{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.60233918128655,
  "eval_steps": 500,
  "global_step": 6500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 2.4548874139785766,
      "epoch": 0.005847953216374269,
      "grad_norm": 17.375,
      "learning_rate": 2e-06,
      "loss": 14.5381,
      "mean_token_accuracy": 0.03773948270827532,
      "num_tokens": 13999.0,
      "step": 5
    },
    {
      "entropy": 2.515810489654541,
      "epoch": 0.011695906432748537,
      "grad_norm": 15.6875,
      "learning_rate": 4.5e-06,
      "loss": 14.4504,
      "mean_token_accuracy": 0.03857097271829844,
      "num_tokens": 25888.0,
      "step": 10
    },
    {
      "entropy": 2.5658308982849123,
      "epoch": 0.017543859649122806,
      "grad_norm": 18.75,
      "learning_rate": 7e-06,
      "loss": 14.3376,
      "mean_token_accuracy": 0.03801379818469286,
      "num_tokens": 37699.0,
      "step": 15
    },
    {
      "entropy": 2.648589539527893,
      "epoch": 0.023391812865497075,
      "grad_norm": 18.0,
      "learning_rate": 9.5e-06,
      "loss": 14.4331,
      "mean_token_accuracy": 0.025170913990586997,
      "num_tokens": 49648.0,
      "step": 20
    },
    {
      "entropy": 3.067568039894104,
      "epoch": 0.029239766081871343,
      "grad_norm": 21.125,
      "learning_rate": 1.2e-05,
      "loss": 13.8106,
      "mean_token_accuracy": 0.042210881039500235,
      "num_tokens": 62255.0,
      "step": 25
    },
    {
      "entropy": 3.5193849325180055,
      "epoch": 0.03508771929824561,
      "grad_norm": 25.125,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 13.4282,
      "mean_token_accuracy": 0.03690041042864323,
      "num_tokens": 76003.0,
      "step": 30
    },
    {
      "entropy": 5.110470724105835,
      "epoch": 0.04093567251461988,
      "grad_norm": 39.25,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 12.305,
      "mean_token_accuracy": 0.032853833958506586,
      "num_tokens": 87585.0,
      "step": 35
    },
    {
      "entropy": 8.690666055679321,
      "epoch": 0.04678362573099415,
      "grad_norm": 4.90625,
      "learning_rate": 1.95e-05,
      "loss": 10.8954,
      "mean_token_accuracy": 0.02273947522044182,
      "num_tokens": 100404.0,
      "step": 40
    },
    {
      "entropy": 10.664226055145264,
      "epoch": 0.05263157894736842,
      "grad_norm": 4.375,
      "learning_rate": 2.2e-05,
      "loss": 10.6246,
      "mean_token_accuracy": 0.01664919788017869,
      "num_tokens": 113130.0,
      "step": 45
    },
    {
      "entropy": 10.546917343139649,
      "epoch": 0.05847953216374269,
      "grad_norm": 3.875,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 10.344,
      "mean_token_accuracy": 0.02105573210865259,
      "num_tokens": 125430.0,
      "step": 50
    },
    {
      "entropy": 10.102593898773193,
      "epoch": 0.06432748538011696,
      "grad_norm": 4.375,
      "learning_rate": 2.7e-05,
      "loss": 10.2003,
      "mean_token_accuracy": 0.02809331100434065,
      "num_tokens": 137550.0,
      "step": 55
    },
    {
      "entropy": 10.388475513458252,
      "epoch": 0.07017543859649122,
      "grad_norm": 2.21875,
      "learning_rate": 2.95e-05,
      "loss": 10.0452,
      "mean_token_accuracy": 0.03612718116492033,
      "num_tokens": 149530.0,
      "step": 60
    },
    {
      "entropy": 10.559987545013428,
      "epoch": 0.07602339181286549,
      "grad_norm": 1.9609375,
      "learning_rate": 3.2e-05,
      "loss": 9.9327,
      "mean_token_accuracy": 0.04171597119420767,
      "num_tokens": 161671.0,
      "step": 65
    },
    {
      "entropy": 10.477953338623047,
      "epoch": 0.08187134502923976,
      "grad_norm": 1.8828125,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.8499,
      "mean_token_accuracy": 0.04532963968813419,
      "num_tokens": 174160.0,
      "step": 70
    },
    {
      "entropy": 10.54740114212036,
      "epoch": 0.08771929824561403,
      "grad_norm": 2.375,
      "learning_rate": 3.7e-05,
      "loss": 9.7707,
      "mean_token_accuracy": 0.06920528337359429,
      "num_tokens": 186969.0,
      "step": 75
    },
    {
      "entropy": 10.44933729171753,
      "epoch": 0.0935672514619883,
      "grad_norm": 1.7578125,
      "learning_rate": 3.95e-05,
      "loss": 9.658,
      "mean_token_accuracy": 0.07367655821144581,
      "num_tokens": 199470.0,
      "step": 80
    },
    {
      "entropy": 10.356026935577393,
      "epoch": 0.09941520467836257,
      "grad_norm": 1.7109375,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.5258,
      "mean_token_accuracy": 0.07762051969766617,
      "num_tokens": 211306.0,
      "step": 85
    },
    {
      "entropy": 10.388982677459717,
      "epoch": 0.10526315789473684,
      "grad_norm": 1.859375,
      "learning_rate": 4.45e-05,
      "loss": 9.521,
      "mean_token_accuracy": 0.06988351345062256,
      "num_tokens": 223323.0,
      "step": 90
    },
    {
      "entropy": 10.449160861968995,
      "epoch": 0.1111111111111111,
      "grad_norm": 1.8515625,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 9.404,
      "mean_token_accuracy": 0.07955777905881405,
      "num_tokens": 235896.0,
      "step": 95
    },
    {
      "entropy": 10.24062376022339,
      "epoch": 0.11695906432748537,
      "grad_norm": 2.265625,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 9.3366,
      "mean_token_accuracy": 0.07595778591930866,
      "num_tokens": 247879.0,
      "step": 100
    },
    {
      "entropy": 10.324538803100586,
      "epoch": 0.12280701754385964,
      "grad_norm": 1.640625,
      "learning_rate": 5.2e-05,
      "loss": 9.2463,
      "mean_token_accuracy": 0.07249736785888672,
      "num_tokens": 259714.0,
      "step": 105
    },
    {
      "entropy": 10.299348545074462,
      "epoch": 0.1286549707602339,
      "grad_norm": 1.6796875,
      "learning_rate": 5.45e-05,
      "loss": 9.1954,
      "mean_token_accuracy": 0.07774566560983658,
      "num_tokens": 272429.0,
      "step": 110
    },
    {
      "entropy": 10.246688842773438,
      "epoch": 0.13450292397660818,
      "grad_norm": 1.5703125,
      "learning_rate": 5.7e-05,
      "loss": 9.1908,
      "mean_token_accuracy": 0.07083970755338669,
      "num_tokens": 285924.0,
      "step": 115
    },
    {
      "entropy": 10.224361610412597,
      "epoch": 0.14035087719298245,
      "grad_norm": 1.5,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 9.0459,
      "mean_token_accuracy": 0.08458271101117135,
      "num_tokens": 299453.0,
      "step": 120
    },
    {
      "entropy": 10.05007495880127,
      "epoch": 0.14619883040935672,
      "grad_norm": 1.3359375,
      "learning_rate": 6.2e-05,
      "loss": 9.0164,
      "mean_token_accuracy": 0.08130386173725128,
      "num_tokens": 312295.0,
      "step": 125
    },
    {
      "entropy": 10.085463523864746,
      "epoch": 0.15204678362573099,
      "grad_norm": 1.4453125,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.8884,
      "mean_token_accuracy": 0.08093869239091873,
      "num_tokens": 324730.0,
      "step": 130
    },
    {
      "entropy": 9.985839366912842,
      "epoch": 0.15789473684210525,
      "grad_norm": 1.21875,
      "learning_rate": 6.7e-05,
      "loss": 8.8096,
      "mean_token_accuracy": 0.08213753812015057,
      "num_tokens": 336481.0,
      "step": 135
    },
    {
      "entropy": 9.958092880249023,
      "epoch": 0.16374269005847952,
      "grad_norm": 1.265625,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.7771,
      "mean_token_accuracy": 0.08334722444415092,
      "num_tokens": 348987.0,
      "step": 140
    },
    {
      "entropy": 9.810273170471191,
      "epoch": 0.1695906432748538,
      "grad_norm": 1.1875,
      "learning_rate": 7.2e-05,
      "loss": 8.6756,
      "mean_token_accuracy": 0.0778221569955349,
      "num_tokens": 360390.0,
      "step": 145
    },
    {
      "entropy": 9.81429967880249,
      "epoch": 0.17543859649122806,
      "grad_norm": 1.109375,
      "learning_rate": 7.45e-05,
      "loss": 8.6545,
      "mean_token_accuracy": 0.07858845368027687,
      "num_tokens": 372456.0,
      "step": 150
    },
    {
      "entropy": 9.693943881988526,
      "epoch": 0.18128654970760233,
      "grad_norm": 1.140625,
      "learning_rate": 7.7e-05,
      "loss": 8.5776,
      "mean_token_accuracy": 0.07394571527838707,
      "num_tokens": 384185.0,
      "step": 155
    },
    {
      "entropy": 9.575428581237793,
      "epoch": 0.1871345029239766,
      "grad_norm": 0.98828125,
      "learning_rate": 7.950000000000001e-05,
      "loss": 8.5065,
      "mean_token_accuracy": 0.07995149567723274,
      "num_tokens": 395399.0,
      "step": 160
    },
    {
      "entropy": 9.371949768066406,
      "epoch": 0.19298245614035087,
      "grad_norm": 0.8828125,
      "learning_rate": 8.2e-05,
      "loss": 8.5001,
      "mean_token_accuracy": 0.0805008016526699,
      "num_tokens": 408676.0,
      "step": 165
    },
    {
      "entropy": 9.294499015808105,
      "epoch": 0.19883040935672514,
      "grad_norm": 0.83203125,
      "learning_rate": 8.450000000000001e-05,
      "loss": 8.4325,
      "mean_token_accuracy": 0.08033415265381336,
      "num_tokens": 421861.0,
      "step": 170
    },
    {
      "entropy": 9.107025909423829,
      "epoch": 0.2046783625730994,
      "grad_norm": 0.80859375,
      "learning_rate": 8.7e-05,
      "loss": 8.4751,
      "mean_token_accuracy": 0.08693947605788707,
      "num_tokens": 434852.0,
      "step": 175
    },
    {
      "entropy": 8.985072135925293,
      "epoch": 0.21052631578947367,
      "grad_norm": 0.6875,
      "learning_rate": 8.95e-05,
      "loss": 8.3966,
      "mean_token_accuracy": 0.0785727895796299,
      "num_tokens": 447313.0,
      "step": 180
    },
    {
      "entropy": 8.946270751953126,
      "epoch": 0.21637426900584794,
      "grad_norm": 1.2109375,
      "learning_rate": 9.2e-05,
      "loss": 8.3362,
      "mean_token_accuracy": 0.08314507529139518,
      "num_tokens": 460271.0,
      "step": 185
    },
    {
      "entropy": 8.779074096679688,
      "epoch": 0.2222222222222222,
      "grad_norm": 0.82421875,
      "learning_rate": 9.45e-05,
      "loss": 8.2913,
      "mean_token_accuracy": 0.0910579614341259,
      "num_tokens": 473006.0,
      "step": 190
    },
    {
      "entropy": 8.667890739440917,
      "epoch": 0.22807017543859648,
      "grad_norm": 1.1015625,
      "learning_rate": 9.7e-05,
      "loss": 8.3421,
      "mean_token_accuracy": 0.08133782222867011,
      "num_tokens": 486100.0,
      "step": 195
    },
    {
      "entropy": 8.728115272521972,
      "epoch": 0.23391812865497075,
      "grad_norm": 1.78125,
      "learning_rate": 9.95e-05,
      "loss": 8.2275,
      "mean_token_accuracy": 0.08423927277326584,
      "num_tokens": 499460.0,
      "step": 200
    },
    {
      "entropy": 8.724910449981689,
      "epoch": 0.23976608187134502,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000102,
      "loss": 8.2439,
      "mean_token_accuracy": 0.07881896123290062,
      "num_tokens": 511795.0,
      "step": 205
    },
    {
      "entropy": 8.58439531326294,
      "epoch": 0.24561403508771928,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00010449999999999999,
      "loss": 8.295,
      "mean_token_accuracy": 0.08954361081123352,
      "num_tokens": 524897.0,
      "step": 210
    },
    {
      "entropy": 8.694450759887696,
      "epoch": 0.25146198830409355,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000107,
      "loss": 8.2492,
      "mean_token_accuracy": 0.08542842417955399,
      "num_tokens": 538325.0,
      "step": 215
    },
    {
      "entropy": 8.672403717041016,
      "epoch": 0.2573099415204678,
      "grad_norm": 0.69921875,
      "learning_rate": 0.0001095,
      "loss": 8.294,
      "mean_token_accuracy": 0.08235822916030884,
      "num_tokens": 550979.0,
      "step": 220
    },
    {
      "entropy": 8.61720962524414,
      "epoch": 0.2631578947368421,
      "grad_norm": 1.03125,
      "learning_rate": 0.000112,
      "loss": 8.2583,
      "mean_token_accuracy": 0.08766907453536987,
      "num_tokens": 563001.0,
      "step": 225
    },
    {
      "entropy": 8.610678768157959,
      "epoch": 0.26900584795321636,
      "grad_norm": 0.68359375,
      "learning_rate": 0.0001145,
      "loss": 8.1767,
      "mean_token_accuracy": 0.08766172006726265,
      "num_tokens": 573600.0,
      "step": 230
    },
    {
      "entropy": 8.663910102844238,
      "epoch": 0.27485380116959063,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00011700000000000001,
      "loss": 8.1709,
      "mean_token_accuracy": 0.0914827212691307,
      "num_tokens": 585287.0,
      "step": 235
    },
    {
      "entropy": 8.543773651123047,
      "epoch": 0.2807017543859649,
      "grad_norm": 0.984375,
      "learning_rate": 0.00011949999999999999,
      "loss": 8.2154,
      "mean_token_accuracy": 0.08472376465797424,
      "num_tokens": 597386.0,
      "step": 240
    },
    {
      "entropy": 8.523555374145507,
      "epoch": 0.28654970760233917,
      "grad_norm": 0.82421875,
      "learning_rate": 0.000122,
      "loss": 8.1833,
      "mean_token_accuracy": 0.09036211296916008,
      "num_tokens": 609256.0,
      "step": 245
    },
    {
      "entropy": 8.552801322937011,
      "epoch": 0.29239766081871343,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001245,
      "loss": 8.1955,
      "mean_token_accuracy": 0.08631851598620414,
      "num_tokens": 621506.0,
      "step": 250
    },
    {
      "entropy": 8.489137554168702,
      "epoch": 0.2982456140350877,
      "grad_norm": 0.921875,
      "learning_rate": 0.000127,
      "loss": 8.0989,
      "mean_token_accuracy": 0.09788421541452408,
      "num_tokens": 633512.0,
      "step": 255
    },
    {
      "entropy": 8.542137622833252,
      "epoch": 0.30409356725146197,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001295,
      "loss": 8.2385,
      "mean_token_accuracy": 0.0867220088839531,
      "num_tokens": 647098.0,
      "step": 260
    },
    {
      "entropy": 8.521434879302978,
      "epoch": 0.30994152046783624,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000132,
      "loss": 8.2165,
      "mean_token_accuracy": 0.08817893266677856,
      "num_tokens": 659559.0,
      "step": 265
    },
    {
      "entropy": 8.531222820281982,
      "epoch": 0.3157894736842105,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013450000000000002,
      "loss": 8.1671,
      "mean_token_accuracy": 0.08971341624855995,
      "num_tokens": 672621.0,
      "step": 270
    },
    {
      "entropy": 8.470209407806397,
      "epoch": 0.3216374269005848,
      "grad_norm": 1.1875,
      "learning_rate": 0.00013700000000000002,
      "loss": 8.1267,
      "mean_token_accuracy": 0.09204860255122185,
      "num_tokens": 685652.0,
      "step": 275
    },
    {
      "entropy": 8.486028385162353,
      "epoch": 0.32748538011695905,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0001395,
      "loss": 8.1752,
      "mean_token_accuracy": 0.09480618685483932,
      "num_tokens": 696910.0,
      "step": 280
    },
    {
      "entropy": 8.44861536026001,
      "epoch": 0.3333333333333333,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014199999999999998,
      "loss": 8.0682,
      "mean_token_accuracy": 0.09902379810810089,
      "num_tokens": 709054.0,
      "step": 285
    },
    {
      "entropy": 8.478011894226075,
      "epoch": 0.3391812865497076,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001445,
      "loss": 8.0885,
      "mean_token_accuracy": 0.09357146173715591,
      "num_tokens": 720902.0,
      "step": 290
    },
    {
      "entropy": 8.469257354736328,
      "epoch": 0.34502923976608185,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000147,
      "loss": 8.153,
      "mean_token_accuracy": 0.09559885188937187,
      "num_tokens": 733319.0,
      "step": 295
    },
    {
      "entropy": 8.412768745422364,
      "epoch": 0.3508771929824561,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0001495,
      "loss": 8.0329,
      "mean_token_accuracy": 0.09888283610343933,
      "num_tokens": 744817.0,
      "step": 300
    },
    {
      "entropy": 8.385979843139648,
      "epoch": 0.3567251461988304,
      "grad_norm": 0.921875,
      "learning_rate": 0.000152,
      "loss": 8.0906,
      "mean_token_accuracy": 0.09885545223951339,
      "num_tokens": 756582.0,
      "step": 305
    },
    {
      "entropy": 8.444335651397704,
      "epoch": 0.36257309941520466,
      "grad_norm": 1.03125,
      "learning_rate": 0.00015450000000000001,
      "loss": 8.0097,
      "mean_token_accuracy": 0.10672620832920074,
      "num_tokens": 769273.0,
      "step": 310
    },
    {
      "entropy": 8.386385726928712,
      "epoch": 0.3684210526315789,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000157,
      "loss": 8.0753,
      "mean_token_accuracy": 0.09999166950583457,
      "num_tokens": 782001.0,
      "step": 315
    },
    {
      "entropy": 8.315172958374024,
      "epoch": 0.3742690058479532,
      "grad_norm": 1.03125,
      "learning_rate": 0.0001595,
      "loss": 8.0586,
      "mean_token_accuracy": 0.10521865040063857,
      "num_tokens": 794575.0,
      "step": 320
    },
    {
      "entropy": 8.437547206878662,
      "epoch": 0.38011695906432746,
      "grad_norm": 0.84375,
      "learning_rate": 0.000162,
      "loss": 8.0377,
      "mean_token_accuracy": 0.10137710124254226,
      "num_tokens": 805583.0,
      "step": 325
    },
    {
      "entropy": 8.299012565612793,
      "epoch": 0.38596491228070173,
      "grad_norm": 1.65625,
      "learning_rate": 0.00016450000000000001,
      "loss": 8.0804,
      "mean_token_accuracy": 0.10292363837361336,
      "num_tokens": 817617.0,
      "step": 330
    },
    {
      "entropy": 8.361987209320068,
      "epoch": 0.391812865497076,
      "grad_norm": 1.046875,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.9841,
      "mean_token_accuracy": 0.10257630646228791,
      "num_tokens": 830043.0,
      "step": 335
    },
    {
      "entropy": 8.306623458862305,
      "epoch": 0.39766081871345027,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.9788,
      "mean_token_accuracy": 0.11217362359166146,
      "num_tokens": 843200.0,
      "step": 340
    },
    {
      "entropy": 8.317947292327881,
      "epoch": 0.40350877192982454,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.9994,
      "mean_token_accuracy": 0.11003749370574951,
      "num_tokens": 856031.0,
      "step": 345
    },
    {
      "entropy": 8.315475559234619,
      "epoch": 0.4093567251461988,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.9637,
      "mean_token_accuracy": 0.11112827882170677,
      "num_tokens": 869119.0,
      "step": 350
    },
    {
      "entropy": 8.308537578582763,
      "epoch": 0.4152046783625731,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000177,
      "loss": 7.984,
      "mean_token_accuracy": 0.1118388794362545,
      "num_tokens": 880802.0,
      "step": 355
    },
    {
      "entropy": 8.310247135162353,
      "epoch": 0.42105263157894735,
      "grad_norm": 1.015625,
      "learning_rate": 0.0001795,
      "loss": 7.9752,
      "mean_token_accuracy": 0.11066636517643928,
      "num_tokens": 894237.0,
      "step": 360
    },
    {
      "entropy": 8.24343318939209,
      "epoch": 0.4269005847953216,
      "grad_norm": 1.03125,
      "learning_rate": 0.000182,
      "loss": 7.99,
      "mean_token_accuracy": 0.10824075117707252,
      "num_tokens": 906695.0,
      "step": 365
    },
    {
      "entropy": 8.380464553833008,
      "epoch": 0.4327485380116959,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001845,
      "loss": 7.9536,
      "mean_token_accuracy": 0.11111580803990365,
      "num_tokens": 918955.0,
      "step": 370
    },
    {
      "entropy": 8.231604385375977,
      "epoch": 0.43859649122807015,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000187,
      "loss": 7.9735,
      "mean_token_accuracy": 0.11058306768536567,
      "num_tokens": 931812.0,
      "step": 375
    },
    {
      "entropy": 8.334205436706544,
      "epoch": 0.4444444444444444,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0001895,
      "loss": 7.9795,
      "mean_token_accuracy": 0.10282116234302521,
      "num_tokens": 945261.0,
      "step": 380
    },
    {
      "entropy": 8.210968399047852,
      "epoch": 0.4502923976608187,
      "grad_norm": 0.984375,
      "learning_rate": 0.000192,
      "loss": 7.8956,
      "mean_token_accuracy": 0.11960833892226219,
      "num_tokens": 958376.0,
      "step": 385
    },
    {
      "entropy": 8.18839659690857,
      "epoch": 0.45614035087719296,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0001945,
      "loss": 7.9245,
      "mean_token_accuracy": 0.11952447071671486,
      "num_tokens": 971709.0,
      "step": 390
    },
    {
      "entropy": 8.367930507659912,
      "epoch": 0.4619883040935672,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.8919,
      "mean_token_accuracy": 0.11701075136661529,
      "num_tokens": 983847.0,
      "step": 395
    },
    {
      "entropy": 8.25608139038086,
      "epoch": 0.4678362573099415,
      "grad_norm": 1.078125,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.9585,
      "mean_token_accuracy": 0.10782488510012626,
      "num_tokens": 997034.0,
      "step": 400
    },
    {
      "entropy": 8.187435054779053,
      "epoch": 0.47368421052631576,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000202,
      "loss": 7.9853,
      "mean_token_accuracy": 0.11163778379559516,
      "num_tokens": 1010213.0,
      "step": 405
    },
    {
      "entropy": 8.300292873382569,
      "epoch": 0.47953216374269003,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020449999999999998,
      "loss": 7.8797,
      "mean_token_accuracy": 0.11556096971035004,
      "num_tokens": 1021618.0,
      "step": 410
    },
    {
      "entropy": 8.165176582336425,
      "epoch": 0.4853801169590643,
      "grad_norm": 1.15625,
      "learning_rate": 0.000207,
      "loss": 7.8907,
      "mean_token_accuracy": 0.11225929781794548,
      "num_tokens": 1033809.0,
      "step": 415
    },
    {
      "entropy": 8.331029987335205,
      "epoch": 0.49122807017543857,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002095,
      "loss": 7.888,
      "mean_token_accuracy": 0.11076272949576378,
      "num_tokens": 1045291.0,
      "step": 420
    },
    {
      "entropy": 8.092021656036376,
      "epoch": 0.49707602339181284,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000212,
      "loss": 7.8613,
      "mean_token_accuracy": 0.11468090787529946,
      "num_tokens": 1057516.0,
      "step": 425
    },
    {
      "entropy": 8.167704010009766,
      "epoch": 0.5029239766081871,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002145,
      "loss": 7.8082,
      "mean_token_accuracy": 0.12245556116104125,
      "num_tokens": 1070675.0,
      "step": 430
    },
    {
      "entropy": 8.19180703163147,
      "epoch": 0.5087719298245614,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00021700000000000002,
      "loss": 7.888,
      "mean_token_accuracy": 0.12304584905505181,
      "num_tokens": 1082965.0,
      "step": 435
    },
    {
      "entropy": 8.234239482879639,
      "epoch": 0.5146198830409356,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0002195,
      "loss": 7.858,
      "mean_token_accuracy": 0.112663733959198,
      "num_tokens": 1095922.0,
      "step": 440
    },
    {
      "entropy": 8.179874038696289,
      "epoch": 0.52046783625731,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000222,
      "loss": 7.8413,
      "mean_token_accuracy": 0.11735650300979614,
      "num_tokens": 1109002.0,
      "step": 445
    },
    {
      "entropy": 8.141633415222168,
      "epoch": 0.5263157894736842,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0002245,
      "loss": 7.8662,
      "mean_token_accuracy": 0.11112897917628288,
      "num_tokens": 1121199.0,
      "step": 450
    },
    {
      "entropy": 8.179853630065917,
      "epoch": 0.5321637426900585,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00022700000000000002,
      "loss": 7.8017,
      "mean_token_accuracy": 0.12308088317513466,
      "num_tokens": 1134721.0,
      "step": 455
    },
    {
      "entropy": 8.180206298828125,
      "epoch": 0.5380116959064327,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022950000000000002,
      "loss": 7.8632,
      "mean_token_accuracy": 0.11824943199753761,
      "num_tokens": 1147954.0,
      "step": 460
    },
    {
      "entropy": 8.030020809173584,
      "epoch": 0.543859649122807,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00023200000000000003,
      "loss": 7.81,
      "mean_token_accuracy": 0.1231809377670288,
      "num_tokens": 1161208.0,
      "step": 465
    },
    {
      "entropy": 8.183233261108398,
      "epoch": 0.5497076023391813,
      "grad_norm": 1.25,
      "learning_rate": 0.00023449999999999998,
      "loss": 7.9011,
      "mean_token_accuracy": 0.11757142841815948,
      "num_tokens": 1173789.0,
      "step": 470
    },
    {
      "entropy": 8.235744667053222,
      "epoch": 0.5555555555555556,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000237,
      "loss": 7.7889,
      "mean_token_accuracy": 0.12023304700851441,
      "num_tokens": 1185765.0,
      "step": 475
    },
    {
      "entropy": 7.98179259300232,
      "epoch": 0.5614035087719298,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002395,
      "loss": 7.8028,
      "mean_token_accuracy": 0.1243036225438118,
      "num_tokens": 1197954.0,
      "step": 480
    },
    {
      "entropy": 8.164046669006348,
      "epoch": 0.5672514619883041,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000242,
      "loss": 7.7434,
      "mean_token_accuracy": 0.12476142644882202,
      "num_tokens": 1211285.0,
      "step": 485
    },
    {
      "entropy": 8.129650735855103,
      "epoch": 0.5730994152046783,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0002445,
      "loss": 7.8072,
      "mean_token_accuracy": 0.1169096663594246,
      "num_tokens": 1223197.0,
      "step": 490
    },
    {
      "entropy": 8.043150568008423,
      "epoch": 0.5789473684210527,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000247,
      "loss": 7.7402,
      "mean_token_accuracy": 0.12265990749001503,
      "num_tokens": 1234817.0,
      "step": 495
    },
    {
      "entropy": 8.130622863769531,
      "epoch": 0.5847953216374269,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0002495,
      "loss": 7.8029,
      "mean_token_accuracy": 0.12631115764379502,
      "num_tokens": 1249305.0,
      "step": 500
    },
    {
      "epoch": 0.5847953216374269,
      "eval_entropy": 7.920137181746221,
      "eval_loss": 7.822256088256836,
      "eval_mean_token_accuracy": 0.1146006511996849,
      "eval_num_tokens": 1249305.0,
      "eval_runtime": 1.74,
      "eval_samples_per_second": 1555.772,
      "eval_steps_per_second": 194.831,
      "step": 500
    },
    {
      "entropy": 8.136149835586547,
      "epoch": 0.5906432748538012,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000252,
      "loss": 7.7717,
      "mean_token_accuracy": 0.12191700711846351,
      "num_tokens": 1262182.0,
      "step": 505
    },
    {
      "entropy": 8.012873411178589,
      "epoch": 0.5964912280701754,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002545,
      "loss": 7.7877,
      "mean_token_accuracy": 0.12733336314558982,
      "num_tokens": 1274616.0,
      "step": 510
    },
    {
      "entropy": 8.046648216247558,
      "epoch": 0.6023391812865497,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000257,
      "loss": 7.7911,
      "mean_token_accuracy": 0.11969717964529991,
      "num_tokens": 1287306.0,
      "step": 515
    },
    {
      "entropy": 8.059616661071777,
      "epoch": 0.6081871345029239,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002595,
      "loss": 7.6474,
      "mean_token_accuracy": 0.12940788641571999,
      "num_tokens": 1299472.0,
      "step": 520
    },
    {
      "entropy": 8.013744401931763,
      "epoch": 0.6140350877192983,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000262,
      "loss": 7.6178,
      "mean_token_accuracy": 0.13267553746700286,
      "num_tokens": 1312636.0,
      "step": 525
    },
    {
      "entropy": 7.913894128799439,
      "epoch": 0.6198830409356725,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026450000000000003,
      "loss": 7.659,
      "mean_token_accuracy": 0.13482385575771333,
      "num_tokens": 1324058.0,
      "step": 530
    },
    {
      "entropy": 8.036643075942994,
      "epoch": 0.6257309941520468,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026700000000000004,
      "loss": 7.7684,
      "mean_token_accuracy": 0.11713232323527337,
      "num_tokens": 1337605.0,
      "step": 535
    },
    {
      "entropy": 8.078272104263306,
      "epoch": 0.631578947368421,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026950000000000005,
      "loss": 7.777,
      "mean_token_accuracy": 0.1210218109190464,
      "num_tokens": 1350658.0,
      "step": 540
    },
    {
      "entropy": 8.057914018630981,
      "epoch": 0.6374269005847953,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027200000000000005,
      "loss": 7.707,
      "mean_token_accuracy": 0.12556772753596307,
      "num_tokens": 1362091.0,
      "step": 545
    },
    {
      "entropy": 7.995908594131469,
      "epoch": 0.6432748538011696,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002745,
      "loss": 7.704,
      "mean_token_accuracy": 0.12518641129136085,
      "num_tokens": 1374055.0,
      "step": 550
    },
    {
      "entropy": 7.98234977722168,
      "epoch": 0.6491228070175439,
      "grad_norm": 1.171875,
      "learning_rate": 0.000277,
      "loss": 7.58,
      "mean_token_accuracy": 0.12927438765764238,
      "num_tokens": 1385269.0,
      "step": 555
    },
    {
      "entropy": 7.961950588226318,
      "epoch": 0.6549707602339181,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0002795,
      "loss": 7.6932,
      "mean_token_accuracy": 0.13011750653386117,
      "num_tokens": 1396559.0,
      "step": 560
    },
    {
      "entropy": 7.937076187133789,
      "epoch": 0.6608187134502924,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028199999999999997,
      "loss": 7.7119,
      "mean_token_accuracy": 0.13076897338032722,
      "num_tokens": 1408838.0,
      "step": 565
    },
    {
      "entropy": 7.980747890472412,
      "epoch": 0.6666666666666666,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0002845,
      "loss": 7.5961,
      "mean_token_accuracy": 0.1342053160071373,
      "num_tokens": 1421023.0,
      "step": 570
    },
    {
      "entropy": 7.89539532661438,
      "epoch": 0.672514619883041,
      "grad_norm": 1.15625,
      "learning_rate": 0.000287,
      "loss": 7.6598,
      "mean_token_accuracy": 0.13019674867391587,
      "num_tokens": 1433912.0,
      "step": 575
    },
    {
      "entropy": 8.000441455841065,
      "epoch": 0.6783625730994152,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002895,
      "loss": 7.6742,
      "mean_token_accuracy": 0.12573233470320702,
      "num_tokens": 1445262.0,
      "step": 580
    },
    {
      "entropy": 7.876591634750366,
      "epoch": 0.6842105263157895,
      "grad_norm": 0.83203125,
      "learning_rate": 0.000292,
      "loss": 7.6415,
      "mean_token_accuracy": 0.12933589220046998,
      "num_tokens": 1457927.0,
      "step": 585
    },
    {
      "entropy": 7.937921953201294,
      "epoch": 0.6900584795321637,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0002945,
      "loss": 7.6915,
      "mean_token_accuracy": 0.12511010393500327,
      "num_tokens": 1470116.0,
      "step": 590
    },
    {
      "entropy": 7.94811863899231,
      "epoch": 0.695906432748538,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000297,
      "loss": 7.5772,
      "mean_token_accuracy": 0.13277097642421723,
      "num_tokens": 1482067.0,
      "step": 595
    },
    {
      "entropy": 7.93737530708313,
      "epoch": 0.7017543859649122,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002995,
      "loss": 7.6747,
      "mean_token_accuracy": 0.12135898694396019,
      "num_tokens": 1493902.0,
      "step": 600
    },
    {
      "entropy": 7.944917917251587,
      "epoch": 0.7076023391812866,
      "grad_norm": 1.1875,
      "learning_rate": 0.000302,
      "loss": 7.6329,
      "mean_token_accuracy": 0.12311855405569076,
      "num_tokens": 1505689.0,
      "step": 605
    },
    {
      "entropy": 7.876118469238281,
      "epoch": 0.7134502923976608,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003045,
      "loss": 7.5735,
      "mean_token_accuracy": 0.1439603090286255,
      "num_tokens": 1518002.0,
      "step": 610
    },
    {
      "entropy": 7.9030755996704105,
      "epoch": 0.7192982456140351,
      "grad_norm": 1.078125,
      "learning_rate": 0.000307,
      "loss": 7.601,
      "mean_token_accuracy": 0.12741394564509392,
      "num_tokens": 1529307.0,
      "step": 615
    },
    {
      "entropy": 7.934239673614502,
      "epoch": 0.7251461988304093,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003095,
      "loss": 7.7544,
      "mean_token_accuracy": 0.11436132788658142,
      "num_tokens": 1542351.0,
      "step": 620
    },
    {
      "entropy": 7.823674011230469,
      "epoch": 0.7309941520467836,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000312,
      "loss": 7.5697,
      "mean_token_accuracy": 0.1321065440773964,
      "num_tokens": 1556263.0,
      "step": 625
    },
    {
      "entropy": 7.873884916305542,
      "epoch": 0.7368421052631579,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0003145,
      "loss": 7.5818,
      "mean_token_accuracy": 0.13355903923511506,
      "num_tokens": 1568906.0,
      "step": 630
    },
    {
      "entropy": 7.839196968078613,
      "epoch": 0.7426900584795322,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000317,
      "loss": 7.5782,
      "mean_token_accuracy": 0.132675963640213,
      "num_tokens": 1580631.0,
      "step": 635
    },
    {
      "entropy": 7.838891458511353,
      "epoch": 0.7485380116959064,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003195,
      "loss": 7.5937,
      "mean_token_accuracy": 0.12946432083845139,
      "num_tokens": 1592369.0,
      "step": 640
    },
    {
      "entropy": 7.938169670104981,
      "epoch": 0.7543859649122807,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000322,
      "loss": 7.6258,
      "mean_token_accuracy": 0.13315243050456047,
      "num_tokens": 1604888.0,
      "step": 645
    },
    {
      "entropy": 7.889529895782471,
      "epoch": 0.7602339181286549,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032450000000000003,
      "loss": 7.5451,
      "mean_token_accuracy": 0.13434720635414124,
      "num_tokens": 1617128.0,
      "step": 650
    },
    {
      "entropy": 7.793774127960205,
      "epoch": 0.7660818713450293,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032700000000000003,
      "loss": 7.5885,
      "mean_token_accuracy": 0.1243424341082573,
      "num_tokens": 1629350.0,
      "step": 655
    },
    {
      "entropy": 7.897194147109985,
      "epoch": 0.7719298245614035,
      "grad_norm": 0.984375,
      "learning_rate": 0.00032950000000000004,
      "loss": 7.6016,
      "mean_token_accuracy": 0.1256876789033413,
      "num_tokens": 1643082.0,
      "step": 660
    },
    {
      "entropy": 7.854056262969971,
      "epoch": 0.7777777777777778,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00033200000000000005,
      "loss": 7.5822,
      "mean_token_accuracy": 0.13232242465019226,
      "num_tokens": 1654785.0,
      "step": 665
    },
    {
      "entropy": 7.879878759384155,
      "epoch": 0.783625730994152,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00033450000000000005,
      "loss": 7.6061,
      "mean_token_accuracy": 0.12869918420910836,
      "num_tokens": 1667704.0,
      "step": 670
    },
    {
      "entropy": 7.890443611145019,
      "epoch": 0.7894736842105263,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000337,
      "loss": 7.5665,
      "mean_token_accuracy": 0.12847531214356422,
      "num_tokens": 1680459.0,
      "step": 675
    },
    {
      "entropy": 7.804240417480469,
      "epoch": 0.7953216374269005,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003395,
      "loss": 7.4724,
      "mean_token_accuracy": 0.13116607591509818,
      "num_tokens": 1692912.0,
      "step": 680
    },
    {
      "entropy": 7.765193080902099,
      "epoch": 0.8011695906432749,
      "grad_norm": 2.28125,
      "learning_rate": 0.000342,
      "loss": 7.5031,
      "mean_token_accuracy": 0.13217658475041388,
      "num_tokens": 1705329.0,
      "step": 685
    },
    {
      "entropy": 7.918375539779663,
      "epoch": 0.8070175438596491,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034449999999999997,
      "loss": 7.5958,
      "mean_token_accuracy": 0.1265776574611664,
      "num_tokens": 1717283.0,
      "step": 690
    },
    {
      "entropy": 7.726377534866333,
      "epoch": 0.8128654970760234,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000347,
      "loss": 7.5333,
      "mean_token_accuracy": 0.13292808830738068,
      "num_tokens": 1730308.0,
      "step": 695
    },
    {
      "entropy": 7.862821102142334,
      "epoch": 0.8187134502923976,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003495,
      "loss": 7.5828,
      "mean_token_accuracy": 0.13102516084909438,
      "num_tokens": 1742716.0,
      "step": 700
    },
    {
      "entropy": 7.830471420288086,
      "epoch": 0.8245614035087719,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000352,
      "loss": 7.5287,
      "mean_token_accuracy": 0.13678008690476418,
      "num_tokens": 1754183.0,
      "step": 705
    },
    {
      "entropy": 7.751367998123169,
      "epoch": 0.8304093567251462,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003545,
      "loss": 7.5464,
      "mean_token_accuracy": 0.12743634656071662,
      "num_tokens": 1766601.0,
      "step": 710
    },
    {
      "entropy": 7.780935764312744,
      "epoch": 0.8362573099415205,
      "grad_norm": 1.7734375,
      "learning_rate": 0.000357,
      "loss": 7.4354,
      "mean_token_accuracy": 0.14405448883771896,
      "num_tokens": 1778756.0,
      "step": 715
    },
    {
      "entropy": 7.539671182632446,
      "epoch": 0.8421052631578947,
      "grad_norm": 1.25,
      "learning_rate": 0.0003595,
      "loss": 7.4979,
      "mean_token_accuracy": 0.13751505464315414,
      "num_tokens": 1791485.0,
      "step": 720
    },
    {
      "entropy": 7.818372631072998,
      "epoch": 0.847953216374269,
      "grad_norm": 0.984375,
      "learning_rate": 0.000362,
      "loss": 7.518,
      "mean_token_accuracy": 0.1304582789540291,
      "num_tokens": 1803712.0,
      "step": 725
    },
    {
      "entropy": 7.721041631698609,
      "epoch": 0.8538011695906432,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0003645,
      "loss": 7.4694,
      "mean_token_accuracy": 0.13335771411657332,
      "num_tokens": 1818387.0,
      "step": 730
    },
    {
      "entropy": 7.743247652053833,
      "epoch": 0.8596491228070176,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000367,
      "loss": 7.4671,
      "mean_token_accuracy": 0.13619914576411246,
      "num_tokens": 1830113.0,
      "step": 735
    },
    {
      "entropy": 7.735281753540039,
      "epoch": 0.8654970760233918,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003695,
      "loss": 7.5761,
      "mean_token_accuracy": 0.13483594059944154,
      "num_tokens": 1841789.0,
      "step": 740
    },
    {
      "entropy": 7.668903923034668,
      "epoch": 0.8713450292397661,
      "grad_norm": 1.078125,
      "learning_rate": 0.000372,
      "loss": 7.4507,
      "mean_token_accuracy": 0.13731630146503448,
      "num_tokens": 1853669.0,
      "step": 745
    },
    {
      "entropy": 7.724067640304566,
      "epoch": 0.8771929824561403,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0003745,
      "loss": 7.5109,
      "mean_token_accuracy": 0.13437143638730048,
      "num_tokens": 1865322.0,
      "step": 750
    },
    {
      "entropy": 7.644190073013306,
      "epoch": 0.8830409356725146,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000377,
      "loss": 7.3942,
      "mean_token_accuracy": 0.1427222914993763,
      "num_tokens": 1879039.0,
      "step": 755
    },
    {
      "entropy": 7.785293626785278,
      "epoch": 0.8888888888888888,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003795,
      "loss": 7.5554,
      "mean_token_accuracy": 0.12891999259591103,
      "num_tokens": 1893078.0,
      "step": 760
    },
    {
      "entropy": 7.748642826080323,
      "epoch": 0.8947368421052632,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000382,
      "loss": 7.5767,
      "mean_token_accuracy": 0.12865981981158256,
      "num_tokens": 1907399.0,
      "step": 765
    },
    {
      "entropy": 7.704698705673218,
      "epoch": 0.9005847953216374,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003845,
      "loss": 7.5235,
      "mean_token_accuracy": 0.12384044975042344,
      "num_tokens": 1919290.0,
      "step": 770
    },
    {
      "entropy": 7.804563283920288,
      "epoch": 0.9064327485380117,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038700000000000003,
      "loss": 7.4842,
      "mean_token_accuracy": 0.12996885776519776,
      "num_tokens": 1931658.0,
      "step": 775
    },
    {
      "entropy": 7.707001209259033,
      "epoch": 0.9122807017543859,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038950000000000003,
      "loss": 7.4773,
      "mean_token_accuracy": 0.1285408228635788,
      "num_tokens": 1943881.0,
      "step": 780
    },
    {
      "entropy": 7.667708969116211,
      "epoch": 0.9181286549707602,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039200000000000004,
      "loss": 7.4053,
      "mean_token_accuracy": 0.1371392272412777,
      "num_tokens": 1956288.0,
      "step": 785
    },
    {
      "entropy": 7.6015739917755125,
      "epoch": 0.9239766081871345,
      "grad_norm": 1.125,
      "learning_rate": 0.00039450000000000005,
      "loss": 7.3066,
      "mean_token_accuracy": 0.13772179931402206,
      "num_tokens": 1967265.0,
      "step": 790
    },
    {
      "entropy": 7.597353601455689,
      "epoch": 0.9298245614035088,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039700000000000005,
      "loss": 7.447,
      "mean_token_accuracy": 0.13603045642375947,
      "num_tokens": 1980039.0,
      "step": 795
    },
    {
      "entropy": 7.690238094329834,
      "epoch": 0.935672514619883,
      "grad_norm": 0.921875,
      "learning_rate": 0.0003995,
      "loss": 7.4093,
      "mean_token_accuracy": 0.13160398453474045,
      "num_tokens": 1992735.0,
      "step": 800
    },
    {
      "entropy": 7.704154777526855,
      "epoch": 0.9415204678362573,
      "grad_norm": 1.0,
      "learning_rate": 0.000402,
      "loss": 7.4332,
      "mean_token_accuracy": 0.13839306458830833,
      "num_tokens": 2004545.0,
      "step": 805
    },
    {
      "entropy": 7.581313180923462,
      "epoch": 0.9473684210526315,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004045,
      "loss": 7.4323,
      "mean_token_accuracy": 0.13345621451735495,
      "num_tokens": 2017100.0,
      "step": 810
    },
    {
      "entropy": 7.68658595085144,
      "epoch": 0.9532163742690059,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00040699999999999997,
      "loss": 7.3636,
      "mean_token_accuracy": 0.14662561491131781,
      "num_tokens": 2029868.0,
      "step": 815
    },
    {
      "entropy": 7.646354150772095,
      "epoch": 0.9590643274853801,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004095,
      "loss": 7.4397,
      "mean_token_accuracy": 0.13421894684433938,
      "num_tokens": 2041334.0,
      "step": 820
    },
    {
      "entropy": 7.6902111053466795,
      "epoch": 0.9649122807017544,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000412,
      "loss": 7.5138,
      "mean_token_accuracy": 0.13264959007501603,
      "num_tokens": 2054797.0,
      "step": 825
    },
    {
      "entropy": 7.602681684494018,
      "epoch": 0.9707602339181286,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004145,
      "loss": 7.4045,
      "mean_token_accuracy": 0.13147154748439788,
      "num_tokens": 2068337.0,
      "step": 830
    },
    {
      "entropy": 7.694418144226074,
      "epoch": 0.9766081871345029,
      "grad_norm": 1.0,
      "learning_rate": 0.000417,
      "loss": 7.4365,
      "mean_token_accuracy": 0.13966115042567254,
      "num_tokens": 2079984.0,
      "step": 835
    },
    {
      "entropy": 7.612453842163086,
      "epoch": 0.9824561403508771,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004195,
      "loss": 7.4319,
      "mean_token_accuracy": 0.13970493376255036,
      "num_tokens": 2093069.0,
      "step": 840
    },
    {
      "entropy": 7.611123704910279,
      "epoch": 0.9883040935672515,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000422,
      "loss": 7.3801,
      "mean_token_accuracy": 0.13889345675706863,
      "num_tokens": 2104753.0,
      "step": 845
    },
    {
      "entropy": 7.604637861251831,
      "epoch": 0.9941520467836257,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004245,
      "loss": 7.391,
      "mean_token_accuracy": 0.13473013713955878,
      "num_tokens": 2116576.0,
      "step": 850
    },
    {
      "entropy": 7.654403877258301,
      "epoch": 1.0,
      "grad_norm": 1.015625,
      "learning_rate": 0.000427,
      "loss": 7.3747,
      "mean_token_accuracy": 0.13166687712073327,
      "num_tokens": 2129484.0,
      "step": 855
    },
    {
      "entropy": 7.537336349487305,
      "epoch": 1.0058479532163742,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004295,
      "loss": 7.2449,
      "mean_token_accuracy": 0.14783877283334732,
      "num_tokens": 2142000.0,
      "step": 860
    },
    {
      "entropy": 7.523657131195068,
      "epoch": 1.0116959064327484,
      "grad_norm": 1.046875,
      "learning_rate": 0.000432,
      "loss": 7.176,
      "mean_token_accuracy": 0.1486393131315708,
      "num_tokens": 2152965.0,
      "step": 865
    },
    {
      "entropy": 7.559857559204102,
      "epoch": 1.0175438596491229,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004345,
      "loss": 7.2756,
      "mean_token_accuracy": 0.13471193611621857,
      "num_tokens": 2165078.0,
      "step": 870
    },
    {
      "entropy": 7.430526161193848,
      "epoch": 1.023391812865497,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000437,
      "loss": 7.2031,
      "mean_token_accuracy": 0.14576031267642975,
      "num_tokens": 2176706.0,
      "step": 875
    },
    {
      "entropy": 7.623335981369019,
      "epoch": 1.0292397660818713,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004395,
      "loss": 7.2517,
      "mean_token_accuracy": 0.13940724432468415,
      "num_tokens": 2190157.0,
      "step": 880
    },
    {
      "entropy": 7.54054594039917,
      "epoch": 1.0350877192982457,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000442,
      "loss": 7.2133,
      "mean_token_accuracy": 0.14537803903222085,
      "num_tokens": 2201821.0,
      "step": 885
    },
    {
      "entropy": 7.468587160110474,
      "epoch": 1.04093567251462,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004445,
      "loss": 7.158,
      "mean_token_accuracy": 0.15055620521306992,
      "num_tokens": 2214504.0,
      "step": 890
    },
    {
      "entropy": 7.467399740219117,
      "epoch": 1.0467836257309941,
      "grad_norm": 0.953125,
      "learning_rate": 0.000447,
      "loss": 7.2665,
      "mean_token_accuracy": 0.14105323553085328,
      "num_tokens": 2226318.0,
      "step": 895
    },
    {
      "entropy": 7.555955314636231,
      "epoch": 1.0526315789473684,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00044950000000000003,
      "loss": 7.2829,
      "mean_token_accuracy": 0.13729312494397164,
      "num_tokens": 2237892.0,
      "step": 900
    },
    {
      "entropy": 7.526016616821289,
      "epoch": 1.0584795321637426,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045200000000000004,
      "loss": 7.2314,
      "mean_token_accuracy": 0.1435258150100708,
      "num_tokens": 2250656.0,
      "step": 905
    },
    {
      "entropy": 7.511065912246704,
      "epoch": 1.064327485380117,
      "grad_norm": 1.0,
      "learning_rate": 0.00045450000000000004,
      "loss": 7.2637,
      "mean_token_accuracy": 0.13697190061211587,
      "num_tokens": 2263685.0,
      "step": 910
    },
    {
      "entropy": 7.494118690490723,
      "epoch": 1.0701754385964912,
      "grad_norm": 1.0,
      "learning_rate": 0.00045700000000000005,
      "loss": 7.232,
      "mean_token_accuracy": 0.1364179164171219,
      "num_tokens": 2275413.0,
      "step": 915
    },
    {
      "entropy": 7.477536964416504,
      "epoch": 1.0760233918128654,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00045950000000000006,
      "loss": 7.2195,
      "mean_token_accuracy": 0.14017109125852584,
      "num_tokens": 2288165.0,
      "step": 920
    },
    {
      "entropy": 7.574978637695312,
      "epoch": 1.0818713450292399,
      "grad_norm": 1.015625,
      "learning_rate": 0.000462,
      "loss": 7.2053,
      "mean_token_accuracy": 0.13687562569975853,
      "num_tokens": 2300574.0,
      "step": 925
    },
    {
      "entropy": 7.416238403320312,
      "epoch": 1.087719298245614,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004645,
      "loss": 7.219,
      "mean_token_accuracy": 0.1380111664533615,
      "num_tokens": 2311631.0,
      "step": 930
    },
    {
      "entropy": 7.547397470474243,
      "epoch": 1.0935672514619883,
      "grad_norm": 1.09375,
      "learning_rate": 0.000467,
      "loss": 7.234,
      "mean_token_accuracy": 0.14326555654406548,
      "num_tokens": 2324312.0,
      "step": 935
    },
    {
      "entropy": 7.512025690078735,
      "epoch": 1.0994152046783625,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004695,
      "loss": 7.2855,
      "mean_token_accuracy": 0.13624050617218017,
      "num_tokens": 2337747.0,
      "step": 940
    },
    {
      "entropy": 7.413490962982178,
      "epoch": 1.1052631578947367,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000472,
      "loss": 7.2546,
      "mean_token_accuracy": 0.1424066536128521,
      "num_tokens": 2349830.0,
      "step": 945
    },
    {
      "entropy": 7.565069675445557,
      "epoch": 1.1111111111111112,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004745,
      "loss": 7.1993,
      "mean_token_accuracy": 0.14769526124000548,
      "num_tokens": 2361825.0,
      "step": 950
    },
    {
      "entropy": 7.369341516494751,
      "epoch": 1.1169590643274854,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000477,
      "loss": 7.1361,
      "mean_token_accuracy": 0.1449691839516163,
      "num_tokens": 2373963.0,
      "step": 955
    },
    {
      "entropy": 7.453041648864746,
      "epoch": 1.1228070175438596,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004795,
      "loss": 7.2795,
      "mean_token_accuracy": 0.13922639787197114,
      "num_tokens": 2386348.0,
      "step": 960
    },
    {
      "entropy": 7.489872980117798,
      "epoch": 1.128654970760234,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000482,
      "loss": 7.1985,
      "mean_token_accuracy": 0.13827377706766128,
      "num_tokens": 2398338.0,
      "step": 965
    },
    {
      "entropy": 7.514949655532837,
      "epoch": 1.1345029239766082,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004845,
      "loss": 7.1533,
      "mean_token_accuracy": 0.14474421963095666,
      "num_tokens": 2411855.0,
      "step": 970
    },
    {
      "entropy": 7.357582998275757,
      "epoch": 1.1403508771929824,
      "grad_norm": 1.0625,
      "learning_rate": 0.000487,
      "loss": 7.2126,
      "mean_token_accuracy": 0.14526815563440323,
      "num_tokens": 2424271.0,
      "step": 975
    },
    {
      "entropy": 7.440734958648681,
      "epoch": 1.1461988304093567,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004895,
      "loss": 7.2256,
      "mean_token_accuracy": 0.13997550830245017,
      "num_tokens": 2436668.0,
      "step": 980
    },
    {
      "entropy": 7.424071168899536,
      "epoch": 1.1520467836257309,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000492,
      "loss": 7.1748,
      "mean_token_accuracy": 0.1401802718639374,
      "num_tokens": 2449451.0,
      "step": 985
    },
    {
      "entropy": 7.5277848720550535,
      "epoch": 1.1578947368421053,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004945,
      "loss": 7.2966,
      "mean_token_accuracy": 0.13498533368110657,
      "num_tokens": 2463107.0,
      "step": 990
    },
    {
      "entropy": 7.464729595184326,
      "epoch": 1.1637426900584795,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000497,
      "loss": 7.2098,
      "mean_token_accuracy": 0.14099321886897087,
      "num_tokens": 2475443.0,
      "step": 995
    },
    {
      "entropy": 7.4347583770751955,
      "epoch": 1.1695906432748537,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004995,
      "loss": 7.1595,
      "mean_token_accuracy": 0.13942319229245187,
      "num_tokens": 2486884.0,
      "step": 1000
    },
    {
      "epoch": 1.1695906432748537,
      "eval_entropy": 7.396586781054472,
      "eval_loss": 7.327540397644043,
      "eval_mean_token_accuracy": 0.13058083411306143,
      "eval_num_tokens": 2486884.0,
      "eval_runtime": 1.7343,
      "eval_samples_per_second": 1560.818,
      "eval_steps_per_second": 195.463,
      "step": 1000
    },
    {
      "entropy": 7.410801887512207,
      "epoch": 1.1754385964912282,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999996875147963,
      "loss": 7.1007,
      "mean_token_accuracy": 0.15155821964144706,
      "num_tokens": 2499310.0,
      "step": 1005
    },
    {
      "entropy": 7.434916257858276,
      "epoch": 1.1812865497076024,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999984180451436,
      "loss": 7.2163,
      "mean_token_accuracy": 0.1340879425406456,
      "num_tokens": 2512230.0,
      "step": 1010
    },
    {
      "entropy": 7.411216306686401,
      "epoch": 1.1871345029239766,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999961720662221,
      "loss": 7.274,
      "mean_token_accuracy": 0.13820524960756303,
      "num_tokens": 2525255.0,
      "step": 1015
    },
    {
      "entropy": 7.425869369506836,
      "epoch": 1.1929824561403508,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999929495877796,
      "loss": 7.1618,
      "mean_token_accuracy": 0.14160262942314147,
      "num_tokens": 2537672.0,
      "step": 1020
    },
    {
      "entropy": 7.341161251068115,
      "epoch": 1.198830409356725,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999887506238018,
      "loss": 7.1885,
      "mean_token_accuracy": 0.14109369292855262,
      "num_tokens": 2550371.0,
      "step": 1025
    },
    {
      "entropy": 7.4244989395141605,
      "epoch": 1.2046783625730995,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999835751925124,
      "loss": 7.2077,
      "mean_token_accuracy": 0.13407954722642898,
      "num_tokens": 2563708.0,
      "step": 1030
    },
    {
      "entropy": 7.464514541625976,
      "epoch": 1.2105263157894737,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999774233163734,
      "loss": 7.1749,
      "mean_token_accuracy": 0.13515729755163192,
      "num_tokens": 2576529.0,
      "step": 1035
    },
    {
      "entropy": 7.382573938369751,
      "epoch": 1.2163742690058479,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999702950220841,
      "loss": 7.189,
      "mean_token_accuracy": 0.13476522713899614,
      "num_tokens": 2588970.0,
      "step": 1040
    },
    {
      "entropy": 7.370054435729981,
      "epoch": 1.2222222222222223,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999621903405819,
      "loss": 7.126,
      "mean_token_accuracy": 0.1400519900023937,
      "num_tokens": 2601134.0,
      "step": 1045
    },
    {
      "entropy": 7.446572351455688,
      "epoch": 1.2280701754385965,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000499953109307042,
      "loss": 7.1968,
      "mean_token_accuracy": 0.13780457451939582,
      "num_tokens": 2612670.0,
      "step": 1050
    },
    {
      "entropy": 7.330738973617554,
      "epoch": 1.2339181286549707,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999430519608762,
      "loss": 7.1517,
      "mean_token_accuracy": 0.1500091567635536,
      "num_tokens": 2625157.0,
      "step": 1055
    },
    {
      "entropy": 7.424869203567505,
      "epoch": 1.239766081871345,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999320183457346,
      "loss": 7.1486,
      "mean_token_accuracy": 0.1392211377620697,
      "num_tokens": 2638318.0,
      "step": 1060
    },
    {
      "entropy": 7.385347890853882,
      "epoch": 1.2456140350877192,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999200085095035,
      "loss": 7.134,
      "mean_token_accuracy": 0.1431312918663025,
      "num_tokens": 2651024.0,
      "step": 1065
    },
    {
      "entropy": 7.197023820877075,
      "epoch": 1.2514619883040936,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999070225043068,
      "loss": 7.0639,
      "mean_token_accuracy": 0.14909755140542985,
      "num_tokens": 2665568.0,
      "step": 1070
    },
    {
      "entropy": 7.542721223831177,
      "epoch": 1.2573099415204678,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998930603865042,
      "loss": 7.1996,
      "mean_token_accuracy": 0.13743996322155,
      "num_tokens": 2678748.0,
      "step": 1075
    },
    {
      "entropy": 7.32189826965332,
      "epoch": 1.263157894736842,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004998781222166929,
      "loss": 7.1173,
      "mean_token_accuracy": 0.14156942144036294,
      "num_tokens": 2690682.0,
      "step": 1080
    },
    {
      "entropy": 7.289649391174317,
      "epoch": 1.2690058479532165,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998622080597052,
      "loss": 7.1589,
      "mean_token_accuracy": 0.13398671224713327,
      "num_tokens": 2703077.0,
      "step": 1085
    },
    {
      "entropy": 7.3387401580810545,
      "epoch": 1.2748538011695907,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998453179846098,
      "loss": 7.1259,
      "mean_token_accuracy": 0.13829808682203293,
      "num_tokens": 2714609.0,
      "step": 1090
    },
    {
      "entropy": 7.396260023117065,
      "epoch": 1.280701754385965,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998274520647109,
      "loss": 7.0918,
      "mean_token_accuracy": 0.1457760065793991,
      "num_tokens": 2726907.0,
      "step": 1095
    },
    {
      "entropy": 7.268807315826416,
      "epoch": 1.286549707602339,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998086103775478,
      "loss": 7.1088,
      "mean_token_accuracy": 0.13768557012081145,
      "num_tokens": 2739248.0,
      "step": 1100
    },
    {
      "entropy": 7.356909608840942,
      "epoch": 1.2923976608187133,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004997887930048947,
      "loss": 7.0384,
      "mean_token_accuracy": 0.1473199874162674,
      "num_tokens": 2753842.0,
      "step": 1105
    },
    {
      "entropy": 7.265995216369629,
      "epoch": 1.2982456140350878,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997680000327607,
      "loss": 7.1389,
      "mean_token_accuracy": 0.13432746306061744,
      "num_tokens": 2767249.0,
      "step": 1110
    },
    {
      "entropy": 7.400834608078003,
      "epoch": 1.304093567251462,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997462315513884,
      "loss": 7.117,
      "mean_token_accuracy": 0.14291927963495255,
      "num_tokens": 2779657.0,
      "step": 1115
    },
    {
      "entropy": 7.343295621871948,
      "epoch": 1.3099415204678362,
      "grad_norm": 1.6484375,
      "learning_rate": 0.000499723487655255,
      "loss": 7.0373,
      "mean_token_accuracy": 0.14398355707526206,
      "num_tokens": 2792760.0,
      "step": 1120
    },
    {
      "entropy": 7.285030221939087,
      "epoch": 1.3157894736842106,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004996997684430703,
      "loss": 7.0528,
      "mean_token_accuracy": 0.14452853202819824,
      "num_tokens": 2805277.0,
      "step": 1125
    },
    {
      "entropy": 7.263354873657226,
      "epoch": 1.3216374269005848,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996750740177775,
      "loss": 7.157,
      "mean_token_accuracy": 0.14185430854558945,
      "num_tokens": 2818341.0,
      "step": 1130
    },
    {
      "entropy": 7.370803260803223,
      "epoch": 1.327485380116959,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996494044865521,
      "loss": 7.118,
      "mean_token_accuracy": 0.1415998138487339,
      "num_tokens": 2830199.0,
      "step": 1135
    },
    {
      "entropy": 7.131627798080444,
      "epoch": 1.3333333333333333,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996227599608018,
      "loss": 7.0109,
      "mean_token_accuracy": 0.1468503713607788,
      "num_tokens": 2842141.0,
      "step": 1140
    },
    {
      "entropy": 7.376796817779541,
      "epoch": 1.3391812865497075,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004995951405561658,
      "loss": 7.087,
      "mean_token_accuracy": 0.1350610874593258,
      "num_tokens": 2854929.0,
      "step": 1145
    },
    {
      "entropy": 7.256033039093017,
      "epoch": 1.345029239766082,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004995665463925142,
      "loss": 7.0946,
      "mean_token_accuracy": 0.14368716776371002,
      "num_tokens": 2867034.0,
      "step": 1150
    },
    {
      "entropy": 7.244560575485229,
      "epoch": 1.3508771929824561,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004995369775939478,
      "loss": 6.9939,
      "mean_token_accuracy": 0.1505236506462097,
      "num_tokens": 2881254.0,
      "step": 1155
    },
    {
      "entropy": 7.286165046691894,
      "epoch": 1.3567251461988303,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004995064342887972,
      "loss": 7.102,
      "mean_token_accuracy": 0.14152786880731583,
      "num_tokens": 2894592.0,
      "step": 1160
    },
    {
      "entropy": 7.240753030776977,
      "epoch": 1.3625730994152048,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004994749166096226,
      "loss": 7.0334,
      "mean_token_accuracy": 0.14414507299661636,
      "num_tokens": 2907196.0,
      "step": 1165
    },
    {
      "entropy": 7.277744579315185,
      "epoch": 1.368421052631579,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004994424246932129,
      "loss": 7.0989,
      "mean_token_accuracy": 0.1371539294719696,
      "num_tokens": 2920398.0,
      "step": 1170
    },
    {
      "entropy": 7.242294979095459,
      "epoch": 1.3742690058479532,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004994089586805856,
      "loss": 7.0167,
      "mean_token_accuracy": 0.14515721797943115,
      "num_tokens": 2932052.0,
      "step": 1175
    },
    {
      "entropy": 7.358625030517578,
      "epoch": 1.3801169590643274,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993745187169852,
      "loss": 7.0886,
      "mean_token_accuracy": 0.1365416131913662,
      "num_tokens": 2943620.0,
      "step": 1180
    },
    {
      "entropy": 7.208013010025025,
      "epoch": 1.3859649122807016,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499339104951884,
      "loss": 7.0413,
      "mean_token_accuracy": 0.1536354199051857,
      "num_tokens": 2956159.0,
      "step": 1185
    },
    {
      "entropy": 7.2751524448394775,
      "epoch": 1.391812865497076,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00049930271753898,
      "loss": 7.0557,
      "mean_token_accuracy": 0.1427343636751175,
      "num_tokens": 2967332.0,
      "step": 1190
    },
    {
      "entropy": 7.205974292755127,
      "epoch": 1.3976608187134503,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004992653566361974,
      "loss": 7.0593,
      "mean_token_accuracy": 0.13768181428313256,
      "num_tokens": 2980168.0,
      "step": 1195
    },
    {
      "entropy": 7.358188152313232,
      "epoch": 1.4035087719298245,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004992270224056849,
      "loss": 7.121,
      "mean_token_accuracy": 0.13874844163656236,
      "num_tokens": 2993509.0,
      "step": 1200
    },
    {
      "entropy": 7.197741508483887,
      "epoch": 1.409356725146199,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004991877150138162,
      "loss": 7.0636,
      "mean_token_accuracy": 0.14075298309326173,
      "num_tokens": 3005827.0,
      "step": 1205
    },
    {
      "entropy": 7.2247803688049315,
      "epoch": 1.4152046783625731,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004991474346311879,
      "loss": 7.0159,
      "mean_token_accuracy": 0.1467432513833046,
      "num_tokens": 3018968.0,
      "step": 1210
    },
    {
      "entropy": 7.296478223800659,
      "epoch": 1.4210526315789473,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004991061814326198,
      "loss": 7.0713,
      "mean_token_accuracy": 0.13609698042273521,
      "num_tokens": 3031892.0,
      "step": 1215
    },
    {
      "entropy": 7.236675786972046,
      "epoch": 1.4269005847953216,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004990639555971539,
      "loss": 6.9919,
      "mean_token_accuracy": 0.1513255462050438,
      "num_tokens": 3045994.0,
      "step": 1220
    },
    {
      "entropy": 7.166168117523194,
      "epoch": 1.4327485380116958,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004990207573080533,
      "loss": 7.0079,
      "mean_token_accuracy": 0.14798654168844222,
      "num_tokens": 3058304.0,
      "step": 1225
    },
    {
      "entropy": 7.150466823577881,
      "epoch": 1.4385964912280702,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004989765867528014,
      "loss": 7.0027,
      "mean_token_accuracy": 0.14707266688346862,
      "num_tokens": 3071102.0,
      "step": 1230
    },
    {
      "entropy": 7.154991912841797,
      "epoch": 1.4444444444444444,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004989314441231019,
      "loss": 7.0306,
      "mean_token_accuracy": 0.139088436961174,
      "num_tokens": 3084653.0,
      "step": 1235
    },
    {
      "entropy": 7.244064474105835,
      "epoch": 1.4502923976608186,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000498885329614877,
      "loss": 7.0558,
      "mean_token_accuracy": 0.1452793374657631,
      "num_tokens": 3097425.0,
      "step": 1240
    },
    {
      "entropy": 7.233583593368531,
      "epoch": 1.456140350877193,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004988382434282668,
      "loss": 7.0555,
      "mean_token_accuracy": 0.14807011559605598,
      "num_tokens": 3111898.0,
      "step": 1245
    },
    {
      "entropy": 7.276054811477661,
      "epoch": 1.4619883040935673,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000498790185767629,
      "loss": 6.9923,
      "mean_token_accuracy": 0.14536793828010558,
      "num_tokens": 3124271.0,
      "step": 1250
    },
    {
      "entropy": 7.214261388778686,
      "epoch": 1.4678362573099415,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000498741156841537,
      "loss": 7.0941,
      "mean_token_accuracy": 0.14358112215995789,
      "num_tokens": 3136344.0,
      "step": 1255
    },
    {
      "entropy": 7.236997652053833,
      "epoch": 1.4736842105263157,
      "grad_norm": 1.0,
      "learning_rate": 0.00049869115686278,
      "loss": 6.983,
      "mean_token_accuracy": 0.14811330884695054,
      "num_tokens": 3148299.0,
      "step": 1260
    },
    {
      "entropy": 7.2414836406707765,
      "epoch": 1.47953216374269,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004986401860483615,
      "loss": 7.0474,
      "mean_token_accuracy": 0.14126647412776946,
      "num_tokens": 3161537.0,
      "step": 1265
    },
    {
      "entropy": 7.195467853546143,
      "epoch": 1.4853801169590644,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004985882446194984,
      "loss": 6.9968,
      "mean_token_accuracy": 0.1453041210770607,
      "num_tokens": 3173676.0,
      "step": 1270
    },
    {
      "entropy": 7.2394012928009035,
      "epoch": 1.4912280701754386,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004985353328016204,
      "loss": 6.9508,
      "mean_token_accuracy": 0.15483074486255646,
      "num_tokens": 3186542.0,
      "step": 1275
    },
    {
      "entropy": 7.1377181053161625,
      "epoch": 1.4970760233918128,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004984814508243685,
      "loss": 6.9925,
      "mean_token_accuracy": 0.1458618648350239,
      "num_tokens": 3197867.0,
      "step": 1280
    },
    {
      "entropy": 7.244440937042237,
      "epoch": 1.5029239766081872,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004984265989215942,
      "loss": 6.9531,
      "mean_token_accuracy": 0.1449991211295128,
      "num_tokens": 3209671.0,
      "step": 1285
    },
    {
      "entropy": 7.145474100112915,
      "epoch": 1.5087719298245614,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004983707773313589,
      "loss": 6.953,
      "mean_token_accuracy": 0.14984701722860336,
      "num_tokens": 3220707.0,
      "step": 1290
    },
    {
      "entropy": 7.075493192672729,
      "epoch": 1.5146198830409356,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004983139862959322,
      "loss": 6.9205,
      "mean_token_accuracy": 0.15351321399211884,
      "num_tokens": 3235122.0,
      "step": 1295
    },
    {
      "entropy": 7.246700382232666,
      "epoch": 1.52046783625731,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004982562260617912,
      "loss": 6.91,
      "mean_token_accuracy": 0.14220256134867668,
      "num_tokens": 3246833.0,
      "step": 1300
    },
    {
      "entropy": 7.16511902809143,
      "epoch": 1.526315789473684,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004981974968796197,
      "loss": 7.0177,
      "mean_token_accuracy": 0.14064179733395576,
      "num_tokens": 3260493.0,
      "step": 1305
    },
    {
      "entropy": 7.1622607707977295,
      "epoch": 1.5321637426900585,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004981377990043064,
      "loss": 6.9473,
      "mean_token_accuracy": 0.14261056780815123,
      "num_tokens": 3272445.0,
      "step": 1310
    },
    {
      "entropy": 7.171639442443848,
      "epoch": 1.5380116959064327,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004980771326949442,
      "loss": 6.9697,
      "mean_token_accuracy": 0.14227717518806457,
      "num_tokens": 3284360.0,
      "step": 1315
    },
    {
      "entropy": 7.176844310760498,
      "epoch": 1.543859649122807,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004980154982148296,
      "loss": 6.95,
      "mean_token_accuracy": 0.1433100312948227,
      "num_tokens": 3296841.0,
      "step": 1320
    },
    {
      "entropy": 7.11176643371582,
      "epoch": 1.5497076023391814,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004979528958314604,
      "loss": 6.9468,
      "mean_token_accuracy": 0.13953165858983993,
      "num_tokens": 3309393.0,
      "step": 1325
    },
    {
      "entropy": 7.238583707809449,
      "epoch": 1.5555555555555556,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004978893258165354,
      "loss": 7.0242,
      "mean_token_accuracy": 0.1429157868027687,
      "num_tokens": 3322094.0,
      "step": 1330
    },
    {
      "entropy": 7.1003436088562015,
      "epoch": 1.5614035087719298,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004978247884459532,
      "loss": 6.9428,
      "mean_token_accuracy": 0.1477564662694931,
      "num_tokens": 3334468.0,
      "step": 1335
    },
    {
      "entropy": 7.157374477386474,
      "epoch": 1.5672514619883042,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004977592839998105,
      "loss": 6.9287,
      "mean_token_accuracy": 0.14433884397149085,
      "num_tokens": 3346367.0,
      "step": 1340
    },
    {
      "entropy": 7.058119916915894,
      "epoch": 1.5730994152046782,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004976928127624013,
      "loss": 6.9132,
      "mean_token_accuracy": 0.15283917635679245,
      "num_tokens": 3357463.0,
      "step": 1345
    },
    {
      "entropy": 7.154469633102417,
      "epoch": 1.5789473684210527,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004976253750222157,
      "loss": 6.9025,
      "mean_token_accuracy": 0.14609844461083413,
      "num_tokens": 3369465.0,
      "step": 1350
    },
    {
      "entropy": 7.102024841308594,
      "epoch": 1.5847953216374269,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004975569710719382,
      "loss": 7.0104,
      "mean_token_accuracy": 0.13702175095677377,
      "num_tokens": 3382859.0,
      "step": 1355
    },
    {
      "entropy": 7.226300191879273,
      "epoch": 1.590643274853801,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004974876012084468,
      "loss": 6.966,
      "mean_token_accuracy": 0.1419912114739418,
      "num_tokens": 3395431.0,
      "step": 1360
    },
    {
      "entropy": 7.105293655395508,
      "epoch": 1.5964912280701755,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004974172657328116,
      "loss": 6.8652,
      "mean_token_accuracy": 0.14728413820266723,
      "num_tokens": 3408388.0,
      "step": 1365
    },
    {
      "entropy": 7.1052686214447025,
      "epoch": 1.6023391812865497,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004973459649502939,
      "loss": 6.8683,
      "mean_token_accuracy": 0.15123746544122696,
      "num_tokens": 3421275.0,
      "step": 1370
    },
    {
      "entropy": 7.077537584304809,
      "epoch": 1.608187134502924,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004972736991703439,
      "loss": 6.9488,
      "mean_token_accuracy": 0.14620793908834456,
      "num_tokens": 3433537.0,
      "step": 1375
    },
    {
      "entropy": 7.12898850440979,
      "epoch": 1.6140350877192984,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004972004687066005,
      "loss": 6.909,
      "mean_token_accuracy": 0.1481366164982319,
      "num_tokens": 3445167.0,
      "step": 1380
    },
    {
      "entropy": 7.038471460342407,
      "epoch": 1.6198830409356724,
      "grad_norm": 0.890625,
      "learning_rate": 0.000497126273876889,
      "loss": 6.9066,
      "mean_token_accuracy": 0.15611805990338326,
      "num_tokens": 3458529.0,
      "step": 1385
    },
    {
      "entropy": 7.139416265487671,
      "epoch": 1.6257309941520468,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004970511150032205,
      "loss": 6.9038,
      "mean_token_accuracy": 0.1489902138710022,
      "num_tokens": 3470315.0,
      "step": 1390
    },
    {
      "entropy": 7.0139429569244385,
      "epoch": 1.631578947368421,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004969749924117896,
      "loss": 6.8256,
      "mean_token_accuracy": 0.14668601006269455,
      "num_tokens": 3481899.0,
      "step": 1395
    },
    {
      "entropy": 7.0898150444030765,
      "epoch": 1.6374269005847952,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004968979064329741,
      "loss": 6.9273,
      "mean_token_accuracy": 0.15040379613637925,
      "num_tokens": 3494076.0,
      "step": 1400
    },
    {
      "entropy": 7.236377954483032,
      "epoch": 1.6432748538011697,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004968198574013323,
      "loss": 6.9515,
      "mean_token_accuracy": 0.13935640305280686,
      "num_tokens": 3507670.0,
      "step": 1405
    },
    {
      "entropy": 7.092464876174927,
      "epoch": 1.6491228070175439,
      "grad_norm": 0.921875,
      "learning_rate": 0.000496740845655603,
      "loss": 6.8558,
      "mean_token_accuracy": 0.14918479546904564,
      "num_tokens": 3519704.0,
      "step": 1410
    },
    {
      "entropy": 7.035297775268555,
      "epoch": 1.654970760233918,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004966608715387029,
      "loss": 6.9247,
      "mean_token_accuracy": 0.1436144933104515,
      "num_tokens": 3531665.0,
      "step": 1415
    },
    {
      "entropy": 6.969947719573975,
      "epoch": 1.6608187134502925,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004965799353977248,
      "loss": 6.9144,
      "mean_token_accuracy": 0.15216819047927857,
      "num_tokens": 3543876.0,
      "step": 1420
    },
    {
      "entropy": 7.078910541534424,
      "epoch": 1.6666666666666665,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004964980375839379,
      "loss": 6.8397,
      "mean_token_accuracy": 0.14977291375398635,
      "num_tokens": 3556517.0,
      "step": 1425
    },
    {
      "entropy": 7.047800302505493,
      "epoch": 1.672514619883041,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004964151784527844,
      "loss": 6.8604,
      "mean_token_accuracy": 0.15354330390691756,
      "num_tokens": 3569378.0,
      "step": 1430
    },
    {
      "entropy": 7.156952142715454,
      "epoch": 1.6783625730994152,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000496331358363879,
      "loss": 6.8541,
      "mean_token_accuracy": 0.14472813308238983,
      "num_tokens": 3581684.0,
      "step": 1435
    },
    {
      "entropy": 6.917730379104614,
      "epoch": 1.6842105263157894,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004962465776810066,
      "loss": 6.7814,
      "mean_token_accuracy": 0.16086088120937347,
      "num_tokens": 3594961.0,
      "step": 1440
    },
    {
      "entropy": 7.072372961044311,
      "epoch": 1.6900584795321638,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004961608367721217,
      "loss": 6.8544,
      "mean_token_accuracy": 0.14478224515914917,
      "num_tokens": 3607197.0,
      "step": 1445
    },
    {
      "entropy": 7.065153741836548,
      "epoch": 1.695906432748538,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004960741360093459,
      "loss": 6.8769,
      "mean_token_accuracy": 0.14579440355300904,
      "num_tokens": 3618728.0,
      "step": 1450
    },
    {
      "entropy": 7.111606073379517,
      "epoch": 1.7017543859649122,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004959864757689669,
      "loss": 6.9783,
      "mean_token_accuracy": 0.13519750013947487,
      "num_tokens": 3630872.0,
      "step": 1455
    },
    {
      "entropy": 7.088176918029785,
      "epoch": 1.7076023391812867,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004958978564314361,
      "loss": 6.8708,
      "mean_token_accuracy": 0.14870110005140305,
      "num_tokens": 3642657.0,
      "step": 1460
    },
    {
      "entropy": 7.056329298019409,
      "epoch": 1.7134502923976607,
      "grad_norm": 0.96875,
      "learning_rate": 0.000495808278381368,
      "loss": 6.8827,
      "mean_token_accuracy": 0.14525013267993928,
      "num_tokens": 3654505.0,
      "step": 1465
    },
    {
      "entropy": 6.989843940734863,
      "epoch": 1.719298245614035,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004957177420075377,
      "loss": 6.7841,
      "mean_token_accuracy": 0.16019067466259002,
      "num_tokens": 3668554.0,
      "step": 1470
    },
    {
      "entropy": 7.060316467285157,
      "epoch": 1.7251461988304093,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004956262477028797,
      "loss": 6.9507,
      "mean_token_accuracy": 0.14162230864167213,
      "num_tokens": 3681448.0,
      "step": 1475
    },
    {
      "entropy": 7.1160413265228275,
      "epoch": 1.7309941520467835,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004955337958644857,
      "loss": 6.8709,
      "mean_token_accuracy": 0.1425687327980995,
      "num_tokens": 3693677.0,
      "step": 1480
    },
    {
      "entropy": 6.9832135200500485,
      "epoch": 1.736842105263158,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004954403868936032,
      "loss": 6.8683,
      "mean_token_accuracy": 0.15228217393159865,
      "num_tokens": 3706204.0,
      "step": 1485
    },
    {
      "entropy": 7.108418083190918,
      "epoch": 1.7426900584795322,
      "grad_norm": 0.984375,
      "learning_rate": 0.000495346021195634,
      "loss": 6.8412,
      "mean_token_accuracy": 0.15044791102409363,
      "num_tokens": 3718255.0,
      "step": 1490
    },
    {
      "entropy": 7.084663152694702,
      "epoch": 1.7485380116959064,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004952506991801319,
      "loss": 6.8399,
      "mean_token_accuracy": 0.1522560551762581,
      "num_tokens": 3730623.0,
      "step": 1495
    },
    {
      "entropy": 7.022539806365967,
      "epoch": 1.7543859649122808,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004951544212608015,
      "loss": 6.8765,
      "mean_token_accuracy": 0.14687300175428392,
      "num_tokens": 3743636.0,
      "step": 1500
    },
    {
      "epoch": 1.7543859649122808,
      "eval_entropy": 6.913726612529923,
      "eval_loss": 6.976031303405762,
      "eval_mean_token_accuracy": 0.13840462224816238,
      "eval_num_tokens": 3743636.0,
      "eval_runtime": 1.7356,
      "eval_samples_per_second": 1559.729,
      "eval_steps_per_second": 195.326,
      "step": 1500
    },
    {
      "entropy": 7.135053968429565,
      "epoch": 1.7602339181286548,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004950571878554956,
      "loss": 6.9312,
      "mean_token_accuracy": 0.13546228855848313,
      "num_tokens": 3755608.0,
      "step": 1505
    },
    {
      "entropy": 7.097068405151367,
      "epoch": 1.7660818713450293,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004949589993862143,
      "loss": 6.8152,
      "mean_token_accuracy": 0.14684600085020066,
      "num_tokens": 3768208.0,
      "step": 1510
    },
    {
      "entropy": 6.97324047088623,
      "epoch": 1.7719298245614035,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004948598562791025,
      "loss": 6.8378,
      "mean_token_accuracy": 0.14705169945955276,
      "num_tokens": 3780550.0,
      "step": 1515
    },
    {
      "entropy": 7.07040581703186,
      "epoch": 1.7777777777777777,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004947597589644484,
      "loss": 6.8259,
      "mean_token_accuracy": 0.14792063385248183,
      "num_tokens": 3792417.0,
      "step": 1520
    },
    {
      "entropy": 7.051655101776123,
      "epoch": 1.7836257309941521,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004946587078766818,
      "loss": 6.807,
      "mean_token_accuracy": 0.14990664571523665,
      "num_tokens": 3803887.0,
      "step": 1525
    },
    {
      "entropy": 7.055767679214478,
      "epoch": 1.7894736842105263,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004945567034543712,
      "loss": 6.9462,
      "mean_token_accuracy": 0.13744124248623849,
      "num_tokens": 3816534.0,
      "step": 1530
    },
    {
      "entropy": 7.054939889907837,
      "epoch": 1.7953216374269005,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004944537461402234,
      "loss": 6.8877,
      "mean_token_accuracy": 0.149474073946476,
      "num_tokens": 3827790.0,
      "step": 1535
    },
    {
      "entropy": 7.062603187561035,
      "epoch": 1.801169590643275,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004943498363810807,
      "loss": 6.8293,
      "mean_token_accuracy": 0.15547692626714707,
      "num_tokens": 3839800.0,
      "step": 1540
    },
    {
      "entropy": 6.857793188095092,
      "epoch": 1.807017543859649,
      "grad_norm": 1.125,
      "learning_rate": 0.0004942449746279184,
      "loss": 6.7816,
      "mean_token_accuracy": 0.1539284199476242,
      "num_tokens": 3852621.0,
      "step": 1545
    },
    {
      "entropy": 7.072565126419067,
      "epoch": 1.8128654970760234,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004941391613358444,
      "loss": 6.8383,
      "mean_token_accuracy": 0.14228285551071168,
      "num_tokens": 3864184.0,
      "step": 1550
    },
    {
      "entropy": 7.071480751037598,
      "epoch": 1.8187134502923976,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000494032396964096,
      "loss": 6.8153,
      "mean_token_accuracy": 0.1449595272541046,
      "num_tokens": 3876750.0,
      "step": 1555
    },
    {
      "entropy": 7.053122901916504,
      "epoch": 1.8245614035087718,
      "grad_norm": 0.984375,
      "learning_rate": 0.000493924681976038,
      "loss": 6.8419,
      "mean_token_accuracy": 0.15216735750436783,
      "num_tokens": 3888788.0,
      "step": 1560
    },
    {
      "entropy": 6.933975648880005,
      "epoch": 1.8304093567251463,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004938160168391614,
      "loss": 6.8979,
      "mean_token_accuracy": 0.14304276704788207,
      "num_tokens": 3901816.0,
      "step": 1565
    },
    {
      "entropy": 7.026661539077759,
      "epoch": 1.8362573099415205,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004937064020250804,
      "loss": 6.7946,
      "mean_token_accuracy": 0.15177498906850814,
      "num_tokens": 3913127.0,
      "step": 1570
    },
    {
      "entropy": 7.026135396957398,
      "epoch": 1.8421052631578947,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004935958380095316,
      "loss": 6.8105,
      "mean_token_accuracy": 0.1479490578174591,
      "num_tokens": 3924878.0,
      "step": 1575
    },
    {
      "entropy": 6.993627643585205,
      "epoch": 1.8479532163742691,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004934843252723702,
      "loss": 6.886,
      "mean_token_accuracy": 0.14974023178219795,
      "num_tokens": 3938334.0,
      "step": 1580
    },
    {
      "entropy": 7.041214179992676,
      "epoch": 1.8538011695906431,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00049337186429757,
      "loss": 6.8473,
      "mean_token_accuracy": 0.14718355536460875,
      "num_tokens": 3949733.0,
      "step": 1585
    },
    {
      "entropy": 7.096500444412231,
      "epoch": 1.8596491228070176,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004932584555732192,
      "loss": 6.872,
      "mean_token_accuracy": 0.14800208285450936,
      "num_tokens": 3963109.0,
      "step": 1590
    },
    {
      "entropy": 7.084401702880859,
      "epoch": 1.8654970760233918,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004931440995915203,
      "loss": 6.845,
      "mean_token_accuracy": 0.14705388098955155,
      "num_tokens": 3975477.0,
      "step": 1595
    },
    {
      "entropy": 6.9311549186706545,
      "epoch": 1.871345029239766,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000493028796848786,
      "loss": 6.8107,
      "mean_token_accuracy": 0.15162300392985345,
      "num_tokens": 3987757.0,
      "step": 1600
    },
    {
      "entropy": 7.065761947631836,
      "epoch": 1.8771929824561404,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004929125478454389,
      "loss": 6.7054,
      "mean_token_accuracy": 0.15114182680845262,
      "num_tokens": 3998807.0,
      "step": 1605
    },
    {
      "entropy": 6.891133451461792,
      "epoch": 1.8830409356725146,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004927953530860077,
      "loss": 6.7646,
      "mean_token_accuracy": 0.15039825066924095,
      "num_tokens": 4010320.0,
      "step": 1610
    },
    {
      "entropy": 7.068708038330078,
      "epoch": 1.8888888888888888,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004926772130791261,
      "loss": 6.8999,
      "mean_token_accuracy": 0.1383692055940628,
      "num_tokens": 4023432.0,
      "step": 1615
    },
    {
      "entropy": 6.984898090362549,
      "epoch": 1.8947368421052633,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004925581283375303,
      "loss": 6.8767,
      "mean_token_accuracy": 0.14132775589823723,
      "num_tokens": 4036282.0,
      "step": 1620
    },
    {
      "entropy": 6.9457885265350345,
      "epoch": 1.9005847953216373,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004924380993780566,
      "loss": 6.7499,
      "mean_token_accuracy": 0.1543327748775482,
      "num_tokens": 4048164.0,
      "step": 1625
    },
    {
      "entropy": 6.9637031078338625,
      "epoch": 1.9064327485380117,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004923171267216392,
      "loss": 6.7939,
      "mean_token_accuracy": 0.14861909225583075,
      "num_tokens": 4059287.0,
      "step": 1630
    },
    {
      "entropy": 6.972118234634399,
      "epoch": 1.912280701754386,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004921952108933083,
      "loss": 6.7804,
      "mean_token_accuracy": 0.14669928476214408,
      "num_tokens": 4071976.0,
      "step": 1635
    },
    {
      "entropy": 7.002675342559814,
      "epoch": 1.9181286549707601,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004920723524221871,
      "loss": 6.8152,
      "mean_token_accuracy": 0.15132055059075356,
      "num_tokens": 4085116.0,
      "step": 1640
    },
    {
      "entropy": 7.055388784408569,
      "epoch": 1.9239766081871346,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004919485518414902,
      "loss": 6.812,
      "mean_token_accuracy": 0.14319210201501847,
      "num_tokens": 4096778.0,
      "step": 1645
    },
    {
      "entropy": 6.902815294265747,
      "epoch": 1.9298245614035088,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004918238096885213,
      "loss": 6.7508,
      "mean_token_accuracy": 0.1475529596209526,
      "num_tokens": 4109490.0,
      "step": 1650
    },
    {
      "entropy": 6.984807109832763,
      "epoch": 1.935672514619883,
      "grad_norm": 0.83984375,
      "learning_rate": 0.00049169812650467,
      "loss": 6.773,
      "mean_token_accuracy": 0.15039625018835068,
      "num_tokens": 4122832.0,
      "step": 1655
    },
    {
      "entropy": 7.0580432415008545,
      "epoch": 1.9415204678362574,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004915715028354102,
      "loss": 6.7956,
      "mean_token_accuracy": 0.1466011457145214,
      "num_tokens": 4134585.0,
      "step": 1660
    },
    {
      "entropy": 6.965006637573242,
      "epoch": 1.9473684210526314,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004914439392302981,
      "loss": 6.7469,
      "mean_token_accuracy": 0.1464242994785309,
      "num_tokens": 4145909.0,
      "step": 1665
    },
    {
      "entropy": 6.98766508102417,
      "epoch": 1.9532163742690059,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004913154362429686,
      "loss": 6.6539,
      "mean_token_accuracy": 0.16458579450845717,
      "num_tokens": 4157360.0,
      "step": 1670
    },
    {
      "entropy": 6.8865578174591064,
      "epoch": 1.95906432748538,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004911859944311341,
      "loss": 6.8376,
      "mean_token_accuracy": 0.14514545798301698,
      "num_tokens": 4169789.0,
      "step": 1675
    },
    {
      "entropy": 7.0449613571167,
      "epoch": 1.9649122807017543,
      "grad_norm": 1.0,
      "learning_rate": 0.0004910556143565812,
      "loss": 6.8308,
      "mean_token_accuracy": 0.1438768610358238,
      "num_tokens": 4183131.0,
      "step": 1680
    },
    {
      "entropy": 6.966651821136475,
      "epoch": 1.9707602339181287,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004909242965851688,
      "loss": 6.8105,
      "mean_token_accuracy": 0.14589049071073532,
      "num_tokens": 4194865.0,
      "step": 1685
    },
    {
      "entropy": 6.988571405410767,
      "epoch": 1.976608187134503,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004907920416868253,
      "loss": 6.8019,
      "mean_token_accuracy": 0.14565620571374893,
      "num_tokens": 4207406.0,
      "step": 1690
    },
    {
      "entropy": 6.944311332702637,
      "epoch": 1.9824561403508771,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004906588502355467,
      "loss": 6.7757,
      "mean_token_accuracy": 0.14800650775432586,
      "num_tokens": 4221623.0,
      "step": 1695
    },
    {
      "entropy": 6.886150646209717,
      "epoch": 1.9883040935672516,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004905247228093933,
      "loss": 6.7147,
      "mean_token_accuracy": 0.16161981076002122,
      "num_tokens": 4233699.0,
      "step": 1700
    },
    {
      "entropy": 6.940931224822998,
      "epoch": 1.9941520467836256,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004903896599904878,
      "loss": 6.7046,
      "mean_token_accuracy": 0.15220564156770705,
      "num_tokens": 4247669.0,
      "step": 1705
    },
    {
      "entropy": 6.927793836593628,
      "epoch": 2.0,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004902536623650125,
      "loss": 6.7571,
      "mean_token_accuracy": 0.14728414714336396,
      "num_tokens": 4258968.0,
      "step": 1710
    },
    {
      "entropy": 6.963106155395508,
      "epoch": 2.0058479532163744,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004901167305232068,
      "loss": 6.5334,
      "mean_token_accuracy": 0.1553110048174858,
      "num_tokens": 4270684.0,
      "step": 1715
    },
    {
      "entropy": 6.848935222625732,
      "epoch": 2.0116959064327484,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004899788650593648,
      "loss": 6.4578,
      "mean_token_accuracy": 0.1595386952161789,
      "num_tokens": 4282567.0,
      "step": 1720
    },
    {
      "entropy": 6.960906648635865,
      "epoch": 2.017543859649123,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004898400665718325,
      "loss": 6.4928,
      "mean_token_accuracy": 0.15572800636291503,
      "num_tokens": 4294803.0,
      "step": 1725
    },
    {
      "entropy": 6.90195779800415,
      "epoch": 2.023391812865497,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004897003356630052,
      "loss": 6.4943,
      "mean_token_accuracy": 0.15569688379764557,
      "num_tokens": 4305898.0,
      "step": 1730
    },
    {
      "entropy": 6.879411506652832,
      "epoch": 2.0292397660818713,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004895596729393252,
      "loss": 6.3867,
      "mean_token_accuracy": 0.1615717150270939,
      "num_tokens": 4317985.0,
      "step": 1735
    },
    {
      "entropy": 6.773409175872803,
      "epoch": 2.0350877192982457,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004894180790112787,
      "loss": 6.4646,
      "mean_token_accuracy": 0.15991110801696778,
      "num_tokens": 4330482.0,
      "step": 1740
    },
    {
      "entropy": 6.953168106079102,
      "epoch": 2.0409356725146197,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004892755544933934,
      "loss": 6.5359,
      "mean_token_accuracy": 0.14696549624204636,
      "num_tokens": 4342782.0,
      "step": 1745
    },
    {
      "entropy": 6.875130081176758,
      "epoch": 2.046783625730994,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004891321000042361,
      "loss": 6.5404,
      "mean_token_accuracy": 0.14955661445856094,
      "num_tokens": 4356204.0,
      "step": 1750
    },
    {
      "entropy": 6.870871734619141,
      "epoch": 2.0526315789473686,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004889877161664096,
      "loss": 6.4454,
      "mean_token_accuracy": 0.1566561982035637,
      "num_tokens": 4370003.0,
      "step": 1755
    },
    {
      "entropy": 6.880638551712036,
      "epoch": 2.0584795321637426,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048884240360655,
      "loss": 6.4973,
      "mean_token_accuracy": 0.15367283672094345,
      "num_tokens": 4381972.0,
      "step": 1760
    },
    {
      "entropy": 6.828345870971679,
      "epoch": 2.064327485380117,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004886961629553242,
      "loss": 6.414,
      "mean_token_accuracy": 0.16296480149030684,
      "num_tokens": 4394033.0,
      "step": 1765
    },
    {
      "entropy": 6.877967786788941,
      "epoch": 2.0701754385964914,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004885489948474272,
      "loss": 6.5394,
      "mean_token_accuracy": 0.15387267172336577,
      "num_tokens": 4406799.0,
      "step": 1770
    },
    {
      "entropy": 6.8831394672393795,
      "epoch": 2.0760233918128654,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004884008999215792,
      "loss": 6.5346,
      "mean_token_accuracy": 0.1527932584285736,
      "num_tokens": 4420182.0,
      "step": 1775
    },
    {
      "entropy": 6.834596395492554,
      "epoch": 2.08187134502924,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004882518788205225,
      "loss": 6.5003,
      "mean_token_accuracy": 0.15515169501304626,
      "num_tokens": 4432014.0,
      "step": 1780
    },
    {
      "entropy": 6.9492169380187985,
      "epoch": 2.087719298245614,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004881019321910198,
      "loss": 6.4922,
      "mean_token_accuracy": 0.15978310108184815,
      "num_tokens": 4444755.0,
      "step": 1785
    },
    {
      "entropy": 6.811837768554687,
      "epoch": 2.0935672514619883,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048795106068384985,
      "loss": 6.5632,
      "mean_token_accuracy": 0.15767932385206224,
      "num_tokens": 4456390.0,
      "step": 1790
    },
    {
      "entropy": 6.876183891296387,
      "epoch": 2.0994152046783627,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000487799264953806,
      "loss": 6.4799,
      "mean_token_accuracy": 0.1628776267170906,
      "num_tokens": 4468567.0,
      "step": 1795
    },
    {
      "entropy": 6.800377607345581,
      "epoch": 2.1052631578947367,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004876465456596926,
      "loss": 6.4521,
      "mean_token_accuracy": 0.1607213571667671,
      "num_tokens": 4479553.0,
      "step": 1800
    },
    {
      "entropy": 6.800175619125366,
      "epoch": 2.111111111111111,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004874929034643223,
      "loss": 6.4485,
      "mean_token_accuracy": 0.16596934646368028,
      "num_tokens": 4491505.0,
      "step": 1805
    },
    {
      "entropy": 6.807722282409668,
      "epoch": 2.116959064327485,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004873383390345132,
      "loss": 6.521,
      "mean_token_accuracy": 0.15679756850004195,
      "num_tokens": 4504896.0,
      "step": 1810
    },
    {
      "entropy": 6.794916152954102,
      "epoch": 2.1228070175438596,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048718285304108613,
      "loss": 6.513,
      "mean_token_accuracy": 0.15284237414598464,
      "num_tokens": 4516319.0,
      "step": 1815
    },
    {
      "entropy": 6.853626441955567,
      "epoch": 2.128654970760234,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048702644615886137,
      "loss": 6.5226,
      "mean_token_accuracy": 0.1524695210158825,
      "num_tokens": 4529132.0,
      "step": 1820
    },
    {
      "entropy": 6.864784622192383,
      "epoch": 2.134502923976608,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048686911906665597,
      "loss": 6.4923,
      "mean_token_accuracy": 0.15571837723255158,
      "num_tokens": 4541080.0,
      "step": 1825
    },
    {
      "entropy": 6.845421934127808,
      "epoch": 2.1403508771929824,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00048671087244728093,
      "loss": 6.4962,
      "mean_token_accuracy": 0.15684320777654648,
      "num_tokens": 4553442.0,
      "step": 1830
    },
    {
      "entropy": 6.731913423538208,
      "epoch": 2.146198830409357,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00048655170698753774,
      "loss": 6.4244,
      "mean_token_accuracy": 0.16705678403377533,
      "num_tokens": 4565617.0,
      "step": 1835
    },
    {
      "entropy": 6.900339508056641,
      "epoch": 2.152046783625731,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000486391623378216,
      "loss": 6.5217,
      "mean_token_accuracy": 0.15130398869514466,
      "num_tokens": 4577418.0,
      "step": 1840
    },
    {
      "entropy": 6.829915714263916,
      "epoch": 2.1578947368421053,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004862306223140901,
      "loss": 6.4447,
      "mean_token_accuracy": 0.15913260877132415,
      "num_tokens": 4589763.0,
      "step": 1845
    },
    {
      "entropy": 6.829157447814941,
      "epoch": 2.1637426900584797,
      "grad_norm": 1.0,
      "learning_rate": 0.00048606870449391604,
      "loss": 6.5366,
      "mean_token_accuracy": 0.15990214347839354,
      "num_tokens": 4602522.0,
      "step": 1850
    },
    {
      "entropy": 6.833024978637695,
      "epoch": 2.1695906432748537,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00048590587062042907,
      "loss": 6.4977,
      "mean_token_accuracy": 0.15934453904628754,
      "num_tokens": 4614680.0,
      "step": 1855
    },
    {
      "entropy": 6.8116669178009035,
      "epoch": 2.175438596491228,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048574212140033965,
      "loss": 6.5341,
      "mean_token_accuracy": 0.15397311598062516,
      "num_tokens": 4628166.0,
      "step": 1860
    },
    {
      "entropy": 6.821261405944824,
      "epoch": 2.181286549707602,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004855774575443311,
      "loss": 6.5328,
      "mean_token_accuracy": 0.15765905529260635,
      "num_tokens": 4640200.0,
      "step": 1865
    },
    {
      "entropy": 6.8316350936889645,
      "epoch": 2.1871345029239766,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048541187976705645,
      "loss": 6.4767,
      "mean_token_accuracy": 0.15292651504278182,
      "num_tokens": 4652968.0,
      "step": 1870
    },
    {
      "entropy": 6.770348358154297,
      "epoch": 2.192982456140351,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048524538878713517,
      "loss": 6.485,
      "mean_token_accuracy": 0.15311889797449113,
      "num_tokens": 4665741.0,
      "step": 1875
    },
    {
      "entropy": 6.82572374343872,
      "epoch": 2.198830409356725,
      "grad_norm": 0.765625,
      "learning_rate": 0.00048507798532715005,
      "loss": 6.5084,
      "mean_token_accuracy": 0.155975741147995,
      "num_tokens": 4680074.0,
      "step": 1880
    },
    {
      "entropy": 6.801500225067139,
      "epoch": 2.2046783625730995,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048490967011364394,
      "loss": 6.5232,
      "mean_token_accuracy": 0.1580188348889351,
      "num_tokens": 4692340.0,
      "step": 1885
    },
    {
      "entropy": 6.821602535247803,
      "epoch": 2.2105263157894735,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004847404438771172,
      "loss": 6.518,
      "mean_token_accuracy": 0.15585452020168306,
      "num_tokens": 4704470.0,
      "step": 1890
    },
    {
      "entropy": 6.843487024307251,
      "epoch": 2.216374269005848,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004845703073520236,
      "loss": 6.5465,
      "mean_token_accuracy": 0.15025353580713272,
      "num_tokens": 4716253.0,
      "step": 1895
    },
    {
      "entropy": 6.897539663314819,
      "epoch": 2.2222222222222223,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048439926127676796,
      "loss": 6.6592,
      "mean_token_accuracy": 0.14622338116168976,
      "num_tokens": 4730083.0,
      "step": 1900
    },
    {
      "entropy": 6.8224715232849125,
      "epoch": 2.2280701754385963,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00048422730639370256,
      "loss": 6.4221,
      "mean_token_accuracy": 0.16505661010742187,
      "num_tokens": 4742770.0,
      "step": 1905
    },
    {
      "entropy": 6.769926834106445,
      "epoch": 2.2339181286549707,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004840544434491238,
      "loss": 6.5511,
      "mean_token_accuracy": 0.15389376133680344,
      "num_tokens": 4754299.0,
      "step": 1910
    },
    {
      "entropy": 6.8316566944122314,
      "epoch": 2.239766081871345,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048388067319326937,
      "loss": 6.5164,
      "mean_token_accuracy": 0.15317742228507997,
      "num_tokens": 4766964.0,
      "step": 1915
    },
    {
      "entropy": 6.77725830078125,
      "epoch": 2.245614035087719,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048370599638031464,
      "loss": 6.5459,
      "mean_token_accuracy": 0.153824120759964,
      "num_tokens": 4779921.0,
      "step": 1920
    },
    {
      "entropy": 6.8391601085662845,
      "epoch": 2.2514619883040936,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048353041376836937,
      "loss": 6.529,
      "mean_token_accuracy": 0.15380112081766129,
      "num_tokens": 4792233.0,
      "step": 1925
    },
    {
      "entropy": 6.785034418106079,
      "epoch": 2.257309941520468,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004833539261194749,
      "loss": 6.5129,
      "mean_token_accuracy": 0.16028628200292588,
      "num_tokens": 4803348.0,
      "step": 1930
    },
    {
      "entropy": 6.850829458236694,
      "epoch": 2.263157894736842,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00048317653419960007,
      "loss": 6.5866,
      "mean_token_accuracy": 0.1491300158202648,
      "num_tokens": 4816574.0,
      "step": 1935
    },
    {
      "entropy": 6.703137397766113,
      "epoch": 2.2690058479532165,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048299823877863857,
      "loss": 6.5065,
      "mean_token_accuracy": 0.15909324884414672,
      "num_tokens": 4828008.0,
      "step": 1940
    },
    {
      "entropy": 6.869985198974609,
      "epoch": 2.2748538011695905,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004828190406304054,
      "loss": 6.5792,
      "mean_token_accuracy": 0.151179438829422,
      "num_tokens": 4840621.0,
      "step": 1945
    },
    {
      "entropy": 6.806095361709595,
      "epoch": 2.280701754385965,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004826389405326334,
      "loss": 6.4768,
      "mean_token_accuracy": 0.15327924191951753,
      "num_tokens": 4853503.0,
      "step": 1950
    },
    {
      "entropy": 6.816493034362793,
      "epoch": 2.2865497076023393,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004824579392669698,
      "loss": 6.4961,
      "mean_token_accuracy": 0.15774255096912385,
      "num_tokens": 4865925.0,
      "step": 1955
    },
    {
      "entropy": 6.774963331222534,
      "epoch": 2.2923976608187133,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048227603761897316,
      "loss": 6.5814,
      "mean_token_accuracy": 0.15094525516033172,
      "num_tokens": 4878704.0,
      "step": 1960
    },
    {
      "entropy": 6.879746818542481,
      "epoch": 2.2982456140350878,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004820932363781096,
      "loss": 6.5312,
      "mean_token_accuracy": 0.15396272242069245,
      "num_tokens": 4891750.0,
      "step": 1965
    },
    {
      "entropy": 6.784665966033936,
      "epoch": 2.3040935672514617,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004819095363377496,
      "loss": 6.5187,
      "mean_token_accuracy": 0.15821436196565627,
      "num_tokens": 4904539.0,
      "step": 1970
    },
    {
      "entropy": 6.7290582180023195,
      "epoch": 2.309941520467836,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048172493829516465,
      "loss": 6.3934,
      "mean_token_accuracy": 0.16299119889736174,
      "num_tokens": 4916210.0,
      "step": 1975
    },
    {
      "entropy": 6.78262882232666,
      "epoch": 2.3157894736842106,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004815394430515234,
      "loss": 6.462,
      "mean_token_accuracy": 0.15879430174827575,
      "num_tokens": 4929011.0,
      "step": 1980
    },
    {
      "entropy": 6.765537166595459,
      "epoch": 2.3216374269005846,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004813530514118884,
      "loss": 6.4498,
      "mean_token_accuracy": 0.15751578211784362,
      "num_tokens": 4942484.0,
      "step": 1985
    },
    {
      "entropy": 6.7109559059143065,
      "epoch": 2.327485380116959,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004811657641852129,
      "loss": 6.427,
      "mean_token_accuracy": 0.15696060806512832,
      "num_tokens": 4955395.0,
      "step": 1990
    },
    {
      "entropy": 6.781054306030273,
      "epoch": 2.3333333333333335,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048097758218433687,
      "loss": 6.4703,
      "mean_token_accuracy": 0.15664945989847184,
      "num_tokens": 4966853.0,
      "step": 1995
    },
    {
      "entropy": 6.6470122814178465,
      "epoch": 2.3391812865497075,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048078850622598355,
      "loss": 6.3438,
      "mean_token_accuracy": 0.16954133957624434,
      "num_tokens": 4978514.0,
      "step": 2000
    },
    {
      "epoch": 2.3391812865497075,
      "eval_entropy": 6.59683396190317,
      "eval_loss": 6.824612617492676,
      "eval_mean_token_accuracy": 0.14168384835222916,
      "eval_num_tokens": 4978514.0,
      "eval_runtime": 1.7431,
      "eval_samples_per_second": 1552.985,
      "eval_steps_per_second": 194.482,
      "step": 2000
    },
    {
      "entropy": 6.78842453956604,
      "epoch": 2.345029239766082,
      "grad_norm": 0.83984375,
      "learning_rate": 0.00048059853713075637,
      "loss": 6.501,
      "mean_token_accuracy": 0.15215353667736053,
      "num_tokens": 4991331.0,
      "step": 2005
    },
    {
      "entropy": 6.730353736877442,
      "epoch": 2.3508771929824563,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004804076757231347,
      "loss": 6.4718,
      "mean_token_accuracy": 0.15783711373806,
      "num_tokens": 5004652.0,
      "step": 2010
    },
    {
      "entropy": 6.759194612503052,
      "epoch": 2.3567251461988303,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048021592283147073,
      "loss": 6.4715,
      "mean_token_accuracy": 0.16659724712371826,
      "num_tokens": 5017204.0,
      "step": 2015
    },
    {
      "entropy": 6.698295402526855,
      "epoch": 2.3625730994152048,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004800232792879858,
      "loss": 6.4975,
      "mean_token_accuracy": 0.16038344353437423,
      "num_tokens": 5029667.0,
      "step": 2020
    },
    {
      "entropy": 6.651339197158814,
      "epoch": 2.3684210526315788,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047982974592876665,
      "loss": 6.3622,
      "mean_token_accuracy": 0.16610048711299896,
      "num_tokens": 5041490.0,
      "step": 2025
    },
    {
      "entropy": 6.777406501770019,
      "epoch": 2.374269005847953,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000479635323593762,
      "loss": 6.4944,
      "mean_token_accuracy": 0.15169765874743463,
      "num_tokens": 5054353.0,
      "step": 2030
    },
    {
      "entropy": 6.768514680862427,
      "epoch": 2.3801169590643276,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047944001312677867,
      "loss": 6.4569,
      "mean_token_accuracy": 0.15618255287408828,
      "num_tokens": 5067322.0,
      "step": 2035
    },
    {
      "entropy": 6.779137372970581,
      "epoch": 2.3859649122807016,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004792438153754782,
      "loss": 6.5077,
      "mean_token_accuracy": 0.16168376058340073,
      "num_tokens": 5079957.0,
      "step": 2040
    },
    {
      "entropy": 6.691183376312256,
      "epoch": 2.391812865497076,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004790467311913727,
      "loss": 6.4856,
      "mean_token_accuracy": 0.15492921620607375,
      "num_tokens": 5091975.0,
      "step": 2045
    },
    {
      "entropy": 6.777608013153076,
      "epoch": 2.39766081871345,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004788487614298219,
      "loss": 6.4976,
      "mean_token_accuracy": 0.1535652071237564,
      "num_tokens": 5103847.0,
      "step": 2050
    },
    {
      "entropy": 6.772614765167236,
      "epoch": 2.4035087719298245,
      "grad_norm": 1.0,
      "learning_rate": 0.0004786499069500286,
      "loss": 6.442,
      "mean_token_accuracy": 0.16196208149194719,
      "num_tokens": 5115153.0,
      "step": 2055
    },
    {
      "entropy": 6.6315269947052,
      "epoch": 2.409356725146199,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047845016861503556,
      "loss": 6.3778,
      "mean_token_accuracy": 0.16543841511011123,
      "num_tokens": 5128008.0,
      "step": 2060
    },
    {
      "entropy": 6.6571738719940186,
      "epoch": 2.415204678362573,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047824954729172136,
      "loss": 6.3427,
      "mean_token_accuracy": 0.1712554380297661,
      "num_tokens": 5140023.0,
      "step": 2065
    },
    {
      "entropy": 6.68211727142334,
      "epoch": 2.4210526315789473,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004780480438507972,
      "loss": 6.3945,
      "mean_token_accuracy": 0.1659012407064438,
      "num_tokens": 5152660.0,
      "step": 2070
    },
    {
      "entropy": 6.691555452346802,
      "epoch": 2.426900584795322,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047784565916680216,
      "loss": 6.4164,
      "mean_token_accuracy": 0.16120534390211105,
      "num_tokens": 5164481.0,
      "step": 2075
    },
    {
      "entropy": 6.8111319065094,
      "epoch": 2.4327485380116958,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004776423941181005,
      "loss": 6.4971,
      "mean_token_accuracy": 0.1621654786169529,
      "num_tokens": 5177178.0,
      "step": 2080
    },
    {
      "entropy": 6.711190795898437,
      "epoch": 2.43859649122807,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047743824958687694,
      "loss": 6.3991,
      "mean_token_accuracy": 0.1595264583826065,
      "num_tokens": 5189385.0,
      "step": 2085
    },
    {
      "entropy": 6.763757753372192,
      "epoch": 2.4444444444444446,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047723322645913354,
      "loss": 6.4655,
      "mean_token_accuracy": 0.16068250834941863,
      "num_tokens": 5201839.0,
      "step": 2090
    },
    {
      "entropy": 6.680694103240967,
      "epoch": 2.4502923976608186,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004770273256246853,
      "loss": 6.4967,
      "mean_token_accuracy": 0.1545516699552536,
      "num_tokens": 5214565.0,
      "step": 2095
    },
    {
      "entropy": 6.766183567047119,
      "epoch": 2.456140350877193,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004768205479771566,
      "loss": 6.4671,
      "mean_token_accuracy": 0.16065627932548524,
      "num_tokens": 5227029.0,
      "step": 2100
    },
    {
      "entropy": 6.767592906951904,
      "epoch": 2.461988304093567,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00047661289441397724,
      "loss": 6.4794,
      "mean_token_accuracy": 0.16503392904996872,
      "num_tokens": 5239208.0,
      "step": 2105
    },
    {
      "entropy": 6.667439317703247,
      "epoch": 2.4678362573099415,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00047640436583637865,
      "loss": 6.4425,
      "mean_token_accuracy": 0.1622099220752716,
      "num_tokens": 5252272.0,
      "step": 2110
    },
    {
      "entropy": 6.799730634689331,
      "epoch": 2.473684210526316,
      "grad_norm": 0.859375,
      "learning_rate": 0.00047619496314938977,
      "loss": 6.5702,
      "mean_token_accuracy": 0.150389401614666,
      "num_tokens": 5264387.0,
      "step": 2115
    },
    {
      "entropy": 6.685615730285645,
      "epoch": 2.47953216374269,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004759846872618332,
      "loss": 6.5309,
      "mean_token_accuracy": 0.15629530549049378,
      "num_tokens": 5276662.0,
      "step": 2120
    },
    {
      "entropy": 6.752291631698609,
      "epoch": 2.4853801169590644,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00047577353908632156,
      "loss": 6.4597,
      "mean_token_accuracy": 0.16320658028125762,
      "num_tokens": 5290734.0,
      "step": 2125
    },
    {
      "entropy": 6.59678897857666,
      "epoch": 2.4912280701754383,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00047556151953925287,
      "loss": 6.3483,
      "mean_token_accuracy": 0.16701145321130753,
      "num_tokens": 5302510.0,
      "step": 2130
    },
    {
      "entropy": 6.849278783798217,
      "epoch": 2.497076023391813,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004753486295408073,
      "loss": 6.4293,
      "mean_token_accuracy": 0.15996883660554886,
      "num_tokens": 5313243.0,
      "step": 2135
    },
    {
      "entropy": 6.5875578880310055,
      "epoch": 2.502923976608187,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00047513487001494257,
      "loss": 6.5105,
      "mean_token_accuracy": 0.15729328095912934,
      "num_tokens": 5325485.0,
      "step": 2140
    },
    {
      "entropy": 6.842334747314453,
      "epoch": 2.5087719298245617,
      "grad_norm": 0.8203125,
      "learning_rate": 0.00047492024188939056,
      "loss": 6.5625,
      "mean_token_accuracy": 0.15200666040182115,
      "num_tokens": 5339335.0,
      "step": 2145
    },
    {
      "entropy": 6.765503215789795,
      "epoch": 2.5146198830409356,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004747047460956525,
      "loss": 6.5084,
      "mean_token_accuracy": 0.15184070765972138,
      "num_tokens": 5351195.0,
      "step": 2150
    },
    {
      "entropy": 6.6956579208374025,
      "epoch": 2.52046783625731,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004744883835689958,
      "loss": 6.4564,
      "mean_token_accuracy": 0.15987571626901625,
      "num_tokens": 5362884.0,
      "step": 2155
    },
    {
      "entropy": 6.669484853744507,
      "epoch": 2.526315789473684,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004742711552484492,
      "loss": 6.4902,
      "mean_token_accuracy": 0.15629828870296478,
      "num_tokens": 5375770.0,
      "step": 2160
    },
    {
      "entropy": 6.7700704574584964,
      "epoch": 2.5321637426900585,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00047405306207679926,
      "loss": 6.4671,
      "mean_token_accuracy": 0.15315858870744706,
      "num_tokens": 5388673.0,
      "step": 2165
    },
    {
      "entropy": 6.570160865783691,
      "epoch": 2.538011695906433,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004738341050005859,
      "loss": 6.3133,
      "mean_token_accuracy": 0.1679084688425064,
      "num_tokens": 5400454.0,
      "step": 2170
    },
    {
      "entropy": 6.723893356323242,
      "epoch": 2.543859649122807,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004736142849700986,
      "loss": 6.4604,
      "mean_token_accuracy": 0.16196997910737992,
      "num_tokens": 5413740.0,
      "step": 2175
    },
    {
      "entropy": 6.702420473098755,
      "epoch": 2.5497076023391814,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004733936029393721,
      "loss": 6.474,
      "mean_token_accuracy": 0.1617956429719925,
      "num_tokens": 5426089.0,
      "step": 2180
    },
    {
      "entropy": 6.7728574752807615,
      "epoch": 2.5555555555555554,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004731720598661821,
      "loss": 6.5213,
      "mean_token_accuracy": 0.1508661463856697,
      "num_tokens": 5438835.0,
      "step": 2185
    },
    {
      "entropy": 6.628914785385132,
      "epoch": 2.56140350877193,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004729496567120415,
      "loss": 6.458,
      "mean_token_accuracy": 0.16138928830623628,
      "num_tokens": 5452054.0,
      "step": 2190
    },
    {
      "entropy": 6.721137571334839,
      "epoch": 2.5672514619883042,
      "grad_norm": 0.8203125,
      "learning_rate": 0.000472726394442196,
      "loss": 6.4259,
      "mean_token_accuracy": 0.15541900396347047,
      "num_tokens": 5466455.0,
      "step": 2195
    },
    {
      "entropy": 6.7194966793060305,
      "epoch": 2.573099415204678,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047250227402561976,
      "loss": 6.4399,
      "mean_token_accuracy": 0.1621753215789795,
      "num_tokens": 5479990.0,
      "step": 2200
    },
    {
      "entropy": 6.7334068775177,
      "epoch": 2.5789473684210527,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004722772964350115,
      "loss": 6.4819,
      "mean_token_accuracy": 0.16410714983940125,
      "num_tokens": 5493035.0,
      "step": 2205
    },
    {
      "entropy": 6.627843904495239,
      "epoch": 2.5847953216374266,
      "grad_norm": 0.9375,
      "learning_rate": 0.00047205146264679015,
      "loss": 6.4972,
      "mean_token_accuracy": 0.15570203065872193,
      "num_tokens": 5506093.0,
      "step": 2210
    },
    {
      "entropy": 6.719814825057983,
      "epoch": 2.590643274853801,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00047182477364109045,
      "loss": 6.4734,
      "mean_token_accuracy": 0.1660175785422325,
      "num_tokens": 5518552.0,
      "step": 2215
    },
    {
      "entropy": 6.664255619049072,
      "epoch": 2.5964912280701755,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047159723040175896,
      "loss": 6.4947,
      "mean_token_accuracy": 0.16128159165382386,
      "num_tokens": 5530123.0,
      "step": 2220
    },
    {
      "entropy": 6.8260692119598385,
      "epoch": 2.60233918128655,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00047136883391634966,
      "loss": 6.5105,
      "mean_token_accuracy": 0.1525448054075241,
      "num_tokens": 5542769.0,
      "step": 2225
    },
    {
      "entropy": 6.516240072250366,
      "epoch": 2.608187134502924,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004711395851761198,
      "loss": 6.3133,
      "mean_token_accuracy": 0.17718010097742082,
      "num_tokens": 5555975.0,
      "step": 2230
    },
    {
      "entropy": 6.68882999420166,
      "epoch": 2.6140350877192984,
      "grad_norm": 1.125,
      "learning_rate": 0.0004709094851760251,
      "loss": 6.422,
      "mean_token_accuracy": 0.15997820794582368,
      "num_tokens": 5569059.0,
      "step": 2235
    },
    {
      "entropy": 6.699761438369751,
      "epoch": 2.6198830409356724,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004706785349147163,
      "loss": 6.488,
      "mean_token_accuracy": 0.15371347069740296,
      "num_tokens": 5581358.0,
      "step": 2240
    },
    {
      "entropy": 6.692812108993531,
      "epoch": 2.625730994152047,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000470446735394534,
      "loss": 6.4376,
      "mean_token_accuracy": 0.15799081325531006,
      "num_tokens": 5594256.0,
      "step": 2245
    },
    {
      "entropy": 6.709306335449218,
      "epoch": 2.6315789473684212,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004702140876215047,
      "loss": 6.4429,
      "mean_token_accuracy": 0.15952685177326204,
      "num_tokens": 5607236.0,
      "step": 2250
    },
    {
      "entropy": 6.751166915893554,
      "epoch": 2.6374269005847952,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004699805926053365,
      "loss": 6.5489,
      "mean_token_accuracy": 0.14810458123683928,
      "num_tokens": 5619550.0,
      "step": 2255
    },
    {
      "entropy": 6.6508769512176515,
      "epoch": 2.6432748538011697,
      "grad_norm": 0.8359375,
      "learning_rate": 0.00046974625135941445,
      "loss": 6.4041,
      "mean_token_accuracy": 0.16714755594730377,
      "num_tokens": 5632721.0,
      "step": 2260
    },
    {
      "entropy": 6.604614305496216,
      "epoch": 2.6491228070175437,
      "grad_norm": 1.0,
      "learning_rate": 0.00046951106490079636,
      "loss": 6.2551,
      "mean_token_accuracy": 0.16622601598501205,
      "num_tokens": 5643672.0,
      "step": 2265
    },
    {
      "entropy": 6.682486724853516,
      "epoch": 2.654970760233918,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00046927503425020847,
      "loss": 6.4072,
      "mean_token_accuracy": 0.15769419074058533,
      "num_tokens": 5656072.0,
      "step": 2270
    },
    {
      "entropy": 6.7076764583587645,
      "epoch": 2.6608187134502925,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00046903816043204045,
      "loss": 6.4216,
      "mean_token_accuracy": 0.16312220990657805,
      "num_tokens": 5668365.0,
      "step": 2275
    },
    {
      "entropy": 6.696519136428833,
      "epoch": 2.6666666666666665,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004688004444743419,
      "loss": 6.3903,
      "mean_token_accuracy": 0.16126345247030258,
      "num_tokens": 5680013.0,
      "step": 2280
    },
    {
      "entropy": 6.579961919784546,
      "epoch": 2.672514619883041,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000468561887408817,
      "loss": 6.4123,
      "mean_token_accuracy": 0.16684895008802414,
      "num_tokens": 5691884.0,
      "step": 2285
    },
    {
      "entropy": 6.68872218132019,
      "epoch": 2.678362573099415,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004683224902708206,
      "loss": 6.4742,
      "mean_token_accuracy": 0.15662514567375183,
      "num_tokens": 5704340.0,
      "step": 2290
    },
    {
      "entropy": 6.756664419174195,
      "epoch": 2.6842105263157894,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004680822540993534,
      "loss": 6.5251,
      "mean_token_accuracy": 0.15745238810777665,
      "num_tokens": 5717792.0,
      "step": 2295
    },
    {
      "entropy": 6.6499731063842775,
      "epoch": 2.690058479532164,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046784117993705753,
      "loss": 6.3996,
      "mean_token_accuracy": 0.15965020656585693,
      "num_tokens": 5730237.0,
      "step": 2300
    },
    {
      "entropy": 6.635748100280762,
      "epoch": 2.6959064327485383,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00046759926883021234,
      "loss": 6.401,
      "mean_token_accuracy": 0.16328736394643784,
      "num_tokens": 5743953.0,
      "step": 2305
    },
    {
      "entropy": 6.630077362060547,
      "epoch": 2.7017543859649122,
      "grad_norm": 0.890625,
      "learning_rate": 0.00046735652182872925,
      "loss": 6.3658,
      "mean_token_accuracy": 0.16114580780267715,
      "num_tokens": 5756148.0,
      "step": 2310
    },
    {
      "entropy": 6.674561071395874,
      "epoch": 2.7076023391812867,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004671129399861476,
      "loss": 6.389,
      "mean_token_accuracy": 0.1633088245987892,
      "num_tokens": 5769163.0,
      "step": 2315
    },
    {
      "entropy": 6.665018463134766,
      "epoch": 2.7134502923976607,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046686852435963016,
      "loss": 6.4305,
      "mean_token_accuracy": 0.15733572244644164,
      "num_tokens": 5781832.0,
      "step": 2320
    },
    {
      "entropy": 6.676378059387207,
      "epoch": 2.719298245614035,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004666232760099583,
      "loss": 6.4262,
      "mean_token_accuracy": 0.15573703944683076,
      "num_tokens": 5794803.0,
      "step": 2325
    },
    {
      "entropy": 6.6417618751525875,
      "epoch": 2.7251461988304095,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046637719600152737,
      "loss": 6.4575,
      "mean_token_accuracy": 0.15561199858784674,
      "num_tokens": 5806768.0,
      "step": 2330
    },
    {
      "entropy": 6.7871111869812015,
      "epoch": 2.7309941520467835,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046613028540234226,
      "loss": 6.4607,
      "mean_token_accuracy": 0.1528087079524994,
      "num_tokens": 5819113.0,
      "step": 2335
    },
    {
      "entropy": 6.690218925476074,
      "epoch": 2.736842105263158,
      "grad_norm": 0.953125,
      "learning_rate": 0.00046588254528401264,
      "loss": 6.437,
      "mean_token_accuracy": 0.1578143686056137,
      "num_tokens": 5830067.0,
      "step": 2340
    },
    {
      "entropy": 6.635287761688232,
      "epoch": 2.742690058479532,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004656339767217485,
      "loss": 6.4848,
      "mean_token_accuracy": 0.16050305515527724,
      "num_tokens": 5842882.0,
      "step": 2345
    },
    {
      "entropy": 6.609443235397339,
      "epoch": 2.7485380116959064,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046538458079435516,
      "loss": 6.4252,
      "mean_token_accuracy": 0.15916400253772736,
      "num_tokens": 5854041.0,
      "step": 2350
    },
    {
      "entropy": 6.7075742244720455,
      "epoch": 2.754385964912281,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004651343585842287,
      "loss": 6.4508,
      "mean_token_accuracy": 0.16283646672964097,
      "num_tokens": 5866027.0,
      "step": 2355
    },
    {
      "entropy": 6.671390342712402,
      "epoch": 2.760233918128655,
      "grad_norm": 0.90625,
      "learning_rate": 0.00046488331117735176,
      "loss": 6.4845,
      "mean_token_accuracy": 0.16220867186784743,
      "num_tokens": 5877064.0,
      "step": 2360
    },
    {
      "entropy": 6.720740461349488,
      "epoch": 2.7660818713450293,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004646314396632878,
      "loss": 6.4449,
      "mean_token_accuracy": 0.1543478935956955,
      "num_tokens": 5889190.0,
      "step": 2365
    },
    {
      "entropy": 6.6629119396209715,
      "epoch": 2.7719298245614032,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004643787451351774,
      "loss": 6.4177,
      "mean_token_accuracy": 0.16133566051721573,
      "num_tokens": 5901615.0,
      "step": 2370
    },
    {
      "entropy": 6.620958757400513,
      "epoch": 2.7777777777777777,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046412522868973297,
      "loss": 6.4197,
      "mean_token_accuracy": 0.16412558257579804,
      "num_tokens": 5913797.0,
      "step": 2375
    },
    {
      "entropy": 6.730884408950805,
      "epoch": 2.783625730994152,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046387089142723386,
      "loss": 6.5005,
      "mean_token_accuracy": 0.15636750161647797,
      "num_tokens": 5927607.0,
      "step": 2380
    },
    {
      "entropy": 6.529775142669678,
      "epoch": 2.7894736842105265,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00046361573445152224,
      "loss": 6.3267,
      "mean_token_accuracy": 0.1680535688996315,
      "num_tokens": 5940254.0,
      "step": 2385
    },
    {
      "entropy": 6.525655937194824,
      "epoch": 2.7953216374269005,
      "grad_norm": 0.90625,
      "learning_rate": 0.00046335975886999754,
      "loss": 6.1682,
      "mean_token_accuracy": 0.17823653072118759,
      "num_tokens": 5953341.0,
      "step": 2390
    },
    {
      "entropy": 6.526809024810791,
      "epoch": 2.801169590643275,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046310296579361225,
      "loss": 6.2108,
      "mean_token_accuracy": 0.1784290373325348,
      "num_tokens": 5965148.0,
      "step": 2395
    },
    {
      "entropy": 6.66150860786438,
      "epoch": 2.807017543859649,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004628453563368666,
      "loss": 6.497,
      "mean_token_accuracy": 0.15572031289339067,
      "num_tokens": 5979143.0,
      "step": 2400
    },
    {
      "entropy": 6.686175584793091,
      "epoch": 2.8128654970760234,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004625869316178043,
      "loss": 6.4467,
      "mean_token_accuracy": 0.15691925287246705,
      "num_tokens": 5992735.0,
      "step": 2405
    },
    {
      "entropy": 6.638707208633423,
      "epoch": 2.818713450292398,
      "grad_norm": 0.90625,
      "learning_rate": 0.000462327692758007,
      "loss": 6.5145,
      "mean_token_accuracy": 0.15147309452295304,
      "num_tokens": 6005033.0,
      "step": 2410
    },
    {
      "entropy": 6.709354877471924,
      "epoch": 2.824561403508772,
      "grad_norm": 0.921875,
      "learning_rate": 0.00046206764088259003,
      "loss": 6.3824,
      "mean_token_accuracy": 0.1630718469619751,
      "num_tokens": 6016746.0,
      "step": 2415
    },
    {
      "entropy": 6.573739004135132,
      "epoch": 2.8304093567251463,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00046180677712019707,
      "loss": 6.4565,
      "mean_token_accuracy": 0.1616368681192398,
      "num_tokens": 6029774.0,
      "step": 2420
    },
    {
      "entropy": 6.69282865524292,
      "epoch": 2.8362573099415203,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004615451026029957,
      "loss": 6.4896,
      "mean_token_accuracy": 0.15256085395812988,
      "num_tokens": 6043773.0,
      "step": 2425
    },
    {
      "entropy": 6.708876514434815,
      "epoch": 2.8421052631578947,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000461282618466672,
      "loss": 6.4152,
      "mean_token_accuracy": 0.15378952920436859,
      "num_tokens": 6055840.0,
      "step": 2430
    },
    {
      "entropy": 6.5623846530914305,
      "epoch": 2.847953216374269,
      "grad_norm": 0.921875,
      "learning_rate": 0.00046101932585042597,
      "loss": 6.3331,
      "mean_token_accuracy": 0.16594051718711852,
      "num_tokens": 6069170.0,
      "step": 2435
    },
    {
      "entropy": 6.63337721824646,
      "epoch": 2.853801169590643,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046075522589696636,
      "loss": 6.3916,
      "mean_token_accuracy": 0.16303064227104186,
      "num_tokens": 6080888.0,
      "step": 2440
    },
    {
      "entropy": 6.606135702133178,
      "epoch": 2.8596491228070176,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000460490319752506,
      "loss": 6.3359,
      "mean_token_accuracy": 0.16366248875856398,
      "num_tokens": 6093035.0,
      "step": 2445
    },
    {
      "entropy": 6.61658935546875,
      "epoch": 2.8654970760233915,
      "grad_norm": 0.80859375,
      "learning_rate": 0.00046022460856675663,
      "loss": 6.4116,
      "mean_token_accuracy": 0.15998516529798507,
      "num_tokens": 6105963.0,
      "step": 2450
    },
    {
      "entropy": 6.775122499465942,
      "epoch": 2.871345029239766,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004599580934929237,
      "loss": 6.4301,
      "mean_token_accuracy": 0.15838743150234222,
      "num_tokens": 6118541.0,
      "step": 2455
    },
    {
      "entropy": 6.554420280456543,
      "epoch": 2.8771929824561404,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004596907756877018,
      "loss": 6.3637,
      "mean_token_accuracy": 0.16104139834642411,
      "num_tokens": 6131317.0,
      "step": 2460
    },
    {
      "entropy": 6.617832326889038,
      "epoch": 2.883040935672515,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00045942265631126935,
      "loss": 6.4151,
      "mean_token_accuracy": 0.1558060348033905,
      "num_tokens": 6143721.0,
      "step": 2465
    },
    {
      "entropy": 6.680070495605468,
      "epoch": 2.888888888888889,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004591537365272837,
      "loss": 6.407,
      "mean_token_accuracy": 0.1582871437072754,
      "num_tokens": 6156497.0,
      "step": 2470
    },
    {
      "entropy": 6.555939054489135,
      "epoch": 2.8947368421052633,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000458884017502876,
      "loss": 6.3992,
      "mean_token_accuracy": 0.16497224867343901,
      "num_tokens": 6168140.0,
      "step": 2475
    },
    {
      "entropy": 6.706430339813233,
      "epoch": 2.9005847953216373,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004586135004086461,
      "loss": 6.3916,
      "mean_token_accuracy": 0.15955118238925933,
      "num_tokens": 6181067.0,
      "step": 2480
    },
    {
      "entropy": 6.61524453163147,
      "epoch": 2.9064327485380117,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004583421864186577,
      "loss": 6.3744,
      "mean_token_accuracy": 0.15712310373783112,
      "num_tokens": 6194072.0,
      "step": 2485
    },
    {
      "entropy": 6.65601954460144,
      "epoch": 2.912280701754386,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004580700767104328,
      "loss": 6.4397,
      "mean_token_accuracy": 0.16010524332523346,
      "num_tokens": 6206711.0,
      "step": 2490
    },
    {
      "entropy": 6.5722051620483395,
      "epoch": 2.91812865497076,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00045779717246494724,
      "loss": 6.3593,
      "mean_token_accuracy": 0.16191931813955307,
      "num_tokens": 6219903.0,
      "step": 2495
    },
    {
      "entropy": 6.632825469970703,
      "epoch": 2.9239766081871346,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004575234748666249,
      "loss": 6.4107,
      "mean_token_accuracy": 0.15712305307388305,
      "num_tokens": 6232947.0,
      "step": 2500
    },
    {
      "epoch": 2.9239766081871346,
      "eval_entropy": 6.474506407712413,
      "eval_loss": 6.6858367919921875,
      "eval_mean_token_accuracy": 0.14477067293769577,
      "eval_num_tokens": 6232947.0,
      "eval_runtime": 1.7302,
      "eval_samples_per_second": 1564.56,
      "eval_steps_per_second": 195.931,
      "step": 2500
    },
    {
      "entropy": 6.636260461807251,
      "epoch": 2.9298245614035086,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000457248985103333,
      "loss": 6.3709,
      "mean_token_accuracy": 0.16393661350011826,
      "num_tokens": 6245597.0,
      "step": 2505
    },
    {
      "entropy": 6.603361558914185,
      "epoch": 2.935672514619883,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004569737043663768,
      "loss": 6.3252,
      "mean_token_accuracy": 0.15618091374635695,
      "num_tokens": 6257745.0,
      "step": 2510
    },
    {
      "entropy": 6.602143621444702,
      "epoch": 2.9415204678362574,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004566976338504945,
      "loss": 6.3858,
      "mean_token_accuracy": 0.1644023910164833,
      "num_tokens": 6268871.0,
      "step": 2515
    },
    {
      "entropy": 6.5644202709198,
      "epoch": 2.9473684210526314,
      "grad_norm": 0.796875,
      "learning_rate": 0.00045642077475385186,
      "loss": 6.3714,
      "mean_token_accuracy": 0.16085665076971054,
      "num_tokens": 6281487.0,
      "step": 2520
    },
    {
      "entropy": 6.543654108047486,
      "epoch": 2.953216374269006,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004561431282780372,
      "loss": 6.3451,
      "mean_token_accuracy": 0.16275432705879211,
      "num_tokens": 6292928.0,
      "step": 2525
    },
    {
      "entropy": 6.575348234176635,
      "epoch": 2.95906432748538,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00045586469562805624,
      "loss": 6.3098,
      "mean_token_accuracy": 0.17834650427103044,
      "num_tokens": 6304340.0,
      "step": 2530
    },
    {
      "entropy": 6.595432710647583,
      "epoch": 2.9649122807017543,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004555854780123265,
      "loss": 6.39,
      "mean_token_accuracy": 0.160810686647892,
      "num_tokens": 6315972.0,
      "step": 2535
    },
    {
      "entropy": 6.6777137279510494,
      "epoch": 2.9707602339181287,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004553054766426725,
      "loss": 6.3921,
      "mean_token_accuracy": 0.16103890985250474,
      "num_tokens": 6327482.0,
      "step": 2540
    },
    {
      "entropy": 6.561354255676269,
      "epoch": 2.976608187134503,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004550246927343203,
      "loss": 6.369,
      "mean_token_accuracy": 0.16236899346113204,
      "num_tokens": 6338932.0,
      "step": 2545
    },
    {
      "entropy": 6.559825801849366,
      "epoch": 2.982456140350877,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00045474312750589226,
      "loss": 6.4044,
      "mean_token_accuracy": 0.16020567417144777,
      "num_tokens": 6350770.0,
      "step": 2550
    },
    {
      "entropy": 6.638618278503418,
      "epoch": 2.9883040935672516,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00045446078217940155,
      "loss": 6.4193,
      "mean_token_accuracy": 0.15757756531238556,
      "num_tokens": 6363472.0,
      "step": 2555
    },
    {
      "entropy": 6.662160015106201,
      "epoch": 2.9941520467836256,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00045417765798024716,
      "loss": 6.3811,
      "mean_token_accuracy": 0.15905229449272157,
      "num_tokens": 6377174.0,
      "step": 2560
    },
    {
      "entropy": 6.551484727859497,
      "epoch": 3.0,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004538937561372086,
      "loss": 6.3637,
      "mean_token_accuracy": 0.157417456805706,
      "num_tokens": 6388452.0,
      "step": 2565
    },
    {
      "entropy": 6.70260968208313,
      "epoch": 3.0058479532163744,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004536090778824398,
      "loss": 6.0832,
      "mean_token_accuracy": 0.17075288593769072,
      "num_tokens": 6401740.0,
      "step": 2570
    },
    {
      "entropy": 6.546187019348144,
      "epoch": 3.0116959064327484,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004533236244514652,
      "loss": 6.0578,
      "mean_token_accuracy": 0.17957376539707184,
      "num_tokens": 6414850.0,
      "step": 2575
    },
    {
      "entropy": 6.6370275020599365,
      "epoch": 3.017543859649123,
      "grad_norm": 0.81640625,
      "learning_rate": 0.00045303739708317303,
      "loss": 6.2139,
      "mean_token_accuracy": 0.15932946354150773,
      "num_tokens": 6429250.0,
      "step": 2580
    },
    {
      "entropy": 6.571931838989258,
      "epoch": 3.023391812865497,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004527503970198105,
      "loss": 6.0364,
      "mean_token_accuracy": 0.17207318246364595,
      "num_tokens": 6440583.0,
      "step": 2585
    },
    {
      "entropy": 6.532599782943725,
      "epoch": 3.0292397660818713,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004524626255069786,
      "loss": 6.0535,
      "mean_token_accuracy": 0.17217930853366853,
      "num_tokens": 6452678.0,
      "step": 2590
    },
    {
      "entropy": 6.606011962890625,
      "epoch": 3.0350877192982457,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00045217408379362635,
      "loss": 6.0997,
      "mean_token_accuracy": 0.16845548003911973,
      "num_tokens": 6464751.0,
      "step": 2595
    },
    {
      "entropy": 6.45399317741394,
      "epoch": 3.0409356725146197,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00045188477313204536,
      "loss": 6.056,
      "mean_token_accuracy": 0.1708752527832985,
      "num_tokens": 6476017.0,
      "step": 2600
    },
    {
      "entropy": 6.657937335968017,
      "epoch": 3.046783625730994,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045159469477786457,
      "loss": 6.0946,
      "mean_token_accuracy": 0.1711534798145294,
      "num_tokens": 6487404.0,
      "step": 2605
    },
    {
      "entropy": 6.580844783782959,
      "epoch": 3.0526315789473686,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004513038499900451,
      "loss": 6.1365,
      "mean_token_accuracy": 0.16481128931045533,
      "num_tokens": 6500027.0,
      "step": 2610
    },
    {
      "entropy": 6.521955156326294,
      "epoch": 3.0584795321637426,
      "grad_norm": 0.890625,
      "learning_rate": 0.000451012240030874,
      "loss": 6.1302,
      "mean_token_accuracy": 0.1614599958062172,
      "num_tokens": 6513067.0,
      "step": 2615
    },
    {
      "entropy": 6.583660364151001,
      "epoch": 3.064327485380117,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045071986616595946,
      "loss": 6.0837,
      "mean_token_accuracy": 0.16450167149305345,
      "num_tokens": 6524932.0,
      "step": 2620
    },
    {
      "entropy": 6.562144422531128,
      "epoch": 3.0701754385964914,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00045042672966422506,
      "loss": 6.1179,
      "mean_token_accuracy": 0.16554533541202546,
      "num_tokens": 6536843.0,
      "step": 2625
    },
    {
      "entropy": 6.591558980941772,
      "epoch": 3.0760233918128654,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004501328317979041,
      "loss": 6.0298,
      "mean_token_accuracy": 0.17564562559127808,
      "num_tokens": 6549414.0,
      "step": 2630
    },
    {
      "entropy": 6.512995338439941,
      "epoch": 3.08187134502924,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004498381738425346,
      "loss": 6.1121,
      "mean_token_accuracy": 0.1698380097746849,
      "num_tokens": 6561492.0,
      "step": 2635
    },
    {
      "entropy": 6.513795471191406,
      "epoch": 3.087719298245614,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00044954275707695304,
      "loss": 6.0179,
      "mean_token_accuracy": 0.17263186872005462,
      "num_tokens": 6572873.0,
      "step": 2640
    },
    {
      "entropy": 6.562921571731567,
      "epoch": 3.0935672514619883,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004492465827832894,
      "loss": 6.1596,
      "mean_token_accuracy": 0.17121457010507585,
      "num_tokens": 6584902.0,
      "step": 2645
    },
    {
      "entropy": 6.5826499462127686,
      "epoch": 3.0994152046783627,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00044894965224696144,
      "loss": 6.1754,
      "mean_token_accuracy": 0.1667486160993576,
      "num_tokens": 6596792.0,
      "step": 2650
    },
    {
      "entropy": 6.559206581115722,
      "epoch": 3.1052631578947367,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00044865196675666886,
      "loss": 6.11,
      "mean_token_accuracy": 0.1693141758441925,
      "num_tokens": 6608559.0,
      "step": 2655
    },
    {
      "entropy": 6.5065614700317385,
      "epoch": 3.111111111111111,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004483535276043881,
      "loss": 6.1453,
      "mean_token_accuracy": 0.16696794629096984,
      "num_tokens": 6619938.0,
      "step": 2660
    },
    {
      "entropy": 6.5608642578125,
      "epoch": 3.116959064327485,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004480543360853665,
      "loss": 6.1206,
      "mean_token_accuracy": 0.1688737764954567,
      "num_tokens": 6632850.0,
      "step": 2665
    },
    {
      "entropy": 6.485915756225586,
      "epoch": 3.1228070175438596,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004477543934981167,
      "loss": 6.0462,
      "mean_token_accuracy": 0.1824930265545845,
      "num_tokens": 6645761.0,
      "step": 2670
    },
    {
      "entropy": 6.5529155254364015,
      "epoch": 3.128654970760234,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044745370114441103,
      "loss": 6.1162,
      "mean_token_accuracy": 0.17532826513051986,
      "num_tokens": 6657861.0,
      "step": 2675
    },
    {
      "entropy": 6.529551315307617,
      "epoch": 3.134502923976608,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044715226032927585,
      "loss": 6.0875,
      "mean_token_accuracy": 0.1792955845594406,
      "num_tokens": 6670495.0,
      "step": 2680
    },
    {
      "entropy": 6.374896669387818,
      "epoch": 3.1403508771929824,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004468500723609859,
      "loss": 6.0383,
      "mean_token_accuracy": 0.174404776096344,
      "num_tokens": 6683127.0,
      "step": 2685
    },
    {
      "entropy": 6.563790512084961,
      "epoch": 3.146198830409357,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00044654713855105864,
      "loss": 6.1114,
      "mean_token_accuracy": 0.16159912645816804,
      "num_tokens": 6696159.0,
      "step": 2690
    },
    {
      "entropy": 6.524526977539063,
      "epoch": 3.152046783625731,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004462434602142486,
      "loss": 6.1785,
      "mean_token_accuracy": 0.16437700539827346,
      "num_tokens": 6708993.0,
      "step": 2695
    },
    {
      "entropy": 6.564835500717163,
      "epoch": 3.1578947368421053,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004459390386685414,
      "loss": 6.2003,
      "mean_token_accuracy": 0.1650101974606514,
      "num_tokens": 6722121.0,
      "step": 2700
    },
    {
      "entropy": 6.484952974319458,
      "epoch": 3.1637426900584797,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044563387523514854,
      "loss": 6.023,
      "mean_token_accuracy": 0.17607033401727676,
      "num_tokens": 6734167.0,
      "step": 2705
    },
    {
      "entropy": 6.459822940826416,
      "epoch": 3.1695906432748537,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004453279712385011,
      "loss": 6.1071,
      "mean_token_accuracy": 0.16553861498832703,
      "num_tokens": 6745720.0,
      "step": 2710
    },
    {
      "entropy": 6.503742504119873,
      "epoch": 3.175438596491228,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004450213280062444,
      "loss": 6.0939,
      "mean_token_accuracy": 0.16853866428136827,
      "num_tokens": 6757523.0,
      "step": 2715
    },
    {
      "entropy": 6.568593549728393,
      "epoch": 3.181286549707602,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044471394686923206,
      "loss": 6.121,
      "mean_token_accuracy": 0.16629476100206375,
      "num_tokens": 6768775.0,
      "step": 2720
    },
    {
      "entropy": 6.525596046447754,
      "epoch": 3.1871345029239766,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004444058291615202,
      "loss": 6.1133,
      "mean_token_accuracy": 0.16562219709157944,
      "num_tokens": 6781547.0,
      "step": 2725
    },
    {
      "entropy": 6.485271024703979,
      "epoch": 3.192982456140351,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004440969762203618,
      "loss": 6.1516,
      "mean_token_accuracy": 0.16345081925392152,
      "num_tokens": 6794708.0,
      "step": 2730
    },
    {
      "entropy": 6.508287048339843,
      "epoch": 3.198830409356725,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004437873893862008,
      "loss": 6.0773,
      "mean_token_accuracy": 0.17581354528665544,
      "num_tokens": 6807738.0,
      "step": 2735
    },
    {
      "entropy": 6.511159420013428,
      "epoch": 3.2046783625730995,
      "grad_norm": 0.984375,
      "learning_rate": 0.00044347707000266627,
      "loss": 6.1269,
      "mean_token_accuracy": 0.1693233847618103,
      "num_tokens": 6820851.0,
      "step": 2740
    },
    {
      "entropy": 6.548748731613159,
      "epoch": 3.2105263157894735,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044316601941656655,
      "loss": 6.0793,
      "mean_token_accuracy": 0.1794822096824646,
      "num_tokens": 6833888.0,
      "step": 2745
    },
    {
      "entropy": 6.494293260574341,
      "epoch": 3.216374269005848,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004428542389778834,
      "loss": 6.1172,
      "mean_token_accuracy": 0.17162996679544448,
      "num_tokens": 6846104.0,
      "step": 2750
    },
    {
      "entropy": 6.45095477104187,
      "epoch": 3.2222222222222223,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004425417300397665,
      "loss": 6.1045,
      "mean_token_accuracy": 0.17195063680410386,
      "num_tokens": 6857458.0,
      "step": 2755
    },
    {
      "entropy": 6.5355345726013185,
      "epoch": 3.2280701754385963,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044222849395852683,
      "loss": 6.074,
      "mean_token_accuracy": 0.17359959185123444,
      "num_tokens": 6870455.0,
      "step": 2760
    },
    {
      "entropy": 6.5262401580810545,
      "epoch": 3.2339181286549707,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044191453209363165,
      "loss": 6.1379,
      "mean_token_accuracy": 0.1655440717935562,
      "num_tokens": 6883359.0,
      "step": 2765
    },
    {
      "entropy": 6.476034498214721,
      "epoch": 3.239766081871345,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004415998458076978,
      "loss": 6.0727,
      "mean_token_accuracy": 0.17608768194913865,
      "num_tokens": 6894940.0,
      "step": 2770
    },
    {
      "entropy": 6.5159460544586185,
      "epoch": 3.245614035087719,
      "grad_norm": 0.984375,
      "learning_rate": 0.00044128443646648634,
      "loss": 6.1268,
      "mean_token_accuracy": 0.16884497851133345,
      "num_tokens": 6905422.0,
      "step": 2775
    },
    {
      "entropy": 6.537970733642578,
      "epoch": 3.2514619883040936,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044096830543889653,
      "loss": 6.1286,
      "mean_token_accuracy": 0.1716626450419426,
      "num_tokens": 6918106.0,
      "step": 2780
    },
    {
      "entropy": 6.5358724117279055,
      "epoch": 3.257309941520468,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004406514540969596,
      "loss": 6.1015,
      "mean_token_accuracy": 0.17332532852888108,
      "num_tokens": 6930975.0,
      "step": 2785
    },
    {
      "entropy": 6.489678764343262,
      "epoch": 3.263157894736842,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00044033388381583324,
      "loss": 6.1935,
      "mean_token_accuracy": 0.16382682770490647,
      "num_tokens": 6944436.0,
      "step": 2790
    },
    {
      "entropy": 6.538244295120239,
      "epoch": 3.2690058479532165,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00044001559597379503,
      "loss": 6.1451,
      "mean_token_accuracy": 0.16451818645000457,
      "num_tokens": 6957095.0,
      "step": 2795
    },
    {
      "entropy": 6.494568872451782,
      "epoch": 3.2748538011695905,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004396965919522373,
      "loss": 6.0881,
      "mean_token_accuracy": 0.17684150040149688,
      "num_tokens": 6969792.0,
      "step": 2800
    },
    {
      "entropy": 6.4879639625549315,
      "epoch": 3.280701754385965,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00043937687313565997,
      "loss": 6.143,
      "mean_token_accuracy": 0.16732580810785294,
      "num_tokens": 6982054.0,
      "step": 2805
    },
    {
      "entropy": 6.493925428390503,
      "epoch": 3.2865497076023393,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004390564409116661,
      "loss": 6.1371,
      "mean_token_accuracy": 0.16817447543144226,
      "num_tokens": 6994057.0,
      "step": 2810
    },
    {
      "entropy": 6.474679803848266,
      "epoch": 3.2923976608187133,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00043873529667095423,
      "loss": 6.1393,
      "mean_token_accuracy": 0.1712675377726555,
      "num_tokens": 7006417.0,
      "step": 2815
    },
    {
      "entropy": 6.477682638168335,
      "epoch": 3.2982456140350878,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00043841344180731346,
      "loss": 6.1547,
      "mean_token_accuracy": 0.16124001741409302,
      "num_tokens": 7018331.0,
      "step": 2820
    },
    {
      "entropy": 6.518821668624878,
      "epoch": 3.3040935672514617,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000438090877717617,
      "loss": 6.04,
      "mean_token_accuracy": 0.17431365847587585,
      "num_tokens": 7031654.0,
      "step": 2825
    },
    {
      "entropy": 6.5531683444976805,
      "epoch": 3.309941520467836,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004377676058018161,
      "loss": 6.1393,
      "mean_token_accuracy": 0.1715486004948616,
      "num_tokens": 7044921.0,
      "step": 2830
    },
    {
      "entropy": 6.42981367111206,
      "epoch": 3.3157894736842106,
      "grad_norm": 1.0,
      "learning_rate": 0.0004374436274629341,
      "loss": 6.0735,
      "mean_token_accuracy": 0.16862040609121323,
      "num_tokens": 7056700.0,
      "step": 2835
    },
    {
      "entropy": 6.495500755310059,
      "epoch": 3.3216374269005846,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004371189441070602,
      "loss": 6.0191,
      "mean_token_accuracy": 0.16978203654289245,
      "num_tokens": 7069196.0,
      "step": 2840
    },
    {
      "entropy": 6.441921043395996,
      "epoch": 3.327485380116959,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004367935571433433,
      "loss": 6.0614,
      "mean_token_accuracy": 0.17559678256511688,
      "num_tokens": 7081528.0,
      "step": 2845
    },
    {
      "entropy": 6.566071462631226,
      "epoch": 3.3333333333333335,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004364674679839864,
      "loss": 6.1552,
      "mean_token_accuracy": 0.16209713369607925,
      "num_tokens": 7093474.0,
      "step": 2850
    },
    {
      "entropy": 6.382375192642212,
      "epoch": 3.3391812865497075,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004361406780442398,
      "loss": 6.0816,
      "mean_token_accuracy": 0.17418255656957626,
      "num_tokens": 7105508.0,
      "step": 2855
    },
    {
      "entropy": 6.453981208801269,
      "epoch": 3.345029239766082,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00043581318874239504,
      "loss": 6.0926,
      "mean_token_accuracy": 0.1744996801018715,
      "num_tokens": 7117952.0,
      "step": 2860
    },
    {
      "entropy": 6.468331241607666,
      "epoch": 3.3508771929824563,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043548500149977946,
      "loss": 6.0771,
      "mean_token_accuracy": 0.1764598608016968,
      "num_tokens": 7130666.0,
      "step": 2865
    },
    {
      "entropy": 6.466050815582276,
      "epoch": 3.3567251461988303,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004351561177407491,
      "loss": 6.1137,
      "mean_token_accuracy": 0.17108856737613679,
      "num_tokens": 7142717.0,
      "step": 2870
    },
    {
      "entropy": 6.552111864089966,
      "epoch": 3.3625730994152048,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004348265388926831,
      "loss": 6.15,
      "mean_token_accuracy": 0.1713667944073677,
      "num_tokens": 7154794.0,
      "step": 2875
    },
    {
      "entropy": 6.496505641937256,
      "epoch": 3.3684210526315788,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004344962663859771,
      "loss": 6.1995,
      "mean_token_accuracy": 0.16334100663661957,
      "num_tokens": 7168907.0,
      "step": 2880
    },
    {
      "entropy": 6.445120096206665,
      "epoch": 3.374269005847953,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00043416530165403777,
      "loss": 6.033,
      "mean_token_accuracy": 0.18498492389917373,
      "num_tokens": 7180828.0,
      "step": 2885
    },
    {
      "entropy": 6.360263347625732,
      "epoch": 3.3801169590643276,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004338336461332756,
      "loss": 6.0388,
      "mean_token_accuracy": 0.17281053215265274,
      "num_tokens": 7193493.0,
      "step": 2890
    },
    {
      "entropy": 6.499506950378418,
      "epoch": 3.3859649122807016,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00043350130126309936,
      "loss": 6.083,
      "mean_token_accuracy": 0.16768689304590226,
      "num_tokens": 7206033.0,
      "step": 2895
    },
    {
      "entropy": 6.505532598495483,
      "epoch": 3.391812865497076,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00043316826848590955,
      "loss": 6.114,
      "mean_token_accuracy": 0.16994736790657045,
      "num_tokens": 7218813.0,
      "step": 2900
    },
    {
      "entropy": 6.444248628616333,
      "epoch": 3.39766081871345,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004328345492470924,
      "loss": 6.0908,
      "mean_token_accuracy": 0.17362091243267058,
      "num_tokens": 7230698.0,
      "step": 2905
    },
    {
      "entropy": 6.474144983291626,
      "epoch": 3.4035087719298245,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004325001449950134,
      "loss": 6.1217,
      "mean_token_accuracy": 0.1663724511861801,
      "num_tokens": 7243354.0,
      "step": 2910
    },
    {
      "entropy": 6.392809629440308,
      "epoch": 3.409356725146199,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004321650571810109,
      "loss": 6.0684,
      "mean_token_accuracy": 0.17750578969717026,
      "num_tokens": 7255855.0,
      "step": 2915
    },
    {
      "entropy": 6.396269178390503,
      "epoch": 3.415204678362573,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004318292872593899,
      "loss": 6.1589,
      "mean_token_accuracy": 0.16955514401197433,
      "num_tokens": 7268921.0,
      "step": 2920
    },
    {
      "entropy": 6.550719833374023,
      "epoch": 3.4210526315789473,
      "grad_norm": 1.0,
      "learning_rate": 0.0004314928366874162,
      "loss": 6.1483,
      "mean_token_accuracy": 0.16568040251731872,
      "num_tokens": 7281467.0,
      "step": 2925
    },
    {
      "entropy": 6.365495777130127,
      "epoch": 3.426900584795322,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004311557069253092,
      "loss": 6.0591,
      "mean_token_accuracy": 0.17555536329746246,
      "num_tokens": 7294887.0,
      "step": 2930
    },
    {
      "entropy": 6.547408819198608,
      "epoch": 3.4327485380116958,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004308178994362364,
      "loss": 6.1935,
      "mean_token_accuracy": 0.17171871960163115,
      "num_tokens": 7307009.0,
      "step": 2935
    },
    {
      "entropy": 6.42487940788269,
      "epoch": 3.43859649122807,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00043047941568630644,
      "loss": 6.1048,
      "mean_token_accuracy": 0.16843656450510025,
      "num_tokens": 7319751.0,
      "step": 2940
    },
    {
      "entropy": 6.515281867980957,
      "epoch": 3.4444444444444446,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043014025714456305,
      "loss": 6.2338,
      "mean_token_accuracy": 0.16237418949604035,
      "num_tokens": 7333952.0,
      "step": 2945
    },
    {
      "entropy": 6.445792150497437,
      "epoch": 3.4502923976608186,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004298004252829787,
      "loss": 5.988,
      "mean_token_accuracy": 0.1804724305868149,
      "num_tokens": 7346834.0,
      "step": 2950
    },
    {
      "entropy": 6.475040483474731,
      "epoch": 3.456140350877193,
      "grad_norm": 0.90625,
      "learning_rate": 0.000429459921576448,
      "loss": 6.1279,
      "mean_token_accuracy": 0.16938423365354538,
      "num_tokens": 7359165.0,
      "step": 2955
    },
    {
      "entropy": 6.506986474990844,
      "epoch": 3.461988304093567,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00042911874750278133,
      "loss": 6.1996,
      "mean_token_accuracy": 0.16553546041250228,
      "num_tokens": 7371585.0,
      "step": 2960
    },
    {
      "entropy": 6.432463550567627,
      "epoch": 3.4678362573099415,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00042877690454269884,
      "loss": 6.0887,
      "mean_token_accuracy": 0.17572804987430574,
      "num_tokens": 7383455.0,
      "step": 2965
    },
    {
      "entropy": 6.420023965835571,
      "epoch": 3.473684210526316,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042843439417982344,
      "loss": 6.1506,
      "mean_token_accuracy": 0.17092279493808746,
      "num_tokens": 7395918.0,
      "step": 2970
    },
    {
      "entropy": 6.524562644958496,
      "epoch": 3.47953216374269,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004280912179006746,
      "loss": 6.1334,
      "mean_token_accuracy": 0.1703528344631195,
      "num_tokens": 7408053.0,
      "step": 2975
    },
    {
      "entropy": 6.497876214981079,
      "epoch": 3.4853801169590644,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000427747377194662,
      "loss": 6.1778,
      "mean_token_accuracy": 0.16454286128282547,
      "num_tokens": 7419947.0,
      "step": 2980
    },
    {
      "entropy": 6.438177871704101,
      "epoch": 3.4912280701754383,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004274028735540789,
      "loss": 6.0593,
      "mean_token_accuracy": 0.16666590571403503,
      "num_tokens": 7433022.0,
      "step": 2985
    },
    {
      "entropy": 6.411357355117798,
      "epoch": 3.497076023391813,
      "grad_norm": 0.96875,
      "learning_rate": 0.00042705770847409596,
      "loss": 6.0595,
      "mean_token_accuracy": 0.17704250365495683,
      "num_tokens": 7445280.0,
      "step": 2990
    },
    {
      "entropy": 6.5110695362091064,
      "epoch": 3.502923976608187,
      "grad_norm": 0.984375,
      "learning_rate": 0.00042671188345275423,
      "loss": 6.0611,
      "mean_token_accuracy": 0.17194437682628633,
      "num_tokens": 7457324.0,
      "step": 2995
    },
    {
      "entropy": 6.45655927658081,
      "epoch": 3.5087719298245617,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004263653999909593,
      "loss": 6.1125,
      "mean_token_accuracy": 0.16834226250648499,
      "num_tokens": 7469070.0,
      "step": 3000
    },
    {
      "epoch": 3.5087719298245617,
      "eval_entropy": 6.256012293441457,
      "eval_loss": 6.655336380004883,
      "eval_mean_token_accuracy": 0.14697449852418495,
      "eval_num_tokens": 7469070.0,
      "eval_runtime": 1.7405,
      "eval_samples_per_second": 1555.284,
      "eval_steps_per_second": 194.77,
      "step": 3000
    },
    {
      "entropy": 6.424974966049194,
      "epoch": 3.5146198830409356,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000426018259592474,
      "loss": 6.1701,
      "mean_token_accuracy": 0.16661889404058455,
      "num_tokens": 7480988.0,
      "step": 3005
    },
    {
      "entropy": 6.492739868164063,
      "epoch": 3.52046783625731,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004256704637639126,
      "loss": 6.0832,
      "mean_token_accuracy": 0.17816389799118043,
      "num_tokens": 7493557.0,
      "step": 3010
    },
    {
      "entropy": 6.454852533340454,
      "epoch": 3.526315789473684,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004253220140147341,
      "loss": 6.129,
      "mean_token_accuracy": 0.17006712257862092,
      "num_tokens": 7507446.0,
      "step": 3015
    },
    {
      "entropy": 6.507586193084717,
      "epoch": 3.5321637426900585,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004249729118572352,
      "loss": 6.0549,
      "mean_token_accuracy": 0.16801738142967224,
      "num_tokens": 7520054.0,
      "step": 3020
    },
    {
      "entropy": 6.39217095375061,
      "epoch": 3.538011695906433,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004246231588065444,
      "loss": 5.9577,
      "mean_token_accuracy": 0.18241367787122725,
      "num_tokens": 7533141.0,
      "step": 3025
    },
    {
      "entropy": 6.451664876937866,
      "epoch": 3.543859649122807,
      "grad_norm": 0.890625,
      "learning_rate": 0.000424272756380615,
      "loss": 6.141,
      "mean_token_accuracy": 0.1678723767399788,
      "num_tokens": 7545544.0,
      "step": 3030
    },
    {
      "entropy": 6.413544464111328,
      "epoch": 3.5497076023391814,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004239217061002188,
      "loss": 6.1115,
      "mean_token_accuracy": 0.17452922016382216,
      "num_tokens": 7558055.0,
      "step": 3035
    },
    {
      "entropy": 6.432606601715088,
      "epoch": 3.5555555555555554,
      "grad_norm": 0.921875,
      "learning_rate": 0.00042357000948893894,
      "loss": 6.0319,
      "mean_token_accuracy": 0.18093366473913192,
      "num_tokens": 7569951.0,
      "step": 3040
    },
    {
      "entropy": 6.4237226963043215,
      "epoch": 3.56140350877193,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042321766807316405,
      "loss": 6.1745,
      "mean_token_accuracy": 0.1635773092508316,
      "num_tokens": 7581810.0,
      "step": 3045
    },
    {
      "entropy": 6.52658429145813,
      "epoch": 3.5672514619883042,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00042286468338208113,
      "loss": 6.1332,
      "mean_token_accuracy": 0.1704472318291664,
      "num_tokens": 7594896.0,
      "step": 3050
    },
    {
      "entropy": 6.439553594589233,
      "epoch": 3.573099415204678,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004225110569476691,
      "loss": 6.1346,
      "mean_token_accuracy": 0.1705726206302643,
      "num_tokens": 7606620.0,
      "step": 3055
    },
    {
      "entropy": 6.401493263244629,
      "epoch": 3.5789473684210527,
      "grad_norm": 1.0,
      "learning_rate": 0.000422156790304692,
      "loss": 6.0067,
      "mean_token_accuracy": 0.1793942779302597,
      "num_tokens": 7619493.0,
      "step": 3060
    },
    {
      "entropy": 6.420085477828979,
      "epoch": 3.5847953216374266,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042180188499069246,
      "loss": 6.0318,
      "mean_token_accuracy": 0.16971791684627532,
      "num_tokens": 7632711.0,
      "step": 3065
    },
    {
      "entropy": 6.460293769836426,
      "epoch": 3.590643274853801,
      "grad_norm": 0.984375,
      "learning_rate": 0.000421446342545985,
      "loss": 6.0863,
      "mean_token_accuracy": 0.1706265017390251,
      "num_tokens": 7645209.0,
      "step": 3070
    },
    {
      "entropy": 6.469331502914429,
      "epoch": 3.5964912280701755,
      "grad_norm": 0.984375,
      "learning_rate": 0.00042109016451364933,
      "loss": 6.1804,
      "mean_token_accuracy": 0.1643272653222084,
      "num_tokens": 7657347.0,
      "step": 3075
    },
    {
      "entropy": 6.553790092468262,
      "epoch": 3.60233918128655,
      "grad_norm": 0.90625,
      "learning_rate": 0.00042073335243952354,
      "loss": 6.1296,
      "mean_token_accuracy": 0.16287589818239212,
      "num_tokens": 7669593.0,
      "step": 3080
    },
    {
      "entropy": 6.452272129058838,
      "epoch": 3.608187134502924,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004203759078721979,
      "loss": 6.1508,
      "mean_token_accuracy": 0.16433351039886473,
      "num_tokens": 7682548.0,
      "step": 3085
    },
    {
      "entropy": 6.484198188781738,
      "epoch": 3.6140350877192984,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004200178323630072,
      "loss": 6.1617,
      "mean_token_accuracy": 0.16616058647632598,
      "num_tokens": 7694554.0,
      "step": 3090
    },
    {
      "entropy": 6.466509199142456,
      "epoch": 3.6198830409356724,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00041965912746602495,
      "loss": 6.193,
      "mean_token_accuracy": 0.17061641663312913,
      "num_tokens": 7707747.0,
      "step": 3095
    },
    {
      "entropy": 6.512132596969605,
      "epoch": 3.625730994152047,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00041929979473805606,
      "loss": 6.1595,
      "mean_token_accuracy": 0.17124320566654205,
      "num_tokens": 7719487.0,
      "step": 3100
    },
    {
      "entropy": 6.508446979522705,
      "epoch": 3.6315789473684212,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00041893983573863044,
      "loss": 6.1994,
      "mean_token_accuracy": 0.16181788444519044,
      "num_tokens": 7732531.0,
      "step": 3105
    },
    {
      "entropy": 6.4622407913208,
      "epoch": 3.6374269005847952,
      "grad_norm": 0.921875,
      "learning_rate": 0.00041857925202999586,
      "loss": 6.115,
      "mean_token_accuracy": 0.17386947870254515,
      "num_tokens": 7745034.0,
      "step": 3110
    },
    {
      "entropy": 6.488437223434448,
      "epoch": 3.6432748538011697,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004182180451771115,
      "loss": 6.1023,
      "mean_token_accuracy": 0.1699843227863312,
      "num_tokens": 7757442.0,
      "step": 3115
    },
    {
      "entropy": 6.378793048858642,
      "epoch": 3.6491228070175437,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004178562167476409,
      "loss": 6.1242,
      "mean_token_accuracy": 0.16865306198596955,
      "num_tokens": 7770358.0,
      "step": 3120
    },
    {
      "entropy": 6.451465606689453,
      "epoch": 3.654970760233918,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00041749376831194565,
      "loss": 6.1071,
      "mean_token_accuracy": 0.16937971562147142,
      "num_tokens": 7783611.0,
      "step": 3125
    },
    {
      "entropy": 6.415663146972657,
      "epoch": 3.6608187134502925,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004171307014430777,
      "loss": 6.1213,
      "mean_token_accuracy": 0.16577081829309465,
      "num_tokens": 7795522.0,
      "step": 3130
    },
    {
      "entropy": 6.512515115737915,
      "epoch": 3.6666666666666665,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00041676701771677343,
      "loss": 6.0944,
      "mean_token_accuracy": 0.17461688667535782,
      "num_tokens": 7806801.0,
      "step": 3135
    },
    {
      "entropy": 6.435156202316284,
      "epoch": 3.672514619883041,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004164027187114463,
      "loss": 6.1822,
      "mean_token_accuracy": 0.16419006139039993,
      "num_tokens": 7821084.0,
      "step": 3140
    },
    {
      "entropy": 6.2991251945495605,
      "epoch": 3.678362573099415,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00041603780600818,
      "loss": 5.9264,
      "mean_token_accuracy": 0.19005913138389588,
      "num_tokens": 7834065.0,
      "step": 3145
    },
    {
      "entropy": 6.449855709075928,
      "epoch": 3.6842105263157894,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004156722811907219,
      "loss": 6.1059,
      "mean_token_accuracy": 0.17338153272867202,
      "num_tokens": 7846378.0,
      "step": 3150
    },
    {
      "entropy": 6.42774338722229,
      "epoch": 3.690058479532164,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041530614584547594,
      "loss": 6.1363,
      "mean_token_accuracy": 0.16735569536685943,
      "num_tokens": 7858933.0,
      "step": 3155
    },
    {
      "entropy": 6.402468252182007,
      "epoch": 3.6959064327485383,
      "grad_norm": 0.9375,
      "learning_rate": 0.00041493940156149577,
      "loss": 5.9806,
      "mean_token_accuracy": 0.17447462528944016,
      "num_tokens": 7870408.0,
      "step": 3160
    },
    {
      "entropy": 6.468065357208252,
      "epoch": 3.7017543859649122,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000414572049930478,
      "loss": 6.1341,
      "mean_token_accuracy": 0.16686153709888457,
      "num_tokens": 7882778.0,
      "step": 3165
    },
    {
      "entropy": 6.392737102508545,
      "epoch": 3.7076023391812867,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041420409254675465,
      "loss": 6.1086,
      "mean_token_accuracy": 0.16954588741064072,
      "num_tokens": 7895169.0,
      "step": 3170
    },
    {
      "entropy": 6.462703657150269,
      "epoch": 3.7134502923976607,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004138355310072876,
      "loss": 6.1637,
      "mean_token_accuracy": 0.16734591722488404,
      "num_tokens": 7908607.0,
      "step": 3175
    },
    {
      "entropy": 6.486843824386597,
      "epoch": 3.719298245614035,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00041346636691166015,
      "loss": 6.118,
      "mean_token_accuracy": 0.17107573002576829,
      "num_tokens": 7922269.0,
      "step": 3180
    },
    {
      "entropy": 6.361353778839112,
      "epoch": 3.7251461988304095,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00041309660186207093,
      "loss": 6.0687,
      "mean_token_accuracy": 0.17154397070407867,
      "num_tokens": 7935205.0,
      "step": 3185
    },
    {
      "entropy": 6.438766384124756,
      "epoch": 3.7309941520467835,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041272623746332687,
      "loss": 6.1277,
      "mean_token_accuracy": 0.17044788748025894,
      "num_tokens": 7947970.0,
      "step": 3190
    },
    {
      "entropy": 6.252483320236206,
      "epoch": 3.736842105263158,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00041235527532283603,
      "loss": 5.9583,
      "mean_token_accuracy": 0.18791067749261856,
      "num_tokens": 7961978.0,
      "step": 3195
    },
    {
      "entropy": 6.507790040969849,
      "epoch": 3.742690058479532,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004119837170506005,
      "loss": 6.1446,
      "mean_token_accuracy": 0.17097347974777222,
      "num_tokens": 7974708.0,
      "step": 3200
    },
    {
      "entropy": 6.442420482635498,
      "epoch": 3.7485380116959064,
      "grad_norm": 0.984375,
      "learning_rate": 0.00041161156425921004,
      "loss": 6.1003,
      "mean_token_accuracy": 0.1739748165011406,
      "num_tokens": 7986179.0,
      "step": 3205
    },
    {
      "entropy": 6.3460376262664795,
      "epoch": 3.754385964912281,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00041123881856383436,
      "loss": 6.1403,
      "mean_token_accuracy": 0.1733413815498352,
      "num_tokens": 7997816.0,
      "step": 3210
    },
    {
      "entropy": 6.4725494384765625,
      "epoch": 3.760233918128655,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004108654815822165,
      "loss": 6.0623,
      "mean_token_accuracy": 0.17147106677293777,
      "num_tokens": 8009407.0,
      "step": 3215
    },
    {
      "entropy": 6.410530233383179,
      "epoch": 3.7660818713450293,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004104915549346658,
      "loss": 6.1453,
      "mean_token_accuracy": 0.16321171075105667,
      "num_tokens": 8021446.0,
      "step": 3220
    },
    {
      "entropy": 6.474694013595581,
      "epoch": 3.7719298245614032,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041011704024405065,
      "loss": 6.1752,
      "mean_token_accuracy": 0.16454457938671113,
      "num_tokens": 8035219.0,
      "step": 3225
    },
    {
      "entropy": 6.454151630401611,
      "epoch": 3.7777777777777777,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00040974193913579174,
      "loss": 5.9859,
      "mean_token_accuracy": 0.18320498317480088,
      "num_tokens": 8047619.0,
      "step": 3230
    },
    {
      "entropy": 6.4610865116119385,
      "epoch": 3.783625730994152,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004093662532378547,
      "loss": 6.1706,
      "mean_token_accuracy": 0.16919512152671815,
      "num_tokens": 8059241.0,
      "step": 3235
    },
    {
      "entropy": 6.434743976593017,
      "epoch": 3.7894736842105265,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004089899841807433,
      "loss": 6.1169,
      "mean_token_accuracy": 0.17338500022888184,
      "num_tokens": 8071543.0,
      "step": 3240
    },
    {
      "entropy": 6.485409736633301,
      "epoch": 3.7953216374269005,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00040861313359749217,
      "loss": 6.1521,
      "mean_token_accuracy": 0.16957753002643586,
      "num_tokens": 8083515.0,
      "step": 3245
    },
    {
      "entropy": 6.370771169662476,
      "epoch": 3.801169590643275,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00040823570312366005,
      "loss": 6.0849,
      "mean_token_accuracy": 0.17670764029026031,
      "num_tokens": 8095703.0,
      "step": 3250
    },
    {
      "entropy": 6.485049057006836,
      "epoch": 3.807017543859649,
      "grad_norm": 0.890625,
      "learning_rate": 0.000407857694397322,
      "loss": 6.1969,
      "mean_token_accuracy": 0.16475343108177185,
      "num_tokens": 8108285.0,
      "step": 3255
    },
    {
      "entropy": 6.389813423156738,
      "epoch": 3.8128654970760234,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004074791090590632,
      "loss": 6.1115,
      "mean_token_accuracy": 0.16721750795841217,
      "num_tokens": 8120943.0,
      "step": 3260
    },
    {
      "entropy": 6.437641382217407,
      "epoch": 3.818713450292398,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004070999487519711,
      "loss": 6.1495,
      "mean_token_accuracy": 0.1652654692530632,
      "num_tokens": 8133086.0,
      "step": 3265
    },
    {
      "entropy": 6.49128098487854,
      "epoch": 3.824561403508772,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00040672021512162864,
      "loss": 6.1745,
      "mean_token_accuracy": 0.16469226330518721,
      "num_tokens": 8147320.0,
      "step": 3270
    },
    {
      "entropy": 6.47356424331665,
      "epoch": 3.8304093567251463,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000406339909816107,
      "loss": 6.121,
      "mean_token_accuracy": 0.16807140707969664,
      "num_tokens": 8160796.0,
      "step": 3275
    },
    {
      "entropy": 6.469601964950561,
      "epoch": 3.8362573099415203,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004059590344859584,
      "loss": 6.0925,
      "mean_token_accuracy": 0.17612387090921403,
      "num_tokens": 8173087.0,
      "step": 3280
    },
    {
      "entropy": 6.361574411392212,
      "epoch": 3.8421052631578947,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004055775907842092,
      "loss": 6.0452,
      "mean_token_accuracy": 0.1729988381266594,
      "num_tokens": 8185274.0,
      "step": 3285
    },
    {
      "entropy": 6.406063556671143,
      "epoch": 3.847953216374269,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004051955803663523,
      "loss": 6.1195,
      "mean_token_accuracy": 0.17484242767095565,
      "num_tokens": 8197753.0,
      "step": 3290
    },
    {
      "entropy": 6.389527368545532,
      "epoch": 3.853801169590643,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004048130048903404,
      "loss": 6.0931,
      "mean_token_accuracy": 0.17322594076395034,
      "num_tokens": 8211882.0,
      "step": 3295
    },
    {
      "entropy": 6.313014936447144,
      "epoch": 3.8596491228070176,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004044298660165786,
      "loss": 5.978,
      "mean_token_accuracy": 0.18418084233999252,
      "num_tokens": 8225089.0,
      "step": 3300
    },
    {
      "entropy": 6.415736484527588,
      "epoch": 3.8654970760233915,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040404616540791683,
      "loss": 6.1589,
      "mean_token_accuracy": 0.16737642288208007,
      "num_tokens": 8238808.0,
      "step": 3305
    },
    {
      "entropy": 6.422418117523193,
      "epoch": 3.871345029239766,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004036619047296435,
      "loss": 6.0743,
      "mean_token_accuracy": 0.17765308767557145,
      "num_tokens": 8252482.0,
      "step": 3310
    },
    {
      "entropy": 6.338311100006104,
      "epoch": 3.8771929824561404,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004032770856494772,
      "loss": 6.1298,
      "mean_token_accuracy": 0.16917635649442672,
      "num_tokens": 8264232.0,
      "step": 3315
    },
    {
      "entropy": 6.450420999526978,
      "epoch": 3.883040935672515,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00040289170983756075,
      "loss": 6.0931,
      "mean_token_accuracy": 0.1663595125079155,
      "num_tokens": 8276784.0,
      "step": 3320
    },
    {
      "entropy": 6.435357093811035,
      "epoch": 3.888888888888889,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004025057789664524,
      "loss": 6.0558,
      "mean_token_accuracy": 0.17435268461704254,
      "num_tokens": 8287887.0,
      "step": 3325
    },
    {
      "entropy": 6.271697330474853,
      "epoch": 3.8947368421052633,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004021192947111201,
      "loss": 5.8968,
      "mean_token_accuracy": 0.18378861248493195,
      "num_tokens": 8301258.0,
      "step": 3330
    },
    {
      "entropy": 6.43449969291687,
      "epoch": 3.9005847953216373,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040173225874893305,
      "loss": 6.1295,
      "mean_token_accuracy": 0.17237055897712708,
      "num_tokens": 8313520.0,
      "step": 3335
    },
    {
      "entropy": 6.544717931747437,
      "epoch": 3.9064327485380117,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040134467275965515,
      "loss": 6.1785,
      "mean_token_accuracy": 0.16572460979223252,
      "num_tokens": 8325597.0,
      "step": 3340
    },
    {
      "entropy": 6.4480797290802006,
      "epoch": 3.912280701754386,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004009565384254374,
      "loss": 6.1892,
      "mean_token_accuracy": 0.16745839416980743,
      "num_tokens": 8337449.0,
      "step": 3345
    },
    {
      "entropy": 6.432670927047729,
      "epoch": 3.91812865497076,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00040056785743081065,
      "loss": 6.1003,
      "mean_token_accuracy": 0.17242465615272523,
      "num_tokens": 8348802.0,
      "step": 3350
    },
    {
      "entropy": 6.422584533691406,
      "epoch": 3.9239766081871346,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004001786314626783,
      "loss": 6.0866,
      "mean_token_accuracy": 0.16576772332191467,
      "num_tokens": 8360284.0,
      "step": 3355
    },
    {
      "entropy": 6.43928074836731,
      "epoch": 3.9298245614035086,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003997888622103088,
      "loss": 6.1398,
      "mean_token_accuracy": 0.16651687473058702,
      "num_tokens": 8372080.0,
      "step": 3360
    },
    {
      "entropy": 6.43187780380249,
      "epoch": 3.935672514619883,
      "grad_norm": 0.921875,
      "learning_rate": 0.0003993985513653289,
      "loss": 6.1035,
      "mean_token_accuracy": 0.17269468009471894,
      "num_tokens": 8385178.0,
      "step": 3365
    },
    {
      "entropy": 6.421070384979248,
      "epoch": 3.9415204678362574,
      "grad_norm": 0.9375,
      "learning_rate": 0.0003990077006217155,
      "loss": 6.0662,
      "mean_token_accuracy": 0.17701358944177628,
      "num_tokens": 8397735.0,
      "step": 3370
    },
    {
      "entropy": 6.418076086044311,
      "epoch": 3.9473684210526314,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000398616311675789,
      "loss": 6.1536,
      "mean_token_accuracy": 0.1681624472141266,
      "num_tokens": 8409916.0,
      "step": 3375
    },
    {
      "entropy": 6.470368814468384,
      "epoch": 3.953216374269006,
      "grad_norm": 1.0,
      "learning_rate": 0.00039822438622620523,
      "loss": 6.1665,
      "mean_token_accuracy": 0.16600227802991868,
      "num_tokens": 8421883.0,
      "step": 3380
    },
    {
      "entropy": 6.345208072662354,
      "epoch": 3.95906432748538,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039783192597394896,
      "loss": 6.0127,
      "mean_token_accuracy": 0.18171416074037552,
      "num_tokens": 8433610.0,
      "step": 3385
    },
    {
      "entropy": 6.3606593132019045,
      "epoch": 3.9649122807017543,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00039743893262232584,
      "loss": 6.1033,
      "mean_token_accuracy": 0.17015521824359894,
      "num_tokens": 8446194.0,
      "step": 3390
    },
    {
      "entropy": 6.440540599822998,
      "epoch": 3.9707602339181287,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0003970454078769551,
      "loss": 6.0653,
      "mean_token_accuracy": 0.17913173586130143,
      "num_tokens": 8458846.0,
      "step": 3395
    },
    {
      "entropy": 6.369477272033691,
      "epoch": 3.976608187134503,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003966513534457623,
      "loss": 6.0808,
      "mean_token_accuracy": 0.1697552502155304,
      "num_tokens": 8470087.0,
      "step": 3400
    },
    {
      "entropy": 6.436861753463745,
      "epoch": 3.982456140350877,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003962567710389721,
      "loss": 6.0962,
      "mean_token_accuracy": 0.16666331142187119,
      "num_tokens": 8481293.0,
      "step": 3405
    },
    {
      "entropy": 6.439307403564453,
      "epoch": 3.9883040935672516,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00039586166236910033,
      "loss": 6.117,
      "mean_token_accuracy": 0.1810487389564514,
      "num_tokens": 8493376.0,
      "step": 3410
    },
    {
      "entropy": 6.3059323787689205,
      "epoch": 3.9941520467836256,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003954660291509469,
      "loss": 5.9915,
      "mean_token_accuracy": 0.17594255656003951,
      "num_tokens": 8506132.0,
      "step": 3415
    },
    {
      "entropy": 6.434258508682251,
      "epoch": 4.0,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039506987310158854,
      "loss": 6.1063,
      "mean_token_accuracy": 0.16990079283714293,
      "num_tokens": 8517936.0,
      "step": 3420
    },
    {
      "entropy": 6.463548040390014,
      "epoch": 4.005847953216374,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00039467319594037076,
      "loss": 5.8035,
      "mean_token_accuracy": 0.17916249930858613,
      "num_tokens": 8529863.0,
      "step": 3425
    },
    {
      "entropy": 6.35048656463623,
      "epoch": 4.011695906432749,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0003942759993889009,
      "loss": 5.8011,
      "mean_token_accuracy": 0.18165926188230513,
      "num_tokens": 8543621.0,
      "step": 3430
    },
    {
      "entropy": 6.366998958587646,
      "epoch": 4.017543859649122,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039387828517104044,
      "loss": 5.7793,
      "mean_token_accuracy": 0.19277448654174806,
      "num_tokens": 8557865.0,
      "step": 3435
    },
    {
      "entropy": 6.335794258117676,
      "epoch": 4.023391812865497,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00039348005501289775,
      "loss": 5.7102,
      "mean_token_accuracy": 0.18506938666105271,
      "num_tokens": 8570203.0,
      "step": 3440
    },
    {
      "entropy": 6.3033116340637205,
      "epoch": 4.029239766081871,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00039308131064282016,
      "loss": 5.7906,
      "mean_token_accuracy": 0.18913604617118834,
      "num_tokens": 8582638.0,
      "step": 3445
    },
    {
      "entropy": 6.415688371658325,
      "epoch": 4.035087719298246,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0003926820537913868,
      "loss": 5.8771,
      "mean_token_accuracy": 0.18149053007364274,
      "num_tokens": 8594973.0,
      "step": 3450
    },
    {
      "entropy": 6.377730035781861,
      "epoch": 4.04093567251462,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0003922822861914012,
      "loss": 5.7352,
      "mean_token_accuracy": 0.19156209081411363,
      "num_tokens": 8607216.0,
      "step": 3455
    },
    {
      "entropy": 6.4414161205291744,
      "epoch": 4.046783625730994,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003918820095778832,
      "loss": 5.9594,
      "mean_token_accuracy": 0.1749463737010956,
      "num_tokens": 8620833.0,
      "step": 3460
    },
    {
      "entropy": 6.354426431655884,
      "epoch": 4.052631578947368,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00039148122568806215,
      "loss": 5.7707,
      "mean_token_accuracy": 0.18958534002304078,
      "num_tokens": 8632931.0,
      "step": 3465
    },
    {
      "entropy": 6.2928613185882565,
      "epoch": 4.058479532163743,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003910799362613688,
      "loss": 5.8491,
      "mean_token_accuracy": 0.18395743668079376,
      "num_tokens": 8644226.0,
      "step": 3470
    },
    {
      "entropy": 6.316759872436523,
      "epoch": 4.064327485380117,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039067814303942806,
      "loss": 5.6716,
      "mean_token_accuracy": 0.2008829265832901,
      "num_tokens": 8656569.0,
      "step": 3475
    },
    {
      "entropy": 6.224283170700073,
      "epoch": 4.0701754385964914,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039027584776605136,
      "loss": 5.7001,
      "mean_token_accuracy": 0.19264741092920304,
      "num_tokens": 8668077.0,
      "step": 3480
    },
    {
      "entropy": 6.354256200790405,
      "epoch": 4.076023391812866,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00038987305218722904,
      "loss": 5.7756,
      "mean_token_accuracy": 0.18626313805580139,
      "num_tokens": 8680227.0,
      "step": 3485
    },
    {
      "entropy": 6.409672498703003,
      "epoch": 4.081871345029239,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003894697580511226,
      "loss": 5.8718,
      "mean_token_accuracy": 0.18492450565099716,
      "num_tokens": 8692354.0,
      "step": 3490
    },
    {
      "entropy": 6.364502048492431,
      "epoch": 4.087719298245614,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003890659671080579,
      "loss": 5.819,
      "mean_token_accuracy": 0.17268625646829605,
      "num_tokens": 8704634.0,
      "step": 3495
    },
    {
      "entropy": 6.3688794612884525,
      "epoch": 4.093567251461988,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003886616811105162,
      "loss": 5.8645,
      "mean_token_accuracy": 0.17775684893131255,
      "num_tokens": 8717253.0,
      "step": 3500
    },
    {
      "epoch": 4.093567251461988,
      "eval_entropy": 6.298222627611638,
      "eval_loss": 6.602290630340576,
      "eval_mean_token_accuracy": 0.14746918304545864,
      "eval_num_tokens": 8717253.0,
      "eval_runtime": 1.9271,
      "eval_samples_per_second": 1404.704,
      "eval_steps_per_second": 175.912,
      "step": 3500
    },
    {
      "entropy": 6.502414894104004,
      "epoch": 4.099415204678363,
      "grad_norm": 1.0,
      "learning_rate": 0.00038825690181312793,
      "loss": 5.8509,
      "mean_token_accuracy": 0.18152675479650499,
      "num_tokens": 8729840.0,
      "step": 3505
    },
    {
      "entropy": 6.222588634490966,
      "epoch": 4.105263157894737,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038785163097266424,
      "loss": 5.716,
      "mean_token_accuracy": 0.19439588040113448,
      "num_tokens": 8742065.0,
      "step": 3510
    },
    {
      "entropy": 6.385587692260742,
      "epoch": 4.111111111111111,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003874458703480296,
      "loss": 5.8821,
      "mean_token_accuracy": 0.1834227293729782,
      "num_tokens": 8754113.0,
      "step": 3515
    },
    {
      "entropy": 6.477599239349365,
      "epoch": 4.116959064327485,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003870396217002542,
      "loss": 5.9333,
      "mean_token_accuracy": 0.16871431171894075,
      "num_tokens": 8767255.0,
      "step": 3520
    },
    {
      "entropy": 6.320106887817383,
      "epoch": 4.12280701754386,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0003866328867924864,
      "loss": 5.8343,
      "mean_token_accuracy": 0.17676957547664643,
      "num_tokens": 8781010.0,
      "step": 3525
    },
    {
      "entropy": 6.413438034057617,
      "epoch": 4.128654970760234,
      "grad_norm": 1.0,
      "learning_rate": 0.0003862256673899848,
      "loss": 5.8609,
      "mean_token_accuracy": 0.1820666030049324,
      "num_tokens": 8794393.0,
      "step": 3530
    },
    {
      "entropy": 6.368429470062256,
      "epoch": 4.1345029239766085,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0003858179652601108,
      "loss": 5.9272,
      "mean_token_accuracy": 0.1709936022758484,
      "num_tokens": 8806496.0,
      "step": 3535
    },
    {
      "entropy": 6.328160095214844,
      "epoch": 4.140350877192983,
      "grad_norm": 0.984375,
      "learning_rate": 0.00038540978217232086,
      "loss": 5.7405,
      "mean_token_accuracy": 0.1859671249985695,
      "num_tokens": 8818411.0,
      "step": 3540
    },
    {
      "entropy": 6.385943603515625,
      "epoch": 4.146198830409356,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038500111989815894,
      "loss": 5.8365,
      "mean_token_accuracy": 0.17890818864107133,
      "num_tokens": 8829714.0,
      "step": 3545
    },
    {
      "entropy": 6.3169248580932615,
      "epoch": 4.152046783625731,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038459198021124853,
      "loss": 5.8154,
      "mean_token_accuracy": 0.18563155084848404,
      "num_tokens": 8842393.0,
      "step": 3550
    },
    {
      "entropy": 6.397756433486938,
      "epoch": 4.157894736842105,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003841823648872852,
      "loss": 5.8286,
      "mean_token_accuracy": 0.18041616380214692,
      "num_tokens": 8855263.0,
      "step": 3555
    },
    {
      "entropy": 6.366588592529297,
      "epoch": 4.16374269005848,
      "grad_norm": 0.9375,
      "learning_rate": 0.0003837722757040289,
      "loss": 5.88,
      "mean_token_accuracy": 0.17474150508642197,
      "num_tokens": 8868220.0,
      "step": 3560
    },
    {
      "entropy": 6.311040830612183,
      "epoch": 4.169590643274854,
      "grad_norm": 1.125,
      "learning_rate": 0.0003833617144412961,
      "loss": 5.8112,
      "mean_token_accuracy": 0.18350528180599213,
      "num_tokens": 8879106.0,
      "step": 3565
    },
    {
      "entropy": 6.356731986999511,
      "epoch": 4.175438596491228,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0003829506828809521,
      "loss": 5.8927,
      "mean_token_accuracy": 0.18050391376018524,
      "num_tokens": 8891358.0,
      "step": 3570
    },
    {
      "entropy": 6.405393075942993,
      "epoch": 4.181286549707602,
      "grad_norm": 1.0,
      "learning_rate": 0.0003825391828069032,
      "loss": 5.8384,
      "mean_token_accuracy": 0.18048698604106903,
      "num_tokens": 8904327.0,
      "step": 3575
    },
    {
      "entropy": 6.333025550842285,
      "epoch": 4.187134502923977,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003821272160050894,
      "loss": 5.7861,
      "mean_token_accuracy": 0.18152665644884108,
      "num_tokens": 8916515.0,
      "step": 3580
    },
    {
      "entropy": 6.332434368133545,
      "epoch": 4.192982456140351,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0003817147842634762,
      "loss": 5.8612,
      "mean_token_accuracy": 0.17733515053987503,
      "num_tokens": 8929482.0,
      "step": 3585
    },
    {
      "entropy": 6.432557201385498,
      "epoch": 4.1988304093567255,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0003813018893720468,
      "loss": 5.9155,
      "mean_token_accuracy": 0.1704854890704155,
      "num_tokens": 8942328.0,
      "step": 3590
    },
    {
      "entropy": 6.37638635635376,
      "epoch": 4.204678362573099,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038088853312279473,
      "loss": 5.8357,
      "mean_token_accuracy": 0.18546354174613952,
      "num_tokens": 8955116.0,
      "step": 3595
    },
    {
      "entropy": 6.36842098236084,
      "epoch": 4.2105263157894735,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003804747173097157,
      "loss": 5.9085,
      "mean_token_accuracy": 0.17164976596832277,
      "num_tokens": 8967459.0,
      "step": 3600
    },
    {
      "entropy": 6.3011571884155275,
      "epoch": 4.216374269005848,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00038006044372880006,
      "loss": 5.8105,
      "mean_token_accuracy": 0.18346855491399766,
      "num_tokens": 8980079.0,
      "step": 3605
    },
    {
      "entropy": 6.385735273361206,
      "epoch": 4.222222222222222,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037964571417802473,
      "loss": 5.8605,
      "mean_token_accuracy": 0.18380159884691238,
      "num_tokens": 8992764.0,
      "step": 3610
    },
    {
      "entropy": 6.314878702163696,
      "epoch": 4.228070175438597,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00037923053045734595,
      "loss": 5.7482,
      "mean_token_accuracy": 0.18109725415706635,
      "num_tokens": 9004904.0,
      "step": 3615
    },
    {
      "entropy": 6.446242046356201,
      "epoch": 4.23391812865497,
      "grad_norm": 0.9375,
      "learning_rate": 0.00037881489436869054,
      "loss": 5.9639,
      "mean_token_accuracy": 0.1697452813386917,
      "num_tokens": 9018411.0,
      "step": 3620
    },
    {
      "entropy": 6.364334154129028,
      "epoch": 4.239766081871345,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003783988077159491,
      "loss": 5.7775,
      "mean_token_accuracy": 0.1888364925980568,
      "num_tokens": 9030593.0,
      "step": 3625
    },
    {
      "entropy": 6.244638919830322,
      "epoch": 4.245614035087719,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037798227230496754,
      "loss": 5.784,
      "mean_token_accuracy": 0.19220828115940095,
      "num_tokens": 9043715.0,
      "step": 3630
    },
    {
      "entropy": 6.249920415878296,
      "epoch": 4.251461988304094,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00037756528994353945,
      "loss": 5.7495,
      "mean_token_accuracy": 0.19220608919858934,
      "num_tokens": 9056384.0,
      "step": 3635
    },
    {
      "entropy": 6.321842241287231,
      "epoch": 4.257309941520468,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000377147862441398,
      "loss": 5.8483,
      "mean_token_accuracy": 0.1788550853729248,
      "num_tokens": 9069444.0,
      "step": 3640
    },
    {
      "entropy": 6.343995809555054,
      "epoch": 4.2631578947368425,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037672999161020873,
      "loss": 5.7927,
      "mean_token_accuracy": 0.1840248689055443,
      "num_tokens": 9080933.0,
      "step": 3645
    },
    {
      "entropy": 6.340948390960693,
      "epoch": 4.269005847953216,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00037631167926356087,
      "loss": 5.8363,
      "mean_token_accuracy": 0.1850079908967018,
      "num_tokens": 9093405.0,
      "step": 3650
    },
    {
      "entropy": 6.297127151489258,
      "epoch": 4.2748538011695905,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003758929272169599,
      "loss": 5.822,
      "mean_token_accuracy": 0.1892406687140465,
      "num_tokens": 9105707.0,
      "step": 3655
    },
    {
      "entropy": 6.3728618144989015,
      "epoch": 4.280701754385965,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00037547373728781996,
      "loss": 5.7988,
      "mean_token_accuracy": 0.18101050555706025,
      "num_tokens": 9119476.0,
      "step": 3660
    },
    {
      "entropy": 6.296815586090088,
      "epoch": 4.286549707602339,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003750541112954552,
      "loss": 5.7751,
      "mean_token_accuracy": 0.1832916870713234,
      "num_tokens": 9131788.0,
      "step": 3665
    },
    {
      "entropy": 6.358005046844482,
      "epoch": 4.292397660818714,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003746340510610727,
      "loss": 5.8688,
      "mean_token_accuracy": 0.17935132831335068,
      "num_tokens": 9144340.0,
      "step": 3670
    },
    {
      "entropy": 6.359562969207763,
      "epoch": 4.298245614035087,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00037421355840776396,
      "loss": 5.8304,
      "mean_token_accuracy": 0.184980046749115,
      "num_tokens": 9156757.0,
      "step": 3675
    },
    {
      "entropy": 6.263106727600098,
      "epoch": 4.304093567251462,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003737926351604971,
      "loss": 5.8014,
      "mean_token_accuracy": 0.1849026069045067,
      "num_tokens": 9169975.0,
      "step": 3680
    },
    {
      "entropy": 6.359688568115234,
      "epoch": 4.309941520467836,
      "grad_norm": 1.0,
      "learning_rate": 0.00037337128314610926,
      "loss": 5.8336,
      "mean_token_accuracy": 0.18788386583328248,
      "num_tokens": 9181891.0,
      "step": 3685
    },
    {
      "entropy": 6.359238338470459,
      "epoch": 4.315789473684211,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003729495041932984,
      "loss": 5.849,
      "mean_token_accuracy": 0.1752017319202423,
      "num_tokens": 9194068.0,
      "step": 3690
    },
    {
      "entropy": 6.3033788204193115,
      "epoch": 4.321637426900585,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003725273001326155,
      "loss": 5.8812,
      "mean_token_accuracy": 0.17873608469963073,
      "num_tokens": 9205530.0,
      "step": 3695
    },
    {
      "entropy": 6.33714747428894,
      "epoch": 4.3274853801169595,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0003721046727964565,
      "loss": 5.8073,
      "mean_token_accuracy": 0.1865734726190567,
      "num_tokens": 9218444.0,
      "step": 3700
    },
    {
      "entropy": 6.330530595779419,
      "epoch": 4.333333333333333,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037168162401905416,
      "loss": 5.8971,
      "mean_token_accuracy": 0.18307720124721527,
      "num_tokens": 9229507.0,
      "step": 3705
    },
    {
      "entropy": 6.394552183151245,
      "epoch": 4.3391812865497075,
      "grad_norm": 1.0,
      "learning_rate": 0.00037125815563647063,
      "loss": 5.9482,
      "mean_token_accuracy": 0.17586733549833297,
      "num_tokens": 9242128.0,
      "step": 3710
    },
    {
      "entropy": 6.343462753295898,
      "epoch": 4.345029239766082,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00037083426948658903,
      "loss": 5.8933,
      "mean_token_accuracy": 0.17846456468105315,
      "num_tokens": 9255598.0,
      "step": 3715
    },
    {
      "entropy": 6.327785062789917,
      "epoch": 4.350877192982456,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037040996740910566,
      "loss": 5.8448,
      "mean_token_accuracy": 0.18709179162979125,
      "num_tokens": 9267568.0,
      "step": 3720
    },
    {
      "entropy": 6.3736433506011965,
      "epoch": 4.356725146198831,
      "grad_norm": 0.90625,
      "learning_rate": 0.000369985251245522,
      "loss": 5.8634,
      "mean_token_accuracy": 0.18130033016204833,
      "num_tokens": 9280223.0,
      "step": 3725
    },
    {
      "entropy": 6.317340278625489,
      "epoch": 4.362573099415204,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00036956012283913655,
      "loss": 5.9097,
      "mean_token_accuracy": 0.18134237676858903,
      "num_tokens": 9292933.0,
      "step": 3730
    },
    {
      "entropy": 6.331889581680298,
      "epoch": 4.368421052631579,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00036913458403503715,
      "loss": 5.8642,
      "mean_token_accuracy": 0.18651467710733413,
      "num_tokens": 9306739.0,
      "step": 3735
    },
    {
      "entropy": 6.375274562835694,
      "epoch": 4.374269005847953,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003687086366800927,
      "loss": 5.836,
      "mean_token_accuracy": 0.18045736104249954,
      "num_tokens": 9319151.0,
      "step": 3740
    },
    {
      "entropy": 6.347825574874878,
      "epoch": 4.380116959064328,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003682822826229453,
      "loss": 5.8978,
      "mean_token_accuracy": 0.17697394788265228,
      "num_tokens": 9332035.0,
      "step": 3745
    },
    {
      "entropy": 6.3848692893981935,
      "epoch": 4.385964912280702,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036785552371400206,
      "loss": 5.9302,
      "mean_token_accuracy": 0.17443449050188065,
      "num_tokens": 9345248.0,
      "step": 3750
    },
    {
      "entropy": 6.335942316055298,
      "epoch": 4.391812865497076,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003674283618054273,
      "loss": 5.8873,
      "mean_token_accuracy": 0.1811821386218071,
      "num_tokens": 9357514.0,
      "step": 3755
    },
    {
      "entropy": 6.384964895248413,
      "epoch": 4.39766081871345,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003670007987511344,
      "loss": 5.9199,
      "mean_token_accuracy": 0.17602879405021668,
      "num_tokens": 9368627.0,
      "step": 3760
    },
    {
      "entropy": 6.293928050994873,
      "epoch": 4.4035087719298245,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0003665728364067776,
      "loss": 5.8353,
      "mean_token_accuracy": 0.18417404145002364,
      "num_tokens": 9380591.0,
      "step": 3765
    },
    {
      "entropy": 6.2800568580627445,
      "epoch": 4.409356725146199,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00036614447662974413,
      "loss": 5.8057,
      "mean_token_accuracy": 0.18520332425832747,
      "num_tokens": 9392873.0,
      "step": 3770
    },
    {
      "entropy": 6.443205976486206,
      "epoch": 4.415204678362573,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003657157212791462,
      "loss": 5.9373,
      "mean_token_accuracy": 0.17334980070590972,
      "num_tokens": 9404252.0,
      "step": 3775
    },
    {
      "entropy": 6.270028972625733,
      "epoch": 4.421052631578947,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003652865722158128,
      "loss": 5.8649,
      "mean_token_accuracy": 0.18557138592004777,
      "num_tokens": 9416091.0,
      "step": 3780
    },
    {
      "entropy": 6.335468578338623,
      "epoch": 4.426900584795321,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036485703130228147,
      "loss": 5.8019,
      "mean_token_accuracy": 0.187326879799366,
      "num_tokens": 9429913.0,
      "step": 3785
    },
    {
      "entropy": 6.346346664428711,
      "epoch": 4.432748538011696,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036442710040279086,
      "loss": 5.7844,
      "mean_token_accuracy": 0.1908934459090233,
      "num_tokens": 9442413.0,
      "step": 3790
    },
    {
      "entropy": 6.143695545196533,
      "epoch": 4.43859649122807,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003639967813832717,
      "loss": 5.732,
      "mean_token_accuracy": 0.19100991040468215,
      "num_tokens": 9453636.0,
      "step": 3795
    },
    {
      "entropy": 6.330870294570923,
      "epoch": 4.444444444444445,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003635660761113394,
      "loss": 5.8177,
      "mean_token_accuracy": 0.18606052845716475,
      "num_tokens": 9466577.0,
      "step": 3800
    },
    {
      "entropy": 6.288309717178345,
      "epoch": 4.450292397660819,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0003631349864562857,
      "loss": 5.8314,
      "mean_token_accuracy": 0.17907752394676207,
      "num_tokens": 9479852.0,
      "step": 3805
    },
    {
      "entropy": 6.326959609985352,
      "epoch": 4.456140350877193,
      "grad_norm": 0.9375,
      "learning_rate": 0.0003627035142890708,
      "loss": 5.7758,
      "mean_token_accuracy": 0.18165678828954696,
      "num_tokens": 9493373.0,
      "step": 3810
    },
    {
      "entropy": 6.310887432098388,
      "epoch": 4.461988304093567,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003622716614823147,
      "loss": 5.8777,
      "mean_token_accuracy": 0.18951957374811174,
      "num_tokens": 9505793.0,
      "step": 3815
    },
    {
      "entropy": 6.298397541046143,
      "epoch": 4.4678362573099415,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003618394299102896,
      "loss": 5.8937,
      "mean_token_accuracy": 0.17274616211652755,
      "num_tokens": 9517810.0,
      "step": 3820
    },
    {
      "entropy": 6.348339557647705,
      "epoch": 4.473684210526316,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036140682144891143,
      "loss": 5.867,
      "mean_token_accuracy": 0.1746555596590042,
      "num_tokens": 9529430.0,
      "step": 3825
    },
    {
      "entropy": 6.341324710845948,
      "epoch": 4.47953216374269,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00036097383797573205,
      "loss": 5.8526,
      "mean_token_accuracy": 0.18363075852394103,
      "num_tokens": 9542385.0,
      "step": 3830
    },
    {
      "entropy": 6.324956035614013,
      "epoch": 4.485380116959064,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036054048136993075,
      "loss": 5.8264,
      "mean_token_accuracy": 0.1813960924744606,
      "num_tokens": 9553955.0,
      "step": 3835
    },
    {
      "entropy": 6.401408958435058,
      "epoch": 4.491228070175438,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036010675351230616,
      "loss": 5.881,
      "mean_token_accuracy": 0.17374949306249618,
      "num_tokens": 9566494.0,
      "step": 3840
    },
    {
      "entropy": 6.2555054187774655,
      "epoch": 4.497076023391813,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003596726562852684,
      "loss": 5.7977,
      "mean_token_accuracy": 0.19336040914058686,
      "num_tokens": 9579122.0,
      "step": 3845
    },
    {
      "entropy": 6.303387212753296,
      "epoch": 4.502923976608187,
      "grad_norm": 0.96875,
      "learning_rate": 0.00035923819157283074,
      "loss": 5.8165,
      "mean_token_accuracy": 0.1852250188589096,
      "num_tokens": 9592144.0,
      "step": 3850
    },
    {
      "entropy": 6.2353424549102785,
      "epoch": 4.508771929824562,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035880336126060074,
      "loss": 5.7102,
      "mean_token_accuracy": 0.18490662425756454,
      "num_tokens": 9604533.0,
      "step": 3855
    },
    {
      "entropy": 6.324378490447998,
      "epoch": 4.514619883040936,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035836816723577354,
      "loss": 5.8863,
      "mean_token_accuracy": 0.18342247605323792,
      "num_tokens": 9617349.0,
      "step": 3860
    },
    {
      "entropy": 6.334613370895386,
      "epoch": 4.52046783625731,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003579326113871222,
      "loss": 5.8618,
      "mean_token_accuracy": 0.1807621881365776,
      "num_tokens": 9629599.0,
      "step": 3865
    },
    {
      "entropy": 6.274072217941284,
      "epoch": 4.526315789473684,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003574966956049904,
      "loss": 5.8305,
      "mean_token_accuracy": 0.1855103477835655,
      "num_tokens": 9642041.0,
      "step": 3870
    },
    {
      "entropy": 6.296230840682983,
      "epoch": 4.5321637426900585,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035706042178128397,
      "loss": 5.8539,
      "mean_token_accuracy": 0.18220545649528502,
      "num_tokens": 9654796.0,
      "step": 3875
    },
    {
      "entropy": 6.326847219467163,
      "epoch": 4.538011695906433,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003566237918094625,
      "loss": 5.8772,
      "mean_token_accuracy": 0.18168586641550064,
      "num_tokens": 9667211.0,
      "step": 3880
    },
    {
      "entropy": 6.286065149307251,
      "epoch": 4.543859649122807,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003561868075845313,
      "loss": 5.7964,
      "mean_token_accuracy": 0.18665794283151627,
      "num_tokens": 9679526.0,
      "step": 3885
    },
    {
      "entropy": 6.367777252197266,
      "epoch": 4.549707602339181,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035574947100303327,
      "loss": 5.9847,
      "mean_token_accuracy": 0.17264424562454223,
      "num_tokens": 9691760.0,
      "step": 3890
    },
    {
      "entropy": 6.3543720722198485,
      "epoch": 4.555555555555555,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035531178396304083,
      "loss": 5.8895,
      "mean_token_accuracy": 0.18378105163574218,
      "num_tokens": 9704154.0,
      "step": 3895
    },
    {
      "entropy": 6.242607641220093,
      "epoch": 4.56140350877193,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035487374836414676,
      "loss": 5.7004,
      "mean_token_accuracy": 0.1956769496202469,
      "num_tokens": 9715850.0,
      "step": 3900
    },
    {
      "entropy": 6.332263517379761,
      "epoch": 4.567251461988304,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035443536610745707,
      "loss": 5.8518,
      "mean_token_accuracy": 0.18083570003509522,
      "num_tokens": 9728202.0,
      "step": 3905
    },
    {
      "entropy": 6.397715377807617,
      "epoch": 4.573099415204679,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00035399663909558226,
      "loss": 5.8672,
      "mean_token_accuracy": 0.17605464458465575,
      "num_tokens": 9741363.0,
      "step": 3910
    },
    {
      "entropy": 6.295465326309204,
      "epoch": 4.578947368421053,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00035355756923262897,
      "loss": 5.8557,
      "mean_token_accuracy": 0.17956543564796448,
      "num_tokens": 9755231.0,
      "step": 3915
    },
    {
      "entropy": 6.272282218933105,
      "epoch": 4.584795321637427,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003531181584241919,
      "loss": 5.699,
      "mean_token_accuracy": 0.1962415874004364,
      "num_tokens": 9768972.0,
      "step": 3920
    },
    {
      "entropy": 6.347984743118286,
      "epoch": 4.590643274853801,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003526784085773455,
      "loss": 5.9018,
      "mean_token_accuracy": 0.1785668686032295,
      "num_tokens": 9781356.0,
      "step": 3925
    },
    {
      "entropy": 6.313095331192017,
      "epoch": 4.5964912280701755,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003522383216006356,
      "loss": 5.8556,
      "mean_token_accuracy": 0.17589664608240127,
      "num_tokens": 9793227.0,
      "step": 3930
    },
    {
      "entropy": 6.297039127349853,
      "epoch": 4.60233918128655,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003517978994040713,
      "loss": 5.8481,
      "mean_token_accuracy": 0.18179278075695038,
      "num_tokens": 9805968.0,
      "step": 3935
    },
    {
      "entropy": 6.335315704345703,
      "epoch": 4.6081871345029235,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035135714389911637,
      "loss": 5.8362,
      "mean_token_accuracy": 0.17830058336257934,
      "num_tokens": 9817864.0,
      "step": 3940
    },
    {
      "entropy": 6.254665088653565,
      "epoch": 4.614035087719298,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003509160569986814,
      "loss": 5.8185,
      "mean_token_accuracy": 0.18883679658174515,
      "num_tokens": 9831192.0,
      "step": 3945
    },
    {
      "entropy": 6.332556056976318,
      "epoch": 4.619883040935672,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035047464061711527,
      "loss": 5.8362,
      "mean_token_accuracy": 0.17580619156360627,
      "num_tokens": 9842813.0,
      "step": 3950
    },
    {
      "entropy": 6.327417659759521,
      "epoch": 4.625730994152047,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035003289667019655,
      "loss": 5.8686,
      "mean_token_accuracy": 0.18022319823503494,
      "num_tokens": 9854500.0,
      "step": 3955
    },
    {
      "entropy": 6.300075960159302,
      "epoch": 4.631578947368421,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003495908270751257,
      "loss": 5.8389,
      "mean_token_accuracy": 0.18082908242940904,
      "num_tokens": 9865729.0,
      "step": 3960
    },
    {
      "entropy": 6.352865219116211,
      "epoch": 4.637426900584796,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003491484337505166,
      "loss": 5.8795,
      "mean_token_accuracy": 0.1775056838989258,
      "num_tokens": 9878306.0,
      "step": 3965
    },
    {
      "entropy": 6.295338106155396,
      "epoch": 4.643274853801169,
      "grad_norm": 0.96875,
      "learning_rate": 0.00034870571861638785,
      "loss": 5.898,
      "mean_token_accuracy": 0.18338036984205247,
      "num_tokens": 9890384.0,
      "step": 3970
    },
    {
      "entropy": 6.323778963088989,
      "epoch": 4.649122807017544,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00034826268359415487,
      "loss": 5.853,
      "mean_token_accuracy": 0.180763578414917,
      "num_tokens": 9902960.0,
      "step": 3975
    },
    {
      "entropy": 6.306023073196411,
      "epoch": 4.654970760233918,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003478193306066215,
      "loss": 5.814,
      "mean_token_accuracy": 0.18519977033138274,
      "num_tokens": 9915143.0,
      "step": 3980
    },
    {
      "entropy": 6.318506240844727,
      "epoch": 4.6608187134502925,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00034737566157797147,
      "loss": 5.9222,
      "mean_token_accuracy": 0.17391184121370315,
      "num_tokens": 9928440.0,
      "step": 3985
    },
    {
      "entropy": 6.36572494506836,
      "epoch": 4.666666666666667,
      "grad_norm": 1.0,
      "learning_rate": 0.00034693167843376016,
      "loss": 5.9352,
      "mean_token_accuracy": 0.1749822050333023,
      "num_tokens": 9940973.0,
      "step": 3990
    },
    {
      "entropy": 6.285900688171386,
      "epoch": 4.6725146198830405,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034648738310090634,
      "loss": 5.8322,
      "mean_token_accuracy": 0.18492650091648102,
      "num_tokens": 9953433.0,
      "step": 3995
    },
    {
      "entropy": 6.21313591003418,
      "epoch": 4.678362573099415,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003460427775076835,
      "loss": 5.7618,
      "mean_token_accuracy": 0.19115083366632463,
      "num_tokens": 9964641.0,
      "step": 4000
    },
    {
      "epoch": 4.678362573099415,
      "eval_entropy": 6.1529634849863415,
      "eval_loss": 6.561431407928467,
      "eval_mean_token_accuracy": 0.15061089429993368,
      "eval_num_tokens": 9964641.0,
      "eval_runtime": 1.7348,
      "eval_samples_per_second": 1560.394,
      "eval_steps_per_second": 195.409,
      "step": 4000
    },
    {
      "entropy": 6.335078954696655,
      "epoch": 4.684210526315789,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00034559786358371187,
      "loss": 5.9449,
      "mean_token_accuracy": 0.17469603419303895,
      "num_tokens": 9978625.0,
      "step": 4005
    },
    {
      "entropy": 6.343950510025024,
      "epoch": 4.690058479532164,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034515264325994975,
      "loss": 5.7975,
      "mean_token_accuracy": 0.18587568253278733,
      "num_tokens": 9991758.0,
      "step": 4010
    },
    {
      "entropy": 6.394791030883789,
      "epoch": 4.695906432748538,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00034470711846868525,
      "loss": 5.9058,
      "mean_token_accuracy": 0.17634283900260925,
      "num_tokens": 10005496.0,
      "step": 4015
    },
    {
      "entropy": 6.2901476383209225,
      "epoch": 4.701754385964913,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034426129114352807,
      "loss": 5.9331,
      "mean_token_accuracy": 0.180510713160038,
      "num_tokens": 10019025.0,
      "step": 4020
    },
    {
      "entropy": 6.2301496982574465,
      "epoch": 4.707602339181286,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003438151632194006,
      "loss": 5.7966,
      "mean_token_accuracy": 0.1861666902899742,
      "num_tokens": 10030741.0,
      "step": 4025
    },
    {
      "entropy": 6.347356271743775,
      "epoch": 4.713450292397661,
      "grad_norm": 1.125,
      "learning_rate": 0.00034336873663252997,
      "loss": 5.8912,
      "mean_token_accuracy": 0.18361773639917373,
      "num_tokens": 10043138.0,
      "step": 4030
    },
    {
      "entropy": 6.297961044311523,
      "epoch": 4.719298245614035,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034292201332043976,
      "loss": 5.8686,
      "mean_token_accuracy": 0.17736987471580506,
      "num_tokens": 10055927.0,
      "step": 4035
    },
    {
      "entropy": 6.271793079376221,
      "epoch": 4.7251461988304095,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003424749952219411,
      "loss": 5.775,
      "mean_token_accuracy": 0.18566307574510574,
      "num_tokens": 10067392.0,
      "step": 4040
    },
    {
      "entropy": 6.317879581451416,
      "epoch": 4.730994152046784,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034202768427712456,
      "loss": 5.8347,
      "mean_token_accuracy": 0.18416796624660492,
      "num_tokens": 10079245.0,
      "step": 4045
    },
    {
      "entropy": 6.2470396041870115,
      "epoch": 4.7368421052631575,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003415800824273518,
      "loss": 5.7541,
      "mean_token_accuracy": 0.18824544101953505,
      "num_tokens": 10091923.0,
      "step": 4050
    },
    {
      "entropy": 6.290208053588867,
      "epoch": 4.742690058479532,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003411321916152469,
      "loss": 5.8311,
      "mean_token_accuracy": 0.18291647881269454,
      "num_tokens": 10105121.0,
      "step": 4055
    },
    {
      "entropy": 6.267402076721192,
      "epoch": 4.748538011695906,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003406840137846881,
      "loss": 5.8876,
      "mean_token_accuracy": 0.18025690764188768,
      "num_tokens": 10117005.0,
      "step": 4060
    },
    {
      "entropy": 6.278704309463501,
      "epoch": 4.754385964912281,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034023555088079924,
      "loss": 5.8116,
      "mean_token_accuracy": 0.1878137931227684,
      "num_tokens": 10129958.0,
      "step": 4065
    },
    {
      "entropy": 6.294871139526367,
      "epoch": 4.760233918128655,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003397868048499416,
      "loss": 5.8426,
      "mean_token_accuracy": 0.180222487449646,
      "num_tokens": 10141537.0,
      "step": 4070
    },
    {
      "entropy": 6.274731922149658,
      "epoch": 4.76608187134503,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033933777763970506,
      "loss": 5.8841,
      "mean_token_accuracy": 0.18863564133644103,
      "num_tokens": 10153882.0,
      "step": 4075
    },
    {
      "entropy": 6.350859117507935,
      "epoch": 4.771929824561403,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003388884711989,
      "loss": 5.867,
      "mean_token_accuracy": 0.17912289798259734,
      "num_tokens": 10165147.0,
      "step": 4080
    },
    {
      "entropy": 6.291949796676636,
      "epoch": 4.777777777777778,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003384388874775486,
      "loss": 5.8992,
      "mean_token_accuracy": 0.17191877216100693,
      "num_tokens": 10177926.0,
      "step": 4085
    },
    {
      "entropy": 6.300592613220215,
      "epoch": 4.783625730994152,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033798902842687627,
      "loss": 5.8388,
      "mean_token_accuracy": 0.18281185925006865,
      "num_tokens": 10190343.0,
      "step": 4090
    },
    {
      "entropy": 6.367020988464356,
      "epoch": 4.7894736842105265,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003375388959993037,
      "loss": 5.9405,
      "mean_token_accuracy": 0.17780277281999587,
      "num_tokens": 10201363.0,
      "step": 4095
    },
    {
      "entropy": 6.1826681137084964,
      "epoch": 4.7953216374269,
      "grad_norm": 0.90625,
      "learning_rate": 0.0003370884921484379,
      "loss": 5.7862,
      "mean_token_accuracy": 0.189967143535614,
      "num_tokens": 10213976.0,
      "step": 4100
    },
    {
      "entropy": 6.296461772918701,
      "epoch": 4.8011695906432745,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003366378188290638,
      "loss": 5.84,
      "mean_token_accuracy": 0.18531502336263656,
      "num_tokens": 10225645.0,
      "step": 4105
    },
    {
      "entropy": 6.35559663772583,
      "epoch": 4.807017543859649,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003361868779971361,
      "loss": 5.8567,
      "mean_token_accuracy": 0.18640678524971008,
      "num_tokens": 10237144.0,
      "step": 4110
    },
    {
      "entropy": 6.384145450592041,
      "epoch": 4.812865497076023,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033573567160977017,
      "loss": 6.0178,
      "mean_token_accuracy": 0.1738305300474167,
      "num_tokens": 10249215.0,
      "step": 4115
    },
    {
      "entropy": 6.277143383026123,
      "epoch": 4.818713450292398,
      "grad_norm": 0.90625,
      "learning_rate": 0.000335284201625234,
      "loss": 5.82,
      "mean_token_accuracy": 0.17938347458839415,
      "num_tokens": 10262428.0,
      "step": 4120
    },
    {
      "entropy": 6.247993087768554,
      "epoch": 4.824561403508772,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033483247000294,
      "loss": 5.86,
      "mean_token_accuracy": 0.1837025299668312,
      "num_tokens": 10275370.0,
      "step": 4125
    },
    {
      "entropy": 6.322489833831787,
      "epoch": 4.830409356725146,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0003343804787034357,
      "loss": 5.8591,
      "mean_token_accuracy": 0.1857644498348236,
      "num_tokens": 10287378.0,
      "step": 4130
    },
    {
      "entropy": 6.328525829315185,
      "epoch": 4.83625730994152,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003339282296883956,
      "loss": 5.8568,
      "mean_token_accuracy": 0.17452150136232375,
      "num_tokens": 10300063.0,
      "step": 4135
    },
    {
      "entropy": 6.3399862289428714,
      "epoch": 4.842105263157895,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000333475724920613,
      "loss": 5.8366,
      "mean_token_accuracy": 0.1851540133357048,
      "num_tokens": 10312095.0,
      "step": 4140
    },
    {
      "entropy": 6.26895432472229,
      "epoch": 4.847953216374269,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003330229663639909,
      "loss": 5.878,
      "mean_token_accuracy": 0.18046684563159943,
      "num_tokens": 10324903.0,
      "step": 4145
    },
    {
      "entropy": 6.423642921447754,
      "epoch": 4.853801169590644,
      "grad_norm": 0.875,
      "learning_rate": 0.000332569955983534,
      "loss": 5.908,
      "mean_token_accuracy": 0.17982545047998427,
      "num_tokens": 10339253.0,
      "step": 4150
    },
    {
      "entropy": 6.343758201599121,
      "epoch": 4.859649122807017,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003321166957453398,
      "loss": 5.9113,
      "mean_token_accuracy": 0.17312162220478058,
      "num_tokens": 10350845.0,
      "step": 4155
    },
    {
      "entropy": 6.278868293762207,
      "epoch": 4.8654970760233915,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003316631876165904,
      "loss": 5.83,
      "mean_token_accuracy": 0.18230910450220109,
      "num_tokens": 10363357.0,
      "step": 4160
    },
    {
      "entropy": 6.218644857406616,
      "epoch": 4.871345029239766,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033120943356554334,
      "loss": 5.7657,
      "mean_token_accuracy": 0.18386161029338838,
      "num_tokens": 10375664.0,
      "step": 4165
    },
    {
      "entropy": 6.320641946792603,
      "epoch": 4.87719298245614,
      "grad_norm": 1.0625,
      "learning_rate": 0.000330755435561524,
      "loss": 5.8243,
      "mean_token_accuracy": 0.18520996272563933,
      "num_tokens": 10387412.0,
      "step": 4170
    },
    {
      "entropy": 6.239684963226319,
      "epoch": 4.883040935672515,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033030119557491604,
      "loss": 5.8747,
      "mean_token_accuracy": 0.18516825437545775,
      "num_tokens": 10399083.0,
      "step": 4175
    },
    {
      "entropy": 6.433531999588013,
      "epoch": 4.888888888888889,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003298467155771537,
      "loss": 5.9233,
      "mean_token_accuracy": 0.17103088796138763,
      "num_tokens": 10412713.0,
      "step": 4180
    },
    {
      "entropy": 6.225171613693237,
      "epoch": 4.894736842105263,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00032939199754071274,
      "loss": 5.7804,
      "mean_token_accuracy": 0.18590821474790573,
      "num_tokens": 10426836.0,
      "step": 4185
    },
    {
      "entropy": 6.224940204620362,
      "epoch": 4.900584795321637,
      "grad_norm": 1.0625,
      "learning_rate": 0.000328937043439102,
      "loss": 5.8487,
      "mean_token_accuracy": 0.183578784763813,
      "num_tokens": 10439817.0,
      "step": 4190
    },
    {
      "entropy": 6.317575359344483,
      "epoch": 4.906432748538012,
      "grad_norm": 0.9375,
      "learning_rate": 0.00032848185524685504,
      "loss": 5.8486,
      "mean_token_accuracy": 0.1808657616376877,
      "num_tokens": 10452509.0,
      "step": 4195
    },
    {
      "entropy": 6.356023597717285,
      "epoch": 4.912280701754386,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00032802643493952106,
      "loss": 5.885,
      "mean_token_accuracy": 0.17912613302469255,
      "num_tokens": 10464593.0,
      "step": 4200
    },
    {
      "entropy": 6.272734117507935,
      "epoch": 4.918128654970761,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000327570784493657,
      "loss": 5.9119,
      "mean_token_accuracy": 0.17638685256242753,
      "num_tokens": 10476868.0,
      "step": 4205
    },
    {
      "entropy": 6.295363521575927,
      "epoch": 4.923976608187134,
      "grad_norm": 0.9375,
      "learning_rate": 0.00032711490588681845,
      "loss": 5.752,
      "mean_token_accuracy": 0.18438016027212142,
      "num_tokens": 10489510.0,
      "step": 4210
    },
    {
      "entropy": 6.327757406234741,
      "epoch": 4.9298245614035086,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003266588010975512,
      "loss": 5.8959,
      "mean_token_accuracy": 0.17960805743932723,
      "num_tokens": 10501901.0,
      "step": 4215
    },
    {
      "entropy": 6.221873140335083,
      "epoch": 4.935672514619883,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003262024721053827,
      "loss": 5.8101,
      "mean_token_accuracy": 0.18644112348556519,
      "num_tokens": 10513604.0,
      "step": 4220
    },
    {
      "entropy": 6.256662082672119,
      "epoch": 4.941520467836257,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00032574592089081375,
      "loss": 5.8288,
      "mean_token_accuracy": 0.18700238019227983,
      "num_tokens": 10527882.0,
      "step": 4225
    },
    {
      "entropy": 6.350762271881104,
      "epoch": 4.947368421052632,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032528914943530905,
      "loss": 5.7872,
      "mean_token_accuracy": 0.19575264006853105,
      "num_tokens": 10539514.0,
      "step": 4230
    },
    {
      "entropy": 6.28397388458252,
      "epoch": 4.953216374269006,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003248321597212896,
      "loss": 5.8466,
      "mean_token_accuracy": 0.18198691606521605,
      "num_tokens": 10551984.0,
      "step": 4235
    },
    {
      "entropy": 6.369769763946533,
      "epoch": 4.95906432748538,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003243749537321236,
      "loss": 5.9254,
      "mean_token_accuracy": 0.17405585050582886,
      "num_tokens": 10564915.0,
      "step": 4240
    },
    {
      "entropy": 6.275656127929688,
      "epoch": 4.964912280701754,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032391753345211767,
      "loss": 5.8755,
      "mean_token_accuracy": 0.184158356487751,
      "num_tokens": 10576614.0,
      "step": 4245
    },
    {
      "entropy": 6.308443403244018,
      "epoch": 4.970760233918129,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003234599008665088,
      "loss": 5.857,
      "mean_token_accuracy": 0.18483080416917802,
      "num_tokens": 10588746.0,
      "step": 4250
    },
    {
      "entropy": 6.317317008972168,
      "epoch": 4.976608187134503,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003230020579614551,
      "loss": 5.8735,
      "mean_token_accuracy": 0.17638473212718964,
      "num_tokens": 10600101.0,
      "step": 4255
    },
    {
      "entropy": 6.296493148803711,
      "epoch": 4.982456140350877,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003225440067240276,
      "loss": 5.8344,
      "mean_token_accuracy": 0.1818413108587265,
      "num_tokens": 10611879.0,
      "step": 4260
    },
    {
      "entropy": 6.316436338424682,
      "epoch": 4.988304093567251,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032208574914220167,
      "loss": 5.8936,
      "mean_token_accuracy": 0.1840697020292282,
      "num_tokens": 10624201.0,
      "step": 4265
    },
    {
      "entropy": 6.300460338592529,
      "epoch": 4.994152046783626,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003216272872048479,
      "loss": 5.8462,
      "mean_token_accuracy": 0.1799156203866005,
      "num_tokens": 10636612.0,
      "step": 4270
    },
    {
      "entropy": 6.325052070617676,
      "epoch": 5.0,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003211686229017241,
      "loss": 5.8392,
      "mean_token_accuracy": 0.18313610255718232,
      "num_tokens": 10647420.0,
      "step": 4275
    },
    {
      "entropy": 6.301950311660766,
      "epoch": 5.005847953216374,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003207097582234662,
      "loss": 5.6832,
      "mean_token_accuracy": 0.18609594106674193,
      "num_tokens": 10660096.0,
      "step": 4280
    },
    {
      "entropy": 6.276100063323975,
      "epoch": 5.011695906432749,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003202506951615798,
      "loss": 5.6204,
      "mean_token_accuracy": 0.1949935957789421,
      "num_tokens": 10672364.0,
      "step": 4285
    },
    {
      "entropy": 6.289952516555786,
      "epoch": 5.017543859649122,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031979143570843145,
      "loss": 5.6091,
      "mean_token_accuracy": 0.19282592684030533,
      "num_tokens": 10685230.0,
      "step": 4290
    },
    {
      "entropy": 6.386991739273071,
      "epoch": 5.023391812865497,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00031933198185724016,
      "loss": 5.7304,
      "mean_token_accuracy": 0.18733056485652924,
      "num_tokens": 10697837.0,
      "step": 4295
    },
    {
      "entropy": 6.2685212135314945,
      "epoch": 5.029239766081871,
      "grad_norm": 1.015625,
      "learning_rate": 0.00031887233560206864,
      "loss": 5.5262,
      "mean_token_accuracy": 0.20623231083154678,
      "num_tokens": 10711549.0,
      "step": 4300
    },
    {
      "entropy": 6.3433928966522215,
      "epoch": 5.035087719298246,
      "grad_norm": 1.0,
      "learning_rate": 0.00031841249893781447,
      "loss": 5.7102,
      "mean_token_accuracy": 0.1871415212750435,
      "num_tokens": 10724897.0,
      "step": 4305
    },
    {
      "entropy": 6.29885516166687,
      "epoch": 5.04093567251462,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003179524738602019,
      "loss": 5.6178,
      "mean_token_accuracy": 0.19183994233608245,
      "num_tokens": 10735808.0,
      "step": 4310
    },
    {
      "entropy": 6.269750356674194,
      "epoch": 5.046783625730994,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00031749226236577277,
      "loss": 5.5785,
      "mean_token_accuracy": 0.19983543157577516,
      "num_tokens": 10749083.0,
      "step": 4315
    },
    {
      "entropy": 6.302289056777954,
      "epoch": 5.052631578947368,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00031703186645187776,
      "loss": 5.6153,
      "mean_token_accuracy": 0.19364545494318008,
      "num_tokens": 10761215.0,
      "step": 4320
    },
    {
      "entropy": 6.1548424243927,
      "epoch": 5.058479532163743,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031657128811666836,
      "loss": 5.5264,
      "mean_token_accuracy": 0.20844042897224427,
      "num_tokens": 10772505.0,
      "step": 4325
    },
    {
      "entropy": 6.342819881439209,
      "epoch": 5.064327485380117,
      "grad_norm": 1.03125,
      "learning_rate": 0.00031611052935908736,
      "loss": 5.7182,
      "mean_token_accuracy": 0.1878417730331421,
      "num_tokens": 10785205.0,
      "step": 4330
    },
    {
      "entropy": 6.261581373214722,
      "epoch": 5.0701754385964914,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000315649592178861,
      "loss": 5.5404,
      "mean_token_accuracy": 0.19956978857517244,
      "num_tokens": 10796632.0,
      "step": 4335
    },
    {
      "entropy": 6.284210252761841,
      "epoch": 5.076023391812866,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003151884785764897,
      "loss": 5.6396,
      "mean_token_accuracy": 0.1910475254058838,
      "num_tokens": 10809466.0,
      "step": 4340
    },
    {
      "entropy": 6.293346643447876,
      "epoch": 5.081871345029239,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003147271905532394,
      "loss": 5.6667,
      "mean_token_accuracy": 0.19342056959867476,
      "num_tokens": 10821894.0,
      "step": 4345
    },
    {
      "entropy": 6.260929727554322,
      "epoch": 5.087719298245614,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003142657301111333,
      "loss": 5.6145,
      "mean_token_accuracy": 0.1909748524427414,
      "num_tokens": 10833612.0,
      "step": 4350
    },
    {
      "entropy": 6.288037395477295,
      "epoch": 5.093567251461988,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0003138040992529428,
      "loss": 5.6669,
      "mean_token_accuracy": 0.1894376441836357,
      "num_tokens": 10845146.0,
      "step": 4355
    },
    {
      "entropy": 6.2804004669189455,
      "epoch": 5.099415204678363,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000313342299982179,
      "loss": 5.6628,
      "mean_token_accuracy": 0.19310903698205947,
      "num_tokens": 10857376.0,
      "step": 4360
    },
    {
      "entropy": 6.31701078414917,
      "epoch": 5.105263157894737,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003128803343030836,
      "loss": 5.6469,
      "mean_token_accuracy": 0.1896333023905754,
      "num_tokens": 10868778.0,
      "step": 4365
    },
    {
      "entropy": 6.216046714782715,
      "epoch": 5.111111111111111,
      "grad_norm": 1.0625,
      "learning_rate": 0.00031241820422062117,
      "loss": 5.674,
      "mean_token_accuracy": 0.19011124819517136,
      "num_tokens": 10881050.0,
      "step": 4370
    },
    {
      "entropy": 6.297796440124512,
      "epoch": 5.116959064327485,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031195591174046914,
      "loss": 5.6839,
      "mean_token_accuracy": 0.192173507809639,
      "num_tokens": 10894468.0,
      "step": 4375
    },
    {
      "entropy": 6.300606346130371,
      "epoch": 5.12280701754386,
      "grad_norm": 1.0625,
      "learning_rate": 0.00031149345886901,
      "loss": 5.5967,
      "mean_token_accuracy": 0.20111036747694017,
      "num_tokens": 10907039.0,
      "step": 4380
    },
    {
      "entropy": 6.217218923568725,
      "epoch": 5.128654970760234,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003110308476133225,
      "loss": 5.6245,
      "mean_token_accuracy": 0.19180447310209275,
      "num_tokens": 10918772.0,
      "step": 4385
    },
    {
      "entropy": 6.2089524269104,
      "epoch": 5.1345029239766085,
      "grad_norm": 1.125,
      "learning_rate": 0.00031056807998117256,
      "loss": 5.5614,
      "mean_token_accuracy": 0.1983884170651436,
      "num_tokens": 10931397.0,
      "step": 4390
    },
    {
      "entropy": 6.301287078857422,
      "epoch": 5.140350877192983,
      "grad_norm": 1.125,
      "learning_rate": 0.000310105157981005,
      "loss": 5.6473,
      "mean_token_accuracy": 0.19144789427518843,
      "num_tokens": 10942728.0,
      "step": 4395
    },
    {
      "entropy": 6.2532453536987305,
      "epoch": 5.146198830409356,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030964208362193436,
      "loss": 5.6067,
      "mean_token_accuracy": 0.19152869731187822,
      "num_tokens": 10954277.0,
      "step": 4400
    },
    {
      "entropy": 6.244741487503052,
      "epoch": 5.152046783625731,
      "grad_norm": 1.125,
      "learning_rate": 0.00030917885891373664,
      "loss": 5.6324,
      "mean_token_accuracy": 0.18773233890533447,
      "num_tokens": 10966894.0,
      "step": 4405
    },
    {
      "entropy": 6.25692777633667,
      "epoch": 5.157894736842105,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003087154858668404,
      "loss": 5.6192,
      "mean_token_accuracy": 0.19250093251466752,
      "num_tokens": 10979229.0,
      "step": 4410
    },
    {
      "entropy": 6.262220859527588,
      "epoch": 5.16374269005848,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003082519664923177,
      "loss": 5.6795,
      "mean_token_accuracy": 0.1928941562771797,
      "num_tokens": 10991999.0,
      "step": 4415
    },
    {
      "entropy": 6.301309108734131,
      "epoch": 5.169590643274854,
      "grad_norm": 1.015625,
      "learning_rate": 0.000307788302801876,
      "loss": 5.6371,
      "mean_token_accuracy": 0.18814297169446945,
      "num_tokens": 11004408.0,
      "step": 4420
    },
    {
      "entropy": 6.2113710880279545,
      "epoch": 5.175438596491228,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030732449680784886,
      "loss": 5.7195,
      "mean_token_accuracy": 0.18902867585420607,
      "num_tokens": 11016322.0,
      "step": 4425
    },
    {
      "entropy": 6.295968961715698,
      "epoch": 5.181286549707602,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003068605505231877,
      "loss": 5.683,
      "mean_token_accuracy": 0.18986148089170457,
      "num_tokens": 11028798.0,
      "step": 4430
    },
    {
      "entropy": 6.232554960250854,
      "epoch": 5.187134502923977,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030639646596145243,
      "loss": 5.5387,
      "mean_token_accuracy": 0.19883911460638046,
      "num_tokens": 11039837.0,
      "step": 4435
    },
    {
      "entropy": 6.236491870880127,
      "epoch": 5.192982456140351,
      "grad_norm": 0.96875,
      "learning_rate": 0.00030593224513680357,
      "loss": 5.5831,
      "mean_token_accuracy": 0.19840813428163528,
      "num_tokens": 11052943.0,
      "step": 4440
    },
    {
      "entropy": 6.3170054912567135,
      "epoch": 5.1988304093567255,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003054678900639927,
      "loss": 5.6619,
      "mean_token_accuracy": 0.18190544545650483,
      "num_tokens": 11065483.0,
      "step": 4445
    },
    {
      "entropy": 6.2382776737213135,
      "epoch": 5.204678362573099,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00030500340275835405,
      "loss": 5.6256,
      "mean_token_accuracy": 0.19440329670906067,
      "num_tokens": 11077743.0,
      "step": 4450
    },
    {
      "entropy": 6.206258106231689,
      "epoch": 5.2105263157894735,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00030453878523579596,
      "loss": 5.646,
      "mean_token_accuracy": 0.18810033351182937,
      "num_tokens": 11089692.0,
      "step": 4455
    },
    {
      "entropy": 6.167195081710815,
      "epoch": 5.216374269005848,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030407403951279174,
      "loss": 5.4871,
      "mean_token_accuracy": 0.21046395599842072,
      "num_tokens": 11104220.0,
      "step": 4460
    },
    {
      "entropy": 6.262682199478149,
      "epoch": 5.222222222222222,
      "grad_norm": 1.125,
      "learning_rate": 0.0003036091676063711,
      "loss": 5.6073,
      "mean_token_accuracy": 0.18952247202396394,
      "num_tokens": 11115599.0,
      "step": 4465
    },
    {
      "entropy": 6.210060262680054,
      "epoch": 5.228070175438597,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003031441715341115,
      "loss": 5.6104,
      "mean_token_accuracy": 0.19400369375944138,
      "num_tokens": 11128125.0,
      "step": 4470
    },
    {
      "entropy": 6.218147802352905,
      "epoch": 5.23391812865497,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003026790533141293,
      "loss": 5.5777,
      "mean_token_accuracy": 0.2003661498427391,
      "num_tokens": 11140105.0,
      "step": 4475
    },
    {
      "entropy": 6.231682777404785,
      "epoch": 5.239766081871345,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003022138149650708,
      "loss": 5.589,
      "mean_token_accuracy": 0.1977740928530693,
      "num_tokens": 11152291.0,
      "step": 4480
    },
    {
      "entropy": 6.1422947406768795,
      "epoch": 5.245614035087719,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003017484585061039,
      "loss": 5.6353,
      "mean_token_accuracy": 0.19870062172412872,
      "num_tokens": 11166330.0,
      "step": 4485
    },
    {
      "entropy": 6.296531867980957,
      "epoch": 5.251461988304094,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030128298595690905,
      "loss": 5.6059,
      "mean_token_accuracy": 0.19451456665992736,
      "num_tokens": 11178737.0,
      "step": 4490
    },
    {
      "entropy": 6.289796686172485,
      "epoch": 5.257309941520468,
      "grad_norm": 0.96875,
      "learning_rate": 0.00030081739933767034,
      "loss": 5.6464,
      "mean_token_accuracy": 0.18791547566652297,
      "num_tokens": 11191179.0,
      "step": 4495
    },
    {
      "entropy": 6.1990679740905765,
      "epoch": 5.2631578947368425,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003003517006690672,
      "loss": 5.6163,
      "mean_token_accuracy": 0.19064578115940095,
      "num_tokens": 11203890.0,
      "step": 4500
    },
    {
      "epoch": 5.2631578947368425,
      "eval_entropy": 6.113778115725447,
      "eval_loss": 6.553351879119873,
      "eval_mean_token_accuracy": 0.15050386196788082,
      "eval_num_tokens": 11203890.0,
      "eval_runtime": 1.7383,
      "eval_samples_per_second": 1557.225,
      "eval_steps_per_second": 195.013,
      "step": 4500
    },
    {
      "entropy": 6.215317296981811,
      "epoch": 5.269005847953216,
      "grad_norm": 1.015625,
      "learning_rate": 0.00029988589197226515,
      "loss": 5.624,
      "mean_token_accuracy": 0.19421444535255433,
      "num_tokens": 11216556.0,
      "step": 4505
    },
    {
      "entropy": 6.258840322494507,
      "epoch": 5.2748538011695905,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029941997526890745,
      "loss": 5.6181,
      "mean_token_accuracy": 0.19635968506336213,
      "num_tokens": 11228352.0,
      "step": 4510
    },
    {
      "entropy": 6.212165307998657,
      "epoch": 5.280701754385965,
      "grad_norm": 1.109375,
      "learning_rate": 0.000298953952581106,
      "loss": 5.618,
      "mean_token_accuracy": 0.19511030614376068,
      "num_tokens": 11239423.0,
      "step": 4515
    },
    {
      "entropy": 6.256838941574097,
      "epoch": 5.286549707602339,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029848782593143246,
      "loss": 5.5767,
      "mean_token_accuracy": 0.19289738386869432,
      "num_tokens": 11251117.0,
      "step": 4520
    },
    {
      "entropy": 6.243201875686646,
      "epoch": 5.292397660818714,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029802159734291015,
      "loss": 5.6315,
      "mean_token_accuracy": 0.19678657948970796,
      "num_tokens": 11264459.0,
      "step": 4525
    },
    {
      "entropy": 6.137487840652466,
      "epoch": 5.298245614035087,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002975552688390045,
      "loss": 5.5349,
      "mean_token_accuracy": 0.20308712869882584,
      "num_tokens": 11276020.0,
      "step": 4530
    },
    {
      "entropy": 6.272948503494263,
      "epoch": 5.304093567251462,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029708884244361474,
      "loss": 5.617,
      "mean_token_accuracy": 0.19261521399021148,
      "num_tokens": 11287412.0,
      "step": 4535
    },
    {
      "entropy": 6.241686105728149,
      "epoch": 5.309941520467836,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029662232018106473,
      "loss": 5.6713,
      "mean_token_accuracy": 0.19322927445173263,
      "num_tokens": 11298607.0,
      "step": 4540
    },
    {
      "entropy": 6.294939613342285,
      "epoch": 5.315789473684211,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029615570407609467,
      "loss": 5.7334,
      "mean_token_accuracy": 0.18689310401678086,
      "num_tokens": 11310943.0,
      "step": 4545
    },
    {
      "entropy": 6.20140643119812,
      "epoch": 5.321637426900585,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029568899615385174,
      "loss": 5.5939,
      "mean_token_accuracy": 0.19311967194080354,
      "num_tokens": 11323684.0,
      "step": 4550
    },
    {
      "entropy": 6.146653270721435,
      "epoch": 5.3274853801169595,
      "grad_norm": 1.0,
      "learning_rate": 0.000295222198439882,
      "loss": 5.5517,
      "mean_token_accuracy": 0.1988784983754158,
      "num_tokens": 11336817.0,
      "step": 4555
    },
    {
      "entropy": 6.2802940845489506,
      "epoch": 5.333333333333333,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002947553129601209,
      "loss": 5.7265,
      "mean_token_accuracy": 0.18717574924230576,
      "num_tokens": 11349246.0,
      "step": 4560
    },
    {
      "entropy": 6.242254400253296,
      "epoch": 5.3391812865497075,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002942883417408849,
      "loss": 5.6258,
      "mean_token_accuracy": 0.19218788295984268,
      "num_tokens": 11360995.0,
      "step": 4565
    },
    {
      "entropy": 6.1783552169799805,
      "epoch": 5.345029239766082,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002938212868088626,
      "loss": 5.5639,
      "mean_token_accuracy": 0.20145443975925445,
      "num_tokens": 11373713.0,
      "step": 4570
    },
    {
      "entropy": 6.221517086029053,
      "epoch": 5.350877192982456,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00029335415019110593,
      "loss": 5.6418,
      "mean_token_accuracy": 0.1871470481157303,
      "num_tokens": 11386532.0,
      "step": 4575
    },
    {
      "entropy": 6.204015254974365,
      "epoch": 5.356725146198831,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0002928869339150212,
      "loss": 5.5389,
      "mean_token_accuracy": 0.20721419751644135,
      "num_tokens": 11400535.0,
      "step": 4580
    },
    {
      "entropy": 6.2049665451049805,
      "epoch": 5.362573099415204,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002924196400083607,
      "loss": 5.6121,
      "mean_token_accuracy": 0.2016605705022812,
      "num_tokens": 11412577.0,
      "step": 4585
    },
    {
      "entropy": 6.204962873458863,
      "epoch": 5.368421052631579,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029195227049921336,
      "loss": 5.6131,
      "mean_token_accuracy": 0.19435095489025117,
      "num_tokens": 11427142.0,
      "step": 4590
    },
    {
      "entropy": 6.296524238586426,
      "epoch": 5.374269005847953,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029148482741599634,
      "loss": 5.6625,
      "mean_token_accuracy": 0.1969881236553192,
      "num_tokens": 11441319.0,
      "step": 4595
    },
    {
      "entropy": 6.18909101486206,
      "epoch": 5.380116959064328,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00029101731278744636,
      "loss": 5.5879,
      "mean_token_accuracy": 0.1971047803759575,
      "num_tokens": 11453947.0,
      "step": 4600
    },
    {
      "entropy": 6.248491525650024,
      "epoch": 5.385964912280702,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002905497286426102,
      "loss": 5.6151,
      "mean_token_accuracy": 0.19610415399074554,
      "num_tokens": 11465463.0,
      "step": 4605
    },
    {
      "entropy": 6.231463098526001,
      "epoch": 5.391812865497076,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00029008207701083677,
      "loss": 5.6281,
      "mean_token_accuracy": 0.19632321149110793,
      "num_tokens": 11477249.0,
      "step": 4610
    },
    {
      "entropy": 6.267431592941284,
      "epoch": 5.39766081871345,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0002896143599217674,
      "loss": 5.688,
      "mean_token_accuracy": 0.19271611273288727,
      "num_tokens": 11490184.0,
      "step": 4615
    },
    {
      "entropy": 6.2184654712677006,
      "epoch": 5.4035087719298245,
      "grad_norm": 1.25,
      "learning_rate": 0.0002891465794053282,
      "loss": 5.6974,
      "mean_token_accuracy": 0.19165746718645096,
      "num_tokens": 11502935.0,
      "step": 4620
    },
    {
      "entropy": 6.227608680725098,
      "epoch": 5.409356725146199,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002886787374917198,
      "loss": 5.6541,
      "mean_token_accuracy": 0.19761152565479279,
      "num_tokens": 11515385.0,
      "step": 4625
    },
    {
      "entropy": 6.283583259582519,
      "epoch": 5.415204678362573,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00028821083621140986,
      "loss": 5.7158,
      "mean_token_accuracy": 0.18800681382417678,
      "num_tokens": 11528786.0,
      "step": 4630
    },
    {
      "entropy": 6.236066150665283,
      "epoch": 5.421052631578947,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002877428775951234,
      "loss": 5.7051,
      "mean_token_accuracy": 0.1876697763800621,
      "num_tokens": 11540401.0,
      "step": 4635
    },
    {
      "entropy": 6.272920274734497,
      "epoch": 5.426900584795321,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002872748636738344,
      "loss": 5.6587,
      "mean_token_accuracy": 0.1919675886631012,
      "num_tokens": 11552920.0,
      "step": 4640
    },
    {
      "entropy": 6.29360146522522,
      "epoch": 5.432748538011696,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002868067964787568,
      "loss": 5.7448,
      "mean_token_accuracy": 0.18205186128616332,
      "num_tokens": 11566237.0,
      "step": 4645
    },
    {
      "entropy": 6.217868232727051,
      "epoch": 5.43859649122807,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028633867804133583,
      "loss": 5.6375,
      "mean_token_accuracy": 0.1906483441591263,
      "num_tokens": 11577380.0,
      "step": 4650
    },
    {
      "entropy": 6.21820650100708,
      "epoch": 5.444444444444445,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002858705103932391,
      "loss": 5.6661,
      "mean_token_accuracy": 0.18434136360883713,
      "num_tokens": 11590708.0,
      "step": 4655
    },
    {
      "entropy": 6.3574059963226315,
      "epoch": 5.450292397660819,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028540229556634785,
      "loss": 5.7064,
      "mean_token_accuracy": 0.19090717136859894,
      "num_tokens": 11603351.0,
      "step": 4660
    },
    {
      "entropy": 6.166899824142456,
      "epoch": 5.456140350877193,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002849340355927477,
      "loss": 5.6307,
      "mean_token_accuracy": 0.19632581770420074,
      "num_tokens": 11615675.0,
      "step": 4665
    },
    {
      "entropy": 6.250672435760498,
      "epoch": 5.461988304093567,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002844657325047208,
      "loss": 5.6613,
      "mean_token_accuracy": 0.18908467143774033,
      "num_tokens": 11628184.0,
      "step": 4670
    },
    {
      "entropy": 6.2795250415802,
      "epoch": 5.4678362573099415,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002839973883347361,
      "loss": 5.6663,
      "mean_token_accuracy": 0.19002282321453096,
      "num_tokens": 11641058.0,
      "step": 4675
    },
    {
      "entropy": 6.091144990921021,
      "epoch": 5.473684210526316,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028352900511544076,
      "loss": 5.5743,
      "mean_token_accuracy": 0.20174657702445983,
      "num_tokens": 11653959.0,
      "step": 4680
    },
    {
      "entropy": 6.26455454826355,
      "epoch": 5.47953216374269,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028306058487965156,
      "loss": 5.6806,
      "mean_token_accuracy": 0.1875467285513878,
      "num_tokens": 11665241.0,
      "step": 4685
    },
    {
      "entropy": 6.318486213684082,
      "epoch": 5.485380116959064,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002825921296603461,
      "loss": 5.6887,
      "mean_token_accuracy": 0.1922642007470131,
      "num_tokens": 11678670.0,
      "step": 4690
    },
    {
      "entropy": 6.206989717483521,
      "epoch": 5.491228070175438,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028212364149065347,
      "loss": 5.5988,
      "mean_token_accuracy": 0.1932536318898201,
      "num_tokens": 11689964.0,
      "step": 4695
    },
    {
      "entropy": 6.231527662277221,
      "epoch": 5.497076023391813,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002816551224038458,
      "loss": 5.6621,
      "mean_token_accuracy": 0.19408544152975082,
      "num_tokens": 11701895.0,
      "step": 4700
    },
    {
      "entropy": 6.262688112258911,
      "epoch": 5.502923976608187,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00028118657443332984,
      "loss": 5.6327,
      "mean_token_accuracy": 0.19257069677114486,
      "num_tokens": 11714335.0,
      "step": 4705
    },
    {
      "entropy": 6.243125295639038,
      "epoch": 5.508771929824562,
      "grad_norm": 0.9375,
      "learning_rate": 0.0002807179996126371,
      "loss": 5.6134,
      "mean_token_accuracy": 0.19889171719551085,
      "num_tokens": 11726828.0,
      "step": 4710
    },
    {
      "entropy": 6.175675392150879,
      "epoch": 5.514619883040936,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028024939997541606,
      "loss": 5.5525,
      "mean_token_accuracy": 0.2006102442741394,
      "num_tokens": 11740729.0,
      "step": 4715
    },
    {
      "entropy": 6.25547456741333,
      "epoch": 5.52046783625731,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00027978077755542264,
      "loss": 5.6917,
      "mean_token_accuracy": 0.18714924305677413,
      "num_tokens": 11753017.0,
      "step": 4720
    },
    {
      "entropy": 6.290017032623291,
      "epoch": 5.526315789473684,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002793121343865119,
      "loss": 5.5773,
      "mean_token_accuracy": 0.19950668811798095,
      "num_tokens": 11766139.0,
      "step": 4725
    },
    {
      "entropy": 6.069074964523315,
      "epoch": 5.5321637426900585,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002788434725026286,
      "loss": 5.4965,
      "mean_token_accuracy": 0.2113112047314644,
      "num_tokens": 11780315.0,
      "step": 4730
    },
    {
      "entropy": 6.2608929634094235,
      "epoch": 5.538011695906433,
      "grad_norm": 1.125,
      "learning_rate": 0.000278374793937799,
      "loss": 5.6059,
      "mean_token_accuracy": 0.19730520844459534,
      "num_tokens": 11792342.0,
      "step": 4735
    },
    {
      "entropy": 6.199027729034424,
      "epoch": 5.543859649122807,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027790610072612174,
      "loss": 5.6204,
      "mean_token_accuracy": 0.19321949928998947,
      "num_tokens": 11804587.0,
      "step": 4740
    },
    {
      "entropy": 6.266590261459351,
      "epoch": 5.549707602339181,
      "grad_norm": 1.09375,
      "learning_rate": 0.00027743739490175903,
      "loss": 5.6792,
      "mean_token_accuracy": 0.19135109186172486,
      "num_tokens": 11815884.0,
      "step": 4745
    },
    {
      "entropy": 6.325189447402954,
      "epoch": 5.555555555555555,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002769686784989276,
      "loss": 5.7124,
      "mean_token_accuracy": 0.18764635622501374,
      "num_tokens": 11827279.0,
      "step": 4750
    },
    {
      "entropy": 6.224939489364624,
      "epoch": 5.56140350877193,
      "grad_norm": 1.03125,
      "learning_rate": 0.00027649995355189026,
      "loss": 5.6249,
      "mean_token_accuracy": 0.1913919195532799,
      "num_tokens": 11838646.0,
      "step": 4755
    },
    {
      "entropy": 6.28067193031311,
      "epoch": 5.567251461988304,
      "grad_norm": 1.015625,
      "learning_rate": 0.000276031222094947,
      "loss": 5.6659,
      "mean_token_accuracy": 0.19607614129781722,
      "num_tokens": 11850565.0,
      "step": 4760
    },
    {
      "entropy": 6.033517169952392,
      "epoch": 5.573099415204679,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002755624861624259,
      "loss": 5.4956,
      "mean_token_accuracy": 0.2141329050064087,
      "num_tokens": 11863244.0,
      "step": 4765
    },
    {
      "entropy": 6.116924619674682,
      "epoch": 5.578947368421053,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0002750937477886745,
      "loss": 5.5879,
      "mean_token_accuracy": 0.20116052776575089,
      "num_tokens": 11876884.0,
      "step": 4770
    },
    {
      "entropy": 6.261211204528808,
      "epoch": 5.584795321637427,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000274625009008051,
      "loss": 5.6075,
      "mean_token_accuracy": 0.1988896518945694,
      "num_tokens": 11889624.0,
      "step": 4775
    },
    {
      "entropy": 6.267670297622681,
      "epoch": 5.590643274853801,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002741562718549152,
      "loss": 5.6554,
      "mean_token_accuracy": 0.189441542327404,
      "num_tokens": 11901304.0,
      "step": 4780
    },
    {
      "entropy": 6.095806121826172,
      "epoch": 5.5964912280701755,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027368753836362023,
      "loss": 5.472,
      "mean_token_accuracy": 0.2054816946387291,
      "num_tokens": 11913188.0,
      "step": 4785
    },
    {
      "entropy": 6.245921802520752,
      "epoch": 5.60233918128655,
      "grad_norm": 0.984375,
      "learning_rate": 0.0002732188105685027,
      "loss": 5.6987,
      "mean_token_accuracy": 0.18979542702436447,
      "num_tokens": 11926771.0,
      "step": 4790
    },
    {
      "entropy": 6.24817008972168,
      "epoch": 5.6081871345029235,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002727500905038752,
      "loss": 5.6796,
      "mean_token_accuracy": 0.19076618999242784,
      "num_tokens": 11939162.0,
      "step": 4795
    },
    {
      "entropy": 6.300720310211181,
      "epoch": 5.614035087719298,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0002722813802040161,
      "loss": 5.7264,
      "mean_token_accuracy": 0.17963754385709763,
      "num_tokens": 11952376.0,
      "step": 4800
    },
    {
      "entropy": 6.161151361465454,
      "epoch": 5.619883040935672,
      "grad_norm": 1.0625,
      "learning_rate": 0.00027181268170316193,
      "loss": 5.5339,
      "mean_token_accuracy": 0.20123673975467682,
      "num_tokens": 11965154.0,
      "step": 4805
    },
    {
      "entropy": 6.202841329574585,
      "epoch": 5.625730994152047,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0002713439970354978,
      "loss": 5.5863,
      "mean_token_accuracy": 0.20163518637418748,
      "num_tokens": 11978492.0,
      "step": 4810
    },
    {
      "entropy": 6.2382164478302,
      "epoch": 5.631578947368421,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002708753282351486,
      "loss": 5.6584,
      "mean_token_accuracy": 0.19390579760074617,
      "num_tokens": 11990603.0,
      "step": 4815
    },
    {
      "entropy": 6.229855346679687,
      "epoch": 5.637426900584796,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002704066773361706,
      "loss": 5.7035,
      "mean_token_accuracy": 0.1881127655506134,
      "num_tokens": 12002608.0,
      "step": 4820
    },
    {
      "entropy": 6.234771108627319,
      "epoch": 5.643274853801169,
      "grad_norm": 1.0,
      "learning_rate": 0.0002699380463725425,
      "loss": 5.7078,
      "mean_token_accuracy": 0.1914609119296074,
      "num_tokens": 12016112.0,
      "step": 4825
    },
    {
      "entropy": 6.273275852203369,
      "epoch": 5.649122807017544,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000269469437378156,
      "loss": 5.7102,
      "mean_token_accuracy": 0.18690686523914338,
      "num_tokens": 12027685.0,
      "step": 4830
    },
    {
      "entropy": 6.159534454345703,
      "epoch": 5.654970760233918,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00026900085238680787,
      "loss": 5.5657,
      "mean_token_accuracy": 0.20058439821004867,
      "num_tokens": 12040524.0,
      "step": 4835
    },
    {
      "entropy": 6.170359659194946,
      "epoch": 5.6608187134502925,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002685322934321907,
      "loss": 5.6216,
      "mean_token_accuracy": 0.1950029268860817,
      "num_tokens": 12052841.0,
      "step": 4840
    },
    {
      "entropy": 6.26184344291687,
      "epoch": 5.666666666666667,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0002680637625478837,
      "loss": 5.6847,
      "mean_token_accuracy": 0.1896748051047325,
      "num_tokens": 12065375.0,
      "step": 4845
    },
    {
      "entropy": 6.3096733570098875,
      "epoch": 5.6725146198830405,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002675952617673446,
      "loss": 5.6738,
      "mean_token_accuracy": 0.19461602717638016,
      "num_tokens": 12078034.0,
      "step": 4850
    },
    {
      "entropy": 6.233367443084717,
      "epoch": 5.678362573099415,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002671267931239006,
      "loss": 5.7142,
      "mean_token_accuracy": 0.19131017327308655,
      "num_tokens": 12090218.0,
      "step": 4855
    },
    {
      "entropy": 6.208056783676147,
      "epoch": 5.684210526315789,
      "grad_norm": 1.078125,
      "learning_rate": 0.00026665835865073896,
      "loss": 5.6528,
      "mean_token_accuracy": 0.18746584802865982,
      "num_tokens": 12101927.0,
      "step": 4860
    },
    {
      "entropy": 6.317375993728637,
      "epoch": 5.690058479532164,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002661899603808989,
      "loss": 5.684,
      "mean_token_accuracy": 0.18479509949684142,
      "num_tokens": 12114887.0,
      "step": 4865
    },
    {
      "entropy": 6.279713439941406,
      "epoch": 5.695906432748538,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0002657216003472626,
      "loss": 5.6759,
      "mean_token_accuracy": 0.18884986788034439,
      "num_tokens": 12127151.0,
      "step": 4870
    },
    {
      "entropy": 6.216506385803223,
      "epoch": 5.701754385964913,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00026525328058254604,
      "loss": 5.6697,
      "mean_token_accuracy": 0.1974136933684349,
      "num_tokens": 12141987.0,
      "step": 4875
    },
    {
      "entropy": 6.174491548538208,
      "epoch": 5.707602339181286,
      "grad_norm": 1.015625,
      "learning_rate": 0.00026478500311929076,
      "loss": 5.5687,
      "mean_token_accuracy": 0.20029060840606688,
      "num_tokens": 12154525.0,
      "step": 4880
    },
    {
      "entropy": 6.12806191444397,
      "epoch": 5.713450292397661,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00026431676998985444,
      "loss": 5.5053,
      "mean_token_accuracy": 0.20539222210645675,
      "num_tokens": 12166599.0,
      "step": 4885
    },
    {
      "entropy": 6.223184537887573,
      "epoch": 5.719298245614035,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002638485832264024,
      "loss": 5.7342,
      "mean_token_accuracy": 0.19104734510183335,
      "num_tokens": 12178658.0,
      "step": 4890
    },
    {
      "entropy": 6.163381338119507,
      "epoch": 5.7251461988304095,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026338044486089895,
      "loss": 5.5816,
      "mean_token_accuracy": 0.19498748034238816,
      "num_tokens": 12190311.0,
      "step": 4895
    },
    {
      "entropy": 6.258320426940918,
      "epoch": 5.730994152046784,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00026291235692509795,
      "loss": 5.69,
      "mean_token_accuracy": 0.1898747056722641,
      "num_tokens": 12202443.0,
      "step": 4900
    },
    {
      "entropy": 6.242038917541504,
      "epoch": 5.7368421052631575,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00026244432145053467,
      "loss": 5.7135,
      "mean_token_accuracy": 0.1846013218164444,
      "num_tokens": 12214245.0,
      "step": 4905
    },
    {
      "entropy": 6.202081966400146,
      "epoch": 5.742690058479532,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002619763404685167,
      "loss": 5.6255,
      "mean_token_accuracy": 0.20055866837501526,
      "num_tokens": 12226752.0,
      "step": 4910
    },
    {
      "entropy": 6.239099788665771,
      "epoch": 5.748538011695906,
      "grad_norm": 0.984375,
      "learning_rate": 0.00026150841601011495,
      "loss": 5.6954,
      "mean_token_accuracy": 0.189306777715683,
      "num_tokens": 12239043.0,
      "step": 4915
    },
    {
      "entropy": 6.205452013015747,
      "epoch": 5.754385964912281,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002610405501061552,
      "loss": 5.673,
      "mean_token_accuracy": 0.1921948418021202,
      "num_tokens": 12250799.0,
      "step": 4920
    },
    {
      "entropy": 6.298740816116333,
      "epoch": 5.760233918128655,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002605727447872091,
      "loss": 5.809,
      "mean_token_accuracy": 0.1768986091017723,
      "num_tokens": 12264503.0,
      "step": 4925
    },
    {
      "entropy": 6.240444087982178,
      "epoch": 5.76608187134503,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002601050020835851,
      "loss": 5.6669,
      "mean_token_accuracy": 0.1936336949467659,
      "num_tokens": 12276106.0,
      "step": 4930
    },
    {
      "entropy": 6.2293507099151615,
      "epoch": 5.771929824561403,
      "grad_norm": 1.046875,
      "learning_rate": 0.00025963732402532026,
      "loss": 5.6037,
      "mean_token_accuracy": 0.1934416338801384,
      "num_tokens": 12288322.0,
      "step": 4935
    },
    {
      "entropy": 6.22146372795105,
      "epoch": 5.777777777777778,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002591697126421707,
      "loss": 5.655,
      "mean_token_accuracy": 0.1945774585008621,
      "num_tokens": 12301059.0,
      "step": 4940
    },
    {
      "entropy": 6.223192501068115,
      "epoch": 5.783625730994152,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002587021699636035,
      "loss": 5.6801,
      "mean_token_accuracy": 0.19261162281036376,
      "num_tokens": 12313206.0,
      "step": 4945
    },
    {
      "entropy": 6.24392409324646,
      "epoch": 5.7894736842105265,
      "grad_norm": 1.109375,
      "learning_rate": 0.00025823469801878734,
      "loss": 5.6154,
      "mean_token_accuracy": 0.19894942939281463,
      "num_tokens": 12325426.0,
      "step": 4950
    },
    {
      "entropy": 6.222641849517823,
      "epoch": 5.7953216374269,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00025776729883658406,
      "loss": 5.6917,
      "mean_token_accuracy": 0.19038689732551575,
      "num_tokens": 12338683.0,
      "step": 4955
    },
    {
      "entropy": 6.180020523071289,
      "epoch": 5.8011695906432745,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002572999744455395,
      "loss": 5.659,
      "mean_token_accuracy": 0.1979015976190567,
      "num_tokens": 12351139.0,
      "step": 4960
    },
    {
      "entropy": 6.203177165985108,
      "epoch": 5.807017543859649,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00025683272687387513,
      "loss": 5.5808,
      "mean_token_accuracy": 0.19831852614879608,
      "num_tokens": 12362804.0,
      "step": 4965
    },
    {
      "entropy": 6.306801700592041,
      "epoch": 5.812865497076023,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0002563655581494789,
      "loss": 5.7364,
      "mean_token_accuracy": 0.1888235554099083,
      "num_tokens": 12375069.0,
      "step": 4970
    },
    {
      "entropy": 6.251407861709595,
      "epoch": 5.818713450292398,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00025589847029989653,
      "loss": 5.6299,
      "mean_token_accuracy": 0.1940399333834648,
      "num_tokens": 12387434.0,
      "step": 4975
    },
    {
      "entropy": 6.188321256637574,
      "epoch": 5.824561403508772,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00025543146535232295,
      "loss": 5.68,
      "mean_token_accuracy": 0.19267833530902861,
      "num_tokens": 12400141.0,
      "step": 4980
    },
    {
      "entropy": 6.209116840362549,
      "epoch": 5.830409356725146,
      "grad_norm": 1.09375,
      "learning_rate": 0.00025496454533359284,
      "loss": 5.6153,
      "mean_token_accuracy": 0.19551173448562623,
      "num_tokens": 12412619.0,
      "step": 4985
    },
    {
      "entropy": 6.258868074417114,
      "epoch": 5.83625730994152,
      "grad_norm": 1.09375,
      "learning_rate": 0.00025449771227017265,
      "loss": 5.6323,
      "mean_token_accuracy": 0.19052976965904236,
      "num_tokens": 12425312.0,
      "step": 4990
    },
    {
      "entropy": 6.260711193084717,
      "epoch": 5.842105263157895,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002540309681881516,
      "loss": 5.6646,
      "mean_token_accuracy": 0.19524475634098054,
      "num_tokens": 12436972.0,
      "step": 4995
    },
    {
      "entropy": 6.324166345596313,
      "epoch": 5.847953216374269,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002535643151132321,
      "loss": 5.7166,
      "mean_token_accuracy": 0.1829158693552017,
      "num_tokens": 12448543.0,
      "step": 5000
    },
    {
      "epoch": 5.847953216374269,
      "eval_entropy": 6.192163904859605,
      "eval_loss": 6.517599582672119,
      "eval_mean_token_accuracy": 0.15165094630160866,
      "eval_num_tokens": 12448543.0,
      "eval_runtime": 1.7436,
      "eval_samples_per_second": 1552.56,
      "eval_steps_per_second": 194.428,
      "step": 5000
    },
    {
      "entropy": 6.229680395126342,
      "epoch": 5.853801169590644,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00025309775507072234,
      "loss": 5.6187,
      "mean_token_accuracy": 0.18957127928733825,
      "num_tokens": 12462292.0,
      "step": 5005
    },
    {
      "entropy": 6.216944169998169,
      "epoch": 5.859649122807017,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002526312900855261,
      "loss": 5.7363,
      "mean_token_accuracy": 0.18410075604915618,
      "num_tokens": 12475930.0,
      "step": 5010
    },
    {
      "entropy": 6.280484533309936,
      "epoch": 5.8654970760233915,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002521649221821349,
      "loss": 5.7373,
      "mean_token_accuracy": 0.18494972586631775,
      "num_tokens": 12487769.0,
      "step": 5015
    },
    {
      "entropy": 6.223931694030762,
      "epoch": 5.871345029239766,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00025169865338461894,
      "loss": 5.6444,
      "mean_token_accuracy": 0.19936908185482025,
      "num_tokens": 12500116.0,
      "step": 5020
    },
    {
      "entropy": 6.137562942504883,
      "epoch": 5.87719298245614,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002512324857166182,
      "loss": 5.5442,
      "mean_token_accuracy": 0.20381303429603576,
      "num_tokens": 12511896.0,
      "step": 5025
    },
    {
      "entropy": 6.2064227104187015,
      "epoch": 5.883040935672515,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002507664212013337,
      "loss": 5.62,
      "mean_token_accuracy": 0.19476156681776047,
      "num_tokens": 12524435.0,
      "step": 5030
    },
    {
      "entropy": 6.277655696868896,
      "epoch": 5.888888888888889,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025030046186151915,
      "loss": 5.6623,
      "mean_token_accuracy": 0.1907810837030411,
      "num_tokens": 12536315.0,
      "step": 5035
    },
    {
      "entropy": 6.2723016262054445,
      "epoch": 5.894736842105263,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002498346097194712,
      "loss": 5.694,
      "mean_token_accuracy": 0.1893911063671112,
      "num_tokens": 12548982.0,
      "step": 5040
    },
    {
      "entropy": 6.286884593963623,
      "epoch": 5.900584795321637,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00024936886679702167,
      "loss": 5.7987,
      "mean_token_accuracy": 0.18214447051286697,
      "num_tokens": 12562746.0,
      "step": 5045
    },
    {
      "entropy": 6.247114658355713,
      "epoch": 5.906432748538012,
      "grad_norm": 1.0625,
      "learning_rate": 0.00024890323511552825,
      "loss": 5.7139,
      "mean_token_accuracy": 0.19190758764743804,
      "num_tokens": 12575054.0,
      "step": 5050
    },
    {
      "entropy": 6.2986006259918215,
      "epoch": 5.912280701754386,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00024843771669586575,
      "loss": 5.7345,
      "mean_token_accuracy": 0.18518062084913253,
      "num_tokens": 12588636.0,
      "step": 5055
    },
    {
      "entropy": 6.204002332687378,
      "epoch": 5.918128654970761,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002479723135584176,
      "loss": 5.6536,
      "mean_token_accuracy": 0.19185338616371156,
      "num_tokens": 12602522.0,
      "step": 5060
    },
    {
      "entropy": 6.251021194458008,
      "epoch": 5.923976608187134,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0002475070277230668,
      "loss": 5.663,
      "mean_token_accuracy": 0.19039535969495774,
      "num_tokens": 12616367.0,
      "step": 5065
    },
    {
      "entropy": 6.226079607009888,
      "epoch": 5.9298245614035086,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002470418612091871,
      "loss": 5.6117,
      "mean_token_accuracy": 0.19544337838888168,
      "num_tokens": 12628469.0,
      "step": 5070
    },
    {
      "entropy": 6.2292975902557375,
      "epoch": 5.935672514619883,
      "grad_norm": 1.0,
      "learning_rate": 0.0002465768160356347,
      "loss": 5.5422,
      "mean_token_accuracy": 0.1984126463532448,
      "num_tokens": 12640506.0,
      "step": 5075
    },
    {
      "entropy": 6.178211402893067,
      "epoch": 5.941520467836257,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002461118942207388,
      "loss": 5.7122,
      "mean_token_accuracy": 0.1901325538754463,
      "num_tokens": 12652862.0,
      "step": 5080
    },
    {
      "entropy": 6.18978590965271,
      "epoch": 5.947368421052632,
      "grad_norm": 1.140625,
      "learning_rate": 0.00024564709778229353,
      "loss": 5.6165,
      "mean_token_accuracy": 0.19258674681186677,
      "num_tokens": 12664390.0,
      "step": 5085
    },
    {
      "entropy": 6.179649114608765,
      "epoch": 5.953216374269006,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00024518242873754887,
      "loss": 5.5731,
      "mean_token_accuracy": 0.20182013362646103,
      "num_tokens": 12676489.0,
      "step": 5090
    },
    {
      "entropy": 6.179190301895142,
      "epoch": 5.95906432748538,
      "grad_norm": 1.09375,
      "learning_rate": 0.00024471788910320163,
      "loss": 5.5556,
      "mean_token_accuracy": 0.19775737226009368,
      "num_tokens": 12688666.0,
      "step": 5095
    },
    {
      "entropy": 6.2486552715301515,
      "epoch": 5.964912280701754,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00024425348089538714,
      "loss": 5.6496,
      "mean_token_accuracy": 0.19471435248851776,
      "num_tokens": 12702159.0,
      "step": 5100
    },
    {
      "entropy": 6.254821634292602,
      "epoch": 5.970760233918129,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002437892061296705,
      "loss": 5.7029,
      "mean_token_accuracy": 0.1909840613603592,
      "num_tokens": 12715528.0,
      "step": 5105
    },
    {
      "entropy": 6.273462295532227,
      "epoch": 5.976608187134503,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00024332506682103722,
      "loss": 5.7483,
      "mean_token_accuracy": 0.18615872859954835,
      "num_tokens": 12727485.0,
      "step": 5110
    },
    {
      "entropy": 6.210376644134522,
      "epoch": 5.982456140350877,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00024286106498388533,
      "loss": 5.6414,
      "mean_token_accuracy": 0.19802975207567214,
      "num_tokens": 12740602.0,
      "step": 5115
    },
    {
      "entropy": 6.240117883682251,
      "epoch": 5.988304093567251,
      "grad_norm": 1.125,
      "learning_rate": 0.0002423972026320162,
      "loss": 5.6959,
      "mean_token_accuracy": 0.19337357729673385,
      "num_tokens": 12752599.0,
      "step": 5120
    },
    {
      "entropy": 6.253961086273193,
      "epoch": 5.994152046783626,
      "grad_norm": 1.09375,
      "learning_rate": 0.00024193348177862551,
      "loss": 5.6951,
      "mean_token_accuracy": 0.1918729364871979,
      "num_tokens": 12764515.0,
      "step": 5125
    },
    {
      "entropy": 6.241068315505982,
      "epoch": 6.0,
      "grad_norm": 1.140625,
      "learning_rate": 0.00024146990443629508,
      "loss": 5.6188,
      "mean_token_accuracy": 0.1931920275092125,
      "num_tokens": 12776904.0,
      "step": 5130
    },
    {
      "entropy": 6.291807460784912,
      "epoch": 6.005847953216374,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002410064726169838,
      "loss": 5.5736,
      "mean_token_accuracy": 0.19911656379699708,
      "num_tokens": 12789313.0,
      "step": 5135
    },
    {
      "entropy": 6.157122182846069,
      "epoch": 6.011695906432749,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00024054318833201894,
      "loss": 5.448,
      "mean_token_accuracy": 0.204440438747406,
      "num_tokens": 12801932.0,
      "step": 5140
    },
    {
      "entropy": 6.23592176437378,
      "epoch": 6.017543859649122,
      "grad_norm": 1.125,
      "learning_rate": 0.00024008005359208774,
      "loss": 5.478,
      "mean_token_accuracy": 0.20389650911092758,
      "num_tokens": 12814530.0,
      "step": 5145
    },
    {
      "entropy": 6.284766817092896,
      "epoch": 6.023391812865497,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023961707040722792,
      "loss": 5.5696,
      "mean_token_accuracy": 0.1926923081278801,
      "num_tokens": 12827053.0,
      "step": 5150
    },
    {
      "entropy": 6.19902868270874,
      "epoch": 6.029239766081871,
      "grad_norm": 1.09375,
      "learning_rate": 0.00023915424078681987,
      "loss": 5.4397,
      "mean_token_accuracy": 0.2031274065375328,
      "num_tokens": 12838875.0,
      "step": 5155
    },
    {
      "entropy": 6.24194016456604,
      "epoch": 6.035087719298246,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00023869156673957733,
      "loss": 5.5287,
      "mean_token_accuracy": 0.19613686203956604,
      "num_tokens": 12851434.0,
      "step": 5160
    },
    {
      "entropy": 6.222735166549683,
      "epoch": 6.04093567251462,
      "grad_norm": 1.125,
      "learning_rate": 0.00023822905027353863,
      "loss": 5.5231,
      "mean_token_accuracy": 0.196422578394413,
      "num_tokens": 12864275.0,
      "step": 5165
    },
    {
      "entropy": 6.184301471710205,
      "epoch": 6.046783625730994,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002377666933960586,
      "loss": 5.4814,
      "mean_token_accuracy": 0.2007116362452507,
      "num_tokens": 12876675.0,
      "step": 5170
    },
    {
      "entropy": 6.13069429397583,
      "epoch": 6.052631578947368,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002373044981137994,
      "loss": 5.3879,
      "mean_token_accuracy": 0.21685769259929658,
      "num_tokens": 12888482.0,
      "step": 5175
    },
    {
      "entropy": 6.180141592025757,
      "epoch": 6.058479532163743,
      "grad_norm": 1.0625,
      "learning_rate": 0.00023684246643272132,
      "loss": 5.4057,
      "mean_token_accuracy": 0.2033799782395363,
      "num_tokens": 12901631.0,
      "step": 5180
    },
    {
      "entropy": 6.266680955886841,
      "epoch": 6.064327485380117,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002363806003580752,
      "loss": 5.5707,
      "mean_token_accuracy": 0.1910759836435318,
      "num_tokens": 12913210.0,
      "step": 5185
    },
    {
      "entropy": 6.1630401611328125,
      "epoch": 6.0701754385964914,
      "grad_norm": 1.09375,
      "learning_rate": 0.00023591890189439296,
      "loss": 5.3993,
      "mean_token_accuracy": 0.21022044867277145,
      "num_tokens": 12924964.0,
      "step": 5190
    },
    {
      "entropy": 6.207860898971558,
      "epoch": 6.076023391812866,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023545737304547893,
      "loss": 5.5046,
      "mean_token_accuracy": 0.19483200311660767,
      "num_tokens": 12936986.0,
      "step": 5195
    },
    {
      "entropy": 6.34308819770813,
      "epoch": 6.081871345029239,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002349960158144015,
      "loss": 5.5653,
      "mean_token_accuracy": 0.19424990564584732,
      "num_tokens": 12949842.0,
      "step": 5200
    },
    {
      "entropy": 6.2509997367858885,
      "epoch": 6.087719298245614,
      "grad_norm": 1.078125,
      "learning_rate": 0.000234534832203484,
      "loss": 5.5399,
      "mean_token_accuracy": 0.19750141501426696,
      "num_tokens": 12962112.0,
      "step": 5205
    },
    {
      "entropy": 6.160948753356934,
      "epoch": 6.093567251461988,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002340738242142965,
      "loss": 5.4586,
      "mean_token_accuracy": 0.20483585745096206,
      "num_tokens": 12973599.0,
      "step": 5210
    },
    {
      "entropy": 6.132392311096192,
      "epoch": 6.099415204678363,
      "grad_norm": 1.0625,
      "learning_rate": 0.00023361299384764677,
      "loss": 5.3822,
      "mean_token_accuracy": 0.21149447560310364,
      "num_tokens": 12987402.0,
      "step": 5215
    },
    {
      "entropy": 6.155905532836914,
      "epoch": 6.105263157894737,
      "grad_norm": 1.078125,
      "learning_rate": 0.00023315234310357147,
      "loss": 5.4023,
      "mean_token_accuracy": 0.20342260897159575,
      "num_tokens": 12999299.0,
      "step": 5220
    },
    {
      "entropy": 6.23543610572815,
      "epoch": 6.111111111111111,
      "grad_norm": 1.0625,
      "learning_rate": 0.00023269187398132802,
      "loss": 5.5343,
      "mean_token_accuracy": 0.19965981543064118,
      "num_tokens": 13011911.0,
      "step": 5225
    },
    {
      "entropy": 6.218723154067993,
      "epoch": 6.116959064327485,
      "grad_norm": 1.125,
      "learning_rate": 0.00023223158847938543,
      "loss": 5.5023,
      "mean_token_accuracy": 0.20291199088096618,
      "num_tokens": 13023493.0,
      "step": 5230
    },
    {
      "entropy": 6.191776704788208,
      "epoch": 6.12280701754386,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023177148859541565,
      "loss": 5.4732,
      "mean_token_accuracy": 0.20537904500961304,
      "num_tokens": 13036364.0,
      "step": 5235
    },
    {
      "entropy": 6.164413738250732,
      "epoch": 6.128654970760234,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00023131157632628525,
      "loss": 5.5268,
      "mean_token_accuracy": 0.20510395169258117,
      "num_tokens": 13048809.0,
      "step": 5240
    },
    {
      "entropy": 6.183512926101685,
      "epoch": 6.1345029239766085,
      "grad_norm": 1.046875,
      "learning_rate": 0.00023085185366804656,
      "loss": 5.4294,
      "mean_token_accuracy": 0.2050108104944229,
      "num_tokens": 13061811.0,
      "step": 5245
    },
    {
      "entropy": 6.268472385406494,
      "epoch": 6.140350877192983,
      "grad_norm": 1.109375,
      "learning_rate": 0.00023039232261592868,
      "loss": 5.5404,
      "mean_token_accuracy": 0.19367864727973938,
      "num_tokens": 13074331.0,
      "step": 5250
    },
    {
      "entropy": 6.215866661071777,
      "epoch": 6.146198830409356,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00022993298516432937,
      "loss": 5.4837,
      "mean_token_accuracy": 0.20793522298336028,
      "num_tokens": 13087642.0,
      "step": 5255
    },
    {
      "entropy": 6.284522247314453,
      "epoch": 6.152046783625731,
      "grad_norm": 1.125,
      "learning_rate": 0.00022947384330680636,
      "loss": 5.5798,
      "mean_token_accuracy": 0.18925997614860535,
      "num_tokens": 13099453.0,
      "step": 5260
    },
    {
      "entropy": 6.235210180282593,
      "epoch": 6.157894736842105,
      "grad_norm": 1.078125,
      "learning_rate": 0.00022901489903606794,
      "loss": 5.4744,
      "mean_token_accuracy": 0.20040415227413177,
      "num_tokens": 13112092.0,
      "step": 5265
    },
    {
      "entropy": 6.152941179275513,
      "epoch": 6.16374269005848,
      "grad_norm": 1.0625,
      "learning_rate": 0.00022855615434396527,
      "loss": 5.3836,
      "mean_token_accuracy": 0.2137580007314682,
      "num_tokens": 13125023.0,
      "step": 5270
    },
    {
      "entropy": 6.16050090789795,
      "epoch": 6.169590643274854,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002280976112214832,
      "loss": 5.537,
      "mean_token_accuracy": 0.19960876405239106,
      "num_tokens": 13137902.0,
      "step": 5275
    },
    {
      "entropy": 6.131514835357666,
      "epoch": 6.175438596491228,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00022763927165873177,
      "loss": 5.4341,
      "mean_token_accuracy": 0.2074015259742737,
      "num_tokens": 13151245.0,
      "step": 5280
    },
    {
      "entropy": 6.049352645874023,
      "epoch": 6.181286549707602,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002271811376449376,
      "loss": 5.3254,
      "mean_token_accuracy": 0.21783139705657958,
      "num_tokens": 13163916.0,
      "step": 5285
    },
    {
      "entropy": 6.248342275619507,
      "epoch": 6.187134502923977,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00022672321116843497,
      "loss": 5.4971,
      "mean_token_accuracy": 0.2049075648188591,
      "num_tokens": 13176265.0,
      "step": 5290
    },
    {
      "entropy": 6.209170866012573,
      "epoch": 6.192982456140351,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00022626549421665788,
      "loss": 5.5336,
      "mean_token_accuracy": 0.1983010560274124,
      "num_tokens": 13188916.0,
      "step": 5295
    },
    {
      "entropy": 6.14793586730957,
      "epoch": 6.1988304093567255,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00022580798877613062,
      "loss": 5.4438,
      "mean_token_accuracy": 0.209325410425663,
      "num_tokens": 13201444.0,
      "step": 5300
    },
    {
      "entropy": 6.271620512008667,
      "epoch": 6.204678362573099,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00022535069683245952,
      "loss": 5.4888,
      "mean_token_accuracy": 0.2017672061920166,
      "num_tokens": 13213390.0,
      "step": 5305
    },
    {
      "entropy": 6.141740846633911,
      "epoch": 6.2105263157894735,
      "grad_norm": 1.0625,
      "learning_rate": 0.00022489362037032446,
      "loss": 5.4103,
      "mean_token_accuracy": 0.2053940102458,
      "num_tokens": 13225321.0,
      "step": 5310
    },
    {
      "entropy": 6.145274019241333,
      "epoch": 6.216374269005848,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002244367613734702,
      "loss": 5.4608,
      "mean_token_accuracy": 0.2004158928990364,
      "num_tokens": 13237737.0,
      "step": 5315
    },
    {
      "entropy": 6.2553142547607425,
      "epoch": 6.222222222222222,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00022398012182469734,
      "loss": 5.5833,
      "mean_token_accuracy": 0.1963038668036461,
      "num_tokens": 13250235.0,
      "step": 5320
    },
    {
      "entropy": 6.2386799335479735,
      "epoch": 6.228070175438597,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002235237037058544,
      "loss": 5.5596,
      "mean_token_accuracy": 0.19574707001447678,
      "num_tokens": 13263537.0,
      "step": 5325
    },
    {
      "entropy": 6.214184093475342,
      "epoch": 6.23391812865497,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00022306750899782875,
      "loss": 5.5071,
      "mean_token_accuracy": 0.2034445062279701,
      "num_tokens": 13275436.0,
      "step": 5330
    },
    {
      "entropy": 6.242596006393432,
      "epoch": 6.239766081871345,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000222611539680538,
      "loss": 5.5677,
      "mean_token_accuracy": 0.19689661711454393,
      "num_tokens": 13287346.0,
      "step": 5335
    },
    {
      "entropy": 6.155242156982422,
      "epoch": 6.245614035087719,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00022215579773292188,
      "loss": 5.4426,
      "mean_token_accuracy": 0.2102979376912117,
      "num_tokens": 13299145.0,
      "step": 5340
    },
    {
      "entropy": 6.2098369121551515,
      "epoch": 6.251461988304094,
      "grad_norm": 1.09375,
      "learning_rate": 0.00022170028513293288,
      "loss": 5.5009,
      "mean_token_accuracy": 0.2047211542725563,
      "num_tokens": 13311709.0,
      "step": 5345
    },
    {
      "entropy": 6.194003057479859,
      "epoch": 6.257309941520468,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002212450038575285,
      "loss": 5.4895,
      "mean_token_accuracy": 0.2053727015852928,
      "num_tokens": 13324441.0,
      "step": 5350
    },
    {
      "entropy": 6.178544044494629,
      "epoch": 6.2631578947368425,
      "grad_norm": 1.046875,
      "learning_rate": 0.00022078995588266227,
      "loss": 5.4209,
      "mean_token_accuracy": 0.21021911799907683,
      "num_tokens": 13337276.0,
      "step": 5355
    },
    {
      "entropy": 6.1617844581604,
      "epoch": 6.269005847953216,
      "grad_norm": 1.125,
      "learning_rate": 0.0002203351431832748,
      "loss": 5.4625,
      "mean_token_accuracy": 0.20122915208339692,
      "num_tokens": 13350254.0,
      "step": 5360
    },
    {
      "entropy": 6.190066242218018,
      "epoch": 6.2748538011695905,
      "grad_norm": 0.96875,
      "learning_rate": 0.0002198805677332858,
      "loss": 5.5761,
      "mean_token_accuracy": 0.20074831694364548,
      "num_tokens": 13363430.0,
      "step": 5365
    },
    {
      "entropy": 6.234248352050781,
      "epoch": 6.280701754385965,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021942623150558556,
      "loss": 5.47,
      "mean_token_accuracy": 0.20248827189207078,
      "num_tokens": 13375350.0,
      "step": 5370
    },
    {
      "entropy": 6.21545820236206,
      "epoch": 6.286549707602339,
      "grad_norm": 1.125,
      "learning_rate": 0.00021897213647202558,
      "loss": 5.5262,
      "mean_token_accuracy": 0.20217197239398957,
      "num_tokens": 13388339.0,
      "step": 5375
    },
    {
      "entropy": 6.097126245498657,
      "epoch": 6.292397660818714,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002185182846034111,
      "loss": 5.3964,
      "mean_token_accuracy": 0.21759578883647918,
      "num_tokens": 13401122.0,
      "step": 5380
    },
    {
      "entropy": 6.185589599609375,
      "epoch": 6.298245614035087,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002180646778694918,
      "loss": 5.4736,
      "mean_token_accuracy": 0.20595618486404418,
      "num_tokens": 13413428.0,
      "step": 5385
    },
    {
      "entropy": 6.209289789199829,
      "epoch": 6.304093567251462,
      "grad_norm": 1.03125,
      "learning_rate": 0.00021761131823895324,
      "loss": 5.5042,
      "mean_token_accuracy": 0.20574771910905837,
      "num_tokens": 13427045.0,
      "step": 5390
    },
    {
      "entropy": 6.215899467468262,
      "epoch": 6.309941520467836,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002171582076794088,
      "loss": 5.5052,
      "mean_token_accuracy": 0.20421712547540666,
      "num_tokens": 13439524.0,
      "step": 5395
    },
    {
      "entropy": 6.24078893661499,
      "epoch": 6.315789473684211,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002167053481573909,
      "loss": 5.5446,
      "mean_token_accuracy": 0.19990173429250718,
      "num_tokens": 13453705.0,
      "step": 5400
    },
    {
      "entropy": 6.168670606613159,
      "epoch": 6.321637426900585,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00021625274163834224,
      "loss": 5.4341,
      "mean_token_accuracy": 0.21321797519922256,
      "num_tokens": 13466129.0,
      "step": 5405
    },
    {
      "entropy": 6.20025143623352,
      "epoch": 6.3274853801169595,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00021580039008660762,
      "loss": 5.5375,
      "mean_token_accuracy": 0.20049335360527037,
      "num_tokens": 13479522.0,
      "step": 5410
    },
    {
      "entropy": 6.195522975921631,
      "epoch": 6.333333333333333,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021534829546542518,
      "loss": 5.4242,
      "mean_token_accuracy": 0.19978488534688948,
      "num_tokens": 13492620.0,
      "step": 5415
    },
    {
      "entropy": 6.177324533462524,
      "epoch": 6.3391812865497075,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002148964597369179,
      "loss": 5.4372,
      "mean_token_accuracy": 0.20522774904966354,
      "num_tokens": 13503573.0,
      "step": 5420
    },
    {
      "entropy": 6.150057363510132,
      "epoch": 6.345029239766082,
      "grad_norm": 1.171875,
      "learning_rate": 0.00021444488486208545,
      "loss": 5.481,
      "mean_token_accuracy": 0.21047003120183944,
      "num_tokens": 13515999.0,
      "step": 5425
    },
    {
      "entropy": 6.209057998657227,
      "epoch": 6.350877192982456,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00021399357280079502,
      "loss": 5.4737,
      "mean_token_accuracy": 0.20550433546304703,
      "num_tokens": 13528480.0,
      "step": 5430
    },
    {
      "entropy": 6.189521455764771,
      "epoch": 6.356725146198831,
      "grad_norm": 1.15625,
      "learning_rate": 0.00021354252551177333,
      "loss": 5.5127,
      "mean_token_accuracy": 0.20278425067663192,
      "num_tokens": 13540606.0,
      "step": 5435
    },
    {
      "entropy": 6.277863502502441,
      "epoch": 6.362573099415204,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00021309174495259814,
      "loss": 5.5187,
      "mean_token_accuracy": 0.20014743953943254,
      "num_tokens": 13552947.0,
      "step": 5440
    },
    {
      "entropy": 6.196978998184204,
      "epoch": 6.368421052631579,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00021264123307968913,
      "loss": 5.4827,
      "mean_token_accuracy": 0.1998566910624504,
      "num_tokens": 13565499.0,
      "step": 5445
    },
    {
      "entropy": 6.102293157577515,
      "epoch": 6.374269005847953,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002121909918483002,
      "loss": 5.3704,
      "mean_token_accuracy": 0.22231035381555558,
      "num_tokens": 13578491.0,
      "step": 5450
    },
    {
      "entropy": 6.153888273239136,
      "epoch": 6.380116959064328,
      "grad_norm": 1.078125,
      "learning_rate": 0.00021174102321251082,
      "loss": 5.4688,
      "mean_token_accuracy": 0.20224058777093887,
      "num_tokens": 13590638.0,
      "step": 5455
    },
    {
      "entropy": 6.223382043838501,
      "epoch": 6.385964912280702,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002112913291252168,
      "loss": 5.5324,
      "mean_token_accuracy": 0.19592385739088058,
      "num_tokens": 13602198.0,
      "step": 5460
    },
    {
      "entropy": 6.203711605072021,
      "epoch": 6.391812865497076,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000210841911538123,
      "loss": 5.477,
      "mean_token_accuracy": 0.19936007857322693,
      "num_tokens": 13615520.0,
      "step": 5465
    },
    {
      "entropy": 6.16109790802002,
      "epoch": 6.39766081871345,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00021039277240173372,
      "loss": 5.5019,
      "mean_token_accuracy": 0.20331777185201644,
      "num_tokens": 13629382.0,
      "step": 5470
    },
    {
      "entropy": 6.038287925720215,
      "epoch": 6.4035087719298245,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0002099439136653452,
      "loss": 5.2765,
      "mean_token_accuracy": 0.2233900710940361,
      "num_tokens": 13642336.0,
      "step": 5475
    },
    {
      "entropy": 6.188841342926025,
      "epoch": 6.409356725146199,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00020949533727703657,
      "loss": 5.5166,
      "mean_token_accuracy": 0.19573960751295089,
      "num_tokens": 13654729.0,
      "step": 5480
    },
    {
      "entropy": 6.176787662506103,
      "epoch": 6.415204678362573,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00020904704518366136,
      "loss": 5.4608,
      "mean_token_accuracy": 0.21132799834012986,
      "num_tokens": 13667068.0,
      "step": 5485
    },
    {
      "entropy": 6.136917209625244,
      "epoch": 6.421052631578947,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00020859903933083952,
      "loss": 5.4859,
      "mean_token_accuracy": 0.20762813836336136,
      "num_tokens": 13680164.0,
      "step": 5490
    },
    {
      "entropy": 6.233702182769775,
      "epoch": 6.426900584795321,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002081513216629486,
      "loss": 5.5786,
      "mean_token_accuracy": 0.19450497031211852,
      "num_tokens": 13693510.0,
      "step": 5495
    },
    {
      "entropy": 6.21709246635437,
      "epoch": 6.432748538011696,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002077038941231153,
      "loss": 5.4664,
      "mean_token_accuracy": 0.20425813049077987,
      "num_tokens": 13705763.0,
      "step": 5500
    },
    {
      "epoch": 6.432748538011696,
      "eval_entropy": 6.072300210463262,
      "eval_loss": 6.5412373542785645,
      "eval_mean_token_accuracy": 0.1516292507273961,
      "eval_num_tokens": 13705763.0,
      "eval_runtime": 1.7429,
      "eval_samples_per_second": 1553.114,
      "eval_steps_per_second": 194.498,
      "step": 5500
    },
    {
      "entropy": 6.196304988861084,
      "epoch": 6.43859649122807,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00020725675865320721,
      "loss": 5.4795,
      "mean_token_accuracy": 0.20288982689380647,
      "num_tokens": 13719223.0,
      "step": 5505
    },
    {
      "entropy": 6.144592523574829,
      "epoch": 6.444444444444445,
      "grad_norm": 1.078125,
      "learning_rate": 0.00020680991719382446,
      "loss": 5.4597,
      "mean_token_accuracy": 0.2067062497138977,
      "num_tokens": 13731571.0,
      "step": 5510
    },
    {
      "entropy": 6.217992305755615,
      "epoch": 6.450292397660819,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00020636337168429092,
      "loss": 5.5942,
      "mean_token_accuracy": 0.19881327599287033,
      "num_tokens": 13744212.0,
      "step": 5515
    },
    {
      "entropy": 6.25103006362915,
      "epoch": 6.456140350877193,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020591712406264613,
      "loss": 5.5837,
      "mean_token_accuracy": 0.19580857306718827,
      "num_tokens": 13756402.0,
      "step": 5520
    },
    {
      "entropy": 6.265035390853882,
      "epoch": 6.461988304093567,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002054711762656369,
      "loss": 5.497,
      "mean_token_accuracy": 0.20531173944473266,
      "num_tokens": 13768386.0,
      "step": 5525
    },
    {
      "entropy": 6.127621364593506,
      "epoch": 6.4678362573099415,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002050255302287084,
      "loss": 5.4543,
      "mean_token_accuracy": 0.21058733016252518,
      "num_tokens": 13778732.0,
      "step": 5530
    },
    {
      "entropy": 6.149897050857544,
      "epoch": 6.473684210526316,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002045801878859967,
      "loss": 5.4593,
      "mean_token_accuracy": 0.2120043382048607,
      "num_tokens": 13791708.0,
      "step": 5535
    },
    {
      "entropy": 6.203775882720947,
      "epoch": 6.47953216374269,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00020413515117031917,
      "loss": 5.5342,
      "mean_token_accuracy": 0.20008005946874619,
      "num_tokens": 13802526.0,
      "step": 5540
    },
    {
      "entropy": 6.175456190109253,
      "epoch": 6.485380116959064,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020369042201316722,
      "loss": 5.5101,
      "mean_token_accuracy": 0.20407217144966125,
      "num_tokens": 13815342.0,
      "step": 5545
    },
    {
      "entropy": 6.262883138656616,
      "epoch": 6.491228070175438,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00020324600234469743,
      "loss": 5.509,
      "mean_token_accuracy": 0.20581623464822768,
      "num_tokens": 13827223.0,
      "step": 5550
    },
    {
      "entropy": 6.1237095355987545,
      "epoch": 6.497076023391813,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020280189409372274,
      "loss": 5.463,
      "mean_token_accuracy": 0.20224383175373079,
      "num_tokens": 13840366.0,
      "step": 5555
    },
    {
      "entropy": 6.175697994232178,
      "epoch": 6.502923976608187,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020235809918770504,
      "loss": 5.5912,
      "mean_token_accuracy": 0.1904074639081955,
      "num_tokens": 13853295.0,
      "step": 5560
    },
    {
      "entropy": 6.231081008911133,
      "epoch": 6.508771929824562,
      "grad_norm": 0.984375,
      "learning_rate": 0.00020191461955274597,
      "loss": 5.5732,
      "mean_token_accuracy": 0.20104458034038544,
      "num_tokens": 13866551.0,
      "step": 5565
    },
    {
      "entropy": 6.253126525878907,
      "epoch": 6.514619883040936,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002014714571135789,
      "loss": 5.5466,
      "mean_token_accuracy": 0.2001429781317711,
      "num_tokens": 13878048.0,
      "step": 5570
    },
    {
      "entropy": 6.191002655029297,
      "epoch": 6.52046783625731,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00020102861379356064,
      "loss": 5.502,
      "mean_token_accuracy": 0.20357385128736497,
      "num_tokens": 13890742.0,
      "step": 5575
    },
    {
      "entropy": 6.125066661834717,
      "epoch": 6.526315789473684,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000200586091514663,
      "loss": 5.4732,
      "mean_token_accuracy": 0.2054262012243271,
      "num_tokens": 13902943.0,
      "step": 5580
    },
    {
      "entropy": 6.164782476425171,
      "epoch": 6.5321637426900585,
      "grad_norm": 1.109375,
      "learning_rate": 0.00020014389219746438,
      "loss": 5.4438,
      "mean_token_accuracy": 0.19984354227781295,
      "num_tokens": 13915732.0,
      "step": 5585
    },
    {
      "entropy": 6.157883834838867,
      "epoch": 6.538011695906433,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001997020177611414,
      "loss": 5.4695,
      "mean_token_accuracy": 0.20199766606092454,
      "num_tokens": 13929173.0,
      "step": 5590
    },
    {
      "entropy": 6.2046833515167235,
      "epoch": 6.543859649122807,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00019926047012346098,
      "loss": 5.4531,
      "mean_token_accuracy": 0.20456216335296631,
      "num_tokens": 13940288.0,
      "step": 5595
    },
    {
      "entropy": 6.117101001739502,
      "epoch": 6.549707602339181,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00019881925120077143,
      "loss": 5.3981,
      "mean_token_accuracy": 0.21367185860872268,
      "num_tokens": 13952633.0,
      "step": 5600
    },
    {
      "entropy": 6.095827341079712,
      "epoch": 6.555555555555555,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00019837836290799454,
      "loss": 5.4019,
      "mean_token_accuracy": 0.2148555263876915,
      "num_tokens": 13964040.0,
      "step": 5605
    },
    {
      "entropy": 6.1827208518981935,
      "epoch": 6.56140350877193,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001979378071586171,
      "loss": 5.4417,
      "mean_token_accuracy": 0.20911121666431426,
      "num_tokens": 13976156.0,
      "step": 5610
    },
    {
      "entropy": 6.186322498321533,
      "epoch": 6.567251461988304,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001974975858646827,
      "loss": 5.4454,
      "mean_token_accuracy": 0.2106892630457878,
      "num_tokens": 13987815.0,
      "step": 5615
    },
    {
      "entropy": 6.211847448348999,
      "epoch": 6.573099415204679,
      "grad_norm": 1.21875,
      "learning_rate": 0.00019705770093678333,
      "loss": 5.5409,
      "mean_token_accuracy": 0.19123295843601226,
      "num_tokens": 13999239.0,
      "step": 5620
    },
    {
      "entropy": 6.1919292449951175,
      "epoch": 6.578947368421053,
      "grad_norm": 1.015625,
      "learning_rate": 0.00019661815428405094,
      "loss": 5.5489,
      "mean_token_accuracy": 0.19548619389533997,
      "num_tokens": 14012146.0,
      "step": 5625
    },
    {
      "entropy": 6.2336047172546385,
      "epoch": 6.584795321637427,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00019617894781414968,
      "loss": 5.4979,
      "mean_token_accuracy": 0.2056644767522812,
      "num_tokens": 14024292.0,
      "step": 5630
    },
    {
      "entropy": 6.199937772750855,
      "epoch": 6.590643274853801,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00019574008343326726,
      "loss": 5.4724,
      "mean_token_accuracy": 0.2025554358959198,
      "num_tokens": 14037011.0,
      "step": 5635
    },
    {
      "entropy": 6.222061681747436,
      "epoch": 6.5964912280701755,
      "grad_norm": 1.109375,
      "learning_rate": 0.00019530156304610623,
      "loss": 5.4578,
      "mean_token_accuracy": 0.19910242259502411,
      "num_tokens": 14049006.0,
      "step": 5640
    },
    {
      "entropy": 6.116253614425659,
      "epoch": 6.60233918128655,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019486338855587683,
      "loss": 5.464,
      "mean_token_accuracy": 0.2030980184674263,
      "num_tokens": 14061594.0,
      "step": 5645
    },
    {
      "entropy": 6.214007329940796,
      "epoch": 6.6081871345029235,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019442556186428762,
      "loss": 5.5444,
      "mean_token_accuracy": 0.19521572887897493,
      "num_tokens": 14073244.0,
      "step": 5650
    },
    {
      "entropy": 6.168609476089477,
      "epoch": 6.614035087719298,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001939880848715378,
      "loss": 5.4261,
      "mean_token_accuracy": 0.21149444878101348,
      "num_tokens": 14086695.0,
      "step": 5655
    },
    {
      "entropy": 6.177991342544556,
      "epoch": 6.619883040935672,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00019355095947630897,
      "loss": 5.4931,
      "mean_token_accuracy": 0.19622178822755815,
      "num_tokens": 14099161.0,
      "step": 5660
    },
    {
      "entropy": 6.213022661209107,
      "epoch": 6.625730994152047,
      "grad_norm": 1.109375,
      "learning_rate": 0.00019311418757575672,
      "loss": 5.4891,
      "mean_token_accuracy": 0.20098488330841063,
      "num_tokens": 14110602.0,
      "step": 5665
    },
    {
      "entropy": 6.158489465713501,
      "epoch": 6.631578947368421,
      "grad_norm": 1.140625,
      "learning_rate": 0.00019267777106550229,
      "loss": 5.4419,
      "mean_token_accuracy": 0.20287867188453673,
      "num_tokens": 14122802.0,
      "step": 5670
    },
    {
      "entropy": 6.109805727005005,
      "epoch": 6.637426900584796,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00019224171183962474,
      "loss": 5.4952,
      "mean_token_accuracy": 0.2055875152349472,
      "num_tokens": 14135296.0,
      "step": 5675
    },
    {
      "entropy": 6.293835783004761,
      "epoch": 6.643274853801169,
      "grad_norm": 1.21875,
      "learning_rate": 0.00019180601179065232,
      "loss": 5.5948,
      "mean_token_accuracy": 0.19030413627624512,
      "num_tokens": 14146825.0,
      "step": 5680
    },
    {
      "entropy": 6.277869606018067,
      "epoch": 6.649122807017544,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001913706728095545,
      "loss": 5.5355,
      "mean_token_accuracy": 0.19599652290344238,
      "num_tokens": 14159655.0,
      "step": 5685
    },
    {
      "entropy": 6.228263807296753,
      "epoch": 6.654970760233918,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00019093569678573375,
      "loss": 5.5728,
      "mean_token_accuracy": 0.19823792427778245,
      "num_tokens": 14171421.0,
      "step": 5690
    },
    {
      "entropy": 6.130559062957763,
      "epoch": 6.6608187134502925,
      "grad_norm": 1.109375,
      "learning_rate": 0.00019050108560701706,
      "loss": 5.5293,
      "mean_token_accuracy": 0.2006700173020363,
      "num_tokens": 14183890.0,
      "step": 5695
    },
    {
      "entropy": 6.1888469696044925,
      "epoch": 6.666666666666667,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019006684115964828,
      "loss": 5.5402,
      "mean_token_accuracy": 0.19505147635936737,
      "num_tokens": 14195715.0,
      "step": 5700
    },
    {
      "entropy": 6.236068487167358,
      "epoch": 6.6725146198830405,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018963296532827935,
      "loss": 5.5497,
      "mean_token_accuracy": 0.1952495187520981,
      "num_tokens": 14207426.0,
      "step": 5705
    },
    {
      "entropy": 6.208843994140625,
      "epoch": 6.678362573099415,
      "grad_norm": 1.0625,
      "learning_rate": 0.00018919945999596244,
      "loss": 5.526,
      "mean_token_accuracy": 0.20125488191843033,
      "num_tokens": 14219491.0,
      "step": 5710
    },
    {
      "entropy": 6.210028076171875,
      "epoch": 6.684210526315789,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001887663270441418,
      "loss": 5.5292,
      "mean_token_accuracy": 0.20554223358631135,
      "num_tokens": 14232130.0,
      "step": 5715
    },
    {
      "entropy": 6.248504734039306,
      "epoch": 6.690058479532164,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00018833356835264576,
      "loss": 5.4995,
      "mean_token_accuracy": 0.19406357556581497,
      "num_tokens": 14243323.0,
      "step": 5720
    },
    {
      "entropy": 6.191807746887207,
      "epoch": 6.695906432748538,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00018790118579967762,
      "loss": 5.5575,
      "mean_token_accuracy": 0.1974978968501091,
      "num_tokens": 14255338.0,
      "step": 5725
    },
    {
      "entropy": 6.222339534759522,
      "epoch": 6.701754385964913,
      "grad_norm": 1.078125,
      "learning_rate": 0.000187469181261809,
      "loss": 5.5464,
      "mean_token_accuracy": 0.19952429234981536,
      "num_tokens": 14268259.0,
      "step": 5730
    },
    {
      "entropy": 6.206273889541626,
      "epoch": 6.707602339181286,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001870375566139706,
      "loss": 5.5429,
      "mean_token_accuracy": 0.20098696798086166,
      "num_tokens": 14280208.0,
      "step": 5735
    },
    {
      "entropy": 6.141180896759034,
      "epoch": 6.713450292397661,
      "grad_norm": 1.15625,
      "learning_rate": 0.00018660631372944403,
      "loss": 5.4322,
      "mean_token_accuracy": 0.21030445396900177,
      "num_tokens": 14294018.0,
      "step": 5740
    },
    {
      "entropy": 6.188878107070923,
      "epoch": 6.719298245614035,
      "grad_norm": 1.09375,
      "learning_rate": 0.00018617545447985461,
      "loss": 5.4486,
      "mean_token_accuracy": 0.20217884182929993,
      "num_tokens": 14307023.0,
      "step": 5745
    },
    {
      "entropy": 6.165043067932129,
      "epoch": 6.7251461988304095,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00018574498073516227,
      "loss": 5.525,
      "mean_token_accuracy": 0.19665981978178024,
      "num_tokens": 14319400.0,
      "step": 5750
    },
    {
      "entropy": 6.2768772602081295,
      "epoch": 6.730994152046784,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00018531489436365395,
      "loss": 5.5824,
      "mean_token_accuracy": 0.19099622666835786,
      "num_tokens": 14331288.0,
      "step": 5755
    },
    {
      "entropy": 6.132233715057373,
      "epoch": 6.7368421052631575,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00018488519723193543,
      "loss": 5.428,
      "mean_token_accuracy": 0.2124825745820999,
      "num_tokens": 14344148.0,
      "step": 5760
    },
    {
      "entropy": 6.2191754341125485,
      "epoch": 6.742690058479532,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001844558912049229,
      "loss": 5.5226,
      "mean_token_accuracy": 0.19880881905555725,
      "num_tokens": 14355906.0,
      "step": 5765
    },
    {
      "entropy": 6.235603904724121,
      "epoch": 6.748538011695906,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018402697814583535,
      "loss": 5.5987,
      "mean_token_accuracy": 0.18818530142307283,
      "num_tokens": 14367028.0,
      "step": 5770
    },
    {
      "entropy": 6.2007850170135494,
      "epoch": 6.754385964912281,
      "grad_norm": 1.0625,
      "learning_rate": 0.0001835984599161864,
      "loss": 5.5879,
      "mean_token_accuracy": 0.19322153329849243,
      "num_tokens": 14380873.0,
      "step": 5775
    },
    {
      "entropy": 6.240662956237793,
      "epoch": 6.760233918128655,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00018317033837577564,
      "loss": 5.5738,
      "mean_token_accuracy": 0.18928497433662414,
      "num_tokens": 14394509.0,
      "step": 5780
    },
    {
      "entropy": 6.204741334915161,
      "epoch": 6.76608187134503,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001827426153826814,
      "loss": 5.5394,
      "mean_token_accuracy": 0.19753802716732025,
      "num_tokens": 14406330.0,
      "step": 5785
    },
    {
      "entropy": 6.266202163696289,
      "epoch": 6.771929824561403,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001823152927932521,
      "loss": 5.5435,
      "mean_token_accuracy": 0.19947395771741866,
      "num_tokens": 14418731.0,
      "step": 5790
    },
    {
      "entropy": 6.25153956413269,
      "epoch": 6.777777777777778,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001818883724620985,
      "loss": 5.5154,
      "mean_token_accuracy": 0.20081432908773422,
      "num_tokens": 14430038.0,
      "step": 5795
    },
    {
      "entropy": 6.2457358837127686,
      "epoch": 6.783625730994152,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00018146185624208527,
      "loss": 5.5863,
      "mean_token_accuracy": 0.1932397797703743,
      "num_tokens": 14444212.0,
      "step": 5800
    },
    {
      "entropy": 6.239344930648803,
      "epoch": 6.7894736842105265,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00018103574598432365,
      "loss": 5.5757,
      "mean_token_accuracy": 0.19280090183019638,
      "num_tokens": 14457462.0,
      "step": 5805
    },
    {
      "entropy": 6.259857606887818,
      "epoch": 6.7953216374269,
      "grad_norm": 1.0625,
      "learning_rate": 0.00018061004353816246,
      "loss": 5.4851,
      "mean_token_accuracy": 0.1988669067621231,
      "num_tokens": 14469356.0,
      "step": 5810
    },
    {
      "entropy": 6.160219669342041,
      "epoch": 6.8011695906432745,
      "grad_norm": 1.109375,
      "learning_rate": 0.000180184750751181,
      "loss": 5.4214,
      "mean_token_accuracy": 0.2097303345799446,
      "num_tokens": 14482555.0,
      "step": 5815
    },
    {
      "entropy": 6.142417287826538,
      "epoch": 6.807017543859649,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0001797598694691804,
      "loss": 5.5532,
      "mean_token_accuracy": 0.19918213337659835,
      "num_tokens": 14495215.0,
      "step": 5820
    },
    {
      "entropy": 6.1205222606658936,
      "epoch": 6.812865497076023,
      "grad_norm": 1.0625,
      "learning_rate": 0.00017933540153617579,
      "loss": 5.4292,
      "mean_token_accuracy": 0.2099185198545456,
      "num_tokens": 14506669.0,
      "step": 5825
    },
    {
      "entropy": 6.188602304458618,
      "epoch": 6.818713450292398,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001789113487943888,
      "loss": 5.4977,
      "mean_token_accuracy": 0.19909754097461702,
      "num_tokens": 14518578.0,
      "step": 5830
    },
    {
      "entropy": 6.210914945602417,
      "epoch": 6.824561403508772,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00017848771308423835,
      "loss": 5.5619,
      "mean_token_accuracy": 0.1985200449824333,
      "num_tokens": 14531008.0,
      "step": 5835
    },
    {
      "entropy": 6.281881046295166,
      "epoch": 6.830409356725146,
      "grad_norm": 1.109375,
      "learning_rate": 0.00017806449624433384,
      "loss": 5.6114,
      "mean_token_accuracy": 0.1914947137236595,
      "num_tokens": 14543299.0,
      "step": 5840
    },
    {
      "entropy": 6.158506441116333,
      "epoch": 6.83625730994152,
      "grad_norm": 1.09375,
      "learning_rate": 0.000177641700111467,
      "loss": 5.4574,
      "mean_token_accuracy": 0.20929526537656784,
      "num_tokens": 14556102.0,
      "step": 5845
    },
    {
      "entropy": 6.232896041870117,
      "epoch": 6.842105263157895,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000177219326520603,
      "loss": 5.5467,
      "mean_token_accuracy": 0.19871536195278167,
      "num_tokens": 14569114.0,
      "step": 5850
    },
    {
      "entropy": 6.207945919036865,
      "epoch": 6.847953216374269,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017679737730487378,
      "loss": 5.4876,
      "mean_token_accuracy": 0.20533139109611512,
      "num_tokens": 14582609.0,
      "step": 5855
    },
    {
      "entropy": 6.209119606018066,
      "epoch": 6.853801169590644,
      "grad_norm": 1.140625,
      "learning_rate": 0.00017637585429556912,
      "loss": 5.5248,
      "mean_token_accuracy": 0.2009544238448143,
      "num_tokens": 14594590.0,
      "step": 5860
    },
    {
      "entropy": 6.159265089035034,
      "epoch": 6.859649122807017,
      "grad_norm": 1.015625,
      "learning_rate": 0.00017595475932212908,
      "loss": 5.5126,
      "mean_token_accuracy": 0.19738391190767288,
      "num_tokens": 14607475.0,
      "step": 5865
    },
    {
      "entropy": 6.148343944549561,
      "epoch": 6.8654970760233915,
      "grad_norm": 1.125,
      "learning_rate": 0.00017553409421213606,
      "loss": 5.4105,
      "mean_token_accuracy": 0.21306047588586807,
      "num_tokens": 14620113.0,
      "step": 5870
    },
    {
      "entropy": 6.184271621704101,
      "epoch": 6.871345029239766,
      "grad_norm": 1.140625,
      "learning_rate": 0.00017511386079130666,
      "loss": 5.5285,
      "mean_token_accuracy": 0.20199768394231796,
      "num_tokens": 14632137.0,
      "step": 5875
    },
    {
      "entropy": 6.214907598495484,
      "epoch": 6.87719298245614,
      "grad_norm": 1.25,
      "learning_rate": 0.00017469406088348427,
      "loss": 5.506,
      "mean_token_accuracy": 0.20427055954933165,
      "num_tokens": 14643841.0,
      "step": 5880
    },
    {
      "entropy": 6.116044425964356,
      "epoch": 6.883040935672515,
      "grad_norm": 1.171875,
      "learning_rate": 0.00017427469631063055,
      "loss": 5.4269,
      "mean_token_accuracy": 0.21602101176977156,
      "num_tokens": 14655349.0,
      "step": 5885
    },
    {
      "entropy": 6.186987018585205,
      "epoch": 6.888888888888889,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00017385576889281774,
      "loss": 5.5417,
      "mean_token_accuracy": 0.1949113428592682,
      "num_tokens": 14668761.0,
      "step": 5890
    },
    {
      "entropy": 6.232372140884399,
      "epoch": 6.894736842105263,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017343728044822093,
      "loss": 5.5074,
      "mean_token_accuracy": 0.20397776663303374,
      "num_tokens": 14680216.0,
      "step": 5895
    },
    {
      "entropy": 6.212569713592529,
      "epoch": 6.900584795321637,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00017301923279310996,
      "loss": 5.4853,
      "mean_token_accuracy": 0.19589865803718567,
      "num_tokens": 14691791.0,
      "step": 5900
    },
    {
      "entropy": 6.2556251049041744,
      "epoch": 6.906432748538012,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017260162774184167,
      "loss": 5.5573,
      "mean_token_accuracy": 0.19589219987392426,
      "num_tokens": 14704591.0,
      "step": 5905
    },
    {
      "entropy": 6.163258218765259,
      "epoch": 6.912280701754386,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00017218446710685176,
      "loss": 5.4329,
      "mean_token_accuracy": 0.20993296951055526,
      "num_tokens": 14717469.0,
      "step": 5910
    },
    {
      "entropy": 6.210109090805053,
      "epoch": 6.918128654970761,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017176775269864763,
      "loss": 5.5168,
      "mean_token_accuracy": 0.19539125859737397,
      "num_tokens": 14730121.0,
      "step": 5915
    },
    {
      "entropy": 6.222295045852661,
      "epoch": 6.923976608187134,
      "grad_norm": 1.125,
      "learning_rate": 0.0001713514863257994,
      "loss": 5.5341,
      "mean_token_accuracy": 0.20109213888645172,
      "num_tokens": 14743443.0,
      "step": 5920
    },
    {
      "entropy": 6.201572513580322,
      "epoch": 6.9298245614035086,
      "grad_norm": 1.09375,
      "learning_rate": 0.00017093566979493302,
      "loss": 5.5758,
      "mean_token_accuracy": 0.19699573814868926,
      "num_tokens": 14757208.0,
      "step": 5925
    },
    {
      "entropy": 6.128821897506714,
      "epoch": 6.935672514619883,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0001705203049107223,
      "loss": 5.4535,
      "mean_token_accuracy": 0.19907963573932647,
      "num_tokens": 14770964.0,
      "step": 5930
    },
    {
      "entropy": 6.193312454223633,
      "epoch": 6.941520467836257,
      "grad_norm": 1.125,
      "learning_rate": 0.00017010539347588023,
      "loss": 5.5496,
      "mean_token_accuracy": 0.19719618409872056,
      "num_tokens": 14784042.0,
      "step": 5935
    },
    {
      "entropy": 6.167773199081421,
      "epoch": 6.947368421052632,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00016969093729115253,
      "loss": 5.4289,
      "mean_token_accuracy": 0.20900112837553025,
      "num_tokens": 14797438.0,
      "step": 5940
    },
    {
      "entropy": 6.19411334991455,
      "epoch": 6.953216374269006,
      "grad_norm": 1.21875,
      "learning_rate": 0.00016927693815530854,
      "loss": 5.5245,
      "mean_token_accuracy": 0.20137338936328888,
      "num_tokens": 14808451.0,
      "step": 5945
    },
    {
      "entropy": 6.129976606369018,
      "epoch": 6.95906432748538,
      "grad_norm": 1.125,
      "learning_rate": 0.00016886339786513428,
      "loss": 5.4782,
      "mean_token_accuracy": 0.20450151860713958,
      "num_tokens": 14821479.0,
      "step": 5950
    },
    {
      "entropy": 6.254252099990845,
      "epoch": 6.964912280701754,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0001684503182154244,
      "loss": 5.5457,
      "mean_token_accuracy": 0.19840530902147294,
      "num_tokens": 14835394.0,
      "step": 5955
    },
    {
      "entropy": 6.111571836471557,
      "epoch": 6.970760233918129,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00016803770099897385,
      "loss": 5.4914,
      "mean_token_accuracy": 0.206533682346344,
      "num_tokens": 14847103.0,
      "step": 5960
    },
    {
      "entropy": 6.101902866363526,
      "epoch": 6.976608187134503,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00016762554800657132,
      "loss": 5.4177,
      "mean_token_accuracy": 0.2131612107157707,
      "num_tokens": 14858950.0,
      "step": 5965
    },
    {
      "entropy": 6.152190732955932,
      "epoch": 6.982456140350877,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001672138610269901,
      "loss": 5.4943,
      "mean_token_accuracy": 0.20474812984466553,
      "num_tokens": 14871272.0,
      "step": 5970
    },
    {
      "entropy": 6.197752666473389,
      "epoch": 6.988304093567251,
      "grad_norm": 1.0625,
      "learning_rate": 0.00016680264184698135,
      "loss": 5.4728,
      "mean_token_accuracy": 0.20604657232761384,
      "num_tokens": 14883260.0,
      "step": 5975
    },
    {
      "entropy": 6.196394062042236,
      "epoch": 6.994152046783626,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016639189225126574,
      "loss": 5.5019,
      "mean_token_accuracy": 0.20529556274414062,
      "num_tokens": 14895235.0,
      "step": 5980
    },
    {
      "entropy": 6.190335988998413,
      "epoch": 7.0,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016598161402252596,
      "loss": 5.5818,
      "mean_token_accuracy": 0.19679760336875915,
      "num_tokens": 14906388.0,
      "step": 5985
    },
    {
      "entropy": 6.239471006393432,
      "epoch": 7.005847953216374,
      "grad_norm": 1.0625,
      "learning_rate": 0.00016557180894139888,
      "loss": 5.3713,
      "mean_token_accuracy": 0.20697255283594132,
      "num_tokens": 14918467.0,
      "step": 5990
    },
    {
      "entropy": 6.082650375366211,
      "epoch": 7.011695906432749,
      "grad_norm": 1.125,
      "learning_rate": 0.00016516247878646805,
      "loss": 5.2858,
      "mean_token_accuracy": 0.218429796397686,
      "num_tokens": 14931457.0,
      "step": 5995
    },
    {
      "entropy": 6.214541435241699,
      "epoch": 7.017543859649122,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00016475362533425574,
      "loss": 5.3858,
      "mean_token_accuracy": 0.20891120433807372,
      "num_tokens": 14944639.0,
      "step": 6000
    },
    {
      "epoch": 7.017543859649122,
      "eval_entropy": 6.0832592027377235,
      "eval_loss": 6.5190277099609375,
      "eval_mean_token_accuracy": 0.1524735024545045,
      "eval_num_tokens": 14944639.0,
      "eval_runtime": 1.7372,
      "eval_samples_per_second": 1558.261,
      "eval_steps_per_second": 195.142,
      "step": 6000
    },
    {
      "entropy": 6.163049554824829,
      "epoch": 7.023391812865497,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00016434525035921517,
      "loss": 5.3824,
      "mean_token_accuracy": 0.21207039207220077,
      "num_tokens": 14957896.0,
      "step": 6005
    },
    {
      "entropy": 6.230481147766113,
      "epoch": 7.029239766081871,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00016393735563372292,
      "loss": 5.3791,
      "mean_token_accuracy": 0.2090053752064705,
      "num_tokens": 14970408.0,
      "step": 6010
    },
    {
      "entropy": 6.216822767257691,
      "epoch": 7.035087719298246,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001635299429280715,
      "loss": 5.3926,
      "mean_token_accuracy": 0.20814293920993804,
      "num_tokens": 14983079.0,
      "step": 6015
    },
    {
      "entropy": 6.202263259887696,
      "epoch": 7.04093567251462,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001631230140104611,
      "loss": 5.3764,
      "mean_token_accuracy": 0.21085111945867538,
      "num_tokens": 14996060.0,
      "step": 6020
    },
    {
      "entropy": 6.14192214012146,
      "epoch": 7.046783625730994,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001627165706469927,
      "loss": 5.3842,
      "mean_token_accuracy": 0.21544072777032852,
      "num_tokens": 15008583.0,
      "step": 6025
    },
    {
      "entropy": 6.203445053100586,
      "epoch": 7.052631578947368,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016231061460165926,
      "loss": 5.4206,
      "mean_token_accuracy": 0.20569537878036498,
      "num_tokens": 15019543.0,
      "step": 6030
    },
    {
      "entropy": 6.217844104766845,
      "epoch": 7.058479532163743,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016190514763633946,
      "loss": 5.3465,
      "mean_token_accuracy": 0.21231979727745057,
      "num_tokens": 15031107.0,
      "step": 6035
    },
    {
      "entropy": 6.174035453796387,
      "epoch": 7.064327485380117,
      "grad_norm": 1.078125,
      "learning_rate": 0.00016150017151078894,
      "loss": 5.3915,
      "mean_token_accuracy": 0.20859073400497435,
      "num_tokens": 15043138.0,
      "step": 6040
    },
    {
      "entropy": 6.174532747268676,
      "epoch": 7.0701754385964914,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00016109568798263285,
      "loss": 5.4222,
      "mean_token_accuracy": 0.20810502469539643,
      "num_tokens": 15056042.0,
      "step": 6045
    },
    {
      "entropy": 6.205597829818726,
      "epoch": 7.076023391812866,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000160691698807359,
      "loss": 5.386,
      "mean_token_accuracy": 0.20664302557706832,
      "num_tokens": 15067702.0,
      "step": 6050
    },
    {
      "entropy": 6.211686420440674,
      "epoch": 7.081871345029239,
      "grad_norm": 1.21875,
      "learning_rate": 0.00016028820573830936,
      "loss": 5.4479,
      "mean_token_accuracy": 0.20614961087703704,
      "num_tokens": 15079091.0,
      "step": 6055
    },
    {
      "entropy": 6.137125158309937,
      "epoch": 7.087719298245614,
      "grad_norm": 1.15625,
      "learning_rate": 0.00015988521052667266,
      "loss": 5.3693,
      "mean_token_accuracy": 0.2162208542227745,
      "num_tokens": 15091711.0,
      "step": 6060
    },
    {
      "entropy": 6.219786643981934,
      "epoch": 7.093567251461988,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00015948271492147707,
      "loss": 5.4203,
      "mean_token_accuracy": 0.2065189450979233,
      "num_tokens": 15103512.0,
      "step": 6065
    },
    {
      "entropy": 6.1149077892303465,
      "epoch": 7.099415204678363,
      "grad_norm": 1.03125,
      "learning_rate": 0.0001590807206695824,
      "loss": 5.2922,
      "mean_token_accuracy": 0.2276335820555687,
      "num_tokens": 15115611.0,
      "step": 6070
    },
    {
      "entropy": 6.223895502090454,
      "epoch": 7.105263157894737,
      "grad_norm": 1.09375,
      "learning_rate": 0.00015867922951567253,
      "loss": 5.4744,
      "mean_token_accuracy": 0.20224761962890625,
      "num_tokens": 15128258.0,
      "step": 6075
    },
    {
      "entropy": 6.092599582672119,
      "epoch": 7.111111111111111,
      "grad_norm": 1.125,
      "learning_rate": 0.00015827824320224781,
      "loss": 5.3627,
      "mean_token_accuracy": 0.22167926281690598,
      "num_tokens": 15141730.0,
      "step": 6080
    },
    {
      "entropy": 6.219390249252319,
      "epoch": 7.116959064327485,
      "grad_norm": 1.15625,
      "learning_rate": 0.00015787776346961764,
      "loss": 5.4312,
      "mean_token_accuracy": 0.2061546191573143,
      "num_tokens": 15154435.0,
      "step": 6085
    },
    {
      "entropy": 6.225492286682129,
      "epoch": 7.12280701754386,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00015747779205589269,
      "loss": 5.4886,
      "mean_token_accuracy": 0.20717781782150269,
      "num_tokens": 15167753.0,
      "step": 6090
    },
    {
      "entropy": 6.216479873657226,
      "epoch": 7.128654970760234,
      "grad_norm": 1.03125,
      "learning_rate": 0.0001570783306969778,
      "loss": 5.4319,
      "mean_token_accuracy": 0.20550419986248017,
      "num_tokens": 15179849.0,
      "step": 6095
    },
    {
      "entropy": 6.1215739250183105,
      "epoch": 7.1345029239766085,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00015667938112656372,
      "loss": 5.3707,
      "mean_token_accuracy": 0.21072355955839156,
      "num_tokens": 15192608.0,
      "step": 6100
    },
    {
      "entropy": 6.169468259811401,
      "epoch": 7.140350877192983,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001562809450761201,
      "loss": 5.4177,
      "mean_token_accuracy": 0.2076058030128479,
      "num_tokens": 15204105.0,
      "step": 6105
    },
    {
      "entropy": 6.2086893081665036,
      "epoch": 7.146198830409356,
      "grad_norm": 1.140625,
      "learning_rate": 0.00015588302427488833,
      "loss": 5.3898,
      "mean_token_accuracy": 0.2116716042160988,
      "num_tokens": 15216970.0,
      "step": 6110
    },
    {
      "entropy": 6.0853291034698485,
      "epoch": 7.152046783625731,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001554856204498728,
      "loss": 5.3311,
      "mean_token_accuracy": 0.21869663745164872,
      "num_tokens": 15229517.0,
      "step": 6115
    },
    {
      "entropy": 6.112571763992309,
      "epoch": 7.157894736842105,
      "grad_norm": 1.0625,
      "learning_rate": 0.0001550887353258349,
      "loss": 5.4175,
      "mean_token_accuracy": 0.21306909769773483,
      "num_tokens": 15243904.0,
      "step": 6120
    },
    {
      "entropy": 6.208440065383911,
      "epoch": 7.16374269005848,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00015469237062528445,
      "loss": 5.3956,
      "mean_token_accuracy": 0.20656993985176086,
      "num_tokens": 15255969.0,
      "step": 6125
    },
    {
      "entropy": 6.145543956756592,
      "epoch": 7.169590643274854,
      "grad_norm": 1.0625,
      "learning_rate": 0.00015429652806847266,
      "loss": 5.4307,
      "mean_token_accuracy": 0.20883027464151382,
      "num_tokens": 15268845.0,
      "step": 6130
    },
    {
      "entropy": 6.200758790969848,
      "epoch": 7.175438596491228,
      "grad_norm": 1.1875,
      "learning_rate": 0.00015390120937338459,
      "loss": 5.431,
      "mean_token_accuracy": 0.20468081384897233,
      "num_tokens": 15281820.0,
      "step": 6135
    },
    {
      "entropy": 6.130422306060791,
      "epoch": 7.181286549707602,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0001535064162557318,
      "loss": 5.3522,
      "mean_token_accuracy": 0.20791111141443253,
      "num_tokens": 15294746.0,
      "step": 6140
    },
    {
      "entropy": 6.154082632064819,
      "epoch": 7.187134502923977,
      "grad_norm": 1.140625,
      "learning_rate": 0.00015311215042894462,
      "loss": 5.4563,
      "mean_token_accuracy": 0.20927639603614806,
      "num_tokens": 15306590.0,
      "step": 6145
    },
    {
      "entropy": 6.225092792510987,
      "epoch": 7.192982456140351,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00015271841360416506,
      "loss": 5.4522,
      "mean_token_accuracy": 0.20625718235969542,
      "num_tokens": 15319318.0,
      "step": 6150
    },
    {
      "entropy": 6.204297685623169,
      "epoch": 7.1988304093567255,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00015232520749023924,
      "loss": 5.4093,
      "mean_token_accuracy": 0.2158847600221634,
      "num_tokens": 15332688.0,
      "step": 6155
    },
    {
      "entropy": 6.120451641082764,
      "epoch": 7.204678362573099,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001519325337937098,
      "loss": 5.2913,
      "mean_token_accuracy": 0.2169152095913887,
      "num_tokens": 15344711.0,
      "step": 6160
    },
    {
      "entropy": 6.133221578598023,
      "epoch": 7.2105263157894735,
      "grad_norm": 1.125,
      "learning_rate": 0.0001515403942188088,
      "loss": 5.3917,
      "mean_token_accuracy": 0.21365178376436234,
      "num_tokens": 15357848.0,
      "step": 6165
    },
    {
      "entropy": 6.185316658020019,
      "epoch": 7.216374269005848,
      "grad_norm": 1.09375,
      "learning_rate": 0.00015114879046745005,
      "loss": 5.4127,
      "mean_token_accuracy": 0.20901222974061967,
      "num_tokens": 15371088.0,
      "step": 6170
    },
    {
      "entropy": 6.151632404327392,
      "epoch": 7.222222222222222,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001507577242392219,
      "loss": 5.4298,
      "mean_token_accuracy": 0.20446956604719163,
      "num_tokens": 15382810.0,
      "step": 6175
    },
    {
      "entropy": 6.114848709106445,
      "epoch": 7.228070175438597,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00015036719723138006,
      "loss": 5.3453,
      "mean_token_accuracy": 0.21931500881910324,
      "num_tokens": 15395752.0,
      "step": 6180
    },
    {
      "entropy": 6.177680253982544,
      "epoch": 7.23391812865497,
      "grad_norm": 1.1875,
      "learning_rate": 0.00014997721113883946,
      "loss": 5.3846,
      "mean_token_accuracy": 0.21278643757104873,
      "num_tokens": 15407318.0,
      "step": 6185
    },
    {
      "entropy": 6.216420745849609,
      "epoch": 7.239766081871345,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00014958776765416776,
      "loss": 5.4663,
      "mean_token_accuracy": 0.20376016348600387,
      "num_tokens": 15420274.0,
      "step": 6190
    },
    {
      "entropy": 6.125368309020996,
      "epoch": 7.245614035087719,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001491988684675778,
      "loss": 5.3758,
      "mean_token_accuracy": 0.2160023793578148,
      "num_tokens": 15433081.0,
      "step": 6195
    },
    {
      "entropy": 6.156131219863892,
      "epoch": 7.251461988304094,
      "grad_norm": 1.1875,
      "learning_rate": 0.00014881051526691963,
      "loss": 5.3846,
      "mean_token_accuracy": 0.216105680167675,
      "num_tokens": 15445317.0,
      "step": 6200
    },
    {
      "entropy": 6.188797044754028,
      "epoch": 7.257309941520468,
      "grad_norm": 1.0625,
      "learning_rate": 0.00014842270973767436,
      "loss": 5.4411,
      "mean_token_accuracy": 0.20467743426561355,
      "num_tokens": 15457146.0,
      "step": 6205
    },
    {
      "entropy": 6.092201995849609,
      "epoch": 7.2631578947368425,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001480354535629454,
      "loss": 5.2337,
      "mean_token_accuracy": 0.22646497935056686,
      "num_tokens": 15471048.0,
      "step": 6210
    },
    {
      "entropy": 6.176982402801514,
      "epoch": 7.269005847953216,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001476487484234526,
      "loss": 5.4085,
      "mean_token_accuracy": 0.2101561263203621,
      "num_tokens": 15484260.0,
      "step": 6215
    },
    {
      "entropy": 6.167039680480957,
      "epoch": 7.2748538011695905,
      "grad_norm": 1.15625,
      "learning_rate": 0.000147262595997524,
      "loss": 5.3715,
      "mean_token_accuracy": 0.20882519334554672,
      "num_tokens": 15495535.0,
      "step": 6220
    },
    {
      "entropy": 6.1125345706939695,
      "epoch": 7.280701754385965,
      "grad_norm": 1.15625,
      "learning_rate": 0.00014687699796108854,
      "loss": 5.3486,
      "mean_token_accuracy": 0.21365023404359818,
      "num_tokens": 15508433.0,
      "step": 6225
    },
    {
      "entropy": 6.175793981552124,
      "epoch": 7.286549707602339,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00014649195598766967,
      "loss": 5.4684,
      "mean_token_accuracy": 0.20011112242937087,
      "num_tokens": 15520871.0,
      "step": 6230
    },
    {
      "entropy": 6.1660017490386965,
      "epoch": 7.292397660818714,
      "grad_norm": 1.109375,
      "learning_rate": 0.00014610747174837702,
      "loss": 5.4109,
      "mean_token_accuracy": 0.20825264006853103,
      "num_tokens": 15532225.0,
      "step": 6235
    },
    {
      "entropy": 6.156658029556274,
      "epoch": 7.298245614035087,
      "grad_norm": 1.171875,
      "learning_rate": 0.00014572354691189977,
      "loss": 5.4023,
      "mean_token_accuracy": 0.214142906665802,
      "num_tokens": 15544147.0,
      "step": 6240
    },
    {
      "entropy": 6.143483352661133,
      "epoch": 7.304093567251462,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001453401831444992,
      "loss": 5.3095,
      "mean_token_accuracy": 0.21573384702205659,
      "num_tokens": 15555788.0,
      "step": 6245
    },
    {
      "entropy": 6.2020348072052,
      "epoch": 7.309941520467836,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014495738211000165,
      "loss": 5.4039,
      "mean_token_accuracy": 0.20989190489053727,
      "num_tokens": 15568268.0,
      "step": 6250
    },
    {
      "entropy": 6.145077800750732,
      "epoch": 7.315789473684211,
      "grad_norm": 1.109375,
      "learning_rate": 0.000144575145469791,
      "loss": 5.3463,
      "mean_token_accuracy": 0.2152687817811966,
      "num_tokens": 15580247.0,
      "step": 6255
    },
    {
      "entropy": 6.085100317001343,
      "epoch": 7.321637426900585,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001441934748828016,
      "loss": 5.366,
      "mean_token_accuracy": 0.2103113055229187,
      "num_tokens": 15592755.0,
      "step": 6260
    },
    {
      "entropy": 6.127097702026367,
      "epoch": 7.3274853801169595,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014381237200551138,
      "loss": 5.3942,
      "mean_token_accuracy": 0.21347033828496934,
      "num_tokens": 15603721.0,
      "step": 6265
    },
    {
      "entropy": 6.067697811126709,
      "epoch": 7.333333333333333,
      "grad_norm": 1.15625,
      "learning_rate": 0.00014343183849193402,
      "loss": 5.3546,
      "mean_token_accuracy": 0.21975378096103668,
      "num_tokens": 15616770.0,
      "step": 6270
    },
    {
      "entropy": 6.097276496887207,
      "epoch": 7.3391812865497075,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00014305187599361223,
      "loss": 5.2998,
      "mean_token_accuracy": 0.22045280635356904,
      "num_tokens": 15630382.0,
      "step": 6275
    },
    {
      "entropy": 6.104697608947754,
      "epoch": 7.345029239766082,
      "grad_norm": 1.171875,
      "learning_rate": 0.00014267248615961047,
      "loss": 5.3066,
      "mean_token_accuracy": 0.21113628000020981,
      "num_tokens": 15642750.0,
      "step": 6280
    },
    {
      "entropy": 6.195154666900635,
      "epoch": 7.350877192982456,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00014229367063650788,
      "loss": 5.4711,
      "mean_token_accuracy": 0.20148903280496597,
      "num_tokens": 15655066.0,
      "step": 6285
    },
    {
      "entropy": 6.1207396507263185,
      "epoch": 7.356725146198831,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00014191543106839116,
      "loss": 5.3719,
      "mean_token_accuracy": 0.2106843739748001,
      "num_tokens": 15667608.0,
      "step": 6290
    },
    {
      "entropy": 6.1363441944122314,
      "epoch": 7.362573099415204,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00014153776909684685,
      "loss": 5.3289,
      "mean_token_accuracy": 0.21957481056451797,
      "num_tokens": 15679457.0,
      "step": 6295
    },
    {
      "entropy": 6.180751132965088,
      "epoch": 7.368421052631579,
      "grad_norm": 1.21875,
      "learning_rate": 0.00014116068636095522,
      "loss": 5.4311,
      "mean_token_accuracy": 0.20529936403036117,
      "num_tokens": 15691542.0,
      "step": 6300
    },
    {
      "entropy": 6.21274037361145,
      "epoch": 7.374269005847953,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00014078418449728236,
      "loss": 5.429,
      "mean_token_accuracy": 0.20624616444110871,
      "num_tokens": 15703411.0,
      "step": 6305
    },
    {
      "entropy": 6.185572624206543,
      "epoch": 7.380116959064328,
      "grad_norm": 1.125,
      "learning_rate": 0.00014040826513987318,
      "loss": 5.4653,
      "mean_token_accuracy": 0.19596865326166152,
      "num_tokens": 15716652.0,
      "step": 6310
    },
    {
      "entropy": 6.144858074188233,
      "epoch": 7.385964912280702,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001400329299202448,
      "loss": 5.3948,
      "mean_token_accuracy": 0.2148033156991005,
      "num_tokens": 15729555.0,
      "step": 6315
    },
    {
      "entropy": 6.161768817901612,
      "epoch": 7.391812865497076,
      "grad_norm": 1.0625,
      "learning_rate": 0.000139658180467379,
      "loss": 5.4004,
      "mean_token_accuracy": 0.2130689114332199,
      "num_tokens": 15742868.0,
      "step": 6320
    },
    {
      "entropy": 6.239882707595825,
      "epoch": 7.39766081871345,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001392840184077152,
      "loss": 5.4669,
      "mean_token_accuracy": 0.20411383956670762,
      "num_tokens": 15754803.0,
      "step": 6325
    },
    {
      "entropy": 6.13710503578186,
      "epoch": 7.4035087719298245,
      "grad_norm": 1.109375,
      "learning_rate": 0.00013891044536514367,
      "loss": 5.3701,
      "mean_token_accuracy": 0.21086589246988297,
      "num_tokens": 15766887.0,
      "step": 6330
    },
    {
      "entropy": 6.172860145568848,
      "epoch": 7.409356725146199,
      "grad_norm": 1.171875,
      "learning_rate": 0.00013853746296099812,
      "loss": 5.422,
      "mean_token_accuracy": 0.20632522255182267,
      "num_tokens": 15778020.0,
      "step": 6335
    },
    {
      "entropy": 6.193747329711914,
      "epoch": 7.415204678362573,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00013816507281404894,
      "loss": 5.3879,
      "mean_token_accuracy": 0.20873268991708754,
      "num_tokens": 15789993.0,
      "step": 6340
    },
    {
      "entropy": 6.182000017166137,
      "epoch": 7.421052631578947,
      "grad_norm": 1.125,
      "learning_rate": 0.00013779327654049606,
      "loss": 5.4538,
      "mean_token_accuracy": 0.21100408732891082,
      "num_tokens": 15802360.0,
      "step": 6345
    },
    {
      "entropy": 6.223145341873169,
      "epoch": 7.426900584795321,
      "grad_norm": 1.234375,
      "learning_rate": 0.000137422075753962,
      "loss": 5.4816,
      "mean_token_accuracy": 0.19950181096792222,
      "num_tokens": 15815526.0,
      "step": 6350
    },
    {
      "entropy": 6.23135027885437,
      "epoch": 7.432748538011696,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001370514720654848,
      "loss": 5.5387,
      "mean_token_accuracy": 0.1967550992965698,
      "num_tokens": 15829207.0,
      "step": 6355
    },
    {
      "entropy": 6.173253345489502,
      "epoch": 7.43859649122807,
      "grad_norm": 1.15625,
      "learning_rate": 0.00013668146708351104,
      "loss": 5.374,
      "mean_token_accuracy": 0.21200289726257324,
      "num_tokens": 15842067.0,
      "step": 6360
    },
    {
      "entropy": 6.185382080078125,
      "epoch": 7.444444444444445,
      "grad_norm": 1.21875,
      "learning_rate": 0.00013631206241388881,
      "loss": 5.3882,
      "mean_token_accuracy": 0.2071247488260269,
      "num_tokens": 15854994.0,
      "step": 6365
    },
    {
      "entropy": 6.1425392627716064,
      "epoch": 7.450292397660819,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00013594325965986088,
      "loss": 5.3765,
      "mean_token_accuracy": 0.21060512363910674,
      "num_tokens": 15867852.0,
      "step": 6370
    },
    {
      "entropy": 6.18590898513794,
      "epoch": 7.456140350877193,
      "grad_norm": 1.125,
      "learning_rate": 0.00013557506042205782,
      "loss": 5.4455,
      "mean_token_accuracy": 0.20474697798490524,
      "num_tokens": 15880815.0,
      "step": 6375
    },
    {
      "entropy": 6.119968461990356,
      "epoch": 7.461988304093567,
      "grad_norm": 1.265625,
      "learning_rate": 0.00013520746629849044,
      "loss": 5.3554,
      "mean_token_accuracy": 0.21304798424243926,
      "num_tokens": 15892515.0,
      "step": 6380
    },
    {
      "entropy": 6.135337924957275,
      "epoch": 7.4678362573099415,
      "grad_norm": 1.078125,
      "learning_rate": 0.00013484047888454387,
      "loss": 5.3409,
      "mean_token_accuracy": 0.21450725495815276,
      "num_tokens": 15905782.0,
      "step": 6385
    },
    {
      "entropy": 6.082184314727783,
      "epoch": 7.473684210526316,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001344740997729697,
      "loss": 5.3007,
      "mean_token_accuracy": 0.22068435847759246,
      "num_tokens": 15918901.0,
      "step": 6390
    },
    {
      "entropy": 6.214112854003906,
      "epoch": 7.47953216374269,
      "grad_norm": 1.140625,
      "learning_rate": 0.00013410833055387943,
      "loss": 5.4128,
      "mean_token_accuracy": 0.20954852551221848,
      "num_tokens": 15931751.0,
      "step": 6395
    },
    {
      "entropy": 6.192382049560547,
      "epoch": 7.485380116959064,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00013374317281473783,
      "loss": 5.3826,
      "mean_token_accuracy": 0.21115387678146363,
      "num_tokens": 15944009.0,
      "step": 6400
    },
    {
      "entropy": 6.152518463134766,
      "epoch": 7.491228070175438,
      "grad_norm": 1.1875,
      "learning_rate": 0.00013337862814035564,
      "loss": 5.3697,
      "mean_token_accuracy": 0.20478594452142715,
      "num_tokens": 15956642.0,
      "step": 6405
    },
    {
      "entropy": 6.186007976531982,
      "epoch": 7.497076023391813,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00013301469811288294,
      "loss": 5.4229,
      "mean_token_accuracy": 0.20723483711481094,
      "num_tokens": 15969164.0,
      "step": 6410
    },
    {
      "entropy": 6.190393781661987,
      "epoch": 7.502923976608187,
      "grad_norm": 1.0625,
      "learning_rate": 0.00013265138431180207,
      "loss": 5.4652,
      "mean_token_accuracy": 0.20336130261421204,
      "num_tokens": 15981650.0,
      "step": 6415
    },
    {
      "entropy": 6.199639892578125,
      "epoch": 7.508771929824562,
      "grad_norm": 1.140625,
      "learning_rate": 0.000132288688313921,
      "loss": 5.3455,
      "mean_token_accuracy": 0.2151162475347519,
      "num_tokens": 15995310.0,
      "step": 6420
    },
    {
      "entropy": 6.091612195968628,
      "epoch": 7.514619883040936,
      "grad_norm": 1.25,
      "learning_rate": 0.00013192661169336638,
      "loss": 5.3196,
      "mean_token_accuracy": 0.2241883546113968,
      "num_tokens": 16007811.0,
      "step": 6425
    },
    {
      "entropy": 6.147864723205567,
      "epoch": 7.52046783625731,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00013156515602157664,
      "loss": 5.4089,
      "mean_token_accuracy": 0.20735065042972564,
      "num_tokens": 16020136.0,
      "step": 6430
    },
    {
      "entropy": 6.172757148742676,
      "epoch": 7.526315789473684,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013120432286729548,
      "loss": 5.4146,
      "mean_token_accuracy": 0.2053794741630554,
      "num_tokens": 16031860.0,
      "step": 6435
    },
    {
      "entropy": 6.102254343032837,
      "epoch": 7.5321637426900585,
      "grad_norm": 1.171875,
      "learning_rate": 0.00013084411379656453,
      "loss": 5.3486,
      "mean_token_accuracy": 0.21875980198383332,
      "num_tokens": 16044544.0,
      "step": 6440
    },
    {
      "entropy": 6.176325464248658,
      "epoch": 7.538011695906433,
      "grad_norm": 1.15625,
      "learning_rate": 0.00013048453037271696,
      "loss": 5.439,
      "mean_token_accuracy": 0.20508948266506194,
      "num_tokens": 16058467.0,
      "step": 6445
    },
    {
      "entropy": 6.218649768829346,
      "epoch": 7.543859649122807,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00013012557415637073,
      "loss": 5.4123,
      "mean_token_accuracy": 0.20697457790374757,
      "num_tokens": 16070379.0,
      "step": 6450
    },
    {
      "entropy": 6.150947046279907,
      "epoch": 7.549707602339181,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001297672467054214,
      "loss": 5.3891,
      "mean_token_accuracy": 0.20964059233665466,
      "num_tokens": 16082949.0,
      "step": 6455
    },
    {
      "entropy": 6.212559175491333,
      "epoch": 7.555555555555555,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001294095495750361,
      "loss": 5.4511,
      "mean_token_accuracy": 0.20171633660793303,
      "num_tokens": 16096281.0,
      "step": 6460
    },
    {
      "entropy": 6.161518859863281,
      "epoch": 7.56140350877193,
      "grad_norm": 1.140625,
      "learning_rate": 0.00012905248431764566,
      "loss": 5.3526,
      "mean_token_accuracy": 0.21033980697393417,
      "num_tokens": 16108248.0,
      "step": 6465
    },
    {
      "entropy": 6.143288087844849,
      "epoch": 7.567251461988304,
      "grad_norm": 1.046875,
      "learning_rate": 0.00012869605248293914,
      "loss": 5.4057,
      "mean_token_accuracy": 0.21002519577741624,
      "num_tokens": 16121659.0,
      "step": 6470
    },
    {
      "entropy": 6.155509805679321,
      "epoch": 7.573099415204679,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001283402556178563,
      "loss": 5.4645,
      "mean_token_accuracy": 0.20809196829795837,
      "num_tokens": 16133049.0,
      "step": 6475
    },
    {
      "entropy": 6.202949857711792,
      "epoch": 7.578947368421053,
      "grad_norm": 1.15625,
      "learning_rate": 0.00012798509526658076,
      "loss": 5.4249,
      "mean_token_accuracy": 0.20736961364746093,
      "num_tokens": 16144821.0,
      "step": 6480
    },
    {
      "entropy": 6.204100799560547,
      "epoch": 7.584795321637427,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00012763057297053427,
      "loss": 5.4553,
      "mean_token_accuracy": 0.20841256082057952,
      "num_tokens": 16158704.0,
      "step": 6485
    },
    {
      "entropy": 6.1512267112731935,
      "epoch": 7.590643274853801,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00012727669026836862,
      "loss": 5.3934,
      "mean_token_accuracy": 0.2098011255264282,
      "num_tokens": 16171500.0,
      "step": 6490
    },
    {
      "entropy": 6.075776767730713,
      "epoch": 7.5964912280701755,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00012692344869596054,
      "loss": 5.2811,
      "mean_token_accuracy": 0.21859795302152635,
      "num_tokens": 16182746.0,
      "step": 6495
    },
    {
      "entropy": 6.2071661949157715,
      "epoch": 7.60233918128655,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001265708497864035,
      "loss": 5.4337,
      "mean_token_accuracy": 0.20379527062177658,
      "num_tokens": 16195226.0,
      "step": 6500
    },
    {
      "epoch": 7.60233918128655,
      "eval_entropy": 6.098659182016829,
      "eval_loss": 6.52553129196167,
      "eval_mean_token_accuracy": 0.1521863601002897,
      "eval_num_tokens": 16195226.0,
      "eval_runtime": 1.7381,
      "eval_samples_per_second": 1557.409,
      "eval_steps_per_second": 195.036,
      "step": 6500
    }
  ],
  "logging_steps": 5,
  "max_steps": 8540,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3479808015544320.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}