      "mean_token_accuracy": 0.1997099608182907,
      "num_tokens": 151795514.0,
      "step": 69285
    },
    {
      "entropy": 5.660791254043579,
      "epoch": 6.876736800317587,
      "grad_norm": 1.4375,
      "learning_rate": 0.00015170539323923165,
      "loss": 4.9155,
      "mean_token_accuracy": 0.2056147798895836,
      "num_tokens": 151806655.0,
      "step": 69290
    },
    {
      "entropy": 5.7046772003173825,
      "epoch": 6.877233028979754,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001516757565329435,
      "loss": 4.9046,
      "mean_token_accuracy": 0.20174833983182908,
      "num_tokens": 151817966.0,
      "step": 69295
    },
    {
      "entropy": 5.7253608226776125,
      "epoch": 6.877729257641922,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015164612288483077,
      "loss": 4.9815,
      "mean_token_accuracy": 0.19676946699619294,
      "num_tokens": 151830303.0,
      "step": 69300
    },
    {
      "entropy": 5.712686395645141,
      "epoch": 6.878225486304089,
      "grad_norm": 1.46875,
      "learning_rate": 0.00015161649229562848,
      "loss": 4.9272,
      "mean_token_accuracy": 0.19343694001436235,
      "num_tokens": 151841124.0,
      "step": 69305
    },
    {
      "entropy": 5.733162212371826,
      "epoch": 6.8787217149662565,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001515868647660713,
      "loss": 4.9509,
      "mean_token_accuracy": 0.20056668072938919,
      "num_tokens": 151852088.0,
      "step": 69310
    },
    {
      "entropy": 5.700450897216797,
      "epoch": 6.879217943628424,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00015155724029689393,
      "loss": 4.9202,
      "mean_token_accuracy": 0.2010957717895508,
      "num_tokens": 151862788.0,
      "step": 69315
    },
    {
      "entropy": 5.712308692932129,
      "epoch": 6.8797141722905915,
      "grad_norm": 1.4375,
      "learning_rate": 0.00015152761888883093,
      "loss": 4.9607,
      "mean_token_accuracy": 0.19486660361289979,
      "num_tokens": 151874385.0,
      "step": 69320
    },
    {
      "entropy": 5.7382975101470945,
      "epoch": 6.880210400952759,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015149800054261702,
      "loss": 4.9882,
      "mean_token_accuracy": 0.20232405215501786,
      "num_tokens": 151885473.0,
      "step": 69325
    },
    {
      "entropy": 5.724369287490845,
      "epoch": 6.880706629614926,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015146838525898659,
      "loss": 4.9179,
      "mean_token_accuracy": 0.20213195234537124,
      "num_tokens": 151896013.0,
      "step": 69330
    },
    {
      "entropy": 5.7018890380859375,
      "epoch": 6.881202858277094,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001514387730386739,
      "loss": 4.9465,
      "mean_token_accuracy": 0.19423777461051941,
      "num_tokens": 151906791.0,
      "step": 69335
    },
    {
      "entropy": 5.814878654479981,
      "epoch": 6.881699086939261,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001514091638824135,
      "loss": 5.0838,
      "mean_token_accuracy": 0.19195429980754852,
      "num_tokens": 151918035.0,
      "step": 69340
    },
    {
      "entropy": 5.710092306137085,
      "epoch": 6.882195315601429,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001513795577909395,
      "loss": 4.8721,
      "mean_token_accuracy": 0.20370811522006987,
      "num_tokens": 151928176.0,
      "step": 69345
    },
    {
      "entropy": 5.642452239990234,
      "epoch": 6.882691544263596,
      "grad_norm": 1.375,
      "learning_rate": 0.000151349954764986,
      "loss": 4.8574,
      "mean_token_accuracy": 0.20475270003080367,
      "num_tokens": 151939652.0,
      "step": 69350
    },
    {
      "entropy": 5.651111888885498,
      "epoch": 6.883187772925764,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015132035480528726,
      "loss": 4.8559,
      "mean_token_accuracy": 0.20553474575281144,
      "num_tokens": 151951459.0,
      "step": 69355
    },
    {
      "entropy": 5.678340196609497,
      "epoch": 6.883684001587932,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015129075791257723,
      "loss": 4.874,
      "mean_token_accuracy": 0.21091117560863495,
      "num_tokens": 151960754.0,
      "step": 69360
    },
    {
      "entropy": 5.702344655990601,
      "epoch": 6.8841802302501,
      "grad_norm": 1.4375,
      "learning_rate": 0.00015126116408758976,
      "loss": 4.8937,
      "mean_token_accuracy": 0.20066849142313004,
      "num_tokens": 151971254.0,
      "step": 69365
    },
    {
      "entropy": 5.670921182632446,
      "epoch": 6.884676458912267,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015123157333105886,
      "loss": 4.877,
      "mean_token_accuracy": 0.2053944230079651,
      "num_tokens": 151982696.0,
      "step": 69370
    },
    {
      "entropy": 5.6492900371551515,
      "epoch": 6.885172687574435,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015120198564371823,
      "loss": 4.9034,
      "mean_token_accuracy": 0.19884195029735566,
      "num_tokens": 151993606.0,
      "step": 69375
    },
    {
      "entropy": 5.681684732437134,
      "epoch": 6.885668916236602,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015117240102630153,
      "loss": 4.9047,
      "mean_token_accuracy": 0.20494787096977235,
      "num_tokens": 152004325.0,
      "step": 69380
    },
    {
      "entropy": 5.738049411773682,
      "epoch": 6.8861651448987695,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015114281947954257,
      "loss": 4.9118,
      "mean_token_accuracy": 0.20721019804477692,
      "num_tokens": 152016233.0,
      "step": 69385
    },
    {
      "entropy": 5.692684364318848,
      "epoch": 6.886661373560937,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015111324100417473,
      "loss": 4.9326,
      "mean_token_accuracy": 0.19849172234535217,
      "num_tokens": 152028089.0,
      "step": 69390
    },
    {
      "entropy": 5.7051104545593265,
      "epoch": 6.8871576022231045,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001510836656009315,
      "loss": 4.9518,
      "mean_token_accuracy": 0.19252048879861833,
      "num_tokens": 152037447.0,
      "step": 69395
    },
    {
      "entropy": 5.708495330810547,
      "epoch": 6.887653830885272,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00015105409327054647,
      "loss": 4.9224,
      "mean_token_accuracy": 0.20124002993106843,
      "num_tokens": 152048926.0,
      "step": 69400
    },
    {
      "entropy": 5.716771268844605,
      "epoch": 6.888150059547439,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001510245240137528,
      "loss": 4.9329,
      "mean_token_accuracy": 0.19884035736322403,
      "num_tokens": 152060185.0,
      "step": 69405
    },
    {
      "entropy": 5.774180793762207,
      "epoch": 6.888646288209607,
      "grad_norm": 1.125,
      "learning_rate": 0.00015099495783128374,
      "loss": 4.9685,
      "mean_token_accuracy": 0.20051054805517196,
      "num_tokens": 152073764.0,
      "step": 69410
    },
    {
      "entropy": 5.6993897438049315,
      "epoch": 6.889142516871774,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001509653947238726,
      "loss": 4.9253,
      "mean_token_accuracy": 0.19555336236953735,
      "num_tokens": 152084195.0,
      "step": 69415
    },
    {
      "entropy": 5.670470952987671,
      "epoch": 6.889638745533942,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015093583469225237,
      "loss": 4.9222,
      "mean_token_accuracy": 0.1989104226231575,
      "num_tokens": 152093481.0,
      "step": 69420
    },
    {
      "entropy": 5.712466239929199,
      "epoch": 6.890134974196109,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015090627773715616,
      "loss": 4.9276,
      "mean_token_accuracy": 0.20291792452335358,
      "num_tokens": 152104365.0,
      "step": 69425
    },
    {
      "entropy": 5.6886570930480955,
      "epoch": 6.890631202858277,
      "grad_norm": 1.4375,
      "learning_rate": 0.00015087672385931674,
      "loss": 4.9101,
      "mean_token_accuracy": 0.2018864318728447,
      "num_tokens": 152115508.0,
      "step": 69430
    },
    {
      "entropy": 5.73196234703064,
      "epoch": 6.891127431520444,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001508471730594672,
      "loss": 4.9192,
      "mean_token_accuracy": 0.1980472907423973,
      "num_tokens": 152126700.0,
      "step": 69435
    },
    {
      "entropy": 5.730788660049439,
      "epoch": 6.891623660182612,
      "grad_norm": 1.46875,
      "learning_rate": 0.00015081762533834014,
      "loss": 4.9238,
      "mean_token_accuracy": 0.20019254684448243,
      "num_tokens": 152136809.0,
      "step": 69440
    },
    {
      "entropy": 5.734889030456543,
      "epoch": 6.89211988884478,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015078808069666846,
      "loss": 4.8935,
      "mean_token_accuracy": 0.20041655153036117,
      "num_tokens": 152148773.0,
      "step": 69445
    },
    {
      "entropy": 5.730102300643921,
      "epoch": 6.892616117506948,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001507585391351848,
      "loss": 4.9262,
      "mean_token_accuracy": 0.19430140703916549,
      "num_tokens": 152158875.0,
      "step": 69450
    },
    {
      "entropy": 5.675013399124145,
      "epoch": 6.893112346169115,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001507290006546216,
      "loss": 4.9007,
      "mean_token_accuracy": 0.19820430129766464,
      "num_tokens": 152169578.0,
      "step": 69455
    },
    {
      "entropy": 5.665196180343628,
      "epoch": 6.8936085748312825,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015069946525571128,
      "loss": 4.8933,
      "mean_token_accuracy": 0.19794905185699463,
      "num_tokens": 152179457.0,
      "step": 69460
    },
    {
      "entropy": 5.657004690170288,
      "epoch": 6.89410480349345,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001506699329391864,
      "loss": 4.8887,
      "mean_token_accuracy": 0.20493444353342055,
      "num_tokens": 152190564.0,
      "step": 69465
    },
    {
      "entropy": 5.640244436264038,
      "epoch": 6.8946010321556175,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015064040370577937,
      "loss": 4.9212,
      "mean_token_accuracy": 0.20027729123830795,
      "num_tokens": 152200497.0,
      "step": 69470
    },
    {
      "entropy": 5.668671751022339,
      "epoch": 6.895097260817785,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00015061087755622237,
      "loss": 4.8914,
      "mean_token_accuracy": 0.19673671573400497,
      "num_tokens": 152212558.0,
      "step": 69475
    },
    {
      "entropy": 5.673755741119384,
      "epoch": 6.895593489479952,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015058135449124758,
      "loss": 4.9222,
      "mean_token_accuracy": 0.20341698676347733,
      "num_tokens": 152224512.0,
      "step": 69480
    },
    {
      "entropy": 5.724456787109375,
      "epoch": 6.89608971814212,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00015055183451158707,
      "loss": 4.9341,
      "mean_token_accuracy": 0.19787950366735457,
      "num_tokens": 152235767.0,
      "step": 69485
    },
    {
      "entropy": 5.776796817779541,
      "epoch": 6.896585946804287,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001505223176179729,
      "loss": 4.978,
      "mean_token_accuracy": 0.19510234892368317,
      "num_tokens": 152246398.0,
      "step": 69490
    },
    {
      "entropy": 5.773206567764282,
      "epoch": 6.897082175466455,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015049280381113696,
      "loss": 4.9854,
      "mean_token_accuracy": 0.1983938157558441,
      "num_tokens": 152258707.0,
      "step": 69495
    },
    {
      "entropy": 5.752345037460327,
      "epoch": 6.897578404128622,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00015046329309181122,
      "loss": 4.9625,
      "mean_token_accuracy": 0.19642294198274612,
      "num_tokens": 152268876.0,
      "step": 69500
    },
    {
      "entropy": 5.804137563705444,
      "epoch": 6.89807463279079,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015043378546072752,
      "loss": 5.0413,
      "mean_token_accuracy": 0.19394744336605071,
      "num_tokens": 152281135.0,
      "step": 69505
    },
    {
      "entropy": 5.710790824890137,
      "epoch": 6.898570861452957,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015040428091861753,
      "loss": 4.9218,
      "mean_token_accuracy": 0.20124260187149048,
      "num_tokens": 152292296.0,
      "step": 69510
    },
    {
      "entropy": 5.70199031829834,
      "epoch": 6.899067090115125,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015037477946621287,
      "loss": 4.9655,
      "mean_token_accuracy": 0.20503420680761336,
      "num_tokens": 152302498.0,
      "step": 69515
    },
    {
      "entropy": 5.705724763870239,
      "epoch": 6.899563318777292,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015034528110424516,
      "loss": 4.9404,
      "mean_token_accuracy": 0.199603608250618,
      "num_tokens": 152313302.0,
      "step": 69520
    },
    {
      "entropy": 5.61656379699707,
      "epoch": 6.90005954743946,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001503157858334458,
      "loss": 4.8156,
      "mean_token_accuracy": 0.2074965462088585,
      "num_tokens": 152324218.0,
      "step": 69525
    },
    {
      "entropy": 5.694287824630737,
      "epoch": 6.900555776101628,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015028629365454628,
      "loss": 4.9252,
      "mean_token_accuracy": 0.20333498120307922,
      "num_tokens": 152335018.0,
      "step": 69530
    },
    {
      "entropy": 5.7535295486450195,
      "epoch": 6.9010520047637955,
      "grad_norm": 1.515625,
      "learning_rate": 0.00015025680456827797,
      "loss": 4.9839,
      "mean_token_accuracy": 0.1944880411028862,
      "num_tokens": 152345462.0,
      "step": 69535
    },
    {
      "entropy": 5.737381172180176,
      "epoch": 6.901548233425963,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015022731857537215,
      "loss": 4.9453,
      "mean_token_accuracy": 0.20285546481609346,
      "num_tokens": 152357497.0,
      "step": 69540
    },
    {
      "entropy": 5.720609855651856,
      "epoch": 6.9020444620881305,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015019783567656,
      "loss": 4.9039,
      "mean_token_accuracy": 0.20143382102251053,
      "num_tokens": 152367526.0,
      "step": 69545
    },
    {
      "entropy": 5.68066897392273,
      "epoch": 6.902540690750298,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015016835587257254,
      "loss": 4.9055,
      "mean_token_accuracy": 0.20829005539417267,
      "num_tokens": 152379340.0,
      "step": 69550
    },
    {
      "entropy": 5.6754951000213625,
      "epoch": 6.9030369194124654,
      "grad_norm": 1.375,
      "learning_rate": 0.00015013887916414078,
      "loss": 4.8368,
      "mean_token_accuracy": 0.20905135422945023,
      "num_tokens": 152389845.0,
      "step": 69555
    },
    {
      "entropy": 5.69297194480896,
      "epoch": 6.903533148074633,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001501094055519959,
      "loss": 4.9871,
      "mean_token_accuracy": 0.19662947654724122,
      "num_tokens": 152400648.0,
      "step": 69560
    },
    {
      "entropy": 5.7644579887390135,
      "epoch": 6.9040293767368,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001500799350368685,
      "loss": 4.9891,
      "mean_token_accuracy": 0.20317918807268143,
      "num_tokens": 152411567.0,
      "step": 69565
    },
    {
      "entropy": 5.693855714797974,
      "epoch": 6.904525605398968,
      "grad_norm": 1.375,
      "learning_rate": 0.0001500504676194897,
      "loss": 4.9256,
      "mean_token_accuracy": 0.19483787417411805,
      "num_tokens": 152421047.0,
      "step": 69570
    },
    {
      "entropy": 5.772578954696655,
      "epoch": 6.905021834061135,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015002100330059,
      "loss": 5.003,
      "mean_token_accuracy": 0.19040878415107726,
      "num_tokens": 152432527.0,
      "step": 69575
    },
    {
      "entropy": 5.775347805023193,
      "epoch": 6.905518062723303,
      "grad_norm": 1.5625,
      "learning_rate": 0.00014999154208090017,
      "loss": 4.907,
      "mean_token_accuracy": 0.20110656321048737,
      "num_tokens": 152442957.0,
      "step": 69580
    },
    {
      "entropy": 5.8090849876403805,
      "epoch": 6.90601429138547,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014996208396115057,
      "loss": 4.9982,
      "mean_token_accuracy": 0.19897086471319197,
      "num_tokens": 152454744.0,
      "step": 69585
    },
    {
      "entropy": 5.639768743515015,
      "epoch": 6.906510520047638,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014993262894207193,
      "loss": 4.882,
      "mean_token_accuracy": 0.21180897057056428,
      "num_tokens": 152465839.0,
      "step": 69590
    },
    {
      "entropy": 5.669004344940186,
      "epoch": 6.907006748709805,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001499031770243946,
      "loss": 4.9127,
      "mean_token_accuracy": 0.20085413604974747,
      "num_tokens": 152478198.0,
      "step": 69595
    },
    {
      "entropy": 5.704383182525635,
      "epoch": 6.907502977371973,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014987372820884893,
      "loss": 4.9488,
      "mean_token_accuracy": 0.20441316515207292,
      "num_tokens": 152488552.0,
      "step": 69600
    },
    {
      "entropy": 5.644125938415527,
      "epoch": 6.90799920603414,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001498442824961652,
      "loss": 4.8371,
      "mean_token_accuracy": 0.20230243504047393,
      "num_tokens": 152499113.0,
      "step": 69605
    },
    {
      "entropy": 5.776827430725097,
      "epoch": 6.908495434696308,
      "grad_norm": 1.375,
      "learning_rate": 0.0001498148398870736,
      "loss": 4.9744,
      "mean_token_accuracy": 0.20165885835886002,
      "num_tokens": 152509811.0,
      "step": 69610
    },
    {
      "entropy": 5.757367086410523,
      "epoch": 6.908991663358476,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001497854003823041,
      "loss": 4.9942,
      "mean_token_accuracy": 0.19742591828107833,
      "num_tokens": 152519565.0,
      "step": 69615
    },
    {
      "entropy": 5.706817770004273,
      "epoch": 6.909487892020643,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014975596398258698,
      "loss": 4.9357,
      "mean_token_accuracy": 0.19878875017166137,
      "num_tokens": 152531140.0,
      "step": 69620
    },
    {
      "entropy": 5.66649055480957,
      "epoch": 6.909984120682811,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014972653068865205,
      "loss": 4.8536,
      "mean_token_accuracy": 0.2047546997666359,
      "num_tokens": 152541174.0,
      "step": 69625
    },
    {
      "entropy": 5.698971796035766,
      "epoch": 6.9104803493449785,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014969710050122915,
      "loss": 4.9109,
      "mean_token_accuracy": 0.1998800665140152,
      "num_tokens": 152550967.0,
      "step": 69630
    },
    {
      "entropy": 5.653390979766845,
      "epoch": 6.910976578007146,
      "grad_norm": 1.5,
      "learning_rate": 0.0001496676734210482,
      "loss": 4.8246,
      "mean_token_accuracy": 0.20980919450521468,
      "num_tokens": 152559887.0,
      "step": 69635
    },
    {
      "entropy": 5.735733699798584,
      "epoch": 6.911472806669313,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014963824944883892,
      "loss": 4.9336,
      "mean_token_accuracy": 0.20071753859519958,
      "num_tokens": 152571380.0,
      "step": 69640
    },
    {
      "entropy": 5.706221246719361,
      "epoch": 6.911969035331481,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014960882858533082,
      "loss": 4.9256,
      "mean_token_accuracy": 0.20365042388439178,
      "num_tokens": 152582112.0,
      "step": 69645
    },
    {
      "entropy": 5.673555707931518,
      "epoch": 6.912465263993648,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001495794108312537,
      "loss": 4.8985,
      "mean_token_accuracy": 0.20293698757886885,
      "num_tokens": 152592728.0,
      "step": 69650
    },
    {
      "entropy": 5.688075017929077,
      "epoch": 6.912961492655816,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001495499961873369,
      "loss": 4.9552,
      "mean_token_accuracy": 0.19539813697338104,
      "num_tokens": 152603422.0,
      "step": 69655
    },
    {
      "entropy": 5.702299404144287,
      "epoch": 6.913457721317983,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001495205846543098,
      "loss": 4.9282,
      "mean_token_accuracy": 0.20214511454105377,
      "num_tokens": 152614823.0,
      "step": 69660
    },
    {
      "entropy": 5.6975585460662845,
      "epoch": 6.913953949980151,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014949117623290188,
      "loss": 4.8735,
      "mean_token_accuracy": 0.20348242968320845,
      "num_tokens": 152625356.0,
      "step": 69665
    },
    {
      "entropy": 5.656325435638427,
      "epoch": 6.914450178642318,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014946177092384233,
      "loss": 4.9015,
      "mean_token_accuracy": 0.19369888305664062,
      "num_tokens": 152636089.0,
      "step": 69670
    },
    {
      "entropy": 5.729442310333252,
      "epoch": 6.914946407304486,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014943236872786033,
      "loss": 5.0131,
      "mean_token_accuracy": 0.19124689400196077,
      "num_tokens": 152647296.0,
      "step": 69675
    },
    {
      "entropy": 5.710827732086182,
      "epoch": 6.915442635966653,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014940296964568504,
      "loss": 4.965,
      "mean_token_accuracy": 0.19685198366641998,
      "num_tokens": 152659767.0,
      "step": 69680
    },
    {
      "entropy": 5.6559864521026615,
      "epoch": 6.915938864628821,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001493735736780455,
      "loss": 4.8988,
      "mean_token_accuracy": 0.20287245213985444,
      "num_tokens": 152671787.0,
      "step": 69685
    },
    {
      "entropy": 5.6337902545928955,
      "epoch": 6.916435093290988,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014934418082567053,
      "loss": 4.8142,
      "mean_token_accuracy": 0.21162479966878892,
      "num_tokens": 152682871.0,
      "step": 69690
    },
    {
      "entropy": 5.701126909255981,
      "epoch": 6.916931321953156,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014931479108928913,
      "loss": 5.0052,
      "mean_token_accuracy": 0.19369610399007797,
      "num_tokens": 152694270.0,
      "step": 69695
    },
    {
      "entropy": 5.669511938095093,
      "epoch": 6.917427550615324,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001492854044696301,
      "loss": 4.924,
      "mean_token_accuracy": 0.19078043550252916,
      "num_tokens": 152704002.0,
      "step": 69700
    },
    {
      "entropy": 5.758019304275512,
      "epoch": 6.917923779277491,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001492560209674222,
      "loss": 4.9927,
      "mean_token_accuracy": 0.19901542067527772,
      "num_tokens": 152714944.0,
      "step": 69705
    },
    {
      "entropy": 5.759227848052978,
      "epoch": 6.918420007939659,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014922664058339403,
      "loss": 4.9832,
      "mean_token_accuracy": 0.2005864441394806,
      "num_tokens": 152725994.0,
      "step": 69710
    },
    {
      "entropy": 5.677873373031616,
      "epoch": 6.918916236601826,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014919726331827417,
      "loss": 4.855,
      "mean_token_accuracy": 0.20115475803613664,
      "num_tokens": 152737011.0,
      "step": 69715
    },
    {
      "entropy": 5.724612474441528,
      "epoch": 6.919412465263994,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014916788917279102,
      "loss": 4.9223,
      "mean_token_accuracy": 0.2010233536362648,
      "num_tokens": 152748312.0,
      "step": 69720
    },
    {
      "entropy": 5.759042549133301,
      "epoch": 6.919908693926161,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014913851814767306,
      "loss": 5.0242,
      "mean_token_accuracy": 0.19872656613588333,
      "num_tokens": 152759646.0,
      "step": 69725
    },
    {
      "entropy": 5.690159034729004,
      "epoch": 6.920404922588329,
      "grad_norm": 1.203125,
      "learning_rate": 0.00014910915024364866,
      "loss": 4.9078,
      "mean_token_accuracy": 0.2075363039970398,
      "num_tokens": 152770912.0,
      "step": 69730
    },
    {
      "entropy": 5.70196385383606,
      "epoch": 6.920901151250496,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001490797854614461,
      "loss": 4.8869,
      "mean_token_accuracy": 0.20186635702848435,
      "num_tokens": 152780624.0,
      "step": 69735
    },
    {
      "entropy": 5.664204216003418,
      "epoch": 6.921397379912664,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014905042380179356,
      "loss": 4.8859,
      "mean_token_accuracy": 0.19491310864686967,
      "num_tokens": 152791226.0,
      "step": 69740
    },
    {
      "entropy": 5.642850399017334,
      "epoch": 6.921893608574831,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001490210652654191,
      "loss": 4.895,
      "mean_token_accuracy": 0.19669412672519684,
      "num_tokens": 152801323.0,
      "step": 69745
    },
    {
      "entropy": 5.69606647491455,
      "epoch": 6.922389837236999,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014899170985305078,
      "loss": 4.898,
      "mean_token_accuracy": 0.20978339314460753,
      "num_tokens": 152812014.0,
      "step": 69750
    },
    {
      "entropy": 5.6739434719085695,
      "epoch": 6.922886065899166,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001489623575654164,
      "loss": 4.8929,
      "mean_token_accuracy": 0.2066900059580803,
      "num_tokens": 152821383.0,
      "step": 69755
    },
    {
      "entropy": 5.717477750778198,
      "epoch": 6.923382294561334,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014893300840324397,
      "loss": 4.9713,
      "mean_token_accuracy": 0.19940100163221358,
      "num_tokens": 152831626.0,
      "step": 69760
    },
    {
      "entropy": 5.682738304138184,
      "epoch": 6.923878523223501,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001489036623672614,
      "loss": 4.8777,
      "mean_token_accuracy": 0.2040489137172699,
      "num_tokens": 152841812.0,
      "step": 69765
    },
    {
      "entropy": 5.744375991821289,
      "epoch": 6.924374751885669,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014887431945819626,
      "loss": 4.9116,
      "mean_token_accuracy": 0.20465498566627502,
      "num_tokens": 152851473.0,
      "step": 69770
    },
    {
      "entropy": 5.706078720092774,
      "epoch": 6.924870980547836,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014884497967677623,
      "loss": 4.9478,
      "mean_token_accuracy": 0.20372089445590974,
      "num_tokens": 152862269.0,
      "step": 69775
    },
    {
      "entropy": 5.685477066040039,
      "epoch": 6.925367209210004,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001488156430237289,
      "loss": 4.9091,
      "mean_token_accuracy": 0.20008111894130706,
      "num_tokens": 152874783.0,
      "step": 69780
    },
    {
      "entropy": 5.729758548736572,
      "epoch": 6.925863437872172,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014878630949978166,
      "loss": 4.9293,
      "mean_token_accuracy": 0.19750369787216188,
      "num_tokens": 152885371.0,
      "step": 69785
    },
    {
      "entropy": 5.7714245319366455,
      "epoch": 6.9263596665343385,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00014875697910566194,
      "loss": 4.9651,
      "mean_token_accuracy": 0.19975923299789428,
      "num_tokens": 152895149.0,
      "step": 69790
    },
    {
      "entropy": 5.680448055267334,
      "epoch": 6.926855895196507,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014872765184209712,
      "loss": 4.8619,
      "mean_token_accuracy": 0.20228481888771058,
      "num_tokens": 152905798.0,
      "step": 69795
    },
    {
      "entropy": 5.716137933731079,
      "epoch": 6.927352123858674,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014869832770981448,
      "loss": 4.9004,
      "mean_token_accuracy": 0.20324934720993043,
      "num_tokens": 152915687.0,
      "step": 69800
    },
    {
      "entropy": 5.6876866817474365,
      "epoch": 6.927848352520842,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014866900670954116,
      "loss": 4.9331,
      "mean_token_accuracy": 0.1989184647798538,
      "num_tokens": 152927101.0,
      "step": 69805
    },
    {
      "entropy": 5.667795658111572,
      "epoch": 6.928344581183009,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014863968884200427,
      "loss": 4.8292,
      "mean_token_accuracy": 0.2048589050769806,
      "num_tokens": 152937956.0,
      "step": 69810
    },
    {
      "entropy": 5.718955659866333,
      "epoch": 6.928840809845177,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014861037410793083,
      "loss": 4.914,
      "mean_token_accuracy": 0.19264829754829407,
      "num_tokens": 152949018.0,
      "step": 69815
    },
    {
      "entropy": 5.671893501281739,
      "epoch": 6.929337038507344,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014858106250804765,
      "loss": 4.9205,
      "mean_token_accuracy": 0.20304385721683502,
      "num_tokens": 152959467.0,
      "step": 69820
    },
    {
      "entropy": 5.670633745193482,
      "epoch": 6.929833267169512,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014855175404308174,
      "loss": 4.9011,
      "mean_token_accuracy": 0.20423428863286971,
      "num_tokens": 152970896.0,
      "step": 69825
    },
    {
      "entropy": 5.655493450164795,
      "epoch": 6.930329495831679,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001485224487137599,
      "loss": 4.8604,
      "mean_token_accuracy": 0.20070720613002777,
      "num_tokens": 152981537.0,
      "step": 69830
    },
    {
      "entropy": 5.674578714370727,
      "epoch": 6.930825724493847,
      "grad_norm": 1.5,
      "learning_rate": 0.00014849314652080886,
      "loss": 4.8485,
      "mean_token_accuracy": 0.20624559372663498,
      "num_tokens": 152990811.0,
      "step": 69835
    },
    {
      "entropy": 5.669294118881226,
      "epoch": 6.931321953156014,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014846384746495514,
      "loss": 4.8781,
      "mean_token_accuracy": 0.2062063530087471,
      "num_tokens": 153001351.0,
      "step": 69840
    },
    {
      "entropy": 5.73016505241394,
      "epoch": 6.931818181818182,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014843455154692537,
      "loss": 4.9159,
      "mean_token_accuracy": 0.19971904009580613,
      "num_tokens": 153013323.0,
      "step": 69845
    },
    {
      "entropy": 5.708191585540772,
      "epoch": 6.932314410480349,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014840525876744587,
      "loss": 4.9311,
      "mean_token_accuracy": 0.19688310623168945,
      "num_tokens": 153023337.0,
      "step": 69850
    },
    {
      "entropy": 5.686050653457642,
      "epoch": 6.932810639142517,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001483759691272433,
      "loss": 4.9443,
      "mean_token_accuracy": 0.19528744220733643,
      "num_tokens": 153033752.0,
      "step": 69855
    },
    {
      "entropy": 5.707064867019653,
      "epoch": 6.933306867804684,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014834668262704375,
      "loss": 4.919,
      "mean_token_accuracy": 0.2004827082157135,
      "num_tokens": 153043871.0,
      "step": 69860
    },
    {
      "entropy": 5.676458883285522,
      "epoch": 6.933803096466852,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014831739926757362,
      "loss": 4.9613,
      "mean_token_accuracy": 0.20169543474912643,
      "num_tokens": 153054972.0,
      "step": 69865
    },
    {
      "entropy": 5.683892869949341,
      "epoch": 6.93429932512902,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014828811904955903,
      "loss": 4.9126,
      "mean_token_accuracy": 0.20352295488119126,
      "num_tokens": 153067032.0,
      "step": 69870
    },
    {
      "entropy": 5.728358173370362,
      "epoch": 6.9347955537911865,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00014825884197372605,
      "loss": 4.9157,
      "mean_token_accuracy": 0.19559111893177034,
      "num_tokens": 153077655.0,
      "step": 69875
    },
    {
      "entropy": 5.651804876327515,
      "epoch": 6.935291782453355,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001482295680408006,
      "loss": 4.8983,
      "mean_token_accuracy": 0.20425187647342682,
      "num_tokens": 153087637.0,
      "step": 69880
    },
    {
      "entropy": 5.677764415740967,
      "epoch": 6.935788011115522,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014820029725150875,
      "loss": 4.9067,
      "mean_token_accuracy": 0.20330181568861008,
      "num_tokens": 153098359.0,
      "step": 69885
    },
    {
      "entropy": 5.651593208312988,
      "epoch": 6.93628423977769,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014817102960657625,
      "loss": 4.8341,
      "mean_token_accuracy": 0.21063929051160812,
      "num_tokens": 153109447.0,
      "step": 69890
    },
    {
      "entropy": 5.61116213798523,
      "epoch": 6.936780468439857,
      "grad_norm": 1.328125,
      "learning_rate": 0.000148141765106729,
      "loss": 4.8208,
      "mean_token_accuracy": 0.20434913784265518,
      "num_tokens": 153121739.0,
      "step": 69895
    },
    {
      "entropy": 5.672968435287475,
      "epoch": 6.937276697102025,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001481125037526926,
      "loss": 4.8479,
      "mean_token_accuracy": 0.2029753088951111,
      "num_tokens": 153133187.0,
      "step": 69900
    },
    {
      "entropy": 5.730524682998658,
      "epoch": 6.937772925764192,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001480832455451927,
      "loss": 4.9432,
      "mean_token_accuracy": 0.20434309244155885,
      "num_tokens": 153143777.0,
      "step": 69905
    },
    {
      "entropy": 5.658534908294678,
      "epoch": 6.93826915442636,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014805399048495478,
      "loss": 4.8855,
      "mean_token_accuracy": 0.20467755645513536,
      "num_tokens": 153155262.0,
      "step": 69910
    },
    {
      "entropy": 5.705502080917358,
      "epoch": 6.938765383088527,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014802473857270437,
      "loss": 4.9132,
      "mean_token_accuracy": 0.20034584701061248,
      "num_tokens": 153167018.0,
      "step": 69915
    },
    {
      "entropy": 5.636715936660766,
      "epoch": 6.939261611750695,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014799548980916683,
      "loss": 4.8126,
      "mean_token_accuracy": 0.20314988046884536,
      "num_tokens": 153177172.0,
      "step": 69920
    },
    {
      "entropy": 5.6663001537322994,
      "epoch": 6.939757840412862,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001479662441950675,
      "loss": 4.8726,
      "mean_token_accuracy": 0.20648206472396852,
      "num_tokens": 153186882.0,
      "step": 69925
    },
    {
      "entropy": 5.66697506904602,
      "epoch": 6.94025406907503,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014793700173113157,
      "loss": 4.8651,
      "mean_token_accuracy": 0.20582276135683059,
      "num_tokens": 153196725.0,
      "step": 69930
    },
    {
      "entropy": 5.685527753829956,
      "epoch": 6.940750297737197,
      "grad_norm": 1.234375,
      "learning_rate": 0.00014790776241808414,
      "loss": 4.8969,
      "mean_token_accuracy": 0.19601932018995286,
      "num_tokens": 153208677.0,
      "step": 69935
    },
    {
      "entropy": 5.695730352401734,
      "epoch": 6.941246526399365,
      "grad_norm": 1.5,
      "learning_rate": 0.00014787852625665045,
      "loss": 4.8842,
      "mean_token_accuracy": 0.20417866259813308,
      "num_tokens": 153219579.0,
      "step": 69940
    },
    {
      "entropy": 5.745663785934449,
      "epoch": 6.941742755061532,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014784929324755533,
      "loss": 4.9585,
      "mean_token_accuracy": 0.1961835503578186,
      "num_tokens": 153231768.0,
      "step": 69945
    },
    {
      "entropy": 5.646796464920044,
      "epoch": 6.9422389837236995,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001478200633915238,
      "loss": 4.8825,
      "mean_token_accuracy": 0.20411352813243866,
      "num_tokens": 153242746.0,
      "step": 69950
    },
    {
      "entropy": 5.751903676986695,
      "epoch": 6.942735212385868,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014779083668928057,
      "loss": 4.9359,
      "mean_token_accuracy": 0.19485824257135392,
      "num_tokens": 153252708.0,
      "step": 69955
    },
    {
      "entropy": 5.673729515075683,
      "epoch": 6.9432314410480345,
      "grad_norm": 1.375,
      "learning_rate": 0.00014776161314155052,
      "loss": 4.8632,
      "mean_token_accuracy": 0.19911507219076158,
      "num_tokens": 153263461.0,
      "step": 69960
    },
    {
      "entropy": 5.654640436172485,
      "epoch": 6.943727669710203,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014773239274905827,
      "loss": 4.8372,
      "mean_token_accuracy": 0.2048186868429184,
      "num_tokens": 153273766.0,
      "step": 69965
    },
    {
      "entropy": 5.61354250907898,
      "epoch": 6.94422389837237,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014770317551252845,
      "loss": 4.8839,
      "mean_token_accuracy": 0.21211165189743042,
      "num_tokens": 153285798.0,
      "step": 69970
    },
    {
      "entropy": 5.613333654403687,
      "epoch": 6.944720127034538,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001476739614326856,
      "loss": 4.8908,
      "mean_token_accuracy": 0.1996266096830368,
      "num_tokens": 153296802.0,
      "step": 69975
    },
    {
      "entropy": 5.684940958023072,
      "epoch": 6.945216355696705,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014764475051025418,
      "loss": 4.9082,
      "mean_token_accuracy": 0.1967089667916298,
      "num_tokens": 153308309.0,
      "step": 69980
    },
    {
      "entropy": 5.662377214431762,
      "epoch": 6.945712584358873,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001476155427459584,
      "loss": 4.9322,
      "mean_token_accuracy": 0.20293236374855042,
      "num_tokens": 153320084.0,
      "step": 69985
    },
    {
      "entropy": 5.76266884803772,
      "epoch": 6.94620881302104,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001475863381405228,
      "loss": 4.9749,
      "mean_token_accuracy": 0.19064807444810866,
      "num_tokens": 153331355.0,
      "step": 69990
    },
    {
      "entropy": 5.710720872879028,
      "epoch": 6.946705041683208,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001475571366946713,
      "loss": 4.8989,
      "mean_token_accuracy": 0.20568737834692002,
      "num_tokens": 153341416.0,
      "step": 69995
    },
    {
      "entropy": 5.721532773971558,
      "epoch": 6.947201270345375,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00014752793840912834,
      "loss": 4.9364,
      "mean_token_accuracy": 0.20659370124340057,
      "num_tokens": 153353174.0,
      "step": 70000
    },
    {
      "entropy": 5.679260206222534,
      "epoch": 6.947697499007543,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014749874328461783,
      "loss": 4.8832,
      "mean_token_accuracy": 0.20409148931503296,
      "num_tokens": 153364877.0,
      "step": 70005
    },
    {
      "entropy": 5.696797943115234,
      "epoch": 6.94819372766971,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014746955132186374,
      "loss": 4.8946,
      "mean_token_accuracy": 0.20136706829071044,
      "num_tokens": 153376648.0,
      "step": 70010
    },
    {
      "entropy": 5.724615669250488,
      "epoch": 6.948689956331878,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014744036252158993,
      "loss": 4.9721,
      "mean_token_accuracy": 0.19621584713459014,
      "num_tokens": 153387646.0,
      "step": 70015
    },
    {
      "entropy": 5.650956010818481,
      "epoch": 6.949186184994045,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001474111768845202,
      "loss": 4.8631,
      "mean_token_accuracy": 0.20510344505310057,
      "num_tokens": 153397945.0,
      "step": 70020
    },
    {
      "entropy": 5.669628524780274,
      "epoch": 6.9496824136562125,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014738199441137839,
      "loss": 4.8821,
      "mean_token_accuracy": 0.19918419867753984,
      "num_tokens": 153409168.0,
      "step": 70025
    },
    {
      "entropy": 5.677310609817505,
      "epoch": 6.95017864231838,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014735281510288816,
      "loss": 4.8721,
      "mean_token_accuracy": 0.20762566328048707,
      "num_tokens": 153420689.0,
      "step": 70030
    },
    {
      "entropy": 5.662202453613281,
      "epoch": 6.9506748709805475,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014732363895977312,
      "loss": 4.9089,
      "mean_token_accuracy": 0.20080201029777528,
      "num_tokens": 153431568.0,
      "step": 70035
    },
    {
      "entropy": 5.679225635528565,
      "epoch": 6.951171099642716,
      "grad_norm": 1.234375,
      "learning_rate": 0.00014729446598275668,
      "loss": 4.8703,
      "mean_token_accuracy": 0.20743533819913865,
      "num_tokens": 153444271.0,
      "step": 70040
    },
    {
      "entropy": 5.739187812805175,
      "epoch": 6.951667328304882,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001472652961725623,
      "loss": 4.9491,
      "mean_token_accuracy": 0.20289022028446196,
      "num_tokens": 153456155.0,
      "step": 70045
    },
    {
      "entropy": 5.712235975265503,
      "epoch": 6.952163556967051,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014723612952991324,
      "loss": 4.8782,
      "mean_token_accuracy": 0.2044766515493393,
      "num_tokens": 153466492.0,
      "step": 70050
    },
    {
      "entropy": 5.66987247467041,
      "epoch": 6.952659785629218,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001472069660555329,
      "loss": 4.899,
      "mean_token_accuracy": 0.20770813375711442,
      "num_tokens": 153477619.0,
      "step": 70055
    },
    {
      "entropy": 5.720232152938843,
      "epoch": 6.953156014291386,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014717780575014447,
      "loss": 4.8941,
      "mean_token_accuracy": 0.2036364495754242,
      "num_tokens": 153488363.0,
      "step": 70060
    },
    {
      "entropy": 5.710514402389526,
      "epoch": 6.953652242953553,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014714864861447102,
      "loss": 4.9163,
      "mean_token_accuracy": 0.2008448362350464,
      "num_tokens": 153498696.0,
      "step": 70065
    },
    {
      "entropy": 5.689071178436279,
      "epoch": 6.954148471615721,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014711949464923563,
      "loss": 4.8961,
      "mean_token_accuracy": 0.20280113369226455,
      "num_tokens": 153510067.0,
      "step": 70070
    },
    {
      "entropy": 5.6611183166503904,
      "epoch": 6.954644700277888,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014709034385516117,
      "loss": 4.9285,
      "mean_token_accuracy": 0.20251651257276534,
      "num_tokens": 153521656.0,
      "step": 70075
    },
    {
      "entropy": 5.685284996032715,
      "epoch": 6.955140928940056,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014706119623297066,
      "loss": 4.8907,
      "mean_token_accuracy": 0.20666783004999162,
      "num_tokens": 153532749.0,
      "step": 70080
    },
    {
      "entropy": 5.7054565906524655,
      "epoch": 6.955637157602223,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001470320517833866,
      "loss": 4.9275,
      "mean_token_accuracy": 0.1978885531425476,
      "num_tokens": 153544536.0,
      "step": 70085
    },
    {
      "entropy": 5.713105773925781,
      "epoch": 6.956133386264391,
      "grad_norm": 1.375,
      "learning_rate": 0.00014700291050713196,
      "loss": 4.8959,
      "mean_token_accuracy": 0.19575773775577546,
      "num_tokens": 153555657.0,
      "step": 70090
    },
    {
      "entropy": 5.697168350219727,
      "epoch": 6.956629614926558,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014697377240492944,
      "loss": 4.8707,
      "mean_token_accuracy": 0.20520184487104415,
      "num_tokens": 153567621.0,
      "step": 70095
    },
    {
      "entropy": 5.717600536346436,
      "epoch": 6.9571258435887255,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00014694463747750143,
      "loss": 4.9435,
      "mean_token_accuracy": 0.19947640150785445,
      "num_tokens": 153577642.0,
      "step": 70100
    },
    {
      "entropy": 5.681799793243409,
      "epoch": 6.957622072250893,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001469155057255705,
      "loss": 4.9231,
      "mean_token_accuracy": 0.20152250528335572,
      "num_tokens": 153587872.0,
      "step": 70105
    },
    {
      "entropy": 5.692588376998901,
      "epoch": 6.9581183009130605,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014688637714985905,
      "loss": 4.9205,
      "mean_token_accuracy": 0.19901377856731414,
      "num_tokens": 153598490.0,
      "step": 70110
    },
    {
      "entropy": 5.667323398590088,
      "epoch": 6.958614529575228,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001468572517510893,
      "loss": 4.8089,
      "mean_token_accuracy": 0.20281816869974137,
      "num_tokens": 153609874.0,
      "step": 70115
    },
    {
      "entropy": 5.677973031997681,
      "epoch": 6.9591107582373954,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014682812952998358,
      "loss": 4.9087,
      "mean_token_accuracy": 0.19952037632465364,
      "num_tokens": 153621963.0,
      "step": 70120
    },
    {
      "entropy": 5.680132436752319,
      "epoch": 6.959606986899564,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014679901048726414,
      "loss": 4.8355,
      "mean_token_accuracy": 0.21035873144865036,
      "num_tokens": 153632867.0,
      "step": 70125
    },
    {
      "entropy": 5.709666776657104,
      "epoch": 6.96010321556173,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014676989462365298,
      "loss": 4.9146,
      "mean_token_accuracy": 0.2005647152662277,
      "num_tokens": 153644176.0,
      "step": 70130
    },
    {
      "entropy": 5.708824872970581,
      "epoch": 6.960599444223899,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001467407819398721,
      "loss": 4.9599,
      "mean_token_accuracy": 0.2019816368818283,
      "num_tokens": 153655769.0,
      "step": 70135
    },
    {
      "entropy": 5.681751918792725,
      "epoch": 6.961095672886066,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014671167243664353,
      "loss": 4.8666,
      "mean_token_accuracy": 0.2044620022177696,
      "num_tokens": 153667201.0,
      "step": 70140
    },
    {
      "entropy": 5.736843204498291,
      "epoch": 6.961591901548234,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014668256611468888,
      "loss": 4.9613,
      "mean_token_accuracy": 0.2025442436337471,
      "num_tokens": 153679079.0,
      "step": 70145
    },
    {
      "entropy": 5.667437934875489,
      "epoch": 6.962088130210401,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014665346297473015,
      "loss": 4.8951,
      "mean_token_accuracy": 0.19782600402832032,
      "num_tokens": 153690822.0,
      "step": 70150
    },
    {
      "entropy": 5.620041704177856,
      "epoch": 6.962584358872569,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014662436301748905,
      "loss": 4.8501,
      "mean_token_accuracy": 0.2089335501194,
      "num_tokens": 153701541.0,
      "step": 70155
    },
    {
      "entropy": 5.731464910507202,
      "epoch": 6.963080587534736,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001465952662436871,
      "loss": 5.0258,
      "mean_token_accuracy": 0.19500327706336976,
      "num_tokens": 153713414.0,
      "step": 70160
    },
    {
      "entropy": 5.699132585525513,
      "epoch": 6.963576816196904,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014656617265404584,
      "loss": 4.871,
      "mean_token_accuracy": 0.20125218629837036,
      "num_tokens": 153724462.0,
      "step": 70165
    },
    {
      "entropy": 5.706181001663208,
      "epoch": 6.964073044859071,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001465370822492868,
      "loss": 4.9793,
      "mean_token_accuracy": 0.1985140323638916,
      "num_tokens": 153737088.0,
      "step": 70170
    },
    {
      "entropy": 5.732731628417969,
      "epoch": 6.964569273521239,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001465079950301312,
      "loss": 4.8916,
      "mean_token_accuracy": 0.2035268560051918,
      "num_tokens": 153749033.0,
      "step": 70175
    },
    {
      "entropy": 5.723353576660156,
      "epoch": 6.965065502183406,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014647891099730053,
      "loss": 4.9634,
      "mean_token_accuracy": 0.20013850182294846,
      "num_tokens": 153760547.0,
      "step": 70180
    },
    {
      "entropy": 5.724623107910157,
      "epoch": 6.9655617308455735,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014644983015151585,
      "loss": 4.9623,
      "mean_token_accuracy": 0.20081560015678407,
      "num_tokens": 153771342.0,
      "step": 70185
    },
    {
      "entropy": 5.704903984069825,
      "epoch": 6.966057959507741,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014642075249349845,
      "loss": 4.917,
      "mean_token_accuracy": 0.18891186118125916,
      "num_tokens": 153782208.0,
      "step": 70190
    },
    {
      "entropy": 5.716779470443726,
      "epoch": 6.9665541881699085,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014639167802396935,
      "loss": 4.9265,
      "mean_token_accuracy": 0.19646005481481552,
      "num_tokens": 153793878.0,
      "step": 70195
    },
    {
      "entropy": 5.694207239151001,
      "epoch": 6.967050416832076,
      "grad_norm": 1.234375,
      "learning_rate": 0.00014636260674364944,
      "loss": 4.9249,
      "mean_token_accuracy": 0.20036870688199998,
      "num_tokens": 153805329.0,
      "step": 70200
    },
    {
      "entropy": 5.741651344299316,
      "epoch": 6.967546645494243,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001463335386532597,
      "loss": 4.9565,
      "mean_token_accuracy": 0.20348783433437348,
      "num_tokens": 153815462.0,
      "step": 70205
    },
    {
      "entropy": 5.7245392322540285,
      "epoch": 6.968042874156412,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014630447375352102,
      "loss": 4.9964,
      "mean_token_accuracy": 0.19999326318502425,
      "num_tokens": 153826608.0,
      "step": 70210
    },
    {
      "entropy": 5.696314764022827,
      "epoch": 6.968539102818578,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014627541204515398,
      "loss": 4.8688,
      "mean_token_accuracy": 0.20163353830575942,
      "num_tokens": 153838112.0,
      "step": 70215
    },
    {
      "entropy": 5.776820039749145,
      "epoch": 6.969035331480747,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001462463535288794,
      "loss": 5.0291,
      "mean_token_accuracy": 0.19687303751707078,
      "num_tokens": 153850116.0,
      "step": 70220
    },
    {
      "entropy": 5.703971815109253,
      "epoch": 6.969531560142914,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00014621729820541785,
      "loss": 4.9311,
      "mean_token_accuracy": 0.1979302480816841,
      "num_tokens": 153860278.0,
      "step": 70225
    },
    {
      "entropy": 5.68034610748291,
      "epoch": 6.970027788805082,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014618824607548971,
      "loss": 4.909,
      "mean_token_accuracy": 0.20177801549434662,
      "num_tokens": 153871681.0,
      "step": 70230
    },
    {
      "entropy": 5.6312202453613285,
      "epoch": 6.970524017467249,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014615919713981557,
      "loss": 4.8339,
      "mean_token_accuracy": 0.20689585506916047,
      "num_tokens": 153883505.0,
      "step": 70235
    },
    {
      "entropy": 5.660091257095337,
      "epoch": 6.971020246129417,
      "grad_norm": 1.515625,
      "learning_rate": 0.00014613015139911572,
      "loss": 4.8679,
      "mean_token_accuracy": 0.20723221004009246,
      "num_tokens": 153894799.0,
      "step": 70240
    },
    {
      "entropy": 5.643140697479248,
      "epoch": 6.971516474791584,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014610110885411042,
      "loss": 4.8944,
      "mean_token_accuracy": 0.19577522277832032,
      "num_tokens": 153905961.0,
      "step": 70245
    },
    {
      "entropy": 5.70575532913208,
      "epoch": 6.972012703453752,
      "grad_norm": 1.5625,
      "learning_rate": 0.00014607206950551982,
      "loss": 4.9199,
      "mean_token_accuracy": 0.19909946173429488,
      "num_tokens": 153916901.0,
      "step": 70250
    },
    {
      "entropy": 5.680959463119507,
      "epoch": 6.972508932115919,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001460430333540641,
      "loss": 4.9226,
      "mean_token_accuracy": 0.20170364230871202,
      "num_tokens": 153926650.0,
      "step": 70255
    },
    {
      "entropy": 5.756147480010986,
      "epoch": 6.9730051607780865,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014601400040046325,
      "loss": 4.9091,
      "mean_token_accuracy": 0.20009371340274812,
      "num_tokens": 153937836.0,
      "step": 70260
    },
    {
      "entropy": 5.674379158020019,
      "epoch": 6.973501389440254,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001459849706454373,
      "loss": 4.8852,
      "mean_token_accuracy": 0.20433256030082703,
      "num_tokens": 153948361.0,
      "step": 70265
    },
    {
      "entropy": 5.609183359146118,
      "epoch": 6.9739976181024215,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001459559440897061,
      "loss": 4.8389,
      "mean_token_accuracy": 0.2023483231663704,
      "num_tokens": 153960034.0,
      "step": 70270
    },
    {
      "entropy": 5.628472566604614,
      "epoch": 6.974493846764589,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001459269207339894,
      "loss": 4.871,
      "mean_token_accuracy": 0.2009868323802948,
      "num_tokens": 153972149.0,
      "step": 70275
    },
    {
      "entropy": 5.654015445709229,
      "epoch": 6.974990075426756,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014589790057900688,
      "loss": 4.8532,
      "mean_token_accuracy": 0.21085764169692994,
      "num_tokens": 153984104.0,
      "step": 70280
    },
    {
      "entropy": 5.693754434585571,
      "epoch": 6.975486304088924,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001458688836254783,
      "loss": 4.9549,
      "mean_token_accuracy": 0.20162637531757355,
      "num_tokens": 153995725.0,
      "step": 70285
    },
    {
      "entropy": 5.682163143157959,
      "epoch": 6.975982532751091,
      "grad_norm": 1.375,
      "learning_rate": 0.00014583986987412307,
      "loss": 4.9008,
      "mean_token_accuracy": 0.20161983370780945,
      "num_tokens": 154008109.0,
      "step": 70290
    },
    {
      "entropy": 5.685020065307617,
      "epoch": 6.976478761413259,
      "grad_norm": 1.203125,
      "learning_rate": 0.00014581085932566086,
      "loss": 4.9341,
      "mean_token_accuracy": 0.1907266318798065,
      "num_tokens": 154020005.0,
      "step": 70295
    },
    {
      "entropy": 5.691484880447388,
      "epoch": 6.976974990075426,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001457818519808109,
      "loss": 4.8782,
      "mean_token_accuracy": 0.2038258746266365,
      "num_tokens": 154031323.0,
      "step": 70300
    },
    {
      "entropy": 5.723736810684204,
      "epoch": 6.977471218737595,
      "grad_norm": 1.515625,
      "learning_rate": 0.00014575284784029263,
      "loss": 4.9597,
      "mean_token_accuracy": 0.19951755553483963,
      "num_tokens": 154041936.0,
      "step": 70305
    },
    {
      "entropy": 5.654895353317261,
      "epoch": 6.977967447399762,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014572384690482521,
      "loss": 4.8421,
      "mean_token_accuracy": 0.20679908543825148,
      "num_tokens": 154052092.0,
      "step": 70310
    },
    {
      "entropy": 5.6420714378356935,
      "epoch": 6.97846367606193,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014569484917512773,
      "loss": 4.8478,
      "mean_token_accuracy": 0.204092575609684,
      "num_tokens": 154061714.0,
      "step": 70315
    },
    {
      "entropy": 5.700956726074219,
      "epoch": 6.978959904724097,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014566585465191937,
      "loss": 4.8514,
      "mean_token_accuracy": 0.20339517593383788,
      "num_tokens": 154072264.0,
      "step": 70320
    },
    {
      "entropy": 5.6302406787872314,
      "epoch": 6.979456133386265,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014563686333591918,
      "loss": 4.8538,
      "mean_token_accuracy": 0.20151280015707015,
      "num_tokens": 154083102.0,
      "step": 70325
    },
    {
      "entropy": 5.648037052154541,
      "epoch": 6.979952362048432,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014560787522784608,
      "loss": 4.8582,
      "mean_token_accuracy": 0.197967004776001,
      "num_tokens": 154092561.0,
      "step": 70330
    },
    {
      "entropy": 5.627038431167603,
      "epoch": 6.9804485907105995,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001455788903284188,
      "loss": 4.8711,
      "mean_token_accuracy": 0.20358560234308243,
      "num_tokens": 154103441.0,
      "step": 70335
    },
    {
      "entropy": 5.691965770721436,
      "epoch": 6.980944819372767,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001455499086383562,
      "loss": 4.9197,
      "mean_token_accuracy": 0.20216268301010132,
      "num_tokens": 154114797.0,
      "step": 70340
    },
    {
      "entropy": 5.690547752380371,
      "epoch": 6.9814410480349345,
      "grad_norm": 1.25,
      "learning_rate": 0.00014552093015837687,
      "loss": 4.9336,
      "mean_token_accuracy": 0.19328244775533676,
      "num_tokens": 154125828.0,
      "step": 70345
    },
    {
      "entropy": 5.704432249069214,
      "epoch": 6.981937276697102,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014549195488919943,
      "loss": 4.9329,
      "mean_token_accuracy": 0.20356693714857102,
      "num_tokens": 154135482.0,
      "step": 70350
    },
    {
      "entropy": 5.62139663696289,
      "epoch": 6.982433505359269,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014546298283154253,
      "loss": 4.8057,
      "mean_token_accuracy": 0.21023996621370317,
      "num_tokens": 154146861.0,
      "step": 70355
    },
    {
      "entropy": 5.688622570037841,
      "epoch": 6.982929734021437,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014543401398612452,
      "loss": 4.8856,
      "mean_token_accuracy": 0.19874528050422668,
      "num_tokens": 154156878.0,
      "step": 70360
    },
    {
      "entropy": 5.693955516815185,
      "epoch": 6.983425962683604,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001454050483536638,
      "loss": 4.9394,
      "mean_token_accuracy": 0.19430848509073256,
      "num_tokens": 154168569.0,
      "step": 70365
    },
    {
      "entropy": 5.6807756423950195,
      "epoch": 6.983922191345772,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001453760859348786,
      "loss": 4.9042,
      "mean_token_accuracy": 0.201180337369442,
      "num_tokens": 154179717.0,
      "step": 70370
    },
    {
      "entropy": 5.711913776397705,
      "epoch": 6.984418420007939,
      "grad_norm": 1.375,
      "learning_rate": 0.00014534712673048712,
      "loss": 4.9529,
      "mean_token_accuracy": 0.20277268439531326,
      "num_tokens": 154191443.0,
      "step": 70375
    },
    {
      "entropy": 5.6679182052612305,
      "epoch": 6.984914648670107,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014531817074120757,
      "loss": 4.891,
      "mean_token_accuracy": 0.20075754970312118,
      "num_tokens": 154203695.0,
      "step": 70380
    },
    {
      "entropy": 5.670199775695801,
      "epoch": 6.985410877332274,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014528921796775785,
      "loss": 4.8588,
      "mean_token_accuracy": 0.20839632749557496,
      "num_tokens": 154214600.0,
      "step": 70385
    },
    {
      "entropy": 5.631595802307129,
      "epoch": 6.985907105994443,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001452602684108561,
      "loss": 4.8383,
      "mean_token_accuracy": 0.20924540907144545,
      "num_tokens": 154224821.0,
      "step": 70390
    },
    {
      "entropy": 5.656076431274414,
      "epoch": 6.98640333465661,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014523132207122015,
      "loss": 4.8744,
      "mean_token_accuracy": 0.2057695657014847,
      "num_tokens": 154236000.0,
      "step": 70395
    },
    {
      "entropy": 5.733483457565308,
      "epoch": 6.986899563318778,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014520237894956778,
      "loss": 4.8943,
      "mean_token_accuracy": 0.20311688333749772,
      "num_tokens": 154245563.0,
      "step": 70400
    },
    {
      "entropy": 5.679761552810669,
      "epoch": 6.987395791980945,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014517343904661673,
      "loss": 4.9,
      "mean_token_accuracy": 0.20033523589372634,
      "num_tokens": 154256871.0,
      "step": 70405
    },
    {
      "entropy": 5.6601114749908445,
      "epoch": 6.9878920206431125,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014514450236308455,
      "loss": 4.9121,
      "mean_token_accuracy": 0.20045748054981233,
      "num_tokens": 154268075.0,
      "step": 70410
    },
    {
      "entropy": 5.627213764190674,
      "epoch": 6.98838824930528,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014511556889968888,
      "loss": 4.8482,
      "mean_token_accuracy": 0.2059463918209076,
      "num_tokens": 154280621.0,
      "step": 70415
    },
    {
      "entropy": 5.67961277961731,
      "epoch": 6.9888844779674475,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014508663865714732,
      "loss": 4.8904,
      "mean_token_accuracy": 0.19844342023134232,
      "num_tokens": 154290212.0,
      "step": 70420
    },
    {
      "entropy": 5.717697477340698,
      "epoch": 6.989380706629615,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001450577116361772,
      "loss": 4.924,
      "mean_token_accuracy": 0.20396513789892196,
      "num_tokens": 154299821.0,
      "step": 70425
    },
    {
      "entropy": 5.596029663085938,
      "epoch": 6.9898769352917824,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00014502878783749582,
      "loss": 4.8084,
      "mean_token_accuracy": 0.21040953546762467,
      "num_tokens": 154312686.0,
      "step": 70430
    },
    {
      "entropy": 5.6174053192138675,
      "epoch": 6.99037316395395,
      "grad_norm": 1.5625,
      "learning_rate": 0.00014499986726182037,
      "loss": 4.7785,
      "mean_token_accuracy": 0.20984114408493043,
      "num_tokens": 154322850.0,
      "step": 70435
    },
    {
      "entropy": 5.696104335784912,
      "epoch": 6.990869392616117,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014497094990986815,
      "loss": 4.9796,
      "mean_token_accuracy": 0.1936451956629753,
      "num_tokens": 154333227.0,
      "step": 70440
    },
    {
      "entropy": 5.69151086807251,
      "epoch": 6.991365621278285,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001449420357823561,
      "loss": 4.9253,
      "mean_token_accuracy": 0.2054791197180748,
      "num_tokens": 154345290.0,
      "step": 70445
    },
    {
      "entropy": 5.715622043609619,
      "epoch": 6.991861849940452,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001449131248800014,
      "loss": 4.9592,
      "mean_token_accuracy": 0.20696377903223037,
      "num_tokens": 154355888.0,
      "step": 70450
    },
    {
      "entropy": 5.71726770401001,
      "epoch": 6.99235807860262,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014488421720352092,
      "loss": 4.931,
      "mean_token_accuracy": 0.20276072174310683,
      "num_tokens": 154367250.0,
      "step": 70455
    },
    {
      "entropy": 5.71823205947876,
      "epoch": 6.992854307264787,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014485531275363145,
      "loss": 4.8876,
      "mean_token_accuracy": 0.19311251640319824,
      "num_tokens": 154377677.0,
      "step": 70460
    },
    {
      "entropy": 5.726369380950928,
      "epoch": 6.993350535926955,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001448264115310497,
      "loss": 5.0,
      "mean_token_accuracy": 0.19617578387260437,
      "num_tokens": 154389342.0,
      "step": 70465
    },
    {
      "entropy": 5.703672742843628,
      "epoch": 6.993846764589122,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014479751353649252,
      "loss": 4.9386,
      "mean_token_accuracy": 0.19852273315191268,
      "num_tokens": 154400698.0,
      "step": 70470
    },
    {
      "entropy": 5.714862060546875,
      "epoch": 6.994342993251291,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014476861877067644,
      "loss": 4.9351,
      "mean_token_accuracy": 0.20124861896038054,
      "num_tokens": 154411862.0,
      "step": 70475
    },
    {
      "entropy": 5.676921939849853,
      "epoch": 6.994839221913458,
      "grad_norm": 1.515625,
      "learning_rate": 0.00014473972723431794,
      "loss": 4.8801,
      "mean_token_accuracy": 0.19948394298553468,
      "num_tokens": 154422376.0,
      "step": 70480
    },
    {
      "entropy": 5.6820684432983395,
      "epoch": 6.995335450575626,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014471083892813357,
      "loss": 4.8659,
      "mean_token_accuracy": 0.20779213905334473,
      "num_tokens": 154433678.0,
      "step": 70485
    },
    {
      "entropy": 5.757376337051392,
      "epoch": 6.995831679237793,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014468195385283967,
      "loss": 4.9924,
      "mean_token_accuracy": 0.19788924753665924,
      "num_tokens": 154443451.0,
      "step": 70490
    },
    {
      "entropy": 5.630037403106689,
      "epoch": 6.9963279078999605,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001446530720091524,
      "loss": 4.8221,
      "mean_token_accuracy": 0.20611171275377274,
      "num_tokens": 154453458.0,
      "step": 70495
    },
    {
      "entropy": 5.671072769165039,
      "epoch": 6.996824136562128,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001446241933977882,
      "loss": 4.9284,
      "mean_token_accuracy": 0.18887531757354736,
      "num_tokens": 154464865.0,
      "step": 70500
    },
    {
      "entropy": 5.716458177566528,
      "epoch": 6.9973203652242955,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014459531801946303,
      "loss": 4.9229,
      "mean_token_accuracy": 0.19782277345657348,
      "num_tokens": 154474772.0,
      "step": 70505
    },
    {
      "entropy": 5.623661661148072,
      "epoch": 6.997816593886463,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014456644587489286,
      "loss": 4.8377,
      "mean_token_accuracy": 0.2099367558956146,
      "num_tokens": 154485381.0,
      "step": 70510
    },
    {
      "entropy": 5.712485885620117,
      "epoch": 6.99831282254863,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014453757696479392,
      "loss": 4.9846,
      "mean_token_accuracy": 0.19722803235054015,
      "num_tokens": 154496600.0,
      "step": 70515
    },
    {
      "entropy": 5.720140838623047,
      "epoch": 6.998809051210798,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014450871128988193,
      "loss": 4.9114,
      "mean_token_accuracy": 0.19971826374530793,
      "num_tokens": 154506441.0,
      "step": 70520
    },
    {
      "entropy": 5.620255088806152,
      "epoch": 6.999305279872965,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014447984885087263,
      "loss": 4.8748,
      "mean_token_accuracy": 0.1970979392528534,
      "num_tokens": 154516785.0,
      "step": 70525
    },
    {
      "entropy": 5.663336324691772,
      "epoch": 6.999801508535133,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001444509896484819,
      "loss": 4.8966,
      "mean_token_accuracy": 0.20686324536800385,
      "num_tokens": 154527214.0,
      "step": 70530
    },
    {
      "entropy": 5.695186805725098,
      "epoch": 7.0002977371973,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014442213368342534,
      "loss": 4.9423,
      "mean_token_accuracy": 0.19731860905885695,
      "num_tokens": 154537819.0,
      "step": 70535
    },
    {
      "entropy": 5.692766857147217,
      "epoch": 7.000793965859468,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014439328095641852,
      "loss": 4.8748,
      "mean_token_accuracy": 0.20848155170679092,
      "num_tokens": 154547241.0,
      "step": 70540
    },
    {
      "entropy": 5.714704513549805,
      "epoch": 7.001290194521635,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014436443146817677,
      "loss": 4.8929,
      "mean_token_accuracy": 0.20028859972953797,
      "num_tokens": 154557928.0,
      "step": 70545
    },
    {
      "entropy": 5.6529582977294925,
      "epoch": 7.001786423183803,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001443355852194157,
      "loss": 4.8021,
      "mean_token_accuracy": 0.21122955828905104,
      "num_tokens": 154568486.0,
      "step": 70550
    },
    {
      "entropy": 5.655238246917724,
      "epoch": 7.00228265184597,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001443067422108505,
      "loss": 4.8532,
      "mean_token_accuracy": 0.2067473202943802,
      "num_tokens": 154578925.0,
      "step": 70555
    },
    {
      "entropy": 5.60662202835083,
      "epoch": 7.002778880508139,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014427790244319656,
      "loss": 4.8169,
      "mean_token_accuracy": 0.20763374716043473,
      "num_tokens": 154589216.0,
      "step": 70560
    },
    {
      "entropy": 5.669947814941406,
      "epoch": 7.003275109170306,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014424906591716896,
      "loss": 4.9096,
      "mean_token_accuracy": 0.2001289516687393,
      "num_tokens": 154601973.0,
      "step": 70565
    },
    {
      "entropy": 5.704417514801025,
      "epoch": 7.0037713378324735,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014422023263348276,
      "loss": 4.9195,
      "mean_token_accuracy": 0.20691292434930803,
      "num_tokens": 154612723.0,
      "step": 70570
    },
    {
      "entropy": 5.741671228408814,
      "epoch": 7.004267566494641,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001441914025928529,
      "loss": 4.9812,
      "mean_token_accuracy": 0.19260264933109283,
      "num_tokens": 154623631.0,
      "step": 70575
    },
    {
      "entropy": 5.653747129440307,
      "epoch": 7.0047637951568085,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014416257579599452,
      "loss": 4.7862,
      "mean_token_accuracy": 0.21279465705156325,
      "num_tokens": 154634607.0,
      "step": 70580
    },
    {
      "entropy": 5.704664993286133,
      "epoch": 7.005260023818976,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001441337522436222,
      "loss": 4.9329,
      "mean_token_accuracy": 0.20457939058542252,
      "num_tokens": 154645739.0,
      "step": 70585
    },
    {
      "entropy": 5.671697807312012,
      "epoch": 7.005756252481143,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001441049319364509,
      "loss": 4.8532,
      "mean_token_accuracy": 0.20386714190244676,
      "num_tokens": 154656072.0,
      "step": 70590
    },
    {
      "entropy": 5.734907627105713,
      "epoch": 7.006252481143311,
      "grad_norm": 1.625,
      "learning_rate": 0.00014407611487519527,
      "loss": 4.9675,
      "mean_token_accuracy": 0.19566509425640105,
      "num_tokens": 154667482.0,
      "step": 70595
    },
    {
      "entropy": 5.715943002700806,
      "epoch": 7.006748709805478,
      "grad_norm": 1.375,
      "learning_rate": 0.0001440473010605699,
      "loss": 4.9283,
      "mean_token_accuracy": 0.19713115841150283,
      "num_tokens": 154679044.0,
      "step": 70600
    },
    {
      "entropy": 5.711881256103515,
      "epoch": 7.007244938467646,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014401849049328915,
      "loss": 4.9242,
      "mean_token_accuracy": 0.19506783336400985,
      "num_tokens": 154689929.0,
      "step": 70605
    },
    {
      "entropy": 5.728475427627563,
      "epoch": 7.007741167129813,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00014398968317406774,
      "loss": 4.8595,
      "mean_token_accuracy": 0.19436549246311188,
      "num_tokens": 154700072.0,
      "step": 70610
    },
    {
      "entropy": 5.713496494293213,
      "epoch": 7.008237395791981,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014396087910361976,
      "loss": 4.9116,
      "mean_token_accuracy": 0.20215059518814088,
      "num_tokens": 154711172.0,
      "step": 70615
    },
    {
      "entropy": 5.776865100860595,
      "epoch": 7.008733624454148,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001439320782826597,
      "loss": 5.0024,
      "mean_token_accuracy": 0.1930355966091156,
      "num_tokens": 154722564.0,
      "step": 70620
    },
    {
      "entropy": 5.724308109283447,
      "epoch": 7.009229853116316,
      "grad_norm": 1.59375,
      "learning_rate": 0.00014390328071190166,
      "loss": 4.918,
      "mean_token_accuracy": 0.2031901314854622,
      "num_tokens": 154732362.0,
      "step": 70625
    },
    {
      "entropy": 5.654094791412353,
      "epoch": 7.009726081778483,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001438744863920598,
      "loss": 4.7539,
      "mean_token_accuracy": 0.21908473819494248,
      "num_tokens": 154745345.0,
      "step": 70630
    },
    {
      "entropy": 5.7312009811401365,
      "epoch": 7.010222310440651,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014384569532384812,
      "loss": 4.8894,
      "mean_token_accuracy": 0.20421424210071565,
      "num_tokens": 154755753.0,
      "step": 70635
    },
    {
      "entropy": 5.684484672546387,
      "epoch": 7.010718539102818,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001438169075079805,
      "loss": 4.9335,
      "mean_token_accuracy": 0.20265607982873918,
      "num_tokens": 154766961.0,
      "step": 70640
    },
    {
      "entropy": 5.711219072341919,
      "epoch": 7.0112147677649865,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014378812294517086,
      "loss": 4.9482,
      "mean_token_accuracy": 0.19911681711673737,
      "num_tokens": 154778502.0,
      "step": 70645
    },
    {
      "entropy": 5.714207458496094,
      "epoch": 7.011710996427154,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014375934163613312,
      "loss": 4.9711,
      "mean_token_accuracy": 0.1922750622034073,
      "num_tokens": 154789293.0,
      "step": 70650
    },
    {
      "entropy": 5.705957412719727,
      "epoch": 7.0122072250893215,
      "grad_norm": 1.421875,
      "learning_rate": 0.000143730563581581,
      "loss": 4.8422,
      "mean_token_accuracy": 0.20081901848316192,
      "num_tokens": 154799592.0,
      "step": 70655
    },
    {
      "entropy": 5.72309193611145,
      "epoch": 7.012703453751489,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014370178878222794,
      "loss": 4.8987,
      "mean_token_accuracy": 0.20398021787405013,
      "num_tokens": 154809311.0,
      "step": 70660
    },
    {
      "entropy": 5.641017246246338,
      "epoch": 7.013199682413656,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014367301723878763,
      "loss": 4.8,
      "mean_token_accuracy": 0.21048857271671295,
      "num_tokens": 154819739.0,
      "step": 70665
    },
    {
      "entropy": 5.697056770324707,
      "epoch": 7.013695911075824,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001436442489519734,
      "loss": 4.8708,
      "mean_token_accuracy": 0.209174482524395,
      "num_tokens": 154830985.0,
      "step": 70670
    },
    {
      "entropy": 5.644635725021362,
      "epoch": 7.014192139737991,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00014361548392249882,
      "loss": 4.8836,
      "mean_token_accuracy": 0.1975934773683548,
      "num_tokens": 154841877.0,
      "step": 70675
    },
    {
      "entropy": 5.691274356842041,
      "epoch": 7.014688368400159,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00014358672215107718,
      "loss": 4.8812,
      "mean_token_accuracy": 0.20399700701236725,
      "num_tokens": 154852259.0,
      "step": 70680
    },
    {
      "entropy": 5.686310482025147,
      "epoch": 7.015184597062326,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014355796363842166,
      "loss": 4.9496,
      "mean_token_accuracy": 0.19635054767131804,
      "num_tokens": 154862525.0,
      "step": 70685
    },
    {
      "entropy": 5.717731857299805,
      "epoch": 7.015680825724494,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014352920838524538,
      "loss": 4.9011,
      "mean_token_accuracy": 0.20028088837862015,
      "num_tokens": 154873123.0,
      "step": 70690
    },
    {
      "entropy": 5.684028482437133,
      "epoch": 7.016177054386661,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014350045639226149,
      "loss": 4.8831,
      "mean_token_accuracy": 0.2044389009475708,
      "num_tokens": 154883759.0,
      "step": 70695
    },
    {
      "entropy": 5.648504543304443,
      "epoch": 7.016673283048829,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001434717076601828,
      "loss": 4.8397,
      "mean_token_accuracy": 0.20495900362730027,
      "num_tokens": 154893908.0,
      "step": 70700
    },
    {
      "entropy": 5.690135431289673,
      "epoch": 7.017169511710996,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001434429621897224,
      "loss": 4.8825,
      "mean_token_accuracy": 0.1998109385371208,
      "num_tokens": 154905052.0,
      "step": 70705
    },
    {
      "entropy": 5.6652039051055905,
      "epoch": 7.017665740373164,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014341421998159305,
      "loss": 4.8407,
      "mean_token_accuracy": 0.21053078174591064,
      "num_tokens": 154915495.0,
      "step": 70710
    },
    {
      "entropy": 5.659006977081299,
      "epoch": 7.018161969035331,
      "grad_norm": 1.375,
      "learning_rate": 0.00014338548103650753,
      "loss": 4.8562,
      "mean_token_accuracy": 0.2028696522116661,
      "num_tokens": 154928425.0,
      "step": 70715
    },
    {
      "entropy": 5.719814443588257,
      "epoch": 7.018658197697499,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001433567453551785,
      "loss": 4.9014,
      "mean_token_accuracy": 0.2084091067314148,
      "num_tokens": 154938094.0,
      "step": 70720
    },
    {
      "entropy": 5.676918601989746,
      "epoch": 7.019154426359666,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001433280129383185,
      "loss": 4.8889,
      "mean_token_accuracy": 0.20360398590564727,
      "num_tokens": 154949771.0,
      "step": 70725
    },
    {
      "entropy": 5.589357185363769,
      "epoch": 7.0196506550218345,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014329928378663996,
      "loss": 4.7826,
      "mean_token_accuracy": 0.22082605063915253,
      "num_tokens": 154961655.0,
      "step": 70730
    },
    {
      "entropy": 5.711060523986816,
      "epoch": 7.020146883684002,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014327055790085553,
      "loss": 4.9495,
      "mean_token_accuracy": 0.2001089483499527,
      "num_tokens": 154971914.0,
      "step": 70735
    },
    {
      "entropy": 5.676310348510742,
      "epoch": 7.0206431123461694,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014324183528167727,
      "loss": 4.8506,
      "mean_token_accuracy": 0.19946676790714263,
      "num_tokens": 154981884.0,
      "step": 70740
    },
    {
      "entropy": 5.639052009582519,
      "epoch": 7.021139341008337,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001432131159298177,
      "loss": 4.8692,
      "mean_token_accuracy": 0.20764423161745071,
      "num_tokens": 154993789.0,
      "step": 70745
    },
    {
      "entropy": 5.6548130989074705,
      "epoch": 7.021635569670504,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001431843998459889,
      "loss": 4.8429,
      "mean_token_accuracy": 0.2050834134221077,
      "num_tokens": 155004352.0,
      "step": 70750
    },
    {
      "entropy": 5.66045389175415,
      "epoch": 7.022131798332672,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001431556870309029,
      "loss": 4.8129,
      "mean_token_accuracy": 0.2056406944990158,
      "num_tokens": 155014576.0,
      "step": 70755
    },
    {
      "entropy": 5.753037405014038,
      "epoch": 7.022628026994839,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001431269774852717,
      "loss": 5.0029,
      "mean_token_accuracy": 0.19656678438186645,
      "num_tokens": 155025254.0,
      "step": 70760
    },
    {
      "entropy": 5.768856525421143,
      "epoch": 7.023124255657007,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014309827120980736,
      "loss": 5.0022,
      "mean_token_accuracy": 0.18991497159004211,
      "num_tokens": 155036304.0,
      "step": 70765
    },
    {
      "entropy": 5.679980278015137,
      "epoch": 7.023620484319174,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014306956820522168,
      "loss": 4.82,
      "mean_token_accuracy": 0.20628341436386108,
      "num_tokens": 155046360.0,
      "step": 70770
    },
    {
      "entropy": 5.656533718109131,
      "epoch": 7.024116712981342,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014304086847222636,
      "loss": 4.8654,
      "mean_token_accuracy": 0.2028188794851303,
      "num_tokens": 155057363.0,
      "step": 70775
    },
    {
      "entropy": 5.667479991912842,
      "epoch": 7.024612941643509,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014301217201153322,
      "loss": 4.8717,
      "mean_token_accuracy": 0.19970032423734665,
      "num_tokens": 155067902.0,
      "step": 70780
    },
    {
      "entropy": 5.697595453262329,
      "epoch": 7.025109170305677,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001429834788238538,
      "loss": 4.9205,
      "mean_token_accuracy": 0.20037121325731277,
      "num_tokens": 155079306.0,
      "step": 70785
    },
    {
      "entropy": 5.730707597732544,
      "epoch": 7.025605398967844,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014295478890989955,
      "loss": 4.9522,
      "mean_token_accuracy": 0.19074240624904631,
      "num_tokens": 155090203.0,
      "step": 70790
    },
    {
      "entropy": 5.637938165664673,
      "epoch": 7.026101627630012,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001429261022703821,
      "loss": 4.8315,
      "mean_token_accuracy": 0.2053218424320221,
      "num_tokens": 155099901.0,
      "step": 70795
    },
    {
      "entropy": 5.739898014068603,
      "epoch": 7.026597856292179,
      "grad_norm": 1.625,
      "learning_rate": 0.00014289741890601273,
      "loss": 4.9528,
      "mean_token_accuracy": 0.19996396750211715,
      "num_tokens": 155110431.0,
      "step": 70800
    },
    {
      "entropy": 5.627730035781861,
      "epoch": 7.027094084954347,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001428687388175026,
      "loss": 4.8471,
      "mean_token_accuracy": 0.20621610730886458,
      "num_tokens": 155120443.0,
      "step": 70805
    },
    {
      "entropy": 5.7569780349731445,
      "epoch": 7.027590313616514,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014284006200556315,
      "loss": 5.0505,
      "mean_token_accuracy": 0.1937011331319809,
      "num_tokens": 155130492.0,
      "step": 70810
    },
    {
      "entropy": 5.640080261230469,
      "epoch": 7.0280865422786825,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014281138847090536,
      "loss": 4.8764,
      "mean_token_accuracy": 0.21189380139112474,
      "num_tokens": 155142042.0,
      "step": 70815
    },
    {
      "entropy": 5.6538715839385985,
      "epoch": 7.02858277094085,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001427827182142402,
      "loss": 4.8082,
      "mean_token_accuracy": 0.2174750715494156,
      "num_tokens": 155152711.0,
      "step": 70820
    },
    {
      "entropy": 5.664878797531128,
      "epoch": 7.029078999603017,
      "grad_norm": 1.1875,
      "learning_rate": 0.00014275405123627882,
      "loss": 4.8599,
      "mean_token_accuracy": 0.20307172536849977,
      "num_tokens": 155163790.0,
      "step": 70825
    },
    {
      "entropy": 5.676343107223511,
      "epoch": 7.029575228265185,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014272538753773203,
      "loss": 4.8769,
      "mean_token_accuracy": 0.19638133943080902,
      "num_tokens": 155174757.0,
      "step": 70830
    },
    {
      "entropy": 5.652895927429199,
      "epoch": 7.030071456927352,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014269672711931058,
      "loss": 4.8808,
      "mean_token_accuracy": 0.207329560816288,
      "num_tokens": 155184940.0,
      "step": 70835
    },
    {
      "entropy": 5.720434856414795,
      "epoch": 7.03056768558952,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014266806998172515,
      "loss": 4.917,
      "mean_token_accuracy": 0.20528952032327652,
      "num_tokens": 155195473.0,
      "step": 70840
    },
    {
      "entropy": 5.67414345741272,
      "epoch": 7.031063914251687,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001426394161256865,
      "loss": 4.8948,
      "mean_token_accuracy": 0.20408122092485428,
      "num_tokens": 155206601.0,
      "step": 70845
    },
    {
      "entropy": 5.727581882476807,
      "epoch": 7.031560142913855,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014261076555190504,
      "loss": 4.9522,
      "mean_token_accuracy": 0.20345655232667922,
      "num_tokens": 155217232.0,
      "step": 70850
    },
    {
      "entropy": 5.662339115142823,
      "epoch": 7.032056371576022,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00014258211826109136,
      "loss": 4.8271,
      "mean_token_accuracy": 0.2044616922736168,
      "num_tokens": 155226462.0,
      "step": 70855
    },
    {
      "entropy": 5.695425796508789,
      "epoch": 7.03255260023819,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014255347425395587,
      "loss": 4.8399,
      "mean_token_accuracy": 0.2066795662045479,
      "num_tokens": 155237465.0,
      "step": 70860
    },
    {
      "entropy": 5.652094602584839,
      "epoch": 7.033048828900357,
      "grad_norm": 1.375,
      "learning_rate": 0.00014252483353120876,
      "loss": 4.8127,
      "mean_token_accuracy": 0.21394740641117097,
      "num_tokens": 155249259.0,
      "step": 70865
    },
    {
      "entropy": 5.637444543838501,
      "epoch": 7.033545057562525,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001424961960935603,
      "loss": 4.8085,
      "mean_token_accuracy": 0.20752732455730438,
      "num_tokens": 155259753.0,
      "step": 70870
    },
    {
      "entropy": 5.659615993499756,
      "epoch": 7.034041286224692,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001424675619417207,
      "loss": 4.899,
      "mean_token_accuracy": 0.2094385251402855,
      "num_tokens": 155270022.0,
      "step": 70875
    },
    {
      "entropy": 5.6738848209381105,
      "epoch": 7.03453751488686,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014243893107639995,
      "loss": 4.8641,
      "mean_token_accuracy": 0.20795340985059738,
      "num_tokens": 155280007.0,
      "step": 70880
    },
    {
      "entropy": 5.651033353805542,
      "epoch": 7.035033743549027,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014241030349830807,
      "loss": 4.8558,
      "mean_token_accuracy": 0.20688396096229553,
      "num_tokens": 155290723.0,
      "step": 70885
    },
    {
      "entropy": 5.641249561309815,
      "epoch": 7.035529972211195,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014238167920815504,
      "loss": 4.8626,
      "mean_token_accuracy": 0.2076795905828476,
      "num_tokens": 155302016.0,
      "step": 70890
    },
    {
      "entropy": 5.6777149677276615,
      "epoch": 7.036026200873362,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014235305820665053,
      "loss": 4.8837,
      "mean_token_accuracy": 0.2015014499425888,
      "num_tokens": 155312842.0,
      "step": 70895
    },
    {
      "entropy": 5.706574964523315,
      "epoch": 7.0365224295355295,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001423244404945043,
      "loss": 4.9257,
      "mean_token_accuracy": 0.19698327630758286,
      "num_tokens": 155323467.0,
      "step": 70900
    },
    {
      "entropy": 5.738660001754761,
      "epoch": 7.037018658197698,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014229582607242614,
      "loss": 4.8815,
      "mean_token_accuracy": 0.21014891862869262,
      "num_tokens": 155334438.0,
      "step": 70905
    },
    {
      "entropy": 5.704851961135864,
      "epoch": 7.037514886859865,
      "grad_norm": 1.5,
      "learning_rate": 0.00014226721494112543,
      "loss": 4.8883,
      "mean_token_accuracy": 0.20731329321861267,
      "num_tokens": 155346173.0,
      "step": 70910
    },
    {
      "entropy": 5.716948413848877,
      "epoch": 7.038011115522033,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014223860710131187,
      "loss": 4.9373,
      "mean_token_accuracy": 0.19528982639312745,
      "num_tokens": 155357469.0,
      "step": 70915
    },
    {
      "entropy": 5.740833330154419,
      "epoch": 7.0385073441842,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001422100025536948,
      "loss": 4.9043,
      "mean_token_accuracy": 0.2074191927909851,
      "num_tokens": 155369886.0,
      "step": 70920
    },
    {
      "entropy": 5.682689762115478,
      "epoch": 7.039003572846368,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001421814012989835,
      "loss": 4.8531,
      "mean_token_accuracy": 0.2042875334620476,
      "num_tokens": 155380843.0,
      "step": 70925
    },
    {
      "entropy": 5.630160665512085,
      "epoch": 7.039499801508535,
      "grad_norm": 1.5,
      "learning_rate": 0.00014215280333788728,
      "loss": 4.7919,
      "mean_token_accuracy": 0.20691651850938797,
      "num_tokens": 155390161.0,
      "step": 70930
    },
    {
      "entropy": 5.686854887008667,
      "epoch": 7.039996030170703,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001421242086711152,
      "loss": 4.9529,
      "mean_token_accuracy": 0.1999336689710617,
      "num_tokens": 155401083.0,
      "step": 70935
    },
    {
      "entropy": 5.701465749740601,
      "epoch": 7.04049225883287,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0001420956172993764,
      "loss": 4.9122,
      "mean_token_accuracy": 0.20156415551900864,
      "num_tokens": 155411983.0,
      "step": 70940
    },
    {
      "entropy": 5.712779760360718,
      "epoch": 7.040988487495038,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014206702922338,
      "loss": 4.8558,
      "mean_token_accuracy": 0.2041281059384346,
      "num_tokens": 155422110.0,
      "step": 70945
    },
    {
      "entropy": 5.678834724426269,
      "epoch": 7.041484716157205,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001420384444438348,
      "loss": 4.9314,
      "mean_token_accuracy": 0.20061158239841462,
      "num_tokens": 155433560.0,
      "step": 70950
    },
    {
      "entropy": 5.651744699478149,
      "epoch": 7.041980944819373,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014200986296144968,
      "loss": 4.8298,
      "mean_token_accuracy": 0.2039400517940521,
      "num_tokens": 155446198.0,
      "step": 70955
    },
    {
      "entropy": 5.688982629776001,
      "epoch": 7.04247717348154,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001419812847769334,
      "loss": 4.9425,
      "mean_token_accuracy": 0.20020297169685364,
      "num_tokens": 155457358.0,
      "step": 70960
    },
    {
      "entropy": 5.66270751953125,
      "epoch": 7.042973402143708,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014195270989099458,
      "loss": 4.8345,
      "mean_token_accuracy": 0.206868714094162,
      "num_tokens": 155468654.0,
      "step": 70965
    },
    {
      "entropy": 5.6732237339019775,
      "epoch": 7.043469630805875,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001419241383043418,
      "loss": 4.8799,
      "mean_token_accuracy": 0.20636239647865295,
      "num_tokens": 155479504.0,
      "step": 70970
    },
    {
      "entropy": 5.749757719039917,
      "epoch": 7.0439658594680425,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014189557001768378,
      "loss": 4.9397,
      "mean_token_accuracy": 0.19819207191467286,
      "num_tokens": 155491539.0,
      "step": 70975
    },
    {
      "entropy": 5.73676028251648,
      "epoch": 7.04446208813021,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014186700503172878,
      "loss": 4.9176,
      "mean_token_accuracy": 0.19814497232437134,
      "num_tokens": 155503884.0,
      "step": 70980
    },
    {
      "entropy": 5.712994003295899,
      "epoch": 7.0449583167923775,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014183844334718514,
      "loss": 4.933,
      "mean_token_accuracy": 0.1997084751725197,
      "num_tokens": 155515352.0,
      "step": 70985
    },
    {
      "entropy": 5.795684862136841,
      "epoch": 7.045454545454546,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001418098849647612,
      "loss": 5.0119,
      "mean_token_accuracy": 0.19760089069604875,
      "num_tokens": 155526847.0,
      "step": 70990
    },
    {
      "entropy": 5.684950447082519,
      "epoch": 7.045950774116713,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014178132988516506,
      "loss": 4.8714,
      "mean_token_accuracy": 0.202969466149807,
      "num_tokens": 155536659.0,
      "step": 70995
    },
    {
      "entropy": 5.7273579120635985,
      "epoch": 7.046447002778881,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014175277810910493,
      "loss": 4.9212,
      "mean_token_accuracy": 0.20919634401798248,
      "num_tokens": 155547644.0,
      "step": 71000
    },
    {
      "entropy": 5.680442142486572,
      "epoch": 7.046943231441048,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014172422963728873,
      "loss": 4.8514,
      "mean_token_accuracy": 0.20883610248565673,
      "num_tokens": 155558117.0,
      "step": 71005
    },
    {
      "entropy": 5.637313747406006,
      "epoch": 7.047439460103216,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001416956844704245,
      "loss": 4.8312,
      "mean_token_accuracy": 0.20677187889814377,
      "num_tokens": 155569833.0,
      "step": 71010
    },
    {
      "entropy": 5.655229806900024,
      "epoch": 7.047935688765383,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014166714260922003,
      "loss": 4.8403,
      "mean_token_accuracy": 0.20654957592487336,
      "num_tokens": 155580474.0,
      "step": 71015
    },
    {
      "entropy": 5.668213748931885,
      "epoch": 7.048431917427551,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014163860405438317,
      "loss": 4.8607,
      "mean_token_accuracy": 0.20471940487623214,
      "num_tokens": 155591522.0,
      "step": 71020
    },
    {
      "entropy": 5.68826732635498,
      "epoch": 7.048928146089718,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014161006880662147,
      "loss": 4.8879,
      "mean_token_accuracy": 0.20967572182416916,
      "num_tokens": 155601848.0,
      "step": 71025
    },
    {
      "entropy": 5.662663078308105,
      "epoch": 7.049424374751886,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014158153686664272,
      "loss": 4.9499,
      "mean_token_accuracy": 0.19553401321172714,
      "num_tokens": 155613389.0,
      "step": 71030
    },
    {
      "entropy": 5.69979157447815,
      "epoch": 7.049920603414053,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014155300823515422,
      "loss": 4.8871,
      "mean_token_accuracy": 0.19975299388170242,
      "num_tokens": 155623587.0,
      "step": 71035
    },
    {
      "entropy": 5.714013004302979,
      "epoch": 7.050416832076221,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001415244829128637,
      "loss": 4.9742,
      "mean_token_accuracy": 0.20027139633893967,
      "num_tokens": 155634471.0,
      "step": 71040
    },
    {
      "entropy": 5.702734422683716,
      "epoch": 7.050913060738388,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014149596090047838,
      "loss": 4.9351,
      "mean_token_accuracy": 0.1990966409444809,
      "num_tokens": 155646260.0,
      "step": 71045
    },
    {
      "entropy": 5.669657611846924,
      "epoch": 7.051409289400556,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014146744219870554,
      "loss": 4.8043,
      "mean_token_accuracy": 0.2151137337088585,
      "num_tokens": 155656335.0,
      "step": 71050
    },
    {
      "entropy": 5.716024446487427,
      "epoch": 7.051905518062723,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014143892680825232,
      "loss": 4.8704,
      "mean_token_accuracy": 0.20839289128780364,
      "num_tokens": 155667230.0,
      "step": 71055
    },
    {
      "entropy": 5.778722620010376,
      "epoch": 7.0524017467248905,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014141041472982602,
      "loss": 4.9671,
      "mean_token_accuracy": 0.20385369956493377,
      "num_tokens": 155678473.0,
      "step": 71060
    },
    {
      "entropy": 5.698121643066406,
      "epoch": 7.052897975387058,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001413819059641336,
      "loss": 4.9243,
      "mean_token_accuracy": 0.2040050894021988,
      "num_tokens": 155689041.0,
      "step": 71065
    },
    {
      "entropy": 5.716518449783325,
      "epoch": 7.0533942040492255,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014135340051188184,
      "loss": 4.8934,
      "mean_token_accuracy": 0.21073468625545502,
      "num_tokens": 155702591.0,
      "step": 71070
    },
    {
      "entropy": 5.710901212692261,
      "epoch": 7.053890432711394,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014132489837377787,
      "loss": 4.9974,
      "mean_token_accuracy": 0.1945952907204628,
      "num_tokens": 155713244.0,
      "step": 71075
    },
    {
      "entropy": 5.745054721832275,
      "epoch": 7.054386661373561,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014129639955052836,
      "loss": 4.9236,
      "mean_token_accuracy": 0.20299219638109206,
      "num_tokens": 155724782.0,
      "step": 71080
    },
    {
      "entropy": 5.66112265586853,
      "epoch": 7.054882890035729,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014126790404284,
      "loss": 4.8423,
      "mean_token_accuracy": 0.21768808215856553,
      "num_tokens": 155735082.0,
      "step": 71085
    },
    {
      "entropy": 5.701023530960083,
      "epoch": 7.055379118697896,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001412394118514195,
      "loss": 4.8665,
      "mean_token_accuracy": 0.20025812238454818,
      "num_tokens": 155745419.0,
      "step": 71090
    },
    {
      "entropy": 5.708082056045532,
      "epoch": 7.055875347360064,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014121092297697334,
      "loss": 4.9754,
      "mean_token_accuracy": 0.19520726352930068,
      "num_tokens": 155755472.0,
      "step": 71095
    },
    {
      "entropy": 5.667200660705566,
      "epoch": 7.056371576022231,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014118243742020797,
      "loss": 4.8257,
      "mean_token_accuracy": 0.2058282047510147,
      "num_tokens": 155766237.0,
      "step": 71100
    },
    {
      "entropy": 5.633888673782349,
      "epoch": 7.056867804684399,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014115395518182988,
      "loss": 4.7851,
      "mean_token_accuracy": 0.21012769192457198,
      "num_tokens": 155776860.0,
      "step": 71105
    },
    {
      "entropy": 5.643973302841187,
      "epoch": 7.057364033346566,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014112547626254527,
      "loss": 4.8512,
      "mean_token_accuracy": 0.20220689475536346,
      "num_tokens": 155788364.0,
      "step": 71110
    },
    {
      "entropy": 5.7435017108917235,
      "epoch": 7.057860262008734,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014109700066306026,
      "loss": 4.9696,
      "mean_token_accuracy": 0.2044576182961464,
      "num_tokens": 155800420.0,
      "step": 71115
    },
    {
      "entropy": 5.645786190032959,
      "epoch": 7.058356490670901,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014106852838408125,
      "loss": 4.7568,
      "mean_token_accuracy": 0.20848638862371444,
      "num_tokens": 155810804.0,
      "step": 71120
    },
    {
      "entropy": 5.661447286605835,
      "epoch": 7.058852719333069,
      "grad_norm": 1.25,
      "learning_rate": 0.0001410400594263141,
      "loss": 4.8843,
      "mean_token_accuracy": 0.19805361479520797,
      "num_tokens": 155822231.0,
      "step": 71125
    },
    {
      "entropy": 5.709858846664429,
      "epoch": 7.059348947995236,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001410115937904647,
      "loss": 4.9409,
      "mean_token_accuracy": 0.1933709964156151,
      "num_tokens": 155833513.0,
      "step": 71130
    },
    {
      "entropy": 5.676025104522705,
      "epoch": 7.0598451766574035,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014098313147723919,
      "loss": 4.8836,
      "mean_token_accuracy": 0.204638209939003,
      "num_tokens": 155844738.0,
      "step": 71135
    },
    {
      "entropy": 5.71778507232666,
      "epoch": 7.060341405319571,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001409546724873432,
      "loss": 4.8426,
      "mean_token_accuracy": 0.207565775513649,
      "num_tokens": 155855487.0,
      "step": 71140
    },
    {
      "entropy": 5.700548887252808,
      "epoch": 7.0608376339817385,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014092621682148244,
      "loss": 4.8355,
      "mean_token_accuracy": 0.20993481427431107,
      "num_tokens": 155867304.0,
      "step": 71145
    },
    {
      "entropy": 5.669704484939575,
      "epoch": 7.061333862643906,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014089776448036268,
      "loss": 4.8502,
      "mean_token_accuracy": 0.2080484837293625,
      "num_tokens": 155877627.0,
      "step": 71150
    },
    {
      "entropy": 5.749571514129639,
      "epoch": 7.061830091306073,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001408693154646894,
      "loss": 4.9805,
      "mean_token_accuracy": 0.1939941644668579,
      "num_tokens": 155887108.0,
      "step": 71155
    },
    {
      "entropy": 5.701310539245606,
      "epoch": 7.062326319968242,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014084086977516803,
      "loss": 4.9316,
      "mean_token_accuracy": 0.19933822900056838,
      "num_tokens": 155898485.0,
      "step": 71160
    },
    {
      "entropy": 5.714568853378296,
      "epoch": 7.062822548630409,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014081242741250394,
      "loss": 4.9315,
      "mean_token_accuracy": 0.19930094331502915,
      "num_tokens": 155908765.0,
      "step": 71165
    },
    {
      "entropy": 5.641532897949219,
      "epoch": 7.063318777292577,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014078398837740258,
      "loss": 4.8756,
      "mean_token_accuracy": 0.21060508936643602,
      "num_tokens": 155920164.0,
      "step": 71170
    },
    {
      "entropy": 5.71754789352417,
      "epoch": 7.063815005954744,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014075555267056906,
      "loss": 4.9227,
      "mean_token_accuracy": 0.2060350313782692,
      "num_tokens": 155929997.0,
      "step": 71175
    },
    {
      "entropy": 5.678124189376831,
      "epoch": 7.064311234616912,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001407271202927086,
      "loss": 4.877,
      "mean_token_accuracy": 0.20675871521234512,
      "num_tokens": 155941289.0,
      "step": 71180
    },
    {
      "entropy": 5.680104780197143,
      "epoch": 7.064807463279079,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014069869124452622,
      "loss": 4.9089,
      "mean_token_accuracy": 0.19453011006116866,
      "num_tokens": 155951231.0,
      "step": 71185
    },
    {
      "entropy": 5.6442476272583,
      "epoch": 7.065303691941247,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014067026552672693,
      "loss": 4.8684,
      "mean_token_accuracy": 0.20487066209316254,
      "num_tokens": 155962034.0,
      "step": 71190
    },
    {
      "entropy": 5.640228176116944,
      "epoch": 7.065799920603414,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014064184314001548,
      "loss": 4.9023,
      "mean_token_accuracy": 0.201419335603714,
      "num_tokens": 155972207.0,
      "step": 71195
    },
    {
      "entropy": 5.6733927726745605,
      "epoch": 7.066296149265582,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014061342408509686,
      "loss": 4.8574,
      "mean_token_accuracy": 0.2095751866698265,
      "num_tokens": 155983885.0,
      "step": 71200
    },
    {
      "entropy": 5.706667518615722,
      "epoch": 7.066792377927749,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001405850083626758,
      "loss": 4.9194,
      "mean_token_accuracy": 0.20175675302743912,
      "num_tokens": 155994558.0,
      "step": 71205
    },
    {
      "entropy": 5.751026248931884,
      "epoch": 7.0672886065899165,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014055659597345688,
      "loss": 4.9544,
      "mean_token_accuracy": 0.196279776096344,
      "num_tokens": 156005434.0,
      "step": 71210
    },
    {
      "entropy": 5.719570827484131,
      "epoch": 7.067784835252084,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014052818691814473,
      "loss": 4.9029,
      "mean_token_accuracy": 0.20419490039348603,
      "num_tokens": 156016235.0,
      "step": 71215
    },
    {
      "entropy": 5.707778024673462,
      "epoch": 7.0682810639142515,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001404997811974437,
      "loss": 4.8904,
      "mean_token_accuracy": 0.20082794427871703,
      "num_tokens": 156026247.0,
      "step": 71220
    },
    {
      "entropy": 5.706471967697143,
      "epoch": 7.068777292576419,
      "grad_norm": 1.375,
      "learning_rate": 0.00014047137881205836,
      "loss": 4.9286,
      "mean_token_accuracy": 0.20064036846160888,
      "num_tokens": 156039117.0,
      "step": 71225
    },
    {
      "entropy": 5.686757946014405,
      "epoch": 7.069273521238586,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014044297976269278,
      "loss": 4.8914,
      "mean_token_accuracy": 0.207351753115654,
      "num_tokens": 156049605.0,
      "step": 71230
    },
    {
      "entropy": 5.683344984054566,
      "epoch": 7.069769749900754,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001404145840500514,
      "loss": 4.9404,
      "mean_token_accuracy": 0.20142990201711655,
      "num_tokens": 156059562.0,
      "step": 71235
    },
    {
      "entropy": 5.610873508453369,
      "epoch": 7.070265978562921,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001403861916748384,
      "loss": 4.7496,
      "mean_token_accuracy": 0.20737636238336563,
      "num_tokens": 156068808.0,
      "step": 71240
    },
    {
      "entropy": 5.692967796325684,
      "epoch": 7.07076220722509,
      "grad_norm": 1.21875,
      "learning_rate": 0.00014035780263775782,
      "loss": 4.8979,
      "mean_token_accuracy": 0.20251226276159287,
      "num_tokens": 156081440.0,
      "step": 71245
    },
    {
      "entropy": 5.647639989852905,
      "epoch": 7.071258435887257,
      "grad_norm": 1.21875,
      "learning_rate": 0.00014032941693951356,
      "loss": 4.8002,
      "mean_token_accuracy": 0.21089763641357423,
      "num_tokens": 156093452.0,
      "step": 71250
    },
    {
      "entropy": 5.675581073760986,
      "epoch": 7.071754664549425,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001403010345808096,
      "loss": 4.9036,
      "mean_token_accuracy": 0.2011298343539238,
      "num_tokens": 156103946.0,
      "step": 71255
    },
    {
      "entropy": 5.749886989593506,
      "epoch": 7.072250893211592,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00014027265556234965,
      "loss": 4.9091,
      "mean_token_accuracy": 0.203841632604599,
      "num_tokens": 156114140.0,
      "step": 71260
    },
    {
      "entropy": 5.6803594589233395,
      "epoch": 7.07274712187376,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001402442798848375,
      "loss": 4.9779,
      "mean_token_accuracy": 0.1943314865231514,
      "num_tokens": 156128868.0,
      "step": 71265
    },
    {
      "entropy": 5.654379653930664,
      "epoch": 7.073243350535927,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014021590754897693,
      "loss": 4.8031,
      "mean_token_accuracy": 0.2066633090376854,
      "num_tokens": 156139169.0,
      "step": 71270
    },
    {
      "entropy": 5.72494592666626,
      "epoch": 7.073739579198095,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014018753855547144,
      "loss": 4.9453,
      "mean_token_accuracy": 0.20626155883073807,
      "num_tokens": 156151197.0,
      "step": 71275
    },
    {
      "entropy": 5.648976564407349,
      "epoch": 7.074235807860262,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014015917290502444,
      "loss": 4.8877,
      "mean_token_accuracy": 0.2064472869038582,
      "num_tokens": 156161846.0,
      "step": 71280
    },
    {
      "entropy": 5.683689975738526,
      "epoch": 7.0747320365224295,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014013081059833948,
      "loss": 4.8786,
      "mean_token_accuracy": 0.2053200751543045,
      "num_tokens": 156173946.0,
      "step": 71285
    },
    {
      "entropy": 5.716583204269409,
      "epoch": 7.075228265184597,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014010245163611966,
      "loss": 4.8505,
      "mean_token_accuracy": 0.20219292342662812,
      "num_tokens": 156184449.0,
      "step": 71290
    },
    {
      "entropy": 5.71894268989563,
      "epoch": 7.0757244938467645,
      "grad_norm": 1.5,
      "learning_rate": 0.00014007409601906844,
      "loss": 4.9069,
      "mean_token_accuracy": 0.20758169889450073,
      "num_tokens": 156195954.0,
      "step": 71295
    },
    {
      "entropy": 5.682262372970581,
      "epoch": 7.076220722508932,
      "grad_norm": 1.5,
      "learning_rate": 0.00014004574374788887,
      "loss": 4.8787,
      "mean_token_accuracy": 0.20349778681993486,
      "num_tokens": 156206477.0,
      "step": 71300
    },
    {
      "entropy": 5.661362504959106,
      "epoch": 7.0767169511710994,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00014001739482328408,
      "loss": 4.8904,
      "mean_token_accuracy": 0.20117424577474594,
      "num_tokens": 156215951.0,
      "step": 71305
    },
    {
      "entropy": 5.7413524150848385,
      "epoch": 7.077213179833267,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013998904924595706,
      "loss": 4.9458,
      "mean_token_accuracy": 0.20089073479175568,
      "num_tokens": 156226069.0,
      "step": 71310
    },
    {
      "entropy": 5.720739126205444,
      "epoch": 7.077709408495434,
      "grad_norm": 1.5,
      "learning_rate": 0.00013996070701661068,
      "loss": 4.937,
      "mean_token_accuracy": 0.19826053231954574,
      "num_tokens": 156236654.0,
      "step": 71315
    },
    {
      "entropy": 5.787364482879639,
      "epoch": 7.078205637157602,
      "grad_norm": 1.265625,
      "learning_rate": 0.00013993236813594773,
      "loss": 5.039,
      "mean_token_accuracy": 0.19929634481668473,
      "num_tokens": 156247688.0,
      "step": 71320
    },
    {
      "entropy": 5.744591140747071,
      "epoch": 7.078701865819769,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013990403260467107,
      "loss": 4.9898,
      "mean_token_accuracy": 0.1976563185453415,
      "num_tokens": 156259264.0,
      "step": 71325
    },
    {
      "entropy": 5.6426831722259525,
      "epoch": 7.079198094481938,
      "grad_norm": 1.203125,
      "learning_rate": 0.00013987570042348318,
      "loss": 4.836,
      "mean_token_accuracy": 0.2087199404835701,
      "num_tokens": 156270746.0,
      "step": 71330
    },
    {
      "entropy": 5.709367275238037,
      "epoch": 7.079694323144105,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001398473715930869,
      "loss": 4.8214,
      "mean_token_accuracy": 0.21125635057687758,
      "num_tokens": 156280911.0,
      "step": 71335
    },
    {
      "entropy": 5.707295608520508,
      "epoch": 7.080190551806273,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013981904611418456,
      "loss": 4.9217,
      "mean_token_accuracy": 0.19793365746736527,
      "num_tokens": 156292330.0,
      "step": 71340
    },
    {
      "entropy": 5.679799127578735,
      "epoch": 7.08068678046844,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013979072398747854,
      "loss": 4.8498,
      "mean_token_accuracy": 0.21233099102973937,
      "num_tokens": 156303989.0,
      "step": 71345
    },
    {
      "entropy": 5.630174350738526,
      "epoch": 7.081183009130608,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001397624052136712,
      "loss": 4.8136,
      "mean_token_accuracy": 0.2128916010260582,
      "num_tokens": 156314202.0,
      "step": 71350
    },
    {
      "entropy": 5.701168155670166,
      "epoch": 7.081679237792775,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013973408979346482,
      "loss": 4.9211,
      "mean_token_accuracy": 0.20183797925710678,
      "num_tokens": 156325033.0,
      "step": 71355
    },
    {
      "entropy": 5.646885967254638,
      "epoch": 7.082175466454943,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001397057777275615,
      "loss": 4.8439,
      "mean_token_accuracy": 0.20862673372030258,
      "num_tokens": 156336781.0,
      "step": 71360
    },
    {
      "entropy": 5.6877271175384525,
      "epoch": 7.08267169511711,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00013967746901666342,
      "loss": 4.8844,
      "mean_token_accuracy": 0.2002812698483467,
      "num_tokens": 156347228.0,
      "step": 71365
    },
    {
      "entropy": 5.704124689102173,
      "epoch": 7.0831679237792775,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013964916366147255,
      "loss": 4.9416,
      "mean_token_accuracy": 0.19662784337997435,
      "num_tokens": 156358634.0,
      "step": 71370
    },
    {
      "entropy": 5.73200626373291,
      "epoch": 7.083664152441445,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001396208616626907,
      "loss": 4.9327,
      "mean_token_accuracy": 0.19816469699144362,
      "num_tokens": 156369042.0,
      "step": 71375
    },
    {
      "entropy": 5.679078388214111,
      "epoch": 7.0841603811036125,
      "grad_norm": 1.546875,
      "learning_rate": 0.00013959256302101974,
      "loss": 4.8747,
      "mean_token_accuracy": 0.20736856907606124,
      "num_tokens": 156379511.0,
      "step": 71380
    },
    {
      "entropy": 5.701576805114746,
      "epoch": 7.08465660976578,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001395642677371615,
      "loss": 4.8736,
      "mean_token_accuracy": 0.20093981623649598,
      "num_tokens": 156390861.0,
      "step": 71385
    },
    {
      "entropy": 5.669820499420166,
      "epoch": 7.085152838427947,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001395359758118176,
      "loss": 4.9273,
      "mean_token_accuracy": 0.19597553461790085,
      "num_tokens": 156401955.0,
      "step": 71390
    },
    {
      "entropy": 5.6207616329193115,
      "epoch": 7.085649067090115,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013950768724568948,
      "loss": 4.7861,
      "mean_token_accuracy": 0.20713740289211274,
      "num_tokens": 156411977.0,
      "step": 71395
    },
    {
      "entropy": 5.722202968597412,
      "epoch": 7.086145295752282,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001394794020394789,
      "loss": 4.9276,
      "mean_token_accuracy": 0.2017260029911995,
      "num_tokens": 156423716.0,
      "step": 71400
    },
    {
      "entropy": 5.679198408126831,
      "epoch": 7.08664152441445,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013945112019388706,
      "loss": 4.8694,
      "mean_token_accuracy": 0.20398113429546355,
      "num_tokens": 156434033.0,
      "step": 71405
    },
    {
      "entropy": 5.691711711883545,
      "epoch": 7.087137753076617,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001394228417096153,
      "loss": 4.874,
      "mean_token_accuracy": 0.214884190261364,
      "num_tokens": 156445872.0,
      "step": 71410
    },
    {
      "entropy": 5.756961727142334,
      "epoch": 7.087633981738785,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000139394566587365,
      "loss": 4.95,
      "mean_token_accuracy": 0.19973174333572388,
      "num_tokens": 156456305.0,
      "step": 71415
    },
    {
      "entropy": 5.634613418579102,
      "epoch": 7.088130210400953,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013936629482783728,
      "loss": 4.8871,
      "mean_token_accuracy": 0.20874116718769073,
      "num_tokens": 156467684.0,
      "step": 71420
    },
    {
      "entropy": 5.622678518295288,
      "epoch": 7.088626439063121,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00013933802643173308,
      "loss": 4.7927,
      "mean_token_accuracy": 0.2132288321852684,
      "num_tokens": 156477919.0,
      "step": 71425
    },
    {
      "entropy": 5.675550699234009,
      "epoch": 7.089122667725288,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013930976139975355,
      "loss": 4.8613,
      "mean_token_accuracy": 0.20694215446710587,
      "num_tokens": 156488914.0,
      "step": 71430
    },
    {
      "entropy": 5.632128286361694,
      "epoch": 7.089618896387456,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001392814997325995,
      "loss": 4.8598,
      "mean_token_accuracy": 0.20885892063379288,
      "num_tokens": 156500457.0,
      "step": 71435
    },
    {
      "entropy": 5.676010322570801,
      "epoch": 7.090115125049623,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001392532414309719,
      "loss": 4.8702,
      "mean_token_accuracy": 0.20904394537210463,
      "num_tokens": 156510578.0,
      "step": 71440
    },
    {
      "entropy": 5.7270121574401855,
      "epoch": 7.0906113537117905,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001392249864955714,
      "loss": 4.9369,
      "mean_token_accuracy": 0.2016648530960083,
      "num_tokens": 156521495.0,
      "step": 71445
    },
    {
      "entropy": 5.7696692943573,
      "epoch": 7.091107582373958,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013919673492709863,
      "loss": 5.0187,
      "mean_token_accuracy": 0.18738416135311126,
      "num_tokens": 156533205.0,
      "step": 71450
    },
    {
      "entropy": 5.755184316635132,
      "epoch": 7.0916038110361255,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013916848672625422,
      "loss": 4.9507,
      "mean_token_accuracy": 0.19818200618028642,
      "num_tokens": 156543939.0,
      "step": 71455
    },
    {
      "entropy": 5.689589977264404,
      "epoch": 7.092100039698293,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013914024189373857,
      "loss": 4.8977,
      "mean_token_accuracy": 0.20134344100952148,
      "num_tokens": 156556660.0,
      "step": 71460
    },
    {
      "entropy": 5.730050802230835,
      "epoch": 7.09259626836046,
      "grad_norm": 1.25,
      "learning_rate": 0.00013911200043025218,
      "loss": 4.9171,
      "mean_token_accuracy": 0.20257658511400223,
      "num_tokens": 156567552.0,
      "step": 71465
    },
    {
      "entropy": 5.695395612716675,
      "epoch": 7.093092497022628,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001390837623364955,
      "loss": 4.8606,
      "mean_token_accuracy": 0.19774342477321624,
      "num_tokens": 156577370.0,
      "step": 71470
    },
    {
      "entropy": 5.733905506134033,
      "epoch": 7.093588725684795,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013905552761316853,
      "loss": 4.9797,
      "mean_token_accuracy": 0.19711003750562667,
      "num_tokens": 156589951.0,
      "step": 71475
    },
    {
      "entropy": 5.646854543685913,
      "epoch": 7.094084954346963,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001390272962609716,
      "loss": 4.7949,
      "mean_token_accuracy": 0.21293821781873704,
      "num_tokens": 156600431.0,
      "step": 71480
    },
    {
      "entropy": 5.693964958190918,
      "epoch": 7.09458118300913,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013899906828060475,
      "loss": 4.9369,
      "mean_token_accuracy": 0.2001310557126999,
      "num_tokens": 156611424.0,
      "step": 71485
    },
    {
      "entropy": 5.651285982131958,
      "epoch": 7.095077411671298,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013897084367276783,
      "loss": 4.7958,
      "mean_token_accuracy": 0.213789638876915,
      "num_tokens": 156622906.0,
      "step": 71490
    },
    {
      "entropy": 5.7359740257263185,
      "epoch": 7.095573640333465,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013894262243816088,
      "loss": 4.9452,
      "mean_token_accuracy": 0.19775111377239227,
      "num_tokens": 156634178.0,
      "step": 71495
    },
    {
      "entropy": 5.711098527908325,
      "epoch": 7.096069868995633,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001389144045774838,
      "loss": 4.8661,
      "mean_token_accuracy": 0.20615654140710832,
      "num_tokens": 156644030.0,
      "step": 71500
    },
    {
      "entropy": 5.67839035987854,
      "epoch": 7.096566097657801,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013888619009143628,
      "loss": 4.9452,
      "mean_token_accuracy": 0.20270509868860245,
      "num_tokens": 156656003.0,
      "step": 71505
    },
    {
      "entropy": 5.673074579238891,
      "epoch": 7.097062326319969,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001388579789807179,
      "loss": 4.8489,
      "mean_token_accuracy": 0.20363759696483613,
      "num_tokens": 156665920.0,
      "step": 71510
    },
    {
      "entropy": 5.693375730514527,
      "epoch": 7.097558554982136,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001388297712460283,
      "loss": 4.8957,
      "mean_token_accuracy": 0.20040686577558517,
      "num_tokens": 156676162.0,
      "step": 71515
    },
    {
      "entropy": 5.751880359649658,
      "epoch": 7.0980547836443035,
      "grad_norm": 1.203125,
      "learning_rate": 0.00013880156688806695,
      "loss": 4.9644,
      "mean_token_accuracy": 0.20050630569458008,
      "num_tokens": 156688461.0,
      "step": 71520
    },
    {
      "entropy": 5.719194793701172,
      "epoch": 7.098551012306471,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013877336590753316,
      "loss": 4.9351,
      "mean_token_accuracy": 0.20335660576820375,
      "num_tokens": 156699367.0,
      "step": 71525
    },
    {
      "entropy": 5.720231056213379,
      "epoch": 7.0990472409686385,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013874516830512636,
      "loss": 4.9332,
      "mean_token_accuracy": 0.19328353703022003,
      "num_tokens": 156712377.0,
      "step": 71530
    },
    {
      "entropy": 5.68320894241333,
      "epoch": 7.099543469630806,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013871697408154588,
      "loss": 4.9274,
      "mean_token_accuracy": 0.2007551833987236,
      "num_tokens": 156722827.0,
      "step": 71535
    },
    {
      "entropy": 5.680389785766602,
      "epoch": 7.100039698292973,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001386887832374908,
      "loss": 4.8791,
      "mean_token_accuracy": 0.19795038998126985,
      "num_tokens": 156733272.0,
      "step": 71540
    },
    {
      "entropy": 5.717280864715576,
      "epoch": 7.100535926955141,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013866059577366015,
      "loss": 4.8769,
      "mean_token_accuracy": 0.20104065984487535,
      "num_tokens": 156743788.0,
      "step": 71545
    },
    {
      "entropy": 5.7254187107086185,
      "epoch": 7.101032155617308,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001386324116907529,
      "loss": 4.9117,
      "mean_token_accuracy": 0.19824205189943314,
      "num_tokens": 156754573.0,
      "step": 71550
    },
    {
      "entropy": 5.709115695953369,
      "epoch": 7.101528384279476,
      "grad_norm": 1.375,
      "learning_rate": 0.00013860423098946794,
      "loss": 4.9462,
      "mean_token_accuracy": 0.20195056200027467,
      "num_tokens": 156766628.0,
      "step": 71555
    },
    {
      "entropy": 5.6498291969299315,
      "epoch": 7.102024612941643,
      "grad_norm": 1.375,
      "learning_rate": 0.0001385760536705042,
      "loss": 4.8491,
      "mean_token_accuracy": 0.20172980278730393,
      "num_tokens": 156777098.0,
      "step": 71560
    },
    {
      "entropy": 5.760821342468262,
      "epoch": 7.102520841603811,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013854787973456033,
      "loss": 4.9859,
      "mean_token_accuracy": 0.19636074602603912,
      "num_tokens": 156788626.0,
      "step": 71565
    },
    {
      "entropy": 5.684071969985962,
      "epoch": 7.103017070265978,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013851970918233512,
      "loss": 4.924,
      "mean_token_accuracy": 0.19663425385951996,
      "num_tokens": 156798810.0,
      "step": 71570
    },
    {
      "entropy": 5.71145224571228,
      "epoch": 7.103513298928146,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00013849154201452696,
      "loss": 4.9344,
      "mean_token_accuracy": 0.2011778399348259,
      "num_tokens": 156810951.0,
      "step": 71575
    },
    {
      "entropy": 5.698038625717163,
      "epoch": 7.104009527590313,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00013846337823183446,
      "loss": 4.8575,
      "mean_token_accuracy": 0.21032427549362182,
      "num_tokens": 156821838.0,
      "step": 71580
    },
    {
      "entropy": 5.702926254272461,
      "epoch": 7.104505756252481,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001384352178349559,
      "loss": 4.804,
      "mean_token_accuracy": 0.21047921031713485,
      "num_tokens": 156832216.0,
      "step": 71585
    },
    {
      "entropy": 5.70343246459961,
      "epoch": 7.105001984914649,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001384070608245897,
      "loss": 4.885,
      "mean_token_accuracy": 0.19903892874717713,
      "num_tokens": 156842173.0,
      "step": 71590
    },
    {
      "entropy": 5.674070835113525,
      "epoch": 7.1054982135768165,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00013837890720143397,
      "loss": 4.8463,
      "mean_token_accuracy": 0.20698819756507875,
      "num_tokens": 156853141.0,
      "step": 71595
    },
    {
      "entropy": 5.675052738189697,
      "epoch": 7.105994442238984,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00013835075696618705,
      "loss": 4.8258,
      "mean_token_accuracy": 0.20787591636180877,
      "num_tokens": 156862961.0,
      "step": 71600
    },
    {
      "entropy": 5.744757461547851,
      "epoch": 7.1064906709011515,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013832261011954692,
      "loss": 4.9374,
      "mean_token_accuracy": 0.19385927617549897,
      "num_tokens": 156873429.0,
      "step": 71605
    },
    {
      "entropy": 5.6590663433074955,
      "epoch": 7.106986899563319,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013829446666221152,
      "loss": 4.8298,
      "mean_token_accuracy": 0.21181602478027345,
      "num_tokens": 156883971.0,
      "step": 71610
    },
    {
      "entropy": 5.66558313369751,
      "epoch": 7.1074831282254864,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001382663265948787,
      "loss": 4.8771,
      "mean_token_accuracy": 0.19652120620012284,
      "num_tokens": 156894803.0,
      "step": 71615
    },
    {
      "entropy": 5.692909288406372,
      "epoch": 7.107979356887654,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013823818991824647,
      "loss": 4.8881,
      "mean_token_accuracy": 0.20714745968580245,
      "num_tokens": 156904924.0,
      "step": 71620
    },
    {
      "entropy": 5.700956153869629,
      "epoch": 7.108475585549821,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00013821005663301232,
      "loss": 4.9629,
      "mean_token_accuracy": 0.20226589292287828,
      "num_tokens": 156915531.0,
      "step": 71625
    },
    {
      "entropy": 5.694294691085815,
      "epoch": 7.108971814211989,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013818192673987407,
      "loss": 4.9179,
      "mean_token_accuracy": 0.2021494358778,
      "num_tokens": 156926409.0,
      "step": 71630
    },
    {
      "entropy": 5.686023712158203,
      "epoch": 7.109468042874156,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013815380023952923,
      "loss": 4.9322,
      "mean_token_accuracy": 0.2024519756436348,
      "num_tokens": 156936729.0,
      "step": 71635
    },
    {
      "entropy": 5.757370662689209,
      "epoch": 7.109964271536324,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013812567713267527,
      "loss": 4.9757,
      "mean_token_accuracy": 0.1936206579208374,
      "num_tokens": 156947470.0,
      "step": 71640
    },
    {
      "entropy": 5.724515390396118,
      "epoch": 7.110460500198491,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001380975574200095,
      "loss": 4.9142,
      "mean_token_accuracy": 0.19266513139009475,
      "num_tokens": 156958536.0,
      "step": 71645
    },
    {
      "entropy": 5.657345628738403,
      "epoch": 7.110956728860659,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00013806944110222942,
      "loss": 4.8155,
      "mean_token_accuracy": 0.20840597599744798,
      "num_tokens": 156970739.0,
      "step": 71650
    },
    {
      "entropy": 5.696716547012329,
      "epoch": 7.111452957522826,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013804132818003202,
      "loss": 4.875,
      "mean_token_accuracy": 0.2074390321969986,
      "num_tokens": 156981371.0,
      "step": 71655
    },
    {
      "entropy": 5.754714727401733,
      "epoch": 7.111949186184994,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00013801321865411465,
      "loss": 4.9954,
      "mean_token_accuracy": 0.1947585478425026,
      "num_tokens": 156991597.0,
      "step": 71660
    },
    {
      "entropy": 5.691407251358032,
      "epoch": 7.112445414847161,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00013798511252517427,
      "loss": 4.8745,
      "mean_token_accuracy": 0.20100143253803254,
      "num_tokens": 157003835.0,
      "step": 71665
    },
    {
      "entropy": 5.732848596572876,
      "epoch": 7.112941643509329,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013795700979390778,
      "loss": 4.8999,
      "mean_token_accuracy": 0.20585773289203643,
      "num_tokens": 157014075.0,
      "step": 71670
    },
    {
      "entropy": 5.689593839645386,
      "epoch": 7.113437872171497,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00013792891046101224,
      "loss": 4.8839,
      "mean_token_accuracy": 0.20167893320322036,
      "num_tokens": 157024092.0,
      "step": 71675
    },
    {
      "entropy": 5.714580297470093,
      "epoch": 7.1139341008336645,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013790081452718433,
      "loss": 4.9149,
      "mean_token_accuracy": 0.19503521919250488,
      "num_tokens": 157034027.0,
      "step": 71680
    },
    {
      "entropy": 5.74271559715271,
      "epoch": 7.114430329495832,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00013787272199312084,
      "loss": 4.9201,
      "mean_token_accuracy": 0.2025743693113327,
      "num_tokens": 157045706.0,
      "step": 71685
    },
    {
      "entropy": 5.737593317031861,
      "epoch": 7.1149265581579995,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013784463285951827,
      "loss": 4.8916,
      "mean_token_accuracy": 0.1961575672030449,
      "num_tokens": 157056608.0,
      "step": 71690
    },
    {
      "entropy": 5.689865493774414,
      "epoch": 7.115422786820167,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001378165471270733,
      "loss": 4.8444,
      "mean_token_accuracy": 0.20852413028478622,
      "num_tokens": 157066461.0,
      "step": 71695
    },
    {
      "entropy": 5.637199878692627,
      "epoch": 7.115919015482334,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013778846479648232,
      "loss": 4.8264,
      "mean_token_accuracy": 0.20776763260364534,
      "num_tokens": 157078130.0,
      "step": 71700
    },
    {
      "entropy": 5.804298782348633,
      "epoch": 7.116415244144502,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001377603858684418,
      "loss": 5.01,
      "mean_token_accuracy": 0.19465696215629577,
      "num_tokens": 157088013.0,
      "step": 71705
    },
    {
      "entropy": 5.755389738082886,
      "epoch": 7.116911472806669,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013773231034364803,
      "loss": 4.9397,
      "mean_token_accuracy": 0.2066171109676361,
      "num_tokens": 157099349.0,
      "step": 71710
    },
    {
      "entropy": 5.71797251701355,
      "epoch": 7.117407701468837,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013770423822279714,
      "loss": 4.8769,
      "mean_token_accuracy": 0.21073654741048814,
      "num_tokens": 157111088.0,
      "step": 71715
    },
    {
      "entropy": 5.7054750442504885,
      "epoch": 7.117903930131004,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013767616950658524,
      "loss": 4.9002,
      "mean_token_accuracy": 0.19998334646224974,
      "num_tokens": 157121966.0,
      "step": 71720
    },
    {
      "entropy": 5.754423522949219,
      "epoch": 7.118400158793172,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001376481041957085,
      "loss": 4.9351,
      "mean_token_accuracy": 0.19766834676265715,
      "num_tokens": 157133531.0,
      "step": 71725
    },
    {
      "entropy": 5.729764699935913,
      "epoch": 7.118896387455339,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00013762004229086273,
      "loss": 4.9154,
      "mean_token_accuracy": 0.1995529979467392,
      "num_tokens": 157145106.0,
      "step": 71730
    },
    {
      "entropy": 5.638950443267822,
      "epoch": 7.119392616117507,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013759198379274397,
      "loss": 4.8207,
      "mean_token_accuracy": 0.21413156092166902,
      "num_tokens": 157155295.0,
      "step": 71735
    },
    {
      "entropy": 5.716288232803345,
      "epoch": 7.119888844779674,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001375639287020479,
      "loss": 4.9224,
      "mean_token_accuracy": 0.202289479970932,
      "num_tokens": 157164766.0,
      "step": 71740
    },
    {
      "entropy": 5.663158941268921,
      "epoch": 7.120385073441842,
      "grad_norm": 1.28125,
      "learning_rate": 0.00013753587701947025,
      "loss": 4.8269,
      "mean_token_accuracy": 0.20694894194602967,
      "num_tokens": 157176005.0,
      "step": 71745
    },
    {
      "entropy": 5.6430457592010494,
      "epoch": 7.120881302104009,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013750782874570667,
      "loss": 4.7987,
      "mean_token_accuracy": 0.21123116016387938,
      "num_tokens": 157186399.0,
      "step": 71750
    },
    {
      "entropy": 5.624517011642456,
      "epoch": 7.121377530766177,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00013747978388145254,
      "loss": 4.7665,
      "mean_token_accuracy": 0.2173677533864975,
      "num_tokens": 157196085.0,
      "step": 71755
    },
    {
      "entropy": 5.739028978347778,
      "epoch": 7.121873759428345,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013745174242740349,
      "loss": 4.9515,
      "mean_token_accuracy": 0.20102251470088958,
      "num_tokens": 157208205.0,
      "step": 71760
    },
    {
      "entropy": 5.722803592681885,
      "epoch": 7.1223699880905125,
      "grad_norm": 1.375,
      "learning_rate": 0.00013742370438425488,
      "loss": 4.921,
      "mean_token_accuracy": 0.20644254237413406,
      "num_tokens": 157218831.0,
      "step": 71765
    },
    {
      "entropy": 5.732047414779663,
      "epoch": 7.12286621675268,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013739566975270198,
      "loss": 4.8906,
      "mean_token_accuracy": 0.2061823323369026,
      "num_tokens": 157228670.0,
      "step": 71770
    },
    {
      "entropy": 5.731135845184326,
      "epoch": 7.123362445414847,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013736763853343994,
      "loss": 4.9425,
      "mean_token_accuracy": 0.19503204971551896,
      "num_tokens": 157239245.0,
      "step": 71775
    },
    {
      "entropy": 5.721837663650513,
      "epoch": 7.123858674077015,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013733961072716388,
      "loss": 4.939,
      "mean_token_accuracy": 0.1985845223069191,
      "num_tokens": 157252279.0,
      "step": 71780
    },
    {
      "entropy": 5.677509689331055,
      "epoch": 7.124354902739182,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001373115863345688,
      "loss": 4.861,
      "mean_token_accuracy": 0.20283709317445756,
      "num_tokens": 157264187.0,
      "step": 71785
    },
    {
      "entropy": 5.727716302871704,
      "epoch": 7.12485113140135,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013728356535634966,
      "loss": 4.9428,
      "mean_token_accuracy": 0.20039424747228624,
      "num_tokens": 157275247.0,
      "step": 71790
    },
    {
      "entropy": 5.694676542282105,
      "epoch": 7.125347360063517,
      "grad_norm": 1.265625,
      "learning_rate": 0.00013725554779320148,
      "loss": 4.9338,
      "mean_token_accuracy": 0.1989286571741104,
      "num_tokens": 157285901.0,
      "step": 71795
    },
    {
      "entropy": 5.673392915725708,
      "epoch": 7.125843588725685,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013722753364581885,
      "loss": 4.8295,
      "mean_token_accuracy": 0.2082594558596611,
      "num_tokens": 157296768.0,
      "step": 71800
    },
    {
      "entropy": 5.658515977859497,
      "epoch": 7.126339817387852,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001371995229148965,
      "loss": 4.91,
      "mean_token_accuracy": 0.20288011133670808,
      "num_tokens": 157308314.0,
      "step": 71805
    },
    {
      "entropy": 5.695727109909058,
      "epoch": 7.12683604605002,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001371715156011291,
      "loss": 4.8615,
      "mean_token_accuracy": 0.2035292312502861,
      "num_tokens": 157319094.0,
      "step": 71810
    },
    {
      "entropy": 5.769219398498535,
      "epoch": 7.127332274712187,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000137143511705211,
      "loss": 5.0085,
      "mean_token_accuracy": 0.1987955555319786,
      "num_tokens": 157329954.0,
      "step": 71815
    },
    {
      "entropy": 5.655066061019897,
      "epoch": 7.127828503374355,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013711551122783683,
      "loss": 4.8679,
      "mean_token_accuracy": 0.20012110322713852,
      "num_tokens": 157340335.0,
      "step": 71820
    },
    {
      "entropy": 5.708222770690918,
      "epoch": 7.128324732036522,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001370875141697008,
      "loss": 4.9382,
      "mean_token_accuracy": 0.20724345743656158,
      "num_tokens": 157351540.0,
      "step": 71825
    },
    {
      "entropy": 5.747264719009399,
      "epoch": 7.12882096069869,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013705952053149733,
      "loss": 4.9452,
      "mean_token_accuracy": 0.19274647980928422,
      "num_tokens": 157361478.0,
      "step": 71830
    },
    {
      "entropy": 5.664245653152466,
      "epoch": 7.129317189360857,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001370315303139205,
      "loss": 4.8369,
      "mean_token_accuracy": 0.20589675456285478,
      "num_tokens": 157374130.0,
      "step": 71835
    },
    {
      "entropy": 5.690146160125733,
      "epoch": 7.129813418023025,
      "grad_norm": 1.375,
      "learning_rate": 0.00013700354351766443,
      "loss": 4.944,
      "mean_token_accuracy": 0.19622268229722978,
      "num_tokens": 157385672.0,
      "step": 71840
    },
    {
      "entropy": 5.757328128814697,
      "epoch": 7.130309646685193,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001369755601434231,
      "loss": 4.8849,
      "mean_token_accuracy": 0.20400647968053817,
      "num_tokens": 157397081.0,
      "step": 71845
    },
    {
      "entropy": 5.807517194747925,
      "epoch": 7.13080587534736,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013694758019189046,
      "loss": 5.0017,
      "mean_token_accuracy": 0.19906863421201706,
      "num_tokens": 157408290.0,
      "step": 71850
    },
    {
      "entropy": 5.758876752853394,
      "epoch": 7.131302104009528,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013691960366376028,
      "loss": 5.0051,
      "mean_token_accuracy": 0.19962619096040726,
      "num_tokens": 157419532.0,
      "step": 71855
    },
    {
      "entropy": 5.719075775146484,
      "epoch": 7.131798332671695,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013689163055972646,
      "loss": 4.9059,
      "mean_token_accuracy": 0.2003992661833763,
      "num_tokens": 157430735.0,
      "step": 71860
    },
    {
      "entropy": 5.743702030181884,
      "epoch": 7.132294561333863,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001368636608804827,
      "loss": 4.9273,
      "mean_token_accuracy": 0.20160183012485505,
      "num_tokens": 157441772.0,
      "step": 71865
    },
    {
      "entropy": 5.67615818977356,
      "epoch": 7.13279078999603,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001368356946267224,
      "loss": 4.8089,
      "mean_token_accuracy": 0.2020708978176117,
      "num_tokens": 157451554.0,
      "step": 71870
    },
    {
      "entropy": 5.7093933582305905,
      "epoch": 7.133287018658198,
      "grad_norm": 1.53125,
      "learning_rate": 0.00013680773179913918,
      "loss": 4.8609,
      "mean_token_accuracy": 0.20757828503847123,
      "num_tokens": 157462792.0,
      "step": 71875
    },
    {
      "entropy": 5.726345777511597,
      "epoch": 7.133783247320365,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013677977239842637,
      "loss": 4.9069,
      "mean_token_accuracy": 0.2027721121907234,
      "num_tokens": 157474628.0,
      "step": 71880
    },
    {
      "entropy": 5.665797281265259,
      "epoch": 7.134279475982533,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00013675181642527738,
      "loss": 4.8672,
      "mean_token_accuracy": 0.19868576973676683,
      "num_tokens": 157485507.0,
      "step": 71885
    },
    {
      "entropy": 5.7664796829223635,
      "epoch": 7.1347757046447,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001367238638803855,
      "loss": 4.9646,
      "mean_token_accuracy": 0.1989065483212471,
      "num_tokens": 157497056.0,
      "step": 71890
    },
    {
      "entropy": 5.73094367980957,
      "epoch": 7.135271933306868,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013669591476444382,
      "loss": 4.9746,
      "mean_token_accuracy": 0.19831025302410127,
      "num_tokens": 157507833.0,
      "step": 71895
    },
    {
      "entropy": 5.680921363830566,
      "epoch": 7.135768161969035,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013666796907814546,
      "loss": 4.8844,
      "mean_token_accuracy": 0.20398773699998857,
      "num_tokens": 157519163.0,
      "step": 71900
    },
    {
      "entropy": 5.680089235305786,
      "epoch": 7.136264390631203,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013664002682218338,
      "loss": 4.8574,
      "mean_token_accuracy": 0.20713352113962175,
      "num_tokens": 157530533.0,
      "step": 71905
    },
    {
      "entropy": 5.666866064071655,
      "epoch": 7.13676061929337,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013661208799725045,
      "loss": 4.8308,
      "mean_token_accuracy": 0.20777108818292617,
      "num_tokens": 157540314.0,
      "step": 71910
    },
    {
      "entropy": 5.669140291213989,
      "epoch": 7.137256847955538,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013658415260403955,
      "loss": 4.88,
      "mean_token_accuracy": 0.2065647065639496,
      "num_tokens": 157551346.0,
      "step": 71915
    },
    {
      "entropy": 5.6891662120819095,
      "epoch": 7.137753076617705,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001365562206432434,
      "loss": 4.9461,
      "mean_token_accuracy": 0.2004583567380905,
      "num_tokens": 157561196.0,
      "step": 71920
    },
    {
      "entropy": 5.684114122390747,
      "epoch": 7.138249305279873,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013652829211555466,
      "loss": 4.9482,
      "mean_token_accuracy": 0.21080276519060134,
      "num_tokens": 157573414.0,
      "step": 71925
    },
    {
      "entropy": 5.658056688308716,
      "epoch": 7.138745533942041,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013650036702166597,
      "loss": 4.8608,
      "mean_token_accuracy": 0.2018075779080391,
      "num_tokens": 157585275.0,
      "step": 71930
    },
    {
      "entropy": 5.729546403884887,
      "epoch": 7.139241762604208,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001364724453622696,
      "loss": 4.9043,
      "mean_token_accuracy": 0.20281825959682465,
      "num_tokens": 157596664.0,
      "step": 71935
    },
    {
      "entropy": 5.6615253448486325,
      "epoch": 7.139737991266376,
      "grad_norm": 1.5,
      "learning_rate": 0.00013644452713805822,
      "loss": 4.7691,
      "mean_token_accuracy": 0.2124841332435608,
      "num_tokens": 157607035.0,
      "step": 71940
    },
    {
      "entropy": 5.6588376522064205,
      "epoch": 7.140234219928543,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013641661234972395,
      "loss": 4.8882,
      "mean_token_accuracy": 0.20029604882001878,
      "num_tokens": 157617505.0,
      "step": 71945
    },
    {
      "entropy": 5.656561279296875,
      "epoch": 7.140730448590711,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013638870099795902,
      "loss": 4.8926,
      "mean_token_accuracy": 0.19898123443126678,
      "num_tokens": 157627230.0,
      "step": 71950
    },
    {
      "entropy": 5.692783689498901,
      "epoch": 7.141226677252878,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013636079308345574,
      "loss": 4.9074,
      "mean_token_accuracy": 0.20760395228862763,
      "num_tokens": 157639083.0,
      "step": 71955
    },
    {
      "entropy": 5.707951927185059,
      "epoch": 7.141722905915046,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013633288860690596,
      "loss": 4.9924,
      "mean_token_accuracy": 0.19544797390699387,
      "num_tokens": 157650603.0,
      "step": 71960
    },
    {
      "entropy": 5.75445294380188,
      "epoch": 7.142219134577213,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013630498756900172,
      "loss": 4.9688,
      "mean_token_accuracy": 0.19584538489580156,
      "num_tokens": 157660594.0,
      "step": 71965
    },
    {
      "entropy": 5.758073186874389,
      "epoch": 7.142715363239381,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013627708997043502,
      "loss": 4.9095,
      "mean_token_accuracy": 0.19836910218000411,
      "num_tokens": 157671133.0,
      "step": 71970
    },
    {
      "entropy": 5.7353862762451175,
      "epoch": 7.143211591901548,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013624919581189753,
      "loss": 4.9172,
      "mean_token_accuracy": 0.2044796735048294,
      "num_tokens": 157683016.0,
      "step": 71975
    },
    {
      "entropy": 5.647454166412354,
      "epoch": 7.143707820563716,
      "grad_norm": 1.375,
      "learning_rate": 0.00013622130509408103,
      "loss": 4.8323,
      "mean_token_accuracy": 0.20905935764312744,
      "num_tokens": 157695108.0,
      "step": 71980
    },
    {
      "entropy": 5.69374270439148,
      "epoch": 7.144204049225883,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013619341781767702,
      "loss": 4.8743,
      "mean_token_accuracy": 0.2119663119316101,
      "num_tokens": 157706627.0,
      "step": 71985
    },
    {
      "entropy": 5.685423088073731,
      "epoch": 7.144700277888051,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013616553398337723,
      "loss": 4.785,
      "mean_token_accuracy": 0.21748417615890503,
      "num_tokens": 157718336.0,
      "step": 71990
    },
    {
      "entropy": 5.722372531890869,
      "epoch": 7.145196506550218,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013613765359187297,
      "loss": 4.9283,
      "mean_token_accuracy": 0.20406254678964614,
      "num_tokens": 157728800.0,
      "step": 71995
    },
    {
      "entropy": 5.639045238494873,
      "epoch": 7.145692735212386,
      "grad_norm": 1.265625,
      "learning_rate": 0.00013610977664385576,
      "loss": 4.8841,
      "mean_token_accuracy": 0.20586117058992387,
      "num_tokens": 157739819.0,
      "step": 72000
    },
    {
      "epoch": 7.145692735212386,
      "eval_entropy": 5.439657088044307,
      "eval_loss": 5.059041500091553,
      "eval_mean_token_accuracy": 0.19909840453673666,
      "eval_num_tokens": 157739819.0,
      "eval_runtime": 20.2559,
      "eval_samples_per_second": 1662.182,
      "eval_steps_per_second": 207.791,
      "step": 72000
    },
    {
      "entropy": 5.691551685333252,
      "epoch": 7.146188963874553,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013608190314001682,
      "loss": 4.8384,
      "mean_token_accuracy": 0.2061251163482666,
      "num_tokens": 157750131.0,
      "step": 72005
    },
    {
      "entropy": 5.669651985168457,
      "epoch": 7.1466851925367205,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001360540330810473,
      "loss": 4.9081,
      "mean_token_accuracy": 0.20311541259288787,
      "num_tokens": 157761436.0,
      "step": 72010
    },
    {
      "entropy": 5.727641010284424,
      "epoch": 7.147181421198889,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013602616646763831,
      "loss": 4.8854,
      "mean_token_accuracy": 0.20912791639566422,
      "num_tokens": 157773608.0,
      "step": 72015
    },
    {
      "entropy": 5.692449140548706,
      "epoch": 7.147677649861056,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00013599830330048102,
      "loss": 4.914,
      "mean_token_accuracy": 0.19883466213941575,
      "num_tokens": 157784302.0,
      "step": 72020
    },
    {
      "entropy": 5.756846189498901,
      "epoch": 7.148173878523224,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013597044358026617,
      "loss": 4.9315,
      "mean_token_accuracy": 0.20134549885988234,
      "num_tokens": 157794932.0,
      "step": 72025
    },
    {
      "entropy": 5.795044374465943,
      "epoch": 7.148670107185391,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013594258730768483,
      "loss": 4.9461,
      "mean_token_accuracy": 0.19726576507091523,
      "num_tokens": 157806636.0,
      "step": 72030
    },
    {
      "entropy": 5.70615406036377,
      "epoch": 7.149166335847559,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013591473448342766,
      "loss": 4.808,
      "mean_token_accuracy": 0.20331944823265075,
      "num_tokens": 157815856.0,
      "step": 72035
    },
    {
      "entropy": 5.67967038154602,
      "epoch": 7.149662564509726,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001358868851081854,
      "loss": 4.9222,
      "mean_token_accuracy": 0.19986500293016435,
      "num_tokens": 157828775.0,
      "step": 72040
    },
    {
      "entropy": 5.672268104553223,
      "epoch": 7.150158793171894,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00013585903918264858,
      "loss": 4.9082,
      "mean_token_accuracy": 0.20444974899291993,
      "num_tokens": 157840601.0,
      "step": 72045
    },
    {
      "entropy": 5.6729473114013675,
      "epoch": 7.150655021834061,
      "grad_norm": 1.21875,
      "learning_rate": 0.00013583119670750778,
      "loss": 4.8645,
      "mean_token_accuracy": 0.20839623212814332,
      "num_tokens": 157851847.0,
      "step": 72050
    },
    {
      "entropy": 5.69614543914795,
      "epoch": 7.151151250496229,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00013580335768345337,
      "loss": 4.917,
      "mean_token_accuracy": 0.2021842271089554,
      "num_tokens": 157862868.0,
      "step": 72055
    },
    {
      "entropy": 5.764385414123535,
      "epoch": 7.151647479158396,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013577552211117583,
      "loss": 4.9506,
      "mean_token_accuracy": 0.20371021032333375,
      "num_tokens": 157875468.0,
      "step": 72060
    },
    {
      "entropy": 5.664150762557983,
      "epoch": 7.152143707820564,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00013574768999136534,
      "loss": 4.8382,
      "mean_token_accuracy": 0.2020119994878769,
      "num_tokens": 157888299.0,
      "step": 72065
    },
    {
      "entropy": 5.619361162185669,
      "epoch": 7.152639936482731,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001357198613247121,
      "loss": 4.7883,
      "mean_token_accuracy": 0.2115718975663185,
      "num_tokens": 157899673.0,
      "step": 72070
    },
    {
      "entropy": 5.649055671691895,
      "epoch": 7.153136165144899,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001356920361119061,
      "loss": 4.8619,
      "mean_token_accuracy": 0.20486299991607665,
      "num_tokens": 157911008.0,
      "step": 72075
    },
    {
      "entropy": 5.66416974067688,
      "epoch": 7.153632393807066,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001356642143536374,
      "loss": 4.8435,
      "mean_token_accuracy": 0.20235202759504317,
      "num_tokens": 157921014.0,
      "step": 72080
    },
    {
      "entropy": 5.7513445854187015,
      "epoch": 7.1541286224692335,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013563639605059595,
      "loss": 4.9667,
      "mean_token_accuracy": 0.19274420738220216,
      "num_tokens": 157933075.0,
      "step": 72085
    },
    {
      "entropy": 5.7214062213897705,
      "epoch": 7.154624851131401,
      "grad_norm": 1.375,
      "learning_rate": 0.00013560858120347164,
      "loss": 4.9762,
      "mean_token_accuracy": 0.20011950880289078,
      "num_tokens": 157944008.0,
      "step": 72090
    },
    {
      "entropy": 5.746220445632934,
      "epoch": 7.1551210797935685,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013558076981295413,
      "loss": 4.9491,
      "mean_token_accuracy": 0.19946557730436326,
      "num_tokens": 157955899.0,
      "step": 72095
    },
    {
      "entropy": 5.7284139633178714,
      "epoch": 7.155617308455737,
      "grad_norm": 1.46875,
      "learning_rate": 0.00013555296187973314,
      "loss": 4.8982,
      "mean_token_accuracy": 0.204998479783535,
      "num_tokens": 157966359.0,
      "step": 72100
    },
    {
      "entropy": 5.65806565284729,
      "epoch": 7.156113537117904,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001355251574044982,
      "loss": 4.8394,
      "mean_token_accuracy": 0.20812029540538787,
      "num_tokens": 157976297.0,
      "step": 72105
    },
    {
      "entropy": 5.679430341720581,
      "epoch": 7.156609765780072,
      "grad_norm": 1.53125,
      "learning_rate": 0.00013549735638793874,
      "loss": 4.8451,
      "mean_token_accuracy": 0.21403334885835648,
      "num_tokens": 157986979.0,
      "step": 72110
    },
    {
      "entropy": 5.756001663208008,
      "epoch": 7.157105994442239,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013546955883074425,
      "loss": 5.0093,
      "mean_token_accuracy": 0.19273797869682313,
      "num_tokens": 157998456.0,
      "step": 72115
    },
    {
      "entropy": 5.678663730621338,
      "epoch": 7.157602223104407,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001354417647336041,
      "loss": 4.8234,
      "mean_token_accuracy": 0.20520367473363876,
      "num_tokens": 158009898.0,
      "step": 72120
    },
    {
      "entropy": 5.683983087539673,
      "epoch": 7.158098451766574,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001354139740972075,
      "loss": 4.8389,
      "mean_token_accuracy": 0.20871348679065704,
      "num_tokens": 158019852.0,
      "step": 72125
    },
    {
      "entropy": 5.704154109954834,
      "epoch": 7.158594680428742,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013538618692224353,
      "loss": 4.9006,
      "mean_token_accuracy": 0.19861896485090255,
      "num_tokens": 158029340.0,
      "step": 72130
    },
    {
      "entropy": 5.6647810459136965,
      "epoch": 7.159090909090909,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013535840320940125,
      "loss": 4.8666,
      "mean_token_accuracy": 0.20994610488414764,
      "num_tokens": 158040649.0,
      "step": 72135
    },
    {
      "entropy": 5.656534719467163,
      "epoch": 7.159587137753077,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013533062295936974,
      "loss": 4.8907,
      "mean_token_accuracy": 0.20333253741264343,
      "num_tokens": 158050903.0,
      "step": 72140
    },
    {
      "entropy": 5.677427101135254,
      "epoch": 7.160083366415244,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013530284617283773,
      "loss": 4.8481,
      "mean_token_accuracy": 0.20818119049072265,
      "num_tokens": 158061807.0,
      "step": 72145
    },
    {
      "entropy": 5.695191764831543,
      "epoch": 7.160579595077412,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013527507285049405,
      "loss": 4.9064,
      "mean_token_accuracy": 0.20628063678741454,
      "num_tokens": 158073446.0,
      "step": 72150
    },
    {
      "entropy": 5.693844223022461,
      "epoch": 7.161075823739579,
      "grad_norm": 1.28125,
      "learning_rate": 0.00013524730299302763,
      "loss": 4.925,
      "mean_token_accuracy": 0.20395153909921646,
      "num_tokens": 158084082.0,
      "step": 72155
    },
    {
      "entropy": 5.718064212799073,
      "epoch": 7.1615720524017465,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013521953660112693,
      "loss": 4.9427,
      "mean_token_accuracy": 0.20154602378606795,
      "num_tokens": 158094635.0,
      "step": 72160
    },
    {
      "entropy": 5.731692790985107,
      "epoch": 7.162068281063914,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013519177367548053,
      "loss": 4.9062,
      "mean_token_accuracy": 0.20086105316877365,
      "num_tokens": 158106091.0,
      "step": 72165
    },
    {
      "entropy": 5.730930423736572,
      "epoch": 7.1625645097260815,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013516401421677677,
      "loss": 4.91,
      "mean_token_accuracy": 0.2040393903851509,
      "num_tokens": 158118537.0,
      "step": 72170
    },
    {
      "entropy": 5.7080165386199955,
      "epoch": 7.163060738388249,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013513625822570425,
      "loss": 4.9482,
      "mean_token_accuracy": 0.2019999608397484,
      "num_tokens": 158129863.0,
      "step": 72175
    },
    {
      "entropy": 5.675544214248657,
      "epoch": 7.1635569670504164,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013510850570295105,
      "loss": 4.9159,
      "mean_token_accuracy": 0.2035668298602104,
      "num_tokens": 158140950.0,
      "step": 72180
    },
    {
      "entropy": 5.806261157989502,
      "epoch": 7.164053195712585,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00013508075664920555,
      "loss": 5.0215,
      "mean_token_accuracy": 0.19181380271911622,
      "num_tokens": 158150530.0,
      "step": 72185
    },
    {
      "entropy": 5.699373197555542,
      "epoch": 7.164549424374752,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001350530110651558,
      "loss": 4.9405,
      "mean_token_accuracy": 0.19439989924430848,
      "num_tokens": 158161446.0,
      "step": 72190
    },
    {
      "entropy": 5.737164688110352,
      "epoch": 7.16504565303692,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001350252689514898,
      "loss": 4.8534,
      "mean_token_accuracy": 0.20463499724864959,
      "num_tokens": 158170784.0,
      "step": 72195
    },
    {
      "entropy": 5.738394784927368,
      "epoch": 7.165541881699087,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013499753030889544,
      "loss": 4.9196,
      "mean_token_accuracy": 0.20100093334913255,
      "num_tokens": 158182247.0,
      "step": 72200
    },
    {
      "entropy": 5.656092262268066,
      "epoch": 7.166038110361255,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00013496979513806073,
      "loss": 4.8525,
      "mean_token_accuracy": 0.20202552974224092,
      "num_tokens": 158193931.0,
      "step": 72205
    },
    {
      "entropy": 5.6800856590271,
      "epoch": 7.166534339023422,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00013494206343967337,
      "loss": 4.853,
      "mean_token_accuracy": 0.20635769367218018,
      "num_tokens": 158206263.0,
      "step": 72210
    },
    {
      "entropy": 5.696887540817261,
      "epoch": 7.16703056768559,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013491433521442098,
      "loss": 4.8954,
      "mean_token_accuracy": 0.20670554041862488,
      "num_tokens": 158218735.0,
      "step": 72215
    },
    {
      "entropy": 5.628895711898804,
      "epoch": 7.167526796347757,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013488661046299122,
      "loss": 4.8337,
      "mean_token_accuracy": 0.21280868947505951,
      "num_tokens": 158230997.0,
      "step": 72220
    },
    {
      "entropy": 5.774912881851196,
      "epoch": 7.168023025009925,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013485888918607166,
      "loss": 5.0039,
      "mean_token_accuracy": 0.19312473386526108,
      "num_tokens": 158241100.0,
      "step": 72225
    },
    {
      "entropy": 5.8052144050598145,
      "epoch": 7.168519253672092,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001348311713843496,
      "loss": 4.9484,
      "mean_token_accuracy": 0.19910383522510527,
      "num_tokens": 158251487.0,
      "step": 72230
    },
    {
      "entropy": 5.679772901535034,
      "epoch": 7.16901548233426,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001348034570585125,
      "loss": 4.8584,
      "mean_token_accuracy": 0.21170160472393035,
      "num_tokens": 158261772.0,
      "step": 72235
    },
    {
      "entropy": 5.687619209289551,
      "epoch": 7.169511710996427,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001347757462092476,
      "loss": 4.8447,
      "mean_token_accuracy": 0.2024623990058899,
      "num_tokens": 158272219.0,
      "step": 72240
    },
    {
      "entropy": 5.71004090309143,
      "epoch": 7.1700079396585945,
      "grad_norm": 1.265625,
      "learning_rate": 0.00013474803883724195,
      "loss": 4.9085,
      "mean_token_accuracy": 0.19872353225946426,
      "num_tokens": 158285468.0,
      "step": 72245
    },
    {
      "entropy": 5.70142297744751,
      "epoch": 7.170504168320762,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001347203349431828,
      "loss": 4.902,
      "mean_token_accuracy": 0.20413182377815248,
      "num_tokens": 158296919.0,
      "step": 72250
    },
    {
      "entropy": 5.714873266220093,
      "epoch": 7.1710003969829295,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013469263452775703,
      "loss": 4.8767,
      "mean_token_accuracy": 0.19993026405572892,
      "num_tokens": 158308559.0,
      "step": 72255
    },
    {
      "entropy": 5.730121803283692,
      "epoch": 7.171496625645097,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013466493759165156,
      "loss": 4.9054,
      "mean_token_accuracy": 0.20254514813423158,
      "num_tokens": 158319023.0,
      "step": 72260
    },
    {
      "entropy": 5.636056423187256,
      "epoch": 7.171992854307264,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013463724413555328,
      "loss": 4.8715,
      "mean_token_accuracy": 0.2092076912522316,
      "num_tokens": 158330981.0,
      "step": 72265
    },
    {
      "entropy": 5.649602365493775,
      "epoch": 7.172489082969433,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00013460955416014892,
      "loss": 4.8453,
      "mean_token_accuracy": 0.2057017907500267,
      "num_tokens": 158342360.0,
      "step": 72270
    },
    {
      "entropy": 5.672864198684692,
      "epoch": 7.1729853116316,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00013458186766612514,
      "loss": 4.8321,
      "mean_token_accuracy": 0.20783007591962815,
      "num_tokens": 158353651.0,
      "step": 72275
    },
    {
      "entropy": 5.679129552841187,
      "epoch": 7.173481540293768,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001345541846541683,
      "loss": 4.8837,
      "mean_token_accuracy": 0.2016866147518158,
      "num_tokens": 158363360.0,
      "step": 72280
    },
    {
      "entropy": 5.636553192138672,
      "epoch": 7.173977768955935,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013452650512496518,
      "loss": 4.7679,
      "mean_token_accuracy": 0.21122974306344985,
      "num_tokens": 158374475.0,
      "step": 72285
    },
    {
      "entropy": 5.682973957061767,
      "epoch": 7.174473997618103,
      "grad_norm": 1.25,
      "learning_rate": 0.00013449882907920192,
      "loss": 4.8866,
      "mean_token_accuracy": 0.20752134174108505,
      "num_tokens": 158386576.0,
      "step": 72290
    },
    {
      "entropy": 5.636138820648194,
      "epoch": 7.17497022628027,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001344711565175651,
      "loss": 4.8576,
      "mean_token_accuracy": 0.21004777848720552,
      "num_tokens": 158396763.0,
      "step": 72295
    },
    {
      "entropy": 5.644560861587524,
      "epoch": 7.175466454942438,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001344434874407407,
      "loss": 4.8609,
      "mean_token_accuracy": 0.20903206318616868,
      "num_tokens": 158408120.0,
      "step": 72300
    },
    {
      "entropy": 5.67270131111145,
      "epoch": 7.175962683604605,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00013441582184941497,
      "loss": 4.8639,
      "mean_token_accuracy": 0.20237279832363128,
      "num_tokens": 158419169.0,
      "step": 72305
    },
    {
      "entropy": 5.710996675491333,
      "epoch": 7.176458912266773,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013438815974427387,
      "loss": 4.914,
      "mean_token_accuracy": 0.19720242470502852,
      "num_tokens": 158428723.0,
      "step": 72310
    },
    {
      "entropy": 5.712594270706177,
      "epoch": 7.17695514092894,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001343605011260035,
      "loss": 4.9343,
      "mean_token_accuracy": 0.19577233791351317,
      "num_tokens": 158438713.0,
      "step": 72315
    },
    {
      "entropy": 5.7368793964385985,
      "epoch": 7.1774513695911075,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013433284599528953,
      "loss": 4.9425,
      "mean_token_accuracy": 0.1978702113032341,
      "num_tokens": 158449602.0,
      "step": 72320
    },
    {
      "entropy": 5.763683700561524,
      "epoch": 7.177947598253275,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013430519435281797,
      "loss": 4.9622,
      "mean_token_accuracy": 0.19755520522594452,
      "num_tokens": 158461060.0,
      "step": 72325
    },
    {
      "entropy": 5.840780019760132,
      "epoch": 7.1784438269154425,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013427754619927442,
      "loss": 5.0579,
      "mean_token_accuracy": 0.18916474431753158,
      "num_tokens": 158472853.0,
      "step": 72330
    },
    {
      "entropy": 5.689894914627075,
      "epoch": 7.17894005557761,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001342499015353445,
      "loss": 4.8321,
      "mean_token_accuracy": 0.20187076181173325,
      "num_tokens": 158484928.0,
      "step": 72335
    },
    {
      "entropy": 5.6384505271911625,
      "epoch": 7.179436284239777,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00013422226036171364,
      "loss": 4.8361,
      "mean_token_accuracy": 0.202097749710083,
      "num_tokens": 158495356.0,
      "step": 72340
    },
    {
      "entropy": 5.742049646377564,
      "epoch": 7.179932512901945,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013419462267906746,
      "loss": 4.9565,
      "mean_token_accuracy": 0.1936546117067337,
      "num_tokens": 158507376.0,
      "step": 72345
    },
    {
      "entropy": 5.619412660598755,
      "epoch": 7.180428741564112,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013416698848809115,
      "loss": 4.844,
      "mean_token_accuracy": 0.20568960905075073,
      "num_tokens": 158520093.0,
      "step": 72350
    },
    {
      "entropy": 5.664939498901367,
      "epoch": 7.18092497022628,
      "grad_norm": 1.46875,
      "learning_rate": 0.00013413935778947011,
      "loss": 4.907,
      "mean_token_accuracy": 0.19895389229059218,
      "num_tokens": 158529993.0,
      "step": 72355
    },
    {
      "entropy": 5.6861106872558596,
      "epoch": 7.181421198888448,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013411173058388952,
      "loss": 4.8425,
      "mean_token_accuracy": 0.20773327499628066,
      "num_tokens": 158540934.0,
      "step": 72360
    },
    {
      "entropy": 5.683526659011841,
      "epoch": 7.181917427550616,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001340841068720344,
      "loss": 4.8595,
      "mean_token_accuracy": 0.2006866991519928,
      "num_tokens": 158552079.0,
      "step": 72365
    },
    {
      "entropy": 5.665769338607788,
      "epoch": 7.182413656212783,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001340564866545898,
      "loss": 4.9145,
      "mean_token_accuracy": 0.19438720792531966,
      "num_tokens": 158563414.0,
      "step": 72370
    },
    {
      "entropy": 5.701188850402832,
      "epoch": 7.182909884874951,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001340288699322405,
      "loss": 4.8664,
      "mean_token_accuracy": 0.20244863182306289,
      "num_tokens": 158575690.0,
      "step": 72375
    },
    {
      "entropy": 5.6759978294372555,
      "epoch": 7.183406113537118,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013400125670567147,
      "loss": 4.8212,
      "mean_token_accuracy": 0.20705084204673768,
      "num_tokens": 158585831.0,
      "step": 72380
    },
    {
      "entropy": 5.742205429077148,
      "epoch": 7.183902342199286,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001339736469755676,
      "loss": 4.9934,
      "mean_token_accuracy": 0.1957031488418579,
      "num_tokens": 158596170.0,
      "step": 72385
    },
    {
      "entropy": 5.711932277679443,
      "epoch": 7.184398570861453,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00013394604074261332,
      "loss": 4.8922,
      "mean_token_accuracy": 0.2045181691646576,
      "num_tokens": 158608070.0,
      "step": 72390
    },
    {
      "entropy": 5.727312850952148,
      "epoch": 7.1848947995236205,
      "grad_norm": 1.515625,
      "learning_rate": 0.00013391843800749333,
      "loss": 4.8965,
      "mean_token_accuracy": 0.21171730607748032,
      "num_tokens": 158618803.0,
      "step": 72395
    },
    {
      "entropy": 5.711637783050537,
      "epoch": 7.185391028185788,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013389083877089202,
      "loss": 4.9025,
      "mean_token_accuracy": 0.19867535382509233,
      "num_tokens": 158629210.0,
      "step": 72400
    },
    {
      "entropy": 5.684913396835327,
      "epoch": 7.1858872568479555,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001338632430334938,
      "loss": 4.9113,
      "mean_token_accuracy": 0.2009398818016052,
      "num_tokens": 158640781.0,
      "step": 72405
    },
    {
      "entropy": 5.6595629215240475,
      "epoch": 7.186383485510123,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013383565079598303,
      "loss": 4.8,
      "mean_token_accuracy": 0.2106298804283142,
      "num_tokens": 158651079.0,
      "step": 72410
    },
    {
      "entropy": 5.721527147293091,
      "epoch": 7.18687971417229,
      "grad_norm": 1.390625,
      "learning_rate": 0.000133808062059044,
      "loss": 4.8812,
      "mean_token_accuracy": 0.20161220878362657,
      "num_tokens": 158662118.0,
      "step": 72415
    },
    {
      "entropy": 5.684679079055786,
      "epoch": 7.187375942834458,
      "grad_norm": 1.375,
      "learning_rate": 0.0001337804768233608,
      "loss": 4.9162,
      "mean_token_accuracy": 0.19730563461780548,
      "num_tokens": 158673134.0,
      "step": 72420
    },
    {
      "entropy": 5.710492038726807,
      "epoch": 7.187872171496625,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00013375289508961753,
      "loss": 4.8893,
      "mean_token_accuracy": 0.20236373245716094,
      "num_tokens": 158684101.0,
      "step": 72425
    },
    {
      "entropy": 5.667779922485352,
      "epoch": 7.188368400158793,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00013372531685849803,
      "loss": 4.8749,
      "mean_token_accuracy": 0.20190877318382264,
      "num_tokens": 158695482.0,
      "step": 72430
    },
    {
      "entropy": 5.70335202217102,
      "epoch": 7.18886462882096,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001336977421306862,
      "loss": 4.9094,
      "mean_token_accuracy": 0.20401819199323654,
      "num_tokens": 158705529.0,
      "step": 72435
    },
    {
      "entropy": 5.705485582351685,
      "epoch": 7.189360857483128,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013367017090686596,
      "loss": 4.9317,
      "mean_token_accuracy": 0.19801025390625,
      "num_tokens": 158717181.0,
      "step": 72440
    },
    {
      "entropy": 5.736115312576294,
      "epoch": 7.189857086145296,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013364260318772086,
      "loss": 4.9592,
      "mean_token_accuracy": 0.1949923664331436,
      "num_tokens": 158727097.0,
      "step": 72445
    },
    {
      "entropy": 5.734600353240967,
      "epoch": 7.190353314807464,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001336150389739347,
      "loss": 5.008,
      "mean_token_accuracy": 0.1919374495744705,
      "num_tokens": 158738326.0,
      "step": 72450
    },
    {
      "entropy": 5.697879600524902,
      "epoch": 7.190849543469631,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001335874782661909,
      "loss": 4.9342,
      "mean_token_accuracy": 0.19331654012203217,
      "num_tokens": 158750210.0,
      "step": 72455
    },
    {
      "entropy": 5.6960591793060305,
      "epoch": 7.191345772131799,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001335599210651729,
      "loss": 4.9274,
      "mean_token_accuracy": 0.20147404372692107,
      "num_tokens": 158761641.0,
      "step": 72460
    },
    {
      "entropy": 5.69707088470459,
      "epoch": 7.191842000793966,
      "grad_norm": 1.375,
      "learning_rate": 0.00013353236737156403,
      "loss": 4.8825,
      "mean_token_accuracy": 0.20330690741539,
      "num_tokens": 158772669.0,
      "step": 72465
    },
    {
      "entropy": 5.619035530090332,
      "epoch": 7.1923382294561335,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001335048171860477,
      "loss": 4.8489,
      "mean_token_accuracy": 0.20867275297641755,
      "num_tokens": 158784160.0,
      "step": 72470
    },
    {
      "entropy": 5.694373369216919,
      "epoch": 7.192834458118301,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001334772705093069,
      "loss": 4.8849,
      "mean_token_accuracy": 0.2078812152147293,
      "num_tokens": 158794040.0,
      "step": 72475
    },
    {
      "entropy": 5.684793901443482,
      "epoch": 7.1933306867804685,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00013344972734202494,
      "loss": 4.8728,
      "mean_token_accuracy": 0.20413217395544053,
      "num_tokens": 158805265.0,
      "step": 72480
    },
    {
      "entropy": 5.691716909408569,
      "epoch": 7.193826915442636,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013342218768488473,
      "loss": 4.8625,
      "mean_token_accuracy": 0.20263309329748153,
      "num_tokens": 158817027.0,
      "step": 72485
    },
    {
      "entropy": 5.689132833480835,
      "epoch": 7.1943231441048034,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00013339465153856917,
      "loss": 4.892,
      "mean_token_accuracy": 0.20987542122602462,
      "num_tokens": 158829019.0,
      "step": 72490
    },
    {
      "entropy": 5.698130750656128,
      "epoch": 7.194819372766971,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00013336711890376106,
      "loss": 4.9022,
      "mean_token_accuracy": 0.20365194082260132,
      "num_tokens": 158841130.0,
      "step": 72495
    },
    {
      "entropy": 5.7220195770263675,
      "epoch": 7.195315601429138,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001333395897811433,
      "loss": 4.8919,
      "mean_token_accuracy": 0.20349656641483307,
      "num_tokens": 158852051.0,
      "step": 72500
    },
    {
      "entropy": 5.558416652679443,
      "epoch": 7.195811830091306,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00013331206417139843,
      "loss": 4.7666,
      "mean_token_accuracy": 0.21543237715959548,
      "num_tokens": 158863215.0,
      "step": 72505
    },
    {
      "entropy": 5.703126668930054,
      "epoch": 7.196308058753473,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013328454207520903,
      "loss": 4.9216,
      "mean_token_accuracy": 0.20334068536758423,
      "num_tokens": 158873899.0,
      "step": 72510
    },
    {
      "entropy": 5.646862459182739,
      "epoch": 7.196804287415641,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013325702349325765,
      "loss": 4.8327,
      "mean_token_accuracy": 0.20936129838228226,
      "num_tokens": 158884380.0,
      "step": 72515
    },
    {
      "entropy": 5.6728440284729,
      "epoch": 7.197300516077808,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013322950842622664,
      "loss": 4.9341,
      "mean_token_accuracy": 0.2036497563123703,
      "num_tokens": 158895163.0,
      "step": 72520
    },
    {
      "entropy": 5.689735078811646,
      "epoch": 7.197796744739976,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013320199687479829,
      "loss": 4.8042,
      "mean_token_accuracy": 0.20642086416482924,
      "num_tokens": 158905784.0,
      "step": 72525
    },
    {
      "entropy": 5.749758911132813,
      "epoch": 7.198292973402144,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013317448883965494,
      "loss": 5.0187,
      "mean_token_accuracy": 0.19021063596010207,
      "num_tokens": 158917913.0,
      "step": 72530
    },
    {
      "entropy": 5.842835283279419,
      "epoch": 7.198789202064312,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00013314698432147866,
      "loss": 5.0402,
      "mean_token_accuracy": 0.19262974560260773,
      "num_tokens": 158928153.0,
      "step": 72535
    },
    {
      "entropy": 5.753598737716675,
      "epoch": 7.199285430726479,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013311948332095141,
      "loss": 4.9281,
      "mean_token_accuracy": 0.1974254921078682,
      "num_tokens": 158938605.0,
      "step": 72540
    },
    {
      "entropy": 5.720532464981079,
      "epoch": 7.199781659388647,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013309198583875533,
      "loss": 4.9385,
      "mean_token_accuracy": 0.1993263527750969,
      "num_tokens": 158948969.0,
      "step": 72545
    },
    {
      "entropy": 5.750428771972656,
      "epoch": 7.200277888050814,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00013306449187557227,
      "loss": 4.9588,
      "mean_token_accuracy": 0.2005851835012436,
      "num_tokens": 158958599.0,
      "step": 72550
    },
    {
      "entropy": 5.72239408493042,
      "epoch": 7.2007741167129815,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013303700143208383,
      "loss": 4.9771,
      "mean_token_accuracy": 0.20451542884111404,
      "num_tokens": 158970922.0,
      "step": 72555
    },
    {
      "entropy": 5.643392181396484,
      "epoch": 7.201270345375149,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00013300951450897198,
      "loss": 4.8034,
      "mean_token_accuracy": 0.20409848242998124,
      "num_tokens": 158980965.0,
      "step": 72560
    },
    {
      "entropy": 5.737605857849121,
      "epoch": 7.2017665740373165,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00013298203110691814,
      "loss": 4.9477,
      "mean_token_accuracy": 0.20015037208795547,
      "num_tokens": 158992678.0,
      "step": 72565
    },
    {
      "entropy": 5.6908043384552,
      "epoch": 7.202262802699484,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013295455122660385,
      "loss": 4.9039,
      "mean_token_accuracy": 0.2055324837565422,
      "num_tokens": 159004572.0,
      "step": 72570
    },
    {
      "entropy": 5.70646448135376,
      "epoch": 7.202759031361651,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013292707486871073,
      "loss": 4.8825,
      "mean_token_accuracy": 0.20521349161863328,
      "num_tokens": 159016259.0,
      "step": 72575
    },
    {
      "entropy": 5.705427551269532,
      "epoch": 7.203255260023819,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00013289960203392,
      "loss": 4.9316,
      "mean_token_accuracy": 0.19716978967189788,
      "num_tokens": 159028050.0,
      "step": 72580
    },
    {
      "entropy": 5.700286245346069,
      "epoch": 7.203751488685986,
      "grad_norm": 1.53125,
      "learning_rate": 0.00013287213272291282,
      "loss": 4.8658,
      "mean_token_accuracy": 0.20163675099611283,
      "num_tokens": 159038064.0,
      "step": 72585
    },
    {
      "entropy": 5.660014533996582,
      "epoch": 7.204247717348154,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001328446669363706,
      "loss": 4.8728,
      "mean_token_accuracy": 0.20610812157392502,
      "num_tokens": 159049690.0,
      "step": 72590
    },
    {
      "entropy": 5.798710823059082,
      "epoch": 7.204743946010321,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00013281720467497432,
      "loss": 4.9709,
      "mean_token_accuracy": 0.19889054596424102,
      "num_tokens": 159061306.0,
      "step": 72595
    },
    {
      "entropy": 5.742213582992553,
      "epoch": 7.205240174672489,
      "grad_norm": 1.578125,
      "learning_rate": 0.00013278974593940496,
      "loss": 4.8642,
      "mean_token_accuracy": 0.20549783557653428,
      "num_tokens": 159072095.0,
      "step": 72600
    },
    {
      "entropy": 5.733544445037841,
      "epoch": 7.205736403334656,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001327622907303434,
      "loss": 4.964,
      "mean_token_accuracy": 0.20555612295866013,
      "num_tokens": 159082032.0,
      "step": 72605
    },
    {
      "entropy": 5.697149324417114,
      "epoch": 7.206232631996824,
      "grad_norm": 1.546875,
      "learning_rate": 0.00013273483904847062,
      "loss": 4.8915,
      "mean_token_accuracy": 0.20855005383491515,
      "num_tokens": 159092277.0,
      "step": 72610
    },
    {
      "entropy": 5.690350723266602,
      "epoch": 7.206728860658991,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00013270739089446721,
      "loss": 4.9126,
      "mean_token_accuracy": 0.19784609973430634,
      "num_tokens": 159102593.0,
      "step": 72615
    },
    {
      "entropy": 5.667830514907837,
      "epoch": 7.20722508932116,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000132679946269014,
      "loss": 4.8546,
      "mean_token_accuracy": 0.2076399639248848,
      "num_tokens": 159113709.0,
      "step": 72620
    },
    {
      "entropy": 5.682183313369751,
      "epoch": 7.207721317983327,
      "grad_norm": 1.375,
      "learning_rate": 0.00013265250517279138,
      "loss": 4.9124,
      "mean_token_accuracy": 0.20536109805107117,
      "num_tokens": 159124772.0,
      "step": 72625
    },
    {
      "entropy": 5.731638765335083,
      "epoch": 7.2082175466454945,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013262506760647997,
      "loss": 4.9763,
      "mean_token_accuracy": 0.2043616831302643,
      "num_tokens": 159135843.0,
      "step": 72630
    },
    {
      "entropy": 5.6849668502807615,
      "epoch": 7.208713775307662,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013259763357075996,
      "loss": 4.8853,
      "mean_token_accuracy": 0.19907640367746354,
      "num_tokens": 159146601.0,
      "step": 72635
    },
    {
      "entropy": 5.732367134094238,
      "epoch": 7.2092100039698295,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013257020306631185,
      "loss": 4.9703,
      "mean_token_accuracy": 0.19690391570329666,
      "num_tokens": 159156341.0,
      "step": 72640
    },
    {
      "entropy": 5.779343223571777,
      "epoch": 7.209706232631997,
      "grad_norm": 1.46875,
      "learning_rate": 0.00013254277609381577,
      "loss": 4.982,
      "mean_token_accuracy": 0.2012724757194519,
      "num_tokens": 159166805.0,
      "step": 72645
    },
    {
      "entropy": 5.665427160263062,
      "epoch": 7.210202461294164,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013251535265395193,
      "loss": 4.8109,
      "mean_token_accuracy": 0.20801002234220506,
      "num_tokens": 159178537.0,
      "step": 72650
    },
    {
      "entropy": 5.7213201999664305,
      "epoch": 7.210698689956332,
      "grad_norm": 1.515625,
      "learning_rate": 0.00013248793274740035,
      "loss": 4.9545,
      "mean_token_accuracy": 0.19621485322713852,
      "num_tokens": 159188054.0,
      "step": 72655
    },
    {
      "entropy": 5.749304342269897,
      "epoch": 7.211194918618499,
      "grad_norm": 1.46875,
      "learning_rate": 0.00013246051637484088,
      "loss": 4.9506,
      "mean_token_accuracy": 0.20111801475286484,
      "num_tokens": 159198766.0,
      "step": 72660
    },
    {
      "entropy": 5.727352142333984,
      "epoch": 7.211691147280667,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001324331035369535,
      "loss": 4.9497,
      "mean_token_accuracy": 0.1956728145480156,
      "num_tokens": 159209519.0,
      "step": 72665
    },
    {
      "entropy": 5.6856513023376465,
      "epoch": 7.212187375942834,
      "grad_norm": 1.375,
      "learning_rate": 0.0001324056942344179,
      "loss": 4.9049,
      "mean_token_accuracy": 0.2014897122979164,
      "num_tokens": 159220907.0,
      "step": 72670
    },
    {
      "entropy": 5.671416187286377,
      "epoch": 7.212683604605002,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013237828846791378,
      "loss": 4.8871,
      "mean_token_accuracy": 0.20207645744085312,
      "num_tokens": 159232080.0,
      "step": 72675
    },
    {
      "entropy": 5.67426495552063,
      "epoch": 7.213179833267169,
      "grad_norm": 1.578125,
      "learning_rate": 0.00013235088623812086,
      "loss": 4.8683,
      "mean_token_accuracy": 0.20511186718940735,
      "num_tokens": 159241548.0,
      "step": 72680
    },
    {
      "entropy": 5.726625776290893,
      "epoch": 7.213676061929337,
      "grad_norm": 1.625,
      "learning_rate": 0.00013232348754571861,
      "loss": 4.9032,
      "mean_token_accuracy": 0.2052326813340187,
      "num_tokens": 159252445.0,
      "step": 72685
    },
    {
      "entropy": 5.651837396621704,
      "epoch": 7.214172290591504,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001322960923913864,
      "loss": 4.8291,
      "mean_token_accuracy": 0.21088571846485138,
      "num_tokens": 159262780.0,
      "step": 72690
    },
    {
      "entropy": 5.698495006561279,
      "epoch": 7.214668519253672,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001322687007758036,
      "loss": 4.9334,
      "mean_token_accuracy": 0.1986249640583992,
      "num_tokens": 159274507.0,
      "step": 72695
    },
    {
      "entropy": 5.775979661941529,
      "epoch": 7.215164747915839,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001322413126996494,
      "loss": 4.9765,
      "mean_token_accuracy": 0.20086997747421265,
      "num_tokens": 159284694.0,
      "step": 72700
    },
    {
      "entropy": 5.758580207824707,
      "epoch": 7.2156609765780075,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000132213928163603,
      "loss": 4.9887,
      "mean_token_accuracy": 0.1981326162815094,
      "num_tokens": 159296638.0,
      "step": 72705
    },
    {
      "entropy": 5.744736289978027,
      "epoch": 7.216157205240175,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001321865471683436,
      "loss": 4.9179,
      "mean_token_accuracy": 0.20281839668750762,
      "num_tokens": 159308973.0,
      "step": 72710
    },
    {
      "entropy": 5.671735572814941,
      "epoch": 7.2166534339023425,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001321591697145501,
      "loss": 4.8984,
      "mean_token_accuracy": 0.20064034610986708,
      "num_tokens": 159320292.0,
      "step": 72715
    },
    {
      "entropy": 5.692672395706177,
      "epoch": 7.21714966256451,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001321317958029014,
      "loss": 4.9281,
      "mean_token_accuracy": 0.19676046073436737,
      "num_tokens": 159331868.0,
      "step": 72720
    },
    {
      "entropy": 5.660838937759399,
      "epoch": 7.217645891226677,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001321044254340763,
      "loss": 4.903,
      "mean_token_accuracy": 0.20377163141965865,
      "num_tokens": 159343823.0,
      "step": 72725
    },
    {
      "entropy": 5.698949956893921,
      "epoch": 7.218142119888845,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013207705860875346,
      "loss": 4.9067,
      "mean_token_accuracy": 0.20640485286712645,
      "num_tokens": 159354518.0,
      "step": 72730
    },
    {
      "entropy": 5.636074638366699,
      "epoch": 7.218638348551012,
      "grad_norm": 1.265625,
      "learning_rate": 0.00013204969532761167,
      "loss": 4.8591,
      "mean_token_accuracy": 0.20021764039993287,
      "num_tokens": 159365329.0,
      "step": 72735
    },
    {
      "entropy": 5.680110836029053,
      "epoch": 7.21913457721318,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013202233559132936,
      "loss": 4.925,
      "mean_token_accuracy": 0.20126355439424515,
      "num_tokens": 159377240.0,
      "step": 72740
    },
    {
      "entropy": 5.698591613769532,
      "epoch": 7.219630805875347,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00013199497940058507,
      "loss": 4.9358,
      "mean_token_accuracy": 0.20094770342111587,
      "num_tokens": 159387998.0,
      "step": 72745
    },
    {
      "entropy": 5.728777551651001,
      "epoch": 7.220127034537515,
      "grad_norm": 1.21875,
      "learning_rate": 0.00013196762675605716,
      "loss": 4.9212,
      "mean_token_accuracy": 0.19665076285600663,
      "num_tokens": 159399807.0,
      "step": 72750
    },
    {
      "entropy": 5.664008665084839,
      "epoch": 7.220623263199682,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013194027765842394,
      "loss": 4.7976,
      "mean_token_accuracy": 0.21057880371809007,
      "num_tokens": 159410352.0,
      "step": 72755
    },
    {
      "entropy": 5.681488275527954,
      "epoch": 7.22111949186185,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013191293210836343,
      "loss": 4.8996,
      "mean_token_accuracy": 0.2017378494143486,
      "num_tokens": 159419992.0,
      "step": 72760
    },
    {
      "entropy": 5.751036882400513,
      "epoch": 7.221615720524017,
      "grad_norm": 1.515625,
      "learning_rate": 0.000131885590106554,
      "loss": 4.9721,
      "mean_token_accuracy": 0.19409172385931014,
      "num_tokens": 159430498.0,
      "step": 72765
    },
    {
      "entropy": 5.68686466217041,
      "epoch": 7.222111949186185,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00013185825165367343,
      "loss": 4.8734,
      "mean_token_accuracy": 0.208520045876503,
      "num_tokens": 159441275.0,
      "step": 72770
    },
    {
      "entropy": 5.664820003509521,
      "epoch": 7.222608177848352,
      "grad_norm": 1.375,
      "learning_rate": 0.00013183091675039987,
      "loss": 4.8991,
      "mean_token_accuracy": 0.20206703394651412,
      "num_tokens": 159452560.0,
      "step": 72775
    },
    {
      "entropy": 5.704058027267456,
      "epoch": 7.22310440651052,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00013180358539741105,
      "loss": 4.9402,
      "mean_token_accuracy": 0.2015734449028969,
      "num_tokens": 159464178.0,
      "step": 72780
    },
    {
      "entropy": 5.675711917877197,
      "epoch": 7.223600635172687,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001317762575953848,
      "loss": 4.8921,
      "mean_token_accuracy": 0.19882894605398177,
      "num_tokens": 159475389.0,
      "step": 72785
    },
    {
      "entropy": 5.700860643386841,
      "epoch": 7.2240968638348555,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00013174893334499862,
      "loss": 4.921,
      "mean_token_accuracy": 0.20226032733917237,
      "num_tokens": 159486464.0,
      "step": 72790
    },
    {
      "entropy": 5.686761903762817,
      "epoch": 7.224593092497023,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013172161264693035,
      "loss": 4.9046,
      "mean_token_accuracy": 0.20018774271011353,
      "num_tokens": 159496163.0,
      "step": 72795
    },
    {
      "entropy": 5.663548851013184,
      "epoch": 7.2250893211591904,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001316942955018572,
      "loss": 4.8133,
      "mean_token_accuracy": 0.20895123183727266,
      "num_tokens": 159507089.0,
      "step": 72800
    },
    {
      "entropy": 5.693067312240601,
      "epoch": 7.225585549821358,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013166698191045684,
      "loss": 4.9135,
      "mean_token_accuracy": 0.20153867602348327,
      "num_tokens": 159517881.0,
      "step": 72805
    },
    {
      "entropy": 5.694004535675049,
      "epoch": 7.226081778483525,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013163967187340648,
      "loss": 4.9442,
      "mean_token_accuracy": 0.19751725494861602,
      "num_tokens": 159527286.0,
      "step": 72810
    },
    {
      "entropy": 5.731995010375977,
      "epoch": 7.226578007145693,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00013161236539138333,
      "loss": 4.9567,
      "mean_token_accuracy": 0.20041866451501847,
      "num_tokens": 159539182.0,
      "step": 72815
    },
    {
      "entropy": 5.668350648880005,
      "epoch": 7.22707423580786,
      "grad_norm": 1.28125,
      "learning_rate": 0.00013158506246506446,
      "loss": 4.8727,
      "mean_token_accuracy": 0.20109773725271224,
      "num_tokens": 159549632.0,
      "step": 72820
    },
    {
      "entropy": 5.706281089782715,
      "epoch": 7.227570464470028,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00013155776309512713,
      "loss": 4.8955,
      "mean_token_accuracy": 0.1979557454586029,
      "num_tokens": 159561164.0,
      "step": 72825
    },
    {
      "entropy": 5.742655277252197,
      "epoch": 7.228066693132195,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013153046728224812,
      "loss": 4.8971,
      "mean_token_accuracy": 0.20039531886577605,
      "num_tokens": 159571815.0,
      "step": 72830
    },
    {
      "entropy": 5.569243001937866,
      "epoch": 7.228562921794363,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013150317502710435,
      "loss": 4.729,
      "mean_token_accuracy": 0.2144664391875267,
      "num_tokens": 159581071.0,
      "step": 72835
    },
    {
      "entropy": 5.694283819198608,
      "epoch": 7.22905915045653,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013147588633037266,
      "loss": 4.8858,
      "mean_token_accuracy": 0.20245189368724822,
      "num_tokens": 159591513.0,
      "step": 72840
    },
    {
      "entropy": 5.670309591293335,
      "epoch": 7.229555379118698,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001314486011927298,
      "loss": 4.9348,
      "mean_token_accuracy": 0.19976669549942017,
      "num_tokens": 159602506.0,
      "step": 72845
    },
    {
      "entropy": 5.652803182601929,
      "epoch": 7.230051607780865,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013142131961485216,
      "loss": 4.889,
      "mean_token_accuracy": 0.20236520171165467,
      "num_tokens": 159615022.0,
      "step": 72850
    },
    {
      "entropy": 5.666239881515503,
      "epoch": 7.230547836443033,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013139404159741647,
      "loss": 4.8491,
      "mean_token_accuracy": 0.20222472697496413,
      "num_tokens": 159625488.0,
      "step": 72855
    },
    {
      "entropy": 5.655373382568359,
      "epoch": 7.2310440651052,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013136676714109913,
      "loss": 4.9108,
      "mean_token_accuracy": 0.19696174710988998,
      "num_tokens": 159636700.0,
      "step": 72860
    },
    {
      "entropy": 5.685163164138794,
      "epoch": 7.231540293767368,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013133949624657642,
      "loss": 4.8642,
      "mean_token_accuracy": 0.1985608622431755,
      "num_tokens": 159647579.0,
      "step": 72865
    },
    {
      "entropy": 5.676514530181885,
      "epoch": 7.232036522429535,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013131222891452467,
      "loss": 4.8943,
      "mean_token_accuracy": 0.20265026539564132,
      "num_tokens": 159657627.0,
      "step": 72870
    },
    {
      "entropy": 5.684318542480469,
      "epoch": 7.2325327510917035,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013128496514562004,
      "loss": 4.8669,
      "mean_token_accuracy": 0.19994088262319565,
      "num_tokens": 159669550.0,
      "step": 72875
    },
    {
      "entropy": 5.661181545257568,
      "epoch": 7.233028979753871,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013125770494053853,
      "loss": 4.7908,
      "mean_token_accuracy": 0.21252347081899642,
      "num_tokens": 159679909.0,
      "step": 72880
    },
    {
      "entropy": 5.718219900131226,
      "epoch": 7.233525208416038,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001312304482999563,
      "loss": 5.0018,
      "mean_token_accuracy": 0.19645613580942153,
      "num_tokens": 159691920.0,
      "step": 72885
    },
    {
      "entropy": 5.708865642547607,
      "epoch": 7.234021437078206,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001312031952245491,
      "loss": 4.9354,
      "mean_token_accuracy": 0.20659729093313217,
      "num_tokens": 159704153.0,
      "step": 72890
    },
    {
      "entropy": 5.6708286762237545,
      "epoch": 7.234517665740373,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00013117594571499287,
      "loss": 4.866,
      "mean_token_accuracy": 0.19763700366020204,
      "num_tokens": 159716513.0,
      "step": 72895
    },
    {
      "entropy": 5.6244225025177,
      "epoch": 7.235013894402541,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013114869977196315,
      "loss": 4.8087,
      "mean_token_accuracy": 0.20697291791439057,
      "num_tokens": 159727779.0,
      "step": 72900
    },
    {
      "entropy": 5.659678363800049,
      "epoch": 7.235510123064708,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013112145739613577,
      "loss": 4.8461,
      "mean_token_accuracy": 0.2041570395231247,
      "num_tokens": 159738359.0,
      "step": 72905
    },
    {
      "entropy": 5.697724342346191,
      "epoch": 7.236006351726876,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013109421858818616,
      "loss": 4.8476,
      "mean_token_accuracy": 0.2103655830025673,
      "num_tokens": 159748745.0,
      "step": 72910
    },
    {
      "entropy": 5.733297443389892,
      "epoch": 7.236502580389043,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013106698334878992,
      "loss": 4.8951,
      "mean_token_accuracy": 0.2018958166241646,
      "num_tokens": 159758991.0,
      "step": 72915
    },
    {
      "entropy": 5.729879093170166,
      "epoch": 7.236998809051211,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013103975167862232,
      "loss": 4.8517,
      "mean_token_accuracy": 0.20919674634933472,
      "num_tokens": 159769924.0,
      "step": 72920
    },
    {
      "entropy": 5.682846546173096,
      "epoch": 7.237495037713378,
      "grad_norm": 1.890625,
      "learning_rate": 0.0001310125235783587,
      "loss": 4.9057,
      "mean_token_accuracy": 0.20106278359889984,
      "num_tokens": 159779790.0,
      "step": 72925
    },
    {
      "entropy": 5.684907293319702,
      "epoch": 7.237991266375546,
      "grad_norm": 1.5,
      "learning_rate": 0.0001309852990486741,
      "loss": 4.8807,
      "mean_token_accuracy": 0.199357008934021,
      "num_tokens": 159791994.0,
      "step": 72930
    },
    {
      "entropy": 5.586900806427002,
      "epoch": 7.238487495037713,
      "grad_norm": 1.46875,
      "learning_rate": 0.00013095807809024383,
      "loss": 4.7335,
      "mean_token_accuracy": 0.22300194650888444,
      "num_tokens": 159802217.0,
      "step": 72935
    },
    {
      "entropy": 5.742870903015136,
      "epoch": 7.238983723699881,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001309308607037429,
      "loss": 4.9993,
      "mean_token_accuracy": 0.2007051706314087,
      "num_tokens": 159812636.0,
      "step": 72940
    },
    {
      "entropy": 5.695516633987427,
      "epoch": 7.239479952362048,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013090364688984616,
      "loss": 4.9113,
      "mean_token_accuracy": 0.19531598091125488,
      "num_tokens": 159822734.0,
      "step": 72945
    },
    {
      "entropy": 5.689391660690307,
      "epoch": 7.239976181024216,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001308764366492285,
      "loss": 4.9373,
      "mean_token_accuracy": 0.20011449754238128,
      "num_tokens": 159834747.0,
      "step": 72950
    },
    {
      "entropy": 5.616680431365967,
      "epoch": 7.240472409686383,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00013084922998256462,
      "loss": 4.7908,
      "mean_token_accuracy": 0.2081227958202362,
      "num_tokens": 159846961.0,
      "step": 72955
    },
    {
      "entropy": 5.709948110580444,
      "epoch": 7.240968638348551,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013082202689052925,
      "loss": 4.8651,
      "mean_token_accuracy": 0.20771242678165436,
      "num_tokens": 159859135.0,
      "step": 72960
    },
    {
      "entropy": 5.696236324310303,
      "epoch": 7.241464867010719,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013079482737379686,
      "loss": 4.9044,
      "mean_token_accuracy": 0.20074516981840135,
      "num_tokens": 159870324.0,
      "step": 72965
    },
    {
      "entropy": 5.739821910858154,
      "epoch": 7.241961095672886,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013076763143304198,
      "loss": 4.937,
      "mean_token_accuracy": 0.20454823225736618,
      "num_tokens": 159882507.0,
      "step": 72970
    },
    {
      "entropy": 5.742320919036866,
      "epoch": 7.242457324335054,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013074043906893918,
      "loss": 4.8835,
      "mean_token_accuracy": 0.2077579379081726,
      "num_tokens": 159893331.0,
      "step": 72975
    },
    {
      "entropy": 5.731770849227905,
      "epoch": 7.242953552997221,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013071325028216258,
      "loss": 4.9531,
      "mean_token_accuracy": 0.2026106372475624,
      "num_tokens": 159905974.0,
      "step": 72980
    },
    {
      "entropy": 5.711605548858643,
      "epoch": 7.243449781659389,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001306860650733865,
      "loss": 4.9062,
      "mean_token_accuracy": 0.19882650375366212,
      "num_tokens": 159916412.0,
      "step": 72985
    },
    {
      "entropy": 5.688416814804077,
      "epoch": 7.243946010321556,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013065888344328502,
      "loss": 4.8593,
      "mean_token_accuracy": 0.2043983280658722,
      "num_tokens": 159926763.0,
      "step": 72990
    },
    {
      "entropy": 5.765195274353028,
      "epoch": 7.244442238983724,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013063170539253218,
      "loss": 4.948,
      "mean_token_accuracy": 0.20079251378774643,
      "num_tokens": 159937782.0,
      "step": 72995
    },
    {
      "entropy": 5.670977687835693,
      "epoch": 7.244938467645891,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013060453092180192,
      "loss": 4.8118,
      "mean_token_accuracy": 0.2131551280617714,
      "num_tokens": 159949346.0,
      "step": 73000
    },
    {
      "entropy": 5.7478172302246096,
      "epoch": 7.245434696308059,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001305773600317682,
      "loss": 4.996,
      "mean_token_accuracy": 0.19485663920640944,
      "num_tokens": 159961456.0,
      "step": 73005
    },
    {
      "entropy": 5.633803606033325,
      "epoch": 7.245930924970226,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013055019272310476,
      "loss": 4.7846,
      "mean_token_accuracy": 0.20526711791753768,
      "num_tokens": 159971836.0,
      "step": 73010
    },
    {
      "entropy": 5.720444917678833,
      "epoch": 7.246427153632394,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013052302899648526,
      "loss": 4.9294,
      "mean_token_accuracy": 0.19933455735445021,
      "num_tokens": 159983142.0,
      "step": 73015
    },
    {
      "entropy": 5.726099824905395,
      "epoch": 7.246923382294561,
      "grad_norm": 1.25,
      "learning_rate": 0.00013049586885258336,
      "loss": 4.8876,
      "mean_token_accuracy": 0.2087286353111267,
      "num_tokens": 159993036.0,
      "step": 73020
    },
    {
      "entropy": 5.628786563873291,
      "epoch": 7.247419610956729,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00013046871229207248,
      "loss": 4.8082,
      "mean_token_accuracy": 0.21579758077859879,
      "num_tokens": 160002889.0,
      "step": 73025
    },
    {
      "entropy": 5.651839733123779,
      "epoch": 7.247915839618896,
      "grad_norm": 1.25,
      "learning_rate": 0.00013044155931562612,
      "loss": 4.9447,
      "mean_token_accuracy": 0.19748458564281463,
      "num_tokens": 160013945.0,
      "step": 73030
    },
    {
      "entropy": 5.700446557998657,
      "epoch": 7.2484120682810635,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013041440992391757,
      "loss": 4.9214,
      "mean_token_accuracy": 0.19962865710258484,
      "num_tokens": 160023747.0,
      "step": 73035
    },
    {
      "entropy": 5.723285818099976,
      "epoch": 7.248908296943231,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013038726411762015,
      "loss": 4.9242,
      "mean_token_accuracy": 0.20951287746429442,
      "num_tokens": 160034103.0,
      "step": 73040
    },
    {
      "entropy": 5.691177845001221,
      "epoch": 7.249404525605399,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013036012189740693,
      "loss": 4.9099,
      "mean_token_accuracy": 0.20019483268260957,
      "num_tokens": 160045950.0,
      "step": 73045
    },
    {
      "entropy": 5.746025705337525,
      "epoch": 7.249900754267567,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013033298326395104,
      "loss": 4.8902,
      "mean_token_accuracy": 0.19873895794153212,
      "num_tokens": 160056916.0,
      "step": 73050
    },
    {
      "entropy": 5.76577935218811,
      "epoch": 7.250396982929734,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001303058482179254,
      "loss": 4.9673,
      "mean_token_accuracy": 0.20157540440559388,
      "num_tokens": 160066338.0,
      "step": 73055
    },
    {
      "entropy": 5.710266351699829,
      "epoch": 7.250893211591902,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013027871676000294,
      "loss": 4.852,
      "mean_token_accuracy": 0.2040468767285347,
      "num_tokens": 160077945.0,
      "step": 73060
    },
    {
      "entropy": 5.703149604797363,
      "epoch": 7.251389440254069,
      "grad_norm": 1.53125,
      "learning_rate": 0.00013025158889085642,
      "loss": 4.9018,
      "mean_token_accuracy": 0.20361482501029968,
      "num_tokens": 160088593.0,
      "step": 73065
    },
    {
      "entropy": 5.640745687484741,
      "epoch": 7.251885668916237,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001302244646111587,
      "loss": 4.8442,
      "mean_token_accuracy": 0.20385676473379136,
      "num_tokens": 160099747.0,
      "step": 73070
    },
    {
      "entropy": 5.706851434707642,
      "epoch": 7.252381897578404,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001301973439215822,
      "loss": 4.9337,
      "mean_token_accuracy": 0.200126351416111,
      "num_tokens": 160110581.0,
      "step": 73075
    },
    {
      "entropy": 5.697217178344727,
      "epoch": 7.252878126240572,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013017022682279967,
      "loss": 4.9257,
      "mean_token_accuracy": 0.2012331858277321,
      "num_tokens": 160122349.0,
      "step": 73080
    },
    {
      "entropy": 5.641365480422974,
      "epoch": 7.253374354902739,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001301431133154833,
      "loss": 4.8914,
      "mean_token_accuracy": 0.20257862210273742,
      "num_tokens": 160133643.0,
      "step": 73085
    },
    {
      "entropy": 5.6674487590789795,
      "epoch": 7.253870583564907,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013011600340030562,
      "loss": 4.8601,
      "mean_token_accuracy": 0.2065005749464035,
      "num_tokens": 160145065.0,
      "step": 73090
    },
    {
      "entropy": 5.66022047996521,
      "epoch": 7.254366812227074,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013008889707793883,
      "loss": 4.8347,
      "mean_token_accuracy": 0.210001839697361,
      "num_tokens": 160156605.0,
      "step": 73095
    },
    {
      "entropy": 5.710678339004517,
      "epoch": 7.254863040889242,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013006179434905518,
      "loss": 4.9206,
      "mean_token_accuracy": 0.20270813256502151,
      "num_tokens": 160167451.0,
      "step": 73100
    },
    {
      "entropy": 5.693026924133301,
      "epoch": 7.255359269551409,
      "grad_norm": 1.59375,
      "learning_rate": 0.00013003469521432676,
      "loss": 4.8643,
      "mean_token_accuracy": 0.2042415753006935,
      "num_tokens": 160178196.0,
      "step": 73105
    },
    {
      "entropy": 5.681305837631226,
      "epoch": 7.255855498213577,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001300075996744255,
      "loss": 4.8307,
      "mean_token_accuracy": 0.20196837186813354,
      "num_tokens": 160189187.0,
      "step": 73110
    },
    {
      "entropy": 5.699272203445434,
      "epoch": 7.256351726875744,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00012998050773002323,
      "loss": 4.8973,
      "mean_token_accuracy": 0.20589228868484497,
      "num_tokens": 160198826.0,
      "step": 73115
    },
    {
      "entropy": 5.646476078033447,
      "epoch": 7.2568479555379115,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000129953419381792,
      "loss": 4.8087,
      "mean_token_accuracy": 0.21167055517435074,
      "num_tokens": 160209322.0,
      "step": 73120
    },
    {
      "entropy": 5.673390293121338,
      "epoch": 7.257344184200079,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012992633463040343,
      "loss": 4.8281,
      "mean_token_accuracy": 0.20868366509675979,
      "num_tokens": 160220600.0,
      "step": 73125
    },
    {
      "entropy": 5.675576210021973,
      "epoch": 7.257840412862247,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00012989925347652904,
      "loss": 4.9026,
      "mean_token_accuracy": 0.20138546973466873,
      "num_tokens": 160231808.0,
      "step": 73130
    },
    {
      "entropy": 5.682719278335571,
      "epoch": 7.258336641524415,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012987217592084062,
      "loss": 4.8361,
      "mean_token_accuracy": 0.19976697713136674,
      "num_tokens": 160241541.0,
      "step": 73135
    },
    {
      "entropy": 5.686421632766724,
      "epoch": 7.258832870186582,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012984510196400946,
      "loss": 4.8584,
      "mean_token_accuracy": 0.2045758157968521,
      "num_tokens": 160251581.0,
      "step": 73140
    },
    {
      "entropy": 5.739494657516479,
      "epoch": 7.25932909884875,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012981803160670697,
      "loss": 4.9543,
      "mean_token_accuracy": 0.18982361257076263,
      "num_tokens": 160262977.0,
      "step": 73145
    },
    {
      "entropy": 5.670151233673096,
      "epoch": 7.259825327510917,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00012979096484960447,
      "loss": 4.9441,
      "mean_token_accuracy": 0.18757739514112473,
      "num_tokens": 160275916.0,
      "step": 73150
    },
    {
      "entropy": 5.737210988998413,
      "epoch": 7.260321556173085,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012976390169337318,
      "loss": 4.8988,
      "mean_token_accuracy": 0.20053697377443314,
      "num_tokens": 160286977.0,
      "step": 73155
    },
    {
      "entropy": 5.68921480178833,
      "epoch": 7.260817784835252,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001297368421386841,
      "loss": 4.8721,
      "mean_token_accuracy": 0.20323170125484466,
      "num_tokens": 160297946.0,
      "step": 73160
    },
    {
      "entropy": 5.727407789230346,
      "epoch": 7.26131401349742,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001297097861862084,
      "loss": 4.8332,
      "mean_token_accuracy": 0.206004236638546,
      "num_tokens": 160308593.0,
      "step": 73165
    },
    {
      "entropy": 5.735522031784058,
      "epoch": 7.261810242159587,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012968273383661682,
      "loss": 4.9372,
      "mean_token_accuracy": 0.19764201790094377,
      "num_tokens": 160319427.0,
      "step": 73170
    },
    {
      "entropy": 5.6660017490386965,
      "epoch": 7.262306470821755,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012965568509058043,
      "loss": 4.8615,
      "mean_token_accuracy": 0.20213250666856766,
      "num_tokens": 160330677.0,
      "step": 73175
    },
    {
      "entropy": 5.666676950454712,
      "epoch": 7.262802699483922,
      "grad_norm": 1.296875,
      "learning_rate": 0.00012962863994876982,
      "loss": 4.9294,
      "mean_token_accuracy": 0.19702020436525344,
      "num_tokens": 160341025.0,
      "step": 73180
    },
    {
      "entropy": 5.6885413646698,
      "epoch": 7.26329892814609,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012960159841185573,
      "loss": 4.8778,
      "mean_token_accuracy": 0.20240346044301988,
      "num_tokens": 160350222.0,
      "step": 73185
    },
    {
      "entropy": 5.698105812072754,
      "epoch": 7.263795156808257,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001295745604805087,
      "loss": 4.9047,
      "mean_token_accuracy": 0.2034572795033455,
      "num_tokens": 160361120.0,
      "step": 73190
    },
    {
      "entropy": 5.683892393112183,
      "epoch": 7.2642913854704245,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012954752615539907,
      "loss": 4.8465,
      "mean_token_accuracy": 0.21324395686388015,
      "num_tokens": 160372209.0,
      "step": 73195
    },
    {
      "entropy": 5.667318439483642,
      "epoch": 7.264787614132592,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012952049543719742,
      "loss": 4.8874,
      "mean_token_accuracy": 0.20380233079195023,
      "num_tokens": 160383191.0,
      "step": 73200
    },
    {
      "entropy": 5.670415830612183,
      "epoch": 7.2652838427947595,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012949346832657408,
      "loss": 4.877,
      "mean_token_accuracy": 0.20681898295879364,
      "num_tokens": 160394604.0,
      "step": 73205
    },
    {
      "entropy": 5.704757785797119,
      "epoch": 7.265780071456927,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012946644482419916,
      "loss": 4.8896,
      "mean_token_accuracy": 0.20550418198108672,
      "num_tokens": 160405509.0,
      "step": 73210
    },
    {
      "entropy": 5.751667261123657,
      "epoch": 7.266276300119095,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012943942493074282,
      "loss": 4.9502,
      "mean_token_accuracy": 0.19673823565244675,
      "num_tokens": 160415698.0,
      "step": 73215
    },
    {
      "entropy": 5.673491907119751,
      "epoch": 7.266772528781263,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001294124086468751,
      "loss": 4.8314,
      "mean_token_accuracy": 0.21378700137138368,
      "num_tokens": 160426194.0,
      "step": 73220
    },
    {
      "entropy": 5.708978891372681,
      "epoch": 7.26726875744343,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012938539597326585,
      "loss": 4.9191,
      "mean_token_accuracy": 0.20079900622367858,
      "num_tokens": 160435560.0,
      "step": 73225
    },
    {
      "entropy": 5.635660934448242,
      "epoch": 7.267764986105598,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000129358386910585,
      "loss": 4.8024,
      "mean_token_accuracy": 0.2140857309103012,
      "num_tokens": 160446638.0,
      "step": 73230
    },
    {
      "entropy": 5.697955465316772,
      "epoch": 7.268261214767765,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001293313814595024,
      "loss": 4.8805,
      "mean_token_accuracy": 0.20214518308639526,
      "num_tokens": 160456812.0,
      "step": 73235
    },
    {
      "entropy": 5.684788990020752,
      "epoch": 7.268757443429933,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012930437962068765,
      "loss": 4.9097,
      "mean_token_accuracy": 0.20260002911090852,
      "num_tokens": 160468293.0,
      "step": 73240
    },
    {
      "entropy": 5.70619764328003,
      "epoch": 7.2692536720921,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012927738139481034,
      "loss": 4.956,
      "mean_token_accuracy": 0.19845926463603974,
      "num_tokens": 160478850.0,
      "step": 73245
    },
    {
      "entropy": 5.626395654678345,
      "epoch": 7.269749900754268,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012925038678253997,
      "loss": 4.8086,
      "mean_token_accuracy": 0.2081681564450264,
      "num_tokens": 160490676.0,
      "step": 73250
    },
    {
      "entropy": 5.753821802139282,
      "epoch": 7.270246129416435,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001292233957845459,
      "loss": 4.9645,
      "mean_token_accuracy": 0.196492500603199,
      "num_tokens": 160501806.0,
      "step": 73255
    },
    {
      "entropy": 5.661699056625366,
      "epoch": 7.270742358078603,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001291964084014975,
      "loss": 4.8057,
      "mean_token_accuracy": 0.20457952618598937,
      "num_tokens": 160511529.0,
      "step": 73260
    },
    {
      "entropy": 5.691360998153686,
      "epoch": 7.27123858674077,
      "grad_norm": 1.296875,
      "learning_rate": 0.000129169424634064,
      "loss": 4.8775,
      "mean_token_accuracy": 0.2066958248615265,
      "num_tokens": 160522708.0,
      "step": 73265
    },
    {
      "entropy": 5.666064739227295,
      "epoch": 7.2717348154029375,
      "grad_norm": 1.265625,
      "learning_rate": 0.00012914244448291456,
      "loss": 4.8512,
      "mean_token_accuracy": 0.21010938435792922,
      "num_tokens": 160534157.0,
      "step": 73270
    },
    {
      "entropy": 5.66432752609253,
      "epoch": 7.272231044065105,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001291154679487182,
      "loss": 4.9154,
      "mean_token_accuracy": 0.2052036255598068,
      "num_tokens": 160544057.0,
      "step": 73275
    },
    {
      "entropy": 5.7196485042572025,
      "epoch": 7.2727272727272725,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001290884950321439,
      "loss": 4.937,
      "mean_token_accuracy": 0.21139743477106093,
      "num_tokens": 160556868.0,
      "step": 73280
    },
    {
      "entropy": 5.701647615432739,
      "epoch": 7.27322350138944,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012906152573386047,
      "loss": 4.93,
      "mean_token_accuracy": 0.20117175728082656,
      "num_tokens": 160567618.0,
      "step": 73285
    },
    {
      "entropy": 5.657708311080933,
      "epoch": 7.273719730051607,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012903456005453672,
      "loss": 4.7947,
      "mean_token_accuracy": 0.20664112865924836,
      "num_tokens": 160578804.0,
      "step": 73290
    },
    {
      "entropy": 5.775219535827636,
      "epoch": 7.274215958713775,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001290075979948413,
      "loss": 5.0253,
      "mean_token_accuracy": 0.19453238546848298,
      "num_tokens": 160589919.0,
      "step": 73295
    },
    {
      "entropy": 5.680251598358154,
      "epoch": 7.274712187375943,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012898063955544298,
      "loss": 4.89,
      "mean_token_accuracy": 0.2039985865354538,
      "num_tokens": 160599420.0,
      "step": 73300
    },
    {
      "entropy": 5.703973960876465,
      "epoch": 7.275208416038111,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012895368473701013,
      "loss": 4.8566,
      "mean_token_accuracy": 0.20560289919376373,
      "num_tokens": 160610502.0,
      "step": 73305
    },
    {
      "entropy": 5.727679252624512,
      "epoch": 7.275704644700278,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001289267335402112,
      "loss": 4.95,
      "mean_token_accuracy": 0.19942206293344497,
      "num_tokens": 160620813.0,
      "step": 73310
    },
    {
      "entropy": 5.656221199035644,
      "epoch": 7.276200873362446,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001288997859657145,
      "loss": 4.8196,
      "mean_token_accuracy": 0.20611786097288132,
      "num_tokens": 160632311.0,
      "step": 73315
    },
    {
      "entropy": 5.622810220718383,
      "epoch": 7.276697102024613,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012887284201418825,
      "loss": 4.9169,
      "mean_token_accuracy": 0.1952848970890045,
      "num_tokens": 160642267.0,
      "step": 73320
    },
    {
      "entropy": 5.65570478439331,
      "epoch": 7.277193330686781,
      "grad_norm": 1.375,
      "learning_rate": 0.00012884590168630062,
      "loss": 4.858,
      "mean_token_accuracy": 0.2099135309457779,
      "num_tokens": 160653947.0,
      "step": 73325
    },
    {
      "entropy": 5.721548938751221,
      "epoch": 7.277689559348948,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001288189649827198,
      "loss": 4.9471,
      "mean_token_accuracy": 0.20398055613040925,
      "num_tokens": 160666656.0,
      "step": 73330
    },
    {
      "entropy": 5.691078424453735,
      "epoch": 7.278185788011116,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00012879203190411365,
      "loss": 4.8874,
      "mean_token_accuracy": 0.20414694845676423,
      "num_tokens": 160677929.0,
      "step": 73335
    },
    {
      "entropy": 5.694706201553345,
      "epoch": 7.278682016673283,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012876510245115003,
      "loss": 4.9209,
      "mean_token_accuracy": 0.19639366418123244,
      "num_tokens": 160690251.0,
      "step": 73340
    },
    {
      "entropy": 5.747551536560058,
      "epoch": 7.2791782453354505,
      "grad_norm": 1.53125,
      "learning_rate": 0.00012873817662449678,
      "loss": 4.9301,
      "mean_token_accuracy": 0.20552220791578293,
      "num_tokens": 160699692.0,
      "step": 73345
    },
    {
      "entropy": 5.661035442352295,
      "epoch": 7.279674473997618,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012871125442482148,
      "loss": 4.8417,
      "mean_token_accuracy": 0.20668209195137024,
      "num_tokens": 160710371.0,
      "step": 73350
    },
    {
      "entropy": 5.68266191482544,
      "epoch": 7.2801707026597855,
      "grad_norm": 1.5,
      "learning_rate": 0.00012868433585279192,
      "loss": 4.8198,
      "mean_token_accuracy": 0.20887982100248337,
      "num_tokens": 160722069.0,
      "step": 73355
    },
    {
      "entropy": 5.719317245483398,
      "epoch": 7.280666931321953,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001286574209090755,
      "loss": 4.9743,
      "mean_token_accuracy": 0.19956245869398118,
      "num_tokens": 160733994.0,
      "step": 73360
    },
    {
      "entropy": 5.7437427043914795,
      "epoch": 7.2811631599841204,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001286305095943398,
      "loss": 4.9709,
      "mean_token_accuracy": 0.1994128406047821,
      "num_tokens": 160745282.0,
      "step": 73365
    },
    {
      "entropy": 5.680620241165161,
      "epoch": 7.281659388646288,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012860360190925202,
      "loss": 4.8199,
      "mean_token_accuracy": 0.2075809970498085,
      "num_tokens": 160756077.0,
      "step": 73370
    },
    {
      "entropy": 5.686081218719482,
      "epoch": 7.282155617308455,
      "grad_norm": 1.375,
      "learning_rate": 0.00012857669785447945,
      "loss": 4.8368,
      "mean_token_accuracy": 0.20883335918188095,
      "num_tokens": 160767309.0,
      "step": 73375
    },
    {
      "entropy": 5.741469240188598,
      "epoch": 7.282651845970623,
      "grad_norm": 1.3125,
      "learning_rate": 0.00012854979743068924,
      "loss": 4.9494,
      "mean_token_accuracy": 0.20063496828079225,
      "num_tokens": 160778359.0,
      "step": 73380
    },
    {
      "entropy": 5.706388092041015,
      "epoch": 7.283148074632791,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012852290063854852,
      "loss": 4.8962,
      "mean_token_accuracy": 0.20450014919042586,
      "num_tokens": 160789552.0,
      "step": 73385
    },
    {
      "entropy": 5.730311250686645,
      "epoch": 7.283644303294959,
      "grad_norm": 1.3125,
      "learning_rate": 0.00012849600747872415,
      "loss": 4.8656,
      "mean_token_accuracy": 0.20472461581230164,
      "num_tokens": 160800456.0,
      "step": 73390
    },
    {
      "entropy": 5.67419662475586,
      "epoch": 7.284140531957126,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001284691179518832,
      "loss": 4.8393,
      "mean_token_accuracy": 0.20362101793289183,
      "num_tokens": 160811453.0,
      "step": 73395
    },
    {
      "entropy": 5.692587995529175,
      "epoch": 7.284636760619294,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012844223205869237,
      "loss": 4.863,
      "mean_token_accuracy": 0.19950709342956544,
      "num_tokens": 160822326.0,
      "step": 73400
    },
    {
      "entropy": 5.667259216308594,
      "epoch": 7.285132989281461,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012841534979981837,
      "loss": 4.9193,
      "mean_token_accuracy": 0.19951068013906478,
      "num_tokens": 160833833.0,
      "step": 73405
    },
    {
      "entropy": 5.686211395263672,
      "epoch": 7.285629217943629,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001283884711759278,
      "loss": 4.9074,
      "mean_token_accuracy": 0.20076024532318115,
      "num_tokens": 160844275.0,
      "step": 73410
    },
    {
      "entropy": 5.664525270462036,
      "epoch": 7.286125446605796,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012836159618768726,
      "loss": 4.896,
      "mean_token_accuracy": 0.20726372301578522,
      "num_tokens": 160854017.0,
      "step": 73415
    },
    {
      "entropy": 5.695429515838623,
      "epoch": 7.286621675267964,
      "grad_norm": 1.59375,
      "learning_rate": 0.00012833472483576316,
      "loss": 4.9235,
      "mean_token_accuracy": 0.20394441932439805,
      "num_tokens": 160864813.0,
      "step": 73420
    },
    {
      "entropy": 5.799733448028564,
      "epoch": 7.287117903930131,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012830785712082178,
      "loss": 5.0301,
      "mean_token_accuracy": 0.1909213989973068,
      "num_tokens": 160875847.0,
      "step": 73425
    },
    {
      "entropy": 5.709377956390381,
      "epoch": 7.2876141325922985,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012828099304352953,
      "loss": 4.8919,
      "mean_token_accuracy": 0.20082005858421326,
      "num_tokens": 160886719.0,
      "step": 73430
    },
    {
      "entropy": 5.7324981689453125,
      "epoch": 7.288110361254466,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012825413260455243,
      "loss": 4.8988,
      "mean_token_accuracy": 0.20813983231782912,
      "num_tokens": 160897477.0,
      "step": 73435
    },
    {
      "entropy": 5.65050482749939,
      "epoch": 7.2886065899166335,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012822727580455667,
      "loss": 4.8426,
      "mean_token_accuracy": 0.20389505922794343,
      "num_tokens": 160907608.0,
      "step": 73440
    },
    {
      "entropy": 5.698640966415406,
      "epoch": 7.289102818578801,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012820042264420826,
      "loss": 4.9192,
      "mean_token_accuracy": 0.20173018872737886,
      "num_tokens": 160917810.0,
      "step": 73445
    },
    {
      "entropy": 5.575117588043213,
      "epoch": 7.289599047240968,
      "grad_norm": 1.375,
      "learning_rate": 0.000128173573124173,
      "loss": 4.7974,
      "mean_token_accuracy": 0.2121240586042404,
      "num_tokens": 160927421.0,
      "step": 73450
    },
    {
      "entropy": 5.66083173751831,
      "epoch": 7.290095275903136,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012814672724511668,
      "loss": 4.9199,
      "mean_token_accuracy": 0.19663075506687164,
      "num_tokens": 160938320.0,
      "step": 73455
    },
    {
      "entropy": 5.69661021232605,
      "epoch": 7.290591504565303,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012811988500770517,
      "loss": 4.8691,
      "mean_token_accuracy": 0.21250872761011125,
      "num_tokens": 160949458.0,
      "step": 73460
    },
    {
      "entropy": 5.694062757492065,
      "epoch": 7.291087733227471,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012809304641260395,
      "loss": 4.886,
      "mean_token_accuracy": 0.20687929838895797,
      "num_tokens": 160960876.0,
      "step": 73465
    },
    {
      "entropy": 5.711702728271485,
      "epoch": 7.291583961889639,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012806621146047863,
      "loss": 4.9067,
      "mean_token_accuracy": 0.1986992672085762,
      "num_tokens": 160972051.0,
      "step": 73470
    },
    {
      "entropy": 5.696146535873413,
      "epoch": 7.292080190551807,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012803938015199472,
      "loss": 4.9567,
      "mean_token_accuracy": 0.19506223499774933,
      "num_tokens": 160984069.0,
      "step": 73475
    },
    {
      "entropy": 5.709176015853882,
      "epoch": 7.292576419213974,
      "grad_norm": 1.28125,
      "learning_rate": 0.00012801255248781748,
      "loss": 4.9307,
      "mean_token_accuracy": 0.2062412440776825,
      "num_tokens": 160996705.0,
      "step": 73480
    },
    {
      "entropy": 5.720725393295288,
      "epoch": 7.293072647876142,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012798572846861223,
      "loss": 4.8731,
      "mean_token_accuracy": 0.2038818284869194,
      "num_tokens": 161007362.0,
      "step": 73485
    },
    {
      "entropy": 5.720604372024536,
      "epoch": 7.293568876538309,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012795890809504405,
      "loss": 4.9849,
      "mean_token_accuracy": 0.19447765946388246,
      "num_tokens": 161018575.0,
      "step": 73490
    },
    {
      "entropy": 5.591070652008057,
      "epoch": 7.294065105200477,
      "grad_norm": 1.375,
      "learning_rate": 0.0001279320913677781,
      "loss": 4.7669,
      "mean_token_accuracy": 0.20855408012866974,
      "num_tokens": 161030101.0,
      "step": 73495
    },
    {
      "entropy": 5.699030780792237,
      "epoch": 7.294561333862644,
      "grad_norm": 1.1875,
      "learning_rate": 0.00012790527828747945,
      "loss": 4.9855,
      "mean_token_accuracy": 0.19154290109872818,
      "num_tokens": 161042434.0,
      "step": 73500
    },
    {
      "entropy": 5.659363698959351,
      "epoch": 7.2950575625248115,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012787846885481294,
      "loss": 4.826,
      "mean_token_accuracy": 0.21204114109277725,
      "num_tokens": 161053663.0,
      "step": 73505
    },
    {
      "entropy": 5.662155437469482,
      "epoch": 7.295553791186979,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001278516630704434,
      "loss": 4.7925,
      "mean_token_accuracy": 0.20686438232660292,
      "num_tokens": 161063919.0,
      "step": 73510
    },
    {
      "entropy": 5.652964878082275,
      "epoch": 7.2960500198491465,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001278248609350355,
      "loss": 4.8482,
      "mean_token_accuracy": 0.20509435832500458,
      "num_tokens": 161074772.0,
      "step": 73515
    },
    {
      "entropy": 5.695656776428223,
      "epoch": 7.296546248511314,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012779806244925384,
      "loss": 4.924,
      "mean_token_accuracy": 0.19996620565652848,
      "num_tokens": 161085105.0,
      "step": 73520
    },
    {
      "entropy": 5.654586601257324,
      "epoch": 7.297042477173481,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00012777126761376307,
      "loss": 4.8453,
      "mean_token_accuracy": 0.20249347388744354,
      "num_tokens": 161096125.0,
      "step": 73525
    },
    {
      "entropy": 5.631401681900025,
      "epoch": 7.297538705835649,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00012774447642922767,
      "loss": 4.8645,
      "mean_token_accuracy": 0.19800270199775696,
      "num_tokens": 161105969.0,
      "step": 73530
    },
    {
      "entropy": 5.671236371994018,
      "epoch": 7.298034934497816,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012771768889631197,
      "loss": 4.8454,
      "mean_token_accuracy": 0.20930688083171844,
      "num_tokens": 161116303.0,
      "step": 73535
    },
    {
      "entropy": 5.722092771530152,
      "epoch": 7.298531163159984,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012769090501568017,
      "loss": 4.9459,
      "mean_token_accuracy": 0.19475499242544175,
      "num_tokens": 161127110.0,
      "step": 73540
    },
    {
      "entropy": 5.6594579219818115,
      "epoch": 7.299027391822151,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012766412478799655,
      "loss": 4.8191,
      "mean_token_accuracy": 0.20868289172649385,
      "num_tokens": 161136198.0,
      "step": 73545
    },
    {
      "entropy": 5.678142070770264,
      "epoch": 7.299523620484319,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012763734821392509,
      "loss": 4.8884,
      "mean_token_accuracy": 0.2034597173333168,
      "num_tokens": 161147730.0,
      "step": 73550
    },
    {
      "entropy": 5.696964120864868,
      "epoch": 7.300019849146487,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001276105752941298,
      "loss": 4.8326,
      "mean_token_accuracy": 0.2065974235534668,
      "num_tokens": 161158284.0,
      "step": 73555
    },
    {
      "entropy": 5.658937931060791,
      "epoch": 7.300516077808655,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00012758380602927478,
      "loss": 4.8208,
      "mean_token_accuracy": 0.2044580399990082,
      "num_tokens": 161170091.0,
      "step": 73560
    },
    {
      "entropy": 5.698244667053222,
      "epoch": 7.301012306470822,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001275570404200237,
      "loss": 4.951,
      "mean_token_accuracy": 0.19692469090223313,
      "num_tokens": 161180524.0,
      "step": 73565
    },
    {
      "entropy": 5.644271564483643,
      "epoch": 7.30150853513299,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001275302784670403,
      "loss": 4.8144,
      "mean_token_accuracy": 0.20528100579977035,
      "num_tokens": 161191159.0,
      "step": 73570
    },
    {
      "entropy": 5.643407392501831,
      "epoch": 7.302004763795157,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012750352017098818,
      "loss": 4.8421,
      "mean_token_accuracy": 0.2014424115419388,
      "num_tokens": 161201713.0,
      "step": 73575
    },
    {
      "entropy": 5.673994064331055,
      "epoch": 7.3025009924573245,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012747676553253099,
      "loss": 4.8659,
      "mean_token_accuracy": 0.20438750684261323,
      "num_tokens": 161211566.0,
      "step": 73580
    },
    {
      "entropy": 5.738550806045533,
      "epoch": 7.302997221119492,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012745001455233203,
      "loss": 4.9122,
      "mean_token_accuracy": 0.19778338372707366,
      "num_tokens": 161222952.0,
      "step": 73585
    },
    {
      "entropy": 5.692842674255371,
      "epoch": 7.3034934497816595,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001274232672310548,
      "loss": 4.8776,
      "mean_token_accuracy": 0.2066318839788437,
      "num_tokens": 161235497.0,
      "step": 73590
    },
    {
      "entropy": 5.7566742420196535,
      "epoch": 7.303989678443827,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012739652356936259,
      "loss": 4.9681,
      "mean_token_accuracy": 0.2030475154519081,
      "num_tokens": 161245807.0,
      "step": 73595
    },
    {
      "entropy": 5.7158918380737305,
      "epoch": 7.304485907105994,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012736978356791856,
      "loss": 4.91,
      "mean_token_accuracy": 0.20329502522945403,
      "num_tokens": 161256053.0,
      "step": 73600
    },
    {
      "entropy": 5.734988021850586,
      "epoch": 7.304982135768162,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012734304722738577,
      "loss": 5.0052,
      "mean_token_accuracy": 0.19294268488883973,
      "num_tokens": 161267848.0,
      "step": 73605
    },
    {
      "entropy": 5.682875394821167,
      "epoch": 7.305478364430329,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012731631454842722,
      "loss": 4.9031,
      "mean_token_accuracy": 0.2007539764046669,
      "num_tokens": 161278421.0,
      "step": 73610
    },
    {
      "entropy": 5.7253107070922855,
      "epoch": 7.305974593092497,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012728958553170577,
      "loss": 4.9834,
      "mean_token_accuracy": 0.20012637078762055,
      "num_tokens": 161288927.0,
      "step": 73615
    },
    {
      "entropy": 5.734542465209961,
      "epoch": 7.306470821754664,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001272628601778843,
      "loss": 4.9275,
      "mean_token_accuracy": 0.20119805932044982,
      "num_tokens": 161299462.0,
      "step": 73620
    },
    {
      "entropy": 5.693387603759765,
      "epoch": 7.306967050416832,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012723613848762566,
      "loss": 4.8839,
      "mean_token_accuracy": 0.20902741104364395,
      "num_tokens": 161310355.0,
      "step": 73625
    },
    {
      "entropy": 5.700574016571045,
      "epoch": 7.307463279078999,
      "grad_norm": 1.28125,
      "learning_rate": 0.00012720942046159236,
      "loss": 4.9259,
      "mean_token_accuracy": 0.20292225182056428,
      "num_tokens": 161321193.0,
      "step": 73630
    },
    {
      "entropy": 5.711646223068238,
      "epoch": 7.307959507741167,
      "grad_norm": 1.265625,
      "learning_rate": 0.00012718270610044695,
      "loss": 4.8475,
      "mean_token_accuracy": 0.20284750461578369,
      "num_tokens": 161331723.0,
      "step": 73635
    },
    {
      "entropy": 5.693671369552613,
      "epoch": 7.308455736403335,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012715599540485197,
      "loss": 4.8434,
      "mean_token_accuracy": 0.20580054968595504,
      "num_tokens": 161342245.0,
      "step": 73640
    },
    {
      "entropy": 5.711370897293091,
      "epoch": 7.308951965065503,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001271292883754696,
      "loss": 4.9158,
      "mean_token_accuracy": 0.1990046977996826,
      "num_tokens": 161354140.0,
      "step": 73645
    },
    {
      "entropy": 5.649596166610718,
      "epoch": 7.30944819372767,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012710258501296237,
      "loss": 4.8138,
      "mean_token_accuracy": 0.20943755954504012,
      "num_tokens": 161364500.0,
      "step": 73650
    },
    {
      "entropy": 5.694247484207153,
      "epoch": 7.3099444223898375,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012707588531799224,
      "loss": 4.8655,
      "mean_token_accuracy": 0.20590068846940995,
      "num_tokens": 161374615.0,
      "step": 73655
    },
    {
      "entropy": 5.6596681594848635,
      "epoch": 7.310440651052005,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012704918929122146,
      "loss": 4.8229,
      "mean_token_accuracy": 0.20779408067464827,
      "num_tokens": 161384426.0,
      "step": 73660
    },
    {
      "entropy": 5.71673903465271,
      "epoch": 7.3109368797141725,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00012702249693331204,
      "loss": 4.9478,
      "mean_token_accuracy": 0.20467217415571212,
      "num_tokens": 161396269.0,
      "step": 73665
    },
    {
      "entropy": 5.656273031234742,
      "epoch": 7.31143310837634,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001269958082449257,
      "loss": 4.8079,
      "mean_token_accuracy": 0.22082410901784896,
      "num_tokens": 161407986.0,
      "step": 73670
    },
    {
      "entropy": 5.705681943893433,
      "epoch": 7.3119293370385074,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001269691232267245,
      "loss": 4.9265,
      "mean_token_accuracy": 0.2022475153207779,
      "num_tokens": 161418615.0,
      "step": 73675
    },
    {
      "entropy": 5.706185293197632,
      "epoch": 7.312425565700675,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012694244187937004,
      "loss": 4.8334,
      "mean_token_accuracy": 0.21059926897287368,
      "num_tokens": 161428683.0,
      "step": 73680
    },
    {
      "entropy": 5.641281652450561,
      "epoch": 7.312921794362842,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012691576420352394,
      "loss": 4.8873,
      "mean_token_accuracy": 0.20773392021656037,
      "num_tokens": 161439762.0,
      "step": 73685
    },
    {
      "entropy": 5.681368637084961,
      "epoch": 7.31341802302501,
      "grad_norm": 1.296875,
      "learning_rate": 0.00012688909019984785,
      "loss": 4.8628,
      "mean_token_accuracy": 0.1975758895277977,
      "num_tokens": 161449533.0,
      "step": 73690
    },
    {
      "entropy": 5.655027914047241,
      "epoch": 7.313914251687177,
      "grad_norm": 1.546875,
      "learning_rate": 0.00012686241986900316,
      "loss": 4.8599,
      "mean_token_accuracy": 0.21295953691005706,
      "num_tokens": 161459532.0,
      "step": 73695
    },
    {
      "entropy": 5.644635438919067,
      "epoch": 7.314410480349345,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001268357532116512,
      "loss": 4.9085,
      "mean_token_accuracy": 0.20483076423406602,
      "num_tokens": 161469690.0,
      "step": 73700
    },
    {
      "entropy": 5.6507350444793705,
      "epoch": 7.314906709011512,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012680909022845334,
      "loss": 4.8587,
      "mean_token_accuracy": 0.20752508640289308,
      "num_tokens": 161480642.0,
      "step": 73705
    },
    {
      "entropy": 5.787677478790283,
      "epoch": 7.31540293767368,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012678243092007074,
      "loss": 5.0285,
      "mean_token_accuracy": 0.19113869071006775,
      "num_tokens": 161492090.0,
      "step": 73710
    },
    {
      "entropy": 5.657116794586182,
      "epoch": 7.315899166335847,
      "grad_norm": 1.296875,
      "learning_rate": 0.00012675577528716448,
      "loss": 4.8384,
      "mean_token_accuracy": 0.20701512992382048,
      "num_tokens": 161502795.0,
      "step": 73715
    },
    {
      "entropy": 5.7445580005645756,
      "epoch": 7.316395394998015,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001267291233303955,
      "loss": 4.9381,
      "mean_token_accuracy": 0.1967364504933357,
      "num_tokens": 161513749.0,
      "step": 73720
    },
    {
      "entropy": 5.636493396759033,
      "epoch": 7.316891623660182,
      "grad_norm": 1.25,
      "learning_rate": 0.00012670247505042483,
      "loss": 4.8989,
      "mean_token_accuracy": 0.20771125108003616,
      "num_tokens": 161525208.0,
      "step": 73725
    },
    {
      "entropy": 5.655841541290283,
      "epoch": 7.31738785232235,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001266758304479131,
      "loss": 4.8539,
      "mean_token_accuracy": 0.20512452423572541,
      "num_tokens": 161536113.0,
      "step": 73730
    },
    {
      "entropy": 5.677980518341064,
      "epoch": 7.317884080984518,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012664918952352128,
      "loss": 4.8602,
      "mean_token_accuracy": 0.2062670946121216,
      "num_tokens": 161547342.0,
      "step": 73735
    },
    {
      "entropy": 5.709439086914062,
      "epoch": 7.3183803096466855,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012662255227790985,
      "loss": 4.9379,
      "mean_token_accuracy": 0.19910078942775727,
      "num_tokens": 161558352.0,
      "step": 73740
    },
    {
      "entropy": 5.696070671081543,
      "epoch": 7.318876538308853,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012659591871173949,
      "loss": 4.9224,
      "mean_token_accuracy": 0.2011604681611061,
      "num_tokens": 161569560.0,
      "step": 73745
    },
    {
      "entropy": 5.67803921699524,
      "epoch": 7.3193727669710205,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012656928882567043,
      "loss": 4.8533,
      "mean_token_accuracy": 0.1937115103006363,
      "num_tokens": 161579700.0,
      "step": 73750
    },
    {
      "entropy": 5.705943632125854,
      "epoch": 7.319868995633188,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012654266262036322,
      "loss": 4.8817,
      "mean_token_accuracy": 0.20721233934164046,
      "num_tokens": 161589843.0,
      "step": 73755
    },
    {
      "entropy": 5.686115169525147,
      "epoch": 7.320365224295355,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012651604009647805,
      "loss": 4.8367,
      "mean_token_accuracy": 0.20934000462293625,
      "num_tokens": 161600953.0,
      "step": 73760
    },
    {
      "entropy": 5.663752889633178,
      "epoch": 7.320861452957523,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00012648942125467516,
      "loss": 4.8024,
      "mean_token_accuracy": 0.2029142275452614,
      "num_tokens": 161611099.0,
      "step": 73765
    },
    {
      "entropy": 5.6151604652404785,
      "epoch": 7.32135768161969,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012646280609561467,
      "loss": 4.8286,
      "mean_token_accuracy": 0.20885894745588302,
      "num_tokens": 161621515.0,
      "step": 73770
    },
    {
      "entropy": 5.657513189315796,
      "epoch": 7.321853910281858,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012643619461995644,
      "loss": 4.9284,
      "mean_token_accuracy": 0.20259661227464676,
      "num_tokens": 161633141.0,
      "step": 73775
    },
    {
      "entropy": 5.728988361358643,
      "epoch": 7.322350138944025,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012640958682836044,
      "loss": 4.984,
      "mean_token_accuracy": 0.19227735698223114,
      "num_tokens": 161643816.0,
      "step": 73780
    },
    {
      "entropy": 5.711668920516968,
      "epoch": 7.322846367606193,
      "grad_norm": 1.5625,
      "learning_rate": 0.00012638298272148657,
      "loss": 4.96,
      "mean_token_accuracy": 0.19943347871303557,
      "num_tokens": 161654382.0,
      "step": 73785
    },
    {
      "entropy": 5.702759170532227,
      "epoch": 7.32334259626836,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00012635638229999442,
      "loss": 4.8681,
      "mean_token_accuracy": 0.2099673107266426,
      "num_tokens": 161664415.0,
      "step": 73790
    },
    {
      "entropy": 5.699808359146118,
      "epoch": 7.323838824930528,
      "grad_norm": 1.671875,
      "learning_rate": 0.00012632978556454376,
      "loss": 4.9302,
      "mean_token_accuracy": 0.19661506563425063,
      "num_tokens": 161673965.0,
      "step": 73795
    },
    {
      "entropy": 5.762392187118531,
      "epoch": 7.324335053592695,
      "grad_norm": 1.5,
      "learning_rate": 0.00012630319251579402,
      "loss": 4.9932,
      "mean_token_accuracy": 0.19786154925823213,
      "num_tokens": 161683842.0,
      "step": 73800
    },
    {
      "entropy": 5.7126811981201175,
      "epoch": 7.324831282254863,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012627660315440477,
      "loss": 4.877,
      "mean_token_accuracy": 0.20906679332256317,
      "num_tokens": 161694620.0,
      "step": 73805
    },
    {
      "entropy": 5.75399808883667,
      "epoch": 7.32532751091703,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012625001748103523,
      "loss": 4.9225,
      "mean_token_accuracy": 0.20598100125789642,
      "num_tokens": 161705249.0,
      "step": 73810
    },
    {
      "entropy": 5.72696008682251,
      "epoch": 7.325823739579198,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012622343549634468,
      "loss": 4.9611,
      "mean_token_accuracy": 0.19764888733625413,
      "num_tokens": 161716022.0,
      "step": 73815
    },
    {
      "entropy": 5.726711320877075,
      "epoch": 7.326319968241366,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012619685720099234,
      "loss": 4.938,
      "mean_token_accuracy": 0.19816607534885405,
      "num_tokens": 161727997.0,
      "step": 73820
    },
    {
      "entropy": 5.7348620891571045,
      "epoch": 7.3268161969035335,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001261702825956374,
      "loss": 4.9239,
      "mean_token_accuracy": 0.20606196522712708,
      "num_tokens": 161740214.0,
      "step": 73825
    },
    {
      "entropy": 5.671978378295899,
      "epoch": 7.327312425565701,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00012614371168093868,
      "loss": 4.8916,
      "mean_token_accuracy": 0.19725461751222612,
      "num_tokens": 161752941.0,
      "step": 73830
    },
    {
      "entropy": 5.635942173004151,
      "epoch": 7.327808654227868,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001261171444575552,
      "loss": 4.8999,
      "mean_token_accuracy": 0.20492536276578904,
      "num_tokens": 161763405.0,
      "step": 73835
    },
    {
      "entropy": 5.649083805084229,
      "epoch": 7.328304882890036,
      "grad_norm": 1.296875,
      "learning_rate": 0.00012609058092614574,
      "loss": 4.8789,
      "mean_token_accuracy": 0.2068229079246521,
      "num_tokens": 161774430.0,
      "step": 73840
    },
    {
      "entropy": 5.692022037506104,
      "epoch": 7.328801111552203,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001260640210873689,
      "loss": 4.8842,
      "mean_token_accuracy": 0.20003947913646697,
      "num_tokens": 161785512.0,
      "step": 73845
    },
    {
      "entropy": 5.721830368041992,
      "epoch": 7.329297340214371,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001260374649418834,
      "loss": 4.9408,
      "mean_token_accuracy": 0.19589996933937073,
      "num_tokens": 161796300.0,
      "step": 73850
    },
    {
      "entropy": 5.696534109115601,
      "epoch": 7.329793568876538,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001260109124903479,
      "loss": 4.922,
      "mean_token_accuracy": 0.19952196031808853,
      "num_tokens": 161807509.0,
      "step": 73855
    },
    {
      "entropy": 5.7489667415618895,
      "epoch": 7.330289797538706,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001259843637334207,
      "loss": 4.9019,
      "mean_token_accuracy": 0.20501121878623962,
      "num_tokens": 161819385.0,
      "step": 73860
    },
    {
      "entropy": 5.739905500411988,
      "epoch": 7.330786026200873,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012595781867176019,
      "loss": 4.8846,
      "mean_token_accuracy": 0.20955090522766112,
      "num_tokens": 161830666.0,
      "step": 73865
    },
    {
      "entropy": 5.719692707061768,
      "epoch": 7.331282254863041,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001259312773060246,
      "loss": 4.8665,
      "mean_token_accuracy": 0.20398680716753007,
      "num_tokens": 161841191.0,
      "step": 73870
    },
    {
      "entropy": 5.672610807418823,
      "epoch": 7.331778483525208,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001259047396368721,
      "loss": 4.8678,
      "mean_token_accuracy": 0.2066289022564888,
      "num_tokens": 161853609.0,
      "step": 73875
    },
    {
      "entropy": 5.681238317489624,
      "epoch": 7.332274712187376,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001258782056649608,
      "loss": 4.8752,
      "mean_token_accuracy": 0.19723902493715287,
      "num_tokens": 161863702.0,
      "step": 73880
    },
    {
      "entropy": 5.733184146881103,
      "epoch": 7.332770940849543,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001258516753909486,
      "loss": 4.9721,
      "mean_token_accuracy": 0.19544972032308577,
      "num_tokens": 161874581.0,
      "step": 73885
    },
    {
      "entropy": 5.658717489242553,
      "epoch": 7.333267169511711,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00012582514881549355,
      "loss": 4.8431,
      "mean_token_accuracy": 0.2033594936132431,
      "num_tokens": 161885439.0,
      "step": 73890
    },
    {
      "entropy": 5.732782888412475,
      "epoch": 7.333763398173878,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012579862593925335,
      "loss": 4.9105,
      "mean_token_accuracy": 0.20388398319482803,
      "num_tokens": 161895787.0,
      "step": 73895
    },
    {
      "entropy": 5.67880802154541,
      "epoch": 7.334259626836046,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012577210676288576,
      "loss": 4.8906,
      "mean_token_accuracy": 0.20520096123218537,
      "num_tokens": 161905507.0,
      "step": 73900
    },
    {
      "entropy": 5.73945164680481,
      "epoch": 7.334755855498214,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012574559128704826,
      "loss": 4.9312,
      "mean_token_accuracy": 0.20022202879190446,
      "num_tokens": 161915574.0,
      "step": 73905
    },
    {
      "entropy": 5.702024841308594,
      "epoch": 7.335252084160381,
      "grad_norm": 1.3125,
      "learning_rate": 0.00012571907951239853,
      "loss": 4.93,
      "mean_token_accuracy": 0.20023505091667176,
      "num_tokens": 161928657.0,
      "step": 73910
    },
    {
      "entropy": 5.700284719467163,
      "epoch": 7.335748312822549,
      "grad_norm": 1.546875,
      "learning_rate": 0.00012569257143959394,
      "loss": 4.8858,
      "mean_token_accuracy": 0.202956885099411,
      "num_tokens": 161938437.0,
      "step": 73915
    },
    {
      "entropy": 5.675711393356323,
      "epoch": 7.336244541484716,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001256660670692919,
      "loss": 4.7946,
      "mean_token_accuracy": 0.21743386834859849,
      "num_tokens": 161947848.0,
      "step": 73920
    },
    {
      "entropy": 5.656577444076538,
      "epoch": 7.336740770146884,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012563956640214964,
      "loss": 4.8773,
      "mean_token_accuracy": 0.20964458733797073,
      "num_tokens": 161959661.0,
      "step": 73925
    },
    {
      "entropy": 5.6496262550354,
      "epoch": 7.337236998809051,
      "grad_norm": 1.5,
      "learning_rate": 0.00012561306943882424,
      "loss": 4.8166,
      "mean_token_accuracy": 0.21034157276153564,
      "num_tokens": 161970542.0,
      "step": 73930
    },
    {
      "entropy": 5.719475030899048,
      "epoch": 7.337733227471219,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012558657617997278,
      "loss": 4.9093,
      "mean_token_accuracy": 0.194912788271904,
      "num_tokens": 161980969.0,
      "step": 73935
    },
    {
      "entropy": 5.681186151504517,
      "epoch": 7.338229456133386,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001255600866262523,
      "loss": 4.9162,
      "mean_token_accuracy": 0.19485195130109786,
      "num_tokens": 161992320.0,
      "step": 73940
    },
    {
      "entropy": 5.713276195526123,
      "epoch": 7.338725684795554,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012553360077831972,
      "loss": 4.8849,
      "mean_token_accuracy": 0.20045779645442963,
      "num_tokens": 162003368.0,
      "step": 73945
    },
    {
      "entropy": 5.752191352844238,
      "epoch": 7.339221913457721,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00012550711863683166,
      "loss": 5.0119,
      "mean_token_accuracy": 0.19733788669109345,
      "num_tokens": 162016125.0,
      "step": 73950
    },
    {
      "entropy": 5.709328079223633,
      "epoch": 7.339718142119889,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012548064020244504,
      "loss": 4.9324,
      "mean_token_accuracy": 0.20427852869033813,
      "num_tokens": 162026467.0,
      "step": 73955
    },
    {
      "entropy": 5.749288940429688,
      "epoch": 7.340214370782056,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012545416547581629,
      "loss": 4.9,
      "mean_token_accuracy": 0.19702374637126924,
      "num_tokens": 162037439.0,
      "step": 73960
    },
    {
      "entropy": 5.74918303489685,
      "epoch": 7.340710599444224,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012542769445760197,
      "loss": 4.9046,
      "mean_token_accuracy": 0.20708560347557067,
      "num_tokens": 162047852.0,
      "step": 73965
    },
    {
      "entropy": 5.654055833816528,
      "epoch": 7.341206828106391,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012540122714845857,
      "loss": 4.8037,
      "mean_token_accuracy": 0.20402350425720214,
      "num_tokens": 162058782.0,
      "step": 73970
    },
    {
      "entropy": 5.693282890319824,
      "epoch": 7.341703056768559,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012537476354904242,
      "loss": 4.9002,
      "mean_token_accuracy": 0.20503951162099837,
      "num_tokens": 162069435.0,
      "step": 73975
    },
    {
      "entropy": 5.708536148071289,
      "epoch": 7.342199285430726,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001253483036600096,
      "loss": 5.0206,
      "mean_token_accuracy": 0.19362933933734894,
      "num_tokens": 162080317.0,
      "step": 73980
    },
    {
      "entropy": 5.7234917163848875,
      "epoch": 7.342695514092894,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012532184748201646,
      "loss": 5.0374,
      "mean_token_accuracy": 0.1977313295006752,
      "num_tokens": 162091589.0,
      "step": 73985
    },
    {
      "entropy": 5.723288249969483,
      "epoch": 7.343191742755062,
      "grad_norm": 1.421875,
      "learning_rate": 0.000125295395015719,
      "loss": 4.8854,
      "mean_token_accuracy": 0.20229700207710266,
      "num_tokens": 162102555.0,
      "step": 73990
    },
    {
      "entropy": 5.672350549697876,
      "epoch": 7.343687971417229,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001252689462617731,
      "loss": 4.7988,
      "mean_token_accuracy": 0.20859242081642151,
      "num_tokens": 162112405.0,
      "step": 73995
    },
    {
      "entropy": 5.7090715885162355,
      "epoch": 7.344184200079397,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012524250122083474,
      "loss": 4.8999,
      "mean_token_accuracy": 0.20164818316698074,
      "num_tokens": 162122925.0,
      "step": 74000
    },
    {
      "entropy": 5.7255815982818605,
      "epoch": 7.344680428741564,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012521605989355966,
      "loss": 4.9297,
      "mean_token_accuracy": 0.20316502898931504,
      "num_tokens": 162133566.0,
      "step": 74005
    },
    {
      "entropy": 5.716795253753662,
      "epoch": 7.345176657403732,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012518962228060343,
      "loss": 4.874,
      "mean_token_accuracy": 0.19559402763843536,
      "num_tokens": 162143906.0,
      "step": 74010
    },
    {
      "entropy": 5.648981809616089,
      "epoch": 7.345672886065899,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001251631883826219,
      "loss": 4.8041,
      "mean_token_accuracy": 0.21393867433071137,
      "num_tokens": 162154364.0,
      "step": 74015
    },
    {
      "entropy": 5.668120527267456,
      "epoch": 7.346169114728067,
      "grad_norm": 1.375,
      "learning_rate": 0.00012513675820027037,
      "loss": 4.7801,
      "mean_token_accuracy": 0.2101615697145462,
      "num_tokens": 162164962.0,
      "step": 74020
    },
    {
      "entropy": 5.664326190948486,
      "epoch": 7.346665343390234,
      "grad_norm": 1.3125,
      "learning_rate": 0.00012511033173420423,
      "loss": 4.8992,
      "mean_token_accuracy": 0.20237720310688018,
      "num_tokens": 162175889.0,
      "step": 74025
    },
    {
      "entropy": 5.707787036895752,
      "epoch": 7.347161572052402,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012508390898507897,
      "loss": 4.9462,
      "mean_token_accuracy": 0.19967334866523742,
      "num_tokens": 162185229.0,
      "step": 74030
    },
    {
      "entropy": 5.6717194557189945,
      "epoch": 7.347657800714569,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012505748995354977,
      "loss": 4.8257,
      "mean_token_accuracy": 0.2079383447766304,
      "num_tokens": 162197025.0,
      "step": 74035
    },
    {
      "entropy": 5.7096508026123045,
      "epoch": 7.348154029376737,
      "grad_norm": 1.3125,
      "learning_rate": 0.00012503107464027167,
      "loss": 4.9249,
      "mean_token_accuracy": 0.19923941195011138,
      "num_tokens": 162208337.0,
      "step": 74040
    },
    {
      "entropy": 5.726832723617553,
      "epoch": 7.348650258038904,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001250046630458997,
      "loss": 4.9588,
      "mean_token_accuracy": 0.19764330834150315,
      "num_tokens": 162219480.0,
      "step": 74045
    },
    {
      "entropy": 5.705218696594239,
      "epoch": 7.349146486701072,
      "grad_norm": 1.375,
      "learning_rate": 0.00012497825517108897,
      "loss": 4.8534,
      "mean_token_accuracy": 0.20538603216409684,
      "num_tokens": 162230021.0,
      "step": 74050
    },
    {
      "entropy": 5.631419372558594,
      "epoch": 7.349642715363239,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001249518510164941,
      "loss": 4.8436,
      "mean_token_accuracy": 0.20499934256076813,
      "num_tokens": 162240538.0,
      "step": 74055
    },
    {
      "entropy": 5.679669713973999,
      "epoch": 7.350138944025407,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012492545058277014,
      "loss": 4.9045,
      "mean_token_accuracy": 0.20038733780384063,
      "num_tokens": 162251276.0,
      "step": 74060
    },
    {
      "entropy": 5.660622215270996,
      "epoch": 7.350635172687574,
      "grad_norm": 1.296875,
      "learning_rate": 0.00012489905387057162,
      "loss": 4.884,
      "mean_token_accuracy": 0.20545292645692825,
      "num_tokens": 162262245.0,
      "step": 74065
    },
    {
      "entropy": 5.690038776397705,
      "epoch": 7.3511314013497415,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001248726608805531,
      "loss": 4.9037,
      "mean_token_accuracy": 0.20444808006286622,
      "num_tokens": 162274871.0,
      "step": 74070
    },
    {
      "entropy": 5.68047981262207,
      "epoch": 7.35162763001191,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012484627161336905,
      "loss": 4.8872,
      "mean_token_accuracy": 0.20528431236743927,
      "num_tokens": 162285532.0,
      "step": 74075
    },
    {
      "entropy": 5.791550731658935,
      "epoch": 7.352123858674077,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012481988606967396,
      "loss": 5.0558,
      "mean_token_accuracy": 0.19495775252580644,
      "num_tokens": 162296766.0,
      "step": 74080
    },
    {
      "entropy": 5.775947046279907,
      "epoch": 7.352620087336245,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012479350425012204,
      "loss": 4.9522,
      "mean_token_accuracy": 0.19678385257720948,
      "num_tokens": 162306290.0,
      "step": 74085
    },
    {
      "entropy": 5.7393162727355955,
      "epoch": 7.353116315998412,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012476712615536759,
      "loss": 4.948,
      "mean_token_accuracy": 0.20212536454200744,
      "num_tokens": 162318228.0,
      "step": 74090
    },
    {
      "entropy": 5.677584934234619,
      "epoch": 7.35361254466058,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012474075178606472,
      "loss": 4.8697,
      "mean_token_accuracy": 0.20250336527824403,
      "num_tokens": 162328941.0,
      "step": 74095
    },
    {
      "entropy": 5.701125383377075,
      "epoch": 7.354108773322747,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012471438114286737,
      "loss": 4.8811,
      "mean_token_accuracy": 0.20813553184270858,
      "num_tokens": 162339353.0,
      "step": 74100
    },
    {
      "entropy": 5.652126216888428,
      "epoch": 7.354605001984915,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001246880142264296,
      "loss": 4.8703,
      "mean_token_accuracy": 0.20341373085975648,
      "num_tokens": 162349124.0,
      "step": 74105
    },
    {
      "entropy": 5.654602956771851,
      "epoch": 7.355101230647082,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001246616510374051,
      "loss": 4.8753,
      "mean_token_accuracy": 0.2038543328642845,
      "num_tokens": 162359679.0,
      "step": 74110
    },
    {
      "entropy": 5.708692455291748,
      "epoch": 7.35559745930925,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012463529157644767,
      "loss": 4.951,
      "mean_token_accuracy": 0.2004231557250023,
      "num_tokens": 162369740.0,
      "step": 74115
    },
    {
      "entropy": 5.707001113891602,
      "epoch": 7.356093687971417,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012460893584421108,
      "loss": 4.9192,
      "mean_token_accuracy": 0.2033076748251915,
      "num_tokens": 162380706.0,
      "step": 74120
    },
    {
      "entropy": 5.707382488250732,
      "epoch": 7.356589916633585,
      "grad_norm": 1.546875,
      "learning_rate": 0.00012458258384134884,
      "loss": 4.899,
      "mean_token_accuracy": 0.20048684179782866,
      "num_tokens": 162391049.0,
      "step": 74125
    },
    {
      "entropy": 5.639961767196655,
      "epoch": 7.357086145295752,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012455623556851438,
      "loss": 4.8288,
      "mean_token_accuracy": 0.21470629274845124,
      "num_tokens": 162401550.0,
      "step": 74130
    },
    {
      "entropy": 5.777278470993042,
      "epoch": 7.35758237395792,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001245298910263611,
      "loss": 4.9899,
      "mean_token_accuracy": 0.19862060695886613,
      "num_tokens": 162411634.0,
      "step": 74135
    },
    {
      "entropy": 5.796222448348999,
      "epoch": 7.358078602620087,
      "grad_norm": 1.28125,
      "learning_rate": 0.00012450355021554222,
      "loss": 5.0195,
      "mean_token_accuracy": 0.1930161014199257,
      "num_tokens": 162423015.0,
      "step": 74140
    },
    {
      "entropy": 5.669342136383056,
      "epoch": 7.3585748312822545,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000124477213136711,
      "loss": 4.9167,
      "mean_token_accuracy": 0.20112192779779434,
      "num_tokens": 162433451.0,
      "step": 74145
    },
    {
      "entropy": 5.751183748245239,
      "epoch": 7.359071059944422,
      "grad_norm": 1.296875,
      "learning_rate": 0.00012445087979052062,
      "loss": 4.9966,
      "mean_token_accuracy": 0.19434526562690735,
      "num_tokens": 162446212.0,
      "step": 74150
    },
    {
      "entropy": 5.7610420227050785,
      "epoch": 7.3595672886065895,
      "grad_norm": 1.4375,
      "learning_rate": 0.000124424550177624,
      "loss": 4.9266,
      "mean_token_accuracy": 0.20265356302261353,
      "num_tokens": 162456892.0,
      "step": 74155
    },
    {
      "entropy": 5.752715063095093,
      "epoch": 7.360063517268758,
      "grad_norm": 1.5,
      "learning_rate": 0.0001243982242986741,
      "loss": 4.968,
      "mean_token_accuracy": 0.1944948613643646,
      "num_tokens": 162468274.0,
      "step": 74160
    },
    {
      "entropy": 5.711588430404663,
      "epoch": 7.360559745930925,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012437190215432373,
      "loss": 4.9425,
      "mean_token_accuracy": 0.20280879437923433,
      "num_tokens": 162478956.0,
      "step": 74165
    },
    {
      "entropy": 5.685707044601441,
      "epoch": 7.361055974593093,
      "grad_norm": 1.375,
      "learning_rate": 0.00012434558374522554,
      "loss": 4.8145,
      "mean_token_accuracy": 0.2107461616396904,
      "num_tokens": 162490272.0,
      "step": 74170
    },
    {
      "entropy": 5.623905086517334,
      "epoch": 7.36155220325526,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00012431926907203227,
      "loss": 4.7342,
      "mean_token_accuracy": 0.20961397886276245,
      "num_tokens": 162500534.0,
      "step": 74175
    },
    {
      "entropy": 5.674023818969727,
      "epoch": 7.362048431917428,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001242929581353964,
      "loss": 4.903,
      "mean_token_accuracy": 0.19982665479183198,
      "num_tokens": 162510896.0,
      "step": 74180
    },
    {
      "entropy": 5.7236682891845705,
      "epoch": 7.362544660579595,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012426665093597048,
      "loss": 5.0033,
      "mean_token_accuracy": 0.1936863213777542,
      "num_tokens": 162521988.0,
      "step": 74185
    },
    {
      "entropy": 5.752522659301758,
      "epoch": 7.363040889241763,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012424034747440685,
      "loss": 5.0018,
      "mean_token_accuracy": 0.19263043701648713,
      "num_tokens": 162533142.0,
      "step": 74190
    },
    {
      "entropy": 5.644054651260376,
      "epoch": 7.36353711790393,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001242140477513577,
      "loss": 4.8743,
      "mean_token_accuracy": 0.19786109030246735,
      "num_tokens": 162544984.0,
      "step": 74195
    },
    {
      "entropy": 5.719528675079346,
      "epoch": 7.364033346566098,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001241877517674752,
      "loss": 4.9171,
      "mean_token_accuracy": 0.2002541869878769,
      "num_tokens": 162555841.0,
      "step": 74200
    },
    {
      "entropy": 5.682117128372193,
      "epoch": 7.364529575228265,
      "grad_norm": 1.5,
      "learning_rate": 0.00012416145952341156,
      "loss": 4.9311,
      "mean_token_accuracy": 0.19865995049476623,
      "num_tokens": 162566423.0,
      "step": 74205
    },
    {
      "entropy": 5.745235109329224,
      "epoch": 7.365025803890433,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001241351710198186,
      "loss": 5.0154,
      "mean_token_accuracy": 0.19478826820850373,
      "num_tokens": 162578576.0,
      "step": 74210
    },
    {
      "entropy": 5.745913124084472,
      "epoch": 7.3655220325526,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001241088862573484,
      "loss": 4.9739,
      "mean_token_accuracy": 0.19701779633760452,
      "num_tokens": 162590314.0,
      "step": 74215
    },
    {
      "entropy": 5.67219090461731,
      "epoch": 7.3660182612147675,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001240826052366527,
      "loss": 4.8542,
      "mean_token_accuracy": 0.20474592447280884,
      "num_tokens": 162600478.0,
      "step": 74220
    },
    {
      "entropy": 5.745263481140137,
      "epoch": 7.366514489876935,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012405632795838318,
      "loss": 5.0014,
      "mean_token_accuracy": 0.19817142933607101,
      "num_tokens": 162612428.0,
      "step": 74225
    },
    {
      "entropy": 5.7108461380004885,
      "epoch": 7.3670107185391025,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001240300544231914,
      "loss": 4.9147,
      "mean_token_accuracy": 0.19479650408029556,
      "num_tokens": 162624558.0,
      "step": 74230
    },
    {
      "entropy": 5.693995189666748,
      "epoch": 7.36750694720127,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012400378463172903,
      "loss": 4.8667,
      "mean_token_accuracy": 0.2035731628537178,
      "num_tokens": 162635438.0,
      "step": 74235
    },
    {
      "entropy": 5.678580570220947,
      "epoch": 7.3680031758634374,
      "grad_norm": 1.546875,
      "learning_rate": 0.00012397751858464741,
      "loss": 4.8536,
      "mean_token_accuracy": 0.20780596733093262,
      "num_tokens": 162646367.0,
      "step": 74240
    },
    {
      "entropy": 5.700334644317627,
      "epoch": 7.368499404525606,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012395125628259787,
      "loss": 4.8661,
      "mean_token_accuracy": 0.2072540521621704,
      "num_tokens": 162657281.0,
      "step": 74245
    },
    {
      "entropy": 5.6139825820922855,
      "epoch": 7.368995633187773,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001239249977262317,
      "loss": 4.771,
      "mean_token_accuracy": 0.21074958443641661,
      "num_tokens": 162667843.0,
      "step": 74250
    },
    {
      "entropy": 5.673219347000122,
      "epoch": 7.369491861849941,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012389874291620013,
      "loss": 4.8291,
      "mean_token_accuracy": 0.20364999920129775,
      "num_tokens": 162678532.0,
      "step": 74255
    },
    {
      "entropy": 5.676128149032593,
      "epoch": 7.369988090512108,
      "grad_norm": 1.34375,
      "learning_rate": 0.000123872491853154,
      "loss": 4.9091,
      "mean_token_accuracy": 0.2022777169942856,
      "num_tokens": 162690289.0,
      "step": 74260
    },
    {
      "entropy": 5.708182764053345,
      "epoch": 7.370484319174276,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012384624453774457,
      "loss": 4.8575,
      "mean_token_accuracy": 0.2102560356259346,
      "num_tokens": 162700342.0,
      "step": 74265
    },
    {
      "entropy": 5.651893949508667,
      "epoch": 7.370980547836443,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012382000097062253,
      "loss": 4.9254,
      "mean_token_accuracy": 0.2058022454380989,
      "num_tokens": 162713011.0,
      "step": 74270
    },
    {
      "entropy": 5.67184042930603,
      "epoch": 7.371476776498611,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001237937611524386,
      "loss": 4.8441,
      "mean_token_accuracy": 0.2124718979001045,
      "num_tokens": 162723499.0,
      "step": 74275
    },
    {
      "entropy": 5.695851850509643,
      "epoch": 7.371973005160778,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001237675250838437,
      "loss": 4.9103,
      "mean_token_accuracy": 0.2036876603960991,
      "num_tokens": 162734292.0,
      "step": 74280
    },
    {
      "entropy": 5.684814596176148,
      "epoch": 7.372469233822946,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00012374129276548822,
      "loss": 4.8504,
      "mean_token_accuracy": 0.20844196677207946,
      "num_tokens": 162744552.0,
      "step": 74285
    },
    {
      "entropy": 5.726684093475342,
      "epoch": 7.372965462485113,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012371506419802276,
      "loss": 5.0069,
      "mean_token_accuracy": 0.19387321472167968,
      "num_tokens": 162757296.0,
      "step": 74290
    },
    {
      "entropy": 5.632974481582641,
      "epoch": 7.373461691147281,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012368883938209774,
      "loss": 4.8132,
      "mean_token_accuracy": 0.20905755311250687,
      "num_tokens": 162768155.0,
      "step": 74295
    },
    {
      "entropy": 5.692772388458252,
      "epoch": 7.373957919809448,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012366261831836347,
      "loss": 4.9236,
      "mean_token_accuracy": 0.19715267270803452,
      "num_tokens": 162779560.0,
      "step": 74300
    },
    {
      "entropy": 5.681787919998169,
      "epoch": 7.3744541484716155,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012363640100747012,
      "loss": 4.8735,
      "mean_token_accuracy": 0.2066667065024376,
      "num_tokens": 162791297.0,
      "step": 74305
    },
    {
      "entropy": 5.661904335021973,
      "epoch": 7.374950377133783,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001236101874500679,
      "loss": 4.882,
      "mean_token_accuracy": 0.2025880753993988,
      "num_tokens": 162802728.0,
      "step": 74310
    },
    {
      "entropy": 5.729058408737183,
      "epoch": 7.3754466057959505,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012358397764680683,
      "loss": 4.9273,
      "mean_token_accuracy": 0.20183738619089125,
      "num_tokens": 162813905.0,
      "step": 74315
    },
    {
      "entropy": 5.677016687393189,
      "epoch": 7.375942834458118,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012355777159833676,
      "loss": 4.8955,
      "mean_token_accuracy": 0.20033892542123793,
      "num_tokens": 162824096.0,
      "step": 74320
    },
    {
      "entropy": 5.699760961532593,
      "epoch": 7.376439063120285,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012353156930530767,
      "loss": 4.9129,
      "mean_token_accuracy": 0.1984574541449547,
      "num_tokens": 162834500.0,
      "step": 74325
    },
    {
      "entropy": 5.662281847000122,
      "epoch": 7.376935291782454,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001235053707683693,
      "loss": 4.8697,
      "mean_token_accuracy": 0.21760421991348267,
      "num_tokens": 162845163.0,
      "step": 74330
    },
    {
      "entropy": 5.722398471832276,
      "epoch": 7.377431520444621,
      "grad_norm": 1.578125,
      "learning_rate": 0.00012347917598817134,
      "loss": 4.9025,
      "mean_token_accuracy": 0.19965894520282745,
      "num_tokens": 162855981.0,
      "step": 74335
    },
    {
      "entropy": 5.759460496902466,
      "epoch": 7.377927749106789,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012345298496536317,
      "loss": 4.9873,
      "mean_token_accuracy": 0.19342369437217713,
      "num_tokens": 162867850.0,
      "step": 74340
    },
    {
      "entropy": 5.666509771347046,
      "epoch": 7.378423977768956,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012342679770059454,
      "loss": 4.8473,
      "mean_token_accuracy": 0.21363096237182616,
      "num_tokens": 162879050.0,
      "step": 74345
    },
    {
      "entropy": 5.754013442993164,
      "epoch": 7.378920206431124,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012340061419451466,
      "loss": 4.94,
      "mean_token_accuracy": 0.19107670187950135,
      "num_tokens": 162890243.0,
      "step": 74350
    },
    {
      "entropy": 5.689929580688476,
      "epoch": 7.379416435093291,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001233744344477729,
      "loss": 4.8806,
      "mean_token_accuracy": 0.20419200509786606,
      "num_tokens": 162899786.0,
      "step": 74355
    },
    {
      "entropy": 5.715045404434204,
      "epoch": 7.379912663755459,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012334825846101847,
      "loss": 4.8832,
      "mean_token_accuracy": 0.21395473778247834,
      "num_tokens": 162911287.0,
      "step": 74360
    },
    {
      "entropy": 5.695003461837769,
      "epoch": 7.380408892417626,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012332208623490044,
      "loss": 4.9488,
      "mean_token_accuracy": 0.19877550899982452,
      "num_tokens": 162922507.0,
      "step": 74365
    },
    {
      "entropy": 5.734168338775635,
      "epoch": 7.380905121079794,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012329591777006775,
      "loss": 4.9256,
      "mean_token_accuracy": 0.19921404868364334,
      "num_tokens": 162932895.0,
      "step": 74370
    },
    {
      "entropy": 5.702882432937622,
      "epoch": 7.381401349741961,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001232697530671695,
      "loss": 4.9051,
      "mean_token_accuracy": 0.20354705154895783,
      "num_tokens": 162943433.0,
      "step": 74375
    },
    {
      "entropy": 5.693787384033203,
      "epoch": 7.3818975784041285,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012324359212685434,
      "loss": 4.8685,
      "mean_token_accuracy": 0.20470433533191681,
      "num_tokens": 162953185.0,
      "step": 74380
    },
    {
      "entropy": 5.78622875213623,
      "epoch": 7.382393807066296,
      "grad_norm": 1.265625,
      "learning_rate": 0.00012321743494977117,
      "loss": 5.0536,
      "mean_token_accuracy": 0.18802090287208556,
      "num_tokens": 162964784.0,
      "step": 74385
    },
    {
      "entropy": 5.741708660125733,
      "epoch": 7.3828900357284635,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012319128153656856,
      "loss": 4.8698,
      "mean_token_accuracy": 0.20602494329214097,
      "num_tokens": 162974836.0,
      "step": 74390
    },
    {
      "entropy": 5.677759313583374,
      "epoch": 7.383386264390631,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012316513188789503,
      "loss": 4.8635,
      "mean_token_accuracy": 0.2054303616285324,
      "num_tokens": 162985225.0,
      "step": 74395
    },
    {
      "entropy": 5.683509349822998,
      "epoch": 7.383882493052798,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012313898600439907,
      "loss": 4.886,
      "mean_token_accuracy": 0.20570487678050994,
      "num_tokens": 162995828.0,
      "step": 74400
    },
    {
      "entropy": 5.656861448287964,
      "epoch": 7.384378721714966,
      "grad_norm": 1.53125,
      "learning_rate": 0.00012311284388672893,
      "loss": 4.9113,
      "mean_token_accuracy": 0.2048454612493515,
      "num_tokens": 163005273.0,
      "step": 74405
    },
    {
      "entropy": 5.678094816207886,
      "epoch": 7.384874950377133,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000123086705535533,
      "loss": 4.8989,
      "mean_token_accuracy": 0.20475785434246063,
      "num_tokens": 163014464.0,
      "step": 74410
    },
    {
      "entropy": 5.708816337585449,
      "epoch": 7.385371179039302,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012306057095145948,
      "loss": 4.979,
      "mean_token_accuracy": 0.20009672045707702,
      "num_tokens": 163025824.0,
      "step": 74415
    },
    {
      "entropy": 5.7292468547821045,
      "epoch": 7.385867407701469,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012303444013515643,
      "loss": 4.9412,
      "mean_token_accuracy": 0.1973734274506569,
      "num_tokens": 163037188.0,
      "step": 74420
    },
    {
      "entropy": 5.715413236618042,
      "epoch": 7.386363636363637,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00012300831308727176,
      "loss": 4.8795,
      "mean_token_accuracy": 0.2053824856877327,
      "num_tokens": 163048637.0,
      "step": 74425
    },
    {
      "entropy": 5.731279945373535,
      "epoch": 7.386859865025804,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012298218980845343,
      "loss": 4.9661,
      "mean_token_accuracy": 0.1959625780582428,
      "num_tokens": 163061006.0,
      "step": 74430
    },
    {
      "entropy": 5.727541828155518,
      "epoch": 7.387356093687972,
      "grad_norm": 1.375,
      "learning_rate": 0.00012295607029934918,
      "loss": 5.0143,
      "mean_token_accuracy": 0.19211636930704118,
      "num_tokens": 163071492.0,
      "step": 74435
    },
    {
      "entropy": 5.679787731170654,
      "epoch": 7.387852322350139,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012292995456060675,
      "loss": 4.8616,
      "mean_token_accuracy": 0.20793869644403457,
      "num_tokens": 163081161.0,
      "step": 74440
    },
    {
      "entropy": 5.632496070861817,
      "epoch": 7.388348551012307,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001229038425928738,
      "loss": 4.8225,
      "mean_token_accuracy": 0.2091659352183342,
      "num_tokens": 163091859.0,
      "step": 74445
    },
    {
      "entropy": 5.773268413543701,
      "epoch": 7.388844779674474,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012287773439679787,
      "loss": 4.9898,
      "mean_token_accuracy": 0.19309359341859816,
      "num_tokens": 163102806.0,
      "step": 74450
    },
    {
      "entropy": 5.722415924072266,
      "epoch": 7.3893410083366415,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001228516299730263,
      "loss": 4.8633,
      "mean_token_accuracy": 0.20344736576080322,
      "num_tokens": 163114895.0,
      "step": 74455
    },
    {
      "entropy": 5.703191137313842,
      "epoch": 7.389837236998809,
      "grad_norm": 1.5,
      "learning_rate": 0.00012282552932220647,
      "loss": 4.8713,
      "mean_token_accuracy": 0.2008776471018791,
      "num_tokens": 163126399.0,
      "step": 74460
    },
    {
      "entropy": 5.691799116134644,
      "epoch": 7.3903334656609765,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012279943244498552,
      "loss": 4.8769,
      "mean_token_accuracy": 0.20636581629514694,
      "num_tokens": 163136964.0,
      "step": 74465
    },
    {
      "entropy": 5.678147888183593,
      "epoch": 7.390829694323144,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012277333934201075,
      "loss": 4.9229,
      "mean_token_accuracy": 0.20216409862041473,
      "num_tokens": 163148008.0,
      "step": 74470
    },
    {
      "entropy": 5.7408020973205565,
      "epoch": 7.391325922985311,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0001227472500139291,
      "loss": 4.9149,
      "mean_token_accuracy": 0.20435035228729248,
      "num_tokens": 163158966.0,
      "step": 74475
    },
    {
      "entropy": 5.679939079284668,
      "epoch": 7.391822151647479,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012272116446138763,
      "loss": 4.8911,
      "mean_token_accuracy": 0.20767390578985215,
      "num_tokens": 163169849.0,
      "step": 74480
    },
    {
      "entropy": 5.716514730453492,
      "epoch": 7.392318380309646,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012269508268503317,
      "loss": 4.9294,
      "mean_token_accuracy": 0.20517384260892868,
      "num_tokens": 163181502.0,
      "step": 74485
    },
    {
      "entropy": 5.628296184539795,
      "epoch": 7.392814608971814,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012266900468551242,
      "loss": 4.8609,
      "mean_token_accuracy": 0.20272858440876007,
      "num_tokens": 163192453.0,
      "step": 74490
    },
    {
      "entropy": 5.629220581054687,
      "epoch": 7.393310837633981,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012264293046347207,
      "loss": 4.8345,
      "mean_token_accuracy": 0.20735965967178344,
      "num_tokens": 163204008.0,
      "step": 74495
    },
    {
      "entropy": 5.747495889663696,
      "epoch": 7.39380706629615,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001226168600195588,
      "loss": 4.9471,
      "mean_token_accuracy": 0.20316196382045745,
      "num_tokens": 163214836.0,
      "step": 74500
    },
    {
      "entropy": 5.677853012084961,
      "epoch": 7.394303294958317,
      "grad_norm": 1.328125,
      "learning_rate": 0.000122590793354419,
      "loss": 4.852,
      "mean_token_accuracy": 0.20667935758829117,
      "num_tokens": 163225631.0,
      "step": 74505
    },
    {
      "entropy": 5.709114217758179,
      "epoch": 7.394799523620485,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012256473046869917,
      "loss": 4.9369,
      "mean_token_accuracy": 0.1999986931681633,
      "num_tokens": 163237872.0,
      "step": 74510
    },
    {
      "entropy": 5.654624652862549,
      "epoch": 7.395295752282652,
      "grad_norm": 1.25,
      "learning_rate": 0.00012253867136304556,
      "loss": 4.8419,
      "mean_token_accuracy": 0.2084093362092972,
      "num_tokens": 163248319.0,
      "step": 74515
    },
    {
      "entropy": 5.776406240463257,
      "epoch": 7.39579198094482,
      "grad_norm": 1.5,
      "learning_rate": 0.0001225126160381044,
      "loss": 4.9714,
      "mean_token_accuracy": 0.2014584481716156,
      "num_tokens": 163259598.0,
      "step": 74520
    },
    {
      "entropy": 5.672357130050659,
      "epoch": 7.396288209606987,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012248656449452165,
      "loss": 4.8463,
      "mean_token_accuracy": 0.21425683796405792,
      "num_tokens": 163270545.0,
      "step": 74525
    },
    {
      "entropy": 5.705460500717163,
      "epoch": 7.3967844382691545,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001224605167329436,
      "loss": 4.9153,
      "mean_token_accuracy": 0.19959664940834046,
      "num_tokens": 163281743.0,
      "step": 74530
    },
    {
      "entropy": 5.6372771739959715,
      "epoch": 7.397280666931322,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001224344727540159,
      "loss": 4.7959,
      "mean_token_accuracy": 0.2098942682147026,
      "num_tokens": 163293487.0,
      "step": 74535
    },
    {
      "entropy": 5.736130809783935,
      "epoch": 7.3977768955934895,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012240843255838463,
      "loss": 4.9091,
      "mean_token_accuracy": 0.1971760004758835,
      "num_tokens": 163303015.0,
      "step": 74540
    },
    {
      "entropy": 5.63513298034668,
      "epoch": 7.398273124255657,
      "grad_norm": 1.21875,
      "learning_rate": 0.00012238239614669545,
      "loss": 4.8565,
      "mean_token_accuracy": 0.20738495886325836,
      "num_tokens": 163314741.0,
      "step": 74545
    },
    {
      "entropy": 5.709094762802124,
      "epoch": 7.3987693529178244,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012235636351959396,
      "loss": 4.9106,
      "mean_token_accuracy": 0.198796047270298,
      "num_tokens": 163325001.0,
      "step": 74550
    },
    {
      "entropy": 5.659395360946656,
      "epoch": 7.399265581579992,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012233033467772567,
      "loss": 4.783,
      "mean_token_accuracy": 0.21008430123329164,
      "num_tokens": 163336275.0,
      "step": 74555
    },
    {
      "entropy": 5.656813764572144,
      "epoch": 7.399761810242159,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012230430962173615,
      "loss": 4.8776,
      "mean_token_accuracy": 0.20453888326883315,
      "num_tokens": 163347608.0,
      "step": 74560
    },
    {
      "entropy": 5.705151796340942,
      "epoch": 7.400258038904327,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001222782883522708,
      "loss": 4.9175,
      "mean_token_accuracy": 0.19903444051742553,
      "num_tokens": 163357799.0,
      "step": 74565
    },
    {
      "entropy": 5.763292646408081,
      "epoch": 7.400754267566494,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012225227086997466,
      "loss": 4.9802,
      "mean_token_accuracy": 0.19799690693616867,
      "num_tokens": 163367874.0,
      "step": 74570
    },
    {
      "entropy": 5.7363323211669925,
      "epoch": 7.401250496228662,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001222262571754932,
      "loss": 4.949,
      "mean_token_accuracy": 0.19784080237150192,
      "num_tokens": 163379592.0,
      "step": 74575
    },
    {
      "entropy": 5.754717302322388,
      "epoch": 7.401746724890829,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001222002472694713,
      "loss": 4.9426,
      "mean_token_accuracy": 0.1958034947514534,
      "num_tokens": 163390706.0,
      "step": 74580
    },
    {
      "entropy": 5.768659400939941,
      "epoch": 7.402242953552998,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012217424115255397,
      "loss": 4.9363,
      "mean_token_accuracy": 0.20200690180063247,
      "num_tokens": 163401072.0,
      "step": 74585
    },
    {
      "entropy": 5.7201845169067385,
      "epoch": 7.402739182215165,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001221482388253862,
      "loss": 4.8586,
      "mean_token_accuracy": 0.2017806053161621,
      "num_tokens": 163410724.0,
      "step": 74590
    },
    {
      "entropy": 5.703323650360107,
      "epoch": 7.403235410877333,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012212224028861277,
      "loss": 4.873,
      "mean_token_accuracy": 0.20460181087255477,
      "num_tokens": 163421310.0,
      "step": 74595
    },
    {
      "entropy": 5.703702449798584,
      "epoch": 7.4037316395395,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012209624554287827,
      "loss": 4.9178,
      "mean_token_accuracy": 0.19477801471948625,
      "num_tokens": 163432211.0,
      "step": 74600
    },
    {
      "entropy": 5.670903253555298,
      "epoch": 7.404227868201668,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012207025458882748,
      "loss": 4.9053,
      "mean_token_accuracy": 0.1989656776189804,
      "num_tokens": 163443540.0,
      "step": 74605
    },
    {
      "entropy": 5.726875257492066,
      "epoch": 7.404724096863835,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012204426742710485,
      "loss": 4.9016,
      "mean_token_accuracy": 0.19480984807014465,
      "num_tokens": 163454312.0,
      "step": 74610
    },
    {
      "entropy": 5.65447964668274,
      "epoch": 7.4052203255260025,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012201828405835473,
      "loss": 4.8816,
      "mean_token_accuracy": 0.1985320895910263,
      "num_tokens": 163465223.0,
      "step": 74615
    },
    {
      "entropy": 5.731775760650635,
      "epoch": 7.40571655418817,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012199230448322156,
      "loss": 4.8671,
      "mean_token_accuracy": 0.2130781665444374,
      "num_tokens": 163476322.0,
      "step": 74620
    },
    {
      "entropy": 5.65556321144104,
      "epoch": 7.4062127828503375,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012196632870234956,
      "loss": 4.8308,
      "mean_token_accuracy": 0.2077892005443573,
      "num_tokens": 163486425.0,
      "step": 74625
    },
    {
      "entropy": 5.631169843673706,
      "epoch": 7.406709011512505,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012194035671638283,
      "loss": 4.7966,
      "mean_token_accuracy": 0.20976672023534776,
      "num_tokens": 163497227.0,
      "step": 74630
    },
    {
      "entropy": 5.673315525054932,
      "epoch": 7.407205240174672,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012191438852596541,
      "loss": 4.9961,
      "mean_token_accuracy": 0.1952238917350769,
      "num_tokens": 163508549.0,
      "step": 74635
    },
    {
      "entropy": 5.669665861129761,
      "epoch": 7.40770146883684,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001218884241317413,
      "loss": 4.8828,
      "mean_token_accuracy": 0.2083701640367508,
      "num_tokens": 163520362.0,
      "step": 74640
    },
    {
      "entropy": 5.728292512893677,
      "epoch": 7.408197697499007,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012186246353435431,
      "loss": 4.955,
      "mean_token_accuracy": 0.1967203676700592,
      "num_tokens": 163531552.0,
      "step": 74645
    },
    {
      "entropy": 5.774665260314942,
      "epoch": 7.408693926161175,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012183650673444832,
      "loss": 4.9764,
      "mean_token_accuracy": 0.19403988867998123,
      "num_tokens": 163542448.0,
      "step": 74650
    },
    {
      "entropy": 5.68231315612793,
      "epoch": 7.409190154823342,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012181055373266688,
      "loss": 4.8657,
      "mean_token_accuracy": 0.2019910305738449,
      "num_tokens": 163555712.0,
      "step": 74655
    },
    {
      "entropy": 5.711365461349487,
      "epoch": 7.40968638348551,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012178460452965366,
      "loss": 4.9079,
      "mean_token_accuracy": 0.2000589907169342,
      "num_tokens": 163566636.0,
      "step": 74660
    },
    {
      "entropy": 5.678678417205811,
      "epoch": 7.410182612147677,
      "grad_norm": 1.59375,
      "learning_rate": 0.00012175865912605202,
      "loss": 4.8407,
      "mean_token_accuracy": 0.21165382713079453,
      "num_tokens": 163577525.0,
      "step": 74665
    },
    {
      "entropy": 5.6690693378448485,
      "epoch": 7.410678840809846,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00012173271752250542,
      "loss": 4.8794,
      "mean_token_accuracy": 0.20712648332118988,
      "num_tokens": 163588352.0,
      "step": 74670
    },
    {
      "entropy": 5.705988216400146,
      "epoch": 7.411175069472013,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012170677971965724,
      "loss": 4.9042,
      "mean_token_accuracy": 0.20501474738121034,
      "num_tokens": 163599796.0,
      "step": 74675
    },
    {
      "entropy": 5.678597736358642,
      "epoch": 7.411671298134181,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012168084571815058,
      "loss": 4.8741,
      "mean_token_accuracy": 0.1982881397008896,
      "num_tokens": 163610892.0,
      "step": 74680
    },
    {
      "entropy": 5.693036603927612,
      "epoch": 7.412167526796348,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00012165491551862861,
      "loss": 4.9331,
      "mean_token_accuracy": 0.19877074360847474,
      "num_tokens": 163621131.0,
      "step": 74685
    },
    {
      "entropy": 5.693021154403686,
      "epoch": 7.4126637554585155,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012162898912173427,
      "loss": 4.8667,
      "mean_token_accuracy": 0.20648714900016785,
      "num_tokens": 163631945.0,
      "step": 74690
    },
    {
      "entropy": 5.641934013366699,
      "epoch": 7.413159984120683,
      "grad_norm": 1.265625,
      "learning_rate": 0.00012160306652811048,
      "loss": 4.8514,
      "mean_token_accuracy": 0.20653954148292542,
      "num_tokens": 163643119.0,
      "step": 74695
    },
    {
      "entropy": 5.65167818069458,
      "epoch": 7.4136562127828505,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012157714773840013,
      "loss": 4.8462,
      "mean_token_accuracy": 0.2009387269616127,
      "num_tokens": 163653452.0,
      "step": 74700
    },
    {
      "entropy": 5.631876850128174,
      "epoch": 7.414152441445018,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012155123275324588,
      "loss": 4.8233,
      "mean_token_accuracy": 0.2055809512734413,
      "num_tokens": 163663562.0,
      "step": 74705
    },
    {
      "entropy": 5.737194347381592,
      "epoch": 7.414648670107185,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012152532157329046,
      "loss": 4.9582,
      "mean_token_accuracy": 0.19842270016670227,
      "num_tokens": 163673963.0,
      "step": 74710
    },
    {
      "entropy": 5.674691820144654,
      "epoch": 7.415144898769353,
      "grad_norm": 1.375,
      "learning_rate": 0.00012149941419917638,
      "loss": 4.8924,
      "mean_token_accuracy": 0.20106081366539003,
      "num_tokens": 163685281.0,
      "step": 74715
    },
    {
      "entropy": 5.701542091369629,
      "epoch": 7.41564112743152,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012147351063154603,
      "loss": 4.888,
      "mean_token_accuracy": 0.20292139798402786,
      "num_tokens": 163696200.0,
      "step": 74720
    },
    {
      "entropy": 5.7448748588562015,
      "epoch": 7.416137356093688,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00012144761087104182,
      "loss": 4.9264,
      "mean_token_accuracy": 0.20262314081192018,
      "num_tokens": 163706853.0,
      "step": 74725
    },
    {
      "entropy": 5.821821498870849,
      "epoch": 7.416633584755855,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012142171491830591,
      "loss": 5.0376,
      "mean_token_accuracy": 0.1993889644742012,
      "num_tokens": 163720184.0,
      "step": 74730
    },
    {
      "entropy": 5.7702733993530275,
      "epoch": 7.417129813418023,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012139582277398052,
      "loss": 4.9806,
      "mean_token_accuracy": 0.20626272708177568,
      "num_tokens": 163732871.0,
      "step": 74735
    },
    {
      "entropy": 5.6525122165679935,
      "epoch": 7.41762604208019,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00012136993443870782,
      "loss": 4.8072,
      "mean_token_accuracy": 0.20623346865177156,
      "num_tokens": 163742394.0,
      "step": 74740
    },
    {
      "entropy": 5.716154146194458,
      "epoch": 7.418122270742358,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001213440499131297,
      "loss": 4.9289,
      "mean_token_accuracy": 0.19627117365598679,
      "num_tokens": 163753886.0,
      "step": 74745
    },
    {
      "entropy": 5.78901743888855,
      "epoch": 7.418618499404525,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012131816919788805,
      "loss": 4.9578,
      "mean_token_accuracy": 0.1940257504582405,
      "num_tokens": 163764178.0,
      "step": 74750
    },
    {
      "entropy": 5.756517601013184,
      "epoch": 7.419114728066694,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012129229229362466,
      "loss": 4.9755,
      "mean_token_accuracy": 0.20137151032686235,
      "num_tokens": 163775382.0,
      "step": 74755
    },
    {
      "entropy": 5.7124011516571045,
      "epoch": 7.419610956728861,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012126641920098111,
      "loss": 4.9552,
      "mean_token_accuracy": 0.1963492676615715,
      "num_tokens": 163787098.0,
      "step": 74760
    },
    {
      "entropy": 5.7443772315979,
      "epoch": 7.4201071853910285,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012124054992059911,
      "loss": 4.9739,
      "mean_token_accuracy": 0.20017934441566468,
      "num_tokens": 163797654.0,
      "step": 74765
    },
    {
      "entropy": 5.721699619293213,
      "epoch": 7.420603414053196,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012121468445312023,
      "loss": 4.9684,
      "mean_token_accuracy": 0.19398384392261506,
      "num_tokens": 163808479.0,
      "step": 74770
    },
    {
      "entropy": 5.765668773651123,
      "epoch": 7.4210996427153635,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012118882279918578,
      "loss": 4.9864,
      "mean_token_accuracy": 0.19458406120538713,
      "num_tokens": 163818771.0,
      "step": 74775
    },
    {
      "entropy": 5.686966753005981,
      "epoch": 7.421595871377531,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00012116296495943712,
      "loss": 4.8619,
      "mean_token_accuracy": 0.2050539508461952,
      "num_tokens": 163830595.0,
      "step": 74780
    },
    {
      "entropy": 5.711833190917969,
      "epoch": 7.422092100039698,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012113711093451541,
      "loss": 4.9208,
      "mean_token_accuracy": 0.20076503455638886,
      "num_tokens": 163840834.0,
      "step": 74785
    },
    {
      "entropy": 5.7405908584594725,
      "epoch": 7.422588328701866,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012111126072506174,
      "loss": 4.9805,
      "mean_token_accuracy": 0.19365494400262834,
      "num_tokens": 163852037.0,
      "step": 74790
    },
    {
      "entropy": 5.6249096393585205,
      "epoch": 7.423084557364033,
      "grad_norm": 1.59375,
      "learning_rate": 0.00012108541433171729,
      "loss": 4.7998,
      "mean_token_accuracy": 0.2124013662338257,
      "num_tokens": 163862532.0,
      "step": 74795
    },
    {
      "entropy": 5.61840558052063,
      "epoch": 7.423580786026201,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012105957175512281,
      "loss": 4.7416,
      "mean_token_accuracy": 0.2116388902068138,
      "num_tokens": 163872752.0,
      "step": 74800
    },
    {
      "entropy": 5.643697500228882,
      "epoch": 7.424077014688368,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012103373299591934,
      "loss": 4.862,
      "mean_token_accuracy": 0.2024365171790123,
      "num_tokens": 163883574.0,
      "step": 74805
    },
    {
      "entropy": 5.699610471725464,
      "epoch": 7.424573243350536,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001210078980547475,
      "loss": 4.9237,
      "mean_token_accuracy": 0.2012356624007225,
      "num_tokens": 163894491.0,
      "step": 74810
    },
    {
      "entropy": 5.633196496963501,
      "epoch": 7.425069472012703,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012098206693224797,
      "loss": 4.8415,
      "mean_token_accuracy": 0.2089397132396698,
      "num_tokens": 163905476.0,
      "step": 74815
    },
    {
      "entropy": 5.689625310897827,
      "epoch": 7.425565700674871,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012095623962906125,
      "loss": 4.8474,
      "mean_token_accuracy": 0.20108177065849303,
      "num_tokens": 163915569.0,
      "step": 74820
    },
    {
      "entropy": 5.6117490291595455,
      "epoch": 7.426061929337038,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012093041614582788,
      "loss": 4.8306,
      "mean_token_accuracy": 0.20096926242113114,
      "num_tokens": 163927709.0,
      "step": 74825
    },
    {
      "entropy": 5.691177225112915,
      "epoch": 7.426558157999206,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00012090459648318815,
      "loss": 4.8703,
      "mean_token_accuracy": 0.20132800936698914,
      "num_tokens": 163937437.0,
      "step": 74830
    },
    {
      "entropy": 5.653995990753174,
      "epoch": 7.427054386661373,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012087878064178248,
      "loss": 4.8632,
      "mean_token_accuracy": 0.21050299555063248,
      "num_tokens": 163948513.0,
      "step": 74835
    },
    {
      "entropy": 5.613763999938965,
      "epoch": 7.4275506153235415,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012085296862225093,
      "loss": 4.8377,
      "mean_token_accuracy": 0.20347936004400252,
      "num_tokens": 163959265.0,
      "step": 74840
    },
    {
      "entropy": 5.705788469314575,
      "epoch": 7.428046843985709,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012082716042523358,
      "loss": 4.9568,
      "mean_token_accuracy": 0.19430903047323228,
      "num_tokens": 163969896.0,
      "step": 74845
    },
    {
      "entropy": 5.675866413116455,
      "epoch": 7.4285430726478765,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012080135605137036,
      "loss": 4.9139,
      "mean_token_accuracy": 0.20052045881748198,
      "num_tokens": 163980159.0,
      "step": 74850
    },
    {
      "entropy": 5.681297636032104,
      "epoch": 7.429039301310044,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012077555550130134,
      "loss": 4.8848,
      "mean_token_accuracy": 0.20245617926120757,
      "num_tokens": 163991110.0,
      "step": 74855
    },
    {
      "entropy": 5.673102760314942,
      "epoch": 7.4295355299722114,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012074975877566625,
      "loss": 4.8427,
      "mean_token_accuracy": 0.20423495024442673,
      "num_tokens": 164000385.0,
      "step": 74860
    },
    {
      "entropy": 5.701217126846314,
      "epoch": 7.430031758634379,
      "grad_norm": 1.296875,
      "learning_rate": 0.00012072396587510464,
      "loss": 4.9208,
      "mean_token_accuracy": 0.1958618849515915,
      "num_tokens": 164012081.0,
      "step": 74865
    },
    {
      "entropy": 5.649904584884643,
      "epoch": 7.430527987296546,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012069817680025632,
      "loss": 4.8568,
      "mean_token_accuracy": 0.20040037482976913,
      "num_tokens": 164022428.0,
      "step": 74870
    },
    {
      "entropy": 5.683264589309692,
      "epoch": 7.431024215958714,
      "grad_norm": 1.296875,
      "learning_rate": 0.00012067239155176073,
      "loss": 4.8811,
      "mean_token_accuracy": 0.20525047183036804,
      "num_tokens": 164033643.0,
      "step": 74875
    },
    {
      "entropy": 5.693092298507691,
      "epoch": 7.431520444620881,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012064661013025721,
      "loss": 4.8369,
      "mean_token_accuracy": 0.21501991003751755,
      "num_tokens": 164043564.0,
      "step": 74880
    },
    {
      "entropy": 5.685124397277832,
      "epoch": 7.432016673283049,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012062083253638526,
      "loss": 4.8643,
      "mean_token_accuracy": 0.20565001666545868,
      "num_tokens": 164054357.0,
      "step": 74885
    },
    {
      "entropy": 5.7852648258209225,
      "epoch": 7.432512901945216,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012059505877078397,
      "loss": 4.9659,
      "mean_token_accuracy": 0.20187194496393204,
      "num_tokens": 164064980.0,
      "step": 74890
    },
    {
      "entropy": 5.7074237823486325,
      "epoch": 7.433009130607384,
      "grad_norm": 1.6875,
      "learning_rate": 0.00012056928883409243,
      "loss": 4.8659,
      "mean_token_accuracy": 0.20332444608211517,
      "num_tokens": 164074572.0,
      "step": 74895
    },
    {
      "entropy": 5.682847738265991,
      "epoch": 7.433505359269551,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012054352272694985,
      "loss": 4.9436,
      "mean_token_accuracy": 0.19857195615768433,
      "num_tokens": 164086064.0,
      "step": 74900
    },
    {
      "entropy": 5.7106575012207035,
      "epoch": 7.434001587931719,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012051776044999502,
      "loss": 4.9128,
      "mean_token_accuracy": 0.2065366104245186,
      "num_tokens": 164096337.0,
      "step": 74905
    },
    {
      "entropy": 5.765903568267822,
      "epoch": 7.434497816593886,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001204920020038669,
      "loss": 4.8967,
      "mean_token_accuracy": 0.20943914204835892,
      "num_tokens": 164107005.0,
      "step": 74910
    },
    {
      "entropy": 5.740025758743286,
      "epoch": 7.434994045256054,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012046624738920421,
      "loss": 4.9566,
      "mean_token_accuracy": 0.1953759178519249,
      "num_tokens": 164118650.0,
      "step": 74915
    },
    {
      "entropy": 5.676563882827759,
      "epoch": 7.435490273918221,
      "grad_norm": 1.203125,
      "learning_rate": 0.00012044049660664558,
      "loss": 4.8273,
      "mean_token_accuracy": 0.2089540734887123,
      "num_tokens": 164129551.0,
      "step": 74920
    },
    {
      "entropy": 5.70098910331726,
      "epoch": 7.4359865025803895,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012041474965682958,
      "loss": 4.9214,
      "mean_token_accuracy": 0.19731417298316956,
      "num_tokens": 164139901.0,
      "step": 74925
    },
    {
      "entropy": 5.663792753219605,
      "epoch": 7.436482731242556,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012038900654039465,
      "loss": 4.8688,
      "mean_token_accuracy": 0.20059208571910858,
      "num_tokens": 164151687.0,
      "step": 74930
    },
    {
      "entropy": 5.663244342803955,
      "epoch": 7.4369789599047245,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012036326725797919,
      "loss": 4.8592,
      "mean_token_accuracy": 0.20733646154403687,
      "num_tokens": 164163364.0,
      "step": 74935
    },
    {
      "entropy": 5.696163177490234,
      "epoch": 7.437475188566892,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012033753181022156,
      "loss": 4.8695,
      "mean_token_accuracy": 0.20765289664268494,
      "num_tokens": 164174867.0,
      "step": 74940
    },
    {
      "entropy": 5.677250337600708,
      "epoch": 7.437971417229059,
      "grad_norm": 1.234375,
      "learning_rate": 0.00012031180019775985,
      "loss": 4.8762,
      "mean_token_accuracy": 0.19963598400354385,
      "num_tokens": 164187181.0,
      "step": 74945
    },
    {
      "entropy": 5.708123683929443,
      "epoch": 7.438467645891227,
      "grad_norm": 1.5,
      "learning_rate": 0.0001202860724212322,
      "loss": 4.8919,
      "mean_token_accuracy": 0.20191635936498642,
      "num_tokens": 164196663.0,
      "step": 74950
    },
    {
      "entropy": 5.6513035774230955,
      "epoch": 7.438963874553394,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012026034848127655,
      "loss": 4.8328,
      "mean_token_accuracy": 0.2113349512219429,
      "num_tokens": 164207089.0,
      "step": 74955
    },
    {
      "entropy": 5.646324586868286,
      "epoch": 7.439460103215562,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012023462837853074,
      "loss": 4.8849,
      "mean_token_accuracy": 0.20318393111228944,
      "num_tokens": 164218270.0,
      "step": 74960
    },
    {
      "entropy": 5.673853254318237,
      "epoch": 7.439956331877729,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001202089121136327,
      "loss": 4.8905,
      "mean_token_accuracy": 0.2091115355491638,
      "num_tokens": 164230615.0,
      "step": 74965
    },
    {
      "entropy": 5.680390453338623,
      "epoch": 7.440452560539897,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012018319968722015,
      "loss": 4.8728,
      "mean_token_accuracy": 0.19829862266778947,
      "num_tokens": 164242445.0,
      "step": 74970
    },
    {
      "entropy": 5.711466932296753,
      "epoch": 7.440948789202064,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00012015749109993064,
      "loss": 4.8942,
      "mean_token_accuracy": 0.20403157472610473,
      "num_tokens": 164254203.0,
      "step": 74975
    },
    {
      "entropy": 5.716122770309449,
      "epoch": 7.441445017864232,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012013178635240171,
      "loss": 4.8655,
      "mean_token_accuracy": 0.20245133489370346,
      "num_tokens": 164264240.0,
      "step": 74980
    },
    {
      "entropy": 5.645616102218628,
      "epoch": 7.441941246526399,
      "grad_norm": 1.3125,
      "learning_rate": 0.00012010608544527074,
      "loss": 4.8715,
      "mean_token_accuracy": 0.2046057477593422,
      "num_tokens": 164275228.0,
      "step": 74985
    },
    {
      "entropy": 5.68099799156189,
      "epoch": 7.442437475188567,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012008038837917503,
      "loss": 4.9068,
      "mean_token_accuracy": 0.20145500898361207,
      "num_tokens": 164286805.0,
      "step": 74990
    },
    {
      "entropy": 5.680743646621704,
      "epoch": 7.442933703850734,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012005469515475196,
      "loss": 4.8673,
      "mean_token_accuracy": 0.20441238433122635,
      "num_tokens": 164297852.0,
      "step": 74995
    },
    {
      "entropy": 5.6578343391418455,
      "epoch": 7.443429932512902,
      "grad_norm": 1.640625,
      "learning_rate": 0.00012002900577263847,
      "loss": 4.8476,
      "mean_token_accuracy": 0.20877961963415145,
      "num_tokens": 164308255.0,
      "step": 75000
    },
    {
      "epoch": 7.443429932512902,
      "eval_entropy": 5.424729228727776,
      "eval_loss": 5.057244777679443,
      "eval_mean_token_accuracy": 0.19942805975725147,
      "eval_num_tokens": 164308255.0,
      "eval_runtime": 20.4745,
      "eval_samples_per_second": 1644.434,
      "eval_steps_per_second": 205.573,
      "step": 75000
    },
    {
      "entropy": 5.639572238922119,
      "epoch": 7.443926161175069,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012000332023347178,
      "loss": 4.8522,
      "mean_token_accuracy": 0.20563351213932038,
      "num_tokens": 164319608.0,
      "step": 75005
    },
    {
      "entropy": 5.679423666000366,
      "epoch": 7.444422389837237,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011997763853788875,
      "loss": 4.877,
      "mean_token_accuracy": 0.20703867673873902,
      "num_tokens": 164330515.0,
      "step": 75010
    },
    {
      "entropy": 5.63789734840393,
      "epoch": 7.444918618499404,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011995196068652627,
      "loss": 4.805,
      "mean_token_accuracy": 0.20364253371953964,
      "num_tokens": 164341784.0,
      "step": 75015
    },
    {
      "entropy": 5.692947292327881,
      "epoch": 7.445414847161572,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00011992628668002105,
      "loss": 4.7971,
      "mean_token_accuracy": 0.21564187556505204,
      "num_tokens": 164353394.0,
      "step": 75020
    },
    {
      "entropy": 5.668346261978149,
      "epoch": 7.44591107582374,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011990061651900967,
      "loss": 4.9238,
      "mean_token_accuracy": 0.2049685925245285,
      "num_tokens": 164365395.0,
      "step": 75025
    },
    {
      "entropy": 5.587914896011353,
      "epoch": 7.446407304485907,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00011987495020412881,
      "loss": 4.771,
      "mean_token_accuracy": 0.209896582365036,
      "num_tokens": 164374621.0,
      "step": 75030
    },
    {
      "entropy": 5.652127122879028,
      "epoch": 7.446903533148075,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011984928773601501,
      "loss": 4.8731,
      "mean_token_accuracy": 0.21115888208150863,
      "num_tokens": 164387053.0,
      "step": 75035
    },
    {
      "entropy": 5.697199821472168,
      "epoch": 7.447399761810242,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011982362911530455,
      "loss": 4.9405,
      "mean_token_accuracy": 0.1957206904888153,
      "num_tokens": 164397788.0,
      "step": 75040
    },
    {
      "entropy": 5.732629299163818,
      "epoch": 7.44789599047241,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00011979797434263371,
      "loss": 5.0006,
      "mean_token_accuracy": 0.19655104875564575,
      "num_tokens": 164410707.0,
      "step": 75045
    },
    {
      "entropy": 5.715654754638672,
      "epoch": 7.448392219134577,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011977232341863865,
      "loss": 4.8836,
      "mean_token_accuracy": 0.20014107674360276,
      "num_tokens": 164421699.0,
      "step": 75050
    },
    {
      "entropy": 5.682710742950439,
      "epoch": 7.448888447796745,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011974667634395544,
      "loss": 4.8604,
      "mean_token_accuracy": 0.20905144810676574,
      "num_tokens": 164432298.0,
      "step": 75055
    },
    {
      "entropy": 5.700390863418579,
      "epoch": 7.449384676458912,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001197210331192201,
      "loss": 4.8684,
      "mean_token_accuracy": 0.21095144152641296,
      "num_tokens": 164443856.0,
      "step": 75060
    },
    {
      "entropy": 5.756387662887573,
      "epoch": 7.44988090512108,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001196953937450686,
      "loss": 4.9521,
      "mean_token_accuracy": 0.20147334039211273,
      "num_tokens": 164455538.0,
      "step": 75065
    },
    {
      "entropy": 5.738417387008667,
      "epoch": 7.450377133783247,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011966975822213671,
      "loss": 4.9733,
      "mean_token_accuracy": 0.19909269213676453,
      "num_tokens": 164467463.0,
      "step": 75070
    },
    {
      "entropy": 5.712104034423828,
      "epoch": 7.450873362445415,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011964412655106011,
      "loss": 4.8476,
      "mean_token_accuracy": 0.20428648740053176,
      "num_tokens": 164478536.0,
      "step": 75075
    },
    {
      "entropy": 5.716276550292969,
      "epoch": 7.451369591107582,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011961849873247439,
      "loss": 4.9136,
      "mean_token_accuracy": 0.2088467448949814,
      "num_tokens": 164489200.0,
      "step": 75080
    },
    {
      "entropy": 5.705645179748535,
      "epoch": 7.45186581976975,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011959287476701504,
      "loss": 4.9492,
      "mean_token_accuracy": 0.20564621686935425,
      "num_tokens": 164499568.0,
      "step": 75085
    },
    {
      "entropy": 5.67967791557312,
      "epoch": 7.452362048431917,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011956725465531757,
      "loss": 4.8769,
      "mean_token_accuracy": 0.20510509759187698,
      "num_tokens": 164511226.0,
      "step": 75090
    },
    {
      "entropy": 5.657925605773926,
      "epoch": 7.4528582770940845,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001195416383980172,
      "loss": 4.8353,
      "mean_token_accuracy": 0.20510906130075454,
      "num_tokens": 164522357.0,
      "step": 75095
    },
    {
      "entropy": 5.705221176147461,
      "epoch": 7.453354505756252,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011951602599574927,
      "loss": 5.007,
      "mean_token_accuracy": 0.2042039304971695,
      "num_tokens": 164534054.0,
      "step": 75100
    },
    {
      "entropy": 5.6151327133178714,
      "epoch": 7.45385073441842,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011949041744914887,
      "loss": 4.76,
      "mean_token_accuracy": 0.22035883069038392,
      "num_tokens": 164545591.0,
      "step": 75105
    },
    {
      "entropy": 5.675327157974243,
      "epoch": 7.454346963080588,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011946481275885102,
      "loss": 4.7951,
      "mean_token_accuracy": 0.20849941074848174,
      "num_tokens": 164556158.0,
      "step": 75110
    },
    {
      "entropy": 5.707766056060791,
      "epoch": 7.454843191742755,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011943921192549059,
      "loss": 4.9235,
      "mean_token_accuracy": 0.19695059359073638,
      "num_tokens": 164568330.0,
      "step": 75115
    },
    {
      "entropy": 5.628185176849366,
      "epoch": 7.455339420404923,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011941361494970257,
      "loss": 4.8085,
      "mean_token_accuracy": 0.2118418663740158,
      "num_tokens": 164579771.0,
      "step": 75120
    },
    {
      "entropy": 5.609923648834228,
      "epoch": 7.45583564906709,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011938802183212158,
      "loss": 4.8202,
      "mean_token_accuracy": 0.21084514558315276,
      "num_tokens": 164590719.0,
      "step": 75125
    },
    {
      "entropy": 5.654656553268433,
      "epoch": 7.456331877729258,
      "grad_norm": 1.375,
      "learning_rate": 0.00011936243257338241,
      "loss": 4.8545,
      "mean_token_accuracy": 0.20857368111610414,
      "num_tokens": 164601426.0,
      "step": 75130
    },
    {
      "entropy": 5.682450246810913,
      "epoch": 7.456828106391425,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011933684717411954,
      "loss": 4.9232,
      "mean_token_accuracy": 0.20123721957206725,
      "num_tokens": 164612490.0,
      "step": 75135
    },
    {
      "entropy": 5.6426482677459715,
      "epoch": 7.457324335053593,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011931126563496742,
      "loss": 4.7942,
      "mean_token_accuracy": 0.21402648389339446,
      "num_tokens": 164623160.0,
      "step": 75140
    },
    {
      "entropy": 5.673897743225098,
      "epoch": 7.45782056371576,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011928568795656037,
      "loss": 4.8694,
      "mean_token_accuracy": 0.20562738925218582,
      "num_tokens": 164634953.0,
      "step": 75145
    },
    {
      "entropy": 5.738275623321533,
      "epoch": 7.458316792377928,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001192601141395328,
      "loss": 4.9853,
      "mean_token_accuracy": 0.19250465482473372,
      "num_tokens": 164646631.0,
      "step": 75150
    },
    {
      "entropy": 5.713188314437867,
      "epoch": 7.458813021040095,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011923454418451881,
      "loss": 4.9032,
      "mean_token_accuracy": 0.19922783374786376,
      "num_tokens": 164657514.0,
      "step": 75155
    },
    {
      "entropy": 5.684090089797974,
      "epoch": 7.459309249702263,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001192089780921524,
      "loss": 4.8273,
      "mean_token_accuracy": 0.20311300605535507,
      "num_tokens": 164670170.0,
      "step": 75160
    },
    {
      "entropy": 5.675091743469238,
      "epoch": 7.45980547836443,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011918341586306771,
      "loss": 4.8645,
      "mean_token_accuracy": 0.20448970496654512,
      "num_tokens": 164682666.0,
      "step": 75165
    },
    {
      "entropy": 5.693013715744018,
      "epoch": 7.460301707026598,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011915785749789846,
      "loss": 4.8863,
      "mean_token_accuracy": 0.2038629546761513,
      "num_tokens": 164693579.0,
      "step": 75170
    },
    {
      "entropy": 5.6918775081634525,
      "epoch": 7.460797935688765,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00011913230299727863,
      "loss": 4.8453,
      "mean_token_accuracy": 0.19897855669260026,
      "num_tokens": 164703222.0,
      "step": 75175
    },
    {
      "entropy": 5.672535943984985,
      "epoch": 7.4612941643509325,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011910675236184182,
      "loss": 4.877,
      "mean_token_accuracy": 0.20584637373685838,
      "num_tokens": 164713561.0,
      "step": 75180
    },
    {
      "entropy": 5.664019203186035,
      "epoch": 7.4617903930131,
      "grad_norm": 1.21875,
      "learning_rate": 0.00011908120559222163,
      "loss": 4.9446,
      "mean_token_accuracy": 0.1999579280614853,
      "num_tokens": 164724531.0,
      "step": 75185
    },
    {
      "entropy": 5.719166040420532,
      "epoch": 7.462286621675268,
      "grad_norm": 1.53125,
      "learning_rate": 0.00011905566268905149,
      "loss": 4.9072,
      "mean_token_accuracy": 0.20144901275634766,
      "num_tokens": 164735248.0,
      "step": 75190
    },
    {
      "entropy": 5.684590625762939,
      "epoch": 7.462782850337436,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011903012365296496,
      "loss": 4.9376,
      "mean_token_accuracy": 0.19624306559562682,
      "num_tokens": 164745432.0,
      "step": 75195
    },
    {
      "entropy": 5.658017492294311,
      "epoch": 7.463279078999603,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001190045884845952,
      "loss": 4.8437,
      "mean_token_accuracy": 0.2022241771221161,
      "num_tokens": 164754878.0,
      "step": 75200
    },
    {
      "entropy": 5.692170143127441,
      "epoch": 7.463775307661771,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001189790571845756,
      "loss": 4.9456,
      "mean_token_accuracy": 0.2003171056509018,
      "num_tokens": 164766522.0,
      "step": 75205
    },
    {
      "entropy": 5.699146556854248,
      "epoch": 7.464271536323938,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00011895352975353916,
      "loss": 4.8617,
      "mean_token_accuracy": 0.21190523952245713,
      "num_tokens": 164777853.0,
      "step": 75210
    },
    {
      "entropy": 5.710839891433716,
      "epoch": 7.464767764986106,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011892800619211892,
      "loss": 4.8962,
      "mean_token_accuracy": 0.20659706890583038,
      "num_tokens": 164789061.0,
      "step": 75215
    },
    {
      "entropy": 5.6829304695129395,
      "epoch": 7.465263993648273,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011890248650094778,
      "loss": 4.8454,
      "mean_token_accuracy": 0.20735769122838973,
      "num_tokens": 164798071.0,
      "step": 75220
    },
    {
      "entropy": 5.709868669509888,
      "epoch": 7.465760222310441,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011887697068065867,
      "loss": 4.9408,
      "mean_token_accuracy": 0.20068377256393433,
      "num_tokens": 164808812.0,
      "step": 75225
    },
    {
      "entropy": 5.679569482803345,
      "epoch": 7.466256450972608,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001188514587318842,
      "loss": 4.8452,
      "mean_token_accuracy": 0.2103036418557167,
      "num_tokens": 164819829.0,
      "step": 75230
    },
    {
      "entropy": 5.648755073547363,
      "epoch": 7.466752679634776,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011882595065525717,
      "loss": 4.8924,
      "mean_token_accuracy": 0.2000469371676445,
      "num_tokens": 164830419.0,
      "step": 75235
    },
    {
      "entropy": 5.712653064727784,
      "epoch": 7.467248908296943,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011880044645141002,
      "loss": 4.929,
      "mean_token_accuracy": 0.19979767948389054,
      "num_tokens": 164841559.0,
      "step": 75240
    },
    {
      "entropy": 5.70214147567749,
      "epoch": 7.467745136959111,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011877494612097524,
      "loss": 4.9186,
      "mean_token_accuracy": 0.19591813087463378,
      "num_tokens": 164852803.0,
      "step": 75245
    },
    {
      "entropy": 5.678170299530029,
      "epoch": 7.468241365621278,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00011874944966458515,
      "loss": 4.9163,
      "mean_token_accuracy": 0.202799192070961,
      "num_tokens": 164863212.0,
      "step": 75250
    },
    {
      "entropy": 5.784892797470093,
      "epoch": 7.4687375942834455,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011872395708287193,
      "loss": 5.0337,
      "mean_token_accuracy": 0.19797939509153367,
      "num_tokens": 164873407.0,
      "step": 75255
    },
    {
      "entropy": 5.738143634796143,
      "epoch": 7.469233822945613,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011869846837646787,
      "loss": 4.8953,
      "mean_token_accuracy": 0.20294498652219772,
      "num_tokens": 164884083.0,
      "step": 75260
    },
    {
      "entropy": 5.7045646667480465,
      "epoch": 7.4697300516077805,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011867298354600502,
      "loss": 4.8941,
      "mean_token_accuracy": 0.19689910262823104,
      "num_tokens": 164894767.0,
      "step": 75265
    },
    {
      "entropy": 5.731856727600098,
      "epoch": 7.470226280269948,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011864750259211534,
      "loss": 4.9047,
      "mean_token_accuracy": 0.2048775225877762,
      "num_tokens": 164906788.0,
      "step": 75270
    },
    {
      "entropy": 5.691495609283447,
      "epoch": 7.470722508932116,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011862202551543069,
      "loss": 4.8703,
      "mean_token_accuracy": 0.2064949244260788,
      "num_tokens": 164918587.0,
      "step": 75275
    },
    {
      "entropy": 5.692820072174072,
      "epoch": 7.471218737594284,
      "grad_norm": 1.171875,
      "learning_rate": 0.00011859655231658285,
      "loss": 4.8895,
      "mean_token_accuracy": 0.20463042706251144,
      "num_tokens": 164930359.0,
      "step": 75280
    },
    {
      "entropy": 5.672866630554199,
      "epoch": 7.471714966256451,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00011857108299620339,
      "loss": 4.8586,
      "mean_token_accuracy": 0.20831997990608214,
      "num_tokens": 164941691.0,
      "step": 75285
    },
    {
      "entropy": 5.653470373153686,
      "epoch": 7.472211194918619,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011854561755492405,
      "loss": 4.8298,
      "mean_token_accuracy": 0.2068220168352127,
      "num_tokens": 164953571.0,
      "step": 75290
    },
    {
      "entropy": 5.6882161617279055,
      "epoch": 7.472707423580786,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011852015599337631,
      "loss": 4.8704,
      "mean_token_accuracy": 0.20240858495235442,
      "num_tokens": 164964596.0,
      "step": 75295
    },
    {
      "entropy": 5.704441404342651,
      "epoch": 7.473203652242954,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011849469831219147,
      "loss": 4.9869,
      "mean_token_accuracy": 0.20011741518974305,
      "num_tokens": 164974895.0,
      "step": 75300
    },
    {
      "entropy": 5.711254072189331,
      "epoch": 7.473699880905121,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00011846924451200094,
      "loss": 4.8839,
      "mean_token_accuracy": 0.20664367973804473,
      "num_tokens": 164985075.0,
      "step": 75305
    },
    {
      "entropy": 5.659528923034668,
      "epoch": 7.474196109567289,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011844379459343583,
      "loss": 4.9183,
      "mean_token_accuracy": 0.19949532598257064,
      "num_tokens": 164996077.0,
      "step": 75310
    },
    {
      "entropy": 5.688484382629395,
      "epoch": 7.474692338229456,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00011841834855712727,
      "loss": 4.8666,
      "mean_token_accuracy": 0.21253384947776793,
      "num_tokens": 165007826.0,
      "step": 75315
    },
    {
      "entropy": 5.676753330230713,
      "epoch": 7.475188566891624,
      "grad_norm": 1.375,
      "learning_rate": 0.00011839290640370619,
      "loss": 4.8493,
      "mean_token_accuracy": 0.20927755534648895,
      "num_tokens": 165018205.0,
      "step": 75320
    },
    {
      "entropy": 5.741189289093017,
      "epoch": 7.475684795553791,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011836746813380355,
      "loss": 4.9175,
      "mean_token_accuracy": 0.20067522674798965,
      "num_tokens": 165028585.0,
      "step": 75325
    },
    {
      "entropy": 5.758553743362427,
      "epoch": 7.4761810242159585,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011834203374805029,
      "loss": 4.9982,
      "mean_token_accuracy": 0.19650593101978303,
      "num_tokens": 165039102.0,
      "step": 75330
    },
    {
      "entropy": 5.693555164337158,
      "epoch": 7.476677252878126,
      "grad_norm": 1.40625,
      "learning_rate": 0.000118316603247077,
      "loss": 4.8977,
      "mean_token_accuracy": 0.1967688500881195,
      "num_tokens": 165050482.0,
      "step": 75335
    },
    {
      "entropy": 5.752546310424805,
      "epoch": 7.4771734815402935,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011829117663151434,
      "loss": 4.9645,
      "mean_token_accuracy": 0.1993579760193825,
      "num_tokens": 165061274.0,
      "step": 75340
    },
    {
      "entropy": 5.722845125198364,
      "epoch": 7.477669710202461,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011826575390199282,
      "loss": 4.8981,
      "mean_token_accuracy": 0.20212198942899703,
      "num_tokens": 165071616.0,
      "step": 75345
    },
    {
      "entropy": 5.67406120300293,
      "epoch": 7.478165938864628,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001182403350591428,
      "loss": 4.8152,
      "mean_token_accuracy": 0.21041377633810043,
      "num_tokens": 165081690.0,
      "step": 75350
    },
    {
      "entropy": 5.7103982925415036,
      "epoch": 7.478662167526796,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011821492010359469,
      "loss": 4.995,
      "mean_token_accuracy": 0.19518216252326964,
      "num_tokens": 165091824.0,
      "step": 75355
    },
    {
      "entropy": 5.704638528823852,
      "epoch": 7.479158396188964,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011818950903597878,
      "loss": 4.881,
      "mean_token_accuracy": 0.20795097798109055,
      "num_tokens": 165102229.0,
      "step": 75360
    },
    {
      "entropy": 5.651724529266358,
      "epoch": 7.479654624851132,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011816410185692517,
      "loss": 4.843,
      "mean_token_accuracy": 0.2064766988158226,
      "num_tokens": 165112447.0,
      "step": 75365
    },
    {
      "entropy": 5.731849479675293,
      "epoch": 7.480150853513299,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011813869856706384,
      "loss": 4.9991,
      "mean_token_accuracy": 0.19603973627090454,
      "num_tokens": 165124235.0,
      "step": 75370
    },
    {
      "entropy": 5.625015449523926,
      "epoch": 7.480647082175467,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011811329916702482,
      "loss": 4.8111,
      "mean_token_accuracy": 0.21034516990184784,
      "num_tokens": 165133945.0,
      "step": 75375
    },
    {
      "entropy": 5.720046138763427,
      "epoch": 7.481143310837634,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011808790365743783,
      "loss": 4.8845,
      "mean_token_accuracy": 0.19619916677474974,
      "num_tokens": 165145175.0,
      "step": 75380
    },
    {
      "entropy": 5.708200216293335,
      "epoch": 7.481639539499802,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011806251203893279,
      "loss": 4.9032,
      "mean_token_accuracy": 0.19174272418022156,
      "num_tokens": 165155028.0,
      "step": 75385
    },
    {
      "entropy": 5.674835681915283,
      "epoch": 7.482135768161969,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001180371243121392,
      "loss": 4.8765,
      "mean_token_accuracy": 0.21272420436143874,
      "num_tokens": 165166011.0,
      "step": 75390
    },
    {
      "entropy": 5.73855185508728,
      "epoch": 7.482631996824137,
      "grad_norm": 1.296875,
      "learning_rate": 0.00011801174047768678,
      "loss": 5.039,
      "mean_token_accuracy": 0.18905191272497177,
      "num_tokens": 165177999.0,
      "step": 75395
    },
    {
      "entropy": 5.698049116134643,
      "epoch": 7.483128225486304,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011798636053620491,
      "loss": 4.9407,
      "mean_token_accuracy": 0.19722990691661835,
      "num_tokens": 165190320.0,
      "step": 75400
    },
    {
      "entropy": 5.636001062393189,
      "epoch": 7.4836244541484715,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011796098448832291,
      "loss": 4.8433,
      "mean_token_accuracy": 0.2109080120921135,
      "num_tokens": 165201061.0,
      "step": 75405
    },
    {
      "entropy": 5.673306608200074,
      "epoch": 7.484120682810639,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001179356123346702,
      "loss": 4.8219,
      "mean_token_accuracy": 0.20694302767515182,
      "num_tokens": 165212733.0,
      "step": 75410
    },
    {
      "entropy": 5.708618402481079,
      "epoch": 7.4846169114728065,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011791024407587582,
      "loss": 4.9342,
      "mean_token_accuracy": 0.19737130999565125,
      "num_tokens": 165223969.0,
      "step": 75415
    },
    {
      "entropy": 5.692445230484009,
      "epoch": 7.485113140134974,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001178848797125688,
      "loss": 4.9342,
      "mean_token_accuracy": 0.195803602039814,
      "num_tokens": 165234021.0,
      "step": 75420
    },
    {
      "entropy": 5.774344062805175,
      "epoch": 7.4856093687971414,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00011785951924537831,
      "loss": 4.9817,
      "mean_token_accuracy": 0.1940490037202835,
      "num_tokens": 165245936.0,
      "step": 75425
    },
    {
      "entropy": 5.68357367515564,
      "epoch": 7.486105597459309,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011783416267493313,
      "loss": 4.9206,
      "mean_token_accuracy": 0.19173689931631088,
      "num_tokens": 165256923.0,
      "step": 75430
    },
    {
      "entropy": 5.684518241882325,
      "epoch": 7.486601826121476,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000117808810001862,
      "loss": 4.872,
      "mean_token_accuracy": 0.20127101391553878,
      "num_tokens": 165267100.0,
      "step": 75435
    },
    {
      "entropy": 5.738047981262207,
      "epoch": 7.487098054783644,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001177834612267937,
      "loss": 4.9573,
      "mean_token_accuracy": 0.2024650275707245,
      "num_tokens": 165279222.0,
      "step": 75440
    },
    {
      "entropy": 5.76999454498291,
      "epoch": 7.487594283445812,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011775811635035685,
      "loss": 4.9119,
      "mean_token_accuracy": 0.20304998010396957,
      "num_tokens": 165289919.0,
      "step": 75445
    },
    {
      "entropy": 5.750783252716064,
      "epoch": 7.48809051210798,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00011773277537317976,
      "loss": 4.9412,
      "mean_token_accuracy": 0.19947172850370407,
      "num_tokens": 165300990.0,
      "step": 75450
    },
    {
      "entropy": 5.664621829986572,
      "epoch": 7.488586740770147,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011770743829589107,
      "loss": 4.7855,
      "mean_token_accuracy": 0.21211886405944824,
      "num_tokens": 165311083.0,
      "step": 75455
    },
    {
      "entropy": 5.682478523254394,
      "epoch": 7.489082969432315,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011768210511911892,
      "loss": 4.8763,
      "mean_token_accuracy": 0.20583308935165406,
      "num_tokens": 165321820.0,
      "step": 75460
    },
    {
      "entropy": 5.701259279251099,
      "epoch": 7.489579198094482,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011765677584349154,
      "loss": 4.8648,
      "mean_token_accuracy": 0.20716896653175354,
      "num_tokens": 165332336.0,
      "step": 75465
    },
    {
      "entropy": 5.604420328140259,
      "epoch": 7.49007542675665,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011763145046963715,
      "loss": 4.8432,
      "mean_token_accuracy": 0.1981372743844986,
      "num_tokens": 165343602.0,
      "step": 75470
    },
    {
      "entropy": 5.6862256050109865,
      "epoch": 7.490571655418817,
      "grad_norm": 1.375,
      "learning_rate": 0.00011760612899818366,
      "loss": 4.8491,
      "mean_token_accuracy": 0.21168037950992585,
      "num_tokens": 165354681.0,
      "step": 75475
    },
    {
      "entropy": 5.72487359046936,
      "epoch": 7.491067884080985,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011758081142975906,
      "loss": 4.9316,
      "mean_token_accuracy": 0.20308834314346313,
      "num_tokens": 165364603.0,
      "step": 75480
    },
    {
      "entropy": 5.761791849136353,
      "epoch": 7.491564112743152,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011755549776499103,
      "loss": 4.9518,
      "mean_token_accuracy": 0.1959439054131508,
      "num_tokens": 165375770.0,
      "step": 75485
    },
    {
      "entropy": 5.638970518112183,
      "epoch": 7.4920603414053195,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011753018800450748,
      "loss": 4.8316,
      "mean_token_accuracy": 0.2054920181632042,
      "num_tokens": 165386632.0,
      "step": 75490
    },
    {
      "entropy": 5.75272650718689,
      "epoch": 7.492556570067487,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011750488214893587,
      "loss": 4.9047,
      "mean_token_accuracy": 0.20175733864307405,
      "num_tokens": 165397220.0,
      "step": 75495
    },
    {
      "entropy": 5.681012773513794,
      "epoch": 7.4930527987296545,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011747958019890389,
      "loss": 4.8967,
      "mean_token_accuracy": 0.2002720892429352,
      "num_tokens": 165408323.0,
      "step": 75500
    },
    {
      "entropy": 5.628765535354614,
      "epoch": 7.493549027391822,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011745428215503885,
      "loss": 4.8528,
      "mean_token_accuracy": 0.21211090832948684,
      "num_tokens": 165418490.0,
      "step": 75505
    },
    {
      "entropy": 5.676031637191772,
      "epoch": 7.494045256053989,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011742898801796818,
      "loss": 4.9198,
      "mean_token_accuracy": 0.1961361512541771,
      "num_tokens": 165429520.0,
      "step": 75510
    },
    {
      "entropy": 5.668372726440429,
      "epoch": 7.494541484716157,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011740369778831899,
      "loss": 4.8047,
      "mean_token_accuracy": 0.20792740881443023,
      "num_tokens": 165440854.0,
      "step": 75515
    },
    {
      "entropy": 5.700179147720337,
      "epoch": 7.495037713378324,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011737841146671857,
      "loss": 4.8999,
      "mean_token_accuracy": 0.20366546064615249,
      "num_tokens": 165451824.0,
      "step": 75520
    },
    {
      "entropy": 5.779961490631104,
      "epoch": 7.495533942040492,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011735312905379383,
      "loss": 4.991,
      "mean_token_accuracy": 0.19289775490760802,
      "num_tokens": 165463057.0,
      "step": 75525
    },
    {
      "entropy": 5.686125755310059,
      "epoch": 7.49603017070266,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00011732785055017185,
      "loss": 4.9151,
      "mean_token_accuracy": 0.19998308420181274,
      "num_tokens": 165475427.0,
      "step": 75530
    },
    {
      "entropy": 5.69255576133728,
      "epoch": 7.496526399364828,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011730257595647946,
      "loss": 4.8835,
      "mean_token_accuracy": 0.20369485914707183,
      "num_tokens": 165486087.0,
      "step": 75535
    },
    {
      "entropy": 5.680118131637573,
      "epoch": 7.497022628026995,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011727730527334333,
      "loss": 4.9215,
      "mean_token_accuracy": 0.2006572961807251,
      "num_tokens": 165496596.0,
      "step": 75540
    },
    {
      "entropy": 5.701265573501587,
      "epoch": 7.497518856689163,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011725203850139021,
      "loss": 4.8829,
      "mean_token_accuracy": 0.1994830310344696,
      "num_tokens": 165507592.0,
      "step": 75545
    },
    {
      "entropy": 5.653654956817627,
      "epoch": 7.49801508535133,
      "grad_norm": 1.265625,
      "learning_rate": 0.00011722677564124652,
      "loss": 4.7834,
      "mean_token_accuracy": 0.2190800204873085,
      "num_tokens": 165519498.0,
      "step": 75550
    },
    {
      "entropy": 5.618029022216797,
      "epoch": 7.498511314013498,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011720151669353883,
      "loss": 4.8189,
      "mean_token_accuracy": 0.2065761297941208,
      "num_tokens": 165529737.0,
      "step": 75555
    },
    {
      "entropy": 5.6195127964019775,
      "epoch": 7.499007542675665,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001171762616588936,
      "loss": 4.8488,
      "mean_token_accuracy": 0.20651358664035796,
      "num_tokens": 165539560.0,
      "step": 75560
    },
    {
      "entropy": 5.689697742462158,
      "epoch": 7.4995037713378325,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011715101053793694,
      "loss": 4.869,
      "mean_token_accuracy": 0.20353133380413055,
      "num_tokens": 165549144.0,
      "step": 75565
    },
    {
      "entropy": 5.663469123840332,
      "epoch": 7.5,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011712576333129513,
      "loss": 4.8731,
      "mean_token_accuracy": 0.20992440432310105,
      "num_tokens": 165560003.0,
      "step": 75570
    },
    {
      "entropy": 5.701990079879761,
      "epoch": 7.5004962286621675,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001171005200395942,
      "loss": 4.861,
      "mean_token_accuracy": 0.19866681545972825,
      "num_tokens": 165571024.0,
      "step": 75575
    },
    {
      "entropy": 5.693926382064819,
      "epoch": 7.500992457324335,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011707528066346006,
      "loss": 4.9387,
      "mean_token_accuracy": 0.1998352587223053,
      "num_tokens": 165580262.0,
      "step": 75580
    },
    {
      "entropy": 5.679874610900879,
      "epoch": 7.501488685986502,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011705004520351863,
      "loss": 4.8658,
      "mean_token_accuracy": 0.21031612604856492,
      "num_tokens": 165591656.0,
      "step": 75585
    },
    {
      "entropy": 5.685077905654907,
      "epoch": 7.50198491464867,
      "grad_norm": 1.375,
      "learning_rate": 0.00011702481366039582,
      "loss": 4.8178,
      "mean_token_accuracy": 0.20868296325206756,
      "num_tokens": 165603284.0,
      "step": 75590
    },
    {
      "entropy": 5.679194211959839,
      "epoch": 7.502481143310837,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011699958603471724,
      "loss": 4.8781,
      "mean_token_accuracy": 0.20288958251476288,
      "num_tokens": 165613365.0,
      "step": 75595
    },
    {
      "entropy": 5.679115962982178,
      "epoch": 7.502977371973005,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011697436232710847,
      "loss": 4.8683,
      "mean_token_accuracy": 0.21449664682149888,
      "num_tokens": 165623656.0,
      "step": 75600
    },
    {
      "entropy": 5.744059610366821,
      "epoch": 7.503473600635172,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011694914253819496,
      "loss": 4.9988,
      "mean_token_accuracy": 0.19241198599338533,
      "num_tokens": 165635494.0,
      "step": 75605
    },
    {
      "entropy": 5.655004024505615,
      "epoch": 7.50396982929734,
      "grad_norm": 1.375,
      "learning_rate": 0.00011692392666860216,
      "loss": 4.8248,
      "mean_token_accuracy": 0.21246956288814545,
      "num_tokens": 165645974.0,
      "step": 75610
    },
    {
      "entropy": 5.609909582138061,
      "epoch": 7.504466057959508,
      "grad_norm": 1.265625,
      "learning_rate": 0.00011689871471895529,
      "loss": 4.7786,
      "mean_token_accuracy": 0.21007352769374849,
      "num_tokens": 165657268.0,
      "step": 75615
    },
    {
      "entropy": 5.620755672454834,
      "epoch": 7.504962286621676,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011687350668987962,
      "loss": 4.8086,
      "mean_token_accuracy": 0.19898166060447692,
      "num_tokens": 165669566.0,
      "step": 75620
    },
    {
      "entropy": 5.690985918045044,
      "epoch": 7.505458515283843,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011684830258200033,
      "loss": 4.9268,
      "mean_token_accuracy": 0.1977507784962654,
      "num_tokens": 165681524.0,
      "step": 75625
    },
    {
      "entropy": 5.713263034820557,
      "epoch": 7.505954743946011,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00011682310239594234,
      "loss": 4.9063,
      "mean_token_accuracy": 0.2047914147377014,
      "num_tokens": 165693882.0,
      "step": 75630
    },
    {
      "entropy": 5.697393321990967,
      "epoch": 7.506450972608178,
      "grad_norm": 1.15625,
      "learning_rate": 0.00011679790613233056,
      "loss": 4.952,
      "mean_token_accuracy": 0.19307398796081543,
      "num_tokens": 165705609.0,
      "step": 75635
    },
    {
      "entropy": 5.773855352401734,
      "epoch": 7.5069472012703455,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011677271379178975,
      "loss": 4.9847,
      "mean_token_accuracy": 0.19370807260274886,
      "num_tokens": 165716603.0,
      "step": 75640
    },
    {
      "entropy": 5.63633017539978,
      "epoch": 7.507443429932513,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011674752537494475,
      "loss": 4.8053,
      "mean_token_accuracy": 0.2077138051390648,
      "num_tokens": 165727820.0,
      "step": 75645
    },
    {
      "entropy": 5.671502017974854,
      "epoch": 7.5079396585946805,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011672234088242007,
      "loss": 4.8948,
      "mean_token_accuracy": 0.19792084097862245,
      "num_tokens": 165740079.0,
      "step": 75650
    },
    {
      "entropy": 5.703852558135987,
      "epoch": 7.508435887256848,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001166971603148403,
      "loss": 4.8165,
      "mean_token_accuracy": 0.21460568010807038,
      "num_tokens": 165750211.0,
      "step": 75655
    },
    {
      "entropy": 5.708106422424317,
      "epoch": 7.508932115919015,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001166719836728299,
      "loss": 4.847,
      "mean_token_accuracy": 0.20193740278482436,
      "num_tokens": 165760636.0,
      "step": 75660
    },
    {
      "entropy": 5.690715217590332,
      "epoch": 7.509428344581183,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011664681095701311,
      "loss": 4.8668,
      "mean_token_accuracy": 0.2097036898136139,
      "num_tokens": 165770136.0,
      "step": 75665
    },
    {
      "entropy": 5.658265542984009,
      "epoch": 7.50992457324335,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011662164216801411,
      "loss": 4.902,
      "mean_token_accuracy": 0.20549207627773286,
      "num_tokens": 165780897.0,
      "step": 75670
    },
    {
      "entropy": 5.683430624008179,
      "epoch": 7.510420801905518,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001165964773064572,
      "loss": 4.8883,
      "mean_token_accuracy": 0.2015915110707283,
      "num_tokens": 165791829.0,
      "step": 75675
    },
    {
      "entropy": 5.723835229873657,
      "epoch": 7.510917030567685,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011657131637296631,
      "loss": 4.941,
      "mean_token_accuracy": 0.20392972081899643,
      "num_tokens": 165801761.0,
      "step": 75680
    },
    {
      "entropy": 5.693230056762696,
      "epoch": 7.511413259229853,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011654615936816533,
      "loss": 4.9143,
      "mean_token_accuracy": 0.19651485085487366,
      "num_tokens": 165811908.0,
      "step": 75685
    },
    {
      "entropy": 5.7261683464050295,
      "epoch": 7.51190948789202,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001165210062926782,
      "loss": 4.9755,
      "mean_token_accuracy": 0.19703267514705658,
      "num_tokens": 165824673.0,
      "step": 75690
    },
    {
      "entropy": 5.706812334060669,
      "epoch": 7.512405716554188,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00011649585714712865,
      "loss": 4.8924,
      "mean_token_accuracy": 0.20597149580717086,
      "num_tokens": 165835565.0,
      "step": 75695
    },
    {
      "entropy": 5.697030830383301,
      "epoch": 7.512901945216356,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011647071193214024,
      "loss": 4.8799,
      "mean_token_accuracy": 0.20269794166088104,
      "num_tokens": 165846690.0,
      "step": 75700
    },
    {
      "entropy": 5.693292140960693,
      "epoch": 7.513398173878524,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011644557064833661,
      "loss": 4.8851,
      "mean_token_accuracy": 0.19841250479221345,
      "num_tokens": 165859149.0,
      "step": 75705
    },
    {
      "entropy": 5.686194133758545,
      "epoch": 7.513894402540691,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011642043329634119,
      "loss": 4.8995,
      "mean_token_accuracy": 0.20587237924337387,
      "num_tokens": 165871121.0,
      "step": 75710
    },
    {
      "entropy": 5.761117267608642,
      "epoch": 7.5143906312028586,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00011639529987677726,
      "loss": 4.9639,
      "mean_token_accuracy": 0.20018037110567094,
      "num_tokens": 165882851.0,
      "step": 75715
    },
    {
      "entropy": 5.768780755996704,
      "epoch": 7.514886859865026,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011637017039026818,
      "loss": 4.9177,
      "mean_token_accuracy": 0.20593856424093246,
      "num_tokens": 165894072.0,
      "step": 75720
    },
    {
      "entropy": 5.669166088104248,
      "epoch": 7.5153830885271935,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011634504483743707,
      "loss": 4.8646,
      "mean_token_accuracy": 0.20785272717475892,
      "num_tokens": 165907118.0,
      "step": 75725
    },
    {
      "entropy": 5.678579664230346,
      "epoch": 7.515879317189361,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001163199232189069,
      "loss": 4.813,
      "mean_token_accuracy": 0.20319856852293014,
      "num_tokens": 165917690.0,
      "step": 75730
    },
    {
      "entropy": 5.7065972805023195,
      "epoch": 7.5163755458515285,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011629480553530078,
      "loss": 4.8822,
      "mean_token_accuracy": 0.20195312947034835,
      "num_tokens": 165928047.0,
      "step": 75735
    },
    {
      "entropy": 5.750730800628662,
      "epoch": 7.516871774513696,
      "grad_norm": 1.28125,
      "learning_rate": 0.00011626969178724149,
      "loss": 5.0251,
      "mean_token_accuracy": 0.19155137687921525,
      "num_tokens": 165941382.0,
      "step": 75740
    },
    {
      "entropy": 5.672338819503784,
      "epoch": 7.517368003175863,
      "grad_norm": 1.5,
      "learning_rate": 0.00011624458197535175,
      "loss": 4.9057,
      "mean_token_accuracy": 0.2029791370034218,
      "num_tokens": 165952543.0,
      "step": 75745
    },
    {
      "entropy": 5.786960506439209,
      "epoch": 7.517864231838031,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011621947610025436,
      "loss": 5.0264,
      "mean_token_accuracy": 0.19588462114334107,
      "num_tokens": 165962931.0,
      "step": 75750
    },
    {
      "entropy": 5.701370096206665,
      "epoch": 7.518360460500198,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011619437416257184,
      "loss": 4.8227,
      "mean_token_accuracy": 0.21559910327196122,
      "num_tokens": 165974215.0,
      "step": 75755
    },
    {
      "entropy": 5.739557838439941,
      "epoch": 7.518856689162366,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011616927616292654,
      "loss": 4.9413,
      "mean_token_accuracy": 0.20032449811697006,
      "num_tokens": 165985262.0,
      "step": 75760
    },
    {
      "entropy": 5.702292108535767,
      "epoch": 7.519352917824533,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011614418210194102,
      "loss": 4.8797,
      "mean_token_accuracy": 0.20082941502332688,
      "num_tokens": 165995040.0,
      "step": 75765
    },
    {
      "entropy": 5.646098852157593,
      "epoch": 7.519849146486701,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011611909198023749,
      "loss": 4.9264,
      "mean_token_accuracy": 0.19823315143585205,
      "num_tokens": 166006164.0,
      "step": 75770
    },
    {
      "entropy": 5.691874694824219,
      "epoch": 7.520345375148868,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011609400579843812,
      "loss": 4.9716,
      "mean_token_accuracy": 0.19963667541742325,
      "num_tokens": 166017518.0,
      "step": 75775
    },
    {
      "entropy": 5.681771516799927,
      "epoch": 7.520841603811036,
      "grad_norm": 1.296875,
      "learning_rate": 0.00011606892355716492,
      "loss": 4.8922,
      "mean_token_accuracy": 0.204949451982975,
      "num_tokens": 166028687.0,
      "step": 75780
    },
    {
      "entropy": 5.7133043766021725,
      "epoch": 7.521337832473204,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011604384525704002,
      "loss": 4.8795,
      "mean_token_accuracy": 0.198341004550457,
      "num_tokens": 166040085.0,
      "step": 75785
    },
    {
      "entropy": 5.671456718444825,
      "epoch": 7.521834061135372,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011601877089868519,
      "loss": 4.8995,
      "mean_token_accuracy": 0.19543635547161103,
      "num_tokens": 166050601.0,
      "step": 75790
    },
    {
      "entropy": 5.681548118591309,
      "epoch": 7.522330289797539,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011599370048272231,
      "loss": 4.8967,
      "mean_token_accuracy": 0.19765037000179292,
      "num_tokens": 166062421.0,
      "step": 75795
    },
    {
      "entropy": 5.689037227630616,
      "epoch": 7.5228265184597065,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011596863400977304,
      "loss": 4.9046,
      "mean_token_accuracy": 0.20387446731328965,
      "num_tokens": 166074165.0,
      "step": 75800
    },
    {
      "entropy": 5.724886083602906,
      "epoch": 7.523322747121874,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011594357148045897,
      "loss": 4.9562,
      "mean_token_accuracy": 0.19590138643980026,
      "num_tokens": 166085335.0,
      "step": 75805
    },
    {
      "entropy": 5.687963581085205,
      "epoch": 7.5238189757840415,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001159185128954015,
      "loss": 4.8885,
      "mean_token_accuracy": 0.20264606475830077,
      "num_tokens": 166094911.0,
      "step": 75810
    },
    {
      "entropy": 5.656887769699097,
      "epoch": 7.524315204446209,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001158934582552222,
      "loss": 4.8496,
      "mean_token_accuracy": 0.21334651857614517,
      "num_tokens": 166104122.0,
      "step": 75815
    },
    {
      "entropy": 5.644549179077148,
      "epoch": 7.524811433108376,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011586840756054224,
      "loss": 4.8327,
      "mean_token_accuracy": 0.2125031903386116,
      "num_tokens": 166115711.0,
      "step": 75820
    },
    {
      "entropy": 5.63635106086731,
      "epoch": 7.525307661770544,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011584336081198292,
      "loss": 4.8593,
      "mean_token_accuracy": 0.20665916204452514,
      "num_tokens": 166126375.0,
      "step": 75825
    },
    {
      "entropy": 5.6670081615448,
      "epoch": 7.525803890432711,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011581831801016531,
      "loss": 4.9104,
      "mean_token_accuracy": 0.201470784842968,
      "num_tokens": 166137398.0,
      "step": 75830
    },
    {
      "entropy": 5.699907207489014,
      "epoch": 7.526300119094879,
      "grad_norm": 1.5,
      "learning_rate": 0.00011579327915571041,
      "loss": 4.8644,
      "mean_token_accuracy": 0.20480111092329026,
      "num_tokens": 166147337.0,
      "step": 75835
    },
    {
      "entropy": 5.663353157043457,
      "epoch": 7.526796347757046,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011576824424923913,
      "loss": 4.8619,
      "mean_token_accuracy": 0.20140828043222428,
      "num_tokens": 166156759.0,
      "step": 75840
    },
    {
      "entropy": 5.635397720336914,
      "epoch": 7.527292576419214,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011574321329137219,
      "loss": 4.8044,
      "mean_token_accuracy": 0.2084490790963173,
      "num_tokens": 166166569.0,
      "step": 75845
    },
    {
      "entropy": 5.629266214370728,
      "epoch": 7.527788805081381,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001157181862827304,
      "loss": 4.8319,
      "mean_token_accuracy": 0.20709496587514878,
      "num_tokens": 166177823.0,
      "step": 75850
    },
    {
      "entropy": 5.651770067214966,
      "epoch": 7.528285033743549,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011569316322393444,
      "loss": 4.8474,
      "mean_token_accuracy": 0.2037935271859169,
      "num_tokens": 166187090.0,
      "step": 75855
    },
    {
      "entropy": 5.710819435119629,
      "epoch": 7.528781262405716,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011566814411560475,
      "loss": 4.8943,
      "mean_token_accuracy": 0.19779685288667678,
      "num_tokens": 166196612.0,
      "step": 75860
    },
    {
      "entropy": 5.604361963272095,
      "epoch": 7.529277491067884,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011564312895836177,
      "loss": 4.8285,
      "mean_token_accuracy": 0.20706521421670915,
      "num_tokens": 166206792.0,
      "step": 75865
    },
    {
      "entropy": 5.723361396789551,
      "epoch": 7.529773719730052,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011561811775282583,
      "loss": 4.9417,
      "mean_token_accuracy": 0.20104961544275285,
      "num_tokens": 166216834.0,
      "step": 75870
    },
    {
      "entropy": 5.685653448104858,
      "epoch": 7.530269948392219,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011559311049961701,
      "loss": 4.9019,
      "mean_token_accuracy": 0.20732731074094773,
      "num_tokens": 166227764.0,
      "step": 75875
    },
    {
      "entropy": 5.641907119750977,
      "epoch": 7.530766177054387,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001155681071993556,
      "loss": 4.8179,
      "mean_token_accuracy": 0.20790619999170304,
      "num_tokens": 166240138.0,
      "step": 75880
    },
    {
      "entropy": 5.709162950515747,
      "epoch": 7.5312624057165545,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00011554310785266164,
      "loss": 4.8919,
      "mean_token_accuracy": 0.2005685970187187,
      "num_tokens": 166250070.0,
      "step": 75885
    },
    {
      "entropy": 5.694460010528564,
      "epoch": 7.531758634378722,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000115518112460155,
      "loss": 4.8721,
      "mean_token_accuracy": 0.20504790395498276,
      "num_tokens": 166260845.0,
      "step": 75890
    },
    {
      "entropy": 5.75649094581604,
      "epoch": 7.532254863040889,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011549312102245551,
      "loss": 4.9368,
      "mean_token_accuracy": 0.19845907986164094,
      "num_tokens": 166272483.0,
      "step": 75895
    },
    {
      "entropy": 5.676439476013184,
      "epoch": 7.532751091703057,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011546813354018293,
      "loss": 4.9103,
      "mean_token_accuracy": 0.20361623167991638,
      "num_tokens": 166284056.0,
      "step": 75900
    },
    {
      "entropy": 5.7095190525054935,
      "epoch": 7.533247320365224,
      "grad_norm": 1.296875,
      "learning_rate": 0.00011544315001395682,
      "loss": 4.974,
      "mean_token_accuracy": 0.2029205173254013,
      "num_tokens": 166294576.0,
      "step": 75905
    },
    {
      "entropy": 5.673703289031982,
      "epoch": 7.533743549027392,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011541817044439681,
      "loss": 4.85,
      "mean_token_accuracy": 0.2035421147942543,
      "num_tokens": 166304802.0,
      "step": 75910
    },
    {
      "entropy": 5.671712350845337,
      "epoch": 7.534239777689559,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011539319483212226,
      "loss": 4.8402,
      "mean_token_accuracy": 0.212987020611763,
      "num_tokens": 166315791.0,
      "step": 75915
    },
    {
      "entropy": 5.662074756622315,
      "epoch": 7.534736006351727,
      "grad_norm": 1.296875,
      "learning_rate": 0.00011536822317775262,
      "loss": 4.866,
      "mean_token_accuracy": 0.20183873921632767,
      "num_tokens": 166327802.0,
      "step": 75920
    },
    {
      "entropy": 5.741487598419189,
      "epoch": 7.535232235013894,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011534325548190708,
      "loss": 4.933,
      "mean_token_accuracy": 0.19931150674819947,
      "num_tokens": 166338996.0,
      "step": 75925
    },
    {
      "entropy": 5.684734964370728,
      "epoch": 7.535728463676062,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011531829174520475,
      "loss": 4.9127,
      "mean_token_accuracy": 0.2090247243642807,
      "num_tokens": 166350714.0,
      "step": 75930
    },
    {
      "entropy": 5.659051370620728,
      "epoch": 7.536224692338229,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00011529333196826465,
      "loss": 4.8321,
      "mean_token_accuracy": 0.20856657326221467,
      "num_tokens": 166361053.0,
      "step": 75935
    },
    {
      "entropy": 5.675377035140992,
      "epoch": 7.536720921000397,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011526837615170583,
      "loss": 4.905,
      "mean_token_accuracy": 0.2033027231693268,
      "num_tokens": 166371766.0,
      "step": 75940
    },
    {
      "entropy": 5.6953941822052006,
      "epoch": 7.537217149662564,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011524342429614704,
      "loss": 4.8843,
      "mean_token_accuracy": 0.19387671649456023,
      "num_tokens": 166381385.0,
      "step": 75945
    },
    {
      "entropy": 5.711613368988037,
      "epoch": 7.537713378324732,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011521847640220716,
      "loss": 4.9337,
      "mean_token_accuracy": 0.19750594943761826,
      "num_tokens": 166392192.0,
      "step": 75950
    },
    {
      "entropy": 5.648209524154663,
      "epoch": 7.5382096069869,
      "grad_norm": 1.296875,
      "learning_rate": 0.00011519353247050475,
      "loss": 4.8924,
      "mean_token_accuracy": 0.21170439720153808,
      "num_tokens": 166402862.0,
      "step": 75955
    },
    {
      "entropy": 5.60919418334961,
      "epoch": 7.538705835649067,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011516859250165838,
      "loss": 4.8652,
      "mean_token_accuracy": 0.21137368232011794,
      "num_tokens": 166413633.0,
      "step": 75960
    },
    {
      "entropy": 5.661532878875732,
      "epoch": 7.539202064311235,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011514365649628644,
      "loss": 4.8528,
      "mean_token_accuracy": 0.20372967123985292,
      "num_tokens": 166424353.0,
      "step": 75965
    },
    {
      "entropy": 5.726870441436768,
      "epoch": 7.539698292973402,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011511872445500742,
      "loss": 4.9277,
      "mean_token_accuracy": 0.20210566073656083,
      "num_tokens": 166434461.0,
      "step": 75970
    },
    {
      "entropy": 5.676540279388428,
      "epoch": 7.54019452163557,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011509379637843948,
      "loss": 4.8405,
      "mean_token_accuracy": 0.20985988825559615,
      "num_tokens": 166445970.0,
      "step": 75975
    },
    {
      "entropy": 5.706249618530274,
      "epoch": 7.540690750297737,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001150688722672009,
      "loss": 4.8997,
      "mean_token_accuracy": 0.2004474937915802,
      "num_tokens": 166458451.0,
      "step": 75980
    },
    {
      "entropy": 5.692497634887696,
      "epoch": 7.541186978959905,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011504395212190962,
      "loss": 4.9129,
      "mean_token_accuracy": 0.20294012427330016,
      "num_tokens": 166470384.0,
      "step": 75985
    },
    {
      "entropy": 5.65823221206665,
      "epoch": 7.541683207622072,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011501903594318368,
      "loss": 4.8171,
      "mean_token_accuracy": 0.2092852383852005,
      "num_tokens": 166481614.0,
      "step": 75990
    },
    {
      "entropy": 5.751466178894043,
      "epoch": 7.54217943628424,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011499412373164085,
      "loss": 5.0033,
      "mean_token_accuracy": 0.19767818301916124,
      "num_tokens": 166492087.0,
      "step": 75995
    },
    {
      "entropy": 5.66689977645874,
      "epoch": 7.542675664946407,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011496921548789903,
      "loss": 4.8619,
      "mean_token_accuracy": 0.203899385035038,
      "num_tokens": 166503190.0,
      "step": 76000
    },
    {
      "entropy": 5.6719207763671875,
      "epoch": 7.543171893608575,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011494431121257584,
      "loss": 4.8106,
      "mean_token_accuracy": 0.20924443155527114,
      "num_tokens": 166513472.0,
      "step": 76005
    },
    {
      "entropy": 5.727767515182495,
      "epoch": 7.543668122270742,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011491941090628877,
      "loss": 4.8934,
      "mean_token_accuracy": 0.20423242151737214,
      "num_tokens": 166524886.0,
      "step": 76010
    },
    {
      "entropy": 5.729034948348999,
      "epoch": 7.54416435093291,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011489451456965544,
      "loss": 4.8942,
      "mean_token_accuracy": 0.20535315871238707,
      "num_tokens": 166535264.0,
      "step": 76015
    },
    {
      "entropy": 5.672491455078125,
      "epoch": 7.544660579595077,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011486962220329312,
      "loss": 4.8097,
      "mean_token_accuracy": 0.2111557111144066,
      "num_tokens": 166546412.0,
      "step": 76020
    },
    {
      "entropy": 5.696306419372559,
      "epoch": 7.545156808257245,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011484473380781909,
      "loss": 4.8694,
      "mean_token_accuracy": 0.2055825725197792,
      "num_tokens": 166558155.0,
      "step": 76025
    },
    {
      "entropy": 5.656610488891602,
      "epoch": 7.545653036919412,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011481984938385062,
      "loss": 4.872,
      "mean_token_accuracy": 0.20253982692956923,
      "num_tokens": 166569770.0,
      "step": 76030
    },
    {
      "entropy": 5.749241828918457,
      "epoch": 7.54614926558158,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011479496893200474,
      "loss": 4.9238,
      "mean_token_accuracy": 0.20707274824380875,
      "num_tokens": 166581314.0,
      "step": 76035
    },
    {
      "entropy": 5.673499679565429,
      "epoch": 7.546645494243748,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011477009245289831,
      "loss": 4.871,
      "mean_token_accuracy": 0.20976967066526414,
      "num_tokens": 166593070.0,
      "step": 76040
    },
    {
      "entropy": 5.742497873306275,
      "epoch": 7.547141722905915,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011474521994714843,
      "loss": 4.9692,
      "mean_token_accuracy": 0.19310105592012405,
      "num_tokens": 166605661.0,
      "step": 76045
    },
    {
      "entropy": 5.667767667770386,
      "epoch": 7.547637951568083,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00011472035141537174,
      "loss": 4.8426,
      "mean_token_accuracy": 0.20855579972267152,
      "num_tokens": 166617324.0,
      "step": 76050
    },
    {
      "entropy": 5.732237768173218,
      "epoch": 7.54813418023025,
      "grad_norm": 1.375,
      "learning_rate": 0.00011469548685818491,
      "loss": 4.9751,
      "mean_token_accuracy": 0.19775811433792115,
      "num_tokens": 166627295.0,
      "step": 76055
    },
    {
      "entropy": 5.679916143417358,
      "epoch": 7.548630408892418,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011467062627620465,
      "loss": 4.8374,
      "mean_token_accuracy": 0.21245890706777573,
      "num_tokens": 166638351.0,
      "step": 76060
    },
    {
      "entropy": 5.71341905593872,
      "epoch": 7.549126637554585,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00011464576967004739,
      "loss": 4.9667,
      "mean_token_accuracy": 0.19136929512023926,
      "num_tokens": 166648532.0,
      "step": 76065
    },
    {
      "entropy": 5.678080701828003,
      "epoch": 7.549622866216753,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011462091704032951,
      "loss": 4.9024,
      "mean_token_accuracy": 0.20483655482530594,
      "num_tokens": 166660196.0,
      "step": 76070
    },
    {
      "entropy": 5.683567094802856,
      "epoch": 7.55011909487892,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001145960683876672,
      "loss": 4.9204,
      "mean_token_accuracy": 0.20105744600296022,
      "num_tokens": 166671117.0,
      "step": 76075
    },
    {
      "entropy": 5.709818744659424,
      "epoch": 7.550615323541088,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00011457122371267688,
      "loss": 4.9006,
      "mean_token_accuracy": 0.20114085972309112,
      "num_tokens": 166681959.0,
      "step": 76080
    },
    {
      "entropy": 5.745361471176148,
      "epoch": 7.551111552203255,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011454638301597439,
      "loss": 4.99,
      "mean_token_accuracy": 0.1997162953019142,
      "num_tokens": 166691825.0,
      "step": 76085
    },
    {
      "entropy": 5.626144599914551,
      "epoch": 7.551607780865423,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011452154629817596,
      "loss": 4.7664,
      "mean_token_accuracy": 0.21599099040031433,
      "num_tokens": 166701689.0,
      "step": 76090
    },
    {
      "entropy": 5.658944177627563,
      "epoch": 7.55210400952759,
      "grad_norm": 1.265625,
      "learning_rate": 0.00011449671355989736,
      "loss": 4.7872,
      "mean_token_accuracy": 0.21279874593019485,
      "num_tokens": 166712830.0,
      "step": 76095
    },
    {
      "entropy": 5.674096965789795,
      "epoch": 7.552600238189758,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011447188480175443,
      "loss": 4.8515,
      "mean_token_accuracy": 0.21307816207408906,
      "num_tokens": 166722766.0,
      "step": 76100
    },
    {
      "entropy": 5.682039737701416,
      "epoch": 7.553096466851925,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011444706002436276,
      "loss": 4.8981,
      "mean_token_accuracy": 0.20015539079904557,
      "num_tokens": 166734417.0,
      "step": 76105
    },
    {
      "entropy": 5.669711923599243,
      "epoch": 7.553592695514093,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011442223922833812,
      "loss": 4.9219,
      "mean_token_accuracy": 0.2093714416027069,
      "num_tokens": 166746150.0,
      "step": 76110
    },
    {
      "entropy": 5.633124542236328,
      "epoch": 7.55408892417626,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00011439742241429587,
      "loss": 4.7771,
      "mean_token_accuracy": 0.2137183353304863,
      "num_tokens": 166758743.0,
      "step": 76115
    },
    {
      "entropy": 5.655492830276489,
      "epoch": 7.554585152838428,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011437260958285153,
      "loss": 4.8317,
      "mean_token_accuracy": 0.21007696390151978,
      "num_tokens": 166770335.0,
      "step": 76120
    },
    {
      "entropy": 5.691648006439209,
      "epoch": 7.555081381500596,
      "grad_norm": 1.375,
      "learning_rate": 0.00011434780073462038,
      "loss": 4.9256,
      "mean_token_accuracy": 0.20077660083770751,
      "num_tokens": 166782366.0,
      "step": 76125
    },
    {
      "entropy": 5.687848281860352,
      "epoch": 7.5555776101627625,
      "grad_norm": 1.265625,
      "learning_rate": 0.00011432299587021758,
      "loss": 4.866,
      "mean_token_accuracy": 0.2098077192902565,
      "num_tokens": 166793227.0,
      "step": 76130
    },
    {
      "entropy": 5.688910007476807,
      "epoch": 7.556073838824931,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011429819499025824,
      "loss": 4.8696,
      "mean_token_accuracy": 0.20396873056888581,
      "num_tokens": 166804518.0,
      "step": 76135
    },
    {
      "entropy": 5.70071268081665,
      "epoch": 7.556570067487098,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011427339809535733,
      "loss": 4.9007,
      "mean_token_accuracy": 0.19769641757011414,
      "num_tokens": 166815129.0,
      "step": 76140
    },
    {
      "entropy": 5.6906485080719,
      "epoch": 7.557066296149266,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011424860518612981,
      "loss": 4.9188,
      "mean_token_accuracy": 0.20651372969150544,
      "num_tokens": 166826685.0,
      "step": 76145
    },
    {
      "entropy": 5.672217607498169,
      "epoch": 7.557562524811433,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00011422381626319055,
      "loss": 4.8386,
      "mean_token_accuracy": 0.21892546415328978,
      "num_tokens": 166839034.0,
      "step": 76150
    },
    {
      "entropy": 5.688381671905518,
      "epoch": 7.558058753473601,
      "grad_norm": 1.5625,
      "learning_rate": 0.00011419903132715422,
      "loss": 4.9451,
      "mean_token_accuracy": 0.19868804663419723,
      "num_tokens": 166850889.0,
      "step": 76155
    },
    {
      "entropy": 5.675950622558593,
      "epoch": 7.558554982135768,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011417425037863541,
      "loss": 4.8285,
      "mean_token_accuracy": 0.20318952798843384,
      "num_tokens": 166862118.0,
      "step": 76160
    },
    {
      "entropy": 5.6226723194122314,
      "epoch": 7.559051210797936,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011414947341824865,
      "loss": 4.7965,
      "mean_token_accuracy": 0.20683836787939072,
      "num_tokens": 166873279.0,
      "step": 76165
    },
    {
      "entropy": 5.675621461868286,
      "epoch": 7.559547439460103,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011412470044660827,
      "loss": 4.8312,
      "mean_token_accuracy": 0.20718860477209092,
      "num_tokens": 166883665.0,
      "step": 76170
    },
    {
      "entropy": 5.610895681381225,
      "epoch": 7.560043668122271,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011409993146432866,
      "loss": 4.8264,
      "mean_token_accuracy": 0.21116240322589874,
      "num_tokens": 166894603.0,
      "step": 76175
    },
    {
      "entropy": 5.584774541854858,
      "epoch": 7.560539896784438,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011407516647202412,
      "loss": 4.7434,
      "mean_token_accuracy": 0.21314240694046022,
      "num_tokens": 166905538.0,
      "step": 76180
    },
    {
      "entropy": 5.657549047470093,
      "epoch": 7.561036125446606,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011405040547030868,
      "loss": 4.9418,
      "mean_token_accuracy": 0.19565881490707399,
      "num_tokens": 166915875.0,
      "step": 76185
    },
    {
      "entropy": 5.673138856887817,
      "epoch": 7.561532354108773,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011402564845979636,
      "loss": 4.8905,
      "mean_token_accuracy": 0.20308040231466293,
      "num_tokens": 166926791.0,
      "step": 76190
    },
    {
      "entropy": 5.665088510513305,
      "epoch": 7.562028582770941,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001140008954411011,
      "loss": 4.8552,
      "mean_token_accuracy": 0.2115411639213562,
      "num_tokens": 166937618.0,
      "step": 76195
    },
    {
      "entropy": 5.687345790863037,
      "epoch": 7.562524811433108,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011397614641483667,
      "loss": 4.9387,
      "mean_token_accuracy": 0.20132783353328704,
      "num_tokens": 166948578.0,
      "step": 76200
    },
    {
      "entropy": 5.687904310226441,
      "epoch": 7.5630210400952755,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001139514013816168,
      "loss": 4.8975,
      "mean_token_accuracy": 0.1942984566092491,
      "num_tokens": 166959654.0,
      "step": 76205
    },
    {
      "entropy": 5.714930772781372,
      "epoch": 7.563517268757444,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011392666034205521,
      "loss": 4.8912,
      "mean_token_accuracy": 0.19562001824378966,
      "num_tokens": 166970911.0,
      "step": 76210
    },
    {
      "entropy": 5.67230339050293,
      "epoch": 7.5640134974196105,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001139019232967654,
      "loss": 4.7826,
      "mean_token_accuracy": 0.21370242536067963,
      "num_tokens": 166981971.0,
      "step": 76215
    },
    {
      "entropy": 5.618020629882812,
      "epoch": 7.564509726081779,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011387719024636073,
      "loss": 4.7714,
      "mean_token_accuracy": 0.2115048050880432,
      "num_tokens": 166991240.0,
      "step": 76220
    },
    {
      "entropy": 5.748172092437744,
      "epoch": 7.565005954743946,
      "grad_norm": 1.375,
      "learning_rate": 0.00011385246119145455,
      "loss": 5.0053,
      "mean_token_accuracy": 0.1939626917243004,
      "num_tokens": 167002649.0,
      "step": 76225
    },
    {
      "entropy": 5.629250240325928,
      "epoch": 7.565502183406114,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00011382773613266,
      "loss": 4.821,
      "mean_token_accuracy": 0.2020006775856018,
      "num_tokens": 167011972.0,
      "step": 76230
    },
    {
      "entropy": 5.604191160202026,
      "epoch": 7.565998412068281,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001138030150705904,
      "loss": 4.7844,
      "mean_token_accuracy": 0.20826795995235442,
      "num_tokens": 167022282.0,
      "step": 76235
    },
    {
      "entropy": 5.6368725299835205,
      "epoch": 7.566494640730449,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001137782980058586,
      "loss": 4.8531,
      "mean_token_accuracy": 0.20760451704263688,
      "num_tokens": 167033593.0,
      "step": 76240
    },
    {
      "entropy": 5.6930407047271725,
      "epoch": 7.566990869392616,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001137535849390777,
      "loss": 4.9287,
      "mean_token_accuracy": 0.20574960112571716,
      "num_tokens": 167045585.0,
      "step": 76245
    },
    {
      "entropy": 5.656587266921997,
      "epoch": 7.567487098054784,
      "grad_norm": 1.203125,
      "learning_rate": 0.00011372887587086041,
      "loss": 4.8125,
      "mean_token_accuracy": 0.20538480430841446,
      "num_tokens": 167057270.0,
      "step": 76250
    },
    {
      "entropy": 5.68033800125122,
      "epoch": 7.567983326716951,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001137041708018195,
      "loss": 4.8962,
      "mean_token_accuracy": 0.20435210317373276,
      "num_tokens": 167069538.0,
      "step": 76255
    },
    {
      "entropy": 5.609844017028808,
      "epoch": 7.568479555379119,
      "grad_norm": 1.5,
      "learning_rate": 0.00011367946973256752,
      "loss": 4.8015,
      "mean_token_accuracy": 0.2055625945329666,
      "num_tokens": 167080026.0,
      "step": 76260
    },
    {
      "entropy": 5.645824384689331,
      "epoch": 7.568975784041286,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011365477266371714,
      "loss": 4.8837,
      "mean_token_accuracy": 0.205758298933506,
      "num_tokens": 167090456.0,
      "step": 76265
    },
    {
      "entropy": 5.694458389282227,
      "epoch": 7.569472012703454,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011363007959588066,
      "loss": 4.9403,
      "mean_token_accuracy": 0.1990431696176529,
      "num_tokens": 167102681.0,
      "step": 76270
    },
    {
      "entropy": 5.5533966541290285,
      "epoch": 7.569968241365621,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011360539052967057,
      "loss": 4.7293,
      "mean_token_accuracy": 0.2230595752596855,
      "num_tokens": 167113981.0,
      "step": 76275
    },
    {
      "entropy": 5.685199308395386,
      "epoch": 7.5704644700277886,
      "grad_norm": 1.625,
      "learning_rate": 0.000113580705465699,
      "loss": 4.8431,
      "mean_token_accuracy": 0.1972173660993576,
      "num_tokens": 167124455.0,
      "step": 76280
    },
    {
      "entropy": 5.667643785476685,
      "epoch": 7.570960698689956,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011355602440457811,
      "loss": 4.8689,
      "mean_token_accuracy": 0.20403318107128143,
      "num_tokens": 167135783.0,
      "step": 76285
    },
    {
      "entropy": 5.737761497497559,
      "epoch": 7.5714569273521235,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011353134734691987,
      "loss": 4.9256,
      "mean_token_accuracy": 0.20492595434188843,
      "num_tokens": 167147334.0,
      "step": 76290
    },
    {
      "entropy": 5.671522808074951,
      "epoch": 7.571953156014292,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00011350667429333634,
      "loss": 4.8753,
      "mean_token_accuracy": 0.1980350658297539,
      "num_tokens": 167157527.0,
      "step": 76295
    },
    {
      "entropy": 5.683262062072754,
      "epoch": 7.5724493846764585,
      "grad_norm": 1.5,
      "learning_rate": 0.00011348200524443933,
      "loss": 4.8872,
      "mean_token_accuracy": 0.20756829380989075,
      "num_tokens": 167167791.0,
      "step": 76300
    },
    {
      "entropy": 5.703725576400757,
      "epoch": 7.572945613338627,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011345734020084046,
      "loss": 4.9277,
      "mean_token_accuracy": 0.20521149933338165,
      "num_tokens": 167179185.0,
      "step": 76305
    },
    {
      "entropy": 5.714132404327392,
      "epoch": 7.573441842000794,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011343267916315155,
      "loss": 4.9154,
      "mean_token_accuracy": 0.20241351574659347,
      "num_tokens": 167191628.0,
      "step": 76310
    },
    {
      "entropy": 5.663579082489013,
      "epoch": 7.573938070662962,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011340802213198401,
      "loss": 4.8472,
      "mean_token_accuracy": 0.2098069116473198,
      "num_tokens": 167201023.0,
      "step": 76315
    },
    {
      "entropy": 5.686571073532105,
      "epoch": 7.574434299325129,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011338336910794928,
      "loss": 4.8934,
      "mean_token_accuracy": 0.20335784405469895,
      "num_tokens": 167211127.0,
      "step": 76320
    },
    {
      "entropy": 5.725111484527588,
      "epoch": 7.574930527987297,
      "grad_norm": 1.546875,
      "learning_rate": 0.00011335872009165882,
      "loss": 4.9381,
      "mean_token_accuracy": 0.19785815924406053,
      "num_tokens": 167222217.0,
      "step": 76325
    },
    {
      "entropy": 5.6599150657653805,
      "epoch": 7.575426756649464,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011333407508372377,
      "loss": 4.8363,
      "mean_token_accuracy": 0.20180163979530336,
      "num_tokens": 167231887.0,
      "step": 76330
    },
    {
      "entropy": 5.74679012298584,
      "epoch": 7.575922985311632,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011330943408475524,
      "loss": 4.9181,
      "mean_token_accuracy": 0.20371322929859162,
      "num_tokens": 167242973.0,
      "step": 76335
    },
    {
      "entropy": 5.589300346374512,
      "epoch": 7.576419213973799,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011328479709536439,
      "loss": 4.8114,
      "mean_token_accuracy": 0.2008885383605957,
      "num_tokens": 167253572.0,
      "step": 76340
    },
    {
      "entropy": 5.675766229629517,
      "epoch": 7.576915442635967,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011326016411616213,
      "loss": 4.8906,
      "mean_token_accuracy": 0.20185467302799226,
      "num_tokens": 167263905.0,
      "step": 76345
    },
    {
      "entropy": 5.652975082397461,
      "epoch": 7.577411671298134,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011323553514775918,
      "loss": 4.8703,
      "mean_token_accuracy": 0.2025991842150688,
      "num_tokens": 167274848.0,
      "step": 76350
    },
    {
      "entropy": 5.642332077026367,
      "epoch": 7.577907899960302,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001132109101907665,
      "loss": 4.8157,
      "mean_token_accuracy": 0.20956458151340485,
      "num_tokens": 167286644.0,
      "step": 76355
    },
    {
      "entropy": 5.736207389831543,
      "epoch": 7.578404128622469,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001131862892457946,
      "loss": 4.9501,
      "mean_token_accuracy": 0.1944169208407402,
      "num_tokens": 167296967.0,
      "step": 76360
    },
    {
      "entropy": 5.69176926612854,
      "epoch": 7.5789003572846365,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011316167231345405,
      "loss": 4.8774,
      "mean_token_accuracy": 0.20747696310281755,
      "num_tokens": 167308659.0,
      "step": 76365
    },
    {
      "entropy": 5.700058078765869,
      "epoch": 7.579396585946804,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011313705939435522,
      "loss": 4.947,
      "mean_token_accuracy": 0.19936044216156007,
      "num_tokens": 167319747.0,
      "step": 76370
    },
    {
      "entropy": 5.69825963973999,
      "epoch": 7.5798928146089715,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011311245048910862,
      "loss": 4.8851,
      "mean_token_accuracy": 0.20550878643989562,
      "num_tokens": 167331274.0,
      "step": 76375
    },
    {
      "entropy": 5.7043780326843265,
      "epoch": 7.58038904327114,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011308784559832431,
      "loss": 4.8951,
      "mean_token_accuracy": 0.2002117559313774,
      "num_tokens": 167340942.0,
      "step": 76380
    },
    {
      "entropy": 5.66337947845459,
      "epoch": 7.580885271933306,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011306324472261263,
      "loss": 4.8345,
      "mean_token_accuracy": 0.2100301966071129,
      "num_tokens": 167351816.0,
      "step": 76385
    },
    {
      "entropy": 5.649212646484375,
      "epoch": 7.581381500595475,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011303864786258354,
      "loss": 4.8261,
      "mean_token_accuracy": 0.213975490629673,
      "num_tokens": 167361676.0,
      "step": 76390
    },
    {
      "entropy": 5.719654130935669,
      "epoch": 7.581877729257642,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011301405501884698,
      "loss": 4.9639,
      "mean_token_accuracy": 0.2038324922323227,
      "num_tokens": 167373399.0,
      "step": 76395
    },
    {
      "entropy": 5.7077216625213625,
      "epoch": 7.58237395791981,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011298946619201277,
      "loss": 4.9158,
      "mean_token_accuracy": 0.20252382308244704,
      "num_tokens": 167383427.0,
      "step": 76400
    },
    {
      "entropy": 5.651013517379761,
      "epoch": 7.582870186581977,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011296488138269064,
      "loss": 4.8422,
      "mean_token_accuracy": 0.205696040391922,
      "num_tokens": 167392967.0,
      "step": 76405
    },
    {
      "entropy": 5.680110120773316,
      "epoch": 7.583366415244145,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011294030059149039,
      "loss": 4.875,
      "mean_token_accuracy": 0.2060295820236206,
      "num_tokens": 167403267.0,
      "step": 76410
    },
    {
      "entropy": 5.7023231983184814,
      "epoch": 7.583862643906312,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001129157238190215,
      "loss": 4.8758,
      "mean_token_accuracy": 0.20562190115451812,
      "num_tokens": 167414713.0,
      "step": 76415
    },
    {
      "entropy": 5.671472120285034,
      "epoch": 7.58435887256848,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011289115106589337,
      "loss": 4.8595,
      "mean_token_accuracy": 0.20559551566839218,
      "num_tokens": 167425491.0,
      "step": 76420
    },
    {
      "entropy": 5.710583686828613,
      "epoch": 7.584855101230647,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001128665823327154,
      "loss": 4.8638,
      "mean_token_accuracy": 0.20871487408876419,
      "num_tokens": 167437141.0,
      "step": 76425
    },
    {
      "entropy": 5.7002678394317625,
      "epoch": 7.585351329892815,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011284201762009674,
      "loss": 4.9095,
      "mean_token_accuracy": 0.20478406846523284,
      "num_tokens": 167448568.0,
      "step": 76430
    },
    {
      "entropy": 5.737886714935303,
      "epoch": 7.585847558554982,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011281745692864671,
      "loss": 4.95,
      "mean_token_accuracy": 0.1946697324514389,
      "num_tokens": 167459921.0,
      "step": 76435
    },
    {
      "entropy": 5.698335790634156,
      "epoch": 7.5863437872171495,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00011279290025897417,
      "loss": 4.9083,
      "mean_token_accuracy": 0.20096564739942552,
      "num_tokens": 167471981.0,
      "step": 76440
    },
    {
      "entropy": 5.650956773757935,
      "epoch": 7.586840015879317,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011276834761168828,
      "loss": 4.9175,
      "mean_token_accuracy": 0.203070704638958,
      "num_tokens": 167483055.0,
      "step": 76445
    },
    {
      "entropy": 5.679297351837159,
      "epoch": 7.5873362445414845,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011274379898739777,
      "loss": 4.8792,
      "mean_token_accuracy": 0.20908067226409913,
      "num_tokens": 167494918.0,
      "step": 76450
    },
    {
      "entropy": 5.752637767791748,
      "epoch": 7.587832473203652,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011271925438671145,
      "loss": 4.9639,
      "mean_token_accuracy": 0.20167342722415924,
      "num_tokens": 167505880.0,
      "step": 76455
    },
    {
      "entropy": 5.734198331832886,
      "epoch": 7.588328701865819,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011269471381023789,
      "loss": 4.9062,
      "mean_token_accuracy": 0.2054770678281784,
      "num_tokens": 167516341.0,
      "step": 76460
    },
    {
      "entropy": 5.778183698654175,
      "epoch": 7.588824930527988,
      "grad_norm": 1.5,
      "learning_rate": 0.00011267017725858566,
      "loss": 4.9762,
      "mean_token_accuracy": 0.19615064114332198,
      "num_tokens": 167526141.0,
      "step": 76465
    },
    {
      "entropy": 5.637139463424683,
      "epoch": 7.589321159190154,
      "grad_norm": 1.296875,
      "learning_rate": 0.00011264564473236323,
      "loss": 4.8468,
      "mean_token_accuracy": 0.20207832604646683,
      "num_tokens": 167538163.0,
      "step": 76470
    },
    {
      "entropy": 5.7132378101348875,
      "epoch": 7.589817387852323,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011262111623217904,
      "loss": 4.9399,
      "mean_token_accuracy": 0.1996671050786972,
      "num_tokens": 167547965.0,
      "step": 76475
    },
    {
      "entropy": 5.695276165008545,
      "epoch": 7.59031361651449,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011259659175864127,
      "loss": 4.8116,
      "mean_token_accuracy": 0.20862601697444916,
      "num_tokens": 167558244.0,
      "step": 76480
    },
    {
      "entropy": 5.712924957275391,
      "epoch": 7.590809845176658,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011257207131235806,
      "loss": 4.8822,
      "mean_token_accuracy": 0.20096946954727174,
      "num_tokens": 167569232.0,
      "step": 76485
    },
    {
      "entropy": 5.642010021209717,
      "epoch": 7.591306073838825,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00011254755489393749,
      "loss": 4.819,
      "mean_token_accuracy": 0.2081494614481926,
      "num_tokens": 167580347.0,
      "step": 76490
    },
    {
      "entropy": 5.698563861846924,
      "epoch": 7.591802302500993,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011252304250398744,
      "loss": 4.9258,
      "mean_token_accuracy": 0.2054936558008194,
      "num_tokens": 167591221.0,
      "step": 76495
    },
    {
      "entropy": 5.673537874221802,
      "epoch": 7.59229853116316,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011249853414311584,
      "loss": 4.8593,
      "mean_token_accuracy": 0.2068338632583618,
      "num_tokens": 167601452.0,
      "step": 76500
    },
    {
      "entropy": 5.7356924533844,
      "epoch": 7.592794759825328,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011247402981193049,
      "loss": 4.9305,
      "mean_token_accuracy": 0.20158857703208924,
      "num_tokens": 167611968.0,
      "step": 76505
    },
    {
      "entropy": 5.696911859512329,
      "epoch": 7.593290988487495,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011244952951103899,
      "loss": 4.8393,
      "mean_token_accuracy": 0.20759871900081633,
      "num_tokens": 167621427.0,
      "step": 76510
    },
    {
      "entropy": 5.706432676315307,
      "epoch": 7.5937872171496625,
      "grad_norm": 1.375,
      "learning_rate": 0.00011242503324104888,
      "loss": 4.8808,
      "mean_token_accuracy": 0.2103072836995125,
      "num_tokens": 167632865.0,
      "step": 76515
    },
    {
      "entropy": 5.700532913208008,
      "epoch": 7.59428344581183,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011240054100256765,
      "loss": 4.887,
      "mean_token_accuracy": 0.2023894101381302,
      "num_tokens": 167643082.0,
      "step": 76520
    },
    {
      "entropy": 5.686709022521972,
      "epoch": 7.5947796744739975,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011237605279620256,
      "loss": 4.8864,
      "mean_token_accuracy": 0.2036646217107773,
      "num_tokens": 167654230.0,
      "step": 76525
    },
    {
      "entropy": 5.659353351593017,
      "epoch": 7.595275903136165,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011235156862256099,
      "loss": 4.8549,
      "mean_token_accuracy": 0.205189149081707,
      "num_tokens": 167665001.0,
      "step": 76530
    },
    {
      "entropy": 5.655055665969849,
      "epoch": 7.595772131798332,
      "grad_norm": 1.375,
      "learning_rate": 0.00011232708848224998,
      "loss": 4.8597,
      "mean_token_accuracy": 0.20702177435159683,
      "num_tokens": 167676433.0,
      "step": 76535
    },
    {
      "entropy": 5.75235333442688,
      "epoch": 7.5962683604605,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011230261237587672,
      "loss": 4.9381,
      "mean_token_accuracy": 0.20039257109165193,
      "num_tokens": 167687068.0,
      "step": 76540
    },
    {
      "entropy": 5.638263177871704,
      "epoch": 7.596764589122667,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001122781403040481,
      "loss": 4.8503,
      "mean_token_accuracy": 0.2164044424891472,
      "num_tokens": 167698723.0,
      "step": 76545
    },
    {
      "entropy": 5.65754804611206,
      "epoch": 7.597260817784836,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011225367226737093,
      "loss": 4.8943,
      "mean_token_accuracy": 0.20324359685182572,
      "num_tokens": 167709699.0,
      "step": 76550
    },
    {
      "entropy": 5.73249454498291,
      "epoch": 7.597757046447002,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011222920826645198,
      "loss": 4.9392,
      "mean_token_accuracy": 0.19228523522615432,
      "num_tokens": 167721104.0,
      "step": 76555
    },
    {
      "entropy": 5.634122037887574,
      "epoch": 7.598253275109171,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011220474830189797,
      "loss": 4.8131,
      "mean_token_accuracy": 0.20773085802793503,
      "num_tokens": 167731887.0,
      "step": 76560
    },
    {
      "entropy": 5.689618921279907,
      "epoch": 7.598749503771338,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011218029237431535,
      "loss": 4.9089,
      "mean_token_accuracy": 0.20045387148857116,
      "num_tokens": 167742591.0,
      "step": 76565
    },
    {
      "entropy": 5.660115146636963,
      "epoch": 7.599245732433506,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00011215584048431067,
      "loss": 4.8785,
      "mean_token_accuracy": 0.2016080990433693,
      "num_tokens": 167755069.0,
      "step": 76570
    },
    {
      "entropy": 5.669845724105835,
      "epoch": 7.599741961095673,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011213139263249028,
      "loss": 4.8934,
      "mean_token_accuracy": 0.20718832165002823,
      "num_tokens": 167765647.0,
      "step": 76575
    },
    {
      "entropy": 5.633862161636353,
      "epoch": 7.600238189757841,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001121069488194604,
      "loss": 4.8109,
      "mean_token_accuracy": 0.21278467625379563,
      "num_tokens": 167776631.0,
      "step": 76580
    },
    {
      "entropy": 5.654821062088013,
      "epoch": 7.600734418420008,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011208250904582711,
      "loss": 4.852,
      "mean_token_accuracy": 0.206762857735157,
      "num_tokens": 167787309.0,
      "step": 76585
    },
    {
      "entropy": 5.654457759857178,
      "epoch": 7.6012306470821756,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001120580733121966,
      "loss": 4.8969,
      "mean_token_accuracy": 0.2033349797129631,
      "num_tokens": 167799333.0,
      "step": 76590
    },
    {
      "entropy": 5.701197385787964,
      "epoch": 7.601726875744343,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011203364161917479,
      "loss": 4.9246,
      "mean_token_accuracy": 0.1997830882668495,
      "num_tokens": 167811236.0,
      "step": 76595
    },
    {
      "entropy": 5.7030809879302975,
      "epoch": 7.6022231044065105,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001120092139673674,
      "loss": 4.9419,
      "mean_token_accuracy": 0.19726648032665253,
      "num_tokens": 167822756.0,
      "step": 76600
    },
    {
      "entropy": 5.7039247989654545,
      "epoch": 7.602719333068678,
      "grad_norm": 1.375,
      "learning_rate": 0.00011198479035738039,
      "loss": 4.9124,
      "mean_token_accuracy": 0.20109868198633193,
      "num_tokens": 167833208.0,
      "step": 76605
    },
    {
      "entropy": 5.7047521591186525,
      "epoch": 7.6032155617308455,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001119603707898193,
      "loss": 4.9147,
      "mean_token_accuracy": 0.19987615048885346,
      "num_tokens": 167845422.0,
      "step": 76610
    },
    {
      "entropy": 5.755177021026611,
      "epoch": 7.603711790393013,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00011193595526528963,
      "loss": 4.9891,
      "mean_token_accuracy": 0.19817183613777162,
      "num_tokens": 167857631.0,
      "step": 76615
    },
    {
      "entropy": 5.720182323455811,
      "epoch": 7.60420801905518,
      "grad_norm": 1.375,
      "learning_rate": 0.00011191154378439697,
      "loss": 4.9157,
      "mean_token_accuracy": 0.2002674847841263,
      "num_tokens": 167868688.0,
      "step": 76620
    },
    {
      "entropy": 5.728912115097046,
      "epoch": 7.604704247717348,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011188713634774658,
      "loss": 4.9106,
      "mean_token_accuracy": 0.20193783044815064,
      "num_tokens": 167879553.0,
      "step": 76625
    },
    {
      "entropy": 5.743295907974243,
      "epoch": 7.605200476379515,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011186273295594368,
      "loss": 4.9007,
      "mean_token_accuracy": 0.19591898769140242,
      "num_tokens": 167890036.0,
      "step": 76630
    },
    {
      "entropy": 5.7378363609313965,
      "epoch": 7.605696705041683,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011183833360959355,
      "loss": 4.8974,
      "mean_token_accuracy": 0.21103255897760392,
      "num_tokens": 167900255.0,
      "step": 76635
    },
    {
      "entropy": 5.642710256576538,
      "epoch": 7.60619293370385,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011181393830930106,
      "loss": 4.836,
      "mean_token_accuracy": 0.19964870810508728,
      "num_tokens": 167911572.0,
      "step": 76640
    },
    {
      "entropy": 5.761328458786011,
      "epoch": 7.606689162366019,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011178954705567136,
      "loss": 5.0019,
      "mean_token_accuracy": 0.1976209133863449,
      "num_tokens": 167922224.0,
      "step": 76645
    },
    {
      "entropy": 5.644684505462647,
      "epoch": 7.607185391028186,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011176515984930922,
      "loss": 4.8531,
      "mean_token_accuracy": 0.20860926806926727,
      "num_tokens": 167932977.0,
      "step": 76650
    },
    {
      "entropy": 5.6695719242095945,
      "epoch": 7.607681619690354,
      "grad_norm": 1.1875,
      "learning_rate": 0.00011174077669081936,
      "loss": 4.8934,
      "mean_token_accuracy": 0.20007054954767228,
      "num_tokens": 167944188.0,
      "step": 76655
    },
    {
      "entropy": 5.66003909111023,
      "epoch": 7.608177848352521,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011171639758080637,
      "loss": 4.8382,
      "mean_token_accuracy": 0.20665666610002517,
      "num_tokens": 167955584.0,
      "step": 76660
    },
    {
      "entropy": 5.742769765853882,
      "epoch": 7.608674077014689,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00011169202251987495,
      "loss": 4.9532,
      "mean_token_accuracy": 0.20726526230573655,
      "num_tokens": 167966638.0,
      "step": 76665
    },
    {
      "entropy": 5.7312439441680905,
      "epoch": 7.609170305676856,
      "grad_norm": 1.375,
      "learning_rate": 0.00011166765150862942,
      "loss": 4.9231,
      "mean_token_accuracy": 0.1980299547314644,
      "num_tokens": 167977648.0,
      "step": 76670
    },
    {
      "entropy": 5.728835248947144,
      "epoch": 7.6096665343390235,
      "grad_norm": 1.234375,
      "learning_rate": 0.00011164328454767423,
      "loss": 4.8736,
      "mean_token_accuracy": 0.2099199652671814,
      "num_tokens": 167989380.0,
      "step": 76675
    },
    {
      "entropy": 5.714833927154541,
      "epoch": 7.610162763001191,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011161892163761359,
      "loss": 4.8917,
      "mean_token_accuracy": 0.20309755057096482,
      "num_tokens": 167999793.0,
      "step": 76680
    },
    {
      "entropy": 5.643592739105225,
      "epoch": 7.6106589916633585,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011159456277905163,
      "loss": 4.8174,
      "mean_token_accuracy": 0.2093827724456787,
      "num_tokens": 168011425.0,
      "step": 76685
    },
    {
      "entropy": 5.633255290985107,
      "epoch": 7.611155220325526,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001115702079725924,
      "loss": 4.8236,
      "mean_token_accuracy": 0.21900554448366166,
      "num_tokens": 168022323.0,
      "step": 76690
    },
    {
      "entropy": 5.689623403549194,
      "epoch": 7.611651448987693,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001115458572188398,
      "loss": 4.8774,
      "mean_token_accuracy": 0.20473720729351044,
      "num_tokens": 168033508.0,
      "step": 76695
    },
    {
      "entropy": 5.730416965484619,
      "epoch": 7.612147677649861,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011152151051839772,
      "loss": 4.9677,
      "mean_token_accuracy": 0.19446810632944106,
      "num_tokens": 168043320.0,
      "step": 76700
    },
    {
      "entropy": 5.715602016448974,
      "epoch": 7.612643906312028,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011149716787186999,
      "loss": 4.9192,
      "mean_token_accuracy": 0.20552789717912673,
      "num_tokens": 168053408.0,
      "step": 76705
    },
    {
      "entropy": 5.692480230331421,
      "epoch": 7.613140134974196,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011147282927986017,
      "loss": 4.8468,
      "mean_token_accuracy": 0.20840200781822205,
      "num_tokens": 168064424.0,
      "step": 76710
    },
    {
      "entropy": 5.691763877868652,
      "epoch": 7.613636363636363,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011144849474297183,
      "loss": 4.8727,
      "mean_token_accuracy": 0.20124118030071259,
      "num_tokens": 168074703.0,
      "step": 76715
    },
    {
      "entropy": 5.720054388046265,
      "epoch": 7.614132592298531,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011142416426180837,
      "loss": 4.864,
      "mean_token_accuracy": 0.20893464386463165,
      "num_tokens": 168084703.0,
      "step": 76720
    },
    {
      "entropy": 5.699337196350098,
      "epoch": 7.614628820960698,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011139983783697313,
      "loss": 4.88,
      "mean_token_accuracy": 0.19862100034952163,
      "num_tokens": 168095315.0,
      "step": 76725
    },
    {
      "entropy": 5.669798183441162,
      "epoch": 7.615125049622867,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011137551546906936,
      "loss": 4.9299,
      "mean_token_accuracy": 0.19789621531963347,
      "num_tokens": 168105462.0,
      "step": 76730
    },
    {
      "entropy": 5.677554655075073,
      "epoch": 7.615621278285034,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00011135119715870032,
      "loss": 4.8436,
      "mean_token_accuracy": 0.20403201431035994,
      "num_tokens": 168115619.0,
      "step": 76735
    },
    {
      "entropy": 5.665165042877197,
      "epoch": 7.616117506947202,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011132688290646892,
      "loss": 4.83,
      "mean_token_accuracy": 0.20743029564619064,
      "num_tokens": 168125482.0,
      "step": 76740
    },
    {
      "entropy": 5.672534704208374,
      "epoch": 7.616613735609369,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011130257271297817,
      "loss": 4.8897,
      "mean_token_accuracy": 0.2044217213988304,
      "num_tokens": 168135743.0,
      "step": 76745
    },
    {
      "entropy": 5.647317695617676,
      "epoch": 7.6171099642715365,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011127826657883086,
      "loss": 4.83,
      "mean_token_accuracy": 0.20653105080127715,
      "num_tokens": 168145435.0,
      "step": 76750
    },
    {
      "entropy": 5.626562452316284,
      "epoch": 7.617606192933704,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00011125396450462977,
      "loss": 4.8046,
      "mean_token_accuracy": 0.20566453784704208,
      "num_tokens": 168155258.0,
      "step": 76755
    },
    {
      "entropy": 5.637254667282105,
      "epoch": 7.6181024215958715,
      "grad_norm": 1.59375,
      "learning_rate": 0.00011122966649097743,
      "loss": 4.8589,
      "mean_token_accuracy": 0.19852512180805207,
      "num_tokens": 168165052.0,
      "step": 76760
    },
    {
      "entropy": 5.742812538146973,
      "epoch": 7.618598650258039,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011120537253847648,
      "loss": 5.0079,
      "mean_token_accuracy": 0.19182489067316055,
      "num_tokens": 168175969.0,
      "step": 76765
    },
    {
      "entropy": 5.674246692657471,
      "epoch": 7.619094878920206,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011118108264772942,
      "loss": 4.8897,
      "mean_token_accuracy": 0.20610811412334443,
      "num_tokens": 168186049.0,
      "step": 76770
    },
    {
      "entropy": 5.693433809280395,
      "epoch": 7.619591107582374,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011115679681933849,
      "loss": 4.8971,
      "mean_token_accuracy": 0.19851886481046677,
      "num_tokens": 168196891.0,
      "step": 76775
    },
    {
      "entropy": 5.706484317779541,
      "epoch": 7.620087336244541,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011113251505390597,
      "loss": 4.9157,
      "mean_token_accuracy": 0.1962538868188858,
      "num_tokens": 168207219.0,
      "step": 76780
    },
    {
      "entropy": 5.747113180160523,
      "epoch": 7.620583564906709,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011110823735203396,
      "loss": 4.9187,
      "mean_token_accuracy": 0.20470038652420045,
      "num_tokens": 168219429.0,
      "step": 76785
    },
    {
      "entropy": 5.675348091125488,
      "epoch": 7.621079793568876,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011108396371432444,
      "loss": 4.7981,
      "mean_token_accuracy": 0.20776496678590775,
      "num_tokens": 168229462.0,
      "step": 76790
    },
    {
      "entropy": 5.7212167263031,
      "epoch": 7.621576022231044,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011105969414137941,
      "loss": 4.8976,
      "mean_token_accuracy": 0.2024093121290207,
      "num_tokens": 168240230.0,
      "step": 76795
    },
    {
      "entropy": 5.611751079559326,
      "epoch": 7.622072250893211,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011103542863380079,
      "loss": 4.823,
      "mean_token_accuracy": 0.2076630711555481,
      "num_tokens": 168251744.0,
      "step": 76800
    },
    {
      "entropy": 5.65147819519043,
      "epoch": 7.622568479555379,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001110111671921902,
      "loss": 4.9073,
      "mean_token_accuracy": 0.1974483072757721,
      "num_tokens": 168262762.0,
      "step": 76805
    },
    {
      "entropy": 5.665113496780395,
      "epoch": 7.623064708217546,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00011098690981714932,
      "loss": 4.8868,
      "mean_token_accuracy": 0.2055704638361931,
      "num_tokens": 168274693.0,
      "step": 76810
    },
    {
      "entropy": 5.740192651748657,
      "epoch": 7.623560936879715,
      "grad_norm": 1.671875,
      "learning_rate": 0.00011096265650927966,
      "loss": 4.9284,
      "mean_token_accuracy": 0.1967868134379387,
      "num_tokens": 168285698.0,
      "step": 76815
    },
    {
      "entropy": 5.65971007347107,
      "epoch": 7.624057165541882,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001109384072691826,
      "loss": 4.8778,
      "mean_token_accuracy": 0.20653839111328126,
      "num_tokens": 168297670.0,
      "step": 76820
    },
    {
      "entropy": 5.665949869155884,
      "epoch": 7.6245533942040495,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00011091416209745954,
      "loss": 4.914,
      "mean_token_accuracy": 0.2009500429034233,
      "num_tokens": 168307707.0,
      "step": 76825
    },
    {
      "entropy": 5.700507545471192,
      "epoch": 7.625049622866217,
      "grad_norm": 1.234375,
      "learning_rate": 0.00011088992099471165,
      "loss": 4.8779,
      "mean_token_accuracy": 0.20423811972141265,
      "num_tokens": 168319538.0,
      "step": 76830
    },
    {
      "entropy": 5.7080244541168215,
      "epoch": 7.6255458515283845,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011086568396154014,
      "loss": 4.9144,
      "mean_token_accuracy": 0.1993437424302101,
      "num_tokens": 168331355.0,
      "step": 76835
    },
    {
      "entropy": 5.68044080734253,
      "epoch": 7.626042080190552,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011084145099854601,
      "loss": 4.8407,
      "mean_token_accuracy": 0.19874810576438903,
      "num_tokens": 168341578.0,
      "step": 76840
    },
    {
      "entropy": 5.740547704696655,
      "epoch": 7.626538308852719,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011081722210633017,
      "loss": 4.94,
      "mean_token_accuracy": 0.20531985610723497,
      "num_tokens": 168353331.0,
      "step": 76845
    },
    {
      "entropy": 5.660848712921142,
      "epoch": 7.627034537514887,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011079299728549336,
      "loss": 4.8431,
      "mean_token_accuracy": 0.2031945616006851,
      "num_tokens": 168365663.0,
      "step": 76850
    },
    {
      "entropy": 5.6645365238189695,
      "epoch": 7.627530766177054,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011076877653663645,
      "loss": 4.9008,
      "mean_token_accuracy": 0.19977969229221343,
      "num_tokens": 168378107.0,
      "step": 76855
    },
    {
      "entropy": 5.666017627716064,
      "epoch": 7.628026994839222,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011074455986035992,
      "loss": 4.8269,
      "mean_token_accuracy": 0.21281147748231888,
      "num_tokens": 168389553.0,
      "step": 76860
    },
    {
      "entropy": 5.671633625030518,
      "epoch": 7.628523223501389,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011072034725726443,
      "loss": 4.8299,
      "mean_token_accuracy": 0.20735543072223664,
      "num_tokens": 168399447.0,
      "step": 76865
    },
    {
      "entropy": 5.677403306961059,
      "epoch": 7.629019452163557,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011069613872795036,
      "loss": 4.8735,
      "mean_token_accuracy": 0.20175670236349105,
      "num_tokens": 168409216.0,
      "step": 76870
    },
    {
      "entropy": 5.626704359054566,
      "epoch": 7.629515680825724,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000110671934273018,
      "loss": 4.7744,
      "mean_token_accuracy": 0.20960666835308076,
      "num_tokens": 168418832.0,
      "step": 76875
    },
    {
      "entropy": 5.6649645328521725,
      "epoch": 7.630011909487892,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001106477338930675,
      "loss": 4.8552,
      "mean_token_accuracy": 0.2013376846909523,
      "num_tokens": 168429677.0,
      "step": 76880
    },
    {
      "entropy": 5.655159187316895,
      "epoch": 7.630508138150059,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011062353758869912,
      "loss": 4.9048,
      "mean_token_accuracy": 0.19804464280605316,
      "num_tokens": 168440029.0,
      "step": 76885
    },
    {
      "entropy": 5.689464092254639,
      "epoch": 7.631004366812227,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001105993453605128,
      "loss": 5.0042,
      "mean_token_accuracy": 0.19650670737028123,
      "num_tokens": 168449982.0,
      "step": 76890
    },
    {
      "entropy": 5.71494426727295,
      "epoch": 7.631500595474394,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011057515720910842,
      "loss": 4.937,
      "mean_token_accuracy": 0.20364655554294586,
      "num_tokens": 168460529.0,
      "step": 76895
    },
    {
      "entropy": 5.6762903213500975,
      "epoch": 7.6319968241365626,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011055097313508589,
      "loss": 4.8749,
      "mean_token_accuracy": 0.20062058717012404,
      "num_tokens": 168472034.0,
      "step": 76900
    },
    {
      "entropy": 5.638419198989868,
      "epoch": 7.63249305279873,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011052679313904479,
      "loss": 4.8205,
      "mean_token_accuracy": 0.20901595056056976,
      "num_tokens": 168483416.0,
      "step": 76905
    },
    {
      "entropy": 5.726557636260987,
      "epoch": 7.6329892814608975,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001105026172215849,
      "loss": 4.8883,
      "mean_token_accuracy": 0.19784980416297912,
      "num_tokens": 168494256.0,
      "step": 76910
    },
    {
      "entropy": 5.632987022399902,
      "epoch": 7.633485510123065,
      "grad_norm": 1.234375,
      "learning_rate": 0.00011047844538330562,
      "loss": 4.8778,
      "mean_token_accuracy": 0.20442206859588624,
      "num_tokens": 168505634.0,
      "step": 76915
    },
    {
      "entropy": 5.68539252281189,
      "epoch": 7.6339817387852325,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011045427762480639,
      "loss": 4.927,
      "mean_token_accuracy": 0.20112885236740113,
      "num_tokens": 168518891.0,
      "step": 76920
    },
    {
      "entropy": 5.723118734359741,
      "epoch": 7.6344779674474,
      "grad_norm": 1.578125,
      "learning_rate": 0.00011043011394668645,
      "loss": 4.9845,
      "mean_token_accuracy": 0.19821908175945283,
      "num_tokens": 168529547.0,
      "step": 76925
    },
    {
      "entropy": 5.64833550453186,
      "epoch": 7.634974196109567,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011040595434954515,
      "loss": 4.8038,
      "mean_token_accuracy": 0.20593770295381547,
      "num_tokens": 168539973.0,
      "step": 76930
    },
    {
      "entropy": 5.674950122833252,
      "epoch": 7.635470424771735,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001103817988339814,
      "loss": 4.8282,
      "mean_token_accuracy": 0.20827759057283401,
      "num_tokens": 168550951.0,
      "step": 76935
    },
    {
      "entropy": 5.725645065307617,
      "epoch": 7.635966653433902,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011035764740059445,
      "loss": 4.9713,
      "mean_token_accuracy": 0.19833364933729172,
      "num_tokens": 168562351.0,
      "step": 76940
    },
    {
      "entropy": 5.692497777938843,
      "epoch": 7.63646288209607,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011033350004998302,
      "loss": 4.9386,
      "mean_token_accuracy": 0.20670733451843262,
      "num_tokens": 168573006.0,
      "step": 76945
    },
    {
      "entropy": 5.651315498352051,
      "epoch": 7.636959110758237,
      "grad_norm": 1.375,
      "learning_rate": 0.00011030935678274597,
      "loss": 4.8859,
      "mean_token_accuracy": 0.20371061861515044,
      "num_tokens": 168584080.0,
      "step": 76950
    },
    {
      "entropy": 5.729304599761963,
      "epoch": 7.637455339420405,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011028521759948195,
      "loss": 4.8713,
      "mean_token_accuracy": 0.2041396304965019,
      "num_tokens": 168594915.0,
      "step": 76955
    },
    {
      "entropy": 5.743035840988159,
      "epoch": 7.637951568082572,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011026108250078964,
      "loss": 4.9889,
      "mean_token_accuracy": 0.1975735381245613,
      "num_tokens": 168606682.0,
      "step": 76960
    },
    {
      "entropy": 5.743520975112915,
      "epoch": 7.63844779674474,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011023695148726747,
      "loss": 4.9831,
      "mean_token_accuracy": 0.19672565907239914,
      "num_tokens": 168618816.0,
      "step": 76965
    },
    {
      "entropy": 5.71779613494873,
      "epoch": 7.638944025406907,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011021282455951392,
      "loss": 4.8505,
      "mean_token_accuracy": 0.20963656306266784,
      "num_tokens": 168631507.0,
      "step": 76970
    },
    {
      "entropy": 5.6503297805786135,
      "epoch": 7.639440254069075,
      "grad_norm": 1.28125,
      "learning_rate": 0.00011018870171812723,
      "loss": 4.8428,
      "mean_token_accuracy": 0.20772385448217393,
      "num_tokens": 168643108.0,
      "step": 76975
    },
    {
      "entropy": 5.695980501174927,
      "epoch": 7.639936482731242,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011016458296370563,
      "loss": 4.828,
      "mean_token_accuracy": 0.2094130665063858,
      "num_tokens": 168654042.0,
      "step": 76980
    },
    {
      "entropy": 5.692969036102295,
      "epoch": 7.6404327113934105,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011014046829684715,
      "loss": 4.8853,
      "mean_token_accuracy": 0.20359814763069153,
      "num_tokens": 168664305.0,
      "step": 76985
    },
    {
      "entropy": 5.716560173034668,
      "epoch": 7.640928940055578,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011011635771814977,
      "loss": 4.9031,
      "mean_token_accuracy": 0.19622959643602372,
      "num_tokens": 168676383.0,
      "step": 76990
    },
    {
      "entropy": 5.691017961502075,
      "epoch": 7.6414251687177455,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001100922512282114,
      "loss": 4.9084,
      "mean_token_accuracy": 0.2058485820889473,
      "num_tokens": 168688409.0,
      "step": 76995
    },
    {
      "entropy": 5.744367694854736,
      "epoch": 7.641921397379913,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011006814882762995,
      "loss": 4.9086,
      "mean_token_accuracy": 0.2079392448067665,
      "num_tokens": 168699230.0,
      "step": 77000
    },
    {
      "entropy": 5.723339080810547,
      "epoch": 7.64241762604208,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011004405051700301,
      "loss": 4.9474,
      "mean_token_accuracy": 0.1942865237593651,
      "num_tokens": 168710569.0,
      "step": 77005
    },
    {
      "entropy": 5.736195611953735,
      "epoch": 7.642913854704248,
      "grad_norm": 1.296875,
      "learning_rate": 0.00011001995629692815,
      "loss": 4.9034,
      "mean_token_accuracy": 0.20823552459478378,
      "num_tokens": 168720919.0,
      "step": 77010
    },
    {
      "entropy": 5.6398828506469725,
      "epoch": 7.643410083366415,
      "grad_norm": 1.546875,
      "learning_rate": 0.00010999586616800289,
      "loss": 4.8428,
      "mean_token_accuracy": 0.21328492611646652,
      "num_tokens": 168731140.0,
      "step": 77015
    },
    {
      "entropy": 5.669364166259766,
      "epoch": 7.643906312028583,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001099717801308245,
      "loss": 4.8174,
      "mean_token_accuracy": 0.20991309434175492,
      "num_tokens": 168743455.0,
      "step": 77020
    },
    {
      "entropy": 5.717079448699951,
      "epoch": 7.64440254069075,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001099476981859904,
      "loss": 4.9455,
      "mean_token_accuracy": 0.1984115183353424,
      "num_tokens": 168755383.0,
      "step": 77025
    },
    {
      "entropy": 5.669408798217773,
      "epoch": 7.644898769352918,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010992362033409778,
      "loss": 4.8517,
      "mean_token_accuracy": 0.2023073434829712,
      "num_tokens": 168766996.0,
      "step": 77030
    },
    {
      "entropy": 5.621461677551269,
      "epoch": 7.645394998015085,
      "grad_norm": 1.328125,
      "learning_rate": 0.00010989954657574366,
      "loss": 4.8264,
      "mean_token_accuracy": 0.2127937853336334,
      "num_tokens": 168778087.0,
      "step": 77035
    },
    {
      "entropy": 5.7250738620758055,
      "epoch": 7.645891226677253,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010987547691152506,
      "loss": 4.8829,
      "mean_token_accuracy": 0.20498938411474227,
      "num_tokens": 168789056.0,
      "step": 77040
    },
    {
      "entropy": 5.669749164581299,
      "epoch": 7.64638745533942,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00010985141134203878,
      "loss": 4.8634,
      "mean_token_accuracy": 0.20724946409463882,
      "num_tokens": 168800488.0,
      "step": 77045
    },
    {
      "entropy": 5.714896392822266,
      "epoch": 7.646883684001588,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010982734986788162,
      "loss": 4.9993,
      "mean_token_accuracy": 0.1967264473438263,
      "num_tokens": 168813223.0,
      "step": 77050
    },
    {
      "entropy": 5.665520811080933,
      "epoch": 7.647379912663755,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00010980329248965024,
      "loss": 4.8459,
      "mean_token_accuracy": 0.20897488296031952,
      "num_tokens": 168823478.0,
      "step": 77055
    },
    {
      "entropy": 5.649406814575196,
      "epoch": 7.647876141325923,
      "grad_norm": 1.375,
      "learning_rate": 0.00010977923920794122,
      "loss": 4.8251,
      "mean_token_accuracy": 0.21125909686088562,
      "num_tokens": 168834340.0,
      "step": 77060
    },
    {
      "entropy": 5.7274798393249515,
      "epoch": 7.64837236998809,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010975519002335112,
      "loss": 4.9266,
      "mean_token_accuracy": 0.19802188724279404,
      "num_tokens": 168845057.0,
      "step": 77065
    },
    {
      "entropy": 5.701469469070434,
      "epoch": 7.6488685986502585,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010973114493647624,
      "loss": 4.8801,
      "mean_token_accuracy": 0.2068081721663475,
      "num_tokens": 168855970.0,
      "step": 77070
    },
    {
      "entropy": 5.697634029388428,
      "epoch": 7.649364827312426,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010970710394791284,
      "loss": 4.9042,
      "mean_token_accuracy": 0.20620231330394745,
      "num_tokens": 168867371.0,
      "step": 77075
    },
    {
      "entropy": 5.7359730243682865,
      "epoch": 7.649861055974593,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00010968306705825709,
      "loss": 4.9054,
      "mean_token_accuracy": 0.20721638947725296,
      "num_tokens": 168878920.0,
      "step": 77080
    },
    {
      "entropy": 5.702080631256104,
      "epoch": 7.650357284636761,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00010965903426810498,
      "loss": 4.9391,
      "mean_token_accuracy": 0.2000475913286209,
      "num_tokens": 168890135.0,
      "step": 77085
    },
    {
      "entropy": 5.655636548995972,
      "epoch": 7.650853513298928,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010963500557805255,
      "loss": 4.8851,
      "mean_token_accuracy": 0.20605571269989015,
      "num_tokens": 168900503.0,
      "step": 77090
    },
    {
      "entropy": 5.664462327957153,
      "epoch": 7.651349741961096,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001096109809886957,
      "loss": 4.838,
      "mean_token_accuracy": 0.2050304025411606,
      "num_tokens": 168909805.0,
      "step": 77095
    },
    {
      "entropy": 5.727297782897949,
      "epoch": 7.651845970623263,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010958696050063017,
      "loss": 4.9319,
      "mean_token_accuracy": 0.20004042088985444,
      "num_tokens": 168920888.0,
      "step": 77100
    },
    {
      "entropy": 5.68484320640564,
      "epoch": 7.652342199285431,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010956294411445155,
      "loss": 4.8819,
      "mean_token_accuracy": 0.19680551290512086,
      "num_tokens": 168931216.0,
      "step": 77105
    },
    {
      "entropy": 5.681425666809082,
      "epoch": 7.652838427947598,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010953893183075544,
      "loss": 4.9526,
      "mean_token_accuracy": 0.19980793744325637,
      "num_tokens": 168942792.0,
      "step": 77110
    },
    {
      "entropy": 5.651960992813111,
      "epoch": 7.653334656609766,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010951492365013723,
      "loss": 4.8191,
      "mean_token_accuracy": 0.21011047959327697,
      "num_tokens": 168952659.0,
      "step": 77115
    },
    {
      "entropy": 5.694907283782959,
      "epoch": 7.653830885271933,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001094909195731924,
      "loss": 4.935,
      "mean_token_accuracy": 0.19869647771120072,
      "num_tokens": 168963430.0,
      "step": 77120
    },
    {
      "entropy": 5.668439149856567,
      "epoch": 7.654327113934101,
      "grad_norm": 1.3125,
      "learning_rate": 0.00010946691960051606,
      "loss": 4.8641,
      "mean_token_accuracy": 0.20677425861358642,
      "num_tokens": 168974591.0,
      "step": 77125
    },
    {
      "entropy": 5.672597408294678,
      "epoch": 7.654823342596268,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010944292373270345,
      "loss": 4.8305,
      "mean_token_accuracy": 0.1999707341194153,
      "num_tokens": 168986003.0,
      "step": 77130
    },
    {
      "entropy": 5.653693723678589,
      "epoch": 7.655319571258436,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010941893197034962,
      "loss": 4.8806,
      "mean_token_accuracy": 0.1965052604675293,
      "num_tokens": 168996785.0,
      "step": 77135
    },
    {
      "entropy": 5.606898164749145,
      "epoch": 7.655815799920603,
      "grad_norm": 1.515625,
      "learning_rate": 0.00010939494431404939,
      "loss": 4.8155,
      "mean_token_accuracy": 0.20309847295284272,
      "num_tokens": 169007508.0,
      "step": 77140
    },
    {
      "entropy": 5.686544942855835,
      "epoch": 7.656312028582771,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010937096076439779,
      "loss": 4.9145,
      "mean_token_accuracy": 0.20171143412590026,
      "num_tokens": 169018411.0,
      "step": 77145
    },
    {
      "entropy": 5.688400983810425,
      "epoch": 7.656808257244938,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00010934698132198944,
      "loss": 4.8834,
      "mean_token_accuracy": 0.20938539057970046,
      "num_tokens": 169031086.0,
      "step": 77150
    },
    {
      "entropy": 5.636014938354492,
      "epoch": 7.657304485907106,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010932300598741896,
      "loss": 4.7481,
      "mean_token_accuracy": 0.21908482909202576,
      "num_tokens": 169040706.0,
      "step": 77155
    },
    {
      "entropy": 5.67721529006958,
      "epoch": 7.657800714569273,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010929903476128097,
      "loss": 4.8687,
      "mean_token_accuracy": 0.2054938480257988,
      "num_tokens": 169050478.0,
      "step": 77160
    },
    {
      "entropy": 5.68352108001709,
      "epoch": 7.658296943231441,
      "grad_norm": 1.375,
      "learning_rate": 0.00010927506764416989,
      "loss": 4.8983,
      "mean_token_accuracy": 0.20099807232618333,
      "num_tokens": 169061119.0,
      "step": 77165
    },
    {
      "entropy": 5.700973892211914,
      "epoch": 7.658793171893609,
      "grad_norm": 1.296875,
      "learning_rate": 0.00010925110463667992,
      "loss": 4.955,
      "mean_token_accuracy": 0.20388998985290527,
      "num_tokens": 169072243.0,
      "step": 77170
    },
    {
      "entropy": 5.736206293106079,
      "epoch": 7.659289400555776,
      "grad_norm": 1.25,
      "learning_rate": 0.0001092271457394055,
      "loss": 4.8959,
      "mean_token_accuracy": 0.198103603720665,
      "num_tokens": 169083878.0,
      "step": 77175
    },
    {
      "entropy": 5.743735599517822,
      "epoch": 7.659785629217944,
      "grad_norm": 1.359375,
      "learning_rate": 0.00010920319095294065,
      "loss": 4.8924,
      "mean_token_accuracy": 0.19764508605003356,
      "num_tokens": 169095942.0,
      "step": 77180
    },
    {
      "entropy": 5.739515590667724,
      "epoch": 7.660281857880111,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010917924027787935,
      "loss": 4.9886,
      "mean_token_accuracy": 0.19866356998682022,
      "num_tokens": 169107682.0,
      "step": 77185
    },
    {
      "entropy": 5.758034658432007,
      "epoch": 7.660778086542279,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001091552937148156,
      "loss": 4.9704,
      "mean_token_accuracy": 0.2015342593193054,
      "num_tokens": 169119785.0,
      "step": 77190
    },
    {
      "entropy": 5.6756244659423825,
      "epoch": 7.661274315204446,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010913135126434326,
      "loss": 4.8776,
      "mean_token_accuracy": 0.20703726410865783,
      "num_tokens": 169130658.0,
      "step": 77195
    },
    {
      "entropy": 5.722485828399658,
      "epoch": 7.661770543866614,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001091074129270559,
      "loss": 4.8841,
      "mean_token_accuracy": 0.20241874009370803,
      "num_tokens": 169141609.0,
      "step": 77200
    },
    {
      "entropy": 5.765383243560791,
      "epoch": 7.662266772528781,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010908347870354729,
      "loss": 4.9627,
      "mean_token_accuracy": 0.19818331599235534,
      "num_tokens": 169151529.0,
      "step": 77205
    },
    {
      "entropy": 5.728743934631348,
      "epoch": 7.662763001190949,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010905954859441089,
      "loss": 4.9484,
      "mean_token_accuracy": 0.19858898520469664,
      "num_tokens": 169162955.0,
      "step": 77210
    },
    {
      "entropy": 5.6766022682189945,
      "epoch": 7.663259229853116,
      "grad_norm": 1.3125,
      "learning_rate": 0.00010903562260024014,
      "loss": 4.8617,
      "mean_token_accuracy": 0.20771554112434387,
      "num_tokens": 169173956.0,
      "step": 77215
    },
    {
      "entropy": 5.745677947998047,
      "epoch": 7.663755458515284,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010901170072162823,
      "loss": 4.949,
      "mean_token_accuracy": 0.20876674354076385,
      "num_tokens": 169186432.0,
      "step": 77220
    },
    {
      "entropy": 5.682797288894653,
      "epoch": 7.664251687177451,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010898778295916856,
      "loss": 4.9217,
      "mean_token_accuracy": 0.19743335545063018,
      "num_tokens": 169198105.0,
      "step": 77225
    },
    {
      "entropy": 5.7605287551879885,
      "epoch": 7.664747915839619,
      "grad_norm": 1.375,
      "learning_rate": 0.00010896386931345406,
      "loss": 4.9406,
      "mean_token_accuracy": 0.197608382999897,
      "num_tokens": 169209122.0,
      "step": 77230
    },
    {
      "entropy": 5.675183200836182,
      "epoch": 7.665244144501786,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010893995978507792,
      "loss": 4.8596,
      "mean_token_accuracy": 0.20373766720294953,
      "num_tokens": 169219431.0,
      "step": 77235
    },
    {
      "entropy": 5.659485816955566,
      "epoch": 7.665740373163954,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010891605437463293,
      "loss": 4.8855,
      "mean_token_accuracy": 0.20912967771291732,
      "num_tokens": 169231476.0,
      "step": 77240
    },
    {
      "entropy": 5.737052869796753,
      "epoch": 7.666236601826121,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010889215308271191,
      "loss": 4.9602,
      "mean_token_accuracy": 0.2005390614271164,
      "num_tokens": 169243907.0,
      "step": 77245
    },
    {
      "entropy": 5.708730792999267,
      "epoch": 7.666732830488289,
      "grad_norm": 1.546875,
      "learning_rate": 0.00010886825590990751,
      "loss": 4.877,
      "mean_token_accuracy": 0.20604358315467836,
      "num_tokens": 169253445.0,
      "step": 77250
    },
    {
      "entropy": 5.733853673934936,
      "epoch": 7.667229059150457,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010884436285681243,
      "loss": 4.8907,
      "mean_token_accuracy": 0.20403937846422196,
      "num_tokens": 169264215.0,
      "step": 77255
    },
    {
      "entropy": 5.702259683609009,
      "epoch": 7.667725287812624,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010882047392401905,
      "loss": 4.8463,
      "mean_token_accuracy": 0.21106126457452773,
      "num_tokens": 169273799.0,
      "step": 77260
    },
    {
      "entropy": 5.698508214950562,
      "epoch": 7.668221516474792,
      "grad_norm": 1.25,
      "learning_rate": 0.00010879658911211992,
      "loss": 4.8859,
      "mean_token_accuracy": 0.20427990853786468,
      "num_tokens": 169285392.0,
      "step": 77265
    },
    {
      "entropy": 5.654837894439697,
      "epoch": 7.668717745136959,
      "grad_norm": 1.765625,
      "learning_rate": 0.00010877270842170724,
      "loss": 4.8275,
      "mean_token_accuracy": 0.2127177819609642,
      "num_tokens": 169296196.0,
      "step": 77270
    },
    {
      "entropy": 5.626491832733154,
      "epoch": 7.669213973799127,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001087488318533732,
      "loss": 4.8318,
      "mean_token_accuracy": 0.20710649639368056,
      "num_tokens": 169308094.0,
      "step": 77275
    },
    {
      "entropy": 5.664405012130738,
      "epoch": 7.669710202461294,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010872495940770988,
      "loss": 4.8551,
      "mean_token_accuracy": 0.20110403895378112,
      "num_tokens": 169319505.0,
      "step": 77280
    },
    {
      "entropy": 5.722268772125244,
      "epoch": 7.670206431123462,
      "grad_norm": 1.375,
      "learning_rate": 0.00010870109108530922,
      "loss": 5.0419,
      "mean_token_accuracy": 0.19637824296951295,
      "num_tokens": 169331779.0,
      "step": 77285
    },
    {
      "entropy": 5.642246103286743,
      "epoch": 7.670702659785629,
      "grad_norm": 1.25,
      "learning_rate": 0.00010867722688676315,
      "loss": 4.8299,
      "mean_token_accuracy": 0.2082362502813339,
      "num_tokens": 169342685.0,
      "step": 77290
    },
    {
      "entropy": 5.713222694396973,
      "epoch": 7.671198888447797,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010865336681266353,
      "loss": 4.9538,
      "mean_token_accuracy": 0.19706801027059556,
      "num_tokens": 169353575.0,
      "step": 77295
    },
    {
      "entropy": 5.750446224212647,
      "epoch": 7.671695117109964,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010862951086360198,
      "loss": 4.9433,
      "mean_token_accuracy": 0.20309385061264038,
      "num_tokens": 169365014.0,
      "step": 77300
    },
    {
      "entropy": 5.689811277389526,
      "epoch": 7.672191345772132,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010860565904017003,
      "loss": 4.8782,
      "mean_token_accuracy": 0.20437295883893966,
      "num_tokens": 169376820.0,
      "step": 77305
    },
    {
      "entropy": 5.744054651260376,
      "epoch": 7.672687574434299,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001085818113429592,
      "loss": 5.0226,
      "mean_token_accuracy": 0.1967771053314209,
      "num_tokens": 169388530.0,
      "step": 77310
    },
    {
      "entropy": 5.646116590499878,
      "epoch": 7.6731838030964665,
      "grad_norm": 1.515625,
      "learning_rate": 0.00010855796777256083,
      "loss": 4.7953,
      "mean_token_accuracy": 0.21184870451688767,
      "num_tokens": 169399162.0,
      "step": 77315
    },
    {
      "entropy": 5.645099449157715,
      "epoch": 7.673680031758634,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010853412832956616,
      "loss": 4.8608,
      "mean_token_accuracy": 0.2076525792479515,
      "num_tokens": 169411854.0,
      "step": 77320
    },
    {
      "entropy": 5.653068590164184,
      "epoch": 7.674176260420802,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010851029301456648,
      "loss": 4.8004,
      "mean_token_accuracy": 0.21001091748476028,
      "num_tokens": 169423193.0,
      "step": 77325
    },
    {
      "entropy": 5.694089889526367,
      "epoch": 7.674672489082969,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010848646182815278,
      "loss": 4.8722,
      "mean_token_accuracy": 0.2059737488627434,
      "num_tokens": 169433624.0,
      "step": 77330
    },
    {
      "entropy": 5.713542222976685,
      "epoch": 7.675168717745137,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010846263477091601,
      "loss": 4.9368,
      "mean_token_accuracy": 0.2009757250547409,
      "num_tokens": 169444145.0,
      "step": 77335
    },
    {
      "entropy": 5.6654698848724365,
      "epoch": 7.675664946407305,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00010843881184344706,
      "loss": 4.892,
      "mean_token_accuracy": 0.20106248408555985,
      "num_tokens": 169455454.0,
      "step": 77340
    },
    {
      "entropy": 5.655514287948608,
      "epoch": 7.676161175069472,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010841499304633667,
      "loss": 4.8737,
      "mean_token_accuracy": 0.20211408138275147,
      "num_tokens": 169466686.0,
      "step": 77345
    },
    {
      "entropy": 5.71259446144104,
      "epoch": 7.67665740373164,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010839117838017543,
      "loss": 4.9689,
      "mean_token_accuracy": 0.19552013128995896,
      "num_tokens": 169477397.0,
      "step": 77350
    },
    {
      "entropy": 5.694039964675904,
      "epoch": 7.677153632393807,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010836736784555393,
      "loss": 4.9467,
      "mean_token_accuracy": 0.19821071922779082,
      "num_tokens": 169489443.0,
      "step": 77355
    },
    {
      "entropy": 5.756268453598023,
      "epoch": 7.677649861055975,
      "grad_norm": 1.328125,
      "learning_rate": 0.00010834356144306275,
      "loss": 4.9737,
      "mean_token_accuracy": 0.19796248525381088,
      "num_tokens": 169499632.0,
      "step": 77360
    },
    {
      "entropy": 5.649510240554809,
      "epoch": 7.678146089718142,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010831975917329209,
      "loss": 4.785,
      "mean_token_accuracy": 0.21915453672409058,
      "num_tokens": 169511124.0,
      "step": 77365
    },
    {
      "entropy": 5.640373754501343,
      "epoch": 7.67864231838031,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010829596103683229,
      "loss": 4.8179,
      "mean_token_accuracy": 0.20795172154903413,
      "num_tokens": 169522076.0,
      "step": 77370
    },
    {
      "entropy": 5.688613843917847,
      "epoch": 7.679138547042477,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001082721670342734,
      "loss": 4.9126,
      "mean_token_accuracy": 0.20412635952234268,
      "num_tokens": 169533175.0,
      "step": 77375
    },
    {
      "entropy": 5.6894927501678465,
      "epoch": 7.679634775704645,
      "grad_norm": 1.375,
      "learning_rate": 0.00010824837716620545,
      "loss": 4.9019,
      "mean_token_accuracy": 0.2001341849565506,
      "num_tokens": 169543655.0,
      "step": 77380
    },
    {
      "entropy": 5.634965658187866,
      "epoch": 7.680131004366812,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010822459143321841,
      "loss": 4.8267,
      "mean_token_accuracy": 0.209472094476223,
      "num_tokens": 169555078.0,
      "step": 77385
    },
    {
      "entropy": 5.738643503189087,
      "epoch": 7.6806272330289795,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010820080983590222,
      "loss": 5.0051,
      "mean_token_accuracy": 0.19460465759038925,
      "num_tokens": 169565438.0,
      "step": 77390
    },
    {
      "entropy": 5.733362054824829,
      "epoch": 7.681123461691147,
      "grad_norm": 1.375,
      "learning_rate": 0.00010817703237484653,
      "loss": 4.9582,
      "mean_token_accuracy": 0.19053809046745301,
      "num_tokens": 169576491.0,
      "step": 77395
    },
    {
      "entropy": 5.725179529190063,
      "epoch": 7.6816196903533145,
      "grad_norm": 1.375,
      "learning_rate": 0.00010815325905064094,
      "loss": 4.9749,
      "mean_token_accuracy": 0.19766023606061936,
      "num_tokens": 169588511.0,
      "step": 77400
    },
    {
      "entropy": 5.697339773178101,
      "epoch": 7.682115919015482,
      "grad_norm": 1.296875,
      "learning_rate": 0.00010812948986387496,
      "loss": 4.9242,
      "mean_token_accuracy": 0.1962566152215004,
      "num_tokens": 169600833.0,
      "step": 77405
    },
    {
      "entropy": 5.685650300979614,
      "epoch": 7.68261214767765,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00010810572481513812,
      "loss": 4.8779,
      "mean_token_accuracy": 0.21230381578207017,
      "num_tokens": 169613319.0,
      "step": 77410
    },
    {
      "entropy": 5.6726758003234865,
      "epoch": 7.683108376339817,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010808196390501961,
      "loss": 4.8691,
      "mean_token_accuracy": 0.2039726957678795,
      "num_tokens": 169624149.0,
      "step": 77415
    },
    {
      "entropy": 5.70033826828003,
      "epoch": 7.683604605001985,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010805820713410878,
      "loss": 4.8708,
      "mean_token_accuracy": 0.20561047345399858,
      "num_tokens": 169635056.0,
      "step": 77420
    },
    {
      "entropy": 5.688794422149658,
      "epoch": 7.684100833664153,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00010803445450299469,
      "loss": 4.866,
      "mean_token_accuracy": 0.20128852427005767,
      "num_tokens": 169646531.0,
      "step": 77425
    },
    {
      "entropy": 5.674498081207275,
      "epoch": 7.68459706232632,
      "grad_norm": 1.5,
      "learning_rate": 0.00010801070601226634,
      "loss": 4.8799,
      "mean_token_accuracy": 0.2050924852490425,
      "num_tokens": 169658019.0,
      "step": 77430
    },
    {
      "entropy": 5.703583478927612,
      "epoch": 7.685093290988488,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010798696166251261,
      "loss": 4.9491,
      "mean_token_accuracy": 0.19974721670150758,
      "num_tokens": 169668250.0,
      "step": 77435
    },
    {
      "entropy": 5.640210485458374,
      "epoch": 7.685589519650655,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010796322145432239,
      "loss": 4.8576,
      "mean_token_accuracy": 0.20921730697155,
      "num_tokens": 169679555.0,
      "step": 77440
    },
    {
      "entropy": 5.7209559917449955,
      "epoch": 7.686085748312823,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010793948538828439,
      "loss": 4.9188,
      "mean_token_accuracy": 0.19874845296144486,
      "num_tokens": 169689692.0,
      "step": 77445
    },
    {
      "entropy": 5.688726854324341,
      "epoch": 7.68658197697499,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00010791575346498706,
      "loss": 4.8607,
      "mean_token_accuracy": 0.1999089926481247,
      "num_tokens": 169698604.0,
      "step": 77450
    },
    {
      "entropy": 5.728805446624756,
      "epoch": 7.687078205637158,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010789202568501906,
      "loss": 4.9174,
      "mean_token_accuracy": 0.20866225510835648,
      "num_tokens": 169709954.0,
      "step": 77455
    },
    {
      "entropy": 5.6915356636047365,
      "epoch": 7.687574434299325,
      "grad_norm": 1.1875,
      "learning_rate": 0.00010786830204896878,
      "loss": 4.9132,
      "mean_token_accuracy": 0.20127541720867156,
      "num_tokens": 169723225.0,
      "step": 77460
    },
    {
      "entropy": 5.731613540649414,
      "epoch": 7.6880706629614926,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010784458255742438,
      "loss": 4.9001,
      "mean_token_accuracy": 0.20631686598062515,
      "num_tokens": 169733837.0,
      "step": 77465
    },
    {
      "entropy": 5.741761350631714,
      "epoch": 7.68856689162366,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001078208672109742,
      "loss": 4.9175,
      "mean_token_accuracy": 0.20025371760129929,
      "num_tokens": 169744042.0,
      "step": 77470
    },
    {
      "entropy": 5.682326555252075,
      "epoch": 7.6890631202858275,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001077971560102063,
      "loss": 4.8842,
      "mean_token_accuracy": 0.2045968860387802,
      "num_tokens": 169755167.0,
      "step": 77475
    },
    {
      "entropy": 5.629179811477661,
      "epoch": 7.689559348947995,
      "grad_norm": 1.5625,
      "learning_rate": 0.00010777344895570857,
      "loss": 4.8755,
      "mean_token_accuracy": 0.2039152830839157,
      "num_tokens": 169765191.0,
      "step": 77480
    },
    {
      "entropy": 5.6673966407775875,
      "epoch": 7.6900555776101625,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010774974604806903,
      "loss": 4.8838,
      "mean_token_accuracy": 0.19857205599546432,
      "num_tokens": 169775590.0,
      "step": 77485
    },
    {
      "entropy": 5.632623243331909,
      "epoch": 7.69055180627233,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001077260472878754,
      "loss": 4.8468,
      "mean_token_accuracy": 0.21174718886613847,
      "num_tokens": 169786228.0,
      "step": 77490
    },
    {
      "entropy": 5.67727746963501,
      "epoch": 7.691048034934498,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001077023526757153,
      "loss": 4.9001,
      "mean_token_accuracy": 0.20247366577386855,
      "num_tokens": 169796817.0,
      "step": 77495
    },
    {
      "entropy": 5.699145269393921,
      "epoch": 7.691544263596665,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00010767866221217639,
      "loss": 4.9158,
      "mean_token_accuracy": 0.1961896762251854,
      "num_tokens": 169807105.0,
      "step": 77500
    },
    {
      "entropy": 5.7082531452178955,
      "epoch": 7.692040492258833,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010765497589784612,
      "loss": 4.8514,
      "mean_token_accuracy": 0.2136620908975601,
      "num_tokens": 169817787.0,
      "step": 77505
    },
    {
      "entropy": 5.719500112533569,
      "epoch": 7.692536720921001,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00010763129373331186,
      "loss": 4.9598,
      "mean_token_accuracy": 0.20236503928899766,
      "num_tokens": 169829595.0,
      "step": 77510
    },
    {
      "entropy": 5.7178911685943605,
      "epoch": 7.693032949583168,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010760761571916081,
      "loss": 4.9588,
      "mean_token_accuracy": 0.20140892565250396,
      "num_tokens": 169839998.0,
      "step": 77515
    },
    {
      "entropy": 5.73994517326355,
      "epoch": 7.693529178245336,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010758394185598023,
      "loss": 4.9618,
      "mean_token_accuracy": 0.19487272799015046,
      "num_tokens": 169851890.0,
      "step": 77520
    },
    {
      "entropy": 5.6563108444213865,
      "epoch": 7.694025406907503,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010756027214435709,
      "loss": 4.8561,
      "mean_token_accuracy": 0.21080141216516496,
      "num_tokens": 169862147.0,
      "step": 77525
    },
    {
      "entropy": 5.6854509830474855,
      "epoch": 7.694521635569671,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010753660658487849,
      "loss": 4.8107,
      "mean_token_accuracy": 0.20604957193136214,
      "num_tokens": 169872888.0,
      "step": 77530
    },
    {
      "entropy": 5.597671365737915,
      "epoch": 7.695017864231838,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010751294517813118,
      "loss": 4.7685,
      "mean_token_accuracy": 0.21174100935459136,
      "num_tokens": 169884951.0,
      "step": 77535
    },
    {
      "entropy": 5.704579448699951,
      "epoch": 7.695514092894006,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010748928792470192,
      "loss": 4.9407,
      "mean_token_accuracy": 0.19818853735923767,
      "num_tokens": 169897048.0,
      "step": 77540
    },
    {
      "entropy": 5.664771795272827,
      "epoch": 7.696010321556173,
      "grad_norm": 1.21875,
      "learning_rate": 0.00010746563482517729,
      "loss": 4.8883,
      "mean_token_accuracy": 0.1945033773779869,
      "num_tokens": 169908327.0,
      "step": 77545
    },
    {
      "entropy": 5.666822242736816,
      "epoch": 7.6965065502183405,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010744198588014398,
      "loss": 4.8857,
      "mean_token_accuracy": 0.20812946408987046,
      "num_tokens": 169918904.0,
      "step": 77550
    },
    {
      "entropy": 5.7024688720703125,
      "epoch": 7.697002778880508,
      "grad_norm": 1.515625,
      "learning_rate": 0.00010741834109018833,
      "loss": 4.9458,
      "mean_token_accuracy": 0.2024140864610672,
      "num_tokens": 169929055.0,
      "step": 77555
    },
    {
      "entropy": 5.658177328109741,
      "epoch": 7.6974990075426755,
      "grad_norm": 1.296875,
      "learning_rate": 0.00010739470045589677,
      "loss": 4.8335,
      "mean_token_accuracy": 0.206898832321167,
      "num_tokens": 169939680.0,
      "step": 77560
    },
    {
      "entropy": 5.642721557617188,
      "epoch": 7.697995236204843,
      "grad_norm": 1.328125,
      "learning_rate": 0.00010737106397785547,
      "loss": 4.8259,
      "mean_token_accuracy": 0.2084597185254097,
      "num_tokens": 169950555.0,
      "step": 77565
    },
    {
      "entropy": 5.703773736953735,
      "epoch": 7.69849146486701,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010734743165665057,
      "loss": 4.9249,
      "mean_token_accuracy": 0.19768306463956833,
      "num_tokens": 169962487.0,
      "step": 77570
    },
    {
      "entropy": 5.704161167144775,
      "epoch": 7.698987693529178,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010732380349286811,
      "loss": 4.8912,
      "mean_token_accuracy": 0.19816820621490477,
      "num_tokens": 169973212.0,
      "step": 77575
    },
    {
      "entropy": 5.687295532226562,
      "epoch": 7.699483922191346,
      "grad_norm": 1.3125,
      "learning_rate": 0.00010730017948709395,
      "loss": 4.8205,
      "mean_token_accuracy": 0.20364097207784654,
      "num_tokens": 169983638.0,
      "step": 77580
    },
    {
      "entropy": 5.696461486816406,
      "epoch": 7.699980150853513,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010727655963991398,
      "loss": 4.8481,
      "mean_token_accuracy": 0.1981531336903572,
      "num_tokens": 169994050.0,
      "step": 77585
    },
    {
      "entropy": 5.660415601730347,
      "epoch": 7.700476379515681,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010725294395191399,
      "loss": 4.8705,
      "mean_token_accuracy": 0.20592766702175141,
      "num_tokens": 170004354.0,
      "step": 77590
    },
    {
      "entropy": 5.656364107131958,
      "epoch": 7.700972608177849,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010722933242367951,
      "loss": 4.8305,
      "mean_token_accuracy": 0.20853037238121033,
      "num_tokens": 170015869.0,
      "step": 77595
    },
    {
      "entropy": 5.617100191116333,
      "epoch": 7.701468836840016,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010720572505579607,
      "loss": 4.7931,
      "mean_token_accuracy": 0.21204425394535065,
      "num_tokens": 170026498.0,
      "step": 77600
    },
    {
      "entropy": 5.656995153427124,
      "epoch": 7.701965065502184,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010718212184884909,
      "loss": 4.8985,
      "mean_token_accuracy": 0.2003520607948303,
      "num_tokens": 170037829.0,
      "step": 77605
    },
    {
      "entropy": 5.689083242416382,
      "epoch": 7.702461294164351,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010715852280342381,
      "loss": 4.9117,
      "mean_token_accuracy": 0.19785570949316025,
      "num_tokens": 170050024.0,
      "step": 77610
    },
    {
      "entropy": 5.664284086227417,
      "epoch": 7.702957522826519,
      "grad_norm": 1.3125,
      "learning_rate": 0.00010713492792010551,
      "loss": 4.8727,
      "mean_token_accuracy": 0.20287209153175353,
      "num_tokens": 170061929.0,
      "step": 77615
    },
    {
      "entropy": 5.711790704727173,
      "epoch": 7.703453751488686,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010711133719947934,
      "loss": 4.9117,
      "mean_token_accuracy": 0.20006005913019181,
      "num_tokens": 170072869.0,
      "step": 77620
    },
    {
      "entropy": 5.714139699935913,
      "epoch": 7.7039499801508535,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010708775064213025,
      "loss": 4.9519,
      "mean_token_accuracy": 0.20409152507781983,
      "num_tokens": 170083478.0,
      "step": 77625
    },
    {
      "entropy": 5.689962577819824,
      "epoch": 7.704446208813021,
      "grad_norm": 1.28125,
      "learning_rate": 0.00010706416824864312,
      "loss": 4.906,
      "mean_token_accuracy": 0.19951433837413787,
      "num_tokens": 170096669.0,
      "step": 77630
    },
    {
      "entropy": 5.682593727111817,
      "epoch": 7.7049424374751885,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001070405900196027,
      "loss": 4.8354,
      "mean_token_accuracy": 0.20450987219810485,
      "num_tokens": 170107518.0,
      "step": 77635
    },
    {
      "entropy": 5.672270345687866,
      "epoch": 7.705438666137356,
      "grad_norm": 1.375,
      "learning_rate": 0.00010701701595559371,
      "loss": 4.8893,
      "mean_token_accuracy": 0.20130468308925628,
      "num_tokens": 170116983.0,
      "step": 77640
    },
    {
      "entropy": 5.720099830627442,
      "epoch": 7.705934894799523,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010699344605720082,
      "loss": 4.8943,
      "mean_token_accuracy": 0.20028685480356218,
      "num_tokens": 170128504.0,
      "step": 77645
    },
    {
      "entropy": 5.753800535202027,
      "epoch": 7.706431123461691,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010696988032500836,
      "loss": 4.8999,
      "mean_token_accuracy": 0.2040368288755417,
      "num_tokens": 170139113.0,
      "step": 77650
    },
    {
      "entropy": 5.807398128509521,
      "epoch": 7.706927352123858,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010694631875960088,
      "loss": 5.0315,
      "mean_token_accuracy": 0.1924394980072975,
      "num_tokens": 170149386.0,
      "step": 77655
    },
    {
      "entropy": 5.629950332641601,
      "epoch": 7.707423580786026,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010692276136156256,
      "loss": 4.8118,
      "mean_token_accuracy": 0.20331616401672364,
      "num_tokens": 170159613.0,
      "step": 77660
    },
    {
      "entropy": 5.7546313285827635,
      "epoch": 7.707919809448194,
      "grad_norm": 1.5,
      "learning_rate": 0.0001068992081314776,
      "loss": 4.9778,
      "mean_token_accuracy": 0.19410065859556197,
      "num_tokens": 170170299.0,
      "step": 77665
    },
    {
      "entropy": 5.704293727874756,
      "epoch": 7.708416038110361,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010687565906992996,
      "loss": 4.9718,
      "mean_token_accuracy": 0.19565879851579665,
      "num_tokens": 170181742.0,
      "step": 77670
    },
    {
      "entropy": 5.658835363388062,
      "epoch": 7.708912266772529,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001068521141775038,
      "loss": 4.8543,
      "mean_token_accuracy": 0.2086360737681389,
      "num_tokens": 170192131.0,
      "step": 77675
    },
    {
      "entropy": 5.762825775146484,
      "epoch": 7.709408495434697,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010682857345478278,
      "loss": 4.9672,
      "mean_token_accuracy": 0.20279531478881835,
      "num_tokens": 170203270.0,
      "step": 77680
    },
    {
      "entropy": 5.670560026168824,
      "epoch": 7.709904724096864,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010680503690235082,
      "loss": 4.8513,
      "mean_token_accuracy": 0.20271459519863128,
      "num_tokens": 170213904.0,
      "step": 77685
    },
    {
      "entropy": 5.697993564605713,
      "epoch": 7.710400952759032,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00010678150452079155,
      "loss": 4.9182,
      "mean_token_accuracy": 0.20136123299598693,
      "num_tokens": 170224967.0,
      "step": 77690
    },
    {
      "entropy": 5.696855545043945,
      "epoch": 7.710897181421199,
      "grad_norm": 1.59375,
      "learning_rate": 0.00010675797631068848,
      "loss": 4.7945,
      "mean_token_accuracy": 0.2079437181353569,
      "num_tokens": 170234602.0,
      "step": 77695
    },
    {
      "entropy": 5.664795017242431,
      "epoch": 7.7113934100833665,
      "grad_norm": 1.515625,
      "learning_rate": 0.00010673445227262502,
      "loss": 4.8639,
      "mean_token_accuracy": 0.21010803580284118,
      "num_tokens": 170244710.0,
      "step": 77700
    },
    {
      "entropy": 5.713927412033081,
      "epoch": 7.711889638745534,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010671093240718457,
      "loss": 4.959,
      "mean_token_accuracy": 0.20644906163215637,
      "num_tokens": 170255807.0,
      "step": 77705
    },
    {
      "entropy": 5.695511960983277,
      "epoch": 7.7123858674077015,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010668741671495034,
      "loss": 4.9551,
      "mean_token_accuracy": 0.19840241968631744,
      "num_tokens": 170267008.0,
      "step": 77710
    },
    {
      "entropy": 5.654291391372681,
      "epoch": 7.712882096069869,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010666390519650557,
      "loss": 4.8528,
      "mean_token_accuracy": 0.20056607127189635,
      "num_tokens": 170277013.0,
      "step": 77715
    },
    {
      "entropy": 5.711384630203247,
      "epoch": 7.713378324732036,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001066403978524332,
      "loss": 4.9466,
      "mean_token_accuracy": 0.2023301303386688,
      "num_tokens": 170287794.0,
      "step": 77720
    },
    {
      "entropy": 5.6921782970428465,
      "epoch": 7.713874553394204,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010661689468331619,
      "loss": 4.8986,
      "mean_token_accuracy": 0.20061463117599487,
      "num_tokens": 170297779.0,
      "step": 77725
    },
    {
      "entropy": 5.764891481399536,
      "epoch": 7.714370782056371,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010659339568973731,
      "loss": 4.9447,
      "mean_token_accuracy": 0.20020258277654648,
      "num_tokens": 170309974.0,
      "step": 77730
    },
    {
      "entropy": 5.673282337188721,
      "epoch": 7.714867010718539,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00010656990087227936,
      "loss": 4.8659,
      "mean_token_accuracy": 0.20368081331253052,
      "num_tokens": 170320695.0,
      "step": 77735
    },
    {
      "entropy": 5.758144521713257,
      "epoch": 7.715363239380706,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010654641023152494,
      "loss": 4.936,
      "mean_token_accuracy": 0.20427474975585938,
      "num_tokens": 170331481.0,
      "step": 77740
    },
    {
      "entropy": 5.721592378616333,
      "epoch": 7.715859468042874,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010652292376805651,
      "loss": 4.8921,
      "mean_token_accuracy": 0.2109535440802574,
      "num_tokens": 170341673.0,
      "step": 77745
    },
    {
      "entropy": 5.776513147354126,
      "epoch": 7.716355696705042,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001064994414824566,
      "loss": 4.9939,
      "mean_token_accuracy": 0.20015773177146912,
      "num_tokens": 170353302.0,
      "step": 77750
    },
    {
      "entropy": 5.705624151229858,
      "epoch": 7.716851925367209,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010647596337530747,
      "loss": 4.8751,
      "mean_token_accuracy": 0.19973436892032623,
      "num_tokens": 170364275.0,
      "step": 77755
    },
    {
      "entropy": 5.648019027709961,
      "epoch": 7.717348154029377,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010645248944719124,
      "loss": 4.8399,
      "mean_token_accuracy": 0.21221620738506317,
      "num_tokens": 170374766.0,
      "step": 77760
    },
    {
      "entropy": 5.721581506729126,
      "epoch": 7.717844382691545,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001064290196986901,
      "loss": 4.9463,
      "mean_token_accuracy": 0.192775921523571,
      "num_tokens": 170385242.0,
      "step": 77765
    },
    {
      "entropy": 5.7399805068969725,
      "epoch": 7.718340611353712,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001064055541303861,
      "loss": 4.9203,
      "mean_token_accuracy": 0.20593204200267792,
      "num_tokens": 170398055.0,
      "step": 77770
    },
    {
      "entropy": 5.7373573780059814,
      "epoch": 7.7188368400158796,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010638209274286096,
      "loss": 4.9587,
      "mean_token_accuracy": 0.1959363713860512,
      "num_tokens": 170409155.0,
      "step": 77775
    },
    {
      "entropy": 5.7494926929473875,
      "epoch": 7.719333068678047,
      "grad_norm": 1.28125,
      "learning_rate": 0.00010635863553669667,
      "loss": 4.975,
      "mean_token_accuracy": 0.19164087474346161,
      "num_tokens": 170421240.0,
      "step": 77780
    },
    {
      "entropy": 5.690513753890992,
      "epoch": 7.7198292973402145,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010633518251247481,
      "loss": 4.8482,
      "mean_token_accuracy": 0.21413697749376298,
      "num_tokens": 170431138.0,
      "step": 77785
    },
    {
      "entropy": 5.712183952331543,
      "epoch": 7.720325526002382,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010631173367077693,
      "loss": 4.9636,
      "mean_token_accuracy": 0.19672538638114928,
      "num_tokens": 170442391.0,
      "step": 77790
    },
    {
      "entropy": 5.671785068511963,
      "epoch": 7.7208217546645495,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010628828901218464,
      "loss": 4.8345,
      "mean_token_accuracy": 0.21038943231105806,
      "num_tokens": 170453441.0,
      "step": 77795
    },
    {
      "entropy": 5.665852642059326,
      "epoch": 7.721317983326717,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00010626484853727922,
      "loss": 4.83,
      "mean_token_accuracy": 0.20718371421098708,
      "num_tokens": 170464534.0,
      "step": 77800
    },
    {
      "entropy": 5.7500707626342775,
      "epoch": 7.721814211988884,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00010624141224664196,
      "loss": 4.9776,
      "mean_token_accuracy": 0.20447972863912584,
      "num_tokens": 170475835.0,
      "step": 77805
    },
    {
      "entropy": 5.6665931224823,
      "epoch": 7.722310440651052,
      "grad_norm": 1.546875,
      "learning_rate": 0.000106217980140854,
      "loss": 4.8809,
      "mean_token_accuracy": 0.2031327098608017,
      "num_tokens": 170486293.0,
      "step": 77810
    },
    {
      "entropy": 5.7140504837036135,
      "epoch": 7.722806669313219,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00010619455222049647,
      "loss": 4.9101,
      "mean_token_accuracy": 0.20591246336698532,
      "num_tokens": 170497096.0,
      "step": 77815
    },
    {
      "entropy": 5.647521877288819,
      "epoch": 7.723302897975387,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00010617112848615026,
      "loss": 4.9007,
      "mean_token_accuracy": 0.20713299810886382,
      "num_tokens": 170508400.0,
      "step": 77820
    },
    {
      "entropy": 5.702852725982666,
      "epoch": 7.723799126637554,
      "grad_norm": 1.359375,
      "learning_rate": 0.00010614770893839631,
      "loss": 4.9757,
      "mean_token_accuracy": 0.2041246622800827,
      "num_tokens": 170519552.0,
      "step": 77825
    },
    {
      "entropy": 5.671231460571289,
      "epoch": 7.724295355299722,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010612429357781537,
      "loss": 4.8881,
      "mean_token_accuracy": 0.20137252956628798,
      "num_tokens": 170529877.0,
      "step": 77830
    },
    {
      "entropy": 5.708114528656006,
      "epoch": 7.724791583961889,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00010610088240498803,
      "loss": 4.9001,
      "mean_token_accuracy": 0.201712067425251,
      "num_tokens": 170540316.0,
      "step": 77835
    },
    {
      "entropy": 5.746037149429322,
      "epoch": 7.725287812624057,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010607747542049478,
      "loss": 4.9782,
      "mean_token_accuracy": 0.19574351161718367,
      "num_tokens": 170551594.0,
      "step": 77840
    },
    {
      "entropy": 5.640881299972534,
      "epoch": 7.725784041286225,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010605407262491626,
      "loss": 4.8156,
      "mean_token_accuracy": 0.209122896194458,
      "num_tokens": 170561475.0,
      "step": 77845
    },
    {
      "entropy": 5.703183889389038,
      "epoch": 7.726280269948393,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001060306740188326,
      "loss": 4.9106,
      "mean_token_accuracy": 0.19467561244964598,
      "num_tokens": 170572214.0,
      "step": 77850
    },
    {
      "entropy": 5.647723150253296,
      "epoch": 7.72677649861056,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010600727960282419,
      "loss": 4.8381,
      "mean_token_accuracy": 0.2033257946372032,
      "num_tokens": 170582611.0,
      "step": 77855
    },
    {
      "entropy": 5.6477946758270265,
      "epoch": 7.7272727272727275,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00010598388937747109,
      "loss": 4.8959,
      "mean_token_accuracy": 0.20316032767295839,
      "num_tokens": 170594257.0,
      "step": 77860
    },
    {
      "entropy": 5.778933382034301,
      "epoch": 7.727768955934895,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010596050334335336,
      "loss": 5.0102,
      "mean_token_accuracy": 0.20340412557125093,
      "num_tokens": 170605671.0,
      "step": 77865
    },
    {
      "entropy": 5.685091257095337,
      "epoch": 7.7282651845970625,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00010593712150105081,
      "loss": 4.8871,
      "mean_token_accuracy": 0.20752983838319777,
      "num_tokens": 170617084.0,
      "step": 77870
    },
    {
      "entropy": 5.677063655853272,
      "epoch": 7.72876141325923,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010591374385114341,
      "loss": 4.8482,
      "mean_token_accuracy": 0.19784586131572723,
      "num_tokens": 170628211.0,
      "step": 77875
    },
    {
      "entropy": 5.7015790939331055,
      "epoch": 7.729257641921397,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010589037039421078,
      "loss": 4.8832,
      "mean_token_accuracy": 0.20331092327833175,
      "num_tokens": 170638758.0,
      "step": 77880
    },
    {
      "entropy": 5.619156694412231,
      "epoch": 7.729753870583565,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001058670011308326,
      "loss": 4.7479,
      "mean_token_accuracy": 0.2104036808013916,
      "num_tokens": 170649020.0,
      "step": 77885
    },
    {
      "entropy": 5.720651054382325,
      "epoch": 7.730250099245732,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010584363606158836,
      "loss": 4.9048,
      "mean_token_accuracy": 0.200547431409359,
      "num_tokens": 170658649.0,
      "step": 77890
    },
    {
      "entropy": 5.664763498306274,
      "epoch": 7.7307463279079,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010582027518705741,
      "loss": 4.9057,
      "mean_token_accuracy": 0.2030136242508888,
      "num_tokens": 170669962.0,
      "step": 77895
    },
    {
      "entropy": 5.696451330184937,
      "epoch": 7.731242556570067,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010579691850781913,
      "loss": 4.9468,
      "mean_token_accuracy": 0.20170529782772065,
      "num_tokens": 170681143.0,
      "step": 77900
    },
    {
      "entropy": 5.678467559814453,
      "epoch": 7.731738785232235,
      "grad_norm": 1.375,
      "learning_rate": 0.00010577356602445259,
      "loss": 4.8928,
      "mean_token_accuracy": 0.20271456837654114,
      "num_tokens": 170693187.0,
      "step": 77905
    },
    {
      "entropy": 5.727177333831787,
      "epoch": 7.732235013894402,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010575021773753698,
      "loss": 4.9135,
      "mean_token_accuracy": 0.1942673146724701,
      "num_tokens": 170703551.0,
      "step": 77910
    },
    {
      "entropy": 5.7037129402160645,
      "epoch": 7.73273124255657,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010572687364765132,
      "loss": 4.9046,
      "mean_token_accuracy": 0.20024407058954238,
      "num_tokens": 170714540.0,
      "step": 77915
    },
    {
      "entropy": 5.664462280273438,
      "epoch": 7.733227471218737,
      "grad_norm": 1.1875,
      "learning_rate": 0.00010570353375537443,
      "loss": 4.8729,
      "mean_token_accuracy": 0.20567940920591354,
      "num_tokens": 170726902.0,
      "step": 77920
    },
    {
      "entropy": 5.7324706554412845,
      "epoch": 7.733723699880905,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00010568019806128512,
      "loss": 4.9343,
      "mean_token_accuracy": 0.20115906894207,
      "num_tokens": 170737169.0,
      "step": 77925
    },
    {
      "entropy": 5.706250333786011,
      "epoch": 7.734219928543073,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00010565686656596202,
      "loss": 4.9079,
      "mean_token_accuracy": 0.2045516014099121,
      "num_tokens": 170749678.0,
      "step": 77930
    },
    {
      "entropy": 5.720998477935791,
      "epoch": 7.7347161572052405,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010563353926998367,
      "loss": 4.8803,
      "mean_token_accuracy": 0.20706553906202316,
      "num_tokens": 170759055.0,
      "step": 77935
    },
    {
      "entropy": 5.578627681732177,
      "epoch": 7.735212385867408,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010561021617392861,
      "loss": 4.6921,
      "mean_token_accuracy": 0.22121264636516572,
      "num_tokens": 170769259.0,
      "step": 77940
    },
    {
      "entropy": 5.679272127151489,
      "epoch": 7.7357086145295755,
      "grad_norm": 1.296875,
      "learning_rate": 0.00010558689727837526,
      "loss": 4.8284,
      "mean_token_accuracy": 0.2049799308180809,
      "num_tokens": 170779380.0,
      "step": 77945
    },
    {
      "entropy": 5.662881898880005,
      "epoch": 7.736204843191743,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010556358258390181,
      "loss": 4.906,
      "mean_token_accuracy": 0.2070533588528633,
      "num_tokens": 170790026.0,
      "step": 77950
    },
    {
      "entropy": 5.695746612548828,
      "epoch": 7.73670107185391,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00010554027209108635,
      "loss": 4.9876,
      "mean_token_accuracy": 0.1991480752825737,
      "num_tokens": 170801239.0,
      "step": 77955
    },
    {
      "entropy": 5.705251741409302,
      "epoch": 7.737197300516078,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010551696580050704,
      "loss": 4.9545,
      "mean_token_accuracy": 0.19469188302755355,
      "num_tokens": 170812695.0,
      "step": 77960
    },
    {
      "entropy": 5.699817371368408,
      "epoch": 7.737693529178245,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010549366371274172,
      "loss": 4.9393,
      "mean_token_accuracy": 0.19786564409732818,
      "num_tokens": 170822168.0,
      "step": 77965
    },
    {
      "entropy": 5.736268377304077,
      "epoch": 7.738189757840413,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010547036582836832,
      "loss": 4.9144,
      "mean_token_accuracy": 0.19975508153438568,
      "num_tokens": 170834604.0,
      "step": 77970
    },
    {
      "entropy": 5.708647394180298,
      "epoch": 7.73868598650258,
      "grad_norm": 1.375,
      "learning_rate": 0.00010544707214796448,
      "loss": 4.94,
      "mean_token_accuracy": 0.19569558054208755,
      "num_tokens": 170845431.0,
      "step": 77975
    },
    {
      "entropy": 5.675007152557373,
      "epoch": 7.739182215164748,
      "grad_norm": 1.296875,
      "learning_rate": 0.00010542378267210795,
      "loss": 4.8711,
      "mean_token_accuracy": 0.21117836982011795,
      "num_tokens": 170857397.0,
      "step": 77980
    },
    {
      "entropy": 5.740943717956543,
      "epoch": 7.739678443826915,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00010540049740137625,
      "loss": 4.9031,
      "mean_token_accuracy": 0.19627435356378556,
      "num_tokens": 170869329.0,
      "step": 77985
    },
    {
      "entropy": 5.725174283981323,
      "epoch": 7.740174672489083,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010537721633634672,
      "loss": 4.8841,
      "mean_token_accuracy": 0.20861314982175827,
      "num_tokens": 170879454.0,
      "step": 77990
    },
    {
      "entropy": 5.710841417312622,
      "epoch": 7.74067090115125,
      "grad_norm": 1.265625,
      "learning_rate": 0.00010535393947759666,
      "loss": 4.8619,
      "mean_token_accuracy": 0.20750389844179154,
      "num_tokens": 170889640.0,
      "step": 77995
    },
    {
      "entropy": 5.682574701309204,
      "epoch": 7.741167129813418,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010533066682570343,
      "loss": 4.8823,
      "mean_token_accuracy": 0.20760975033044815,
      "num_tokens": 170900577.0,
      "step": 78000
    },
    {
      "epoch": 7.741167129813418,
      "eval_entropy": 5.423865316259122,
      "eval_loss": 5.054303169250488,
      "eval_mean_token_accuracy": 0.1998592189582404,
      "eval_num_tokens": 170900577.0,
      "eval_runtime": 20.4045,
      "eval_samples_per_second": 1650.08,
      "eval_steps_per_second": 206.278,
      "step": 78000
    }
  ],
  "logging_steps": 5,
  "max_steps": 100750,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 3000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.256593280335872e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}